基于改进k-means算法的可见光通信非线性补偿技术

可见光通信系统中,正交频分复用技术的高峰均功率比会加大发射信号产生非线性失真的概率。为了降低系统的误码率,提出了一种改进的k均值(k-means)算法的非线性补偿技术。该算法考虑了信号的传输特性,运用Dijkstra算法优化了k-means算法中初始点的选择,能够准确地找到接收信号星座图的聚类中心,避免了传统k-means算法在选择初始聚类点时陷入局部最优解。仿真结果表明,相比于传统的k-means算法,改进的k-means算法能更准确地找到非线性失真后信号的星座点位置,因此能有效地解决可见光通信系统的非线性问题,降低了系统的误码率。

Nonlinear compensation technology based on improved k-means algorithm for visible light communication

The peak-to-average power ratio of orthogonal frequency division multiplexing increased the probability of nonlinear signal generation in visible light communication systems. In order to reduce the bit error rate, an improved k-means algorithm of the nonlinear compensation technology was proposed. The transmission characteristics of signals was considered, and the selection of initial points in k-means algorithm was optimized via Dijkstra. The cluster center of the received signal constellation was found, and the traditional k-means algorithm from falling into the local optimal solution was avoided. The simulation results show that this method is more accurate than the traditional algrithm in finding the constellation position of the signal after nonlinear distortion. Hence, it can effectively solve the nonlinear problem of visible light communication and decrease the bit error rate.