超高速相干光通信两步步长优化CMA算法  

Two-step stage optimized CMA algorithm in super-high coherent optical telecommunications

作　　者：钟昆 杨怀栋[1] ZHONG Kun;YANG Huaidong(State Key Laboratory of Precision Measurement Technology and Instruments, Department ofPrecision Instrument, Tsinghua University, Beijing 100084, China)

机构地区：[1]清华大学精密仪器系精密测试技术与仪器国家重点实验室,北京100084

出　　处：《应用光学》2019年第3期505-510,共6页Journal of Applied Optics

基　　金：国家重大科学仪器设备开发专项(2014YQ510403)

摘　　要：在超高速相干光通信中,对偏振解复用损伤和偏振模色散的补偿一般采用恒模算法(constant modulus algorithm, CMA)。标准的CMA算法具有自适应动态均衡的功能,加权信号根据输入信号进行不断更新,是一种目前流行的盲均衡算法。通常在CMA算法中,步长系数的取值严重影响着算法收敛、误差矢量幅度(error vector magnitude, EVM)和误码率(bit error ratio, BER)性能评价。提出了一种优化的CMA算法,具有两步步长系数,可标记误差矢量函数曲线稳定时的区间。经过计算验证表明,优化的CMA算法具有不同程度的解调性能改善,能够降低X、Y偏振态的EVM,具备较好的应用性。In ultra-high coherent optical telecommunications, constant modulus algorithm (CMA) is usually adopted for compensation of impairment of polarization demultiplexing and polarization mode dispersion (PMD). Standard CMA has a function of self-adapt dynamic equalizer, with coefficient signal constantly updating according to input signal. It is a popular blind- equalizing algorithm at present. In CMA algorithm, the value of step-size usually severely influences the performance evaluation of algorithm convergence, error vector magnitude (EVM), and bit error ratio (BER). This paper presents an optimized CMA algorithm of two-step size coefficients, which is capable of labeling stable span of error vector function curve. The verification of calculation shows, the optimized CMA algorithm can improve the demodulation performance, decrease the EVM of X and Y polarization state, validating its preferable applicability.

关 键 词：通信技术 相干光通信 偏振复用 恒模算法 

分 类 号：TN913.[电子电信—通信与信息系统]