Discrete Time Synchronization Systems for Burst Mode Communication

本课题是对离散时间突发模式同步算法的研究。这项工作的目标是最小化同步捕获时间，以使应用程序能够应用于突发模式系统，如多用户体系结构。主要的设计工具是离散时间最大似然估计器。寻找未知相位或频率的符号定时的下界，以及利用多普勒联合符号时间和载波相位和频率估计的性能。这将建立收购绩效的理论下限。然后将寻求用于离散时间突发模式同步系统的最大似然估计器结构。这项工作将集中在递归和非递归离散时间实现上。在最大似然方程的求解中将使用随机逼近技术。这将导致具有时变增益的数字锁相环系统，大大改善了恒定增益实现的捕获性能。将开发用于载波相位和符号定时的联合和边缘同步的递归结构。结果估计者的获取过程的统计特征是一个主要的重点。马尔可夫理论和高效的蒙特卡罗模拟是工具。将开发和研究用于检测突发开始的最佳技术和最佳前导设计。

This project is an investigation of discrete time burst mode synchronization algorithms. The goal of the work is to minimize acquisition time in synchronization to enable applications to burst mode systems such as multi-user architectures. The primary design tool is the discrete time maximum likelihood estimator. Lower bounds are sought for symbol timing with an unknown phase or frequency and for the performance of joint symbol time and carrier phase and frequency estimation wit doppler. This will establish theoretical lower limits of the acquisition performance. Maximum likelihood estimator structures for discrete time burst mode synchronization systems will then be sought. The work will focus on recursive and nonrecursive discrete time implementations. Stochastic approximation techniques will be used in the solution of the maximum likelihood equations. This will result in digital phase lock loop systems with time varying gains, greatly improving the acquisition performance over constant gain realizations. Recursive structures will be developed for both joint and marginal synchronization of carrier phase and symbol timing. Statistical characterization of the acquisition process of the resulting estimators is a major emphasis. Markov theory and efficient Monte Carlo simulation are tools. Optimal techniques for detecting the start of a burst and optimal preamble designs will be developed and investigated.