Performance Evaluation of Channel Decoding With Deep Neural Networks
This work addresses the need for high data rate and low latency in 5G communication, but it is incremental as it compares existing neural network types without introducing a new method.
The paper tackled the problem of channel decoding for 5G by evaluating three deep neural network decoders (MLP, CNN, RNN) with the same parameter magnitude, finding that RNN achieves the best decoding performance but with the highest computational overhead and identifying saturation lengths due to restricted learning abilities.
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.