examples/Sionna_tutorial_part1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f543de9b-3858-45dd-bddf-cddddcc60da3",
   "metadata": {},
   "source": [
    "# Part 1: Getting Started with Sionna"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c55670f1-3ee6-481d-9661-d59e61074818",
   "metadata": {},
   "source": [
    "This tutorial will guide you through Sionna, from its basic principles to the implementation of a point-to-point link with a 5G NR compliant code and a 3GPP channel model.\n",
    "You will also learn how to write custom trainable layers by implementing a state of the art neural receiver, and how to train and evaluate end-to-end communication systems.\n",
    "\n",
    "The tutorial is structured in four notebooks:\n",
    "\n",
    "- **Part I: Getting started with Sionna**\n",
    "\n",
    "- Part II: Differentiable Communication Systems\n",
    "\n",
    "- Part III: Advanced Link-level Simulations\n",
    "\n",
    "- Part IV: Toward Learned Receivers\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b226f39",
   "metadata": {},
   "source": [
    "The [official documentation](https://nvlabs.github.io/sionna) provides key material on how to use Sionna and how its components are implemented."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d7c4211f-beee-4142-9289-c385d0180877",
   "metadata": {},
   "source": [
    "* [Imports & Basics](#Imports-&-Basics)\n",
    "* [Sionna Data-flow and Design Paradigms](#Sionna-Data-flow-and-Design-Paradigms)\n",
    "* [Hello, Sionna!](#Hello,-Sionna!)\n",
    "* [Communication Systems as Keras Models](#Communication-Systems-as-Keras-Models)\n",
    "* [Forward Error Correction](#Forward-Error-Correction-(FEC))\n",
    "* [Eager vs. Graph Mode](#Eager-vs-Graph-Mode)\n",
    "* [Exercise](#Exercise)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74a184ac-1f64-407f-9a24-c53d40799be2",
   "metadata": {},
   "source": [
    "## Imports & Basics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d84bd69e-59f2-4a42-8dd3-730c8c1d821e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import TensorFlow and NumPy\n",
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "\n",
    "# Import Sionna\n",
    "try:\n",
    "    import sionna as sn\n",
    "except ImportError as e:\n",
    "    # Install Sionna if package is not already installed\n",
    "    import os\n",
    "    os.system(\"pip install sionna\")\n",
    "    import sionna as sn\n",
    "\n",
    "# For plotting\n",
    "%matplotlib inline \n",
    "# also try %matplotlib widget\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# for performance measurements \n",
    "import time\n",
    "\n",
    "# For the implementation of the Keras models\n",
    "from tensorflow.keras import Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42d471a3",
   "metadata": {},
   "source": [
    "We can now access Sionna functions within the `sn` namespace.\n",
    "\n",
    "**Hint**: In Jupyter notebooks, you can run bash commands with `!`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ec4eb0ab-a34e-45e7-9055-a1555d6f0775",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tue Mar 15 14:47:45 2022       \n",
      "+-----------------------------------------------------------------------------+\n",
      "| NVIDIA-SMI 470.103.01   Driver Version: 470.103.01   CUDA Version: 11.4     |\n",
      "|-------------------------------+----------------------+----------------------+\n",
      "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
      "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
      "|                               |                      |               MIG M. |\n",
      "|===============================+======================+======================|\n",
      "|   0  NVIDIA GeForce ...  Off  | 00000000:01:00.0 Off |                  N/A |\n",
      "| 30%   51C    P8    23W / 350W |     53MiB / 24265MiB |      0%      Default |\n",
      "|                               |                      |                  N/A |\n",
      "+-------------------------------+----------------------+----------------------+\n",
      "|   1  NVIDIA GeForce ...  Off  | 00000000:4C:00.0 Off |                  N/A |\n",
      "|  0%   33C    P8    24W / 350W |      8MiB / 24268MiB |      0%      Default |\n",
      "|                               |                      |                  N/A |\n",
      "+-------------------------------+----------------------+----------------------+\n",
      "                                                                               \n",
      "+-----------------------------------------------------------------------------+\n",
      "| Processes:                                                                  |\n",
      "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
      "|        ID   ID                                                   Usage      |\n",
      "|=============================================================================|\n",
      "+-----------------------------------------------------------------------------+\n"
     ]
    }
   ],
   "source": [
    "!nvidia-smi"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6294919-0e78-4d6b-835f-55631e6264f6",
   "metadata": {},
   "source": [
    "## Sionna Data-flow and Design Paradigms\n",
    "\n",
    "Sionna inherently parallelizes simulations via *batching*, i.e., each element in the batch dimension is simulated independently.\n",
    "\n",
    "This means the first tensor dimension is always used for *inter-frame* parallelization similar to an outer *for-loop* in Matlab/NumPy simulations, but operations can be operated in parallel.\n",
    "\n",
    "To keep the dataflow efficient, Sionna follows a few simple design principles:\n",
    "\n",
    "* Signal-processing components are implemented as an individual [Keras layer](https://keras.io/api/layers/).\n",
    "* `tf.float32` is used as preferred datatype and `tf.complex64` for complex-valued datatypes, respectively.  \n",
    "This allows simpler re-use of components (e.g., the same scrambling layer can be used for binary inputs and LLR-values).\n",
    "* `tf.float64`/`tf.complex128` are available when high precision is needed.\n",
    "* Models can be developed in *eager mode* allowing simple (and fast) modification of system parameters.\n",
    "* Number crunching simulations can be executed in the faster *graph mode* or even *XLA* acceleration (experimental) is available for most components.\n",
    "* Whenever possible, components are automatically differentiable via [auto-grad](https://www.tensorflow.org/guide/autodiff) to simplify the deep learning design-flow.\n",
    "* Code is structured into sub-packages for different tasks such as channel coding, mapping,... (see [API documentation](http://nvlabs.github.io/sionna/api/sionna.html) for details).\n",
    "\n",
    "These paradigms simplify the re-useability and reliability of our components for a wide range of communications related applications."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fce869c5-5169-4c66-833c-8144414cdda4",
   "metadata": {},
   "source": [
    "## Hello, Sionna!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b52449ca-532c-4c45-8e03-464832717200",
   "metadata": {},
   "source": [
    "Let's start with a very simple simulation: Transmitting QAM symbols over an AWGN channel. We will implement the system shown in the figure below."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "576ea7ca-3dda-4c09-868d-1e2440d35b00",
   "metadata": {},
   "source": [
    "![QAM AWGN](data:image/png;base64,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)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "777704cf-0855-4984-86df-b615890811ae",
   "metadata": {},
   "source": [
    "We will use upper case for naming simulation parameters that are used throughout this notebook\n",
    "\n",
    "Every layer needs to be initialized once before it can be used.\n",
    "\n",
    "**Tip**: Use the [API documentation](http://nvlabs.github.io/sionna/api/sionna.html) to find an overview of all existing components.\n",
    "You can directly access the signature and the docstring within jupyter via `Shift+TAB`.\n",
    "\n",
    "*Remark*: Most layers are defined to be complex-valued.\n",
    "\n",
    "We first need to create a QAM constellation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "21ccb7b5-b590-4abb-95f7-d939cb283f02",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "NUM_BITS_PER_SYMBOL = 2 # QPSK\n",
    "constellation = sn.mapping.Constellation(\"qam\", NUM_BITS_PER_SYMBOL)\n",
    "\n",
    "constellation.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ded88c1-1576-42a5-a180-5bd8da533470",
   "metadata": {
    "tags": []
   },
   "source": [
    "**Task:** Try to change the modulation order, e.g., to 16-QAM."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dcb7b6f1-5dbc-4335-b633-177da8e8a5e3",
   "metadata": {
    "tags": []
   },
   "source": [
    "We then need to setup a mapper to map bits into constellation points. The mapper takes as parameter the constellation.\n",
    "\n",
    "We also need to setup a corresponding demapper to compute log-likelihood ratios (LLRs) from received noisy samples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ae9da488-bd0b-40d7-b9fd-31bc6ad7ca3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "mapper = sn.mapping.Mapper(constellation=constellation)\n",
    "\n",
    "# The demapper uses the same constellation object as the mapper\n",
    "demapper = sn.mapping.Demapper(\"app\", constellation=constellation) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "980fe9ce",
   "metadata": {},
   "source": [
    "**Tip**: You can access the signature+docstring via `?` command and print the complete class definition via `??` operator.\n",
    "\n",
    "Obviously, you can also access the source code via [https://github.com/nvlabs/sionna/](https://github.com/nvlabs/sionna/)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "10f6b1c2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[0;31mInit signature:\u001b[0m \u001b[0msn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmapping\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mMapper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mSource:\u001b[0m        \n",
      "\u001b[0;32mclass\u001b[0m \u001b[0mMapper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mLayer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m    \u001b[0;31m# pylint: disable=line-too-long\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m    \u001b[0;34mr\"\"\"\u001b[0m\n",
      "\u001b[0;34m    Mapper(constellation_type=None, num_bits_per_symbol=None, constellation=None, dtype=tf.complex64, **kwargs)\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m    Maps binary tensors to points of a constellation.\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m    This class defines a layer that maps a tensor of binary values\u001b[0m\n",
      "\u001b[0;34m    to a tensor of points from a provided constellation.\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m    Parameters\u001b[0m\n",
      "\u001b[0;34m    ----------\u001b[0m\n",
      "\u001b[0;34m    constellation_type : One of [\"qam\", \"pam\", \"custom\"], str\u001b[0m\n",
      "\u001b[0;34m        For \"custom\", an instance of :class:`~sionna.mapping.Constellation`\u001b[0m\n",
      "\u001b[0;34m        must be provided.\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m    num_bits_per_symbol : int\u001b[0m\n",
      "\u001b[0;34m        The number of bits per constellation symbol, e.g., 4 for QAM16.\u001b[0m\n",
      "\u001b[0;34m        Only required for ``constellation_type`` in [\"qam\", \"pam\"].\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m    constellation :  Constellation\u001b[0m\n",
      "\u001b[0;34m        An instance of :class:`~sionna.mapping.Constellation` or\u001b[0m\n",
      "\u001b[0;34m        `None`. In the latter case, ``constellation_type``\u001b[0m\n",
      "\u001b[0;34m        and ``num_bits_per_symbol`` must be provided.\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m    dtype : One of [tf.complex64, tf.complex128], tf.DType\u001b[0m\n",
      "\u001b[0;34m        The output dtype. Defaults to tf.complex64.\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m    Input\u001b[0m\n",
      "\u001b[0;34m    -----\u001b[0m\n",
      "\u001b[0;34m    : [..., n], tf.float or tf.int\u001b[0m\n",
      "\u001b[0;34m        Tensor with with binary entries.\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m    Output\u001b[0m\n",
      "\u001b[0;34m    ------\u001b[0m\n",
      "\u001b[0;34m    : [...,n/Constellation.num_bits_per_symbol], tf.complex\u001b[0m\n",
      "\u001b[0;34m        The mapped constellation symbols.\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m    Note\u001b[0m\n",
      "\u001b[0;34m    ----\u001b[0m\n",
      "\u001b[0;34m    The last input dimension must be an integer multiple of the\u001b[0m\n",
      "\u001b[0;34m    number of bits per constellation symbol.\u001b[0m\n",
      "\u001b[0;34m    \"\"\"\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m    \u001b[0;32mdef\u001b[0m \u001b[0m__init__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m                 \u001b[0mconstellation_type\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m                 \u001b[0mnum_bits_per_symbol\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m                 \u001b[0mconstellation\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m                 \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcomplex64\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m                 \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m                \u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0msuper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__init__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;32massert\u001b[0m \u001b[0mdtype\u001b[0m \u001b[0;32min\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcomplex64\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcomplex128\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\\\n",
      "            \u001b[0;34m\"dtype must be tf.complex64 or tf.complex128\"\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;31m#self._dtype = dtype\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;31m#print(dtype, self._dtype)\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;32mif\u001b[0m \u001b[0mconstellation\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m            \u001b[0;32massert\u001b[0m \u001b[0mconstellation_type\u001b[0m \u001b[0;32min\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"custom\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \\\n",
      "                \u001b[0;34m\"\"\"`constellation_type` must be \"custom\".\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m            \u001b[0;32massert\u001b[0m \u001b[0mnum_bits_per_symbol\u001b[0m \u001b[0;32min\u001b[0m \\\n",
      "                     \u001b[0;34m[\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconstellation\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnum_bits_per_symbol\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \\\n",
      "                \u001b[0;34m\"\"\"`Wrong value of `num_bits_per_symbol.`\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m            \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_constellation\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mconstellation\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m            \u001b[0;32massert\u001b[0m \u001b[0mconstellation_type\u001b[0m \u001b[0;32min\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m\"qam\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"pam\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \\\n",
      "                \u001b[0;34m\"Wrong constellation type.\"\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m            \u001b[0;32massert\u001b[0m \u001b[0mnum_bits_per_symbol\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \\\n",
      "                \u001b[0;34m\"`num_bits_per_symbol` must be provided.\"\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m            \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_constellation\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mConstellation\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconstellation_type\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m                                                \u001b[0mnum_bits_per_symbol\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m                                                \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dtype\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_binary_base\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m                        \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstellation\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnum_bits_per_symbol\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m    \u001b[0;34m@\u001b[0m\u001b[0mproperty\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m    \u001b[0;32mdef\u001b[0m \u001b[0mconstellation\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;34m\"\"\"The Constellation used by the Mapper.\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_constellation\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m    \u001b[0;32mdef\u001b[0m \u001b[0mcall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebugging\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0massert_greater_equal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrank\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m            \u001b[0mmessage\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"The input must have at least rank 2\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;31m# Reshape inputs to the desired format\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0mnew_shape\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \\\n",
      "           \u001b[0;34m[\u001b[0m\u001b[0mint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstellation\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnum_bits_per_symbol\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m            \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstellation\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnum_bits_per_symbol\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0minputs_reshaped\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcast\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnew_shape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mint32\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;31m# Convert the last dimension to an integer\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0mint_rep\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreduce_sum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs_reshaped\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_binary_base\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;31m# Map integers to constellation symbols\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgather\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstellation\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpoints\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mint_rep\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\n",
      "\u001b[0;34m\u001b[0m        \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mFile:\u001b[0m           ~/.local/lib/python3.8/site-packages/sionna/mapping.py\n",
      "\u001b[0;31mType:\u001b[0m           type\n",
      "\u001b[0;31mSubclasses:\u001b[0m     \n"
     ]
    }
   ],
   "source": [
    "# print class definition of the Constellation class\n",
    "sn.mapping.Mapper??"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41c23abb-5163-4e08-a2c8-a4b912046b30",
   "metadata": {},
   "source": [
    "As can be seen, the `Mapper` class inherits from `Layer`, i.e., implements a Keras layer.\n",
    "\n",
    "This allows to simply built complex systems by using the [Keras functional API](https://keras.io/guides/functional_api/) to stack layers."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5d88fc6-f18d-49e7-b66c-98c6361f4a31",
   "metadata": {},
   "source": [
    "Sionna provides as utility a binary source to sample uniform i.i.d. bits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8c64222f-1a4b-42da-a37a-3379034de9cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "binary_source = sn.utils.BinarySource()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bdaf6cb-5de8-4306-8701-b0e75bbc3fcf",
   "metadata": {},
   "source": [
    "Finally, we need the AWGN channel."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2665b1a1-945a-4b8c-bc35-49af1b0c26ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "awgn_channel = sn.channel.AWGN()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "261ad816-d4cf-424a-9dff-61444637e4dd",
   "metadata": {},
   "source": [
    "Sionna provides a utility function to compute the noise power spectral density ratio $N_0$ from the energy per bit to noise power spectral density ratio $E_b/N_0$ in dB and a variety of parameters such as the coderate and the nunber of bits per symbol."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4feb569f-dcc3-465f-94c5-e5d25fad97b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "no = sn.utils.ebnodb2no(ebno_db=10.0,\n",
    "                        num_bits_per_symbol=NUM_BITS_PER_SYMBOL,\n",
    "                        coderate=1.0) # Coderate set to 1 as we do uncoded transmission here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "efcbe1b0-39d9-4d0f-98bc-84343a46099c",
   "metadata": {},
   "source": [
    "We now have all the components we need to transmit QAM symbols over an AWGN channel.\n",
    "\n",
    "Sionna natively supports multi-dimensional tensors.\n",
    "\n",
    "Most layers operate at the last dimension and can have arbitrary input shapes (preserved at output)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "5747f93e-c89e-41b5-bc77-b6eb13aba4d1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of bits:  (64, 1024)\n",
      "Shape of x:  (64, 512)\n",
      "Shape of y:  (64, 512)\n",
      "Shape of llr:  (64, 1024)\n"
     ]
    }
   ],
   "source": [
    "BATCH_SIZE = 64 # How many examples are processed by Sionna in parallel\n",
    "\n",
    "bits = binary_source([BATCH_SIZE,\n",
    "                      1024]) # Blocklength\n",
    "print(\"Shape of bits: \", bits.shape)\n",
    "\n",
    "x = mapper(bits)\n",
    "print(\"Shape of x: \", x.shape)\n",
    "\n",
    "y = awgn_channel([x, no])\n",
    "print(\"Shape of y: \", y.shape)\n",
    "\n",
    "llr = demapper([y, no])\n",
    "print(\"Shape of llr: \", llr.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aca7a98b",
   "metadata": {},
   "source": [
    "In *Eager* mode, we can directly access the values of each tensor. This simplify debugging."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "abffb7f3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "First 8 transmitted bits: [0. 1. 1. 1. 1. 0. 1. 0.]\n",
      "First 4 transmitted symbols: [ 0.71-0.71j -0.71-0.71j -0.71+0.71j -0.71+0.71j]\n",
      "First 4 received symbols: [ 0.65-0.61j -0.62-0.69j -0.6 +0.72j -0.86+0.63j]\n",
      "First 8 demapped llrs: [-36.65  34.36  35.04  38.83  33.96 -40.5   48.47 -35.65]\n"
     ]
    }
   ],
   "source": [
    "num_samples = 8 # how many samples shall be printed\n",
    "num_symbols = int(num_samples/NUM_BITS_PER_SYMBOL)\n",
    "\n",
    "print(f\"First {num_samples} transmitted bits: {bits[0,:num_samples]}\")\n",
    "print(f\"First {num_symbols} transmitted symbols: {np.round(x[0,:num_symbols], 2)}\")\n",
    "print(f\"First {num_symbols} received symbols: {np.round(y[0,:num_symbols], 2)}\")\n",
    "print(f\"First {num_samples} demapped llrs: {np.round(llr[0,:num_samples], 2)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d92339a-44ae-45d9-a59c-bde99e3bd656",
   "metadata": {},
   "source": [
    "Let's visualize the received noisy samples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "682509f0-e734-4aec-bc05-773bd576de16",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,8))\n",
    "plt.axes().set_aspect(1)\n",
    "plt.grid(True)\n",
    "plt.title('Channel output')\n",
    "plt.xlabel('Real Part')\n",
    "plt.ylabel('Imaginary Part')\n",
    "plt.scatter(tf.math.real(y), tf.math.imag(y))\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d415e3e0-65ee-488f-a29b-5909dce48ede",
   "metadata": {},
   "source": [
    "**Task:** One can play with the SNR to visualize the impact on the received samples.\n",
    "\n",
    "**Advanced Task:** Compare the LLR distribution for \"app\" demapping with \"maxlog\" demapping.\n",
    "The [Bit-Interleaved Coded Modulation](https://nvlabs.github.io/sionna/examples/Bit_Interleaved_Coded_Modulation.html) example notebook can be helpful for this task.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfab1249-a6c1-430a-a204-a9c631a82700",
   "metadata": {},
   "source": [
    "## Communication Systems as Keras Models"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bfc184ba-c090-4443-9cd6-c217b3f64052",
   "metadata": {},
   "source": [
    "It is typically more convenient to wrap a Sionna-based communication system into a [Keras models](https://keras.io/api/models/model/).\n",
    "\n",
    "These models can be simply built by using the [Keras functional API](https://keras.io/guides/functional_api/) to stack layers.\n",
    "\n",
    "The following cell implements the previous system as a Keras model.\n",
    "\n",
    "The key functions that need to be defined are `__init__()`, which instantiates the required components, and `__call()__`, which performs forward pass through the end-to-end system."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ee71bdbf-14c7-464f-b90d-450cb7b070b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "class UncodedSystemAWGN(Model): # Inherits from Keras Model\n",
    "    def __init__(self, num_bits_per_symbol, block_length):\n",
    "        \"\"\"\n",
    "        A keras model of an uncoded transmission over the AWGN channel.\n",
    "\n",
    "        Parameters\n",
    "        ----------\n",
    "        num_bits_per_symbol: int\n",
    "            The number of bits per constellation symbol, e.g., 4 for QAM16.\n",
    "\n",
    "        block_length: int\n",
    "            The number of bits per transmitted message block (will be the codeword length later).\n",
    "\n",
    "        Input\n",
    "        -----\n",
    "        batch_size: int\n",
    "            The batch_size of the Monte-Carlo simulation.\n",
    "\n",
    "        ebno_db: float\n",
    "            The `Eb/No` value (=rate-adjusted SNR) in dB.\n",
    "\n",
    "        Output\n",
    "        ------\n",
    "        (bits, llr):\n",
    "            Tuple:\n",
    "\n",
    "        bits: tf.float32\n",
    "            A tensor of shape `[batch_size, block_length] of 0s and 1s\n",
    "            containing the transmitted information bits.\n",
    "\n",
    "        llr: tf.float32\n",
    "            A tensor of shape `[batch_size, block_length] containing the\n",
    "            received log-likelihood-ratio (LLR) values.\n",
    "        \"\"\"\n",
    "\n",
    "        super().__init__() # Must call the Keras model initializer\n",
    "\n",
    "        self.num_bits_per_symbol = num_bits_per_symbol\n",
    "        self.block_length = block_length\n",
    "        self.constellation = sn.mapping.Constellation(\"qam\", self.num_bits_per_symbol)\n",
    "        self.mapper = sn.mapping.Mapper(constellation=self.constellation)\n",
    "        self.demapper = sn.mapping.Demapper(\"app\", constellation=self.constellation)\n",
    "        self.binary_source = sn.utils.BinarySource()\n",
    "        self.awgn_channel = sn.channel.AWGN()\n",
    "\n",
    "    # @tf.function # Enable graph execution to speed things up\n",
    "    def __call__(self, batch_size, ebno_db):\n",
    "\n",
    "        # no channel coding used; we set coderate=1.0\n",
    "        no = sn.utils.ebnodb2no(ebno_db,\n",
    "                                num_bits_per_symbol=self.num_bits_per_symbol,\n",
    "                                coderate=1.0)\n",
    "\n",
    "        bits = self.binary_source([batch_size, self.block_length]) # Blocklength set to 1024 bits\n",
    "        x = self.mapper(bits)\n",
    "        y = self.awgn_channel([x, no])\n",
    "        llr = self.demapper([y,no])\n",
    "        return bits, llr"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2397775-1a9d-467c-8048-0949a25bda02",
   "metadata": {},
   "source": [
    "We need first to instantiate the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "3212d737-3ae0-4158-983b-6d0667cc1dd8",
   "metadata": {},
   "outputs": [],
   "source": [
    "model_uncoded_awgn = UncodedSystemAWGN(num_bits_per_symbol=NUM_BITS_PER_SYMBOL, block_length=1024)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e449509-4ba0-4e14-8a6a-91e55a7efec4",
   "metadata": {},
   "source": [
    "Sionna provides a utility to easily compute and plot the bit error rate (BER)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "34534be8-947d-4d29-8e58-1c7d0dce96c9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EbNo [dB] |        BER |       BLER |  bit errors |    num bits | block errors |  num blocks | runtime [s] |    status\n",
      "---------------------------------------------------------------------------------------------------------------------------------------\n",
      "     -3.0 | 1.5825e-01 | 1.0000e+00 |      324099 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "   -2.579 | 1.4687e-01 | 1.0000e+00 |      300799 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "   -2.158 | 1.3528e-01 | 1.0000e+00 |      277061 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "   -1.737 | 1.2323e-01 | 1.0000e+00 |      252373 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "   -1.316 | 1.1246e-01 | 1.0000e+00 |      230320 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "   -0.895 | 1.0107e-01 | 1.0000e+00 |      206992 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "   -0.474 | 9.0021e-02 | 1.0000e+00 |      184362 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "   -0.053 | 8.0165e-02 | 1.0000e+00 |      164177 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    0.368 | 6.9933e-02 | 1.0000e+00 |      143222 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    0.789 | 6.0897e-02 | 1.0000e+00 |      124717 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    1.211 | 5.2020e-02 | 1.0000e+00 |      106537 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    1.632 | 4.3859e-02 | 1.0000e+00 |       89823 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    2.053 | 3.6686e-02 | 1.0000e+00 |       75132 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    2.474 | 3.0071e-02 | 1.0000e+00 |       61586 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    2.895 | 2.4304e-02 | 1.0000e+00 |       49775 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    3.316 | 1.9330e-02 | 1.0000e+00 |       39588 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    3.737 | 1.4924e-02 | 1.0000e+00 |       30565 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    4.158 | 1.1227e-02 | 1.0000e+00 |       22992 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "    4.579 | 8.2632e-03 | 1.0000e+00 |       16923 |     2048000 |         2000 |        2000 |         0.0 |reached target block errors\n",
      "      5.0 | 5.9722e-03 | 9.9850e-01 |       12231 |     2048000 |         1997 |        2000 |         0.0 |reached target block errors\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1152x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "EBN0_DB_MIN = -3.0 # Minimum value of Eb/N0 [dB] for simulations\n",
    "EBN0_DB_MAX = 5.0 # Maximum value of Eb/N0 [dB] for simulations\n",
    "BATCH_SIZE = 2000 # How many examples are processed by Sionna in parallel\n",
    "\n",
    "ber_plots = sn.utils.PlotBER(\"AWGN\")\n",
    "ber_plots.simulate(model_uncoded_awgn,\n",
    "                  ebno_dbs=np.linspace(EBN0_DB_MIN, EBN0_DB_MAX, 20),\n",
    "                  batch_size=BATCH_SIZE,\n",
    "                  num_target_block_errors=100, # simulate until 100 block errors occured\n",
    "                  legend=\"Uncoded\",\n",
    "                  soft_estimates=True,\n",
    "                  max_mc_iter=100, # run 100 Monte-Carlo simulations (each with batch_size samples)\n",
    "                  show_fig=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70e75524",
   "metadata": {},
   "source": [
    "The `sn.utils.PlotBER` object stores the results and allows to add additional simulations to the previous curves.\n",
    "\n",
    "*Remark*: In Sionna, a block error is defined to happen if for two tensors at least one position in the last dimension differs (i.e., at least one bit wrongly received per codeword).\n",
    "The bit error rate the total number of erroneous positions divided by the total number of transmitted bits."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7c8f75c-65c8-4440-a8bd-fca7e3d8f35f",
   "metadata": {},
   "source": [
    "## Forward Error Correction (FEC)\n",
    "\n",
    "We now add channel coding to our transceiver to make it more robust against transmission errors. For this, we will use [5G compliant low-density parity-check (LDPC) codes and Polar codes](https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationId=3214).\n",
    "You can find more detailed information in the notebooks [Bit-Interleaved Coded Modulation (BICM)](https://nvlabs.github.io/sionna/examples/Bit_Interleaved_Coded_Modulation.html) and [5G Channel Coding and Rate-Matching: Polar vs. LDPC Codes](https://nvlabs.github.io/sionna/examples/5G_Channel_Coding_Polar_vs_LDPC_Codes.html)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "b6800065",
   "metadata": {},
   "outputs": [],
   "source": [
    "k = 12\n",
    "n = 20\n",
    "\n",
    "encoder = sn.fec.ldpc.LDPC5GEncoder(k, n)\n",
    "decoder = sn.fec.ldpc.LDPC5GDecoder(encoder, hard_out=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39673cbe",
   "metadata": {},
   "source": [
    "Let us encode some random input bits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6e940732",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Input bits are: \n",
      " [[1. 1. 0. 0. 1. 1. 0. 0. 0. 0. 1. 1.]]\n",
      "Encoded bits are: \n",
      " [[1. 1. 0. 0. 0. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 0. 0. 1. 1.]]\n"
     ]
    }
   ],
   "source": [
    "BATCH_SIZE = 1 # one codeword in parallel\n",
    "u = binary_source([BATCH_SIZE, k])\n",
    "print(\"Input bits are: \\n\", u.numpy())\n",
    "\n",
    "c = encoder(u)\n",
    "print(\"Encoded bits are: \\n\", c.numpy())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99a5303a",
   "metadata": {},
   "source": [
    "One of the fundamental paradigms of Sionna is batch-processing.\n",
    "Thus, the example above could be executed with for arbitrary batch-sizes to simulate `batch_size` codewords in parallel.\n",
    "\n",
    "However, Sionna can do more - it supports *N*-dimensional input tensors and, thereby, allows the processing of multiple samples of multiple users and several antennas in a single command line.\n",
    "Let's say we want to encoded `batch_size` codewords of length `n` for each of the `num_users` connected to each of the `num_basestations`. \n",
    "This means in total we transmit `batch_size` * `n` * `num_users` * `num_basestations` bits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e89a7fc4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of u:  (10, 4, 5, 500)\n",
      "Shape of c:  (10, 4, 5, 1000)\n",
      "Total number of processed bits:  200000\n"
     ]
    }
   ],
   "source": [
    "BATCH_SIZE = 10 # samples per scenario\n",
    "num_basestations = 4 \n",
    "num_users = 5 # users per basestation\n",
    "n = 1000 # codeword length per transmitted codeword\n",
    "coderate = 0.5 # coderate\n",
    "\n",
    "k = int(coderate * n) # number of info bits per codeword\n",
    "\n",
    "# instantiate a new encoder for codewords of length n\n",
    "encoder = sn.fec.ldpc.LDPC5GEncoder(k, n)\n",
    "\n",
    "# the decoder must be linked to the encoder (to know the exact code parameters used for encoding)\n",
    "decoder = sn.fec.ldpc.LDPC5GDecoder(encoder,\n",
    "                                    hard_out=True, # binary output or provide soft-estimates\n",
    "                                    return_infobits=True, # or also return (decoded) parity bits\n",
    "                                    num_iter=20, # number of decoding iterations\n",
    "                                    cn_type=\"boxplus-phi\") # also try \"minsum\" decoding\n",
    "\n",
    "# draw random bits to encode\n",
    "u = binary_source([BATCH_SIZE, num_basestations, num_users, k])\n",
    "print(\"Shape of u: \", u.shape)\n",
    "\n",
    "# We can immediately encode u for all users, basetation and samples \n",
    "# This all happens with a single line of code\n",
    "c = encoder(u)\n",
    "print(\"Shape of c: \", c.shape)\n",
    "\n",
    "print(\"Total number of processed bits: \", np.prod(c.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f19f80a",
   "metadata": {},
   "source": [
    "This works for arbitrary dimensions and allows a simple extension of the designed system to multi-user or multi-antenna scenarios.\n",
    "\n",
    "Let us now replace the LDPC code by a Polar code. The API remains similar."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b05b3fd2",
   "metadata": {},
   "outputs": [],
   "source": [
    "k = 64\n",
    "n = 128\n",
    "\n",
    "encoder = sn.fec.polar.Polar5GEncoder(k, n)\n",
    "decoder = sn.fec.polar.Polar5GDecoder(encoder,\n",
    "                                      dec_type=\"SCL\") # you can also use \"SCL\" "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f00d4b2",
   "metadata": {},
   "source": [
    "*Advanced Remark:* The 5G Polar encoder/decoder class directly applies rate-matching and the additional CRC concatenation. \n",
    "This is all done internally and transparent to the user.\n",
    "\n",
    "In case you want to access low-level features of the Polar codes, please use `sionna.fec.polar.PolarEncoder` and the desired decoder (`sionna.fec.polar.PolarSCDecoder`, `sionna.fec.polar.PolarSCLDecoder` or `sionna.fec.polar.PolarBPDecoder`).\n",
    "\n",
    "Further details can be found in the tutorial notebook on [5G Channel Coding and Rate-Matching: Polar vs. LDPC Codes](https://nvlabs.github.io/sionna/examples/5G_Channel_Coding_Polar_vs_LDPC_Codes.html).\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "708efce7-2145-4637-981a-5a38a315a9b6",
   "metadata": {},
   "source": [
    "![QAM FEC AWGN](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAmkAAACtCAYAAADrl+hZAAAABGdBTUEAALGPC/xhBQAAACBjSFJNAAB6JgAAgIQAAPoAAACA6AAAdTAAAOpgAAA6mAAAF3CculE8AAAAaGVYSWZNTQAqAAAACAAEAQYAAwAAAAEAAgAAARIAAwAAAAEAAQAAASgAAwAAAAEAAgAAh2kABAAAAAEAAAA+AAAAAAADoAEAAwAAAAEAAQAAoAIABAAAAAEAAAJpoAMABAAAAAEAAACtAAAAAF4rGUEAAALkaVRYdFhNTDpjb20uYWRvYmUueG1wAAAAAAA8eDp4bXBtZXRhIHhtbG5zOng9ImFkb2JlOm5zOm1ldGEvIiB4OnhtcHRrPSJYTVAgQ29yZSA2LjAuMCI+CiAgIDxyZGY6UkRGIHhtbG5zOnJkZj0iaHR0cDovL3d3dy53My5vcmcvMTk5OS8wMi8yMi1yZGYtc3ludGF4LW5zIyI+CiAgICAgIDxyZGY6RGVzY3JpcHRpb24gcmRmOmFib3V0PSIiCiAgICAgICAgICAgIHhtbG5zOnRpZmY9Imh0dHA6Ly9ucy5hZG9iZS5jb20vdGlmZi8xLjAvIgogICAgICAgICAgICB4bWxuczpleGlmPSJodHRwOi8vbnMuYWRvYmUuY29tL2V4aWYvMS4wLyI+CiAgICAgICAgIDx0aWZmOkNvbXByZXNzaW9uPjE8L3RpZmY6Q29tcHJlc3Npb24+CiAgICAgICAgIDx0aWZmOlJlc29sdXRpb25Vbml0PjI8L3RpZmY6UmVzb2x1dGlvblVuaXQ+CiAgICAgICAgIDx0aWZmOk9yaWVudGF0aW9uPjE8L3RpZmY6T3JpZW50YXRpb24+CiAgICAgICAgIDx0aWZmOlBob3RvbWV0cmljSW50ZXJwcmV0YXRpb24+MjwvdGlmZjpQaG90b21ldHJpY0ludGVycHJldGF0aW9uPgogICAgICAgICA8ZXhpZjpQaXhlbFhEaW1lbnNpb24+NjE3PC9leGlmOlBpeGVsWERpbWVuc2lvbj4KICAgICAgICAgPGV4aWY6Q29sb3JTcGFjZT4xPC9leGlmOkNvbG9yU3BhY2U+CiAgICAgICAgIDxleGlmOlBpeGVsWURpbWVuc2lvbj4xNzM8L2V4aWY6UGl4ZWxZRGltZW5zaW9uPgogICAgICA8L3JkZjpEZXNjcmlwdGlvbj4KICAgPC9yZGY6UkRGPgo8L3g6eG1wbWV0YT4KfYoiHAAAQABJREFUeAHtnQl8VNXZ/5+QlewBAmFNwr6DIogKgogbdVcU19rWtvbtv6/Wau3H9q22tdr66Vu1tta3thbbWrVSVNyoCyCLgMgq+75DWLIRkpCN//kevGEYJmQmmczcyTyHzzCTO3c593vmnvM7z/Occ2KOmySalIASUAJKQAkoASWgBFxFoI2rcqOZUQJKQAkoASWgBJSAErAEVKTpD0EJKAEloASUgBJQAi4koCLNhYWiWVICSkAJKAEloASUgIo0/Q0oASWgBJSAElACSsCFBFSkubBQNEtKQAkoASWgBJSAElCRpr8BJaAElIASUAJKQAm4kICKNBcWimZJCSgBJaAElIASUAIq0vQ3oASUgBJQAkpACSgBFxJQkebCQtEsKQEloASUgBJQAkpARZr+BpSAElACSkAJKAEl4EICKtJcWCiaJSWgBJSAElACSkAJxCkCJaAElIAbCSxatEicF/njs6aGCXz961+X9PR0Oeuss2T06NGSmJjY8M76jRJQAhFBIEYXWI+IctJMKoGoIjBlyhSprKyU4cOHW9HBzfNZU8MEpk2bJkeOHJGXX35Zbr/9dvn+979vRVvDR+g3SkAJuJ2AijS3l5DmTwlEEQGsZQi0W265Re677z5rDUpKSrIEnPcowhHQrZaWlkpdXZ3w/vrrr8tf//pXeeKJJ+TSSy9Vq1pAJHVnJeAeAirS3FMWmhMlEPUEcnJy5JlnnpHLLrtMMjMzo55HUwEg1GbPni0PP/ywzJw5U7p3797UU+lxSkAJhJGAirQwwtdLKwElcJLA008/LTt27LDWH7WaneTS1E/Hjh2TX/3qV7Jp0yZ56qmnJDs7u6mn0uOUgBIIEwEd3Rkm8HpZJaAEThJAoGHxeeSRR0QF2kkuzfnEwAHi0qqqqmTp0qWCaNOkBJRAZBFQkRZZ5aW5VQKtkkBxcbEdIKAuzuAWL6M9R4wYIcuXL7cDMYJ7dj2bElACLU1ARVpLE9bzKwEl0CgBBgyMGzeu0f10h8AJMB2HirTAuekRSsANBFSkuaEUNA9KIMoJLF682M7tFeUYWuT2mTdt7dq11u3ZIhfQkyoBJdBiBFSktRhaPbESUAL+EsDdqa5Of2kFth8uz7KyMjs9R2BH6t5KQAmEm4CKtHCXgF5fCSgBJaAElIASUAI+CKhI8wFFNykBJaAElIASUAJKINwEVKSFuwT0+kpACSgBJaAElIAS8EFARZoPKLpJCSgBJaAElIASUALhJqAiLdwloNdXAkrAdQSY+LWkpMRnvmpra3ViWJ9kdKMSUALBJqAiLdhE9XxKQAlELAFm5//tb38rV155pX3ddtttdrZ+Fi530oIFC+zi5ezrpD/+8Y8yf/58QcCRXnzxRZkzZ079385++q4ElIASCISAirRAaOm+SkAJtGoCr7zyikybNk0QZ3fffbeduuKBBx6Qw4cP19/3yy+/LC+88IJUVFTUb/v444/lzTfflOrqarvtX//6l8ybN09qamrq99EPSkAJKIFACahIC5SY7q8ElECrJfD888/L2WefLddcc41MnjxZHn74YVm1apXMnTvXCq49e/ZY8VVQUCDLli2zlrLjx4/L6tWrrUirrKwU9lm5cqV8/vnnaklrtb8UvTElEBoCKtJCw1mvogSUgMsJEIO2ceNGGTt2rKSlpUlycrIVbF27drUz9uPKfOutt6Rv374yYcIEK8pwee7bt08KCwutFW3FihUya9YsOzEvQk0taS4vdM2eEnA5ARVpLi8gzZ4SUAKhIYAIQ1S1bdu2/oLx8fESFxdnBRiuTFydV111lVx//fUyc+ZMKSoqknXr1kmnTp3kiiuukA8//FDefvttufXWW+XIkSNWwGFp06QElIASaAoBFWlNoabHKAEl0OoIsHwSS1PhunQGABw6dMi6Lzt37iwLFy60FrUNGzZYd+bu3bvt4ADcnv369bPu0Xfffdfud91110mvXr3qLXCtDpbekBJQAiEhoCItJJj1IkpACbidABYzrGQzZsyQzZs3S2lpqfz5z3+WnJwcmThxojAYgMXKu3fvbrfhFmWQASKtf//+1jWamJgovXv3ltzcXBkyZIiNZ3MEn9vvX/OnBJSA+wjEuS9LmiMloASUQHgI3HvvvfKTn/xEvvnNb0qbNm1k27Zt8vOf/1xwe37wwQfy2GOPWcGGoCNWjf1SU1PtdB28X3LJJZKdnS2ItUGDBsmnn35ab5ULzx3pVZWAEohkAjEmXkIDJiK5BDXvSqAVEIiJiRE3VEXkYdOmTbJr1y7B1cmcaRkZGXLuuefaeLX777/fijCQl5WVCQMFYmNjpU+fPtK+fXvZvn27JCUlWUsbAwqYugOxhuALZ8rLy5NPPvnEWvjCmQ+9thJQAoERUJEWGC/dWwm0OIGq2qPy9xU3tPh13HSBn9/9H9mx3D39RSavZeTm2rVrrTVsy5YtMnjwYLnjjjskISGhQXSO0ER0uildcmueTPp6nqRnJbkpWy2Wl85pw+T8Ht+VzKQeLXYNPbESCAUBdXeGgrJeQwkEQKDueI1sOvxhAEdE/q6pWe66ByxfWMSIQcvPz7ejOBn1idvzTMlt4szJa0ZHkZ2l8yTx5MIJzlet8r22rkbo7GhSApFOQEVapJeg5r9VEjguUdKaOqXnLsOTkytBdGVlZdlX/cZI/GD48ptyj62yZSEet3caLXfbsiz17OEloCItvPz16krgjAQSYlPk9mHTzrhPa/jyDyuuaNZtED+WkpJyyhxnzTqhHwcfOHBAWGGA0Z4HDx6s/xyINY2Rn8S0tXTa+KnI//vaS9K5izGptdK078hKWbjrOSmu3NlK71BvKxoJqEiLxlLXe44YAm1i4qRv+0sjJr9NzejRwqYdSTD8M888I/v377fuye985zt2vrKmne3EUcSiLV26VEaNGnVGAcWktTt27LBLR3l+xk3qK7GiAaNFmZoDYcbfd911lzz55JN22o5AxJ2v859pW2mBSF76WMltn3um3SL6u9g28ZKwNyWi70EzrwS8CahI8yaifysBlxGIiQnvyECX4ajPDks4fe9735PLL79cvv3tb9tJZ5lAtrmJOdJ++tOfyhtvvGGn12jofKzfiUhjkIHn54b2Z6Jb1vP85S9/aS1+LDtF3pmyoyUFmpMffket+7cUI/zTpARaEwEVaa2pNPVelEAUEXjxxRfl6NGjct9999kpLyoqKuwSToywfOGFFwRRxEoB1157rRVDf/nLX6z1CisZiaWbLrvsMuE8LOfEPGeIvqefftquGnDnnXfKjTfeaC1zCxYskOnTp0txcbGMGzdOpkyZ0iBppu/4zW9+I6xM0LNnT7n77rulS5cu8uyzzwqrFDjWN5aYYkmpG264wS4h9fvf/95elyk72MZC7//zP/9jl5yaN2+edOzYUb71rW9ZS1y4p/Ro8Ob1CyWgBIJKQLvoQcWpJ1MCSiBUBObOnSvjx4+3AgjRQkwak8gifBA8kyZNkpEjR1p35M6dO+3s/wi3m266yU5E+9RTT1n3I+8jRoywooiloVg8HcH2ta99zc6PtmTJEnn00Udl4MCBVtSx8gBu1oYWT+cczK128803C8Jx6tSp1hXLNViN4Bvf+Iadrwyxt3jxYhvL9s4778js2bPtmqBY2H784x9LeXm5cI9ffPGFFYXMufbqq6/aqUFCxVivowSUQHgJqEgLL3+9uhJQAk0kwALmLGzunVjWCcsV1i4sUkybgaWMBdJZrgnr2e23327X3zx27JgVUFjKcJWyigBLPCH2xowZYy1hLKTOJLXMmca6nkxSyzU4n6+UlpZmrV3En61fv14QYMSgMaEsrk2Wk+rQoYMVhs6SUSwvxYS5WP2++tWvypo1a+yLgQlY1MjzxRdfLPPnz29QHPrKi25TAkogsgmoSIvs8tPcK4GoJYCgwsrlnXAp4ubEmsWC6UyhwcACEtsI7EesIbIQcH/4wx+sleyJJ56QWbNm2Rgz4swQVsSKIco43+jRo+Wcc86RH/zgB8IC6g3NmcZSUM8//7w9Bpcqa4ByPqx9uGKd0Zyex+/du1e6detmrYEITPbDtco7+SfPbGeVA7ZpUgJKIDoIqEiLjnLWu1QCrY4AAmjRokXCyErEC++vvPKKnHfeeXa5JoL5t27dat2FF1xwwSn37wglZyOxXljAWJ8TtymxbggnhBxuTkRejx49hPNgETv//POlXbt2dvkorF24OIlF4zPWLlYluOKKK+w758DtiUDE7co+3q5S3LIITqyDuD3J34ABA05ZToptKtCcEtN3JRAdBFSkRUc5610qgVZH4KqrrrJB+cSLIYhYGJ2RmZdeeql1gxL0f9ttt9mpNBBBJO9RlFi5WJ+T0aEIMcQXwf7Ei7F4OnFpV155pT3/448/bs9HDBsi7qKLLrKCC6sZxyG++Ix7srCwUB588EG7DijWPNypuE+5PvliYAGuVifhfmXpqeuvv14eeughex0GCrC/Z549PzvH6rsSUAKtl4Cu3dl6y1bvLEIJVNaUyCOzMm3uk+Iy5GcTiiP0TvzPNuKjKVYirF2IGyaTZYHzPGPlwuJF/NiePXusJYq/+/btawUTLkdiz4gFW7hwoQwdOtRuZzJcxFS/fv3sOyMziSnD+sV8aZwLSxnB/GwbNmyYFU/EjsXFxdkF1hmwwGeWkeL6iLCcnBxh0ltcs7ycmDaEIPthPeP83DvHIBRZfor84mJdvny5nSyXGLaioiJrGWSpKm9LYGOk4dLaF1jfXDhLZqz7nhQcXSs9sybIdQOflY4pAxtDo98rAVcTUJHm6uLRzEUjARVpgZc6E9B6L3xOHBjJn+kqfM38770NIcXL+3xscyxcvj57H4erEzHnK3lf09c+TdmmIq0p1PQYJRB+Ar5rivDnS3OgBJSAEvCbgLdA40BvMXWmk/myTHlvQ4g5YszzXJ7bfH32Pq4hgcY5va/peR39rASUQPQR0Ji06CtzvWMloASUgBJQAkogAgioSIuAQtIsKgEloASUgBJQAtFHQEVa9JW53rESUAJKQAkoASUQAQRUpEVAIWkWlYASUAJKQAkogegjoCIt+spc71gJuI7A8OHD7QS0rstYK8gQU3uwZqivwRWt4Pb0FpRAqyagIq1VF6/enBKIDAIslD5nzpzIyGyE5ZJVGRDBLC2lSQkogcgioCItsspLc6sEWiUBZvhXkdYyRcuqCqmpqT6nD2mZK+pZlYASCBYBFWnBIqnnUQJKoMkEsKSVlJTI1KlTm3wOPfB0AtOmTbPrmrIUFUJNkxJQApFFQEVaZJWX5lYJtEoCLMl01113yaNmHc4VK1a0ynsM9U0Ri/bkk0/KxIkTpX///g2uchDqfOn1lIAS8J+AijT/WemeSkAJtCCBa665xi5ajlhTi1rzQGNBu+eee+yAgT59+qhAax5OPVoJhI2ALgsVNvR6YSWgBDwJYE175JFHZMeOHTJ79ux6oUbQOy9NZyaA5Yw1TBFo69evl0mTJgnClwXhNSkBJRCZBFSkRWa5aa6VQKskwELgvHJzc+tFGgMKtm/f3irvN5g31a1bN2sxGz16tBVnPXv2lKysrIDWMA1mfvRcSkAJNJ+AirTmM9QzKAElEGQCCDXcniQGFBQXF9vP+l/DBNLT060gQ6ypOGuYk36jBCKJgIq0SCotzasSiCICCDVNSkAJKIFoJqADB6K59PXelYASUAJKQAkoAdcSUJHm2qLRjCkBJaAElIASUALRTEBFWjSXvt67ElACSkAJKAEl4FoCKtJcWzSaMSWgBJSAElACSiCaCahIi+bS13tXAkpACSgBJaAEXEtARZpri0YzpgSUgBJQAkpACUQzARVp0Vz6eu9KQAkoASWgBJSAawmoSHNt0WjGlIASUAJKQAkogWgmoCItmktf710JKAEloASUgBJwLQEVaa4tGs2YElACSkAJKAElEM0EVKRFc+nrvSsBJaAElIASUAKuJaAizbVFoxlTAkpACSgBJaAEopmAirRoLn29dyWgBJSAElACSsC1BFSkubZoNGNKQAkoASWgBJRANBNQkRbNpa/3rgSUgBJQAkpACbiWgIo01xaNZkwJKAEloASUgBKIZgIq0qK59PXelYASUAJKQAkoAdcSUJHm2qLRjCkBJaAElIASUALRTEBFWjSXvt67ElACSkAJKAEl4FoCKtJcWzSaMSWgBJSAElACSiCaCahIi+bS13tXAkpACSgBJaAEXEtARZpri0YzpgSUgBJQAkpACUQzARVp0Vz6eu9KQAkoASWgBJSAawmoSHNt0WjGlIASUAJKQAkogWgmoCItmktf710JKAEloASUgBJwLQEVaa4tGs2YElACSkAJKAElEM0E4qL55vXelYBbCNQdr5HNhz+WY7VHpKq2vD5bdcer5YuCaebvGGkblyW920+o/04/KIFoJ1BVWyY7iz+TippCKShbLZU1pRbJ0eoDsvHQB2bbWslsmyc5KQMlPjY52nHp/UcgARVpEVhomuXWR6DueK1sL/lUVhdMl+PH6+pvsLquUj7c8jOJjUmQ4Z1vUZFWT0Y/KAEIxMj6Q+/L1qLZUl5dKEerD1osRRXb5dNdf5CENikyNu8H0jGln+JSAhFJQN2dEVlsmunWRiA2Jk7yMi+Q4oqdcuDo2vrbQ7BhISg7tl/ys8bWb9cPSkAJiCTEpkinlP5SXnVIiiq2SU3dMYsFC9vh8s1SayzU7ZPzJb6NWtH09xKZBFSkRWa5aa5bGYGYmFjJzTzPWMomnnZnsW3ipU+Hy6Rb+jmnfacblEC0E+iXPUk6pw2TuDaJp6CIkVgZmnOjdEjuK23M86VJCUQiARVpkVhqmudWSSDB9PZHd/+2JMamn3J/KXHZdrs2NKdg0T+UgCWQnthZBmZfLWkJOacQyTYWtj6m05Mc3/6U7fqHEogkAirSIqm0NK+tmoAva5pa0Vp1kevNBYmAtzVNrWhBAqunCTsBFWlhLwLNgBI4ScDbmqZWtJNs9JMSaIiAtzUt2wwUIHRArWgNEdPtkUIgZKM7d+/eLTU1NcL79u3b7WcgOdu7desmcXFx4rzzXU5OjuTl5UlSUhJ/alICrZ6ApzVt/aG3NRat1Ze43mCwCGBNW3twhpRVHTSxaJMlW2PRgoVWzxNGAi0q0hBg8+fPl8rKynox5txrTEyM81H4vGfPHvv3jh07zBQEx+3n/fv3S2ZmpowePdq+Dx8+3L7XH6gflEArJIA17bwe98iBI6s1Fq0Vlq/eUssQcKxptXXVakVrGcR61jAQiDGC6IQiCtLFHWG2efPmemsZIsyxkPGem5trrWZc0tnuWNScd75bsWKFbNiwwQqzoqKiesE2fvz4ViHWiouLZdu2bcayuFUKzefSkhIpKT4xGSP33xpTRmaGDBkyWFKSUyQ/v6e1lrr1PrH8YvXl5fwueQ9Fqq6rkDWHXpeh2beFdGQa1uwxY8ZImzZtJM9YsXv06BGK223SNZyy4Z3kvNs/Wul/lI3jcaB8+Oy2xDNCWfByvCe8hyKVHtsnhcc2mGk5hkrb2HahuKS9BmXBC6OC8x6yi/t5IYwllMm2bVtl3/69UlF5TA4WnJhXzs9TRNxueOFGjDxbkhKSbHuD3oi0FFSRNm3aNPnoo4+s5ax79+72x0qlkpiYWC/GAgWENY0fFi+sbPzQ0JVY1SJZrC1fvlxmzf5ICg7vN5MwlklNbZUwoSmv1pxizVD4hLgUMdMXSWpCuky+6SYZcfYI191yWVmZvPrqq/UW3nBkMDZepLY6HFc2162tFZ69O++804q28OSi4atOnTrV1gns4WmVb/iI1vMN9R9C6Nprr5WJEye6KhyENmD9+vWndNBDTT7W6Fbz8xUJqvmh8bugXOh4I9IoG97dklasWC6fLvxUdu7eLuU1R82qJrSjtVJbFxrxHC4OMWay46T4NIk5HiN1FW3kllunyAXnjwlXdpp03aCItEWLFsmcOXME61nnzp1tpT5u3LgWqTywrnGtEmN1ikSxxkNM/md+9J4cT6ySnv17SOdu2ZKanvLlK7VJBRkpB5WVlsmmtdvkSOlR2bZup6QkZkrf3AG2UiMGMdyJHj+NzGuvvSaHDh2yv2Usv47FNxJ7YoEwRZwRogAH3nnGEhIS5Pbbb3eFVY3OGgKNDhudNMrGaQyd90DuN9L2dcqGOvfYsWO2fG688Ubp379/WK1qiEY6NTw7I0eOlN69e9d7TJxnJ9JYB5JffpfOywnZcYMhwbO9Kasqlv7DekmnrtnSPjtTEhITpF12ViC3GXH7Vh2rkjXLN0jVsWpZu3yjaW8ypFuHPJkyZYqt0yPhhpol0ngwqTQQHVi4Jk+eLBdffHGLiDNvmN5i7fLLL7fWNbcPMnjzzTfl5X/9Tbr2yZYLLz1Xcrp18r61qPkbwTb3g8XyybuL5CuXXi0/+tGPwn7vCLMHH3xQUlNT5bbbbrPxkGHPVJgygGDDMv7rX//aioDnnnsuTDk5edm77rrLWs6++tWvWpF28pvo+4RQ+9WvfiVpaWny1FNPSYcOHcIG4YEHHrCdmlC2AWG72UYu7HQkVq5cKfxOsaqFK9He/PP1v0uX3u3lrNFDpc/A/HBlJezXRbAt/mSZTP/b+zL23PHym9/8Jux58icDsY+a5M+O3vtQQVBpr1u3Ts455xzb0x41apTtdXvv2xJ/Y3VxBhTMnDnTikUn9s2tQm2riQV4693p0rZDrHzlxomtvhfTWLnTk8vp1lFiYmNkyaKlkpXRTvr07tPYYS32PZaJ2bNny5IlS+Sxxx6TgQMHtti1IuHETlxaly5d5L333pPY2FgTTzgkbFmnM0ij8/TTT9sOWdgy4pILY6HCejhr1iwpLy+XQYMGhaz+9URAWzBjxgxBqBHeguU1mpMTl0a4AJZFvEvh8BJsM7HOb74zXZJNe3OFaW86m7o2mlNsXKxlkJaZKp/Mmi+pyWkyaOAg1yNpkkjjofzHP/4hR48elRtuuEGuv/56+yOkUg91cqbpwKqHWIuPj7cVl9uEGr2rf78xTcpqi+SCS84x5ubWbWb293eQ+KVQq66plsVzl8q5o861Vix/jw/Wfgg0rEb/+c9/5J577ol6geZwdYQaz9kLL7wgY8eOlYyMDOfrkL0j0F566SW57777VKB5UHeE2t/+9jcr0jp27GgHfXjs0qIfnbbg5ptvtp3maBdoDmxHqCHSCgoKrCU6lG2SbW+mT5OjdUVy/kRtb5xycYRacmpbeWf6f2TsmAslPf3UFV6cfd3yHrCqch5KetQPPfSQ4GYM5Y/PFzh8/1TemJUZLUmFji/eTYl8bdyyTjp0zjBqPnpdnL7KJC09VQYO7yN1sdWyfPkyX7u0+Dbc9fy2+S1jodV0kgAjCC+88ELp2rWrtVif/CZ0n3imiY8jDk3TqQT4vRIDhgWY33EoE88M1zz33HPD3g6E8r79uRZWTtomZiZANIUybd9u2pvN66W9bW+i24LmzR0PztBzBkjb9HhZaAZTuD0FJNJ4IKks+eERrxPuYFVPuPT0sYAg1MgnFhE3CbWi4kJJSI6TTmaQgKbTCaSYgRPtO6fLsjCJNALlDx48qALt9KKxW3B1IgaIQQ1HcgYKhOPakXBNyoY4XSzCoUx4MBgoEO6OeijvOZBrIdRI/H5DmQqNMExIjZUcM0hA0+kEEGr5/bvJgoULTv/SZVv8FmkIH1ycWNAY6UVgtRuTM6Jm4cKFrhJqO3bstMOe1Yrm+1eTZkRadpcOsmrVKt87tPBWrAE0cmpF8w0aa1o4RRqdQ7Wi+S4btjoiLRyWNOpcFWm+y8YRaaG2pO00orCq9lhUD0zzXSIntiYkxkuvAfmy5LMlZ9rNFd/5JdIQaIwiGjBggLTU1BrBpOGM9HzllVdk8eLFIXcB+LqXkuISOwcaU21oOp0APZvU9LZSsP/A6V+GYAuWNAJ9W/sUG01FSWwabHbu3NnUUzTrOBo5p8Fr1ola6cGUDb9ffsehTFjS8GK4cVLdUHJo6FrEppFC7dUpKSm1c6DR+dV0OgFi05iGZM/uvad/6bItfk1XzWgqhnkTHOpWC5o3V4QaFTtCjV6exkx4E9K/lYASUAJKQAkoATcTaNSShhUNE/q9994bUUsxOTFqCLQ33njDjrBxc0Fo3pSAElACSkAJKAEl4EngjCLNcXN+7Wtfk8GDB0ecSdsRapiaGQod6ngNT9D6WQkoASWgBJSAElACgRBoVKQhbCLZVeg5kID5ajQpASWgBJSAElACSiASCDQo0hw3J/OPOcGPZ7oh5jCqqqqS6urTV4Tmu3Am4tOc2Z/VmhbOktBrKwF3EqDeak49xdQXoQ7adydJzZUSUALBJNCgSJs2zcxWbFYUYGi3P8OrX3/9dXn88cfl5z//uTz//PN2sXXW/tuyZYv83//9n5SWlgYz3wGdC7cn98G0HGpNCwid7qwEooLABx98YEMi6urqpKSkRJ544gnbsUO4UY8xoz+dUBITUz/55JN2KSZGu1K//eIXv7D137///W+7H/PJMbksx5I4JyPkWb5JkxJQAkrAXwINijQsaYHMf4OoQwTRm/z444/l5ZdftiKPig1h5FRW/mYs2PupNS3YRPV8SqB1EMCK9uyzzwodTccT8OKLL8qCBQtsvbV582YrwPbu3WutbSwf9tprr8mePXvs4vMffvihPa6wsNCsmLHcHjN37lwrynbt2mWPoZOKSKPjq0kJKAEl4C8Bn1NwINA6dOggF1xwgV9WNOdiLB3z7W9/W5566ilZuXKlncGdSSjbtm1rF2eml8lnKr3k5GS56KKLpEePHnbx5o0bN9qFebF4nXXWWXafTZs2CRUfvVDSlVdeKT179rS90XfeecfGyuXm5vq1Vh3WNCyCuD1xefpjHXTuS9+VgBJovQSYxBjr2L59++Q73/mOZGdn24XCqbsmTZpk13Rlxng6ocxHxmomTKxLZ5SR41jS+BsrGfWLs34lx/Pdj3/8YwsPq1xzXKqttwT0zpSAEmiIgE9LGlaxK664IuA50Zg49rnnnrMTyDqDDdasWWMrNSqwv//97zJ9+nQ5cOCAILKo4OhZfvHFF3aSTITd7373OyvKli1bJs8884x8+umnVqi9/fbbtlJEsLE/eQzUdYllkMrYEX0NQdHtSkAJRA8B5lJkku6ysrJ6SxjiDJcl9dZ7771nV1qZOXOmrXOWLl0ql1xyifD30KFDZcKECXYeSYi1a9eu3lrWr18/K+zefffdegtd9FDVO1UCSiAYBHyKtEBdnU5GcBUw3QVWMmI1WO8P6xquTnqQVIL0RP/rv/7LVoqffPKJXWsOQTdixAi7Gj2uBCahZV9cpV/5ylfkv//7v+Xqq6+24g9hNmvWLOnevbt07NjRLyuakz9cnvSaeekAAoeKviuB6CVAfYK7csqUKdayTyeSemfs2LG2LsMaRlwt80QSY8a+WMoQZ4cOHRKEGPUciWNxac6bN8+GffTp08cuoYfrdPXq1WEP+WhqKW8+/JEcKN9gVkwJ7WoGTc2vHqcEWhOB00Sa4+rEjRjoUh+ILUaDPvDAA9ZShmWMc8TExNQzY2mXrKwsu8QUlRw9VdwIW7dulc6dO9t9qTipKPPz84WKDtfrZZddZt0RWNjYH6sYIi2Q5O3yDOTYaNu3prpGSgpLpbIitAs2Rxtnvd/wEsB1ST2EmzIlJUUYQED906lTJ1vHEKvWq1cvW/+kp6fLn//8ZzsIiTqMUe+4QZ14W45BoGHpd7Zde+211gr3pz/9KeSLnweL7NK9f5ePN/9CFu/+kxQcXdtssUaHndhlBmlocjkBU1Z1tXVSa8qrJedoOF53XCrKK6T48InQJpdTCWn2fIq0QAYMeOaWHmViYqLEx8dby1lFRUWDMRjsQ0KcEbB7zjnnyMUXX2wrSo6jkiNuzOml9u7d28ajsS/uStYRpVINNHFvnL8lLGkbDs2UjYf/I0WVOwLNlt/7b1y7VV578S158el/1r8+n7/CPEQnRpH5faJGdiw6XCxzP1gkWze23L00koWI/prfGFZh3Oskfm9YgLHKaDqVwLqDb0thxdZTN4bgLyz/DHA6++yzrdhChCEeCPrnHSs+1jNcn3xHCAiuTjqMdD55//zzz60wo7N5/vnn2/08O6V4Fb773e9agRap8WhHju2TLwr+JbO2/FI+3Pxos8QaDKZOnSqPPvqoPPbYYzZmb+3ateoObsLv/VD5RtlZskgqqouacLQfh5iy2rv7gPzlGdPWPPOK/PP5aTLvw0VypKTMj4MD24VncdOarTLzzdmBHejCvavrKmSj0QLFlTuDkjufAweacmYqJnqRuDtpkHBzDhkyxJr5PSst73Mj1tq3b28rO9YHzcjIEB5aFnT2TFSKEydOtEG4DDjo0qXLKRY6z33P9BmRRiwJQg/LWjDThsMzZX/pSslqmyc92423r6yk3GBewv6Q5/5noWR3ai8dOrWz5+YHHuxeDg/i8sWrJTm1rQwc1jeo9xDqk/HQbCv8RDqmDpTMpB4huTzuehoj3GYMpmF03wsvvGAbeiwzmk4S+HTnc5Ka0NE8L+Okl3lu2rXtefLLFvyEpZ+Y2T/84Q/Wak+dgxWNjuANN9xgY82oI4g5o5665ppr7FQcDG6iPsJKRhwuUw7Nnj3bljHl3rVrV9v5dLKOR+DBBx+0g6HOVBc6+zf1Pcc8pkv2/1G21WU19RQ+jyus2GE627VSWrXXiLXXZUfRAtlSOMeWVVL7Mok54e31eaz3RkTaSy+9ZAeQYXlE5DIi9nvf+57teHvX+97H698nCewqWSKrC96QTqZeo73pmnaWtI0PXtnTpuzfXSAz/jlTzh0/QmqqamT1sg1SsOegXH3rZfUDZE7mqOmfaoyRYdfWPfLZ3OUy5RvXNv1ELjjyWM0RWbDzWUlP7GbrtLzM860maGrWThNpu3fvti5GKqFA0o033igbNpi4BWPCZgkpRmgycpMe6Z133mkDa3knloMKj4rr7rvvtn8To4bFgcoNdykWNFwJw4YNs65OJx+4Q+nRIgARdk1JVLqOkGzK8Y0ds/fICtlSNEe2mkrMEWq8BzN1ze0sl15zkQwcfkI84U3esGqT7N6xT6qOVZsyqDXCqp/k9+shRYeKZe3yjbJ3135JTEqQIecMlPbZWbJu5SbZs3OftOuQZc7TRzp17WRMzcWyY/Nu2WSsdUWFJcbdecL0XFFeacXhtk07bOPUd3AvyenSUTat3yqH9hdKedlRSctMlwmTxkhMG5MZlyXvhyY/a0yLizXc9TRAWFi+8Y1v2MafQTD8xunEEI+JW4znjGcFMcfzgxUHazQDarDwEKuJ9c17+8iRI21HBqGBdQehzrPFdq7L1A90RHj+7rnnntM6PW4qopLKXcYCPdM8M7Nla7uLQibWiC2jvmEQABYvEmX12Wef2XIhNIPQDeoqxNXAgQOt2CKulr+Ji/3Rj35kLaZMz0FYBqzHjBlj6z46qY7HAKH3wx/+sEnWf3/Lqkt/kYX7npakI8F9BuvqqqTO/HOSp1gbdFWxbC9/X3Lq7pCENv57Nq666iobA8hyfcwxh9UZ0QZHrJe0I/CmU81UKPAmzpm2gXaAuGWeLX7vtDeEwPA8sR/PE88NswPwTFHfI/7YjzaFmQR8bUdArlq1ysYsYzDg+cODw7mLiorsgDee1/POO8/mzeERrvfKmhLZXjRPNhd+aJ6dT+xz44i1YOYJY8ANd15p6vY2Mvf9T+WDt+bI4LP6S+cenUw7slF2b99r2pR20ndIb9MuZMv+PQdkxaIvDLMj0iE7U4aMHGifg2ULVpo66Yj06NlV+g3pI5lZaVKw77AsW7BCDh8qkr07TqwIhFfoUEGhrFm5QQ6b9x75XaX3gHxJSk6UmdNnGQNFB9m/74Ccde4Q6TMwNB26QHgSu3mofLP1qm0zdVpu1tgTZZN1YZPE2mlKDJFGgxGoSJs8ebLP+yCwlhcJkeYkptLgRbrppptsI0OFxoNCA0cFygPnJCpOeriDBg2yDy5TeTQ1VdWWyWe7/iwHUzs09RQ+j9tVsliq6yrtd4WV26Vw79R6sba1aLupxLr4PC7QjYf2H5ZFsz8zgmqH9B/aR/L79pDP5q+QtSs2SJ8BPY2JukAKDxYbUZYoK5eskaXm4cjKzpCUtBQpKSqVXdv2yvyPFktqmhngYXovPCAXTBhpeknrjTl7odkvVSB/tKzcxqR9sWStfPzuPGnXMUuqKqpk++adpmc1UhYYdyju1yHnDJD4xASxBwV6MyHY/5SHxgjo/CysnOPMe8uINRqYw4cPC88SLnkaC4QUs9LTkCCoaCTYzm8c8UZjwIAWRjcjFGiEGF2IBYZ9vbcjDkhMd4NY4LkhnhSBR1zVm2++aa12WKb5rrmpzlhS9pSbuROvOSZztv26uac75fijVYfs38XHdsmyfX+TraaMnEanJS1riFpenom4Wl5O+vrXv26Z8jd1EusY8+4kGn5eWNAQEE69mWcEnmdiO+Xqeazn98H43MZYtGrrzMoHJ/VUME7b4DmO1ZVJfLK5WBvzMjFFcqrzo8Hj+AIO/G6xKlOvY9Hkd4tVzfme5wJXMSP5eU7wnvDM8PzAF0GHyGNQB8KLjj6dGzopTKOCuEL89e3b1x5Dx+j++++3zxuWT+/tnItjyRttEM8qovv999+3+aIdQ7DzXSAp3YROby+aH/TnZmfxQqmqLZcqUw7biufKvrKV5tkxYi1rnOwpXWWy6L9oPtP9IM7STNvRJTdHjpkY5aULV8oS82q3IV1WGwNAZrt06xbdvGG7TLrhYpn17gLT0d8snUxH3hSb7cjvMkJugWlzOvfIkf27DlhhN/HqcTLPtCHLzLnaG8/QwYKDJhtt5NCBQnnv9Q+NoaDUCLMk2bOrwLQz22TiNRfKS8+9LqPGniVduufYDuiZ8t3Yd7Em4iqzZ2nQy+WYEc/lxgVNnXmwYpMcqtxs67SeX7Y7PQMUaz5FGtYmp7Jp7EaD9b3T4+RhpKHxTlSC5AuXAw93U5NjSVuw7S3ZnNDyo5VOijWR/h1ubWq2Tzmu2gT1F5teCsLoaBlxf2KsWeXSzvRarph8sSw3vZi1qzbah2mLiSnr3rOLXHXL5aY3E2fjCRZ9skwystLlhq9eKWuWrpMVRsjFm0Zkx9bdUlVZK1+/7xpjgSuR16fOsKJuwxebZLN5SC7plyvFdWYKlM/XSVLbJCM6qiTDPKDX3zHJWtI8RfUpGfbzjwRjzIjvvKtlHxoTx3G4Yots8xJrfmbRr91oRIi1pKEgdpLpZrCGYbFxxBoxmDTqNC40TlgPOI5td911l22AEGg0EPzevbfPmDHD5mXOnDny/e9/335mqgeuhRikLHCzEuDuxHX6lfkGdkLo7i6bJyMmV8qHWx5tYK+mbUZYeCZiOTzF2qCO14StA+BdF3n/7eQ7NTXV+djge0sKNC66f4PItZO/Y6wa6Q3mwfMLf6X7yv2vSWnFznprWmJcunTPGGXEwHh585H/lbxbL5eEuMbv3/PazmeY4B1B+DA6lrnqeE7Yxm8coYULmgFkt9xyixV0bKMdwEL261//2naIsJBRBog9RB0dmyNHjtjwGYwAWEqZJYDOPh0oOi+e2wnVeeutt+xzRmw0ljM6OoTWEKpAZ4sRwFj3Ao2FzuhkhMrROea5me/cdlDeeSZxQzsJy9q2IiPWjiDWRHIzvuJ8FZR3rJFJxhuT2DZR9hmvzefzVph2IEF69c+T7Zt2yuql620IzvLFX8jl119kxNTZ1iJasO+grFy0Wgad3V8uuWa8LPx4icz5zwLp0auHLP10pZw3YYSxig2VWe/MkxWfrZEdW3bLh+bz2ecONlazdvKFaaOWL1wlI8cMM+1TlQwe3s8ItnE2H825sTjT12rXuzjo9ZmZDdHc90khTyeZmNuiym1WrPXKMl4wU6cN6Gisk35YNlwj0hqDPWrUKDu5Lg91c8QAIq38SI1xC7a8QGvsnpr6fSdjUh4/6QLpb8zLyaanQU+HlNkuQzp1zpaOpleyyQzi2L5pl1Qb92d+31zpmNPe7lNqLGlHS49KXp8e0i23szUnixFp640Qg2teX9O77ZcnW2W7pKSmmBE3lXKg4LCkGdM0D2lapnEDnD9E0jPSrEDMN8Kwo7lmMFKSqefjO29v8YfG9nA8xFrPdhMktXx40IQADQ49ftwuxFEyoSmdnocffljWrTOi2FjMaCxoZGg8cN1jBcBdxt/MtYWlmJ4+8ZmEDXhvd1yl/J4pF6x3BLrjMmImfCzOeV7WnOaVkVkeyTQIVGw1X1qLm3e+xo+m0SmvKbQ9UtMh19QIgb1GpI3OuU9yc3s0sqfztX8ybY+JfTpSuUcSY1PrxVmvzHHSNXOElO56QY7XBhCU5lz6y3c631icsSTzbOBmpB6iY8FADVybJMIEeE74mxfPCKP7cedzDo5FzNEpocPPiF2eQ8QeggyLHc8RzwadGO/tdKKwXjP4g2eJDhbPrmOJxlrK88T5A03mduR4TI15bkLT5vB81tQF/4nBDVls2o+y0jJJTG5rLF+HrdsTXp26djQd/zQpNcYD2peh5wyyFjZYFew9aOY6LZYJ/S+07VCPXl3l6JFy0z7tkGJjLRs20ogx0z7RRlGehNlUmpGeGe2NF8Ac39MYBzqaNq/WjDJNNIaJYecNMV6gIFkJY81I4xDVZ4g16rSK2hNWNmtd4cfRSDpNpBFvwXB0HopArGlkAFOzs7pAI9cN+Gt/eqr+nJR7I97tgp73SN9eposTxERczV4zcMCz0O0gAtPj3NduW9CuFBvbRtoaV2ZKSlt7TqbL8ExtzPcxxveRmpFiK6oDpifDg8VDVmfcEknmuGLzIDDcGZ8/2zp1zTYBoQeMafqwjU3jgaiuqjam/3hTUaUZd4bYGICEhFgTC3LMVoQFb8ySRPN90JL5vcYY14knv6Cd28eJjh+vkwrz0FTWGLEktX70aXycxMcmGgdW1SCeBosZMTe4SohRoqLH7cLqG4g2/qYXz7NDI4Q1jZHN/E45B7E1PFve2zk3sW00ToQn8JvG/UMIARNKB9rb93Ebp2xqExMnXVJGy4q3EuVeE8cVzLRk91+Me+CEy5PzJsVlSJ6J4+hlnhtcN13Sz7L1WTCviZUF6yQNOo0MjTd1HoI3FAmhgEUHwdGcTqdnXk0oqsTHmtH1sUmem5v9OSk2S3p3uMRYZs4XR5zFtzlR9zT15MSTIaqwIPO7ZaAGq8tgwcLdzHOB+EJYYeHylZyyQuSxD5YuLGqONZnniGeRd2I0KWO+P3jQ1Ifm3J7bmcwY0Yd44zOdIn4XuDf5jfBbaapFuvSAsdhUjZFx+Rf4uo0mb9tbuty6UavrTqwHG9cmUTqkGJds5oWyL2tjk8/rfSBTcBwx7UflxkpZbLwwySnJxprVRzaajn1GuzTrfqwzPvZa8wM8bFyVn89faUNiiDmrMHVUbXWt6dSnyoE9BcaTc1T2GfdlWyPyGPwWG9fGuj4zszJMWVVLGyPO25r2KTk1WXJMnPTwUQOl4miFxBsLXlysmc7LxDwnmbYvGKmmWqR4iynv/O8E43T156iqMYPu9r8ildWF9duS49ub8Bpi08ZLfsaF0jl9qHnu/atrGhRpKFp/RRq9etwuPGw0NN/61rfsj70+hy764Ii0Mb3ukWG9hgU1ZxWmsT9YtsGKDEec9TKxT7gF1mZNld0lm4JyPQYIvP3ah7JwzlJ7viEjTMSwSaeYTo3g6Z7fRdq1zzSxahvlH8//25RnrAweMUB6EsM2d5n89dlXrFuT/cZOHC3LjJt0yYLlZt/p5mGJsT0a3Jn4/z96+xP5aMYn9qFqZ4JER405K2iWJ5t5898xs6xh9b7u5qEJjlvYOa/3QwOnFDOa8ORDM1YqC1ONEHjSOaRZ7zQMjnsTSxeVPgKNjgYWA1wxbCdImWBorMNY13Br0pA8/vjj1rpAQ8JSaAgx7+3XXXednUoG986rr75qO0c8s7hIg9Xoe0JApHVPHiNLX0+SS373iOdXzf687sAMK9J8ibM2gQwd9CMnsMQtTD1Ah/LSSy+1ZcMktIhpLDShSMyvNtWM/v3pT39qBUAortnUa5zd9XZJje8kXTKGS3PFmZMHhBSdECxelMH48eOtYGVNVFz8iCSsVrfeeuspwtnXbxsRRblhdUbkId6witG54Vmj04IoRJRh3SZ2k3hPz+0sgcg1CUegA4VAY1+eSUcMOnkP9L3EiLQOdePkkp4nlgcL9PiG9l+8+/+MUWC5sTbV1osz2pseGefJiow/ya6SzQ0dGtB2gvjf+Md71r2IBewrN10iQ0cOMRayQ8YDs9kG8yNge5ng/oHD+tgY5VnvzZdVn62VlPRk6W08MwOMi3LB7M9lp4mHPrDvkJw9erD0NwPfBq3qZ9qWeXZwGxa3uPhYyevdw3w/xITVrDUD3vZKQryxno0cJDndA5sXtbGbrDUeyaItmaZcgluflZopa5iK65iJS6Odyc087xRxFmiddppIa+zGvL+nQfrjH/9oHwJ6PfSGaCzcmhyR1lBsSXPz3Tl9uGQl5Znh6SfEGWItmKnPoJ5y4ZHzjGXs5ELNBFeONMGU9DJijRBj9Od5488x7zm2F4ircr/pvbAfLtEePbuZ7XFmNOhe6dw1R4aeO9CMBM2VZDOQIMeYrXdt220HHeD3ZxRPB2OGTjCxB9s37pQaU9Fl57SzVrpzzTWDGWdzzEy/U70/r8UfGsRZPuVjejSd0gYJD82Ooh1BKyasWAykwYqG+4bgczovNAKILhpotiOwsN5885vftI0KGXAsO4g4YmNwWyIsvLcT8E6DxehDJnimEaLXT2VJHA2WuWCn2DZxZhg+1prmWVG889U2rp30zzbr8hqrGdYzLGeBVmTe5/T1N/FOP/vZz2zjjVjG3cwIWOa0+8c//iEPPfRQyEQa9RBigAEgWGncnAZmXxM04c9v9i4Tc4mI4hlgOhNGw/Ic8JtHOCOU6KDguqeeZqoTLI48P1iQeZYIDUA8MRiD0ZYIOkQa1jCeJ+olRj7zjjhj0AErzuC2xHrnvZ1OVP/+/a1blM4Tx3AdDBUcxzPdVG+OMdib33N80J8bYgM7pw2TbGM9w0KTa8RZeiKWWeNJ8dNKc6bfnennG2HUSa4zMcdwSM9KlVzTdgwybUKcaT8uvmqcaRvaGUvYPhOb1layTIeemQMuuXa8rDZxy4cOHJZM47Ls1K2jHXRgNLOUFpfKWecNtZ38Dmbfy667SFaZfbHUDRrRz/wmUs2Agw5y7W2T5Itl6+Sg8exkmvjpFOMVSjPWuJvvvk7S0oPj6uTe62pjgl4uCXHJkhbfWbp0Gm7bmbz0C6zlrKl12mkiDXcnlZm/Qov9GCnzgx/8wAZ20jjgx8ekzBIqWAAI+uRhpJdDz4aeEw8cPSYeKvz/c4wljgeUhxcrAY0OjQ/rfPKwMKEkPxRG32Cl4AHmnDx8gZihHZHWEhVjv/aXC6+OKf2bNNT2TA+M811fM+SY15kSIo2Xk3KMK7POPCFYkNp8OUVGZ/Pg4NKk0sR9SiJGrasZfYP707pMnROY9zEXm5jACaNMA1dXz7sd1rQgp+O1bVr8ofEUZ0HOvj0dLhOC9p3E75QGhRcNEi8SMTeILxKNEmvZ0ljhBiXxHW5OjvPebncw/xE3M8FM70Cj4pyLvyMpndfDLGqe3K/FxJnDgsB06hnWAaaxp0OJCIAxiUB16h1nhQE6oGzDQ0B9A9c844bDcokoQDxT1pQBddi//vUvW/dhIaLDSuNOXUgdiPUGcU5diNUmkpIv61VT8++INF/HU/9jdXb4OJ4c2gMnYd3i5SQ6QCTc/LQrTltAmcKdMANGaMLdOR/7N7SduE4GLrAvzx2J59SNqXvGSCPKukhOqokR/lKcBTWfpm3o0q2TfOM+354NxNQlV4+37QhtiPM7ye/dXfKMmKs1bUWc6TQ6abJpX3CLxiYYt+WXG5lag4EHWC6dsuMr4tDy+nS3bRGGB2f/W+6+1jmda98TY9Pkgrz/J9lt+9cbAZqT2dNEGg0Kw5axBPjTc6BhYD4ZhBo9Ix4UftyMjEGkUVlRafHAcD5cDQRnItKoNIkRYATc//7v/9pYBCo1YjVY9JieLpUjPV22IeoQkDzMuFgRcOQToeeG1K/D5W7Ixml5sELMPHCeiW24P72T3de4Or0T20+c4vRjvPd129/BfmgCvT/PxsHzWEdUsY3fNOLAU7ixnWfJ13a+cxL7OA2Ksy2S3ofn3FpfwbdkvqmTsLowtxaJuoVEh5LEFAzUUdR/iC6sbUxWSznROcQ1x7xqLA1Fo88+rCJBfBsjBZ944gm5/vrr7bmYMBtrDzGFv/3tb+390TFEiNNZRXRr8k2goefF994nt3oeh6UMqxll7FjEnD2xTtNmeG93vufYSEiIs04pg0Py7DTEo8F2xBgD4uTUtsLWUwknhK/n+U60Oafuy/cn6jXPPSPjc0JsigzpODlo5XIaMWfggL+WNCowejqIJ4Y+01NlyDJijN4NvRiEGgKLyglLG6qZxGd6q1yLCpDhzUxsi1UAVxDuImaiRohxDL1aBBsjPbkuFaRTwfpbfFjS6AW3hCXN3zzofqEl4Dw0BGs21eTc0jmmQqKzwqTQnqmh7Z77RPpnKulQJALVGTHbUMKCw7xcxDIhyuh00gFl+SfcXazzSV2F1RMxR92Gi4z4Jiw31C24oTkHiThDPAMvm6WnOBdWIuo8gtwd611DedHtzSeAxQ3mDKrxTA1t99wnUj6H6tmJFB5uyWcwy+U0kYb5noYBoeWIqcZunIbl97//vY0bwFXDJITEFWAVyMvLs2Z/3JwILCpAX4lKkLlviKthtBs9IGJrEI2INyxxxI+QL2agppJjhBSCzt9E5UrAKq4O4h80RQ+BYD40LUGN/NF58BYRDW1viTy09nPSMcPq1VBCpBEoTueQKVSw5LM/nUE6mMQ4OZ1X9qGO4hjCL5x6iL+psxACbEfs0RHFtYpnAXFHPdRQPdhQ3nR74ASwdFIW3paxhrYHfgU9Qgm0PIHT3J0IIn7UTLBJgCfiqbFEBYb5n6BcphfAvYmli8qISg1XJe5NXvQkEVhMH0D8hyMEuQ7HkLg++3C8sy/+ar7HisayUGyn0sUl4W+iZ8sxLTVowN986H5KQAmEngB1B/UaVn/cXVjhqUe86wPcZtRLuCz/8pe/yNNPP23FGJOoOmLMyT31iS/BxXbqL96ptxgEQgeUbXSEqTO5Nh1NTUpACSiBhgicZkljR3ofxE5QoTSWqKAw3+POxB1JBYcLgOBL/sZFQHwZZmcqSaxpxK+xP71Vp2fqeR0Cbjkvx1Kpcn6GWCMEmckdF4KzCLtnHILnOXx9xi3hWOd8fa/blIASaL0EWNKJhOhCgLHUFqM6sXT5Sli/iAekg0i8GgmLvtOx9HWM9zYsbogzLGq4Q7HGIdIY8ETdRafWVx3ofR79Wwkogegk4FOkMWKGisQflye9QZa2oeJD2GF9IziXuAx6ic8884x1GTDfDUG0xJcxFw7rtRG/QbwaAZx85/RoEXnkgVFVzGXD4tL0SBlBijXuT3/6kx10gGDj+v4kBCeuTtyc3jEK/hyv+ygBJRDZBKhjnnzySRvv+stf/tKGZBDcj+WeEbdO/YOowu3MIAA6h6wYwcAOYtNYGYJwCSxxuKL5noEI1E/e53DWemQONgY58U7HE0seYSCsZfzee+81KBIjm7bmXgkogWAQOM3dyUlxeVJ5EUNBj+9MLk8qOCbfRAQhoIhnIyHApprJGtnO8U5MEJNtMhiACtHZxv6MmHIS27GmYdjQUDkAAArYSURBVI2j10qP09mXChFrHNf1HLLrHNvQu+PqZGQPcSGalIASiD4CzLlF7CtWMjprTh3CQCcnOdM68PeLL75oLV3UN1i8cHdeffXVzq7WO4CHgOR5DubvctL48eOFF3WhZ93zk5/8xAo0zq1JCSgBJeCLgE9LGjsylQa9PKxZ/pj3qXwcgeZ5IbY7AsvZTq/Te5vznee7E4fmvS+9Wqdy9dy/oc+4GRhWf9VVV9kebEP76XYloARaPwHqE6z3/tQh1GmOiKKz6FjbmkLJU6A5xzvndv7WdyWgBJSAJ4EGRRojNnE7fvzxx9Zy5XlQpH3GokdcG2vC6ajOSCs9za8SUAJKQAkogegk0KBIAwdmfwL3/bWmuREhVjRm+8bFkWfiQDQpASWgBJSAElACSiASCJxRpBGbxoS0DAgg2NUft6ebbpoYEGLdiGO7++671YrmpsLRvCgBJaAElIASUAJnJHBGkcaRWNNYSYDRmATsR1Jiyg3ctbm5uTqiM5IKTvOqBJSAElACSkAJSKMiDWvaD3/4Q/nss89kyZIlfk95EW62TKjLcHcWzGUBZF9Bu6HMY2KSGc1q/lUd8z0nUyjz4sZr1dbUWjYpqY1PntwS+SdAnN8I085oOp0A8xbChrV1w5EYickKAJp8E6Bs+P36Grzl+4jgbGWqJjwWkeZlCc7dN34WJm0nMVgulMmZPUHbG9/Uj9cdl/Kj5ZKWnuZ7BxdtbVSkkVcGETBnEPOTsbC52ydfJA7t+eeft4sY33777a5wc/bI7SHxsYmyb/cBFxW/e7JypPSoHNhbKEOGDg5LpqhEWcqH+EVNpxNg6gnYMBdYOBKTYc8x6wJr8k2AsmH+tuaMPvV95jNvHT16tCxfvtyuIHPmPaPzWyZeJ+HNCWXqkdvdtDdJ2t40AP2YMZZsWbdTzhk1ooE93LPZL5FGdh2h9uqrr8q6detcK9QQaMwizoSUbhFo8Ms3gxbaxqXIfhVp4DgtVZZXSumhMunTu89p34ViA40bjZyKNN+0EWms8jFkyBDfO7TwVuYZY3Z+Tb4JUDYskRfqKT0QacxBqctb+S4X2iNSqGcVyM/LN+1NsrY3votFaqprZN+OAzJ44KAG9nDPZr9FGllGqGFSJz4N96e/s/2H6nZxcb700kt26RZmBw/1g3Gm++yZ30sy0zrIrq17pay08eW2znSu1vYdJvnCQyWSnpIlXbt0C8vtYUk7++yzraV49+7dYcmDWy+Kq3Pnzp12bkNWCAlHQqQhBhzLRDjy4NZr8nvFu8FyfoEskxeM+0GksfILL7d7WIJxv4GcA/c8L1z1oV7lJj+/p7T7sr3BS6HpJAFCaw7uN+uJJ6VJbo/8k1+49FPsoyYFkjfcDiyYjhiiYWNZFCaX9Z5wNpBzNndfYiJWrVpll28hb/fdd19AC6839/r+HI+4JW/z5y2Qmrpq6ZbXWWLjYv05tFXvwwOzZcN2mfvuYrlswiT5yqQrw3K/TGyalpZmrTWIAZY2C7XrKCw33shFEWgFBQXyi1/8wi7dxvJu4Uh0uHB3Yk1DsIU6xicc9+zPNYlFe+KJJ+zKMJQN9UwoU3Z2to1VpmwIFyBmMZxtQSjv/UzXIhaNZQjXrFljV8/BSh/KVN/ezP3UtDdV2t58CZ9YtP17CuTff31PLh57qdx805RQFkuTrhWwSOMqDCag8v7Zz34mR48elU6dOklWVlbIg1bJC+ZkVkZgjdABAwbI/fffb9cD5Tu3JRhtXLdZliz+XJLTkyWzXZokJEbvkjBY0Dav3y6z31kg1SUx8uOHfxLWIsMKQSA0azWylBm/a/6O1sQzvmXLFiHE4f3337fr6IaTBfMcsv4lnUJEW6itE+G8d1/XxoLG9EishUxfm6X8wpGw4L3wwgs2Li2cbUE47t3XNbGeMbMA5dKrVy9hKbJwpKzML9ubz5ZKSlpbaZvSVpLNK1oTBoEdW3fLLNPebFu9V5767dMRgaJJIo07c9b3ZL06Zw41elA0aqEYYYT1DBO7s1g7a+WxYDGWPbcmejcdOrSX3Tv2yZoV6yQuIU7iE+Ol2ixEj1UtGixrPCjlZeXW5btm+SaZ+e85UnG4Vr71rW8a62d44tGc3wvWNBoZ3mfMmGGnnOnatasd1chvOxosawizkpISe++rV6+2gpUl1R544IGwxaM55eMIM0QjDSGijXdStFjWKBusNE5ox/Tp0229h+U31PFoTrlgTaMD6rQFrPdcXl5uR3zyzESDZY0yoU3CaMCsAi+//LJdeowQoXCF3ZxobzrI7u17ZM3KDVJhRjOmZ6WZUY0V0iY2RuLj450ibLXvWM6OHjlq7r1CNq3bLu++9rHsXLtf7r33v42GcX88GgUTYyrl480tIeZR+93vfifEJ0yYMEH69OljR7O0RHwEDwI9SMzI9FaIi7vyyivtqLNQm/qbw+3zZUvk/Q/flf2H9kjb1ETJ7tJeUtNbfy+n6liNFJn4s4ojlbJ/22Hp32+Q3Dz5Zte5p/md0QHgt8bvGEGAxaC1J6ZSIP6M52rBggVCA3zHHXeEXaB5c59qlnpz4tMYJET5REOibCgjBrjQyN500012kfdwCTRv5rQF69evr4+Ro2xaoh3wvm64/6azwIvfJEYL2sLLL788bALNm8fSZZ/LJwtmy9adm6VNQox07JIlWR3CM52Od95a8u/amjo5VFAkNVW1sn3NHhuDdsdtd1ojU0teN5jnDopII0OMLiKW58knn7TzkjE3GWINq0RzA1odYUZMF40mFVSkijPPwoPX0mVLZe/+PfLFyi/kQMFBz69b5WfmQcvvmSddOneRm1wozjyh87uj0amqqrK/bX7jrT3xvDKCk0bfjeLMkz9CjcRzxCsaEmVDGWGhGTVqVNisZ2dizTPDs+O0CTw/rT1hLeNF7JmbxJknd6x87898X4pLimTlypWyfesOz69b5WfqsT79e0tGWobcftsdESXOnAIJmkhzTug8oLwPHjzYuiHoVTiCjf0cl6jz7hxLEKwzKSIPOT0TzPuOMGMyXYaZT5kyJeIsZ8496rsSUAJKQAkoASWgBPwhEHSR5nlRX4KN77GsYar3trAhxpxh3Kj+bdu2WfMx+zm+fURaJLk1PXnoZyWgBJSAElACSkAJ+EugRUWaZyYcwcY2zOCYwJ13Zz8GIzixFYgxTMe8VJQ5hPRdCSgBJaAElIASiBYCIRNp0QJU71MJKAEloASUgBJQAsEgENCKA8G4oJ5DCSgBJaAElIASUAJKoHECKtIaZ6R7KAEloASUgBJQAkog5ARUpIUcuV5QCSgBJaAElIASUAKNE1CR1jgj3UMJKAEloASUgBJQAiEnoCIt5Mj1gkpACSgBJaAElIASaJyAirTGGekeSkAJKAEloASUgBIIOQEVaSFHrhdUAkpACSgBJaAElEDjBFSkNc5I91ACSkAJKAEloASUQMgJqEgLOXK9oBJQAkpACSgBJaAEGiegIq1xRrqHElACSkAJKAEloARCTkBFWsiR6wWVgBJQAkpACSgBJdA4ARVpjTPSPZSAElACSkAJKAElEHICKtJCjlwvqASUgBJQAkpACSiBxgmoSGucke6hBJSAElACSkAJKIGQE1CRFnLkekEloASUgBJQAkpACTROQEVa44x0DyWgBJSAElACSkAJhJyAirSQI9cLKgEloASUgBJQAkqgcQL/H9DhU++tqriMAAAAAElFTkSuQmCC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "39b597a6-8eef-4ba5-91e7-f030c5fedae4",
   "metadata": {},
   "outputs": [],
   "source": [
    "class CodedSystemAWGN(Model): # Inherits from Keras Model\n",
    "    def __init__(self, num_bits_per_symbol, n, coderate):\n",
    "        super().__init__() # Must call the Keras model initializer\n",
    "\n",
    "        self.num_bits_per_symbol = num_bits_per_symbol\n",
    "        self.n = n\n",
    "        self.k = int(n*coderate)\n",
    "        self.coderate = coderate\n",
    "        self.constellation = sn.mapping.Constellation(\"qam\", self.num_bits_per_symbol)\n",
    "       \n",
    "        self.mapper = sn.mapping.Mapper(constellation=self.constellation)\n",
    "        self.demapper = sn.mapping.Demapper(\"app\", constellation=self.constellation)\n",
    "        \n",
    "        self.binary_source = sn.utils.BinarySource()\n",
    "        self.awgn_channel = sn.channel.AWGN()\n",
    "\n",
    "        self.encoder = sn.fec.ldpc.LDPC5GEncoder(self.k, self.n)\n",
    "        self.decoder = sn.fec.ldpc.LDPC5GDecoder(self.encoder, hard_out=True)\n",
    "    \n",
    "    #@tf.function # activate graph execution to speed things up\n",
    "    def __call__(self, batch_size, ebno_db):\n",
    "        no = sn.utils.ebnodb2no(ebno_db, num_bits_per_symbol=self.num_bits_per_symbol, coderate=self.coderate)\n",
    "\n",
    "        bits = self.binary_source([batch_size, self.k])\n",
    "        codewords = self.encoder(bits)\n",
    "        x = self.mapper(codewords)\n",
    "        y = self.awgn_channel([x, no])\n",
    "        llr = self.demapper([y,no])\n",
    "        bits_hat = self.decoder(llr)\n",
    "        return bits, bits_hat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "056e25da-3d0a-416d-ac6c-98b15bc042e7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EbNo [dB] |        BER |       BLER |  bit errors |    num bits | block errors |  num blocks | runtime [s] |    status\n",
      "---------------------------------------------------------------------------------------------------------------------------------------\n",
      "     -3.0 | 2.7896e-01 | 1.0000e+00 |      571312 |     2048000 |         2000 |        2000 |         1.8 |reached target block errors\n",
      "   -2.429 | 2.6327e-01 | 1.0000e+00 |      539174 |     2048000 |         2000 |        2000 |         1.7 |reached target block errors\n",
      "   -1.857 | 2.4783e-01 | 1.0000e+00 |      507546 |     2048000 |         2000 |        2000 |         1.7 |reached target block errors\n",
      "   -1.286 | 2.2687e-01 | 1.0000e+00 |      464636 |     2048000 |         2000 |        2000 |         1.7 |reached target block errors\n",
      "   -0.714 | 2.0312e-01 | 1.0000e+00 |      415988 |     2048000 |         2000 |        2000 |         1.7 |reached target block errors\n",
      "   -0.143 | 1.7154e-01 | 1.0000e+00 |      351316 |     2048000 |         2000 |        2000 |         1.7 |reached target block errors\n",
      "    0.429 | 1.1296e-01 | 9.9050e-01 |      231337 |     2048000 |         1981 |        2000 |         1.7 |reached target block errors\n",
      "      1.0 | 2.0114e-02 | 4.7150e-01 |       41193 |     2048000 |          943 |        2000 |         1.7 |reached target block errors\n",
      "    1.571 | 1.7617e-04 | 1.1633e-02 |        5412 |    30720000 |          349 |       30000 |        25.2 |reached max iter       \n",
      "    2.143 | 0.0000e+00 | 0.0000e+00 |           0 |    30720000 |            0 |       30000 |        25.3 |reached max iter       \n",
      "\n",
      "Simulation stopped as no error occurred @ EbNo = 2.1 dB.\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9gAAAJ/CAYAAACDRHuTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAACUNUlEQVR4nOzdd3gc1b3G8fesurTqsoplufduLDA2LhI9ENPCBRISSMGEJISQSkluQm7uhVCSEEIaIZAQU0N36E0uYGzLuPdeZEtWsbply9LcP2YlS7JkS7akmd39fp5nn5VmZ3d/6x+b+PU5c46xLEsAAAAAAOD0eJwuAAAAAACAQEDABgAAAACgGxCwAQAAAADoBgRsAAAAAAC6AQEbAAAAAIBuQMAGAAAAAKAbELABAAAAAOgGBGwAAPyYMSbRGHPIGGP5bsPaOefzLR7PPsFr3dLivH+c4LxQY0yV77z7TnBeljHmv40xHxpj9vrqPGSMKTDGvG+M+ZUxZtIJnp/Xop7PjDHmBOd+tencjs4BAKCnEbABAPBv10uKbPH719s5Z4GkBt/POSd4rdwWP5/ovDMleX0/f9j2QWNMuDHmd5K2Sfof3+tmSmqUVCcpQ9J5kn4m6TNjzAJjzIATvJ8kTZJ03UnOAQDAUQRsAAD82zd893/w3d9ojAlpeYJlWZWSPvP9mnOC15rluy+UNMAYM7CD85pe44ikj1s+YIyJlPSepNslhUl6V9JlkhIsy4qxLCtRUrjskP7fkvZKmiFp1AnqavIrY0xYJ84DAMARBGwAAPyUMeYMSRMllUv6iaQdskeHL2nn9I989zPaBnDfa42SlCZpk6TXfYdz257X5vhSy7Jq2zz2qKSZvp+/a1nWRZZlzbMsq6LpBMuyjlqWlW9Z1v9KGiTpTkmHOvqckj6QPfI9RNI3T3AeAACOImADAOC/mkavn7csq07SU77f25sm3hSw4ySd0c7jOb77PEnz2xxr5htBntbmNZseG9fivf9sWdajJ6xezWH7fsuy5p/gtL06NkL/38YY7wnOBQDAMQRsAAD8kG8q9pd8vz7V4t6S9HljTFqbpyySdNT3c047L9l0bL5OELBlT+2O8f38UZvHvi3J+N7n/05U/ym4T/ZIfaqkH3TzawMA0C0I2AAA+KcvSEqQtNWyrE8kybKs7bKDdKikG1qebFlWtaRlvl/bm/rddP11nmVZBbIXKOtvjBnc5rym5x6WtLjNY+f57j/zvUa3sSzroKRf+379kTGmT3e+PgAA3YGADQCAf2qaHv5Um+OdmSY+3RgT2nSwxfXXWyzL2u873NEodtPvi33T0pteI0xS0xZhK09e/il5RFKBpFjZK5ADAOAqBGwAAPyMb1Q5R/Z08H+1efgF2QuGjTTGTGvzWFPAjpU0ucXxplHpvBbH5rd5TMaYcHVw/bWkpBY/l52g9r8YYwrbub3c0XOaWJZ1SNI9vl9vOcEq5wAAOIKADQCA//ma7GudF1qWtbPlA74tuV71/fqN1k/TJ7K31pJaj0w3/dxyobGmn2e1OHaWpGjfz20DdmclyB4tb3tLOsFzWnpS0kbZW3396hRrAACgRxCwAQDwI8YYj6Sv+n5tOz28yT9999e0XHHbt6XWUt+vOS3Ob77+usW5uyTtkpRljBnS5jm1kpa0ec+Wo9YdhmXLsq6zLMs03VrU2imWZTVI+qnv1y8ZY8Z35fkAAPQkAjYAAP7lIkn9fD8/boyx2t4kve173CvpmjbPb3UdtjFmtOyVube1szBZ22niTfcfW5Z1pOWJlmXVS9ri+3XiqXywzrIs62XZAd+jYwufAQDgOAI2AAD+pe20766e3xSwvZKy1Xr/67aaFzrzXX89tc1rtPWB7/4MY0xmF+vsqjt8958zxsw64ZkAAPQSAjYAAH7CtzXVZb5fr5a9WFlHt7N8500zxoxo8TKLZW+xJdkj0jm+n1tef602x3IknS0pyvd7RwH7T7IXXgvVsWncPcKyrPmS3vL9en9PvhcAAJ1FwAYAwH98RVKYpApJ8yzLqj7BbZnsxcCkFqPYvq21mvavzlE711+3OHeb7G2xMiXN8R2ulpTfXnGWZa2R9ITv128ZY249tY/ZaXdJapQ0RdJVPfxeAACcFAEbAAD/0RSUX2t7DXQH/u27v6Hlvtc6NgJ9nuzrr3dYlrWng9doGsX+ou9+oWVZR0/wnrdKWuD7+Q/GmHeMMbONMfFNJxhjPMaY4caYn0i6tBOfo12WZa2S9Izv19mn+joAAHQXAjYAAH7AGHO2pNG+X/99onNbaDovTa2DbFPADvHd553gNZoCdtO5J9yeyzdCfoGk30uql3ShpNcllRtjqo0xxbL36d4ke2p3iqSFkn50ks/SkZ/r2NZjAAA4ioANAIB/aBq9rpD0bmee4JuyvaHN8yV7Be5DLX5v7/rrjh476f7XlmUdsSzrdklDJf3C95x9sq/N9koq9R27T9IEy7JmWpbV7rTzTrzXDkl/OZXnAgDQ3YxlWU7XAAAAAACA32MEGwAAAACAbkDABgAAAACgGxCwAQAAAADoBgRsAAAAAAC6AQEbAAAAAIBuEOp0AYEoJSXFGjhwoNNldKimpkYxMTFOl4E26Iv70BN3oi/uQ0/cib64Dz1xJ/riPv7Qk+XLl5dYltWn7XECdjcyxsyWNDszM1MPPfSQ0+V0qLq6Wl6v1+ky0AZ9cR964k70xX3oiTvRF/ehJ+5EX9zHH3qSm5u7q73j7IPdA7Kzs638/Hyny+hQXl6ecnJynC4DbdAX96En7kRf3IeeuBN9cR964k70xX38oSfGmOWWZWW3Pc412AAAAAAAdAMCNgAAAAAA3YCADQAAAABANyBgAwAAAADQDQjYAAAAAAB0AwI2AAAAAADdgIANAAAAAEA3CHW6AAAAAABocvjwYZWVlamqqkoNDQ09/n7x8fHasGFDj78POq+3exISEqLY2FglJSUpIiLitF6LgA0AAADAFQ4fPqzdu3crMTFRAwcOVFhYmIwxPfqeVVVVio2N7dH3QNf0Zk8sy1J9fb0qKyu1e/du9e/f/7RCNlPEAQAAALhCWVmZEhMTlZKSovDw8B4P14AxRuHh4UpJSVFiYqLKyspO6/UI2AAAAABcoaqqSnFxcU6XgSAVFxenqqqq03oNAjYAAAAAV2hoaFBYWJjTZSBIhYWFnfZ1/wRsAAAAAK7BtHA4pTv+2yNgAwAAAADQDQjYAAAAAAB0AwI2AAAAAKBX5eXlyRije+65p8feY+fOnTLG6Ktf/WqPvUdbBGwAAAAAcJGm8JmTk9PhOU3hceDAgb1WF04u1OkCAokxZrak2ZmZmcrLy3O6nA5VV1e7ur5gRV/ch564E31xH3riTvTFfejJycXHx5/2Nkld1dDQ0OvveTK1tbWSTlxbdXW1JMmyLNfV3xlNn/Hw4cPH1d9dPWn6M6qvr+/069XV1Z3W95SA3Y0sy5onaV52dvacE/1rk9Py8vJO+K9hcAZ9cR964k70xX3oiTvRF/ehJye3YcMGxcbG9up7VlVV9fp7nkx0dLQkKSQkpMPavF6vJHvla7fV3xlNnzEiIuK4+rurJ01/RmFhYZ1+vcjISE2aNOmU35Mp4gAAAAAQAO655x4ZY5SXl6cXX3xRZ511lqKjo5WUlKTrrrtOBQUF7T6vrKxMP/3pTzV27FhFR0crPj5eEyZM0J133qmamppW527ZskU33HCDMjMzFR4err59++qGG27Qli1b2n3toqIifeMb31BaWpqioqI0ceJE/fOf/zzh5ygrK9Ndd92lUaNGKSoqSvHx8TrvvPP07rvvtnt+VVWVfvCDH6hfv36KjIzUyJEj9dvf/laNjY2d+FPrXoxgB5t9K5VUmi/tiZaiEqXIBCkqQQoJc7oyAAAAAN3gT3/6k15//XVddtllmjVrlpYsWaLnn39eq1at0sqVKxUREdF87o4dO5Sbm6tdu3Zp8uTJ+ta3vqXGxkZt3rxZv/vd73TLLbcoJiZGkrRs2TKdf/75qqqq0mWXXabRo0dr48aNmjt3rl577TW9//77OvPMM5tfu6SkRNOmTdP27ds1ffp0TZ8+Xfv379ctt9yiCy+8sN3ad+3apVmzZmnXrl2aMWOGLr74YtXU1Og///mPLr74Yv31r3/VnDlzms8/fPiwzjvvPC1btkwTJkzQ9ddfr/Lycv3qV7/S/Pnze+hPuGME7GCz9DGNX/O0tKbN8bAYO2g3Be6W4bvtfavH4gnnAAAAgIu8/fbbWrZsmcaNG9d87Etf+pKeffZZvfbaa7rmmmuaj19//fXatWuX7r33Xt11112tXqekpKR5mrVlWbrhhhtUWVmpuXPn6vrrr28+7/nnn9d1112nr3zlK1q/fr08Hnui9N13363t27fr9ttv1+9+97vm82+99VZNnTq13dpvvPFG7d69W88++6yuu+665uPl5eXKycnRbbfdpssuu0xpaWmSpN/85jdatmyZrrrqKv373/9ufu8777xTkydPPqU/v9NBwA42qaNVljhJSVFGqiuXDpVLdRVSfY19q2x/2sgJhXvbhO/4doJ5ou/nxGPHIuOlEP4TBAAAwMkNvPMNp0volJ2/vtTpEnTbbbe1CteSNGfOHD377LNaunRpc8Bevny5Fi9erIkTJ+qOO+447nVSUlKaf/7kk0+0ceNGTZ06tVW4lqRrr71Wjz76qBYtWqRFixZp5syZqq+v19NPP63Y2NjjtuLKzs7W9ddff9xU8VWrVmn+/Pm64oorWoVrSUpISNAvf/lLXXHFFXrppZf07W9/W5L05JNPyuPx6IEHHmgO15I0aNAg3XbbbfrlL3/ZyT+17kG6CTbTbtXqI2NbL7BhWdLhqmOB+9DBFuG7zX3bx+oqpCPV9q1yb9frCY/1BfOETo6YJxwL8Z6QU/gDAAAAAAJbdnb2cceysrIkSQcPHmw+9umnn0qSLrroolbhtD2fffaZJOncc89t9/Fzzz1XixYt0ooVKzRz5kxt3LhRtbW1mjFjhuLj4487Pycn57iAvXjxYklSRUVFu/tjFxcXS7IXw5Psa6+3bt2qrKwsDRkypN33IGCj9xkjRcbZt4T+XXtuY6N0pKrrwbxp5PxIlX2r2NP1uiPiTh7M27uFe+3PDAAAAL/RUyPDblxFvCnsnmiRrqbH2gvGCQkJxx0LDbWjX0NDQ/Ox8vJySVJmZuZJa6qoqJAkZWRktPt40/Gm12w6v2kqd1vp6enHHSstLZUkffTRR/roo486rKVp+61TeY+eRsDG6fF47NHkyHgpcUDXntvYKB2uPIVgXi7VVdrPPVwpaXcXaw5tMW29M7cERs0BAADQa5pGfJsCZ3tKSkoktR+mO6vpuR2tLt5eTYWFhe0+vn///lbnNd0XFRW1e357r9P0nPvvv18/+clPOl1TV96jpxGw4RyP59j08K5qbLBHwDsTzA8dbHF/0L7WvLbEvnVVZHznQnnbAB8a3vX3AgAAQFAaMWKEIiIitHnzZpWWlio5Ofm4c5qmU0+YMOGU3+fss8+WJL3zzju69957TzhNvGlv6Ly8vHYfbxpxPuOMMyRJI0eOVHR0tFauXKmKiorjpom39zpN9TR9tpOJjY3V0KFDtX37dm3btu24aeId1dqTCNjwT54QKTrJvnXV0cOtA3fTra6dY60erzh2O7iza+8ZFnP8iHibUfKU4n3SDk/r42HRTGcHAAAIMpGRkbruuuv0z3/+Uz/+8Y/197//XabF3wn37t2rBx98UJL01a9+9ZTfZ/LkyZo2bZo++eQT3X///cetIl5aWqqYmBhFRkbqnHPO0YgRI7Ro0SK9+OKLuvrqq5vPe/HFF7Vw4UINHz5c06dPlySFhYXp+uuv19/+9jfdc889rVYRz8/P19NPP31cPdnZ2ZoxY4Zef/11PfHEE/r6179+3Dlr1qxRWlqaUlNTJUlf+9rX9NOf/lR33HGHXnjhheZ/JNixY4ceeeSRU/6zOVUEbASf0AgpNs2+dUXTqHnbEfGT3erKW6zS3vFCcGMlad39rQ+GhJ9kdDzB/keG2AwpNt2+D4vq2ucCAACA6zRtP/Xkk09q8eLFuuCCCxQXF6ddu3bptddeU1VVle644w7NmjXrtN5n7ty5ysnJ0d13362XXnpJOTk5sixLW7Zs0bvvvquNGzdq4MCBMsbon//8py644AJde+21uvzyyzVy5Eht2rRJr776qmJjY/XUU0+1GgW/99579cEHH+jhhx9Wfn5+8z7Yzz//vC655BK9/vrrx9XzzDPPKCcnR9/4xjf0yCOPaMqUKUpISNDevXu1evVqrV27VosXL24O2D/84Q/16quv6qWXXtIZZ5yhiy66SOXl5XrhhRc0c+bMdt+jJxGwgc461VFzy7JXWT9JEC/es0V9YkJbh/ejh6TqIvvWWZEJrQN3y/u4vva9N439ywEAAFwsOTlZS5Ys0SOPPKJXXnlF//jHP3To0CElJydr1qxZ+ta3vqVLLrnktN9n0KBB+uyzz/TAAw/o1Vdf1aOPPqrIyEgNHDhQP/zhD5uDrCRNmTJFy5Yt0//+7//q/fff17x585SSkqIvfvGL+u///m+NGDGi1WunpKTo448/1t1336158+YpPz9fI0aM0J///GcNHDiw3fDbr18/LViwQP/4xz/00ksv6emnn1ZDQ4PS09M1evRoffe73221BVlERITef/993XPPPXr++ef1+9//XgMHDtTPfvYzXXnllb0esI1lWb36hsEgOzvbys/Pd7qMDuXl5bXepguu0G5f6g91PFreNKW9pliqKpKqCqWq/VJjfSfezUgxKS3Cd0b7oTwmJagXduO74k70xX3oiTvRF/ehJye3YcMGjRo1qlff042riAc7J3vS2f8GjTHLLcs6bj80RrABNwuLsm9x7W+HcJzGRjt0V+1vcStsfV+5X6o5YAfzmmKpcE3Hr2dCfGG7ZfBuJ5RHJXKtOAAAAIIeARsIJB6PFJNs39LHdnxew1HfyHfbAL7Pd+87VlsqVRbYtxMJiWh/KnrbUfEI/nUYAAAAgYuADQSjkFB7VPxkI+NHD9vXf1d2MBre9PPhSql8l307kfDYdkbEfXU0/e5Nl8Iiu++zAgAAAL2EgA2gY6ERUkJ/+3Yih6vtIN4UuCv3tR/Gj1RJpVVS6ZYTv15UYosA7hsNj8uQEgb4blmslg4AAADXIWADOH0RXvuWPKTjcyzLXpjtuODdMpAXStWFxxZyO7C+49fzptlhO7EpdPc/9nN8P1ZJBwAAQK8jYAPoHcYc27879QQrMzY22td+t70uvGKvVL7bnoZesffY9mV7l7bzXh4pLrPjAB6bYV+vDgAAAHQjAjYAd/F4JG8f+5Yxvv1zGo7aAbx8l3RwV4t7XwCv3CdV7LFvuxYd//yQcCk+q3Xobg7iA+ztyVgVHQAAAF1EwAbgf0JC7euwE7KkgdOPf/zoYd+IdwcBvKZYKttm39oTFiMl9Ne4hhip9o3jR8KjEnr04wEAAMA/EbABBJ7QCPt68I6uCT9S4wvbu1sE8J3HAnhdhVS8QcmStDT/+OdHxh8/6t0ygIdH9+CHAwAAgFsRsAEEn/AY+zrwjq4FP1Qule/S2kVvamxmbOuR8PLddgAvXG3f2hPTp0Xo7t86gMdnSaHhPfbRAAAA4BwCNgC0FZUgRSWopM9BaVpO68csS6opOX7UuzmA77GnoNcUSwXtjH7LSHF9O16ALa6v5Anp+c8IAACAbkfABoCuMObYImz9so9/vLHBXvW87ah308+VBcduuz85/vkhEVLaGCljwrFb6mgpLLLnPxsAAABOCwEbALqTJ0SKz7RvA6Yd/3hD/fELsLUM4NVF0r7P7Fvza4ba09kzJkgZE+37tDH2VHcAAAC4hrEsy+kaAoYxZrak2ZmZmXPmzp3rdDkdqq6ultfrdboMtEFf3MeJnoTWV8tbvUPe6u2Krdomb/U2RdcWyKj1/1Zb8qg2OlNVsUNU7R3sux+khtDAD918V9yHnrgTfXEfenJy8fHxGjp0aK++Z0NDg0JCuDyrpVtuuUXPPPOM1qxZowEDBvTIe9x777369a9/rTfeeEMzZsxo9ZiTPdm6dasqKipOel5ubu5yy7KOm87ICHY3sixrnqR52dnZc3Jycpwup0N5eXlyc33Bir64j2t6crhaKlon7V8l7V8p7V8lc2CDYmr3KKZ2j1SUd+zcpMGtp5enT5Bikp2qvEe4pi9oRk/cib64Dz05uQ0bNig2NrZX37OqqqrX37OzNm7cqD/+8Y/66KOPtGfPHh06dEgpKSmaNGmSrrrqKn35y19WREREt79vWFiYJMnr9fbYn01T3dHR0ce9h5M9iYyM1KRJk075+QRsAHC7CK/Uf4p9a1JfJx1oCt2+W9E6qWy7fVv3yrFz47Nah+6MCVJseu9/DgAA0Gn/8z//o1/+8pdqbGzU1KlTdeONN8rr9aqoqEh5eXm66aab9Oc//1n5+e0tqgqnELABwB+FRUqZk+1bk4Z6qXhj69BduEaq2GPfNv7n2LnetGPXczfd4vvZi7gBAABH3XvvvfrFL36hrKws/fvf/9aUKVOOO+c///mPfvOb3zhQHU7E43QBAIBuEhImpY+TJn1ZuuRB6RvvSnftlb6zVLrqb9LUW6WBM6SIOHsxtS3vSAsekJ6/Xnp4rPTAYOmpK6T3fiGtfVkq3SY1Njr9qQAACCo7d+7UPffco7CwML355pvthmtJ+vznP6+333671bEXXnhBM2fOVHx8vKKiojRu3Djdd999Onz4cLuv8f7772vGjBmKiYlRUlKSrrjiCm3cuPGE9S1ZskRXX3210tPTFR4erqysLH3zm9/Uvn372j1/+fLluvjiixUbG6u4uDidf/75Wrx4cSf+JPwTI9gAEMg8IVKfEfZt/DX2scZGqXxn65HufSulQ2XS9o/sW5OIOCl9fOuR7pRh7NUNAEAPefLJJ1VfX6/rrrtOY8eOPeG5La+/vvvuu3XfffcpJSVFX/rSl+T1evXWW2/p7rvv1jvvvKN3331X4eHhzee/+OKLuvbaaxUeHq5rr71WGRkZWrRokaZOnarx48e3+35PPPGEbr75ZkVEROiyyy5TVlaWtmzZoscff1zz5s3Tp59+qv79+zef/8knn+j888/XkSNHdNVVV2no0KFauXKlcnJydO65557mn5Q7EbABINh4PPZiaEmDpTFX2scsy96bu23ori6Udi2yb03CoqW0sa1Dd5+RUmh4u28HAAA6b9Ei+/9zzzvvvE4/Z/HixbrvvvuUlZWlpUuXKj3dXmvlvvvu05VXXqn//Oc/euihh3T33XdLsle0/+Y3vymPx6OFCxcqO/vYYtjf//739fDDDx/3Hps3b9Ytt9yigQMHav78+crMzGx+7IMPPtCFF16o733ve3rlFXsdGMuy9PWvf12HDh3Sq6++qssvv7z5/N///ve6/fbbO/35/AkBGwBgX3sd38++jbz02PGqQmn/6hYrmK+WKnZLe5fatyYh4VLqaDts951o36eOsa8VBwCgO9wT3yMv2+1rVd9z8i2eTmT//v2SpH79+nX6OU888YQk6Wc/+1lzuJak0NBQ/eY3v9Gbb76pxx9/vDlgv/baayorK9MNN9zQKlxL0j333KMnn3zyuK2q/vznP6u+vl6///3vW4Vryf7HgMsuu0zz5s1rXgH8k08+0aZNmzRz5sxW4VqSbr31Vv3hD3/Qtm3bOv0Z/QUBGwDQsdh0+zb8wmPHastaj3TvXyWVbfMF8JXSZ/+0zzMhUuqo1iPdaWPtVdEBAEC3+eyzzySp3WnXw4cPV79+/bRjxw5VVFQoPj6++fxZs2Ydd358fLwmTpyo+fPntzredN30/PnztWzZsuOed+DAATU0NGjz5s2aPHnyCd8jJCRE06dPJ2ADAKDoJGlIrn1rUldpr1jeMnSXbJKK1tq3lU/7TjT2NdxZZ0lTbrEXZQMAoDNOc2S4I27bBzsjI0MbNmxQQUFBp5/TNNqckZHR4Wvu3r1b5eXlio+Pbz4/LS2t3fNbjoI3KS0tlSQ9+OCDJ6ylurq6VU1deY9AQMAGAJy+yDhp4Dn2rcmRWntv7v0rj4XuAxukks32bcVcafjnpJk/lvpN7vClAQAIJtOnT9eHH36oDz74QN/4xjc69Zz4eHv6fGFhoYYMGXLc403TzpvOa7ovKipq9/UKCws7fI+KigrFxcV1uqauvEcgYJsuAEDPCI+Wss6UzpojXf6odMtC6e4C6eY86exvS6FR0ua3pMfPtbcH2/mx0xUDAOC4r33tawoLC9NLL72k9evXn/Dcpu23Jk2aJEnKy8s77pytW7dq7969GjRokBISEiRJZ5xxhiQdNw1csgP0ypUrjzt+9tlnS5IWLlzYqc9xovdoaGhoXswt0BCwAQC9JzRC6jtJuvg+6fY10vTvS+Gx9tZg/7hEeuJiaev79qrmAAAEoYEDB+qee+7RkSNHdOmllyo/P7/d895++2197nOfkyR9/etflyT97//+r4qLi5vPaWho0I9+9CM1Nja2Gg2//PLLlZiYqGeeeea417/nnnuOW+BMshcmCwsL0/e//31t3rz5uMePHDnSKnxPmzZNI0aM0IIFC/Taa6+1OvfRRx8NyOuvJaaIAwCc4u0jnX+PdM73pCWPSZ/+Sdq9WJr7BTuEz/yxPYXcw78FAwCCy913362jR4/ql7/8pc4880xNmzZN2dnZ8nq9Kioq0oIFC7Rly5bmFcCnTZumn/zkJ3rggQc0duxYXX311YqJidFbb72ltWvXavr06frxj3/c/Pper1ePPfaYrr32Ws2YMaPVPthr167VzJkztWDBglY1jRw5Uk888YS+/vWva8yYMbr44os1fPhw1dfXa/fu3Vq4cKH69OmjjRs3SpKMMfr73/+uCy64QF/4whda7YP9wQcf6OKLL9bbb7/de3+ovYS/tQSZ9fsqtb60QTtLalRX3+B0OQAgRSVKOXdI318rnf9LKaaPtG+F9NyXpL+cI615UWrkf68AAMHl5z//udauXatbb71VFRUVevLJJ/Xggw/qjTfe0JAhQ/T444+3mmZ9//3369lnn9WwYcP01FNP6ZFHHlFjY6P+93//V++9957Cw8Nbvf7VV1+tt99+W5MnT9YLL7ygv/zlL0pKStLixYs1aNCgdmv68pe/rOXLl+v666/X6tWr9eijj2ru3LnaunWrrr76av3pT39qdf4555yjhQsX6vzzz9dbb72lP/zhDzpy5Ijy8vI0ZcqU7v9DcwFjMQ2v22VnZ1sdTeVw2h0vrtbz+Xuaf0/xRigzIVJ9E6Kaby1/T44JlzHGwYqDR15ennJycpwuAy3QE4ccqZVW/Eta9LBUtc8+ljREmvFDafw1ylv4MX1xGb4r7kRf3IeenNyGDRs0atSoXn1Pt60iDmd70tn/Bo0xyy3Lym57nCniQWZQnxiNSPSoRhEqrKhTSfVhlVQf1qq97W97EB7qUWZClPomRKpvfJQy2gTwvvFRigoP6eVPASCghUdLU74pTf6qtOpZaeFv7X22X/u2lPdr9U29RKo/WwqLdLpSAACAVgjYQeaWWUM00tqjnJwcNTRaKq46rILyQ9rX4lZQXmf/XHFI5bX12lFSox0lNR2+ZlJMeHMAt0fAm0bDI5WZEKUUb4Q8HkbBAXRRaIQdsid+WVr7krTwIalks4ZX/EX6/avStO9K2V+TwmOcrhQAAEASATuohXiM0uMjlR4fqckDEts9p+bwUe2vaBG6yw+1COR12l9xSGU1R1RWc0RrCyrbfY2wEKOMeN8oeKsAfmw0PDqc/xQBdCAkVJpwrTTuv6QNr6v6zXvkrd4hvftTadFv7S2/zpojRcY7XSkAAAhypBqcUExEqIamxmpoavvXQDQ2WiqpbhoFr2sdwCvsY2U1R7S7rFa7y2o7fJ+E6LAWI+DHpqBnJkYpKzFaKV6uBQeCnscjjblC+QfildP3sLTgQakgX/rwV9LHj0hTbpamfEuKSXa6UgAAEKQI2DgtHo9RalykUuMiNal/++ccOtLgC9vHRr5bBvCCcnsqenltvdbvb38UPCosRP0So5SVFK0s332/xGhlJdk/x0WG9eCnBOAqxkgjLpaGXyTtmC8teEjaudAO3Iv/JJ35dWnqd6XYNKcrBQAAQYaAjR4XFR6iIX28GtLH2+7jlmWptObIcdeAFxw8pL3ltdpTdkgVh+q15UC1thyobvc14qPC7LCdGN0cwvslRSsrMVr9EqMUGcZCbEDAMUYanGPfdn9qB+2t70mf/MHeV/uMG+w9thOynK4UAAAECQI2HGeMUYo3QineCI3vl9DuOZV19dpTZoftvQdr7Z8PHvLd16riUL0qCuo7vA48NTaiOXj3T4puDt9ZSVHKiI9SCIuwAf6t/9nSl1+0989e8JC08T/Ssr9Jy5+UJnxRmv59KXmI01UCAIAAR8CGX4iLDNOYvvEa0/f4RYwsy1JJ9RHt8QXvvS2C954ye1T8QNVhHag6rOW7Dh73/FCPUd+EqFYj4Memo3P9N+BX+k6SrntaKlpvL4C29iV7T+2VT0tjv2DvpZ3au/urAgC6xrIs/u4FR1iWddqvQcCG3zPGqE9shPrERuiM/sevhn60oVGFlXXaU3ZIew7Waq9vwbWmEfADVYdbLMJWetzzuf4b8ENpo6UvPC7l3GUH7VXPSWv+bd9GzZZm/EjqO9HpKgEAbYSEhKi+vl7h4eFOl4IgVF9fr5CQ07u0lICNgBca4lG/RDsUT9XxqwvX1TfYo96+8L2nzQh4Z6//HpAco8EpMRqYHKNBfWI0KDlGiTH8nwPgqOQh0uV/lGbdYa80/tlT0oZ59m3oBdLMH0v9pzhdJQDAJzY2VpWVlUpJSXG6FAShyspKxca2v3tSZxGwEfQiw0I0NNWroantL8J2Otd/J0SHaWBT8E6J0SDfbWBKjLwRfP2AXpPQX7r0IWnmj+xF0PKftBdE2/qeNHCGHbQHzbQXTgMAOCYpKUm7d++WJMXFxSksLIzp4uhRlmWpvr5elZWVOnjwoPr372BrpE7ib/jASXTm+u/dZbXaVVqjnSU12l5So52lNdpRXKPy2nqtrC3Xyj3lxz23T2yEHbh9I941RUfVt6hK/ZOiWfUc6Cmx6dJF/ydN/4G05M/Skr/aW3ztXCj1O9MO2sMuJGgDgEMiIiLUv39/lZWVaefOnWpoaOjx96yrq1NkZGSPvw86r7d7EhISotjYWPXv318RERGn9VoEbOA0tLz+e/KA1td/W5al4qrD2lFSY998oXtnaY12ltaquOqwiqsOa+mOsubn/GHFAhkj9Y2P0uA+vunmLUa++yVGKTTE09sfEwg8McnSuT+Tpt5qrza++E/S3mXSM9dI6ePsa7RHXSZ5+L4BQG+LiIhQRkaGMjIyeuX98vLyNGnSpF55L3SOP/eEgA30EGOMUuMilRoXqSmDW1/73dBoaX/FIe0oOTbqvXzzHlU2RmjPwUMqKLdvC7eUtHpeqMeof1J083TzgSnHpp9nxEXKw3ZjQNdEJdij1lO+JS3/h/TJI1LhGunfN0opI+xVx8d+QQrh/y4BAMDJ8TcGwAEhHtO88NqMYX0kSXl5xcrJyVF9Q6P2lNUeG/luMeV8X0WdtvsCeVsRoZ7mEe+WwXtQSgxbjQEnE+GVpt0qnXmTva3Xx7+XSjZJr9ws5d1r76M94YtS6OlNGwMAAIGNgA24TFiIR4P7eDW4z/GLrh060qBdZS2u9W4O4bUqqT6sTUVV2lRUddzzvBGhzcHbnm4erYHJ9hR0VjoHWgiLlM6aI51xo7TmBWnhb6Sy7dK870nzH5DO+Z50xg1SWJTTlQIAABciYAN+JCo8RCPT4zQyPe64xyrr6rWrpFbbS6q1s6RWO0qqm0fAK+uOak1BhdYUVBz3vPioMDt4J0drQIsR8IHJ0UqIJnwjSIWGS5O+bI9ar3tFWvCQVLxBeusn9s9TvyOd+Q0p4vS28gAAAIGFgA0EiLjIMI3rF69x/Vqvdm5ZlspqjmhnaY22Ny2yVlLbPPW84lC9Vu0p16p2Vjpv2mZsUEqMBiRH2+E72Q7g8VFhvfTJAAd5QqRxV0tjrpI2vSkteFDav1J6/xfSot9JZ39bmnKzFJV40pcCAACBj4ANBDhjjJK9EUr2RmjygKRWj1mWpeLqw9pZUqudJU0rnNtTzneVnnibscToppFvO3A3B/CUGMVFEr4RYDweadTnpZGXSls/sIP2nk/t67M/+YN01k3S2d+RvH2crhQAADiIgA0EMWOMUmMjlRobqbMGtRO+fduMtQzdO0pqtKu0Vgdr63Vwd7lW7C4/7nWTY8I1IDm6VQC3R76jFUv4hj8zRhp2vjT0PGnXx3bQ3p5nj2Z/+hcp+2vStO9KcX2drhQAADiAgA2gXSfaZsyyLBVVHvZNN7f3+N7pC947S2tUWnNEpTVH9Fk74TvFG66ByTG+672jW4TvGHkj+J8k+AljpIHT7dvefPu67M1vSZ/+SVr2uDTxemn67VLiQKcrBQAAvYi/zQLoMmOM0uMjlR4fqbPbhO/GRktFVXX2tPPSYyudN4XvkuojKqk+ovxdB4973RRvxLEVzn0rng9P82pAcozCQjy99fGArumXLX3pOWn/anvV8fWvScuflD57SjrjK9IlD0khzNwAACAYELA7YIy5RtJtkiZKKrEsa6CjBQF+wuMxyoiPUkZ8lKYOOT58F1bW+a73rvVNPfeNfpfZW42VVB/Wsp2tw3dYiNGglBgNS4vV8NRYDU/zalharAYmRyuU4A23yBgvXfNPqXiTPWV89QvS8n9I6ePtFccBAEDAI2B37KCkRyWlSfq+w7UAAcHjMeqbEKW+CVGaNrT1Y42NlvY3h287dG8rrtHmoirtPXhIm4uqtbmoWm9of/NzwkM8GtynKXjbobtpxDvEY3r50wE+fUZIV/5FGnaB9OLXpYW/tbf8Co1wujIAANDDCNgdsCzrPUkyxlzhcClAUPB4jDITopSZEKVzhqa0eqzm8FFtK7YD9paiKm0uqtLmomoVlB/SxsIqbSysanV+eKhHQ/p4NTzNq+FpsRrmC9/9k6IJ3ug9o6+UUh+SDqyXVvxLOvMmpysCAAA9zLUB2xhzl6QzJE2WNEjSro6maRtjPJK+J+mbkgZKKpb0gqSfW5ZV0xv1Aug5MRGhGt8vQeP7JbQ6Xn34qLYeqNbmoipf8LYD+L6KOm3YX6kN+ytbnR/RInjbo932iHdWYrQ8BG90N49HmnWH9O8bfaPYX2EUGwCAAOfagC3pXkllkj6TlHCSc38n+3rpVyT9RtIo3++TjDHnW5bV2HSiMeY5Sdee4LVyLcvKO/WyAfQWb0SoJmYlaGJWQqvjVXX12nLADttbiqq12ffz/oo6rd9fqfVtgndkmEdDU70anhrbPM18eFqsGi2rFz8NAtKoy6TU0YxiAwAQJNwcsIdYlrVdkowxayV52zvJGDNG0nclvWxZ1hdaHN8h6RFJ10l6psVT5ki69QTvW3GadQNwWGxkmM7on6gz+ie2Ol5ZV68tzdPMq7XlgD3dvKjysNYWVGptQevgHR4ijVy3yA7fvuA9LDVWmQlRjHijcxjFBgAgqLg2YDeF6074oiQj6eE2x/8m6deSvqwWAduyrCpJVQIQdOIiwzR5QKImD2gdvCtq631h2zfd/IA98n2g6rBW763Q6r2t/90tOjyk+brupunmI9JilREfKWMI3mhj1GVS6hjpwDp7666z5jhdEQAA6CGuDdhdcKakRklLWx60LKvOGLPS93iXGWNCJIX5bsYYE2m/rHX49MoF4Dbx0WHKHpik7IFJrY7/592PlDZ8gu8a7+rmxdVKqg9r1d4KrWoTvOOjwjQqI1Yj0+M0OiNOIzPs67wjw0J68+PAbTweKecO6YUb7FHsM25gFBsAgAAVCAG7r+x9qtsLvgWSphljwi3LOtLF1/2KpCdb/H5I0i7Zi6gBCALecKMzBybpzDbB+2DNETtsHzi2qvmmwiodrK3Xp9vL9On2suZzPUYalBKjURlxvpsdwBntDjIjZzOKDQBAEDCWHyzi03QNdnuriBtjtkkKsyyrfzuPPSU7KCdallXewzXeLOlmSUpLS5v83HPP9eTbnZbq6mp5ve1e0g4H0Rf36UpPLMtS+WFLe6oatbuqUXt994U1lhrb+Z/ZmDCpn9ej/nEe9Yv1qH+sR5lej8JDCN0n46/flZTiTzR23f06HJ6sT8/+qyxPmNMldRt/7Umgoy/uQ0/cib64jz/0JDc3d7llWdltjwfCCHatpNQOHotscU6PsizrMUmPSVJ2draVk5PT0295yvLy8uTm+oIVfXGf7uhJXX2Dth6o9m0bVqWNhfb2YQdr67XpYKM2HWze5IDR7k7y2+9K40yp5A1FFK3VLO/OgBrF9tueBDj64j70xJ3oi/v4c08CIWDvkzTaGBPRzjTxTNnTx7s6PRwAukVkWIjGZsZrbGZ88zHLslRUeVgbfGF74/4qbdhfqe0lNdpWbN/+s3p/8/nxUWEamR7bHLpHZcRxbbc/alpR/IWvHFtRPCzy5M8DAAB+IxAC9jJJF0o6S9LCpoO+RckmSlrgTFkA0D5jjNLjI5UeH6ncEccm4JxotHvJjjIt2XH8td0jM+wF1Rjt9hMjPy+ljZWK1tr7YgfQKDYAAAiMgP28pLsl3a4WAVv2ftfRkp52oCYA6LJTHe1+g9Fu/8EoNgAAAc21AdsY8xVJA3y/9pEUboz5me/3XZZl/UuSLMtaY4z5o6RbjTEvS3pT0ihJt0marxZ7YAOAv+mJ0e6mAM5ot0MYxQYAIGC5NmBL+oakWW2O/cp3P1/Sv1ocv13STtmreF8qqUTSHyT93LKsRgFAgGG02495PFLOndLzX2YUGwCAAOMX23T5C2PMbEmzMzMz58ydO9fpcjrkD8veByP64j6B0pMjDZb2VTdqT1XrW3X98ecaSekxRlmxnuZb/ziPEiOMa0a7A6IvVqOy838gb80ObRl6swr6Xep0RaclIHoSgOiL+9ATd6Iv7uMPPelomy4Cdg/Izs628vPznS6jQ/687H0goy/uE8g9aTvavWF/lTb6Rrsb2tm4OyG6xWh3ur2N2LA0ryOj3QHTlw3z7FHs2AzptpV+PYodMD0JMPTFfeiJO9EX9/GHnhhjAnYfbABAF53o2u4tRdWtp5kXVqq8tl6fbi/Tp9tbX9s9uI9Xo3zXdY/OiNPIjFilx3Ftd6eMuFRKGycVrZE+e0qacrPTFQEAgNNEwAYANIsMC9G4fvEa16/1td2FlXXauL9K6/f7gndhlbYXV2vrAfs2b9Wx10iIDtOodDtsN414OzXa7Woej5Rzhz2Kvei30hk3+PUoNgAAIGADAE7CGKOM+ChlxEcpd2Tr0e7NRVXNwdteybxK5bX1Wry9VIu3lzafG+IxGuxbyXxUi+CdFhcR3KPdIz8vpY+TCtdIn/1TmvJNpysCAACngYANADglkWEhGt8vQeP7JTQfsyxL+yvqmsP2+v2V2ri/UjtKarTlQLW2tBntTowO00jfNd1NwXtoahCNdhsjzbpTev56e0XxM25kFBsAAD9GwAYAdBtjjPomRKlvQpTOHZnWfPzQkQZtOVDVvKDahv3H9u3uaLR7dN84je+XoIlZCRrTNy5wQ/fISxnFBgAgQLCKeDdimy6cDvriPvSkZ1mWpbI6S3uqGrW7qlF7ffdFNZba/j9TiJH6x3o0OMGjzMh6jU6LVmq0kSdAppenFH+qsevu0+HwRC2Z8pgaQ8KdLqlL+K64E31xH3riTvTFffyhJ2zT1YvYpgungr64Dz1xxqEj9rXda/dVaNWecq3cU64tB6rV9v+u4iJDNSErQZOyEjSxf4ImZiUqKca/gmkzy5L+OsMexf7cA343is13xZ3oi/vQE3eiL+7jDz1hmy4AgF+ICg/RhKwETchK0PVTBkiSqurqtWZvhVbsKdf7K7Zq76FQFVcd1sItJVq4paT5uf2TojUxy55WPrF/gkZn+MnUcmOknLuk577kuxb7BiksyumqAABAFxGwAQCuFxsZpmlDUzRtaIrGmL2aNWuW9lXUaeXucq3cc1Cr9lRodUG5dpfVandZrV5ftU+SFBZiNCoj7ljozkrQoJQYd65cPuKSY9diL/+ndPYtTlcEAAC6iIANAPA7xhhlJkQpMyFKl47PkCQdbWjUpqIqrdxT7gve5dpaXK3Veyu0em+Fnlq8S5IUHxWmCb6wPck3Uu6KqeUtR7EX/U6afCOj2AAA+BkCNgAgIISGeDSmb7zG9I1vNbV89d4KO3T7bsVVh7Vgc7EWbC5ufu6A5GhN6HdsavmYvnGKCHVgavmIS6T08VLhakaxAQDwQwRsAEDAio0M0zlDU3TO0BRJ9srlLaeWr9xTrjUFFdpVWqtdpa2nlo9umlruW0BtYHJ0z08tbx7F/qK06LeMYgMA4GcI2ACAoNHe1PL6hkZtKrSnljetWr61uFqr9lZo1d4K/bO3p5aP+FyLUex/SGd/q/vfAwAA9AgCNgAgqIWFeDQ2M15jM+P15bPtqeWVvlXLV+4p1wrf9dwl1e1PLW9aPG1S/0SNzohTeKjn9ApqNYr9O2nyVxnFBgDAT7APdjcyxsyWNDszM3PO3LlznS6nQ/6wcXswoi/uQ0/cyYm+WJal0jpL2ysatb28QdsrGrWzolFHGlufF2qkAXEeDUnwaHBCiIbEe5QSZbo+tdyyNHn5DxVbvU1bht6kgn6zu+/D9AC+K+5EX9yHnrgTfXEff+hJbm5uu/tgE7B7QHZ2tpWfn+90GR3yh43bgxF9cR964k5u6UvLqeX2SPdBbSuuOe68FG+Eb4Tbnlo+PitB3ohOTCDb9Jb07HWSN0363ipXj2K7pSdojb64Dz1xJ/riPv7QE2NMuwGbKeIAAJyC9qaWVxyqb76Oe8Xug1rhm1r+/oYivb+hSJI9A3xEWuyx0N0/UUP6eBXiaTPKPfxiKWOCtH+VlP+kNPXbvf0RAQBAFxGwAQDoJvFRYZo5vI9mDu8jyZ5avqu0Viv2HNTK3eVasadc6/dVamNhlTYWVum5ZXskSd6IUE3IivctoJaoif0TlOKNsK/FfvY66eOHpeyvuXoUGwAAELABAOgxxhgNTInRwJQYXTmpnySprr5B6/ZVaIUvcK/cXa6C8kP6eGupPt5a2vzcrKQoTeqXpp/HjlJK1QYdXfp3hZ5zq1MfBQAAdAIBGwCAXhQZFqLJA5I0eUBS87EDlXVa0bxi+UGt3luhPWWHtKfskGo8l+jv4RtU9u6DunXFaI0ZkKZJ/RM1KStB/RKjen5vbgAA0GkEbAAAHJYaF6mLxqTrojHpkqSjDY3acqDaHuXelanNG1/T8MatGrv/ZT2x93N68uOdkqQUb3jzFmFdWkANAAD0CP5fGAAAlwkN8WhURpxGZcTpS1P6S5v+T3r2Wt0Z+5b6nPFNLdtX59ub+4je33BA7284IMleQG14aqwm9T+2N/fQ1HYWUAMAAD2CgA0AgNsNv0jKmKjw/Sv1rdgF+tZXvyPLsrS7rNa3Yrm9avn6/ZXaVFSlTUWtF1Ab3y/eF7oTNXlAopJiwh3+QAAABCYCNgAAbmeMb0Xxa6VFD0uTvyYTHq0ByTEakByjyydmSmpaQK3y2DZhvgXUPtlWqk+2HVtAbURarKYMTtKUQcmaMjjJXrEcAACcNgI2AAD+YPhFUt9J0r4V0vInpanfOe4UewE1e5RaGiRJOlBVp5W77b25P/OF7qZR7qcW75IkDU31asqgJJ092A7cqbGRvfnJAAAIGMayLKdrCBjGmNmSZmdmZs6ZO3eu0+V0qLq6Wl6v1+ky0AZ9cR964k7B3Jek0nyNX/MrHQlL0KdnP6bGkK6PPNc3WtpR0aiNZQ3aVNagLeWNOtLQ+pz0aKORSSEakRSikUkeJUZ6TviawdwTN6Mv7kNP3Im+uI8/9CQ3N3e5ZVnZbY8TsHtAdna2lZ+f73QZHcrLy1NOTo7TZaAN+uI+9MSdgrovliX9Ldcexb7w/6Rpp78v9pGjjVpTUKFPt5dqyY4yLd9Zppo2iXtgcnTzdPKzByerb0JUq8eDuicuRl/ch564E31xH3/oiTGm3YDNFHEAAPxF07XYz1wjffywlP11KTz6tF4yPNTTPK38O7lSfUOj1hZUaMmOMi3ZXqplOw9qZ2mtdpbW6vl8e+G0rKQoTRmUbE8pH5R0kncAACB4ELABAPAnwy6U+p4h7ftMyn+iW0axWwoL8dj7avdP1C2zhuhoQ6PW76/Uku1l+nR7qZbuLNOeskPaU7ZXLy7fK0lKjjSadWClzvaNcvdPipYxbA0GAAg+BGwAAPxJ8yj2f3XbKPaJhIZ4NL5fgsb3S9CcmYPV0Ghpw/7K5inlS3eUqfRQvV7+rEAvf1YgScqIj9SUQUmaMtge5R6YTOAGAAQHAjYAAP5m2AUtRrH/Lk37bq+9dYjHaGxmvMZmxuumGYPV2Ghp7n8+VEPyYC3ZXqYlO0q1v6JOr67cp1dX7pMkpcZG+MK2vTXYkD4xBG4AQEAiYAMA4G9ajWL/3jeKHeNIKR6PUf+4EOWcM0hfO2eQGhstbTlQrSU7Su1R7u1lOlB1WPNW7dO8VXbgTvFG2Aum+Ua5h6V6CdwAgIBAwAYAwB8Nu0DKnCwVLPddi917o9gn4vEYjUiP1Yj0WN0wdaAsy9K24mot3m4vmrZkR5mKqw7rjdX79cbq/ZKk5JhwnTUoyd6Le0iyhqfGyuMhcAMA/A8BGwAAf9Q0iv301Y6PYp+IMUZDU2M1NDVWXzl7gCzL0vaSmubp5J9uL1VR5WG9tbZQb60tlCQlRofprEFJmjYkRdOHpWhwClPKAQD+gYANAIC/Gnq+K0exT8QYoyF9vBrSx6svTekvy7K0q7TWF7btUe59FXV6Z12R3llXJEnqGx+pc4baYfucoSlK8UY4/CkAAGgfARsAAH/VchR70cOuHcU+EWOMBqbEaGBKjK490w7cew8e0uJtpVq0tUQfby3Rvoo6/Xv5Xv3bty3YqIw4zfCF7bMGJikqPMThTwEAgI2ADQCAP2s5ir3s79I5tzld0WkxxigrKVpZSdG65swsNTZa2lBYqY+3lmjhlhIt3VGmDfsrtWF/pR5bsF3hIR5NHpCo6cNSNGNYisb0jVcI128DABxiLMtyuoaAYYyZLWl2ZmbmnLlz5zpdToeqq6vl9XqdLgNt0Bf3oSfuRF+Ol1S6XOPX/I+OhMXr07MfU2NIZK++f2/25EiDpa3ljVpf2qB1JQ3aWdmoln+TiQmTRiWFaExyiMakhCg12tMrdbkR3xX3oSfuRF/cxx96kpubu9yyrOy2xwnYPSA7O9vKz893uowO5eXlKScnx+ky0AZ9cR964k70pR2WJT1+vlSQL13wq14fxXayJwdrjmjx9lIt3FKiRVuLtafsUKvH+ydF65yh9uj2tCHJSogOd6ROJ/BdcR964k70xX38oSfGmHYDNlPEAQDwd83XYn/BXlH8zG/43bXYpyoxJlyXjMvQJeMyJEm7S2u1cGuxPt5aoo+3lmp3Wa12L92tZ5fuljHSuMx4O3APTdHkgYmKCOX6bQBA9yFgAwAQCIaeJ2Vm26PYyx6Xzvme0xU5on9ytK5PHqDrpwxQQ6OltQUVWrS1RIu2lGj5roNavbdCq/dW6M952xQZ5tGZA5OaF0wblR7H/tsAgNNCwAYAIBAcN4p9U9CMYnckxGM0IStBE7IS9J3coTp0pEFLd5Zp0ZZiLdpaqg37K7Vwi714miQlx4Tb24H5tgTrmxDl8CcAAPgbAjYAAIFi6HlSvzOlvcuCehS7I1HhIZo1vI9mDe8jSSquOqxPttmj24u2lmh/RZ1eX7VPr6/aJ0kanBLTvPf21CHJiosMc7J8AIAfIGADABAojJFy7pTmMordGX1iI3T5xExdPjFTlmVpe0mNFvlGtD/dXqrtJTXaXlKjpxbvskfD+8X7Rrf7aGJWgsJDg3eFcgBA+wjYAAAEkiEtRrGX/k2afrvTFfkFY4yG9PFqSB+vbpw2UEcbGrVqb7kWbSnVoq3FWrG7XJ/5bo98uFUx4SGaMji5eTr5sFSvjOH6bQAIdgRsAAACSctR7E8esUexI9y9l6gbhYZ4NHlAkiYPSNL3zh+m6sNHtcS3HdjHW0u05UC1Ptx4QB9uPCBJyoiPbJ5+Pm1oiuKjmE4OAMGIgA0AQKAZ0uZabEaxT5s3IlTnjUrTeaPSJEmFFXVatNUO2wu3FGt/RZ2eW7ZHzy3boxCP0aSsBDtwj+ijsX3jWZ0cAIIEARsAgEDTtKL43KsYxe4h6fGRunpyP109uZ8aGy2t31+p+ZuLNX9zsT7bdVD5vttv3tuspJhwzRiWolnD+2jGsD7qExvhdPkAgB5CwAYAIBANOVfqd5a0dymj2D3M4zEamxmvsZnx+k7uUFXV1euTbaV24N5UrILyQ3pt5T69ttJenXxsZpxmDrOnk58xIFFhISyWBgCBgoANAEAgar4Wm1Hs3hYbGaaLxqTrojHpsixL24prtMA3uv3p9lKtLajU2oJK/Slvm7wRoTpnaLJmDU/VzOEp6pcY7XT5AIDTQMAGACBQtRrF/ps0/ftOVxR0jDEamurV0FSvvj59kOrqG7RkR5nmbyrWgi3F2nqgWu+sK9I764okSUP6xGjW8FTNGtFHUwYlKTIsxOFPAADoCgI2AACByhgp9y7pX1dKHz8inTmHUWyHRYaFNK82Lkl7D9ZqweYSzd98QB9vLdW24hptK96hJz7eoYhQj6YMTm4+f0ifGLYCAwCXM5ZlOV1DwDDGzJY0OzMzc87cuXOdLqdD1dXV8nr5C5bb0Bf3oSfuRF+6yLI0acWdiq/cqG2Db9Ce/l/o9regJ93jaKOlbeWNWlPSoDUlDdpV2djq8eRIo3EpIRrXJ0Sjk0MUFXrisE1f3IeeuBN9cR9/6Elubu5yy7Ky2x4nYPeA7OxsKz8/3+kyOpSXl6ecnByny0Ab9MV96Ik70ZdTsO1DexQ7Kkm6fbUUEdutL09PekZx1WEt2lrsm05eorKaI82PhXqMzhiQ2Dy6PToj7ritwOiL+9ATd6Iv7uMPPTHGtBuwmSIOAECgG5wrZU2R9iyRlv5NmvEDpytCJ/SJjdCVk/rpykn2VmBr91U0X7v92e5yLd1RpqU7yvTgO5uU4g23VyYf0UfTh6Yo2ctWYADgBAI2AACBrmlF8X9dKX3yB+msOd0+io2e5fEYje+XoPH9EvTd84ap4lC9Ptla0rz39v6KOr28okAvryiQMdK4zHgNiDgi78AyTcxKUChbgQFAryBgAwAQDAbnSllnS3s+ZRQ7AMRHhelz4zL0uXEZsixLWw9UN4ftJdvLtHpvhVZLmrdtsWIjQzV9aIpmDe+jnBGpSo+PdLp8AAhYBGwAAIJB8yj2FYxiBxhjjIalxWpYWqxumjFYtUeOasn2Mj3z0Qptq4nQ9pIavbW2UG+tLZQkTegXrwvHpOvC0WkamuplZXIA6EYEbAAAgsXgnBaj2I9JM37odEXoAdHhocodmSpTGKGcnBztLq3V/C32YmmLthZr1d4KrdpboQff2aRBKTG6cHSaLhyTpolZiQrxELYB4HQQsAEACBbHjWLfzCh2EOifHK2vJA/QV84eoENHGrRwS7HeW1+k9zcUaUdJjf66YLv+umC7UrwRumB0qi4YnaZpQ1IUGRbidOkA4HcI2AAABJPBOVL/qdLuxYxiB6Go8BB7eviYdB1taNTyXQf17voivbOuUHsPHtKzS/fo2aV7FB0eopwRfXTh6HTljkhVfHSY06UDgF8gYAMAEEyaRrGfupxR7CAXGuLRlMHJmjI4WT+7dJQ2Flbp3XVFend9odbtq9Sbawr15ppChXqMzh6crAvHpOmC0WnKiI9yunQAcC0CNgAAwWbQrGOj2Ev+Ks38kdMVwWHGGI3KiNOojDh97/xh2nuwVu+tL9K764q0dGeZFm0t0aKtJfr5a+s0vl+877rtdA1jkTQAaIWADQBAsGk5ir34UXsUOzLO6argIv0So/W1cwbpa+cM0sGaI/pw4wG9t75I8zcX21uA7a3QQ+9u1sDkaF04Jl0XjE7TGf1ZJA0ACNgAAASjlqPYSx9jFBsdSowJ1xcm99MXJvdTXX2DFm0p0bvrC/X+hgPaWVqrxxZs12MLtis5Jlznj7JXJD9nKIukAQhOBGwAAIKRMVLOXdJTlzGKjU6LDAvR+aPTdP7oNDU0WvYiaesK9e76Iu0uq9Xz+Xv0fL69SNqs4X104Zg0nTsijUXSAAQNAjYAAMFq0Eyp/zRp9yeMYqPLQjxGZw1K0lmDkvTTS0dpU9GxRdLWFlTqrbWFemttoUI8RmcPTtKFo+2p5H0TWCQNQOAiYAMAEKyar8W+7NiK4oxi4xQYYzQyPU4j0+N023nDVFB+SO+vt8P2p9vL9PHWUn28tVS/eH2dxmXai6RdMCZNI9JiWSQNQEAhYAMAEMxajWL/VZr5Y6crQgDITIjSjdMG6sZpA1Vee0QfbTqgd9cVKW9TsdYUVGhNQYV+895m9U+Kbl6RfPIAFkkD4P8I2AAABLNWo9iPSmd9k1FsdKuE6HBdOamfrpxkL5L28dYSvbuuSO9vsK/bfnzRDj2+aIeSY8J13qhUXTg6XdOHsUgaAP9EwAYAINgNmikNOEfa9TGj2OhRkWEhOm9Ums4bZS+S9tnuY4uk7Sqt1Qv5e/VC/l5FhYUod2QfXTqur3JH9lF0OH9lBeAfjGVZTtcQMIwxsyXNzszMnDN37lyny+lQdXW1vF6v02WgDfriPvTEnehLz0g4uFoTV/236kO9+vTsv6khNLrTz6Un7uRPfbEsS/uqLS0/cFQrihq0o7Kx+bHwEGlCnxCdlR6q8X1CFBHiv9PI/aknwYS+uI8/9CQ3N3e5ZVnZbY8TsHtAdna2lZ+f73QZHcrLy1NOTo7TZaAN+uI+9MSd6EsPsSzpH5fao9i5P5NmdX4Um564kz/3paD8kN5as19vrNmvFbvLm49HhYXo3FGp+vy4DOWMSFVUuH9NI/fnngQy+uI+/tATY0y7AZv5NgAA4Ni12P+cbe+LPeVmKTLe6aoQpDITonTTjMG6acbg5rD9n9X7tXJPud5YvV9vrN6v6PAQnTsyVZf6adgGEJgI2AAAwDZopjRgurRrkbTksS6NYgM9pWXY3nuwVm+tKdR/1uzXqj3l+s9qO3hHh9vXdl86Ll05I1JZIA2AYwjYAADgmJw7pX9+nlFsuFK/xGjNmTlYc2YO1p6yWr211h7NXrW3QvNW7dO8VfsU4wvbl4zLUM6IPoRtAL2KgA0AAI4ZNKPFKPZfpVk/cboioF1ZSdG6eeYQ3TxziPaU1erNNfv15ho7bL++ap9ebxG2Lx2foVnDCdsAeh4BGwAAtNZqFPubjGLD9bKSovXNWUP0zVl22H7DF7ZXtwnb549O06XjMjSTsA2ghxCwAQBAa4xiw49lJUXrlllDdMusIdpdWqs3fdPI1xRU6LWV+/Tayn3yRoTq/FGpuoSwDaCbEbABAMDxGMVGAOif3Dpsv7Fmv95Ys09rCyr16sp9etUXti8YbV+zPWNYCmEbwGkhYAMAgOMNmiENnCHtXMgoNgJC/+RofStniL6VM0S7SmvssL16v9btq9QrKwr0yooCxUaENk8jnzE8RRGhhG0AXUPABgAA7Zt1hx2wFz8qnXWzFJXgdEVAtxiQHKNv5wzVt3OGamdJTfM1223D9gWj7QXSpg8jbAPoHAI2AABoX9tR7Jw7nK4I6HYDU2L0ndyh+k7uUO0oqdGbvpHt9fsr9fKKAr28okCxkb6wPY6wDeDECNgAAKBjOXdK/1goffpH+1psRrERwAa1E7b/s3q/Nuyv1MufFejlz+ywfeHodF06Pl3Th/ZReKjH6bIBuAgBGwAAdGzgdEaxEZRahu3txdXNYXtjYZVe+myvXvpsr+IiQ3XhmHRdPrGvpg1JUYjHOF02AIcRsAEAwIkxio0gN7iPV7eeO0y3njtM24qr9ebq/XpjjR22X1y+Vy8u36vU2AhdNqGvrpiUqTF942QMYRsIRgRsAABwYq1Gsf9iB24gSA3p49V3zxum7543TFsPVGveqn16dWWBdpXW6vFFO/T4oh0alurVFZMydfnEvuqXGO10yQB6EReNAACAk8u5y75f/CfpULmjpQBuMTTVq+9fMFx5P8rRy9+ephumDlBidJi2HKjWg+9s0vT7P9I1f12s55buVsWheqfLBdALGMEGAAAnN/AcadBMaccCRrGBNowxOqN/os7on6j//vxoLdhcrFdWFOi99UVauqNMS3eU6eevrdP4FKPDfQqVM6IPK5EDAYqADQAAOmfWnXbAXvwnacotXIsNtCMsxKPzRqXpvFFpqqqr19trC/XqygJ9sq1U+UVS/r+WKz4qTJeOz9CVkzI1uX+iPCyOBgQMAjYAAOiclqPYn/5Zyr3L6YoAV4uNDNN/ZWfpv7KzVFhRp9+9vFCrKyO1YX+lnlmyW88s2a1+iVG6YmKmrpiUqaGpXqdLBnCauAYbAAB0XtO12J/+mWuxgS5Ij4/U5waF6a3vzdDbt8/QLbOGKCM+UnsPHtKjH23V+b+dr9l/WKS/L9qhA1V1TpcL4BQRsAEAQOcNmCYNmiUdrrBDNoAuG5kepzs/N1If33Gunp1ztq7NzlJsRKjWFFToV/9Zr7Pv/UA3PLFUr6zYq5rDR50uF0AXMEUcAAB0Tc6d0o75dsA++1tOVwP4LY/HaOqQZE0dkqxfXj5GH248oFdWFChv0wEt2FysBZuLFRW2VheNSdMVkzI1fWiKQkMYHwPcjIANAAC6pmkUuylkm6lOVwT4vciwEF0yLkOXjMvQwZojemPNfr26okD5uw7q1ZX79OrKfUrxhmv2hL66clKmxmXGyxgWRwPchoANAAC6rnkU+08KzR7ndDVAQEmMCdeXzx6gL589QLtLa/XaygK9sqJA20tq9OTHO/Xkxzs1pE+MrpyUqcsnZiorKdrpkgH4ELABAEDXtRjF7rf3dUmfd7oiICD1T47Wd88bplvPHao1BRV6ZUWB5q3ap23FNXro3c166N3NOnNgoq6YlKlLx2UoITrc6ZKBoEbABgAApybnLl/AnifVVUiR8U5XBAQsY4zG90vQ+H4J+uklo7Rwa4leXVGgd9YVatnOg1q286DueX2dckek6spJmcodmarIsBCnywaCjrEsy+kaAoYxZrak2ZmZmXPmzp3rdDkdqq6ultfLPotuQ1/ch564E31xlzOW/0hxVVu0avwvdDDpDKfLQQt8V9ynJ3py6Kilz4qOavG+Bq0rbVDT3+yjQqUz00M1rW+ohid65OF67Q7xXXEff+hJbm7ucsuystseJ2D3gOzsbCs/P9/pMjqUl5ennJwcp8tAG/TFfeiJO9EXl3nzx9LSx6TzfylNv93patAC3xX36emeHKis0+ur9unVlQVaW1DZfLxvfKQun5SpqyZlalhabI+9v7/iu+I+/tATY0y7AZsp4gAA4NSljbHvi9Y5WwcApcZF6qYZg3XTjMHaUlSlV1cW6NUV+1RQfkh/ztumP+dt05kDE3X9lAG6eGw6U8iBHkDABgAApy7Nt4J40Vpn6wDQyrC0WP34opH64QUjlL/roF5ZsVfzVu1vvl47cV6Y/is7S188q78GpcQ4XS4QMAjYAADg1KWOkiUjU7JZOnpYCo1wuiIALXg8RmcNStJZg5L0s0tH67WV+/T0kl1at69Sjy3YrscWbNf0oSm6fkp/nT86TWEhHqdLBvwaARsAAJy68GgdispQ9KF9UvEmKWO80xUB6EBMRKi+NKW/vnhWllbtrdAzS3bp9VX7tGhriRZtLVGf2Ahdd2aWrjurvzITopwuF/BL/BMVAAA4LdXeQfYPTBMH/IIxRhOzEvTA1RO05O7zdc/s0RqW6lVx1WH94cOtmnH/h/rGP5bpw41FamhkQWSgKxjBBgAAp6XaO1CpxR9LhQRswN/ER4Xpq+cM0o3TBmrZzoN6eskuvbWmUB9sPKAPNh5QZkKUvnhWlq7JzlJqXKTT5QKuR8AGAACnpSaGEWzA3xlz7Frtn3/+sF5cvlfPLN2tXaW1eujdzXr4/S26cEyarp8yQFMHJ8vjYV9toD0EbAAAcFqqvQPtH4rWSpYlGf7iDfizZG+EvjlriObMGKyPt5Xo6U93670NRXpzTaHeXFOoQSkx+tJZ/fWFyf2UFBPudLmAqxCwAQDAaTkckSJFxku1pVJVoRSX4XRJALqBx2M0Y1gfzRjWR0WVdXp+2R49u3S3dpTU6P/e3KAH392kS8dl6Pop/TV5QKIM/7gGsMgZAAA4TcawHzYQ4NLiInXbecO08Ce5evyGbOWO6KP6hka9sqJAV/9lsS5+eKGeWrxTlXX1TpcKOIqADQAATl/aGPuegA0EtNAQj84fnaYnv3aWFvw4V9/JHaIUb7g2FVXp56+t05T/+0B3vLhaq/eWO10q4AimiAMAgNOXPta+ZyVxIGhkJUXrxxeN1PfOG6731hfp6SW79Mm2Uj2fv0fP5+/RuMx4XT+lvy6b2FfR4cQOBAf+SwcAAKcvzRewGcEGgk54qEeXjs/QpeMztK24Ws8u2a1/L9+rNQUVuvPlNfq/NzboyjMydf2UARqRHut0uUCPImADAIDTlzpKMh6pZItUXyeFsV8uEIyG9PHqZ58frR9dNEJvrtmvp5fs1vJdB/XU4l16avEuZQ9I1PVn99fnxmYoMizE6XKBbkfABgAApy8sSkoeKpVsloo3Sn0nOl0RAAdFhoXoqjP66aoz+mnD/ko9s2S3XllRoPxdB5W/66B+OW+9/mtyP31pygANSolxulyg27DIGQAA6B5MEwfQjlEZcfrVFWO15O7zdN9V4zSmb5zKa+v1t4U7lPtQnq5//FO9uWa/6hsanS4VOG2MYAMAgO6RPlZa9zILnQFoV0xEqL54Vn9dd2aWVu+t0NNLdun1Vfv08dZSfby1VH1iI3RtdpauOytL/RKjnS4XOCUEbAAA0D0YwQbQCcYYTchK0ISsBP300tF65bO9enrJbm05UK1HP9qqP+Zt1Xkj03TTjEGaMihJxhinSwY6jYANAAC6R8uAbVkSfykGcBLxUWH66jmDdOO0gVq286CeXrJLb60p1PsbivT+hiKNy4zXTTMG6ZJxGQoL4epWuB//lQIAgO4R11eKSpQOHZQq9zldDQA/YozRWYOS9PvrJumTu87V984bpqSYcK0pqND3nlupmQ98pL/O36aKQ/VOlwqcEAEbAAB0D2NajGKvc7YWAH4rxRuh718wXJ/cea7uu2qchvSJ0f6KOt331kZNu+8D/XLeOu0pq3W6TKBdBGwAANB9mgP2GmfrAOD3IsNC9MWz+uu978/Sk189U+cMTVbNkQY9+fFOzXrwI3376eVavuug02UCrXANNgAA6D7pvoDNSuIAuonHY5Q7MlW5I1O1bl+F/r5oh+at2qc31xTqzTWFmtQ/QXNmDNaFo9MUynXacBj/BQIAgO7DFHEAPWhM33j99pqJWnTHufp2zhDFR4Vpxe5yffvpz5TzUJ6eWLRD1YePOl0mghgBGwAAdJ8+IyUTIpVukeoPOV0NgACVFhepn1w8UovvOlf/c/kYDUiO1t6Dh/Q//1mvqfd9oPve3KB95fxvEHofARsAAHSfsEgpZZhkNUoHNjhdDYAAFx0eqhumDtSHP8zRX78yWWcNTFJV3VH9dcF2zXjgI9327Aqt2VvhdJkIIgRsAADQvVruhw0AvSDEY3TRmHS9cMtUvfadczR7Ql9J0uur9mn2o4t0zV8X6731RWpstByuFIGOgA0AALpX2hj7nuuwAThgQlaC/vDFSZr/4xzNmTFIsRGhWrqjTHOeytd5v52vf326S4eONDhdJgIUARsAAHSv9HH2PSuJA3BQv8Ro/fTS0frkrnP1s0tHKTMhSjtKavTfr67V1F9/oIfe2aQDlXVOl4kAwzZdAACge7XcC9uyJGOcrQdAUIuNDNNNMwbrq9MG6u11hfrbwh1atadcj360VY8t2K6z0j1KG1GpURlxTpeKAMAINgAA6F6x6VJUklRXIVUWOF0NAEiSQkM8+vz4vnr129P04i1TdfGYdNU3NmpRwVF97vcL9eXHlyhv0wFZFtdp49Qxgg0AALqXMVL6WGnHAnuaeHw/pysCgGbGGGUPTFL2wCTtKq3Rr55fpE8KLS3aWqJFW0s0LNWrm2YM0uUTMxUZFuJ0ufAzjGADAIDul+a7DrtojbN1AMAJDEiO0ZdHR2jxnefpjotHKi0uQlsOVOuOl9Zo+v0f6vfvb1Fp9WGny4QfIWADAIDul950HTYriQNwv/joMH0rZ4gW/uRc/e7aCRqdEaeS6iP63fubNe3XH+qul9do64Eqp8uEH2CKOAAA6H5NW3WxkjgAPxIe6tGVk/rpiomZWry9VH9fuEMfbDygZ5fu1rNLdyt3RB/NmTFYU4cky7CAI9pBwAYAAN2vz0jJEyqVbZOO1Erh0U5XBACdZozRtCEpmjYkRVsPVOuJj3fopeV79dGmYn20qVijM+J004xB+vz4vgoPZVIwjuG/hnYYYyKMMX8zxmw3xlQZYzYbY77rdF0AAPiN0AgpZbhkNUoHNjhdDQCcsqGpXt175Tgtvus8/eCC4Urxhmv9/kr94IVVmvHAh3pswTbVHD7qdJlwCQJ2+0IlFUq6UFK8pGsk/cwYc42jVQEA4E+apokXMU0cgP9LignXbecN06I7ztUDXxiv4WleFVUe1r1vbtQ5vgXRKmrrnS4TDiNgt8OyrBrLsv7bsqytlmU1Wpa1UtLrkqY7XBoAAP4jrWmhMwI2gMARGRaia87M0ju3z9STXz1Tkwckqry2Xr97f7POuf9D3f/2RpWw8njQcm3ANsbcZYz5t2+atmWM2XmCcz3GmO8bYzYaY+qMMXuMMb8xxsR0Uy1hkmZIWt0drwcAQFBoWkmchc4ABCBjjHJHpurFW6bq2Tln65yhyao+fFR/ztum6fd/qHteX6f9FYecLhO9zLUBW9K9ks6VtE3SwZOc+ztJv5W0XtJ3Jf1b0m2S5hljWn1GY8xzvsDe0S2nndd/VFKVpKdO6xMBABBM0lps1WVZztYCAD3EGKOpQ5L19E1n65VvT9P5o1JVV9+of3yyUzMf+Eh3vbxau0prnC4TvcTNq4gPsSxruyQZY9ZK8rZ3kjFmjOxQ/bJlWV9ocXyHpEckXSfpmRZPmSPp1hO8b0Wb1/+tpKmSzrUs68gpfA4AAIKTN02KTpFqS6SKPVJCf6crAoAeNal/oh6/8Uxt2F+pP360VW+s2a9nl+7R88v26LIJffXt3KEanhbrdJnoQa4dwW4K153wRUlG0sNtjv9NUq2kL7d53SrLskpOcGtemcAY87CkCySdZ1lWyal+FgAAgpIxTBMHEJRGZcTp0S+dofd/MEtXT+4njzF6deU+Xfi7Bfrmv/K1Zm/FyV8Efsm1AbsLzpTUKGlpy4OWZdVJWul7vMuMMY9IOl/2yHXxadYIAEBwYqEzAEFsSB+vHvqvCcr7cY6+cvYAhYd69M66Is1+dJFufGKplu0sc7pEdLNACNh9JZVYltXeUn0FklKMMeFdeUFjzADZ086HStphjKn23d46/XIBAAgiBGwAUL/EaP3qirFa9JNc3TxzsKLDQzR/c7H+6y+Ldc1fF2vB5mJZrFUREIw/NLLpGmzLsga289g2SWGWZR13YZcx5ilJX5GUaFlWeQ/XeLOkmyUpLS1t8nPPPdeTb3daqqur5fW2e0k7HERf3IeeuBN9cZ8T9SSmeofOzL9dtVF9tXTKn3u5suDGd8V96Ik7OdGX6iOW3t1Vr/d31av2qH1sULxHsweHaWJqiDzG9Go9buMP35Xc3NzllmVltz0eCAF7jaRUy7LS2nnsBUn/JSmiNxcoy87OtvLz83vr7bosLy9POTk5TpeBNuiL+9ATd6Iv7nPCnhw9It2bITU2SHcXSOHdsoMmOoHvivvQE3dysi9VdfX616e79PeFO1RaY8eVEWmx+nbuEH1+fF+FeIIzaPvDd8UY027ADoQp4vtkTwOPaOexTNnTx1n9GwAAJ4SGSykjJFnSgQ1OVwMArhIbGaZv5wzVojvO1c8/P1rpcZHaVFSl7z23Uuf9Jk/PL9utI0cbnS4TXdBrAdsYk2KMeagHXnqZ7M9xVpv3i5Q0UZJ7h5IBAAgGzSuJr3G2DgBwqajwEH19+iDN/0mO7rtqnPonRWtnaa3ueGmNch78SP/8ZKfq6hucLhOd0OMB2xiTaIy5V9J2Sd/vgbd4XpIl6fY2x+dIipb0dA+8JwAA6CwWOgOATokIDdEXz+qvD384Sw9fO1HDUr3aV1GnX7y+TtPv/0h/nb9N1YePOl0mTiD0VJ5kjBkqaYykEEnbLcta2c45Xkk/kh18Y2XvVd3eSt8dvcdXJA3w/dpHUrgx5me+33dZlvUvSbIsa40x5o+SbjXGvCzpTUmjJN0mab6kZ7r6+QAAQDdqGsEuWudsHQDgJ0JDPLpiUqYum9BX764v1KMfbdXagkrd99ZG/Slvm752zkB9ddpAJUR3abMk9IIuBWxjTJakuZKmtzm+UtL1lmVt9P1+g6QHZAdjI6lW0t8kdWWK+DckzWpz7Fe++/mS/tXi+O2SdspexftSSSWS/iDp55ZlcdECAABOSmsRsC1LCvLVcQGgszweo4vHZuiiMenK21ysP364Vfm7Durh97fobwu268tTB+im6YPVJ7a95ajghE6vIm6MiZa0WtIg2aG5rX2SRkq6T9K3fedUSPqjpIctyyrpjoLdzBgzW9LszMzMOXPnznW6nA75w7L3wYi+uA89cSf64j6d6cm0j29UeH25Pp3ymOqijtv4Az2A74r70BN38qe+WJalTQcbNW/bEa0rtccRwzzSrH6h+tygMCVHBcIa1v7Rk9PepssY831Jv5F9vfOTkt6WHaIvlvQ13/H3JF0o6ZCkByX91rKsyu74AP6EbbpwKuiL+9ATd6Iv7tOpnvzrSmnbh9K1T0ujPt8rdQU7vivuQ0/cyV/7snJPuR79cKve31AkSQoLMbpqUj99K2eIBqb495aI/tCT7tim6zLZIfr/LMu6ybKsFy3L+rdlWd+Q9H+yw/YFkvZIOsOyrHuCMVwDAIB2pI2x77kOGwC6xcSsBD1+Y7bevn2GZk/oq4ZGS8/n79G5v8nT955boU2FVU6XGJS6ErBH++4faeex37f4+SeWZW069ZIAAEDASRtn3xexVRcAdKeR6XH6wxcn6YMf5uia7H7yGKPXVu7TRQ8v0M1P5Wv13nKnSwwqXQnYiZKq2ruW2nes6Z9IPuiOwgAAQABp3gubrboAoCcMSonRA1dPUN6Pc3TD1AEKD/Xo3fVFuuzRj/WVvy/R0h1lTpcYFLoSsENlrwbekVqpOWwDAAAckzxM8oRJB3dIh6udrgYAAla/xGj9z+VjteiOXH1z5mDFhIdo4ZYSXfPXxbr+8U+1fBdBuycFxjJzAADA3ULDpT4j7Z8PrHe2FgAIAqmxkbrrklH6+M5z9b3zhik2MlQfby3VF/68WDc+sVQr95Q7XWJA6tI+2JJCfHtht7dNV4jUvFd2hxtcWpa1u4vvCQAAAkH6WPsa7MI1UtZZTlcDAEEhITpc379guL5+ziA9vmi7nli0Q/M3F2v+5mKdPypVt58/XGMz450uM2B0ZZuuRtmriJ8Oy7KsroZ6v8E+2Dgd9MV96Ik70Rf36WxP+u15TUO3PaGCvp/TluG39EJlwY3vivvQE3cKtr5UHbH01o56vb+7Xkca7GOT00J0xdBwZcW6Y4KzP/SkO/bBbuyGOizLskK64XVcjX2wcSroi/vQE3eiL+7T6Z5s+0j61xVS1hTpG+/2dFlBj++K+9ATdwrWvpRUH9Zf8rbpX5/u0uGjdtS7dHyGvn/+MA1NjXW0Nn/oSUf7YHdlNPlr3VgPAAAINulNW3WtkxobJY87RkoAIBileCP0s8+P1s0zB+tPedv0zJLdemP1fr21Zr8un5ip284bpkEpMU6X6Xc6HbAty/pnTxYCAAACXEyK5E2Xqgul8p1S0mCnKwKAoJcaF6l7Lhujb84arEc/3KoX8vfolRUFen3VPl01yQ7aWUnRTpfpN/inYwAA0HvSxtj3ReucrQMA0EpGfJT+78px+vCHObo2O0uS9O/le5X7UJ7uenmNCsoPOVyhfyBgAwCA3pM+1r4vXOtsHQCAdmUlRev+q8frwx/O0lVnZKrRsvTs0t3KfTBPP39trQor6pwu0dU6HbCNMTONMWef6hsZY35gjPn5qT4fAAAEgLSm67AJ2ADgZgOSY/TbaybqvR/M0mUT+qq+sVFPLd6lmQ9+pP+Zt14Hqgja7enKCHaepJfae8AYs9QYs+0kz/+xpF904f0AAECgaZ4iTsAGAH8wpI9Xj3xxkt65faYuGZeuI0cb9cTHOzTzgY9035sbVFZzxOkSXaWrU8RNB8ezJA08vVIAAEDASxkmhYRLB3dKdZVOVwMA6KThabH60/WT9eZtM3TB6DTV1Tfqrwu2a8b9H+rBdzaqvJagLXV9H+xCy7L6tvPYfkmpJ9rjujPn+DtjzGxJszMzM+fMnTvX6XI65A8btwcj+uI+9MSd6Iv7dLUnk/O/r9jq7fps0q9VGT+qBysLbnxX3IeeuBN9OTU7Kxr08tZ6rS5ukCRFhUoXDgjThQPDFBPW0bhs5/hDT3Jzc097H2ychGVZ8yTNy87OnuPmjdH9YeP2YERf3IeeuBN9cZ8u96R8qrRyu87ICJXO6sLz0CV8V9yHnrgTfTl1X5X02e6D+t17m7VwS4le21avjwos3TxzsL56ziB5I04tbvpzT1hFHAAA9C626gKAgHFG/0T96xtT9MI3p+rswUmqrDuqh97drBn3f6i/zN+m2iNHnS6xVxGwAQBA70rzbdXFQmcAEDDOGpSk526eqmdumqLJAxJ1sLZev35ro2Y+8JEeX7hddfUNTpfYKwjYAACgdzUH7PVSY6OztQAAutW0oSl68Zap+ufXz9KErASVVB/R/76xQTMf+Ej//GSnDh8N7KBNwAYAAL0rJlmKzZDqa6SDO5yuBgDQzYwxmjW8j1799jQ98dVsjc2M04Gqw/rF6+uU82Cenl6yS0eOBuY/sHb1qvNwY8wMHb9dV7gkdfBYq3MAAACUNlaq2m9PE08e4nQ1AIAeYIzRuSPTlDsiVe+uL9Lv3tusjYVV+ukra/XnvG267dxhuuqMTIWGBM64b1cDdqKkvBM8fqLHjKTO7QkGAAACW/pYaet7UuFaafTlTlcDAOhBxhhdNCZdF4xK01trC/W79zdr64Fq/eSl1fpT3lbddt4wXT4xUyGe09veyw26+k8F5jRuAAAAtubrsFlJHACChcdjdOn4DL1z+0w9fO1EDUqJ0c7SWv3ghVW68Hfz9fqqfWps9O8x2a6MYOf2WBUAACC4NAfsNc7WAQDodSEeoysmZerz4zP0yooCPfLhFm0rrtFtz67Qox9u0QUZRzWz0ZLHD0e0jWX5978QuIkxZrak2ZmZmXPmzp3rdDkdqq6ultfrdboMtEFf3IeeuBN9cZ9T6YlpbNCMhdfKY9Vr4fRn1BAa00PVBS++K+5DT9yJvjjvaKOlRQVH9fq2epXVWTKydN+MaKXHuPfa7Nzc3OWWZWW3PU7A7gHZ2dlWfn6+02V0KC8vTzk5OU6XgTboi/vQE3eiL+5zyj356yxp/0rpa29JA6Z1d1lBj++K+9ATd6Iv7nH4aINeyN+rRSs36q+3XOR0OSdkjGk3YHd1kbPOvJFX0iWSBkmqkZRvWdan3f0+AADAz6WNtQN20ToCNgBAEaEh+srZA5RV579bOHY5YBtjzpZ0uSSvpE2SnrIsq9L32GWS/iEpvs1zFku62rKswtMtGAAABIh033XYhVyHDQAIDF0K2MaYn0v6RZvDdxtjpkhKl/S8pIh2njpV0lvGmDMtyzp6SpUCAIDA0rzQ2Vpn6wAAoJt0+qpxY8xUSffo2LZbZb77NEn/J+lnssP1K7KniI+RdJGkp3znjZf0le4rHQAA+LW0Mfb9gQ1SY4OztQAA0A26sizbN333+ZKyLMvqIynL9/tsSRdLetyyrC9YlvW2ZVkbLMt6z7Ksr0q6V3bI/q/uKx0AAPi16CQpLlOqr5XK/Pd6OwAAmnQlYE+TZEn6nmVZBZLku79d9jXXoZIe7eC5f/DdTzi1MgEAQEBiP2wAQADpSsDOlNQgqe2K4Et8xyVpY3tPtCyrSFKFpOSuFggAAAJY80JnXIcNAPB/XQnYUZLKrDYbZ1uW1Sj7emxZlnXkBM+vkxTW5QoBAEDgaroOu2ids3UAANANuhKwJamxi8cBAAA6ljbOvmclcQBAAOhqwAYAAOg+yUOk0EipYo906KDT1QAAcFpMmxnfHZ9oTKOkQsuy+rbz2H5JqZZlhZzg+Sc9x98ZY2ZLmp2ZmTln7ty5TpfToerqanm9XqfLQBv0xX3oiTvRF/c53Z6csfyHiqvaqhUT71VFwphurCy48V1xH3riTvTFffyhJ7m5ucsty8puezy0i6+TZozpcKPKEz0me5uuzqV5P2VZ1jxJ87Kzs+fk5OQ4XU6H8vLy5Ob6ghV9cR964k70xX1OuycVU6UVWzUpI1Sachqvg1b4rrgPPXEn+uI+/tyTrgZs0yNVAACA4JXuuw67kK26AAD+rSsB+5c9VgUAAAherCQOAAgQnQ7YlmURsAEAQPdrCtgHNkiNDZInYJdrAQAEOFYRBwAAzopKlOKzpKOHpNJtTlcDAMApI2ADAADnpY2174u4DhsA4L8I2AAAwHlchw0ACAAEbAAA4Lx03wh24Vpn6wAA4DQQsAEAgPPSfFt1FRGwAQD+i4ANAACclzRICo2SKguk2jKnqwEA4JQQsAEAgPM8IVLaaPtnrsMGAPgpAjYAAHCH5pXEmSYOAPBPBGwAAOAOBGwAgJ8jYAMAAHdgJXEAgJ8zlmU5XUPAMMbMljQ7MzNzzty5c50up0PV1dXyer1Ol4E26Iv70BN3oi/u0109CTlaoxmLvqRGE6aFM56X5QnphuqCF98V96En7kRf3McfepKbm7vcsqzstscJ2D0gOzvbys/Pd7qMDuXl5SknJ8fpMtAGfXEfeuJO9MV9urUnD4+TyndL314ipY7sntcMUnxX3IeeuBN9cR9/6Ikxpt2AzRRxAADgHlyHDQDwYwRsAADgHgRsAIAfI2ADAAD3YKEzAIAfI2ADAAD3aB7BXudsHQAAnAICNgAAcI/EQVJYjFS1T6otc7oaAAC6hIANAADcw+OR0kbbPxeucbYWAAC6iIANAADcJW2Mfc9CZwAAP0PABgAA7sJ12AAAP0XABgAA7pI+zr5nijgAwM8QsAEAgLuk+q7BLt4oNdQ7WwsAAF1AwAYAAO4SGSclDJAajkilW52uBgCATiNgAwAA92meJs5CZwAA/0HABgAA7tO80BnXYQMA/AcBGwAAuE/zVl2sJA4A8B8EbAAA4D7pvhFspogDAPwIARsAALhPwkAp3CtVF0o1JU5XAwBApxCwAQCA+3g8x6aJsx82AMBPELABAIA7cR02AMDPGMuynK4hYBhjZkuanZmZOWfu3LlOl9Oh6upqeb1ep8tAG/TFfeiJO9EX9+mpnvQteEvDt/xFhWm52jjq9m5//UDHd8V96Ik70Rf38Yee5ObmLrcsK7vtcQJ2D8jOzrby8/OdLqNDeXl5ysnJcboMtEFf3IeeuBN9cZ8e68mepdLfL5DSxknfWtT9rx/g+K64Dz1xJ/riPv7QE2NMuwGbKeIAAMCdUkfZ98UbpYZ6Z2sBAKATCNgAAMCdImKlxEFSY71UstnpagAAOCkCNgAAcC/2wwYA+BECNgAAcK80X8AuImADANyPgA0AANyLgA0A8CMEbAAA4F5MEQcA+BECNgAAcK+EAVJ4rFRzQKo+4HQ1AACcEAEbAAC4lzFS2hj7Z6aJAwBcjoANAADcjWniAAA/QcAGAADuxkJnAAA/QcAGAADu1hyw1zlbBwAAJ0HABgAA7pY2WpKRijdJR484XQ0AAB0iYAMAAHcLj5GSBkuN9VLJJqerAQCgQwRsAADgfk0ribPQGQDAxQjYAADA/dLH2fcsdAYAcDECNgAAcD9WEgcA+AECNgAAcL+We2FblrO1AADQAQI2AABwv/gsKSJeqi2Rqg84XQ0AAO0iYAMAAPcz5thCZ0VrnK0FAIAOELABAIB/aDlNHAAAFyJgAwAA/9A8gr3O2ToAAOgAARsAAPiHNLbqAgC4GwEbAAD4h9RRkoxUslk6etjpagAAOA4BGwAA+IfwaCl5iNR4VCre6HQ1AAAch4ANAAD8R5pvoTOuwwYAuJCxLMvpGgKGMWa2pNmZmZlz5s6d63Q5HaqurpbX63W6DLRBX9yHnrgTfXGf3uxJ/10vaPCOp7Wn32XaNvQbvfKe/orvivvQE3eiL+7jDz3Jzc1dbllWdtvjBOwekJ2dbeXn5ztdRofy8vKUk5PjdBlog764Dz1xJ/riPr3ak01vS89eKw2aKd04r3fe00/xXXEfeuJO9MV9/KEnxph2AzZTxAEAgP9o2qqrcK3EIAEAwGUI2AAAwH/E95Mi46VDZVJVodPVAADQCgEbAAD4D2PYDxsA4FoEbAAA4F+apokTsAEALkPABgAA/iXdt1VXIQEbAOAuBGwAAOBfmvfCJmADANyFgA0AAPxL6ijJeKSSLVJ9ndPVAADQjIANAAD8S1iUlDxUshqk4o1OVwMAQDMCNgAA8D9MEwcAuBABGwAA+B8WOgMAuBABGwAA+B9GsAEALkTABgAA/qdlwLYsZ2sBAMCHgA0AAPxPXF8pKlE6dFCq3Od0NQAASCJgAwAAf2QM08QBAK5DwAYAAP6JgA0AcBkCNgAA8E+sJA4AcBkCNgAA8E+MYAMAXIaADQAA/FOfkZIJkUq3SvWHnK4GAAACNgAA8FNhkVLKMMlqlA5scLoaAAAI2AAAwI8xTRwA4CIEbAAA4L/Sxtj3ReucrQMAABGwAQCAP0sfZ9+zkjgAwAUI2AAAwH81TxFfI1mWs7UAAIIeARsAAPiv2HQpKkmqq5Aq9jpdDQAgyBGwAQCA/zJGSm8axeY6bACAswjYAADAv6X5rsMuWuNsHQCAoEfABgAA/q1pBJuFzgAADiNgAwAA/8ZWXQAAlyBgAwAA/9ZnpOQJlcq2SUdqna4GABDECNgAAMC/hUZIKcMlq1E6sMHpagAAQYyADQAA/F/zNHGuwwYAOIeADQAA/F9a01ZdBGwAgHMI2AAAwP+xkjgAwAUI2AAAwP81j2CvkyzL2VoAAEGLgA0AAPyfN02KTpEOV0gVe5yuBgAQpAjYAADA/xnDNHEAgOMI2B0wxvzJGLPHGFNpjCkwxjxsjAl3ui4AANABFjoDADiMgN2xRyWNtCwrTtIE3+1uZ0sCAAAdImADABwW6nQBbmVZ1voWvxpJjZKGOVQOAAA4GaaIAwAc5toRbGPMXcaYfxtjthtjLGPMzhOc6zHGfN8Ys9EYU+eb2v0bY0zMadZwpzGmWtIB2SPYD5/O6wEAgB6UMkLyhEpl26UjNU5XAwAIQq4N2JLulXSupG2SDp7k3N9J+q2k9ZK+K+nfkm6TNM8Y0+ozGmOe8wX2jm45TedalvVry7K8kkZL+ouk/d302QAAQHcLDbdDtiypaP1JTwcAoLu5eYr4EMuytkuSMWatJG97JxljxsgO1S9blvWFFsd3SHpE0nWSnmnxlDmSbj3B+1a0PWBZ1gZjzCpJ/5KU28XPAQAAekv6WOnAOvs67Kwzna4GABBkXDuC3RSuO+GLsq+RfrjN8b9JqpX05TavW2VZVskJbvUdvE+YpOFd+AgAAKC3sdAZAMBBrg3YXXCm7AXIlrY8aFlWnaSVvse7xBgTb4z5qjEmwdjGS/qZpHe6oV4AANBTWOgMAOAgY1mW0zWcVNMUccuyBrbz2BpJqZZlpbXz2AuS/ktShGVZR7rwfnGSXpZ0hqRw2YucvSzpF5ZltbtqijHmZkk3S1JaWtrk5557rrNv1+uqq6vl9bY74x4Ooi/uQ0/cib64j5t6EnakXOd8cqOOhkRp0fRnJWOcLskxbuoLbPTEneiL+/hDT3Jzc5dblpXd9ribr8HurGhJhzt4rK7FOZ0O2JZlVUo6vytFWJb1mKTHJCk7O9vKycnpytN7VV5entxcX7CiL+5DT9yJvriP63qyKlWhNQeUM3GQlDjQ6Woc47q+gJ64FH1xH3/uSSBMEa+VFNHBY5EtzgEAAMGAaeIAAIcEQsDeJynFGNNeyM6UVNKV6eEAAMDPpY2x74vWOVsHACDoBELAXib7c5zV8qAxJlLSREn5DtQEAACckjbOvi9a42wdAICgEwgB+3lJlqTb2xyfI/va66d7uyAAAOAgpogDABzi2kXOjDFfkTTA92sfSeHGmJ/5ft9lWda/JMmyrDXGmD9KutUY87KkNyWNknSbpPmSnundygEAgKOSh0meMOngDulwlRQR63RFAIAg4dptuowxeZJmdfDwfMuyclqcGyJ7BPtmSQMllcge2f65ZVnVPVlnS8aY2ZJmZ2Zmzpk7d25vvW2X+cOy98GIvrgPPXEn+uI+buxJ9rLb5a3Zoc8m3a/K+JFOl+MIN/Yl2NETd6Iv7uMPPelomy7XBmx/lp2dbeXnu/fSb39e9j6Q0Rf3oSfuRF/cx5U9eeUWadWz0qW/lc78htPVOMKVfQly9MSd6Iv7+ENPjDHtBuxAuAYbAACgtTTfddhFXIcNAOg9BGwAABB42KoLAOAAAjYAAAg86U1bda2TGhudrQUAEDQI2AAAIPDEpEjedOlItVS+0+lqAABBgoANAAACE9PEAQC9jIANAAACU7pvobNCFjoDAPQOtunqRuyDjdNBX9yHnrgTfXEft/YktWi+Rm/4rYpTpmjd2LudLqfXubUvwYyeuBN9cR9/6An7YPci9sHGqaAv7kNP3Im+uI9re1K0XvrzVClhgHT7aqer6XWu7UsQoyfuRF/cxx96wj7YAAAguKQMk0LCpfJdUl2l09UAAIIAARsAAASmkDCpz0j75wPrna0FABAUCNgAACBwNe2HXbjG2ToAAEGBgA0AAAIXW3UBAHoRARsAAASuNN9WXUVs1QUA6HkEbAAAELiaA/Z6qbHR2VoAAAGPgA0AAAJXTLIUmyHV10gHdzhdDQAgwLEPdjcyxsyWNDszM3PO3LlznS6nQ/6wcXswoi/uQ0/cib64j9t7Mm71/yi5bLnWjrlDJX2mOV1Or3F7X4IRPXEn+uI+/tCT3NzcdvfBDnWimEBlWdY8SfOys7PnuHljdH/YuD0Y0Rf3oSfuRF/cx/U9OZonLVquscmW5OY6u5nr+xKE6Ik70Rf38eeeMEUcAAAENhY6AwD0EgI2AAAIbARsAEAvIWADAIDAljxUComQyndLdRVOVwMACGAEbAAAENhCQqXUUfbPReucrQUAENAI2AAAIPA1TxMnYAMAeg4BGwAABL50X8AuXONsHQCAgEbABgAAgY+FzgAAvYCADQAAAl/aGPu+aL3U2OBsLQCAgEXABgAAgS86SYrLlI4eksp2OF0NACBAGcuynK4hYBhjZkuanZmZOWfu3LlOl9Oh6upqeb1ep8tAG/TFfeiJO9EX9/GXnoxb/Ssll+Vr3eifqDj1HKfL6XH+0pdgQk/cib64jz/0JDc3d7llWdltj4c6UUygsixrnqR52dnZc3Jycpwup0N5eXlyc33Bir64Dz1xJ/riPn7Tk4YF0sJ8jUlulPyh3tPkN30JIvTEneiL+/hzT5giDgAAgkPzddhs1QUA6BkEbAAAEBzSxtn3rCQOAOghBGwAABAckodIoZFSxR7p0EGnqwEABCACNgAACA6eECl1lP0z08QBAD2AgA0AAIJH2lj7noANAOgBBGwAABA80n3XYReucbYOAEBAImADAIDg0bySOAudAQC6HwEbAAAEj6aAfWCD1NjgbC0AgIBDwAYAAMEjKlGKz5KO1kml25yuBgAQYAjYAAAguDQvdMZ12ACA7mUsy3K6hoBhjJktaXZmZuacuXPnOl1Oh6qrq+X1ep0uA23QF/ehJ+5EX9zH33oyaPtcDdj9b+3qf7V2DP6K0+X0GH/rSzCgJ+5EX9zHH3qSm5u73LKs7LbHQ50oJlBZljVP0rzs7Ow5OTk5TpfToby8PLm5vmBFX9yHnrgTfXEfv+tJn4PS7n9rQESVBvhT3V3kd30JAvTEneiL+/hzT5giDgAAgkuab6suVhIHAHQzAjYAAAguSYOk0CipskCqLXO6GgBAACFgAwCA4OIJkdJG2z8XrXO2FgBAQCFgAwCA4NO8kjjTxAEA3YeADQAAgk9TwC4kYAMAug8BGwAABJ90RrABAN2PgA0AAIJP2hj7/sAGqeGos7UAAAIGARsAAASfyHgpob/UcFgq3ep0NQCAAEHABgAAwYmFzgAA3YyADQAAghMBGwDQzQjYAAAgOKWzkjgAoHsRsAEAQHBiBBsA0M0I2AAAIDglDpLCYqSq/VJNqdPVAAACgLEsy+kaAoYxZrak2ZmZmXPmzp3rdDkdqq6ultfrdboMtEFf3IeeuBN9cR9/7smkz36i+MpNWjnhVypPHO90Od3Kn/sSqOiJO9EX9/GHnuTm5i63LCu77fFQJ4oJVJZlzZM0Lzs7e05OTo7T5XQoLy9Pbq4vWNEX96En7kRf3Meve1I1VVq+SRPTQ6SpOU5X0638ui8Bip64E31xH3/uCVPEAQBA8Gq+Dnuds3UAAAICARsAAASv9HH2feEaZ+sAAAQEAjYAAAheqaPt++KNUkO9s7UAAPweARsAAASvyDgpYYDUcEQq2eJ0NQAAP0fABgAAwa1pmjjXYQMAThMBGwAABLfmhc64DhsAcHoI2AAAILiljbHvC9c6WwcAwO8RsAEAQHBLZ6suAED3IGADAIDgljBQCvdK1YVSTYnT1QAA/BgBGwAABDeP59h2XeyHDQA4DQRsAAAApokDALoBARsAAKB5JXEWOgMAnDoCNgAAQNNe2KwkDgA4DQRsAACA1FH2ffFG6egRZ2sBAPgtAjYAAEBErJQ4SGqsl0q3OF0NAMBPEbABAACkYwudMU0cAHCKCNgAAABSi4XO2KoLAHBqjGVZTtcQMIwxsyXNzszMnDN37lyny+lQdXW1vF6v02WgDfriPvTEneiL+wRKT1KKP9XYdfepLHGiVk/4pdPlnLZA6UsgoSfuRF/cxx96kpubu9yyrOy2xwnYPSA7O9vKz893uowO5eXlKScnx+ky0AZ9cR964k70xX0CpicHd0q/nyDFpEo/9v/rsAOmLwGEnrgTfXEff+iJMabdgM0UcQAAAElKGCCFx0o1B6TqA05XAwDwQwRsAAAASTJGShtj/1zEQmcAgK4jYAMAADRhJXEAwGkgYAMAADRpXkmcgA0A6DoCNgAAQJM0RrABAKeOgA0AANAkbbQkI5Vsko4ecboaAICfIWADAAA0CY+RkgZLjUftkA0AQBcQsAEAAFpqWkmcaeIAgC4iYAMAALSUPs6+Z6EzAEAXEbABAABaYiVxAMApImADAAC01HIvbMtythYAgF8hYAMAALQUnyVFxEu1JVJ1kdPVAAD8CAEbAACgJWOOLXTGNHEAQBcQsAEAANpqOU0cAIBOImADAAC0xQg2AOAUELABAADaSmvaqmuds3UAAPwKARsAAKCt1FGSjFSyWTp62OlqAAB+goANAADQVni0lDxEajwqFW90uhoAgJ8gYAMAALQnzbfQGdPEAQCdRMAGAABoDyuJAwC6iIANAADQnuaFztY4WwcAwG8QsAEAANrTtFVX4VrJspytBQDgFwjYAAAA7YnvJ0XGS4fKpKpCp6sBAPgBAjYAAEB7jGkxTZzrsAEAJ0fABgAA6EjzNHGuwwYAnJyxuKao2xhjZkuanZmZOWfu3LlOl9Oh6upqeb1ep8vA/7d352GTleWdx7+/sC8jazdrRlSISzBA0mJkJoZ2iCYagVzExCguUdRcURIdZ3Q0MaKScSZEQiLZIIkg0VEzBqPBuCQGJGREGkQctwAOrYLsgmytLPf8cU7RRXVVvd1vn/c91V3fz3XVdepUPeecu+vpt6ruerYR1svssU5mk/Uye7bmOtn3O5/mCV8/k5tWPp2vPun1fYezSbbmetlSWSezyXqZPVtCnaxevfryqlo1+rgJ9hJYtWpVrVmzpu8wJrrwwgs5+uij+w5DI6yX2WOdzCbrZfZs1XVy/RVw9mpY8QR49aV9R7NJtup62UJZJ7PJepk9W0KdJBmbYNtFXJIkaZKVT4T8ENx6Ndy/ru9oJEkzzgRbkiRpku12gr0OhnoQbvla39FIkmacCbYkSdI0+xzabJ1JXJK0ABNsSZKkafZtE+wbTbAlSdOZYEuSJE1jC7YkaSOZYEuSJE0znGC7+ookaQoTbEmSpGketT/stAfc91246zt9RyNJmmEm2JIkSdMksMvK5v667/UbiyRppplgS5IkSZLUARNsSZIkSZI6YIItSZIkSVIHTLAlSZIkSeqACbYkSZIkSR0wwZYkSZIkqQMm2JIkSZIkdcAEW5IkSZKkDphgS5IkSZLUARNsSZIkSZI6YIItSZIkSVIHTLAlSZI2WvUdgCRphplgS5IkLSTpOwJJ0hbABFuSJEmSpA6YYEuSJEmS1AETbEmSJEmSOmCCLUmSJElSB0ywJUmSJEnqgAm2JEmSJEkdMMGWJEmSJKkDJtiSJEmSJHXABFuSJEmSpA6YYEuSJEmS1AETbEmSpI1V1XcEkqQZlvKDonNJbgHW9h3HFHsDt/YdhDZgvcwe62Q2WS+zxzqZTdbL7LFOZpP1Mnu2hDp5dFWtGH3QBHsOJVlTVav6jkOPZL3MHutkNlkvs8c6mU3Wy+yxTmaT9TJ7tuQ6sYu4JEmSJEkdMMGWJEmSJKkDJtjz6ay+A9BY1svssU5mk/Uye6yT2WS9zB7rZDZZL7Nni60Tx2BLkiRJktQBW7AlSZIkSeqACbYkSZIkSR0wwZ5zSVYmeU+Sq5LcnmRdkmuS/GWSg/uObx4lOSDJm5JclOQ7Se5J8uUkpyXZq+/45lmSVyV5X5KvJXkwiWNslkGSH0ryuvZ1X5fkW0nelWSXvmObZ+371N8k+UaSSnJd3zHNsyQ/kuTtST6X5JYkdyW5Mslv+bfSnySPbz83vprkziT3tu9lpyfZr+/41Eiy89B72Zl9xzOv2td/3O3uvmPbFNv2HYB6twfwI8CngLXAfcAhwMuA5yX5yar6So/xzaPnAqcAFwCnAXcBRwKvBZ6f5ClVdWNv0c23NwF7AV8AdgEO7DecufEHwG8A5wPvAp7Y7h+R5JiqeqjP4ObYfwduB64Adu83FNF8br8a+CjwPuB+YDVwKvBL7ef5fT3GN68OBPajef/6NvAA8GTglTSf6YdX1c09xqfG24EVfQchAC5mwwnO7u8jkMVykjONleQpwOeBP62qX+87nnmS5EeB20aT6CQnAWcD76qq/9JLcHMuyUHAN6vqoSR/DzynqtJzWFu19u/hS8D5VXXC0OMnA38EvLCq3t9XfPMsyWOr6hvt/f8L7FpVB/Ub1fxKsgq4uqruHHn8VOC3gJOrypa5GZHkecCHgDdW1e/1Hc88S/LjNN9530DzI+4fV9Vr+o1qPrU9A8+tqpf2HcvmsIu4JlnbbvfoNYo5VFVfntBC/cF2e+hyxqP1quo6W0uX3a8AAc4Yefxs4F7gxOUOSI1Bcq3ZUFVrRpPrlp8ds8nvWTMgyTY0nyefAP6253DUSrJ9kl37jmOxTLAFQJLtkuydZL8kPwX8r/apj/cZlx5h0B35pl6jkJbXU4CHaFoXHlZV64Ar2+clTeZnxwxIsmP7PevAJM8E/rx9yu9Z/Xod8ATAFuvZ8Ys0P6DfleTmJO9OslvfQW0Kx2Br4FnAx4b2bwJeX1Xn9RSPNvS2dntur1FIy2t/4Naq+v6Y564HjkqyfVX9YJnjkmZe2zr3Fppxvw6l6NdJwLuH9q8DTqyqi/sJR0keQ/Pd6u1VdV07DEz9+jzwN8A1wKOAZ9P8+PHTSY6qqi1isjMT7K1Ekt1pJsHaWH9UVbcP7X8O+BlgJ+BJwPOBPZJsW1UPdBXnPOmgTobP9XrgecBZVfWZzY9ufnVZL1oWOwPjkmuAdUNlTLClDZ0BPA14c1V9vedY5t1HgK8BuwJHAMcCe/cZkPgz4BvA6X0HokZVPXXkofcmuQr4XeA32+3Mc5KzrUT7q9v/24RDDqmqa6acb3/gKuDDVfWqzQxvLnVVJ+3kZmfRdCP7haraomZSnDUd1ouTnC2DJF8CVlbVPmOe+xDND0872ILdLyc5mz1J3gH8Ns0Ps36Oz5gkPwZcBpxSVe/sO555k+RE4L3A06vqX9rHDqL5fuAkZzMkyXbA3cDlVXVU3/FsDMdgbyXayZeyCbeJyXV7vhuAfwRenmSH5flXbF26qJMkL6NJrj8FnGByvfm6/lvRkrsB2HvC+9ABNN3HTa6lIUlOoUmu3wP8Wr/RaJyquopmyUdXallm7efJ6TQNFzcmOTjJwcCj2yK7tY/t3leMWq/97nsDW1CPDxNsTbMTsA3NGAgtsza5/guaHzqOnzAGVdraXUbzWXXk8INJdgQOB9b0EJM0s9rk+q0083WcVHZVnGU7AXv2HcQc2olmzevnAFcP3S5snz+x3T+pj+D0SO3n/YFsQRM1OgZ7ziXZp6o2+A+b5EnAfwKurapblj+y+ZbkpTTLRnwGOK6dMVmaRx8E3kwzbn54MqBX0Iy9fl8PMUkzKcnv0CTX5wEvc1nB/iXZd9zSm0lW0yydduGyB6V7aIYXjVoB/AnNkl1/STNUUsskyV5VdduYp95Bk7N+bMxzM8kx2HMuyRk0k5tdQDOjZWje8F8EbAccW1Wf7Cu+eZTkWOB84HvAG4D7RorcXVUfWe64BEmeCxzW7p4IPJ5mhl6AO6rqzF4C28oleTfNLKLn03TpeyLwG8AlwDNMIvqR5EWs71J5MrA98K52f62rUCyvJK8GzgS+SfO+NPp3cVNVfXrZA5tzSc4H9qP5wXwtsCPwEzSTyd4LHF1VV/YWoB7mGOx+JfkD4CeBf6Z5H9uVZhbx1cClwOqqGv1OPJNMsOdckmNoxmetAlbSdAm/HrgI+P2q+nKP4c2loe59k6x1IqF+JDkHeMmEp62XJdIuNfRa4JXAQcCtNC3bv7OlLNmxNUpyIfDTE56+qKqOXr5otMD7E1gnvUjyS8CLaX6cXQEUTaL9aeC0qvpmj+FpiAl2v5IcRzMnwaHAXsCDNF31PwScviX15jTBliRJkiSpA05yJkmSJElSB0ywJUmSJEnqgAm2JEmSJEkdMMGWJEmSJKkDJtiSJEmSJHXABFuSJEmSpA6YYEuSJEmS1AETbEmSJEmSOmCCLUmSepfk1CSV5I2LPP6c9vhzOg5tcP5PtOd/xlKcX5K0dTDBliRpiSRZ2yZlG3P7646ueWV7vmMWiOeCBc7zprbcbV3EtcC1DgT+M3ALcOYSnP+UKa/7vUmuTnJukqOmnOaUdvv7Sfz+JEkayw8ISZKWQJK9gX/f7n4XuGmB20UdXPPRwGHAHcCFU+IBOCbJblNO9xPt9gubG9dG+F1gJ+D3quqeJb7W8Gt+C7A9cDDwYuCSJKeMO6iqPgd8EjgCOHGJY5QkbaFMsCVJWho/PnT/qKrad4Hb2R1c87h2+/GqemBKPLfTJJbPnXKuQYJ9eQdxTZTkAOCFwA+Av1rKawGMvOYrgR2A/8j6f+dbp7Rk/1m7fcNSxylJ2jKZYEuStDQGCepdwNeX6ZqDBPvvxjw3iOdW1ieyJ4w7SZI9gYPa3Su6Cm6CVwDb0PwocPsSX2sDVfVgVV0CHD/08HETin+c5seJH03yH5Y6NknSlscEW5KkpTFoMb68qmqpL5ZkD+DpNC3B/zAlniuAD7f3n5Vk1yllYQlbsJMEeHm7+/6NKP/CJJckuSvJnUkuTfLK9jybpaq+DQzGm497TaiqH7D+tXvl5l5TkrT1McGWJGlpDFqM1yzT9Z4DbAt8pqrumhLPFcClwPU0456fPabsIMG+E7i24ziHHQoc2N6/eFKhNP4K+GvgKGAX4EFgFfDnbERyvpC2q/pe7e60HgefbbfP2txrSpK2PibYkiR1rG1Nfky7e9kyXXZi9/CReK5oW9TPb/d/ccy5Hp7gbIlb35/ebr9VVTdOKXcy8Kvt/TOBlVW1J7Anzezev8zkbt1TJdkmydNY/3rcDLx3yiGXttt9kjxhMdeUJG29TLAlSerecBfrP01y45TbhZt7sSQ7AD8LFPDRBeIZjKkedHV+dpKdJpRf6vHXT223X5xUIMmOwFvb3fOq6uSquhWgqu6sqrcB/xPYfWMuOPLa3wx8H/hX4PHA+4Ajq+qOScdX1dXA3e3u0zbmmpKk+WGCLUlS94YT2j2Bfabcvjl6cJLrFlqnesQzaMYNr6mqG6bEc0dVDbp8X0yzTNUuDHV3bpfuely7u9QJ9v7t9pYpZZ5J8xoCvH1Cmf8BrNvIaw6/9itoJlgD2BnYrX18IYOx2vtPLSVJmjsm2JIkdW+Q0P5TVWWB24uHD0yyO/Bo4KpNuN7x7Xbc7OHD8Ty8pnVVPThUfrib+BHAYNKwR0xw1nanfmOSa5J8P8m1SX47ybabEOuwFe122uzhq9rtt6rqmnEFqurO0VgnGX39acahHwGcC/w88Nkkxy9wmkG8K6aWkiTNHRNsSZK6NxjDvClJ8sCPtdsvbUzhdgbtwXrWH1kgntEW6UE38Z9Psv1I2buBfxsp/26a1uJ/AV4N/DPwDpqJxhZjx3b7/SllVrbb6xc417cXE0BVrauqK6vqJJpx2DsA5yR51JTD7mu3O04pI0maQybYkiR1qE3MDm53F5NgH7aJxx4J7AdcW1VfXiCe0QT7n2hmCt8N+Jn2sUFr95VV9dDQeZ4M/BrwJ1X10qr6izYpPR14WZJVbLpBV+s9FnHsUji73e7G+NnVBwZd1m+bUkaSNIdMsCVJ6tZwF+vFJtg/AO5Jcl6S25LcnOScdnz0qImzh4+J5xEJdlXdD3ys3T2h3U5q7X5+e54zRh4f7P/KhOtPMxh7veeUMje32wMWONdCz2+MtUP3HzOx1Pp4p40dlyTNIRNsSZK6NWgBfgDYoEV5IxxG06r8mXb7JuBTwEuA94wpf3y7XWj89bgu37C+m/hx7XJeh7T7o2OaVwG3tbNoP6yqvgXcwPrEfFN8pd0+dkqZwTriP5zkceMKtK30i7n+qAOH7t8z4Vr/Dti73f1qB9eUJG1FTLAlSerWINH7t6qaNrZ4A0m2AQ6lSeBeVVWvqaqzqupE4H8Dv5Bk36HyhwBPBG4FLlkgnkd0+R7ySZpkck/gdaz/bjDagr0/k8dBX8/iWpA/224Pa5caG+fTwHfb+2+ZUOYNNJOVba4XDN1fM6HMKprX6AEmv+aSpDllgi1JUrcGLcaL6R7+eJqJs86vqk+NPPeJdjvc2jvoHv737azg0+IZu+RWVd0H/EO7+7p2ex8bts7uzOTJyNaxuAT3EppEdXvg8CnxvaPdfUmSM5LsBU3LdZK3AG8G7ljE9WnPs2+SU2l6CQB8Dvg/E4oP1u6+oqrunlBGkjSnFrushiRJGpFkF5okGeDYJDcucMhFVfXLQ/uDCc7+cEzZarfDXZenjr8eiWfamtYfplmqa9d2/4tjEvZ7aWbYHmdH1s+svdGq6nvtet/HAccCl04o+oc0Y8lfBPwmcHKSO4FH0axj/QGa5P8lE45/2Jg62ZFmUrOBLwEnVFUx3rHt9v0LXUuSNH9swZYkqTuHs/6zdWdgnwVu3xg5fpBgj+ue/FSaJPfrAElWAEfRJLajrd3j4pmWYF/AI1unx5W9gcndwA9g4WW0Jhks8fWCdsmxDVTVQ+164S+maV2+j6aR4Aqamc1fMO64CUbrYGfgRpqu8q8AVlXVDeMOTPJY4Gnt9d+7CdeUJM0JW7AlSepIVV3C+hm7F2OQYD9irHTbJfoFwEeral378HNpkud/rKp7NyeeqrqLhdd0vhx4ZpJDhic6S/LDNOOzP7DQdSb4JHAt8Djgp1g/LntcnOcB5014+qXtbdxxpwCnLDK+YSe22w9U1XenlpQkzSVbsCVJmh2DBHv1yOPvBLZj/VhkWN89/CNLHNPAB2m6qb925PHB/qIS7HbitcHkZf9tMedYDm13+5NpWvrf1nM4kqQZZQu2JEkzoO3yvR/weeC8JKcBtwPPA44BXlFVXxk65BLgC0xenqtTVfXFJGcBv94mmxfTdJd+OXBuVV22Gaf/AE2i/nNJjqyqz292wN17Dc3s7qdV1dqFCkuS5lMmz+EhSZKWS5JjaJak+jngyTQJ5540XbNPrapPTD56eSTZFvivwEk0a0bfQLM29zur6v7NPPfhNGt6X1ZVF2xepN1L8hpgL+CMqrqz73gkSbPJBFuSJEmSpA44BluSJEmSpA6YYEuSJEmS1AETbEmSJEmSOmCCLUmSJElSB0ywJUmSJEnqgAm2JEmSJEkdMMGWJEmSJKkDJtiSJEmSJHXABFuSJEmSpA78f4cQFtbjK77/AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1152x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "CODERATE = 0.5\n",
    "BATCH_SIZE = 2000\n",
    "\n",
    "model_coded_awgn = CodedSystemAWGN(num_bits_per_symbol=NUM_BITS_PER_SYMBOL,\n",
    "                                   n=2048,\n",
    "                                   coderate=CODERATE)\n",
    "ber_plots.simulate(model_coded_awgn,\n",
    "                   ebno_dbs=np.linspace(EBN0_DB_MIN, EBN0_DB_MAX, 15),\n",
    "                   batch_size=BATCH_SIZE,\n",
    "                   num_target_block_errors=500,\n",
    "                   legend=\"Coded\",\n",
    "                   soft_estimates=False,\n",
    "                   max_mc_iter=15,\n",
    "                   show_fig=True,\n",
    "                   forward_keyboard_interrupt=False);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df537cad",
   "metadata": {},
   "source": [
    "As can be seen, the `BerPlot` class uses multiple stopping conditions and stops the simulation after no error occured at a specifc SNR point."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e7902e8",
   "metadata": {},
   "source": [
    "**Task**: Replace the coding scheme by a Polar encoder/decoder or a convolutional code with Viterbi decoding."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c02ebb11",
   "metadata": {},
   "source": [
    "## Eager vs Graph Mode\n",
    "\n",
    "So far, we have executed the example in *eager* mode. \n",
    "This allows to run TensorFlow ops as if it was written NumPy and simplifies development and debugging.\n",
    "\n",
    "However, to unleash Sionna's full performance, we need to activate *graph* mode which can be enabled with the function decorator *@tf.function()*.\n",
    "\n",
    "We refer to [TensorFlow Functions](https://www.tensorflow.org/guide/function) for further details.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "b3bc7a86",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function() # enables graph-mode of the following function\n",
    "def run_graph(batch_size, ebno_db):\n",
    "    # all code inside this function will be executed in graph mode, also calls of other functions\n",
    "    print(f\"Tracing run_graph for values batch_size={batch_size} and ebno_db={ebno_db}.\") # print whenever this function is traced\n",
    "    return model_coded_awgn(batch_size, ebno_db)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "64a7c186",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tracing run_graph for values batch_size=10 and ebno_db=1.5.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(<tf.Tensor: shape=(10, 1024), dtype=float32, numpy=\n",
       " array([[1., 1., 0., ..., 0., 1., 0.],\n",
       "        [1., 1., 1., ..., 0., 1., 1.],\n",
       "        [1., 1., 0., ..., 0., 1., 1.],\n",
       "        ...,\n",
       "        [0., 1., 0., ..., 0., 0., 0.],\n",
       "        [0., 1., 0., ..., 0., 1., 0.],\n",
       "        [0., 1., 1., ..., 0., 1., 1.]], dtype=float32)>,\n",
       " <tf.Tensor: shape=(10, 1024), dtype=float32, numpy=\n",
       " array([[1., 1., 0., ..., 0., 1., 0.],\n",
       "        [1., 1., 1., ..., 0., 1., 1.],\n",
       "        [1., 1., 0., ..., 0., 1., 1.],\n",
       "        ...,\n",
       "        [0., 1., 0., ..., 0., 0., 0.],\n",
       "        [0., 1., 0., ..., 0., 1., 0.],\n",
       "        [0., 1., 1., ..., 0., 1., 1.]], dtype=float32)>)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "batch_size = 10 # try also different batch sizes\n",
    "ebno_db = 1.5\n",
    "\n",
    "# run twice - how does the output change?\n",
    "run_graph(batch_size, ebno_db)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e76ffd1d",
   "metadata": {},
   "source": [
    "In graph mode, Python code (i.e., *non-TensorFlow code*) is only executed whenever the function is *traced*.\n",
    "This happens whenever the input signature changes.\n",
    "\n",
    "As can be seen above, the print statement was executed, i.e., the graph was traced again.\n",
    "\n",
    "To avoid this re-tracing for different inputs, we now input tensors.\n",
    "You can see that the function is now traced once for input tensors of same dtype.\n",
    "\n",
    "See [TensorFlow Rules of Tracing](https://www.tensorflow.org/guide/function#rules_of_tracing) for details.\n",
    "\n",
    "**Task:** change the code above such that tensors are used as input and execute the code with different input values. Understand when re-tracing happens.\n",
    "\n",
    "*Remark*: if the input to a function is a tensor its signature must change and not *just* its value. For example the input could have a different size or datatype.\n",
    "For efficient code execution, we usually want to avoid re-tracing of the code if not required."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "b5f3704f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "run_graph(batch_size=10, ebno_db=1.5)\n",
      "  Returns:\n",
      "    (<1>, <2>)\n",
      "      <1>: float32 Tensor, shape=(10, 1024)\n",
      "      <2>: float32 Tensor, shape=(10, 1024)\n"
     ]
    }
   ],
   "source": [
    "# You can print the cached signatures with\n",
    "print(run_graph.pretty_printed_concrete_signatures())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b3664c47",
   "metadata": {},
   "source": [
    "We now compare the throughput of the different modes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "d3f22327",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Throughput in Eager mode: 1.212 Mbit/s\n",
      "Tracing run_graph for values batch_size=Tensor(\"batch_size:0\", shape=(), dtype=int32) and ebno_db=Tensor(\"ebno_db:0\", shape=(), dtype=float32).\n",
      "Throughput in graph mode: 8.623 Mbit/s\n"
     ]
    }
   ],
   "source": [
    "repetitions = 4 # average over multiple runs\n",
    "batch_size = BATCH_SIZE # try also different batch sizes\n",
    "ebno_db = 1.5\n",
    "\n",
    "# --- eager mode ---\n",
    "t_start = time.perf_counter()\n",
    "for _ in range(repetitions):\n",
    "    bits, bits_hat = model_coded_awgn(tf.constant(batch_size, tf.int32),\n",
    "                                tf.constant(ebno_db, tf. float32))\n",
    "t_stop = time.perf_counter()\n",
    "# throughput in bit/s\n",
    "throughput_eager = np.size(bits.numpy())*repetitions / (t_stop - t_start) / 1e6\n",
    "\n",
    "print(f\"Throughput in Eager mode: {throughput_eager :.3f} Mbit/s\")\n",
    "# --- graph mode ---\n",
    "# run once to trace graph (ignored for throughput)\n",
    "run_graph(tf.constant(batch_size, tf.int32),\n",
    "          tf.constant(ebno_db, tf. float32))\n",
    "\n",
    "t_start = time.perf_counter()\n",
    "for _ in range(repetitions):\n",
    "    bits, bits_hat = run_graph(tf.constant(batch_size, tf.int32),\n",
    "                                tf.constant(ebno_db, tf. float32))\n",
    "t_stop = time.perf_counter()\n",
    "# throughput in bit/s\n",
    "throughput_graph = np.size(bits.numpy())*repetitions / (t_stop - t_start) / 1e6\n",
    "\n",
    "print(f\"Throughput in graph mode: {throughput_graph :.3f} Mbit/s\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "749d2355",
   "metadata": {},
   "source": [
    "Let's run the same simulation as above in graph mode."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "1de9bbf6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EbNo [dB] |        BER |       BLER |  bit errors |    num bits | block errors |  num blocks | runtime [s] |    status\n",
      "---------------------------------------------------------------------------------------------------------------------------------------\n",
      "     -3.0 | 2.7922e-01 | 1.0000e+00 |      571845 |     2048000 |         2000 |        2000 |         0.2 |reached target block errors\n",
      "   -2.273 | 2.5948e-01 | 1.0000e+00 |      531422 |     2048000 |         2000 |        2000 |         0.2 |reached target block errors\n",
      "   -1.545 | 2.3550e-01 | 1.0000e+00 |      482301 |     2048000 |         2000 |        2000 |         0.2 |reached target block errors\n",
      "   -0.818 | 2.0768e-01 | 1.0000e+00 |      425335 |     2048000 |         2000 |        2000 |         0.2 |reached target block errors\n",
      "   -0.091 | 1.6918e-01 | 1.0000e+00 |      346477 |     2048000 |         2000 |        2000 |         0.2 |reached target block errors\n",
      "    0.636 | 7.6115e-02 | 9.1650e-01 |      155883 |     2048000 |         1833 |        2000 |         0.2 |reached target block errors\n",
      "    1.364 | 1.7544e-03 | 7.2125e-02 |       14372 |     8192000 |          577 |        8000 |         1.0 |reached target block errors\n",
      "    2.091 | 7.8125e-08 | 2.0000e-05 |          16 |   204800000 |            4 |      200000 |        24.3 |reached max iter       \n",
      "    2.818 | 0.0000e+00 | 0.0000e+00 |           0 |   204800000 |            0 |      200000 |        24.4 |reached max iter       \n",
      "\n",
      "Simulation stopped as no error occurred @ EbNo = 2.8 dB.\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1152x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ber_plots.simulate(run_graph,\n",
    "                   ebno_dbs=np.linspace(EBN0_DB_MIN, EBN0_DB_MAX, 12),\n",
    "                   batch_size=BATCH_SIZE,\n",
    "                   num_target_block_errors=500,\n",
    "                   legend=\"Coded (Graph mode)\",\n",
    "                   soft_estimates=True,\n",
    "                   max_mc_iter=100,\n",
    "                   show_fig=True,\n",
    "                   forward_keyboard_interrupt=False);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "57cdb153",
   "metadata": {},
   "source": [
    "**Task:** TensorFlow allows to *compile* graphs with [XLA](https://www.tensorflow.org/xla). Try to further accelerate the code with XLA (`@tf.function(jit_compile=True)`).\n",
    "\n",
    "*Remark*: XLA is still an experimental feature and not all TensorFlow (and, thus, Sionna) functions support XLA.\n",
    "\n",
    "**Task 2:** Check the GPU load with `!nvidia-smi`. Find the best tradeoff between batch-size and throughput for your specific GPU architecture."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d4765a0",
   "metadata": {},
   "source": [
    "## Exercise\n",
    "\n",
    "Simulate the coded bit error rate (BER) for a Polar coded and 64-QAM modulation.\n",
    "Assume a codeword length of n = 200 and coderate = 0.5.\n",
    "\n",
    "**Hint**: For Polar codes, successive cancellation list decoding (SCL) gives the best BER performance.\n",
    "However, successive cancellation (SC) decoding (without a list) is less complex.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "cf855e19",
   "metadata": {},
   "outputs": [],
   "source": [
    "n = 200\n",
    "coderate = 0.5\n",
    "\n",
    "# *You can implement your code here*\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}