Research On Equalization Technology In Non-cooperative Communication

Non-cooperative communication is widely used in the fields of electronic counterwork for military communications and surveillance by government agencies in the rapidly developing information age.When signals are transmitted in a wireless environment,multi-path effects and channel bandwidth will cause inter-symbol interference.At the same time,in noncooperative communication,the information such as the symbol period,carrier frequency,and training sequence is unknown.When the training sequence cannot be used,the channel needs to be corrected by blind equalization technology.Therefore,the research on blind equalization technology is of great significance in the development of non-cooperative communication.The main research work of this paper is as follows:This paper first analyzes the principle that the blind equalization algorithm that introduces the momentum term can accelerate the convergence speed.At the same time,the traditional CMA algorithm uses a fixed step size to contradict the convergence speed and the steady state error.Function variable step size improved constant modulus(CMA)blind equalization algorithm.This method adds momentum factor and smoothes the mean square error(MSE)based on the traditional CMA algorithm,and then controls it by the improved Tansig function Step size factor.After simulation verification,the improved algorithm has a certain degree of improvement in convergence speed.The idea of variable step size is further introduced into the MMA algorithm and fractional interval sampling is added.After simulation verification,the improved algorithm has a very high degree of constellation point convergence.When the same residual inter-symbol interference is achieved,it will be improved by about 3000 iterations compared to the improved variable-step MMA+DD dualmode equalization algorithm.Secondly,since the a priori information on the origin of the signal is not known in non-cooperative communication,even if the signal passes through the carrier recovery algorithm,it still cannot complete perfect synchronization,and there are remaining carrier deviations.For the blind equalization problem with residual carrier deviations In this paper,the use of the CMA algorithm to blur the carrier frequency offset characteristics,proposed a combination of improved CMA algorithm and phase-locked loop structure,the improved algorithm uses the improved CMA algorithm as the equalization module in the loop,using DD phase detection The phase error obtained by the algorithm compensates the equalizer,and the blind equalization method with carrier frequency offset is realized.The combination of the improved CMA algorithm and the phase-locked loop is verified by simulation.Compared with the traditional CMA algorithm and the phase-locked loop The combination can increase the loop convergence speed by about 700 symbol points,and the constellation of the improved algorithm is more closely integrated.Finally,when using the Gardner synchronization algorithm for timing recovery,it is necessary to accurately know the symbol period of the transmitted signal,but in non-cooperative communication,the symbol rate of the transmitting end is unknown.Although some symbol rate estimation algorithms have passed,there will still be residual deviations At this time,the problem of sampling deviation will occur in the algorithm,causing a lot of inter-symbol interference.For the blind equalization problem with sampling deviation,this paper uses the characteristic that the sampling frequency is proportional to the steady-state error of blind equalization,and proposes the use of blind Balanced steady-state error feedback method for adjusting the sampling frequency.This method first uses the initial sampling frequency for sampling.After obtaining the equilibrium steady-state error,the size of the steady-state error is used to adjust the sampling frequency.It is verified by simulation that this method exists within 0.1% The sampling deviation can still achieve a good balance effect.