Research On Satellite Navigation Signal Acquisition Algorithm Based On Compressive Sensing

At present,the application of satellite navigation system is very extensive,portability and low cost are the new trend of navigation receivers in the future.Signal acquisition is the core component of GPS baseband signal processing,which is an important impact factor affecting the power consumption and cost of GPS receiver module.The traditional acquisition algorithm is divided into two processing modes: time domain and frequency domain.In order to obtain all the information of the signal,the signal sampling is based on the Nyquist sampling theorem,which leads to a large amount of data.So a novel GPS signal acquisition method based on the theory of compressive sensing is proposed.Compared with the existing methods in this field,the proposed one can accomplish the acquisition with much less measurements and at a sampling rate far lower than Nysquist criteria.On the basis of the satellite signal acquisition,thesis proposes an efficient parallel GPS signal acquisition algorithm utilizing compressive sensing instead of circular correlation.The main research work is as follows:Firstly,thesis proposes an efficient parallel GPS signal acquisition algorithm utilizing compressive sensing instead of circular correlation.Specifically,we first represent the GPS signals in sparse form by projecting the signal onto a base matrix consisting of the orthogonal C/A codes.Then,the reconstruction of the compressive signal is analyzed.Because of the high computational complexity of the typical orthogonal matching pursuit algorithm,the idea of distributed processing is proposed,so fitting the compressive sensing problem into coordinate descent(CD)framework and the alternating direction method of multiplies(ADMM)framework.For the coordinate descent method,the original problem is decomposed into a number of related sub problems,and solved one by one along the axis direction.For the alternating direction method of multipliers method,the original compressive sensing problem is decomposed into several independent sub problems.Each step of ADMM can be computed in closed form,thus giving it very low complexity.Finally,thesis compares the traditional acquisition method and the acquisition method based on compressive sensing from two aspects: acquisition time and acquisition accuracy.