Development of an efficient hybrid model for range sidelobe suppression in pulse compression radar

An efficient hybrid model that substantially reduces the sidelobe of the compressed output of the binary phase coded waveforms is suggested by suitably combining a matched filter (MF) and a radial function (RF). The sidelobe suppression is achieved by modulating the MF output by the RF output. Simulation study is carried out to evaluate the performance of standard MF, multilayer artificial neural network (MLANN) and radial basis function neural network (RBFNN) based pulse compressors for binary phase coded pulse compression. The evaluation is based on comparative analysis of the peak to sidelobe ratio (PSR) of the compressed output under noisy as well as Doppler shift conditions. The experimental results demonstrate that the performance of proposed method is significantly superior compared to that of the other standard methods. Further, the hardware requirement of the proposed model is significantly less and unlike other neural networks it does not require training operation.