Reduced complexity acquisition of GPS signals for software embedded applications

Abstract

Location technologies have many possible applications in wireless communications, surveillance, military equipment, gaming industry, tracking and safety applications to name a few. Wireless operators expect annual revenues to be tens of billions from potential location-based services. The paper focuses on the positioning in cellular wireless networks using Assisted Global Positioning System (A-GPS) technology. State-of-the-art A-GPS receivers are implemented using hardware correlators, as the computational complexity is very high. Positioning receivers are operated in many operational modes depending on the environment, provided assistance, internally available data and measurements. Hardware implementations constrain the performance of A-GPS receivers and as a result their operation is not robust indoors and in urban canyons. Highly reconfigurable receivers can accommodate all the possible modes of operation, and receivers implemented in 'software' are the most promising and flexible ones. To implement A-GPS receivers in 'software', one should drastically reduce the complexity of algorithms. The authors suggest a novel fast Fourier transform (FFT)-based correlator which is good for software implementations. As the focus is on A-GPS, the assistance information from aiding sources reduces the operational range of the correlator and this approach combines gains due to FFT processing and aiding. The complexity is further reduced by employing a search in frequency in FFT domain which is called the shifting replica method. Extensive simulations on several platforms confirm the validity of the proposed method.