One of the important issues in laser radar (Lidar) systems that work based on photon counting photodetectors is the time of arrival (delay) estimation of received optical pulses. It has been a significant research concept from the early days using these detectors. Based upon previous works it seems that further studies in two aspects are reasonable. First,although finding error bounds is necessary, it is insufficient to design practical estimators; therefore finding practical manners toachievedesirable estimates is necessary. Second, we must try to adjust estimationconditions to real environments. In this thesis, we present the maximum likelihood (ML) delay estimation formulation by assuming received Gaussian optical pulse and considering detector dead time under no-background condition. We examine three approximation techniques to solve this equation. Becauseour focus lies on long-range laser radars, we use appropriate delay values according to this system’s operation ranges. Furthermore we consider intuitive “mean” and “count rate based” methods. Simulation results verify that these intuitive methods show success in some complex, such as inno-background noise, conditions when our ML approximation methods fail to work successfully. We also identify the suitable method for each delayaccording to the pulse widths. Background noise and detector jitter effects are considered next. At first we neglect the background noise eventsoutside signal duration, and later by considering the symmetric exterior to the background noise events distribution around the actual delay. Due to undesirable results that were attained from previous solutions in nonsymmetrical background noise events distribution, we apply a new track type ad hoc method to overcome thisproblem. Simulation results show that this method can be successful inthis case. Finally in a different attitude, we propose a new technique for detecting significant delay variations. Simulation results show thatourproposed algorithm can detect delay variations that are greater than or equal to 1.5 times the primary delay. Keywords: 1-Lidar 2-dead time 3-background noise 4-photon counting photodetector 5-jitter
