Fast Acquisition Method And The Service Testing Model For Burst DSSS Signals

Acquisition process is the primary stage of the burst Direct Se-quence Spread Spectrum(DSSS)communication receivers,and the key techniques are to achieve fast signal acquisition and to build a service-testing model against the acquisition system.The former determines whether the signal can estimate the pseudocode phase delay and carrier Doppler shift to a certain accuracy within a specified time,allowing the receiver to access the next stage successfully.The latter is an important way to evaluate the service performance of the acquisition system.Inbound signals of the Radio Determination Satellite Service(RDSS)belong to a typical class of burst DSSS signals.The RDSS closed-loop testing system is an important platform to verify the acquisition performance.In this platform environment,the testing signals have the characteristics of large signal-to-noise ratio,and the fast capture technology has been a research hotspot in the related industry.Grounded in the above requirements,this thesis selects the strong inbound signal as the research object under the RDSS closed-loop testing system environment,and investigates several theoretical and technical problems of the acqusition system for burst DSSS signals.For the problem that the traditional acquisition algorithms neglect the sparse characteristics in the time domain of the capture result,an acquisition method based on sparse recovery is proposed to improve the acquisition speed.By taking advantage of the sparsity of the acquisition result in the time domain,we introduce sparse factor to deal with the signals via sparsification,and then apply the sparse recovery to search and estimate the acquisition result.The computation complexity and mean acquisition time,as well as the relationship between the inbound capacity and the acquisition performance are provided.We present the theoretical analysis for the effect of the sparse factor on the acquisition performance.The estimation errors verify our analysis,and the simulations show the remarkable effectiveness of our method compared to the advanced parallel code-phase acquisition.Compared with non-downsampling acquisition methods,using downsampling techniques is an important way to achieve fast acquisition.However,the legacy acquisition methods via downsampling all suffer from the problem of downsampling loss.In addition,GPU-accelerated signal processing has already become attractive for application.Therefore,a GPU-based acquisition method with minimum loss and downsampling techniques(MLD)is proposed,which introduces a preprocessing mechanism to obtain the maximum signal-to-noise gain of the downsampled output while not changing the noise energy,and with low computational complexity and good acquisition probability.The experimental results show that the MLD acquisition outperforms other downsampling schemes and the capture speed worked on the GPU is greatly improved compared to the CPU platform.In view of the issue that the general service-testing model of acquisition system cannot meet the requirements of burst DSSS communication,we provide a novel approach based on multiserver queueing model with loss mechanism.First,we use a probability model of burst DSSS signals based on the Poisson distribution by using the theorem of the binomial distribution to approximate the Poisson distribution,which can simulate the stochastic process of the signals entering the acquisition system.Second,based on the signals occupying system processing time obeys the negative exponential distribution,a multiserver and loss-existed queueing model,named as M/M/n/n queuing model,is constructed through the above probability model combined with the first-come-first-served principle of the acquisition channels,which can analyze the service performance for a given call loss rate.Thereby,a service-testing method of acquisition system is portrayed through a global discrete-event simulation model with stochastic and dynamic characteristics from the signal to the system level.Finally,the Monte-Carlo method is used to simulate the signals inbound time and the acquisition system processing time,which can reach the same conclusions as the theoretical analysis.Thus,using M/M/n/n queueing model can test the service performance of acquisition system for burst DSSS system.