基于改进PSO的连续相位调制训练序列优化  

TRAINING SEQUENCE OPTIMIZATION FOR CONTINUOUS PHASE MODULATION BASED ON MODIFIED PSO

作　　者：王乐[1] Wang Le(School of Information Science and Engineering,North China University of Technology,Beijing 100144,China)

机构地区：[1]北方工业大学信息学院,北京100144

出　　处：《计算机应用与软件》2020年第7期227-231,共5页Computer Applications and Software

基　　金：2019北京市属高校基本科研业务费项目(110052971921/011)。

摘　　要：用于连续相位调制信号同步的最优训练序列具有自相关函数旁瓣较高的特点,增加了序列起始位置的误检概率。针对该问题,提出基于多约束条件的粒子群算法,搜索旁瓣较低且同步参数估计性能仍然保持最优的训练序列。通过在粒子群算法中引入基因突变,使其尽可能收敛于全局最优,搜索到的训练序列其自相关函数旁瓣得到有效降低,且该搜索方法可以扩展到任意训练序列长度。仿真结果表明,和传统最优训练序列相比,该训练序列能够降低帧起始位置的误检,同时同步参数的估计性能不下降。误码率性能测试表明,该序列的解调性能优于传统最优训练序列约2 dB。The optimal training sequence for the synchronization of continuous phase modulation has the feature of high side lobe,which increases the false detection probability of the initial position of the sequence.Aiming at this problem,this paper proposes a particle swarm optimization(PSO)based on multiple constraints,which searches for training sequences with low side lobes and still maintains the optimal performance of synchronous parameter estimation.By introducing gene mutation into PSO,it converged to the global optimum as much as possible.The side-lobe of the searched training sequence was effectively reduced,and the search method could be extended to any length of the training sequence.The simulation results show that compared with the traditional optimal training sequence,the training sequence can reduce the false detection of the frame initial position.Meanwhile,the estimation performance of synchronization parameters does not decrease.The bit error rate(BER)performances test shows that the demodulation performance of this sequence is better than the traditional optimal training sequence about 2 dB.

关 键 词：粒子群算法 遗传算法 连续相位调制 同步 最优训练序列设计 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术] TN939[电子电信—信号与信息处理]