针对OFDM系统中峰值平均功率比(峰均比)偏高的问题,提出了一种基于粒子群优化的限幅算法。该算法利用粒子群算法优秀的搜索寻优能力来迭代计算最佳限幅门限,结合具有对峰均比控制能力强、实现简单等优点的限幅算法对采样序列进行限幅,通过搜索迭代、限幅、滤波等措施来寻找最优粒子,即在峰均比满足高于门限值限定概率的前提下实现系统最小峰均比的最佳限幅门限,以达到降低系统峰均比的目的。对比仿真了线性递减的惯性权重因子和固定惯性权重因子对系统误码率以及峰均比分布曲线的影响,仿真结果表明:相对于简单的限幅算法,该算法没有带来误码率的额外增加,而能在较低的信噪比时实现降低系统峰均比的目的,对于改善较低信噪比情况下的系统性能有着重要的现实意义。
As to the high Peak to Average Power Ratio (PAPR) issue in OFDM system, the paper has proposed a clipping algorithm based on the Particle Swarm Optimization (PSO). The algorithm takes advantage of the splendid searching ability of the PSO to calculate the best threshold and takes advantages of strong controlling ability and easy implementation of the clipping to clip the signal sequences. It takes the steps of iterative searching, clipping, filtering etc. to search the best particle, i.e. the excellent threshold to obtain the lowest PAPR with the precondition of the set CCDF, achieving the target of reducing the system PAPR. It presents comparisons of BER and CCDF between the linearly decreasing weight (LDW) and fixed inertia weight. The simulations show that, compared with the simple clipping algorithm, by using the proposed algorithm no extra BER is brought but the target of reducing PAPR in the low SNR area is achieved, so the proposed algorithm is of vital significance to improving the performance of the whole system.