低复杂度的多符号差分空时分组编码在快速瑞利衰落信道中的研究

Reduced complexity MSDD of STBC over fast Rayleigh fading channel

为了降低多符号差分检测(MSDD)算法的高计算复杂度,提出了2个低复杂度的算法,分别是简单化的M算法(SMA)和改进的多符号反馈检测(IMDFD)算法.SMA是在M算法的基础上,通过删除对后续搜索过程不起作用的路径数,来达到和M算法性能相同,但复杂度降低的目的;而IMDFD算法结合了SMA和MDFD算法的思想,通过联合估计最大化值和设置序列长度值的方法,不仅达到低复杂度,而且还克服了MDFD算法的错误传播和低信噪比下低性能的问题.仿真结果显示,这2种算法通过增加观察间隔,都能很好的填补差分检测和相关检测之间的间距,并且随着星座点数的增加,SMA算法和IMDFD算法之间的性能间距在逐渐拉大,相对来说SMA算法具有稍高的复杂度但却有更好的性能.

In order to reduce the high computation complexity of multiple symbol differential detection （MSDD） algorithm two algorithms with reduced complexity are presented, i.e., simplified M-algorithm （SMA） and improved multiple differential feedback detection （IMDFD）. SMA has the same performance with M algorithm, but has lower complexity by deleting some paths that have no effect on the later process. And IMDFD combines the ideas of simplified M-algorithm and MDFD algorithm. By allying estimating max and setting the length of signal sequence, it not only reduces the complexity, but also can solve the problems of error propagation and performance degradation at low signal-noise ratio. Simulation results show that these two algorithms are effective to fill the gap by extending the observation interval. With increase of the number of constellation dots, the difference of performance between SMA and IMDFD algorithm is enlarged, the SMA algorithm has better performance with higher complexity.