A new lattice polynomial perceptron and its applications to fading channel equalization and adjacent-channel interference suppression

Abstract

Introduces a new lattice polynomial perceptron (LPP) with a faster convergence rate. The adaptation of the parameters in each lattice stage can be carried out independently, which is very suitable for VLSI implementations. Applications of the LPP for frequency-selective fading channel equalization and adjacent-channel interference suppression in 16-level and 64-level quadrature amplitude modulation (QAM) receiver systems are considered. Computer simulations are used to evaluate and compare the performance of LPP with that of the polynomial perceptron (PP), fractionally spaced bilinear perceptron (FSBLP), fractionally spaced decision feedback multilayer perceptron (FSDFMLP), and the conventional decision feedback equalizer (DFE). The results show that in 16QAM system, the performance of LPP is better than that of PP and FSDFMLP. FSBLP can achieve almost the same performance as that of LPP. While in 64QAM system, the performance of LPP is clearly superior to that of the other structures.