Performance Evaluation of Channel Decoding with Deep Neural Networks

Abstract

With the demand of high data rate and low latency in fifth generation (5G), deep neural network decoder (NND) has become a promising candidate due to its capability of one-shot decoding and parallel computing. In this paper, three types of NND, i.e., multi-layer perceptron (MLP), convolution neural network (CNN) and recurrent neural network (RNN), are proposed with the same parameter magnitude. The performance of these deep neural networks are evaluated through extensive simulation. Numerical results show that RNN has the best decoding performance, yet at the price of the highest computational overhead. Moreover, we find there exists a saturation length for each type of neural network, which is caused by their restricted learning abilities.

Deep Learning-Based Decoding for Constrained Sequence Codes

“…We then investigate employing CNNs for this task. With ECC decoding, CNNs and MLP networks have similar number of weights in order to achieve similar performance [31]. In the following we outline our findings that are unique to CS decoding.…”

“…We use the notation h = [h 1 , h 2 , ..., h L ] to represent a network with L hidden layers, where h l denotes the number of neurons in the fully connected layer l, or the number of kernels in the convolutional layer l. In recent works that apply DNNs to decode ECCs, the training set explodes rapidly as the source word length grows. For example, with a rate 0.5 (n = 1024, k = 512) ECC, one epoch consists of 2 512 possibilities of codewords of length 1024, which results in very large complexity and makes it difficult to train and implement DNN-based decoding in practical systems [28], [29], [31], [32]. However, we note that in FL CS decoding, this problem does not exist since CS source words are typically considerably shorter, possibly only up to a few dozen symbols [1], [6]- [17].…”

“…Recently several works have reported the application of DNNs to the decoding of error control codes (ECCs) [28]- [33]. A DNN enables low-latency decoding since it enables one-shot decoding, where the DNN finds its estimate by passing each layer only once [28], [31], [32]. In addition, DNNs can efficiently execute in parallel and be implemented with low-precision data types on a graphical processing unit (GPU), field programmable gate array (FPGA), or application specific integrated circuit (ASIC) [28], [31]- [33], [35].…”

“…A DNN enables low-latency decoding since it enables one-shot decoding, where the DNN finds its estimate by passing each layer only once [28], [31], [32]. In addition, DNNs can efficiently execute in parallel and be implemented with low-precision data types on a graphical processing unit (GPU), field programmable gate array (FPGA), or application specific integrated circuit (ASIC) [28], [31]- [33], [35]. It has been shown that, with short-to-medium length (i.e., up to a few hundred bits) codewords, DNN-based decoding can achieve competitive bit error rate (BER) performance.…”

Learning to Flip Successive Cancellation Decoding of Polar Codes with LSTM Networks

“…However, the optimal flipping strategy (which bit to flip) is still an open problem due to the lack of a complete mathematical characterization. When an optimal solution is unavailable, deep learning algorithms are worth trying [9]- [12]. Recently, deep learning [13] has achieved tremendous progress in tasks such as Go [14], image classification [15], machine translation [16].…”