Postgraduate research opportunities Advanced Imaging algorithms for distributed spaceborne SAR

Usually, Synthetic Aperture Radars have been deployed on a single platform with the transmitter and the receiver co-located, thus providing a relatively simpler sensor configuration when compared with the case of distributed transmitters and receivers. 

However, spaceborne distributed SAR can offer a number of potential technical benefits, such as very-high azimuth resolution exploiting waveform diversity, single pass interferometry and tomography, 3D target point cloud generation, advanced clutter removal for Ground Moving Target Indication, etc., all available in any weather and light condition. 

These are some of the reasons why the UK MOD is investigating this concept with the project Oberon, and this project fits within the remits of Dstl to address technical challenges related to this peculiar sensing configuration. 

Indeed, several challenges need to be addressed in order to make possible the integration of the multistatic SAR return in the formation of a coherently integrated radar image products that could be then fed to state of the art information/intelligence extraction algorithms. 

For example, assuming a sufficiently accurate synchronization between the local oscillators on board of the platforms the coherent integration challenge translates in a phase centre estimation problem, meaning that it is fundamental to have accurate knowledge of the instantaneous baseline. This is not always possible and advanced signal processing are required to compensate in uncertainties in the 3D baseline between a Tx/Rx pairs. 

This project will investigate state of the art techniques in radar signal processing to develop accurate multi-static SAR imaging. The techniques investigated will be applied in different domains, such as: raw, range compressed and range-Doppler in order to identify the best solution to estimate baseline errors to be compensated in order to make the imaging coherent. Example of techniques considered will be Digital Beamforming, Generalized MIMO ambiguity function [1] , sub-sample registration techniques in the range-time domain [2], fine bistatic Doppler estimation using Fractional Fourier transform and Chebyshev based bistatic phase history [3]. The output of this project will be a framework able to integrate a number of solutions applicable to different of imaging modes of the multi-static SAR system. 

The aim of the project is to develop a robust processing framework applicable to multi static distributed SAR; the aim will be achieved through the following objectives:

• Development of a signal model for space borne multistatic SAR ; 
• Understanding of practical uncertainty in orbit control and formation flying challenges and their effect on the radar signal; 
• Simulation of space borne SAR system; 
• Developing of processing framework based on MIMO ambiguity function and post processing refinement techniques based on sub-sample registration. 
• Development of proof of concept imaging demonstrator in Inverse SAR equivalent scenario to demonstrate algorithms capabilities. 

References

[1] C. V. Ilioudis, C. Clemente, I. K. Proudler and J. Soraghan, "Generalized Ambiguity Function for MIMO Radar Systems," in IEEE Transactions on Aerospace and Electronic Systems, vol. 55, no. 6, pp. 2629-2646, Dec. 2019, doi: 10.1109/TAES.2019.2907390.

[2] L. Pallotta, G. Giunta and C. Clemente, "SAR Image Registration in the Presence of Rotation and Translation: A Constrained Least Squares Approach," in IEEE Geoscience and Remote Sensing Letters, vol. 18, no. 9, pp. 1595-1599, Sept. 2021, doi: 10.1109/LGRS.2020.3005198.

[3] C. Clemente and J. J. Soraghan, "Approximation of the Bistatic Slant Range Using Chebyshev Polynomials," in IEEE Geoscience and Remote Sensing Letters, vol. 9, no. 4, pp. 682-686, July 2012, doi: 10.1109/LGRS.2011.2178812.