Performance of systems using adaptive arrays in mobile ad hoc networks

In the design of future mobile ad hoc networks, it has been widely assumed that adaptive antenna arrays could significantly improve the overall network throughput. However, implementing adaptive antenna systems is neither simple nor inexpensive. In order to make reasonable decisions regarding the use of such technology, the achievable performance improvement using adaptive arrays must be quantified. This paper studies the system throughput of mobile ad hoc networks with various antenna configurations. Computer simulations are used to calculate these throughputs. The computer simulation allows control of various parameters, such as antenna geometry, number of antenna elements, node movement model, propagation model, and channel characteristics. The performance of systems using adaptive arrays is compared to the performance of a baseline system which uses omnidirectional antennas. In particular, the performance of systems that steer only the main beam is compared with systems that steer nulls as well. It is shown that steering only the main beam of an adaptive antenna array, so that it points in the direction of the signal of interest, results in performance that is only slightly better than that of an omnidirectional antenna. However, steering the nulls along with the main beam, such that the signal to interference ratio is maximized, results in system throughput that is several times greater than that obtained with omnidirectional antennas.