A HEALING SYSTEM FOR FAILED ANTENNA ARRAY USING EVOLUTIONARY COMPUTATIONAL TECHNIQUES

Abstract

Antenna arrays are an integral part of many signal acquisition systems such as sonar, radar, and satellite communication, since they provide additional dimensions of flexibility and control to system designer. The arrays used for these applications utilize the spatial diversity effectively, known as beam-forming in array terminology, for improving the quality of signal and reducing interference. Because of the presence of the large number of radiating elements in an antenna array, there is always a possibility that some of the elements may malfunction. One of the reasons for this is that the active components like transistor and switches, T/R modules, power supplies used in phased array antennas have a finite lifetime. Alternatively, the degradation in array performance may occur due to some unforeseen reasons like vagaries of weather or natural calamities. Thus, over a period of time, as the components of the antenna fail, the antenna performance will degrade which is a matter of concern to the system designer. Faults in an array degrade the far field radiation pattern of the antenna. This degradation may be in the form of increased side lobe levels (SLL), decreased gain and directivity, and the removal of nulls. Thus, entire system performance is affected due to element failure. One possible solution to this problem could be the replacement of the defective element(s). However, this increases the overall cost of the antenna and system downtime. Furthermore, replacement of the faulty elements is not always possible, particularly when the array is on a space platform or placed in a difficult geographical location. Thus, it is a tremendous challenge for the engineers to establish an uninterrupted and reliable communication by maintaining the radiation properties of the array. Therefore, methods need to be developed to tackle this problem of element failure by means of remote handling in an antenna array, so that the antenna system can heal itself as much and as fast as possible, till more elaborate repairs can be undertaken. The possibility of failure correction for digital beam-forming arrays by remotely changing the excitation of the functioning elements without removing the faulty elements provides a costiv effective alternative to hardware replacements. This extends the effective usefulness of the phased array and its dependent systems. The malfunctioning of one or more radiating elements makes the array asymmetric and hence, it becomes difficult to handle the problem of compensation analytically. It was, therefore, proposed to use computational techniques after converting the compensation problem to an optimization problem. However, classical optimization techniques, like conjugate-gradient method increase in complexity for multi-variable systems. A good alternative is the use of evolutionary computing techniques which have gained currency in recent years. These tools fall under the broad category of soft-computing methods. Over the years, biologically inspired evolutionary computational techniques have been used in all engineering branches for design and optimization. The methods can tolerate imprecision, uncertainty and approximation to achieve robust and low cost solution in a small time frame. Researchers have successfully used techniques like Neural Networks (NN), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), Bacteria Foraging Optimization (BFO) and many more for finding an easy solution for their problems. The robustness of these techniques has been tested for the problems encompassing every engineering field. For the last decade or so, antenna engineers have also frequently used these techniques. The aim of the present research work is to develop methods of compensation for SLL suppression, null steering, DoA estimation and beamforming for a failed phased antenna array with the help of evolutionary computational techniques, viz. particle swarm optimization and bacteria foraging optimization techniques. Although the results of compensation are presented for a typical array structure, through extensive simulation it has been found that it is equally applicable for other arrays also. The overall aim is to make the faulty array to work as a normal one.