Home > Generic Ground Clutter Simulation for Radar Testing and Evaluation

Generic Ground Clutter Simulation for Radar Testing and Evaluation


Jurgen Strydom Jurgen Strydom

Click here to read the interview

Jurgen Strydom works at the CSIR in Pretoria, within the Radar and Electronic Warfare group, which is part of the department of Defence, Peace, Safety and Security. He obtained a BEng (Electronic Engineering) from the University of Pretoria in 2009, while completing an internship at the CSIR the same year.

He joined the CSIR as a full time employee from the start of 2010 and has since grown to the level of Senior Engineer within the Experimental Electronic Warfare team.

He enrolled as a part-time student for the Radar Masters programme during 2011, and graduated from the University of Cape Town with a MEng (Radar and Electronic Defence) the following year.

Click on the photo to read an interview with him, which is part of the ‘Meet our Alumni‘ series.



Strydom, Johannes Jurgen. Generic Ground Clutter Simulation for Radar Testing and Evaluation. MEng (specialising in Radar and Electronic Defence) Dissertation. Department of Electrical Engineering, University of Cape Town, 2012.



Download PDF of Dissertation

Hardware in the loop (HIL) simulation is often used to reduce the cost of evaluating the performance of a radar [1]. These hardware in the loop simulators aim to recreate an accurate representation of the environment of a radar. Traditionally radar environment simulators focused on single scatterer targets, and electronic countermeasures, but as the fidelity of the radar increases these simulation methods alone become less useful. Thus the fidelity of radar environment simulators has to be increased if they are to remain relevant. Targets modelled using multiple complex scatterers have been incorporated into HIL simulators, and radar clutter simulation is the next step in this process of increasing fidelity, which is the focus of this study.

The design challenge is to balance the contradictory requirements of complexity and fidelity of a clutter simulation system. The fidelity has to be high enough to accurately simulate cutter with realistic properties so as to be indistinguishable from true clutter returns for the radar under test, but the complexity has to be low enough to execute the clutter simulation on a field programmable gate array (FPGA) platform in real time. Before the FPGA simulation can be performed, the algorithms have to be designed and tested in software. This project aims to investigate the real time approximation of relevant distributions that should be used when testing and evaluating a radar in clutter scenarios. Simulation of these relevant distributions are investigated to determine their accuracy.

First a literature survey is presented to analyse the aspects of clutter that should be considered for incorporation into a radar environment. The most commonly used statistical models for surface clutter were chosen and methods for their simulation from literature were identified that could potentially be implemented on an FPGA. The actual FPGA implementation did not form part of this project, but was a constraint during the evaluation of the methods. The methods identified were implemented in software for analysis. The probability density function (PDF), spectral shape, autocorrelation function, and higher order moments were analysed.

Furthermore, the trends pertaining to the effects of correlation in the samples, and the effects of varying the most important distribution parameters were analysed. A new approach to simulate the statistics of radar clutter was then developed to answer the question of whether the methods identified in literature could be improved upon. The new approach was tested and compared to the methods in literature, and a path for possible future research in this regard emerged.