Multi-Objective Optimization Approach in Channel Equalization Problems
Pradyumna Kumar Mohapatra, Saroja Kumar Rout, Kottu Santosh Kumar, Anudeep Meda, Ravi Narayan Panda · 2022 8th International Conference on Signal Processing and Communication (ICSC) · 2022
The use of training sequences used by conventional adaptive filters for least mean square algorithms (LMS) is not preferable for researchers. Linear equalizers based on LMS have poor performance for severely nonlinear and dispersive channels. The development of blind adaptive channel equalization algorithms without training signals provides a solution to this problem. In the absence of transmitter assistance, individual receivers can begin self-adapting based on these “blind” algorithms. In this paper, fast blind channel equalization is achieved by using multiple objective optimizations. An examination of this equalizer in the presence of noise shows that fractionally spaced constant modulus algorithms (FS-CMA) have local minima near the mean square error (MSE) equalizers. Due to this, fractional spaced CMA with minimum mean square error (MMSE) receivers that are poorly designed may produce a local minimum with large MSE. In the latter part of the simulation, we considered symbols transmitted using quadrature amplitude modulation (QAM) and analyzed the performance of the