An Image Formation Algorithm for Bistatic SAR Using GNSS Signal With Improved Range Resolution

Abstract

Passive synthetic aperture radar (SAR) using Global Navigation Satellite System (GNSS) signals as the illuminators of opportunity has a relative coarse range resolution due to the narrow bandwidth of the navigation signals and the bi-static acquisition geometry, limiting its application for high-resolution requirement. In this paper, an image formation algorithm for range resolution improvement is investigated. A modified second order differentiation based method is applied on the range compressed signal of reference channel and surveillance channel, improving range resolution and keeping side-lobes at a lower level simultaneously, and the phase of processed signal is preserved by phase multiplication compensation for the following azimuth compression processing. To validate the effectiveness and benefits of the proposed algorithm in terms of range resolution and targets side-lobes, point targets simulation under different situations and real field experiment with strong scatters are conducted and discussed. Compared with the traditional GNSS-based passive SAR imaging algorithm, the results of the proposed method show that the imaging performance can be improved for better range resolution and lower side-lobes, benefitting for better distinguishing nearby targets and recognizing weak targets.

Decimetre Level Range Resolution for Beidou B3I Signal-Based Passive SAR Imaging—Proof-of-Concept Experimental Demonstrations

“…Different from our previous work [23] and the work in [24], in order to avoid the unnecessary computations in the regions without reflected signals, the range domains which have the obvious correlated pulses are detected and selected for further processing. The respective rule for threshold setting can be expressed as follows:…”

“…For resolution issue with respect to passive SAR using other types of GNSS signals i.e. GPS C/A signal, GLONASS signal, Galileo signal and Beidou B1 signal, there already has several related works [8]- [24]. Among them, for bistatic GNSS-SAR system, range resolution (the main-lobe width of range compressed pulse) is determined by signal bandwidth and bi-static angle [8], [20], [21], while azimuth resolution is determined by dwell time and the trace for forming synthetic aperture imaging [12], [21].…”

“…Moreover, [23] outer-performs [22], as the method in [23] will not degrade image signal-to-noise (SNR) level. At the same time, the literature [24] improves the algorithm in [23] by reducing side-lobe magnitude for range profiles.…”

“…To date, the best range resolution reported in the GNSS-SAR-related literature only reached the meter level, indicating considerable room for improvement in terms of remote sensing applications for high-resolution imaging. Moreover, although single Diff2 operator [23], [24] can greatly enhance range resolution without distorting SNR with respect to numbers of sampling points (determined by sampling frequency), its impact is not unlimited. Through a respective investigation based on the methods in [23], [24], when sampling frequency reaches a certain level, the value of range resolution will not increase any more.…”

“…Moreover, although single Diff2 operator [23], [24] can greatly enhance range resolution without distorting SNR with respect to numbers of sampling points (determined by sampling frequency), its impact is not unlimited. Through a respective investigation based on the methods in [23], [24], when sampling frequency reaches a certain level, the value of range resolution will not increase any more. For instance, in GPS C/A code signal based SAR, when sampling frequency increases to the level no less than 1.6 × 10 7 Hz, the range resolution value using the methods in [23] and [24] remains at 40 m. In summary, an evaluation of the performance of Beidou B3I SAR imaging and attempts to further improve the range resolution are strongly warranted, and will increase the applicability of this novel technique for environmental surveillance.…”