Research On Digital Signal Detection Based On Nonlinear Stochastic Resonance

With the rapid development of wireless communication,there are a large number of mobile communication terminals,more and more crowded wireless spectrum resources and complex communication modulations.These factors pose a threat to the service quality of wireless communication.In order to reduce the influence of adverse factors on the service quality of communication,how to detect useful signals in complex interference environments,recover useful original signals,and improve the service quality of communication is the important problem that we need to solve now.In view of the need for detection of wireless communication signals which are submerged in noise,this dissertation will further study the theory of nonlinear stochastic resonance(SR),focusing on array SR,parameterturning SR,adaptive SR and other issues.SR is a nonlinear resonance phenomenon in nonlinear subject,mainly describe the weak signal and noise in nonlinear system under the action of three to achieve a synergistic resonance phenomenon,namely,under the effect of nonlinear system,noise energy transfers to the energy of the signal,thus the output signal energy of nonlinear system is enhanced.This dissertation mainly studies the waveform detection method of binary pulse amplitude modulated signal and theoretically analyzes detection performance in respect of the output signal-to-noise ratio(SNR),the SNR gain and error the rate(BER),using the array SR theory,parameter-turning array SR theory and SR theory based on variable step size adaptive particle swarm optimization(PSO)and genetic algorithm(GA)in the complex interference environment.The research contents and main contributions are summarized as follows:1.Under the condition of gaussian and non-gaussian noise,combining the array SR theory with the multiple antenna reception and diversity combining algorithm,a combined algorithm of branch output based on array SR system is proposed.The main research is based on the array SR output SNR gain of maximum ratio combining(MRC),equal gain combining(EGC)and selection combining(SC)algorithm,and put forward the array SR as the receiver of digital communication system.The BER of the communication system based on the three combining algorithms is compared and analyzed,and the optimal combining algorithm of the array SR system is explored.The theoretical analysis and numerical simulation are used to analyze the output SNR gain and BER of array SR.The conclusion is that the performance of array SR system is better than that of single branch SR system.With the same number of branches,the performance of the MRC algorithm is better,the performance of the EGC algorithm is the second,and the SC algorithm is the worst.Under the non-gaussian condition,the performance of the array SR is better than that of the gaussian noise,and the detection performance of the weak signal is greatly improved.In conclusion,these conclusions provide a theoretical basis for the selection of each branch output combining algorithm of array SR.2.Combining the parameter tuning SR theory with the parallel array idea,a parameter tuning array SR system is proposed.Parameter tuning SR is a resonance phenomenon of noise,signal and nonlinear SR system by adjusting system parameters a and b.This dissertation uses parallel array ideas to further improve the parameter tuning SR detection performance of weak signal and the main way is to add the output adjustable parameters of each branch of SR system together and average the summation.By modeling the probability density distribution of single branch output signals as the gaussian distribution,SNR gain and BER expressions of parameter tuning array SR system are derived.The relationship between the number of branches of parameter tuning array SR and the SNR gain and the BER are also simulated and analyzed.The conclusion is that for the detection of weak binary pulse amplitude modulation signal,parameter tuning array stochastic resonance system has better performance compared with a single parameter tuning SR system,and with the increase of the number of branches,the output SNR gain of parameter tuning array SR is significantly improved and the BER is significantly reduced.It is of great significance to make the parameter tuning SR system in the research of signal detection.3.A variable step size adaptive stochastic resonance weak signal detection method based on PSO is proposed.With the SNR and the BER of the system as the fitness function of the PSO,using PSO algorithm to search for the best value of the whole population,the problem of adaptive selection of h of a,b and Runge Kutta method is solved.The adaptive stochastic resonance problem is transformed into a multi parameter global optimization problem,and the optimal parameters of adaptive stochastic resonance system are obtained,which ensures that the system produces better resonance characteristics,so that the weak signal in the background of noise can be detected optimally.The conclusion is that for the detection of weak binary pulse amplitude modulation signal,stochastic resonance system based on PSO has better performance compared with parameter tuning stochastic resonance system and constant parameter stochastic resonance system,and the output SNR ratio gain of the variable step size adaptive stochastic resonance system based on PSO is greatly improved,and the BER is obviously reduced.4.A variable step size adaptive SR detection method based on GA is proposed.By using GA selection,crossover and mutation operators,the adaptive synchronization optimization of a,b and Runge Kutta h for SR system is achieved.The adaptive stochastic resonance problem is transformed into a multi parameter global optimization problem,and the optimal parameters of adaptive stochastic resonance system are obtained,so that the adaptive stochastic resonance can detect weak signal better.The parameter tuning stochastic resonance system and the constant parameter stochastic resonance system are compared in the detection of weak binary pulse amplitude modulation signals.From the two aspects of output signal to noise ratio and bit error rate,the variable step size adaptive stochastic resonance system based on genetic algorithm is analyzed.It is concluded that the variable step size adaptive stochastic resonance system based on genetic algorithm makes the stochastic resonance system always in the best resonance state,which also maximizes the output SNR and reduces the BER to the minimum.By comparing the output BER of PSO algorithm and GA,it is proved that the adaptive stochastic resonance based on the optimized algorithm has reached the optimal detection performance of weak signal.