Dbf Sensor Gain And Phase Error The Calibration Robustness Of The Algorithm

Adaptive digital beamforming(ADBF) is a perfect combination of array antenna and modern signal processing, it is also called adaptive spatial filtering. Compared with the traditional array antenna technology, ADBF has better performance in multi-beam, flexibility, low sidelobe, it can effectively control beampattern and cancel the interferences according to the external electromagnetic environment. However the presence of gain and phase error in practical engineering would seriously influence the static beampattern, causing antenna gain lose, sidelobe increase and pointing precision degrade. And at the same time when adaptive algorithms are adopted the errors would also make signal models mismatch, giving rise to degradation of SINR, response speed of the algorithm, resolution of direction-finding and depth of nulls. In this thesis, the array calibration and robust algorithms are researched based on the analysis of error influence on DBF system. Main works of this thesis include:(1). The effect of gain and phase error on static beampattern and adaptive algorithms are studied on the basis of setting up the models of narrowband signal and gain and phase error. Through simulation, the former gives some parameters of static beampattern under different mean square error, and the latter shows the adaptive beampattern and the changing gains with different SNRs under certain error;(2). Focused on the problem of imprecise location of auxiliary source in far field calibration, calibration of linear array based on timesharing operation source is studied. Its calibration requirements and related improved method are imposed with its performance simulation followed by. Through adding the times of turning the turntable the method of timesharing operation source is introduced to apply to planner array, both the theory and simulation certify the accuracy and effectiveness;(3). The MCT (Mutual Coupling Technique) is applied to the calibration of DBF array here. The influences of errors such as array edge error, uncertainty confirming error, array position error and random error on calibration results and final pattern are studied using corresponding simulations to provide references for error analysis during actual calibration, simultaneously ways to reduce these errors are given followed by. At last some matters are given in practical operation;(4). Beside calibration robust algorithm against gain and phase error is another way to overcome error influences. In this thesis the white noise gain based algorithms are introduced, robust variable load based RLS(RLS-VL), positive load based SMI(PSMI) and negative load based SMI(NSMI) are especially discussed. The simulation results show that the robustness of these traditional algorithms can be significantly improved with the white noise gain constraints when the errors exist.