一种改进的分布式天线系统帧同步度量函数与算法

An improved frame synchronization metric function and algorithm for MIMO-OFDM with distributed antenna system

传统分布式多输入多输出正交频分复用(Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing,MIMO-OFDM)系统定时同步算法中,帧同步算法都是通过直接选取经验值作为阈值进行帧头捕获,而当发射天线数改变时同步度量值变化较大,这将大大增加漏捕获和误捕获的概率.针对这一问题,通过对传统度量函数进行统计特性分析,提出一种修正的定时同步度量函数,推导了其统计特性,并且基于‘3σ’阈值选取准则,形成一种新的基于计算阈值的帧同步(Frame Synchronization Algorithm based on Computed Threshold,CTFS)算法.进行了大量的仿真实验,在不同MIMO发射天线数目下,比较了提出的CTFS帧同步算法与传统的不等周期同步(Unequal Period Synchronization Pattern,UPSP)方案和正交变周期同步(Orthogonal Variable Period Synchronization Pattern,OVPSP)方案的帧同步性能,实验证明相较传统算法,提出的算法大大提高了同步度量值及同步性能的鲁棒性,解决了传统同步度量函数在发射天线变化时出现的问题.

In the traditional timing synchronization algorithms for multiple input multi pie output-orthogonal frequency division multiplexing （MIMO-OFDM）, the synchroni- zation metric performs no robustly in distributed antenna system. For example, when the transmitting antenna number increases, synchronization metric value will decrease considerahly. But traditional frame capture methods are based on the threshold decision with experience value. According to the theoretical analysis, change of transmitting an- tenna number greatly increases the probability of miss capture and mis-capture in tradit-ional synchronization metric function. In order to solve this problem, after analyzing the statistical characteristics of metric function, we modified the timing synchronization metric function and deduced its statistical properties. And then, a frame synchronization algorithm based on computing threshold is proposed using the 3a principle. At last, the proposed algorithm is compared with the traditional Unequal Period Synchronization Pattern （UPSP） and Orthogonal Variable Period Synchronization Pattern （OVPSP） al- gorithm by Monte-Carlo simulation in the different systems with different MIMO trans- mitting antenna numbers. The numerous simulations proves that the proposed algorithm has greatly improved the frame synchronization performance and robustness of synchro- nization metric value. This work has basically solved the problem in traditional synchro- nization function when the number of transmit antennas changes.