Wideband smart antenna design using vector space projection methods.

摘要

Smart antenna arrays are versatile systems used in wireless communication for overcoming fading due to multipath reception, and increasing communication capacity by spatial filtering. There are two major ways of exploiting smart antennas. In the first spatial diversity techniques are used to combine multipath signals in such a way as to ameliorate the effect of multipath fading. In the second adaptive beamforming is exploited to reduce the effect of multipath and create nulls in the direction of interfering signals. In the latter scheme, multiple subscribers using the same radio channels can communicate in the same zone.; We use a well-known iterative method, vector-space projection (VSP), as a design tool to design wideband adaptive beamforming and self-healing arrays. Two types of array structures are discussed: a two-dimensional planar array and a one-dimensional antenna array with tapped-delay lines (TDL) structure. For each design, the motivation for solving the specific design problem, the mathematical model and the derivation of the projection operators are given. We formulate a set of design constraints and then iteratively improve on a trial solution by operations known as "alternating projections". When all of the constraint sets are convex and the intersection of these sets is not the empty set, we can meet the design specifications in a finite dimensional setting. In our simulations we show that reasonable design constraints are readily met. We demonstrate that VSP is useful for broadband self-healing i.e., the reconfiguration of the array when some broadband elements fail to operate. Finally we compare our results with known design procedures for broadband antenna arrays. Our results enable us to conclude that VSP provides a very efficient way to design wideband smart antenna array.