Finished initial (non-working) implementation of proximal decoder

sw/decoders/__init__.py

@@ -0,0 +1,2 @@
+"""This package contains a number of different decoder implementations for LDPC codes
+"""

sw/decoders/proximal.py

@@ -0,0 +1,86 @@
+import numpy as np
+from tqdm import tqdm
+
+
+class ProximalDecoder:
+    """Class implementing the Proximal Decoding algorithm. See "Proximal Decoding for LDPC Codes" by Tadashi
+    Wadayama, and Satoshi Takabe.
+    """
+    def __init__(self, H: np.array, K: int = 10, step_size: float = 0.1, gamma: float = 0.05):
+        """Construct a new ProximalDecoder Object.
+
+        :param H: Parity Check Matrix
+        :param K: Max number of iterations to perform when decoding
+        :param step_size: Step size for the gradient descent process
+        :param gamma: Positive constant. Arises in the approximation of the prior PDF
+        """
+        self._H = H
+        self._K = K
+        self._step_size = step_size
+        self._gamma = gamma
+
+    @staticmethod
+    def _L_awgn(s: np.array, y: np.array) -> np.array:
+        """Variation of the negative log-likelihood for the special case of AWGN noise. See 4.1, p. 4."""
+        return s - y
+
+    def _grad_h(self, x: np.array) -> np.array:
+        """Gradient of the code-constraint polynomial. See 2.3, p. 2."""
+        # Calculate first term
+
+        result = 4 * (x**2 - 1) * x
+
+        # Calculate second term
+
+        for k, x_k in enumerate(x):
+            # TODO: Perform this operation for each row simultaneously
+            B_k = np.argwhere(self._H[:, k] == 1)
+            B_k = B_k[:, 0]  # Get rid of one layer of arrays
+
+            # TODO: Perform the summation with np.sum()
+            sum_result = 0
+            for i in B_k:
+                # TODO: Perform this operation for each column simultaneously
+                A_i = np.argwhere(self._H[i] == 1)
+                A_i = A_i[:, 0]  # Get rid of one layer of arrays
+
+                prod = 1
+                for j in A_i:
+                    prod *= x[j]
+
+                sum_result += prod**2 - prod
+
+            term_2 = 2 / x_k * sum_result
+
+            result[k] += term_2
+
+        return np.array(result)
+
+    def _check_parity(self, y_hat: np.array) -> bool:
+        """Perform a parity check for a given codeword.
+
+        :param y_hat: codeword to be checked
+        :return: True if the parity check passes, i.e. the codeword is valid. False otherwise
+        """
+        syndrome = np.dot(self._H, y_hat) % 2
+        return not np.any(syndrome)
+
+    def decode(self, y: np.array) -> np.array:
+        """Decode a received signal. The algorithm is detailed in 3.2, p.3.
+
+        This function assumes an AWGN channel.
+
+        :param y: Vector of received values
+        :return: Most probably sent symbol
+        """
+        s = 0
+        x_hat = 0
+        for k in range(self._K):
+            r = s - self._step_size * self._L_awgn(s, y)
+            s = r - self._gamma * self._grad_h(r)
+            x_hat = (np.sign(s) == 1) * 1
+
+            if self._check_parity(x_hat):
+                break
+
+        return x_hat

sw/decoders/utility.py

@@ -0,0 +1,80 @@
+"""This file contains various utility functions that can be used in combination with the decoders.
+"""
+
+import numpy as np
+import typing
+from tqdm import tqdm
+
+
+def _get_noise_amp_from_SNR(SNR: float, signal_amp: float = 1) -> float:
+    """Calculate the amplitude of the noise from an SNR and the signal amplitude
+
+    :param SNR: Signal-to-Noise-Ratio in dB
+    :param signal_amp: Signal Amplitude (linear)
+    :return: Noise Amplitude (linear)
+    """
+    SNR_linear = 10 ** (SNR / 10)
+    noise_amp = (1 / np.sqrt(SNR_linear)) * signal_amp
+
+    return noise_amp
+
+
+def add_awgn(c: np.array, SNR: float) -> np.array:
+    """Add Additive White Gaussian Noise to a data vector. As this function adds random noise to the input,
+    the output changes, even if it is called multiple times with the same input.
+
+    :param c: Binary vector representing the data to be transmitted
+    :param SNR: Signal-to-Noise-Ratio in dB
+    :return: Data vector with added noise
+    """
+    noise_amp = _get_noise_amp_from_SNR(SNR, signal_amp=1)
+    y = c + np.random.normal(scale=noise_amp, size=c.size)
+    return y
+
+
+def count_bit_errors(d: np.array, d_hat: np.array) -> int:
+    """Count the number of wrong bits in a decoded codeword.
+
+    :param d: Originally sent data
+    :param d_hat: Received data
+    :return: Number of bit errors
+    """
+    return np.sum(d != d_hat)
+
+
+def test_decoder(decoder: typing.Any,
+                 c: np.array,
+                 SNRs: typing.Sequence[float] = np.linspace(1, 4, 7),
+                 N=10000) \
+        -> typing.Tuple[np.array, np.array]:
+    """Calculate the Bit Error Rate (BER) for a given decoder for a number of SNRs.
+
+    This function prints it's progress to stdout
+
+    :param decoder: Instance of the decoder to be tested
+    :param c: Codeword whose transmission is to be simulated
+    :param SNRs: List of SNRs for which the BER should be calculated
+    :param N: Number of iterations to perform for each SNR
+    :return: Tuple of numpy arrays of the form (SNRs, BERs)
+    """
+    BERs = []
+    for SNR in tqdm(SNRs, desc="Calculating Bit-Error-Rates",
+                    position=0,
+                    bar_format="{l_bar}{bar}| {n_fmt}/{total_fmt}"):
+
+        total_bit_errors = 0
+
+        for n in tqdm(range(N), desc=f"Simulating for SNR = {SNR} dB",
+                      position=1,
+                      leave=False,
+                      bar_format="{l_bar}{bar}| {n_fmt}/{total_fmt}"):
+
+            y = add_awgn(c, SNR)
+            y_hat = decoder.decode(y)
+
+            total_bit_errors += count_bit_errors(c, y_hat)
+
+        total_bits = c.size * N
+        BERs.append(total_bit_errors / total_bits)
+
+    return np.array(SNRs), np.array(BERs)