A Study Of Array Adaptive Digital Beamforming

Adaptive array (AA) technique is one of the branches in array signal processing. It has played important roles in both military and commercial application fields, such as radar, communication, sonar, navigation, acoustics and speech signal processing, and so on. Especially, it has become a key technique in adaptive active phased array multifunction radars and smart antennas for mobile communication systems. With about half a century's development, numbers of scholars have made great progress in AA techniques. Nowadays, AA has become applied in real systems. However, the research in this field is far from perfect. The array signal adaptive digital beamforming is studied in this dissertation. The common DBF algorithms and error model are summarized in the article and the method of suppressing the main-beam disturb and stationary interferers is investigated. This dissertation analyses the application of virtual transformation in array signal and the adaptive pattern control is studied. The main contributions of this dissertation can be summarized as follows:1. When target signal is in the sample data, the performance of adaptive digital beamformer (ADBF) will be degraded. If the interference falls into mainlobe, the mainbeam will be distorted with conventional ADBF. A new Block Matrix method is proposed in chapter 3 which preprocesses the sample data via a block matrix so that the influence of signal-of-interest(SOI) and mainlobe jamming is diminished. This method avoids mainlobe distortion and has the better performance of interference cancelling than the direct sample matrix inversion (SMI) method. This method can deal with multiple mainbeam interferer and coherent interferer. For non-uniform array, we also present relevant block matrix, The theoretical analysis and results of simulation demonstrate that the better performance will be obtained .2. In adaptive arrays the notch position formed by conventional ADBF method is usually sensitive to the direction of the interference. In some special environment the interferences can't be effectively suppressed such as the moving interference. In practical application the optimal weight gained with the snapshot in a period (weight training) is used for the next period (weight applied) because of the limitation of the calculation speed and the weight does not change in applied period until the next update. The shortcoming of this method is that the weights and the data applied are possibly mismatched. An effective method is to widen the notch so that the interference is always laid in it. Using the conventional ADBF method to cancel the fixed strong interference perturbations and the random interference the computational complexity is large and the freedom of the system is consumed. We proposed an approach that can reduce the computation burden and the approach don't occupy the extra freedom of system in chapter 4.3. The ADBF techniques play an important role in a variety of applications such as radar, sonar and communications systems. Most of the ADBF algorithms require that the array should be uniform linear array(ULA). In practice, the actual array may be non-uniform linear array(NULA), uniform circular array(UCA) or arbitrary geometry array. The array interpolation method can solve this problem and that most of paper in this area investigate how to reduce the interpolation error to a minimum level, however, the sensor number of the virtual array affects the system performance too. In the dissertation some effective conclusion are obtained.4. This paper describes a simple adaptive pattern control method based on linearly constrained minimum variance algorithm in chapter 6. The desired static pattern weights are imposed on constrain set which make the overall beamformer sidelobe response equal to the desired quiescent response and simultaneously form nulls in the direction of interference. The sidelobe has many ripples in condition of small training sample size and the diagonal loading(DT) technology is applied to this method, which improve the convergence speed of sidelobe. We also study the pattern control of area array. The effectiveness of this new method is illustrated by the designed simulations.5. Adaptive sidelobe cancellation is a important part of phase array radar. To realize low sidelobe and suppressing the interferer the antenna technique and signal process are combined. The assistant array also has different geometry and position and the performance of system will be different. In chapter 7 we discuss the canceling performance in several typical form and position. For area arry we use reduceing dimention method to process SLC and study the performance of non-uniform auxiliary channel.