The Method Based On Stap Eliminate Clutter Interference Study

Array signal processing is an important branch of modern signal processing. It isnow widely used in many fields, such as biomedical engineering, biomedical science,speech processing and radar. If the platform where processer is located is in the air andkinetic, the atmosphere it is exposed to is usually a complex electromagneticenvironment both with strong clutter and jamming, while it has to face with the problemof clutter spectrum broadening due to the platform’s movement. In theory, the optimalprocessor of space-time adaptive signal processing(STAP), a key technology to solvethe above problems, can perfectly suppress the clutter and jamming to achieve the goalof target detection through a space-time two-dimensional joint adaptive processing filter.It is, however, limited by the complexity of the real working environment, thelimitations of platform and the level of technology, and its own high need forcomputation. Therefore, the optimal processor cannot be achieved in reality.In view of this situation, taking practical engineering background into account, thisdissertation firstly preprocesses the sampling data to decrease the degrees offreedom(DOF), occupied by the clutter, of adaptive sidelobe canceller(ASLC), and thenfilter the remaining clutter with the help of STAP. So this dissertation can be mainlysummarized two part:1. A technique to preprocess the sampling data is proposed here. It can perfectlydetect the range bins which include the clutter data and remove them from the trainingdata of ASLC, ultimately releases the DOFs originally occupied by the clutter. Thesimulation shows that a ASLC with a small number of auxiliary channels cansuccessfully suppress the jamming with a nice mainbeam after this clutter preprocessingtechnique.2. As a good technique to reduce the rank of STAP, multistage Wienerfilter(MWF) is used here to filter the remaining clutter after the ASLC, and a variablediagonal loading technique is introduced in the backward recursive proceeding of MWF.The simulation shows that the new algorithm is not only feasible to be realized inpractice due to its reduced rank and low computation load, but also is more robust than MWF only.