The problem of tracking radar targets in the low-angle region Where conventional monopulse radars face difficulties due to the presence of multipath waves is considered in this thesis. The emphasis of the presentation is mainly directed towards finding a new simple closed-form solution to the coherent multipath problem over a smooth surface. Another concern is to improve the performance of the three-subapertures maximum-likelihood estimator when the two received signals are in-phase or anti-phase at the centre of the array. The multipath phenomenon and its modelling for smooth and rough surfaces are discussed and simulation results obtained for different surfaces. subsequently the following are treated First. a new four-subapertures technique to improve the in-phase and anti-phase performance of the maximum likelihood estimator above is derived and simulation results are shown. Then. an improved version of this technique is introduced as a part of the new algorithm. Second. a new three-subapertures trigonometric solution to solve the coherent multipath problem is derived and demonstrated by simulation results. This new method is simpler than the maximum likelihood estimator above and very similar in its estimation accuracy. Third. the performance of the maximum entropy method is tested for the coherent multipath problem by using the three-subapertures arrangement of a linear array. Finally the performances of the above three methods and the normal phase monopulse radar are tested and compared to different surfaces when the coherent and noncoherent multipath exist together. Simulation results show that the performances of the maximum entropy method and phase monopulse are much better than the others when the target is low-tlyine over a rough surface.

Abstract Because of the rapid development of mobile communications, the bandwidth efficiency and the communication quality have become interesting issues for researchers. Multipath fading and multiple-access interference (MAI) are the main barriers of performance improvement. Fortunately, spread spectrum techniques have an inherent ability of anti-interference and anti-multipath. Thus, the commercialized CDMA systems are developed on the basis of spread spectrum techniques. Recently, smart antennas using adaptive array techniques have been applied in CDMA communications. There are two different approaches to realize smart antenna systems: one is named as ¡¥switching-beam arrays¡¦; the other is called ¡¥tracking-beam arrays¡¦. Switching-beam arrays use a number of fixed beams at an antenna site. The mobiles select the beam that offers the best signal enhancement and interference reduction. On the other hand, tracking-beam arrays can adjust their pattern to track desired signals, reduce interference, and collect correlated multipath power. In this thesis, several smart antenna systems based on tracking-beam arrays in DS-CDMA systems are developed. The proposed smart antennas comprise two main structures, i.e., the direction-of-arrival (DOA) tracking and the dynamic beamforming. The improved recursively updated method and the Kalman filter algorithm are employed to track the mobiles¡¦ DOAs. By referring to the estimated DOA, the minimum- variance-distortionless-response (MVDR) dynamic beamformer produces a main beam to target on a certain mobile and simultaneously, interferences from other mobiles are suppressed. Therefore, the MAI in DS-CDMA systems can be suppressed properly at all times. Besides, the inherent processing gain in DS-CDMA systems is utilized to break through the limit that the number of mobiles must be smaller than that of sensors. The proposed smart antennas work properly even the number of mobiles exceeds that of sensor. Simulation results show that the advantages of the proposed technique over the conventional tracking-beam arrays.