Correntropy Based Parameter Estimation And Source Location On Radio Signal

With the continuous development of radio technology and the popularization of radio applications,it has not only brought convenience to our daily life,but brought a lot of problems in radio monitoring as the same time.For example,illegal broadcasts can spread false advertising and rumors,and illegal radio transmissions can interfere with legal radio communication systems.All of these have seriously negative impacts on the safety of the country,the civilian life and the property.Detecting and locating unknown or illegal radio transmissions are the focus of radio monitoring.In the current complex electromagnetic environment,where pulse noise and co-channel interference signals coexist,most of the traditional locating methods cannot deal with the complex environment,and the performance of the traditional algorithm is degraded.The localization method based on time delay estimation has been widely used because of its simple principle,convenient realization and high localization accuracy.In this paper,we proposed TDOA algorithms based on correntropy in order to adapt the current complex electromagnetic environment.Aiming at the above problems,this paper using the noise model of ? stable distribution,which can simulate a strong impulse noise environment.In this paper we introduce Correntropy and Maximum Correntropy Criterion,combine with Affine Projection algorithm and time delay estimation algorithm to propose a novel adaptive filtering algorithm for channel equalization and time delay estimation.As well as using the cyclostationary property of the signal and combining cyclic correntropy,a novel time delay estimation algorithm which can suppress signal interference in the co-channel interference environment is proposed.The specific research details are as follows:Firstly,for the problem that the minimum mean square error criterion of traditional affine projection algorithm fails under impulsive noise,the Maximum Correntropy Criterion is used as the cost function of the adaptive filter,and the iterative error is substituted into the Correntropy function to effectively control wild point problem of the impulse noise.The new algorithm is applied to the channel equalization experiment,and experiment shows that the learning curve converges fast and the steady-state error is small.Secondly,based on the principle of adaptive filtering time delay estimation,a new time delay estimation method based on Maximum Correntropy Criterion is proposed.The reference signal and the time delayed signal are used as the standard input matrix and the ideal output matrix of the adaptive filter.The estimated value of the time delay is determined by the position of the maximum value of the filter weight.Experiments show that the new algorithm can obtained high accuracy in impulse noise environment.Thirdly,according to the cyclostationary property of the signal,the time delay estimation in the co-channel interference environment is estimated by the method of cyclic correntropy and cyclic correntropy spectrum.The target signal and the interference signal will be separated in the cyclic frequency domain and get the correct estimation result.