Array Antenna Model-based Radar Signal Separation Algorithm

With the increasing of air traffic, the response signal of aircraft arrive at the receiver usually have those problems which including overlapping, code garbling, confusion, fruit, targets out of range and so on. In order to achieve the precise position of aircraft and extract useful information, we need to separate the mixed signal. In order to solve this problem, in this dissertation some effecvie algorithms have been proposed based on the existing algorithms. First, Reasearched the PA (Project Algorithm) method based on the array antenna signal processing. Through the simulation we can know this algorithm has low computation cost, good separateion efficiency and high stability. Secnode, aiming at the shortmings of PA and bases on the PA extending the research, then obtains the EPA. The application scope of EPA is more than PA. Third, reasearched blind source separation algorithm, choosing appropriate number of iteration and compared function, the mixed signal can be separated. From the simulateion, we know the separation result is reliably and stably.The paimary work in this dissertation:1. Reasearched the PA (Project Algorithm) method based on the array antenna signal processing which is usedful for separating when two signals overlapping. The mixed signal has some characteristics on the time domain, and then uses SVD mathematical principia and Moore-Penrose generalized inverse matrix and some related mathematical knowledge to obtain separateion matrix of the mixed-signal matrix. Thus we can estimate the source signals.2. Reasearched EPA (Extended Project Algorithm) method. It is based on the array antenna signal processing. When the number of source signals is more than 2, PA can not be used to separate the mixed signal; when the time difference of arrival is small, PA can not be used to separate the mixed signal. EPA has to pre-processes the mixed signal and then based on the principle of iteration, separte the overlapping signal one by one.3. Reasearched blind source separation algorithm. Blind source separation is a hot spot in signal processing and neural network in recent years, and it is mainly used for voice signal processing, which is rarely used in radar signal processing. This dissertation researched the signal whitening based on PCA. The contrast functions have been analyzed, the selected methodology for contrast function has been summarized, and several contrast functions have been proposed. The result shows that the blind source separateion can excellently separate the signals, which their components are overlapped in frequency domain.