Research On Parameter Estimation In Ultra Wideband Communication System

Ultra Wideband (UWB) is a kind of wireless communication technology which transmitsinformation by nanoseconds or even sub-nanoseconds impulse. Due to its high time resolution, theaccuracy of UWB positioning based on time of arrival (TOA) estimation for UWB signals can reachcentimeter or even millimeter level. Furthermore, if the direction of arrival (DOA) can be estimated, itwill contribute to accurate positioning. So the research on TOA and DOA parameter estimation hasgreat significance in theory and application. The main contents of the thesis are listed as follows:(1) A propagator method (PM) based on the rotational invariance technique is proposed for jointTOA and DOA estimation in UWB system. This algorithm obtains the TOA estimates by fullyexploiting the rotation invariance property of the propagator, and then gets the DOA estimates via thedifference of the TOAs. Compared with the traditional PM algorithm, the proposed algorithm does notrequire exhaustive spectral peak searching. Furthermore, it achieves automatically paired parameters.(2) A successive PM algorithm for joint TOA and DOA estimation in UWB system isinvestigated. The algorithm first obtains the initial TOA estimates by fully exploiting the rotationinvariance property of the propagator, and then achieves the exact TOA estimates by simplifying thetwo-dimensional global searching in two-dimensional propagator method (2D-PM) algorithm into twosuccessively one-dimensional local searching. DOA parameters can be estimated via the difference ofthe TOAs. This algorithm not only has lower complexity than2D-PM algorithm, but has very closeparameter estimation performance to2D-PM algorithm. Meanwhile, it can estimate automaticallypaired parameters.(3) Joint TOA and DOA estimation based on root multiple signal classification (MUSIC)algorithm is presented in UWB system. The algorithm obatains the closed-form solution of the TOAsby simplifying the two-dimensional global searching in two-dimensional multiple signal classification(2D-MUSIC) algorithm into two one-dimensional root MUSIC algorithm. The problem of TOAspairing is settled. And then gets the DOA estimates via the difference of the paried TOAs. Thealgorithm ont only has much lower complexity than2D-MUSIC algorithm, but has very closeparameter estimation performance to2D-MUSIC algorithm.(4) Joint TOA and DOA estimation in UWB system using a successive MUSIC algorithm isproposed. The proposed algorithm gets the initial TOA estimates corresponding to the first antenna viaroot MUSIC algorithm, and obtains the exact TOA estimates in the two antennas via two successively one-dimensional searching. It then estimates the DOA parameters via the difference of the TOAsbetween the two antennas. The proposed algorithm not only has lower complexity than2D-MUSICalgorithm, but has very close parameter estimation performance to2D-MUSIC algorithm. Meanwhile,it can obtain the parameters paired automatically.(5) A joint TOA and DOA estimation algorithm in UWB system based on the sparse representationis proposed. This algorithm utilizes thel0norm as the sparsity constraint in the sparse representationframework, and employs the classical orthogonal matching pursuit (OMP) algorithm to obtain the TOAestimates. Then TOA pairing method is given. DOA parameters are estimated via the the difference ofthe TOAs. The advantage of this algorithm is that it has better parameter estimation performance withone single snapshot.