ADBF And Pattern Contorl At Subarry Level

The study on ADBF (Adaptive Digital BeamForming) approaches has been carried through for more than thirty years and lots of effective approaches have been proposed, but they are basically based on element level. ADBF techniques have important applications in phased array system, now the array usually comprises thousands of elements, so it usually adopts subarray configuration. Thus we need adopt ADBF methods at subarray level.Monopulse angle estimation is usually used in modern phased array radar. Monopulse is a radar technique through which the angular location of a target can be exactly determined. This technique can estimate the angle based on outputs of sum and difference beams. The performance of monopulse angle estimation degrades severely when there are jammers. If not effectively countered, electronic jamming can prevent successful radar target detection and tracking. Thus, as an important component of monopulse angle estimation, the difference beam also needs adaptive technique to cancel the jammer to maintain target detection and tracking exactly. Study on adaptive difference digital beamforming at subarray level has practical and academic value although it was rarely carried through.Firstly, based on the structure of the subarray, we construct the difference signal model at subarray level. By generalizing it to 2-D, we construct 2-D signal model at subarray level, this model has no limitation to the array configuration and can be applied to any plane array. The above models are appropriate to overlapped or non-overlapped subarrays.Combining the above signal model and the conventional LCMV method at element level we present a LCMV method at subarray level. In all kinds of jammer scenarios, the algorithm can suppress jammers well, but the disadvantage is that the SLL (SideLobe Level) of adaptive patterns is much higher.Quiescent pattern control is important for radar systems equipped ADBF and spatial adaptivity. We present some algorithms: the normalization algorithm; the MOD algorithm by introducing mismatched control vector and the SSP algorithm by partitioning subspaces to reduce the dimensions of adaptive process.The approach combining SSP and conventional approach can suppress SLL of adaptive pattern effectively and reduce adaptive performance loss compared with SSP.In monopulse angle estimation system, the LCMV approach can null the jammer effectively when there are jammers. But the monopulse technique for DOA estimation degrades severely when there is jammer in mainlobe. We study on ADBF approach for canceling MLJ (MainLobe Jammer) at subarray level while preserving the radar's ability to estimate the target angle accurately using monopulse techniques. Then we combine an SLC (SideLobe Cancellation) approach with MLC (MainLobe Cancellation). A mainlobe maintenance (MLM) technique or constrained adaptation during the sidelobe cancellation process is imposed so that the results of the SLJ (SideLobe Jammer) cancellation process do not distort the subsequent mainlobe cancellation process.Simultaneous, this paper study on DOA ability, when there is sensor position uncertainty in superresolution spatial spectrum estimation algorithm.We simulate and analyze all the presented algorithms. The simulation results demonstrate the efficiency of the algorithms.