UC Irvine

With the advancement of signal processing and other enabling technologies, new products and services have appeared in large numbers over the past decade, and are changing people's daily lives quickly and profoundly. They could not have occurred without the rapid development in areas like digital communications, information theory, detection and estimation, where resource allocation plays a crucial role.

In this work, we study resource allocation for two classes of problems. The first class is rate allocation in digital systems. The functionality of modern digital systems can be broadly divided into two parts: communications and source coding. For communications, we systematically study the allocation problem from a game theory perspective for the multiuser downlink broadcast channel, and apply the solutions to the special case where spatial block diagonalization is combined with time-sharing to multiplex a subset of the users. For source coding, we consider the achievable sum-rate/distortion tradeoff for the Gaussian central estimation officer problem with a scalar source having arbitrary memory. We formulate the variational problem of minimizing the sum rate subject to a distortion constraint, and the conventional Lagrange method is extended to solve the problem. A sufficient condition is also found that can be used to verify if the necessary solution results in the minimal sum rate.

The second class of problems is target localization. We analyze the performance of a reduced-dimension separable space-time adaptive processing algorithm for radar systems under the large array assumption. The study shows that in the asymptotic sense the simplified scheme performs as well as the fully adaptive algorithm with a significant saving in computational complexity. For target localization in wireless systems, we propose a two-stage approach in order to handle non-line-of-sight transmission based on mild assumptions regarding the propagation environments. For the first stage of positioning, we analyze the cases of scattering and reflection respectively, and propose methods to estimate the position and velocity of the moving target. Once the estimation is done, the results can be used as the initial values of extended Kalman filters in the second stage to track the subsequent movements of the target.