Spatial Spectrum Estimation Under Amplitude Limited Condition

Array signal processing is a very hot research topic in the field of modern signalprocessing, an important branch of array signal processing is spatial spectrumestimation, which is also called DOA estimation. Since array signal processingtechnology developed quickly, the electromagnetic environment has becomeincreasingly complex, the existence of noise and the strong interference makes theamplitude value of the receiver may exceed the threshold of the receiver. How toimprove the performance of spectrum estimation under amplitude limited condition isalso a worthy study.This article focuses on the work of the spatial spectrum estimation under amplitudelimited condition. In the traditional spatial spectrum estimation methods, theauto-correlation matrix of the observation vector either make the eigenvaluedecomposition MUSIC, or make the generalized eigenvalue decomposition ESPRIT,or make matrix inversion MVDR to complete the corresponding spatial spectrumestimation. When the received signals by receiver involve strong interferences or strongnoises or signal sources, the received signal value by receiver exceeds the receiverdynamic threshold, so the observation vector in such conditions will inevitably lead to alot of invalid data, it will directly lead to the autocorrelation matrix emerges a largenumber of invalid data. In order to avoid the autocorrelation matrix for eigenvaluedecomposition or inversion and since the signal sources in the space are sparse. In thispaper, we use the sparse signal reconstruction algorithm to estimate the angle of signalsources under amplitude limited condition.In this paper, include the following content:The third chapter of this paper analysis some classic space spectrum estimationalgorithms. We analysis these algorithms from the mathematical models and theprinciple of algorithms and compare the performance of the algorithms by simulation.The main work of the fourth chapter analysis performance between classic spatialspectrum estimation algorithms (Music, MVDR) and sparse signal reconstructionalgorithm used under the amplitude limited condition. Analyzing the performance of several algorithms in different signal-to-noise ratio and considering the existence of thecoherent signal and the number of snapshots impact the performance of thesealgorithms. then comparing resolution of these algorithms under the amplitude limitedcondition. In order to reduce the amount of calculation, finally discussing the1-SVDdecomposition algorithm under amplitude limited condition.The last chapter of this paper discuss the robust beamforming under the amplitudelimited condition. Assuming the received signals by receiver have strong interferencesignals. Through adjusting the weighting vector of the array make the main beam willbe aligned with the desired signal and the desired signal will be reserved withoutdistortion. But the beam in the direction of interferences generated nulling sufficiently,so the strong interfering signals and noise can be suppressed. This chapter do diagonalloading of observation vector autocorrelation matrix to improve the performance of thebeam and finally consider the disturbance of the covariance matrix to obtain robustadaptive beamforming algorithms.