点击次数：

发布时间：2020-09-06

论文类型：期刊论文

发表时间：2020-08-01

发表刊物：IEEE COMMUNICATIONS LETTERS

收录刊物：SCIE

卷号：24

期号：8

页面范围：1812-1815

ISSN号：1089-7798

关键字：Blind equalizers; Convergence; Gaussian noise; Cost function; Interference; Acceleration; Simulation; Blind equalization; bounded nonlinear function (BNF); quasi-Newton; non-Gaussian noise

摘要：Blind equalization is widely utilized to eliminate inter-symbol interference in communication system. It is still a challenge to support blind equalization method within non-Gaussian noise environments. Aiming at improving its convergence speed and robustness performance, in this letter, a novel fast blind equalization method is proposed by using bounded nonlinear function (BNF) and quasi-Newton method. Firstly, BNF-based cost function is proposed to effectively eliminate non-Gaussian noise and realize equalization. Next, quasi-Newton method is developed as the iteration method, which can accelerate the convergence speed. Moreover, theoretical analysis is provided to illustrate that the proposed algorithm has a robust convergence performance. Simulation results show that the proposed method enables evident performance improvement in terms of convergence speed and robustness with non-Gaussian noise.