Advanced~neural network receiver design~to combat multiple channel~impairments

Abstract

In communication systems, the channel noise is usually assumed to be white and Gaussian distributed. Therefore, an optimum receiver structure designed for the additive white Gaussian noise (AWGN) channel is employed in applications. However, in wireless communication systems, noise is often caused by strong interferences. Moreover, there are other effects such as phase offset that degrade the performance of the receiver. Designing the optimum receiver for different channel models is difficult and not reasonable because channel model and channel statistics are not known at the receiver. In this paper, we propose a neural network-based approach to demodulate the transmitted signal over unknown channels. Naturally, the collection of the training data, design and training of the neural network, and finally reconfiguration of the system according to the designed neural network are implemented on software-defined digital signal processing facilities. In particular, we show that the proposed receiver is capable of jointly canceling the strong interferences and phase offset. Simulation results in various signal environments are presented to illustrate the performance of the proposed system. It is shown that the proposed approach has the same performance as the correlation demodulator structure for AWGN channels, while it has a clear advantage for unknown channel models. Moreover, it is shown that the neural network-based receiver may be used for channel estimation and equalization over Rayleigh channels. Numerical results indicate that the performance of the proposed receiver is very close to the Rayleigh theoretical bound.

DeepDeMod: BPSK Demodulation Using Deep Learning Over Software-Defined Radio

“…Nakayama and Imai, [9] proposed an amplitude shift keying demodulator based on a neural network to combine the wideband noise rejection, pulse waveform shaping, and decoding into a single neural network. Multi-layer perceptron (MLP) based demodulator was proposed in [10], [11]. Moreover, He et al [10] used multiple MLPs to construct a demodulator named MaxMLP classifier, which automatically detected different modulated signals (BPSK, QPSK, and GMSK) without using complex signal processing algorithms.…”

“…Önder et.al. [11] proposed a NN-based receiver to decode multiple phase-shift keying (MPSK) signal using a three-layer MLP structure. Fan and Wu [12] proposed a deep belief network based demodulator for BPSK and QPSK.…”

“…2, output of ADC samples are in the form of a row vector. From known bit boundaries each set of oversampled received data is adjusted as shown in (11) and combined together to form a 2-D array as input of the DNN named as T rain DN N shown in (12). Label set for training is named as Label DN N shown in (13).…”

End to End Deep Neural Network Frequency Demodulation of Speech Signals

“…Except for the traditional methods for radio transmission decoding and for minimum mean square error (MMSE) based speech enhancement techniques, several, though not many, neural networks based methods have been proposed for each of the two problems separately. For example the radio transmission decoding: [1], [14] and recently [2], and for channel noise estimation [13], however, these works deal with digital communication for which bit-streams are mapped to symbols, moreover, traditionally the symbols are precoded and scrambled before transmitted, therefore effectively the coded data stream is uncorrelated from time-sample to time-sample [21], and use of the prior speech data to overcome the noise in the transmission channel is not possible. The fact that the modulating input is proportional only to the instantaneous frequency of the received FM signal has driven the development of traditional FM demodulators to rely on very short time frame processing in order to extract the modulating signal, disregarding long range dependencies that are present in the transmitted voice.…”

FPGA Implementation of a BPSK 1D-CNN Demodulator

“…Recently, some scholars have attempted to introduce machine learning into demodulation technology. For the neural network demodulators mentioned in [21][22][23], the demodulation principle is to analyse the modulated data in every symbol period by neural network. The modulated data is divided into symbols according to the number of samplings.…”