NEURAL NETWORK BASED DIRECTION• O.F ARRIVAL ESTIMATION

Abstract

In order to reduce the time of computation and to mitigate the problem of array failure, the direction of arrival problem have been approached here using neural network technique, instead of analytical techniques. In the first phase of work a NN-based DoA estimator was designed for switched beam antenna system which can be applied in DS-CDMA. In this we found the Sol AoA in the presence of a varying number of mobile users and a varying Sol to MA! ratio. In the second phase of the work the problem of multiple source tracking was accomplished with a NN-based smart antenna which is robust to sensor failure condition also. A 16 element uniform array was taken as the test antenna to validate the developed technique under different failure condition. The performance of the developed model was tested for multiple source directions