ISM的2种改进算法

目前,对宽带信号处理算法的研究主要分为2类:一类是基于非相干信号的处理算法(incoherent signal-subspacemethod,ISM);另一类是基于相干信号的处理算法(coherent signal-subspace method,CSM.)传统的ISM算法由于需要对每个子带做窄带空间谱估计,导致运算量大,而且由于宽带信号能量分布的不均匀,使得估计效果并不理想.针对运算量大和精确度低这2点不足提出了2种改进方法:一是从提高精确性的角度对信号能量做加权处理;二是从减少计算量的角度对信号结构进行改善.通过仿真实验,从均方根误差和分辨概率2个方面验证了2种改进算法的优越性.

Two improved algorithms for ISM

At present, the broadband signal processing algorithms are mainly classified into two classes, which are based on the non-coherent signal （ISM, incoherent signal-subspace method） and coherent signal （CSM, coherent signal-subspace method） separately. Traditional ISM algorithm needs a large amount of computation, for it needs to make narrow-band spatial spectrum estimation in each sub-band, and the estimation results are not satisfying due to the uneven energy distribution of broadband signals. Concerning the low accuracy and heavy computation mentioned above, this paper gives two improved methods one is to improve the signal structure from the perspective of reducing computation load, and the other is to process signal energy by the weighted method from the perspective of improving estimation accuracy. Simulation results show the superiority of the two improved algorithms from two aspects of rootmean-square error and resolution probability.