Pre-Correlation Noise And Interference Suppression For Use In Direct-Sequence Spread Spectrum Systems With Periodic Prn Codes

Suppression of pre-correlation noise is decisive to enable GPS acquisition in weak signal situations. In addition, the residual frequency offset introduced by satellite and receiver motion and receiver reference oscillator errors should be compensated to avoid further signal loss. Stand-alone high sensitivity GPS (HSGPS) utilizes short coherent integration times followed by long noncoherent integration. On the other hand, assisted GPS (AGPS) uses external aiding information to allow for a longer coherent integration beyond data bit boundaries. Additionally, the external information can also aid in narrowing the frequency/code search space. However, the aforementioned techniques are potentially limited by additional system complexity needed to provide assistance data, limiting effect of oscillator stability and SNR loss introduced during squaring operation.Differential detection schemes have been proposed to deal with some of these limitations. For instance, post-detection differential and pre-detection differential techniques yield enhancements in terms of acquisition sensitivity and robustness to frequency errors respectively. Nevertheless, they are still limited by the same fundamental problems of frequency error and SNR loss. Substantial performance improvement can be accomplished by combining the merits of coherent and differential/noncoherent techniques while compensating for their weaknesses. This paper proposes an innovative GPS acquisition, which utilizes coherent integration in the form of pre-filtering and differential detection in the form of multi-correlation differential detection. This technique allows for weak signal acquisition even with very large frequency offsets. Preliminary analysis using hardware simulated and live GPS data corroborate the superior acquisition performance of the proposed technique over traditional noncoherent scheme.