Multi-Layered Interval Search Tree Approach on Radar Identification

Radar identification algorithms used in Electronic Warfare (EW) systems try to match the digitized parameters of radar signal waves with radars data in a pre-prepared mission data file (MDF). In this way, critical information that cannot be understood from the signal alone, such as who the signal belongs to, whether the source of the signal is friendly or hostile, can be obtained from the MDF through past experience and observations. As the number and diversity of active radars and radars in the MDF increase, the number of comparisons, so, the time spent for comparison increases. For time-critical EW systems, the tolerable comparison time is very short. The performance of these systems may also depend on the accuracy of the identification algorithm. In this work, the information about the radars in the MDFs is represented by a multi-layer, as balanced as possible, interval tree structure. The goal is to find the radars in the MDFs that match the incoming signal accurately, while making the comparison process fast. The proposed structures are compared with brute-force and improved brute-force algorithms. Experiments are conducted in terms of performance, memory and accuracy. For a MDF with about 10000 radar modes, the tree methods used up to 50% more memory. On the other hand, they found the correct radars in the MDF about 14 times faster than the improved brute force algorithm.