Passive Global Navigation Satellite System (GNSS)-based Synthetic Aperture Radar (SAR), known as GNSS-SAR, is a currently developing passive radar sensing system. As the system works in passive mode, GNSS-SAR is much cheaper with a much smaller size, thus it is more flexible to be installed under many application scenarios. However because of the restriction of GNSS signal bands, the resolution of GNSS-SAR is lower than conventional SAR. Also, the weak reflected GNSS signals is another limitation for the application of GNSS-SAR. Due to the fact that GNSS signals are low Equivalent Isotropically Radiated Power (EIRP) sources, the signal strength after reflection will be very weak.
In this study, a new GNSS-SAR imaging algorithm is proposed to improve object detectability under weak reflected signals. Both theoretical analysis and experimental study show that the proposed algorithm can result in obviously enhanced imaging detectability. For instance, using GPS C/A code signal receiver, the proposed algorithm can detect the object with the signal strength as low as -160 dBm, while the conventional algorithm cannot. Meanwhile, computation with the proposed imaging algorithm is significantly more efficient than with conventional GNSS-SAR imaging algorithm.
To enhance range resolution, two new range compression algorithms are proposed to reduce the compressed ambiguity of main-lobe due to chip rate of the respective pseudo-random noise (PRN) code, respectively. In first proposed algorithm (see Chapter 4), range compression is carried out by correlating a reflected GNSS intermediate frequency (IF) signal with a synchronized direct GNSS base-band signal at range domain, where the main lobe ambiguity of the compressed pulse is narrowed down. Thereafter, spectrum equalization technique is applied to the compressed results for suppressing side lobes to obtain a final range-compressed signal. In the second proposed algorithm (see Chapter 5), the main-lobe ambiguity of range compressed signal is deduced by applying Diff2 peak extraction method. Both simulation and field experimental results demonstrate that the proposed range compression algorithms contribute to the resolution enhancement very significantly. For example, on the basis of GPS C/A code receiver platform with the IF value 4 MHz and sampling rate 16 MHz, the first proposed algorithm can improve the best attainable range resolution to 40 m level, while the second proposed algorithm can enhance the best attainable range resolution to 36 m level, compared to the best attainable range resolution 171 m provided by conventional range compression algorithm. In contrast with many current GNSS-SAR research works, the major novelty of the proposed range compression algorithms is that range compressed pulse ambiguity caused by PRN code correlation function has been addressed successfully.