新的混合型盲均衡.doc

新的混合型盲均衡算法 摘 要: 针对恒模算法(cma)仅适用于模值为常数的信号,对高阶qam信号的均衡效果差这一问题,结合cma与判决引导(dd)算法各自的优点,提出一种cma与dd算法相结合的改进混合型算法。在算法初期,利用cma+dd并行算法使系统眼图睁开;然后,再转换到dd算法来进一步减小剩余误差。仿真结果表明,采用新算法对高阶信号均衡有较快的收敛速度和较小的剩余误差,星座图恢复很紧凑,并且改进算法能修正相位和跟踪信道。 关键词:恒模算法;判决引导算法;混合型算法;均方误差;高阶qam信号  new mixed blind equalization algorithm  geng tianyu*,shu qin,ying dali ( school of electrical engineering and information technology, sichuan university, chengdu sichuan 610065, china ) abstract: constant modulus algorithm (cma) is only suited to balance the signals with modulus of the same value, but not suited to balance the highorder quadrature amplitude modulation (qam) signals. combined with the virtues of cma and decisiondirected (dd) algorithm, an improved mixed blind equalization algorithm had been presented. first, the improved algorithm would make the system eye open using the cma+dd parallel algorithm, and then switch to the dd algorithm to further reduce the residual error. the simulation results indicate that the improved algorithm has a very fast convergence speed, as well as a very small residual error for highorder qam signals, and constellation recovery plan is very tight. meanwhile, the improved algorithm can correct phase, and trace channel. key words: constant modulus algorithm (cma); decisiondirected (dd) algorithm; mixed algorithm; mean square error (mse); highorder quadrature amplitude modulation (qam) signal  0 引言 由于信道失真导致接收信号存在码间串扰,这就需要用信道均衡技术来减小或者消除码间串扰的影响。盲均衡技术与传统的自适应均衡技术相比,不需要发送序列,从而节省了系统带宽,使得信道的利用率得到提高,而在盲均衡算法中bussgang算法鲁棒性强、原理简单,其中的恒模算法(constant modulus algorithm, cma)具有计算简单、容易实现等优点,因此是近几年来研究的热点。但cma并不对相位信息进行处理,而且当需要较高通信速率时(如图像传输),需采用信道率较高的信号形式(如正交幅度调制信号)[1],对于这类非常模信号,cma误差始终不为零,因此cma虽然可以稳定收敛,但收敛速度慢,而且稳态误差大,尤其是对高阶正交幅度调制(quadrature amplitude modulation, qam)信号均衡效果不理想。而判决引导(decisiondirected, dd)算法可以对相位信息进行处理,但应用的前提是此时信息中大部分码间串扰(inter symbol interference, isi)已经被消除,这样才能保证该算法对代价函数梯度的估计方向在统计上是正确的,此时也能充分利用它优良的剩余误差特性。所以文献[3]提出了用cma与dd算法相结合的cma+dd并行算法,该算法充分利用cma和dd各自的优势,用cma和