Research On Key Technologies Of Reliable Communication Reception In Complex Battlefield Environment

The modern battlefield environment has gradually evolved into the multi-dimensional and complex battlefield environment of land,sea,air and sky.The joint operation of multi-arms cooperation has become the mainstream trend of future wars,and wireless communication is the bridge and link of their cooperation.Wireless communication system needs higher transmission rate,better communication reliability and better frequency band utilization to meet the requirements of complex battlefield environment.At present,various complicated external noise sources and electromagnetic interference in wireless channels,as well as intersymbol interference caused by multipath propagation,have become the bottleneck problems that limit reliable reception and broadband transmission of wireless communication systems.Therefore,filter technology and channel equalization technology become important research contents of wireless communication reliable receiving system in complex battlefield environment as important means to filter out communication frequency band interference and suppress inter-symbol interference.In this paper,the wireless communication system architecture of a missile in the joint combat system is analyzed.The key of reliable communication reception lies in the problem of filtering the noise out of frequency band and suppressing the inter-symbol interference in frequency band.The problem of eliminating out-band noise is mainly solved by band-pass filter.Cavity filters are widely used because of their higher Q value,larger power capacity,more stable structure and better heat dissipation performance compared with other types of filters.The traditional design of dual-passband filter is mainly realized by multi-mode resonator or combination of resonators,and the design is carried out by network synthesis method.The design process is complex,requires multiple iterations and corrections,and has limited flexibility.The problem of suppressing inter-symbol interference is mainly solved by channel equalizer.Blind equalizer is widely used because of its better equalization performance compared with other types of equalizers.However,traditional blind equalization algorithm has disadvantages of slow convergence speed and limited equalization performance.This paper mainly focuses on the above problems,and the specific work is as follows:1.In view of the problems of complex synthesis and poor flexibility of the coupling matrix of the existing band-pass cavity filter,this paper studies a method of extracting the coupling matrix of the double-pass cavity filter based on the theory of diplexer,and designs a doublepass cavity filter.In the design process,the extraction of coupling matrix is a very key link.At present,the extraction of coupling matrix is mainly assisted by computer,and the optimized coupling matrix can be obtained after several iterations of optimization.In this paper,the coupling matrix of double-passband cavity filter can be transformed from the coupling matrix of double-passband cavity filter by analyzing the similarity between the diplexer and double-passband filter.Based on the above theory,this paper introduces the duplexer synthesis method and its coupling matrix design idea into the dual-passband cavity filter.Firstly,the desired dual-pass band cavity filter index is converted to the diplexer index,and its coupling matrix is obtained.The coupling matrix of double-pass cavity filter is obtained by matrix transformation.This design method is not affected by passband parameters,which can greatly improve the design efficiency and flexibility of doublepassband cavity filters,and can be applied to all band-pass cavity filters.2.To solve the problem that the traditional blind equalization algorithm has slow convergence speed and is not suitable for high-order quadrature amplitude modulation signal,a modified constant modulus blind equalization algorithm based on bat algorithm is studied.The initial weight matrix of the traditional modified constant modulus blind equalization algorithm starts with zero matrix because there is no reference value,which ultimately leads to slow convergence and affects equalization performance.Considering the similarity between the iterative process of swarm intelligence algorithm and the iterative process of equalizer,this paper combines bat algorithm and blind equalization algorithm in swarm intelligence algorithm,and replaces bat algorithm’s evaluation function with the error function in blind equalization algorithm.Then the bat algorithm adaptively calculates the initial value of the weight vector in the blind equalization algorithm.After obtaining the optimized initial value,the modified constant modulus blind equalization algorithm is implemented.Both theoretical and software tests prove that the proposed equalization algorithm has good convergence and equalization performance.Based on 64-QAM signal,the proposed algorithm is about 1000 steps faster than the modified constant modulus blind equalization algorithm,and the algorithm can be applied to almost all blind equalization algorithms.3.Aiming at the problems of slow convergence speed and high steady mean square error of traditional blind equalization algorithm,a concurrent modified constant modulus blind equalization algorithm and decision directed algorithm based on two-step gradient method are proposed in this paper.Considering similarity of the equalization process and optimization process,existing blind equalization algorithm and two-point step size gradient method in optimization algorithm be combined.First,the modified constant modulus blind equalizer and the decision directed equalizer are run simultaneously.Then,the iterative step size of the decision directed equalizer is optimized by two-point step size gradient method,so that the algorithm can converge faster and have lower steady mean square error.Finally,software simulations have further confirmed that the proposed algorithm has lower steadystate mean square error and residual inter-symbol interference.Compared with the original algorithm,the convergence speed of the proposed algorithm is about 100 steps faster,and the steady-state mean square error is reduced by 5dB.