邻近支持向量机的阵列波束优化

摘要

For the shortcomings of standard support vector regression Beamformer, such as high computational complexity, memory overhead of large and slow training, it Proposes the beam optimization method based on Proximal Support Vector Machine (PSVM) . The method of PSVM breaks the conventional thinking of solving the original problem through dual problem, sets the constraints of support vector regression equations, analyses and solves the original problem directly. Finally, the optimization model of beamformer based on PSVM is given and its concrete realization of the process and numerical simulation experiment are done. The results show that the method reduces the computational complexity and the memory occupancy and improves the training speed. Compared with the traditional support vector regression beamforming, it provides a new and effective ways for the optimization design of beamformer.%针对标准支持向量回归波束形成器的计算复杂度高、内存开销大、训练速度慢的缺点,提出了邻近支持向量机(Proximal Support Vector Machine,PSVM)波束优化方法.PSVM打破了通过对偶问题求解原问题的传统思维,将支持向量回归的约束条件等式化,直接对原问题进行分析与求解,给出了基于PSVM波束形成器的优化模型及具体实现过程,并进行了数值仿真实验.研究结果表明,在保持波束形成器性能基本不变的情况下,降低了计算复杂度,减少了内存开销,提高了训练速度.与传统的支持向量回归波束形成相比,具有良好的快速性,为波束形成器的优化设计提供了一种新的有效方法.