The Key Technologies About Dectection And Estimation Under Complex Electromagnetic Enviroment

With the rapid development of modern wireless communication, communication systems and modulation schemes is gradually complicated and diversified. Spectrum was increasingly congested and overlapping, which lead to an increase of background noise and interference significantly. The electromagnetic environment is extremely complex. This complex electromagnetic environment has badly restricted transmission quality and reception performance of communication system on civil and military fields, which even caused connection broken suddenly. Thus, it offers higher demands and severer challenges for wireless communication, especially signal detection and estimation in communication terminal. Due to the nonlinear stochastic resonance (SR) technology has good effects in noise suppression and weak signal detection, it is widely applied in many different areas such as biology, chemistry, electronics, image processing. However, the application of stochastic resonance in communications is still facing many key problem including model design, complicated signal process, parameter optimization, adaptive control, et al. Therefore, on the basis of in-depth research on typical SR systems and solution to their belated bottle-neck problems, this thesis proposed some important technologies about weak signal detection, and parameter estimation based on SR, which further promoted the reception performance of wireless communications systems under complex electromagnetic environment.First, the SR phenomenon and its mechanism in typical systems are researched and analyzed in details, which provides the theorotic basis for its subsequent actual applications. Then, the periodic signal and noise are put into bistable stochastic resonance (BSR) and suprathreshold stochastic resonance (SSR) mode, and the SR phenomena caused to synergistic effect between signal, noise and nonlinear systems and their mechanism are research and analyzed. Moreover, the adaptive parameters adjust strategy based on linear transformation is proposed, and system parameter optimization and complex signal process problem are solved for optimal performance in SR systems.Next, nonlinear SR process technology is applied in weak signal detection, to improve detection performance and reduce detection time. An energy detection (ED) based on BSR is proposed under the complex environment. By utilizing the SR process of received signal, proposed algorithm can achieve energy transfer from noise to signal, thus effectively eliminates the influence of interference noise and improves 1～2dB detection performance. Then, a truncated sequential probability ratio test (SPRT) algorithm based on BSR is introduced and the optimal truncated threshold is derived to further shorten detection time about 50%. Moreover, we propose an ED algorithm based on generalized stochastic resonance (GSR), which can improve the detection performance by adding SR noise with special probability density function.Then, nonlinear SR system is applied in signal reception under Generalized Gaussian noise conditions, to reduce bit error rate (BER) and improve reception performance. Because non-Gaussian interference noise brings to performance degradation of linear best receiving algorithms designed under Gaussian noise, on the basis of analysis on the output signal feature of SSR system, a nonlinear receiving algorithm is proposed based on SSR. Then, a strategy for parameters adaptive adjustment is put forward to ensure the optimization of output SNR and BER. Theory deduction and simulation results show that under non-Gaussian interference noise, e.g. Laplacian noise, the received performance of proposed algorithm is superior to linear best receiving algorithm based on matched filter (MF) for 1-3dB, thus it guarantees the reception performance of wireless communications systems under non-Gaussian noise.Finally, according to parameter estimation demand of spread spectrum (DS) signal, we propose the maximum Doppler shift estimation based on GSR and the PN estimation based on eigenanalysis. The SR phenomenon is discovered in Doppler shift estimation, and new Doppler shift estimation is proposed based on this SR mode. Through a low-pass filtering processing of the received signal, interference noise and the Doppler estimator can be matched, thus the estimation performance of the maximum Doppler shift can be improved more than 1.5dB. Meanwhile, a blind estimation for PN sequence is proposed based on the similarities among the DS signals and the eigenanalysis technique. Simulation results show that compared with the existing algorithms, the proposed algorithm not only overcomes the partial-encode problem, but also improves estimation performance by 2dB. Thus, it achieves an accurate estimation of PN sequence in low SNR conditions.