Research On Localization Method Of Mixed Far-field And Near-field Sources With Gain-phase Errors

Sources localization is one of the important research contents in array signal processing field.The traditional sources localization techniques research mainly focus on pure near-field or pure far-field signal models,but in practical far-field sources and near-field sources often coexist,in which case traditional estimation methods will fail.At present,the estimation results of traditional mixed far-field and near-field signal sources positioning algorithms under the gain-phase errors will produce large deviations or even fail.Therefore,researching the novel algorithms to locate the mixed far-field and near-field signal sources under gain-phase errors can improve the array signal processing technology and promote the development of spatial spectrum researches.First of all,This paper Carries out data analysis and simulation experiments on a representative mixed far-field and near-field signal source location algorithm to explore the advantages and disadvantages of various algorithms.Simultaneously,the effects of gain-phase errors on representative algorithms are simulated.The effects of different errors on the source angle and near-field source distance are compared respectively.Secondly,in order to solve the problem of mixed source localization under gain-phase errors,this paper proposes to use a high-power calibrated source to calibrate gain-phase errors and use weighted?₁-norm constraint and MUSIC joint algorithm get signal parameters.The algorithm uses a high-power calibration source and array covariance element calibrates the array gain-phase errors,and then uses the weighted?₁-norm with MUSIC to classify the mixed source and complete the far and near field sources estimation of angle and near-field distance.The results of data derivation and simulations show that this algorithm has a better estimation effect under the gain-phase errors than the classic algorithm of mixed source positioning.At last,this paper proposes weighted sparse total least squares algorithm for partially calibrated arrays,which can solve the problem of mixed source localization under gain-phase errors.The algorithm first uses partially calibrated symmetric uniform arrays and M-STLS algorithm to estimate all source angles and gain-phase errors.Then compensating the gain-phase errors,constructing a discriminant function to distinguish the angle of the mixed source,and using the spatial differencing method technique to separate the mixed source.Finally,the spectral peak searching is used to estimate the near-field sources distance parameters.Compared with the classical mixed source positioning algorithm,this algorithm has better estimation effect and more practicality under the background of gain-phase errors in the array.