光学学报, 2012, 32 (10): 1006004, 网络出版: 2012-08-13
高速相干光通信系统中的自适应步长恒模算法
Adaptive Step-Size Constant-Modulus Algorithm for High-Speed Optical Coherent Communication System
光通信 恒模算法 相干检测 偏分复用 正交幅度调制 optical communications constant-modulus algorithm coherent detection polarization multiplexing quadrature amplitude modulation
摘要
对于高速光通信系统中的相干检测来说,盲均衡是一项常用的恢复传输符号的技术。目前应用最广泛的盲均衡算法是恒模算法,但是一方面它不能兼顾收敛速度和收敛精度,另一方面在偏分复用系统中很容易陷入奇异性。因此提出了一种改进型的恒模算法自适应步长恒模算法,并在112 Gb/s偏分复用16进制正交幅度调制系统中对提出的算法进行了测试。与传统的恒模算法相比,改进后的算法收敛速度仅为恒模算法的1/20;收敛稳定后误差函数的方差比恒模算法降低了0.7 dB;对偏分复用系统中的奇异性有很好的抵抗能力;在不考虑奇异性问题时,光信噪比代价与恒模算法相比降低了1.5 dB。
Abstract
In coherent detection of a high-speed optical communication system, blind equalization is a common technique to recover transmitted symbols, and one of the most widely used algorithms is constant-modulus algorithm (CMA). But on one hand, it can not converge fast and precisely enough; on the other hand, it can be caught in singularity problem while operating on a polarization division multiplexed system. A modified CMA, which is adaptive step-size CMA, is proposed. The performance of the proposed algorithm is numerically tested in a 112 Gb/s polarization division multiplexed 16 quadrature amplitude modulation (QAM) optical coherent system. Compared with conventional CMA, the convergence rate of the modified CMA is 20 times faster, and the variance of the error function is 0.7 dB lower. Besides, the modified CMA can overcome the singularity problem quite well. Without considering the singularity problem, the optical signal noise rate (OSNR) penalty is 15 dB lower than that of conventional CMA.
邸雪静, 童程, 张霞, 张晓光, 席丽霞. 高速相干光通信系统中的自适应步长恒模算法[J]. 光学学报, 2012, 32(10): 1006004. Di Xuejing, Tong Cheng, Zhang Xia, Zhang Xiaoguang, Xi Lixia. Adaptive Step-Size Constant-Modulus Algorithm for High-Speed Optical Coherent Communication System[J]. Acta Optica Sinica, 2012, 32(10): 1006004.