摘要
为解决DS-CDMA系统的多址干扰问题,提出了一种将自适应惯性权重的混沌粒子群算法与独立分量分析方法结合(CAPSO-ICA)的盲多用户检测算法。该算法首先将自适应的非线性递减惯性权重w和混沌运动引入到粒子群(PSO)算法中,有效地避免了传统粒子群算法易陷入局部最优的问题,从而使寻优结果更为准确。然后根据各用户相互独立这一特点,将改进的粒子群算法与独立分量分析方法结合起来进行盲多用户检测。仿真结果表明,在相同的条件下,相对于已有的FICA算法和PSO-ICA算法,基于CAPSO-ICA的盲多用户检测算法有更小的误码率,这说明改进算法的多用户检测性能更为优越。
To solve the multiple access interference problem in DS-CDMA system, a kind of CAPSO-ICA blind multi-user detection algorithm was proposed. This algorithm introduced adaptive nonlinear decreasing inertia weight and chaotic motion into classic particle swarm optimization（PSO） algorithm firstly, in this way, the local optimum problem can be avoided and the optimization result is more accurate.As the users are independent, the algorithm then combined the chaos particle swarm optimizer using adaptive inertia weight（CAPSO） algorithm with independent component analysis（ICA）. Simulation results show that the CAPSO-ICA algorithm has a smaller bit error rate in comparison with the FICA and PSO-ICA algorithm under same conditions, which indicates that the improved algorithm performs better in blind multi-user detection.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2014年第6期1285-1290,共6页
Journal of System Simulation
基金
中央高校基本科研业务费专项资金资助项(N100404018)
关键词
盲多用户检测
独立分量分析
粒子群
混沌
自适应非线性递减惯性权重
blind multi-user detection
independent component analysis
particle swarm optimization
chaos
adaptive nonlinear decreasing inertia weight
