Universal Signal Matching for RF Threat Classification

The U.S. Navy uses digital wideband electronic warfare (EW) receivers for RF threat warning and cueing of countermeasures to protect Navy aircraft. The receivers accomplish these functions by processing RF waveforms from enemy radars, extracting pulse parameters, and conducting emitter identification. The receiver''s ability to accurately identify the emitter type (necessary for warning and countermeasure cueing) depends completely on its ability to form accurate parameter estimates from waveform pulse trains. In particular, the radio frequency (RF) and pulse width (PW) parameters are typically measured directly, but the pulse repetition interval (PRI) must be derived based on the output of a deinterleaving algorithm. This algorithm searches through the pulse input buffer associating pulses together using a variety of matching techniques. Missing pulses, receiver blanking, and mis-associations can cause the PRI estimate produced by the deinterleaver to be incorrect. Corruption of the PRI estimate is the primary source of the emitter identification problems. To deal with emitter identification issues, Navy EW engineers modify the emitter identification (EID) table that is used with EW receivers to accommodate known failure modes of the deinterleaving algorithm. However, doing so increases work load and leads to sub-optimal receiver performance in certain conditions. To mitigate the need for human intervention, a new approach is needed for EW receiver processing. The objective of this project is to develop a new corrupted pulse train resolution algorithm that will process the data produced by the deinterleaving algorithm and attempt to identify corruptions in the pulse train data that could lead to an incorrect identification. In taking this approach, knowledge of the scenarios that cause corrupted pulse trains to be generated by the deinterleaver can be embedded in the algorithm instead of being encoded in the EID table. Thus, the receiver will be able to maintain a universal table, and the need for adapting the EID table to specific missions will be mitigated.