Performance comparison of blind and non-blind channel equalizers using artificial neural networks

Abstract

Abstract-In digital communication systems, multipath propagation induces Inter Symbol Interference (ISI).To reduce the effect of ISI different channel equalization algorithms are used. Complex equalization algorithms allow for achieving the best performance but they do not meet the requirements for implementation of real-time detection at low complexity, thus limiting their application. In this paper, we present different blind and non-blind equalization structures based on Artificial Neural Networks (ANNs) and, also, we analyze their complexity versus performance. The simulated network is based on multilayer feedforward perceptron ANN, which is trained by utilizing the error back-propagation algorithm. The weights of the network are updated in accordance with training of the network to improve the convergence speed. Simulation results demonstrate that the implementation of equalizers using ANN provides an upper hand over the performance and computational complexity with respect to conventional methods.

Blind Channel Equalization Using Variational Autoencoders

“…In our first set of experiments, we compared our model to the baseline algorithms at various noise levels, using the following non-minimum phase channels taken from [11], [24], [26], We generated 2000 QPSK random symbols as the training set. Then we applied convolution with the channel impulse response, and added white Gaussian noise at a signal to noise ratio (SNR) in the range 0dB -10dB.…”

“…In our first set of experiments, we compared our model to the baseline algorithms at various noise levels, using the following non-minimum phase channels taken from [11], [24], [26], present SER results for h 1 , h 2 and h 3 , respectively. As can be seen, the new VAEBCE significantly outperforms the baseline blind equalizers, and is quite close to the performance of the non-blind adaptive MMSE.…”

Performance Evaluation of Machine Learning-Based Channel Equalization Techniques: New Trends and Challenges

“…In the last few years, FLANN is very famous [52]. It is a single-layer NN that can form complex decision boundaries.…”

Developing an Artificial Neural Network Algorithm for Generalized Singular Value Decomposition-based Linear Discriminant Analysis

“…The use of ANN in developing the algorithm has the benefit of accuracy. Also, ANN is equipped with the uniqueness of concurrent processing, can learn and recall data relationships, and mapping of non-linear instances [34,35]. It is also used in several mathematical calculations [36].…”