Neural Network Detection of Data Sequences in Communication Systems

Abstract

We consider detection based on deep learning, and show it is possible to train detectors that perform well without any knowledge of the underlying channel models. Moreover, when the channel model is known, we demonstrate that it is possible to train detectors that do not require channel state information (CSI). In particular, a technique we call a sliding bidirectional recurrent neural network (SBRNN) is proposed for detection where, after training, the detector estimates the data in realtime as the signal stream arrives at the receiver. We evaluate this algorithm, as well as other neural network (NN) architectures, using the Poisson channel model, which is applicable to both optical and molecular communication systems. In addition, we also evaluate the performance of this detection method applied to data sent over a molecular communication platform, where the channel model is difficult to model analytically. We show that SBRNN is computationally efficient, and can perform detection under various channel conditions without knowing the underlying channel model. We also demonstrate that the bit error rate (BER) performance of the proposed SBRNN detector is better than that of a Viterbi detector with imperfect CSI as well as that of other NN detectors that have been previously proposed. Finally, we show that the SBRNN can perform well in rapidly changing channels, where the coherence time is on the order of a single symbol duration.

Deep Soft Interference Cancellation for MIMO Detection

“…To identify the advantages of DeepSIC over previously proposed ML-based detectors, we first recall that, as discussed in the introduction, these data-driven receivers can be divided into two main types: The first family of deep receivers implements symbol detection using a single conventional network, generally treated as a block box. While their architecture can account for some a-priori knowledge of the scenario, such as OFDM signaling [23], [25], channel memory [21], [22], and the presence of low resolution quantizers [9], [39], the design is typically not based on established detection algorithms. Compared to such deep receivers, DeepSIC, which learns to implement only the model-based computations of an established algorithm, has fewer hyperparameters, and can thus be trained using smaller training sets, allowing it to be quickly retrained in the presence of dynamic environment.…”

Deep Learning for Communication over Dispersive Nonlinear Channels: Performance and Comparison with Classical Digital Signal Processing

“…After training of the autoencoder, we employ it in the sliding window sequence estimation algorithm proposed in [5]. It is important to mention that for the training set a Mersenne twister was used as a random number generator.…”

“…, W´1 and ř W´1 q"0 a pqq " 1 are the weighting coefficients for the softmax probability output of the receiver BRNN. Equal weights a pqq " 1 W were previously assumed in both [5] and [12]. Note that the final W´1 blocks y T`1 , .…”

Deep Learning-Based Channel Estimation Algorithm Over Time Selective Fading Channels

“…Research on such channel using deep learning is somehow only a little. Sliding Bidirectional Recurrent Neural Network(SBRNN) has been put forward in [11] and works as a detector to learn rapid varying optical and molecular channel. A simple application of neural network to Rayleigh fading channel is given in [12].…”