Adaptive Antennas in Multipath Wireless Communications

grantor: University of TorontoWith the ability to null interference and reduce fading in multipath environments, adaptive antennas have emerged recently as an attractive means to increase the capacity of wireless systems without consuming extra spectrum. Typically, the signals received at the different branches of an adaptive antenna are the multipath superposition of signals transmitted from a desired user and many co-channel interferers, plus thermal noise. For a small adaptive antenna, the branch signals are statistically correlated because of lack of independent wave propagation paths, direct electromagnetic coupling between antenna elements, and electromagnetic proximity effects of nearby disturbing objects. The successful design of an adaptive antenna in a cellular or personal communication application depends on how realistically its multipath performance can be predicted in the presence of many interferers and under the condition of correlated fadings. In this thesis, adaptive antennas are studied as diversity antennas with optimum combining (i.e. with adjustment of both amplitude and phase). First, an approximate analytical cumulative distribution function (CDF) of the antenna output signal-to-interference-plus-noise ratio (SINR) is derived and expressed in terms of elementary functions, which enables quick and accurate evaluation of the antenna SINR performance at any probability level in most cases. The electromagnetic interaction between antenna elements enters into the CDF primarily through the correlations among the branch signals. With the CDF thus available in terms of elementary functions, the average bit error rate (BER) performance for coherent binary phase shift keying (BPSK) is derived, a technique that can be extended to other modulations. Furthermore, a simple approximation and an optimization criterion for antenna performance are developed, which are useful in the analysis and design of adaptive antennas where correlations between branch signals are non-zero. The impacts of antenna-element mutual coupling and array geometry on the signal correlations and hence the SINR performance are studied for small adaptive antenna arrays. When integrated into the electronics of modern transceivers or operating within buildings, adaptive antennas are subject to proximity effects of nearby disturbing objects, which may severely degrade antenna performance. The mechanisms of proximity effects are identified and several case studies with basic configurations are carried out to illustrate the phenomenon, quantify it, and point the way to its mitigation. Combined with electromagnetic modelling tools, the obtained analytical results are used in the design of a small three-element adaptive antenna for an indoor base station, and in the study of proximity effects on a small two-element antenna. These theoretical results are verified experimentally at 900 MHz in an indoor environment using a digital beamforming testbed especially developed for this purpose.Ph.D