A Hopfield network based adaptation algorithm for phased antenna arrays

One of the problems of adaptive antennas is to find the weight factors for an array pattern optimizing the signal to noise and interference ratio for the actual signal situation. A neural Hopfield network is able to find the optimal factors, if the direction to the desired transmitter and the interfering transmitters are known. To actualize altering directions, the proposed random search algorithm analyses the signal power of the antenna output. In combination with the Hopfield network it can track the desired signal and suppress interfering sources. This is shown in simulations, which were carried out using a digital controller of an array antenna (algorithm and Hopfield network) and a host computer (signal situation, antenna pattern and output power)