Using a hybrid data generator for testing of ABF-algorithms

Abstract

Testing of Algorithms for Adaptive Beamforming (ABF) is always a critical issue. With flight trials data, the conditions and parameters for targets, jammers and clutter are often not well known. In contrast, with simulated data, all parameters and conditions are well defined. However, the value of simulated data in this field is often poor due to the absence of real-world effects such as non-linearities. Especially for evaluation of algorithms for airborne radars, the use of flight trials data with various scenarios is essential. The disadvantage of using flight trials data is that, in most cases, the performance improvement when applying the algorithms is not unambiguously evident. For these reasons, a hybrid data generator has been developed which can combine measured data from flight trials with simulated data from complicated target or jammer manoeuvers. The measured data includes clutter and noise signals as well as targets of opportunity and can also contain different types of jammer signals. The synthetic target generator (STG) is able to simulate targets as well as CW and broadband noise jammers. The advantage of the hybrid data generator is that the Signal-to-Noise and Jammer-to-Noise ratios of synthetic targets and jammers can be exactly adapted to the measured data. The quality criterion of ABF algorithms for noise jammers is the so called burn-through range [2]. For CW jammers it is easy to evaluate the jamming signal reduction after applying the ABF algorithms. This paper is not concerned with the different ABF algorithms, it only illustrates the method of evaluating the numerous algorithms. Further, the influence of non-linear effects are theoretically described in [3].