Study On The Smart Antenna's DOA Estimation And Adaptive Beamforming Techniques

Smart Antenna (SA) is one of the key techniques in the future mobile communications. The mobile subscribers hit such a large number, made the Multi-access mode indispensable. While TDMA, CDMA and CDMA methods greatly enlarged the system's capacity, severe interferences were introduced. Also, radio waves are subject to numerous fadings due to the complex channel environment. Hence signal enhancement techniques are necessary. SA can adjust its beaming patterns based on the arrival direction of signals, track the expected signal, cut or counteract the interferences and improve the SIR (Signal to Interference Ratio). Introducing smart antenna to mobile communication systems boosts spectrum utilities, enlarges the system's capability, expands base stations'coverage, reduces the electromagnetic pollution and improves the system's overall service quality. SDMA (Spatial Division Multiple Access) based on SA effectively supplements the traditional multi-access techniques and maximize the customer's number.SA has two patterns: Beam-switching and Adaptive Array Antenna (AAA). The latter gains more interests because of its flexibility. Given the sampled data of each element, signal processing of AAA consists of three parts: to determine quantity of the sources, to detect Direction of Arrival (DOA) and Digital Beamforming (DBF). They are not independent; rather, they are highly related.The major work and endeavors are shown below:①Two structures of the Smart Antenna was related, its mathematical model, geometric patterns, and key techniques, mainly the Adaptive Array one.②The conventional Capon method and MUSIC algorithm was analyzed and simulated. Moreover, the comparison between them was made.③MUSIC algorithm was adapted to CDMA signal's 2-D DOA estimation. It made full use of CDMA signal's spatial information and enhanced DOA estimation's performance, greatly supplemented the traditional MUSIC algorithm.④The traditional Maximum Eigenvalue Optimal Weight-Vector algorithm was revised and applied to the practical model with colored pollution. By correcting the interference-noise subspace and altering the scaling factor, this thesis successfully dismissed the direction error caused by the uncertain and time-varying colored noise. Computer simulation convinced the algorithm's value. This thesis studied the traditional MUSIC algorithm and Optimal Weight-Vector adaptive algorithm, extended their applicabilities. However, these works are accomplished under some assumptions. How to estimate DOA information in the colored noise environment and how to form decent beams without knowledge of the noise's covariance matrix require further research.