Research On Mininum Module Casecaded Canceller Algorithm

The aim of adaptive array processing is to constrain the interference, which couldn't separate from time-frequency domain but from space domain, and get some information from useful signal. In order to meet the swift shifting environment, adaptive array processing must deal with the problems of calculation complexity and convergence speed. In application, it is hard to get large number of samples that meet for the requirement of stationary. The non-stationary data, especially the non-stationary outliers or impulsive noise spikes which can not be separate from expect signals in space domain, have a severe influence to the convergence of sample covariance matrices used in block processed adaptive algorithms. While there exists the bigger outliers from expect signal direction of arrival, the nulls will occur to the expect signal direction of arrival, and that will eliminate the expect signal. Often, the outliers occur in airborne radar and GPS receivers applications, especially under complex electromagnetic interference and intended interference environment. When the outliers exist, the adaptive processor could be invalid. The paper is concerned on the outliers influence, and novel minimum module cascaded canceller (MMCC) are introduced here that perform well in outliers contaminated data environments. Convergence performance is shown to be commensurate with SMI in non-contaminated environments as well. The novel algorithms are applied to GPS receivers to show the good performances in interference suppression.The main work of this paper is concluded as the following aspect:1. Due to the defects of large operation and instability convergence performance in the array signal adaptive processor under the non-stationary impulsive noise (outliers) from signal direction of arrival, a minimum module cascaded canceller is proposed by substituting the median of the median cascaded canceller (MCC) for the minimum module value. The enhanced algorithm has the ability of fast convergence, less operation and performs well with non-stationary samples. Simulation results indicated the algorithm is shown to reach same convergent performance using fewer samples as the kind of sample matrix inversion (SMI) algorithm.2. Due to suppression the influence of impulsive noise spikes (outliers) from desired signal direction of arrival to adaptive weights calculation, an multistage minimum module cascaded canceller (MMMCC), householder-based multistage minimum module cascaded canceller (HMMMCC) and improved multistage wiener filter (IMSWF) are proposed by substituting the weights or some weights of the multistage winner filter for the minimal module weights, which have the minimal module of the samples quotient between channels. The enhanced new algorithms retain the good performance of the multistage winner filter and the minimum module value methods, and avoid the influence of impulsive noise spikes (outliers) from desired signal direction of arrival to adaptive weights calculation effectively.3. Due to suppression the influence of impulsive noise spikes (outliers) from desired signal direction of arrival in GPS signals. The Householder multistage wiener filter was improved, a sample selection Householder multistage wiener filter was proposed. The enhanced algorithm employed by space-time anti-jamming processing in GPS receiver, with the ability of avoiding the impulsive noise spikes (outliers) influence of weights calculation, and hold the desired signal unchanged. Simulation results indicate the algorithm achieved favorable anti-jamming performance.4. Due to suppression the influence of impulsive noise spikes (outliers) from desired signal direction of arrival in GPS signals, the MMMCC, HMMMCC and IMSWF algorithms are adopted in space-time anti-jamming processing in GPS receiver. These methods have the prominent ability of restricting the influence of the impulsive noise spikes (outliers) from desired signal direction of arrival to adaptive weights, and the ability of low complexity, also fast convergence. Computational simulation results indicate the method achieved better fast convergence ability and a good performance in reducing the influence of the impulsive noise spikes (outliers) to adaptive weights.