Radar Signal Recognition Based on Multilayer Perceptron Neural Network

Low Probability of Intercept (LPI) radars are developed on an advanced architecture by making use coded waveforms. Detection and classification radar waveforms important in many critical applications like electronic warfare, threat to surveillance. Precise estimation parameter the type waveform will provide information about also helps develop sophisticated intercept receiver. The present work is modulation LPI using multilayer perceptron neural (MLPN) network. approach based following two steps. In first step, analysed cyclstationary technique which models signal bi-frequency (BF) plane. Using this algorithm, BF images signals obtained. second fed a feature extraction unit get salient features then network for classification. Nine types noise free (Frank, four polyphase codes poly time codes) classified obtained step. success rate achieved 100 % signals. experiment repeated various levels up -12dB SNR. noisy signals, before feeding MLPN network, denoised denoising filters connected cascade 93.3%