The practical implementation of a retrodirective cross-eye jammer by using software defined radio (SDR)

Abstract

Radar-guided missiles have the potential to cause extreme damage to vital military assets. Although traditional deception techniques can deceive radars in range and Doppler shift, only a few methods can deceive them in angle. Cross-eye jamming was identified as a possible countermeasure against angular radar threats. This electronic attack (EA) method works by artificially creating the worst case of glint in angular radars.

Numerous analyses of cross-eye jamming exist in the literature. The earlier analyses were derivative glint analyses that made two incorrect assumptions. The first was to use linear fits to the monopulse antenna patterns, which is only valid when the target platform is on broadside of the radar. The second was to assume that the target platform is an infinite distance from the radar, which is not possible. The analyses also did not consider retrodirectivity. It was only during a later cross-eye jamming analysis that the limitations were identified and corrected. The limitations in the analysis could have been identified much sooner if practical measurements were made. The extended cross-eye jamming analysis made fewer assumptions and was proven accurate by numerous simulations and some experimental results. However, the only available experiments where the radar rotation was considered did not implement true retrodirectivity but simulated it by combining isolated channel measurements. A need was identified for the development of a truly-retrodirective cross-eye jammer in a laboratory environment to expand the body of knowledge available about cross-eye jamming. The cost-effective jammer would be used to identify any real-world effects or anomalies that could not be predicted by the extended analysis or identified by simulation.

This dissertation presents the development of a truly-retrodirective cross-eye jammer by using a software-defined radio (SDR). The development is accompanied by a method of calibrating the cross-eye jammer to obtain the ideal magnitude factor and phase difference between the retrodirective paths by minimising the magnitude of the sum-channel return of a monopulse radar. The developed system was tested in an anechoic environment against a self-implemented phase-comparison monopulse radar. It was shown that significant angular errors could be induced. The angular errors were larger than 10° at broadside of the radar. This equated to a minimum miss-distance of around 1 m at a range of 6 m. It was shown that a cross-eye gain of around ten was obtained, which resulted in the indicated angle of the radar never becoming zero, regardless of the radar rotation. This suggested that tracking radars, such as that used by active homing missiles, would lose lock on the target platform. Further experiments also proved the jammer to be retrodirective, with large angular errors for all rotations of the jammer antennas. All results correlated very well with that predicted by the extended analysis, with only minor deviations between radar rotations of 0° and 5°. After further investigation, it was concluded that the deviations were most likely caused by mutual coupling between the radar antennas and were not caused by a reduction in the performance of the jammer.