Title :
Stochastic quasi-Newton method for minimum bit error rate nonlinear equalizers online training
Author :
Renxiang, Zhu ; Lenan, Wu
Author_Institution :
Sch. of Electron. & Inf. Eng., Ningbo Univ. of Technol., Ningbo, China
Abstract :
A sliding window hybrid quasi Newton algorithm is proposed in this paper for minimum bit error rate nonlinear equalizers online training. Switching between sliding window stochastic gradient algorithm and sliding window quasi Newton algorithm makes the new algorithm be stable and converge fast. Moreover, by modifying the quasi Newton method, the new algorithm can be applied to high-dimensional parameters. In simulations, the new algorithm is used for training nonlinear equalizers in direct sequence spread spectrum communications and its high efficiency is proved by simulation results.
Keywords :
Newton method; code division multiple access; error statistics; gradient methods; spread spectrum communication; stochastic processes; direct sequence spread spectrum communication; high-dimensional parameter; minimum bit error rate nonlinear equalizer online training; sliding window hybrid quasiNewton algorithm; sliding window stochastic gradient algorithm; stochastic quasiNewton method; Algorithm design and analysis; Bit error rate; Equalizers; Matrices; Newton method; Signal processing algorithms; Training; communication signal processing; learning algorithm; minimum bit error rate; nonlinear equalizers;
Conference_Titel :
Computer and Information Application (ICCIA), 2010 International Conference on
Conference_Location :
Tianjin
Print_ISBN :
978-1-4244-8597-0
DOI :
10.1109/ICCIA.2010.6141545
