Comparative Survey of Signal Processing and Artificial Intelligence Based Channel Equalization Techniques and Technologies

Abstract

This paper presents an inspection and review of wired and wireless channel equalization techniques and their existing hardware implementations in terms of features, similarities, and differences. The authors begin with the theory behind channel equalization followed by techniques, and the technological realizations for achieving the proper filter in response to variations of the channel. Included in both the techniques and realizations are the rebirth of the use of artificial intelligence as a self-learning filter for the weights to use by the filtering structure of channel equalizers. These equalizers were compared, contrasted, and their key differentiation was identified. It was found that gaps such as complexity and convergence time are potential areas for extending the performance and limits of existing channel equalizers.

A Comparative Survey of Coding, Multiplexing, and Equalization Techniques Used in Coherent Optical Fiber Communications

“…How does transmission and reception of data or information work, and how do noise and intersymbol interference be introduced or developed in a channel and then seen at the receiver? The transmitted signal and waveform in transmitting symbols are depicted by '()* and +()* respectively, while the interval of a symbol and the nth symbol is depicted by , andrespectively, thus the transmitted signal is shown as -+() − ,* in (2) [69], [70].…”

“…On the other hand, the received signal is shown in (3), where the received signal and received waveform are depicted by /()* and 0()* respectively, while the nth symbol received and intersymbol interference plus the noise are depicted by and ()* respectively [69], [70].…”

“…Thus, equalizers are implemented to retrieve the correct transmitted symbols. Equalizers are also called filters, as their primary purpose is to remove unwanted channel effects [69].…”

“…12, it is used to mitigate nonlinear channel effects and intersymbol interference. An adaptive equalizer is a self-learning implementation that removes unwanted attributes in the input signal and channel impairments by updating the weights to eliminate the error signal [69]. In Fig.…”