The Algorithm of Blind MLSE Based on Advanced MSE Sigmoid Variable Step-size
Aimed at the disadvantage of convergence performance of blind maximum likelihood sequence estimation (MLSE) and its cause, this paper develops advanced Sigmoid function combined mean square error applied in control of step size of blind MLSE based on Newton gradient algorithm. This algorithm can avoid maladjustment and divergence of algorithm because of burst error. Thus, it has well tracing performance. Bit error performance analysis is carried out for several different adaptive equalizers in trap channel. Theoretic analysis and simulation validates the superior performance of the algorithm of blind MLSE based on advanced MSE Sigmoid variable step-size Newton gradient.
Variable step-size Newton gradient MLSE Viterbi
Chen Xing Li Chuantao Yuan Kuo Li Jinjiang
Major of communication Engineering,Shandong Institute of Business and Technology, China,264005 Major of electronic technology, Yantai Dongfang Wisdom Electric Co., Ltd, China, 264000
国际会议
第八届国际测试技术研讨会(8th International Symposium on Test and Measurement)
重庆
英文
1045-1048
2009-08-01（万方平台首次上网日期，不代表论文的发表时间）