几种Bussgang族盲均衡算法收敛性能仿真研究

Simulations of the Convergence Performance of Several Bussgang Family Blind Equalization Algorithms

盲均衡是数字通信中的热点问题。收敛速度和剩余均方误差是衡量盲均衡算法性能优劣的重要指标。常数模算法是目前流行的盲均衡算法,有很多优点,但在时变多径衰落无线信道均衡中,该算法存在着收敛速度过慢和剩余均方误差较大的问题。该文研究了三种Bussgang族盲均衡算法,即常数摸算法、归一化常数模算法和超指数迭代算法,在无线信道均衡中的应用,并通过计算机仿真对其性能进行了分析比较。仿真结果表明,在算法剩余均方误差一致的情况下,超指数迭代算法收敛速度最快,归一化常数模算法次之,常数模算法最慢。因此,超指数迭代算法的性能优于归一化常数模算法和传统的常数模算法。该研究结果在工程实践中具有一定的指导意义与应用价值。

Blind equalization, using convergence rate and residual mean square error （MSE） as a main performance criterion, is a topic problem in digital communications. Constant modulus algorithm （CMA） with many advantages is a popular blind equalization algorithm, but its slower convergence rate and higher residual MSE are disadvantages for equalizing the time varying multipath fading wireless channels. Three Bussgang family blind equalization algorithms applied to wireless channel equalization are discussed comparably, they are CMA, normalized constant modulus algorithm （NCMA）, and super - exponential iterative （SEI） algorithm. Performance of the three algorithms is analyzed by the computer simulation. The simulation results show that the convergence rate of the SEI algorithm is the fastest of all, then is the NCMA, and the CMA is the lowest one. Therefore the performance of the SEI algorithm is prior to that of the NCMA and CMA. The conclusion proposed in this paper has important guiding significance and application value in engineering practice.