Research On The Application Of Neural Network In DOA Estimation

Direction finding is one of the major problems for many applications such as radar,navigation, mobile communications, electronic warfare systems, sonar and seismology.Direction finding algorithms have also been known as spectral estimation,direction-of-arrival (DOA) estimation, angle of arrival (AOA) estimation, or bearingestimation. In fact, the goal of DOA estimation algorithm is to estimate the direction of thesignal of interest from a collection of noise ‘‘contaminated’’ set of received signals.Direction finding have followed an evolutionary trend. In the previous decade, somepowerful and high-resolution methods for DOA estimation such as MUSIC and ESPRIThave already been developed. However, these conventional methods usually consumed a lotof time, because they use the method of linear algebra as to require the calculation of amatrix inversion. Therefore, they were not able to meet real-time requirements.With the rapid development of computational intelligence technology, people begin toresearch to solve the direction of arrival estimation problem by learning a lot of samples.Neural Network is considered to be a powerful tool to solve this problem as result of thenonlinear mapping and generalization ability. The advantage of this method lies thatmodeling process is using the training samples to structure neural network instead ofaccurate mathematical equations. In practice, the collected training samples can take thenoise, signal noise ratio, transmission channel model, and other factors into account,without the need for eigen value decomposition, spectral peak searching, and calculationscan be fast parallel implementation, which is expected to be applied to practicalengineering.The main contributions of this paper are as follows.1. This thesis studied a method for the direction of arrival estimation about one signalusing Selective Neural Network Ensemble (SNNE). Selective Neural Network Ensemblebased on Particle Swarm Optimization (PSO) is proposed to solve the direction of arrivalestimation of one signal. The basic idea of the method is to optimally select NeuralNetwork to construct Neural Network Ensemble with the aid of PSO. This may maintainthe diversity of Neural Network and decrease the effect of co-linearity and noise of sample.The computer simulation shows that the method is more excellent compared with BPNN,RBFNN, GRNN and MUSIC algorithms about the one signal of DOA estimation, which makes it feasible to carry out in practical interference location system.2. Particle swarm optimization is used for optimization of BP neural network toimprove the performance of direction of arrival estimation. Due to the fact that BP neuralnetwork is inclined to be trapped in local extreme, a novel network, particle swarmoptimization based BP neural network, is proposed to solve the above shortcoming, it isapplied to direction of arrival estimation for study. This thesis only presents using the firstrow of correlation matrix instead of common used upper triangular half of the covariancematrix, therefore the feature dimension is largely reduced without losing any DOAinformation. Experimental results show that the performance of the method in this papercompared to classic RBF method is much better in terms of neural network size,generalization and estimation precision, so it has a broad application value.3. As the different SNRS, antenna numbers, snapshots and signal angle intervalchanges, the DOA estimation performance between PSO-BP neural network and the classicRBF method is compared and analysised. Experiment shows that the performance ofPSO-BP neural network method is significantly better than that of RBF method under thesame experimental conditions.