System and method for detecting and de-interleaving radar emitters
View Patent ↗A method of classifying radar emitters includes the steps of: (a) receiving pulses from multiple radar emitters; (b) categorizing received pulses based on pulse data descriptors (PDWs); (c) forming clusters of received pulses based on the PDWs; and (d) de-interleaving pulses from the cluster using one of a differential time of arrival (DTOA) histogram technique, a spectrum estimation technique, or a Hough transform technique. Step (a) includes receiving the pulses during a predetermined time interval and storing the received pulses as a snapshot representing the pulses present during the predetermined time interval.
1. A method of classifying radar emitters comprising the steps of:
(a) receiving pulses from multiple radar emitters;
(b) categorizing received pulses based on pulse data descriptors (PDWs);
(c) forming clusters of received pulses based on the PDWs; and
(d) extracting pulses from at least one of the formed clusters, based on pulse repetition intervals (PRIs) between the extracted pulses, to form a pulse train belonging to a radar emitter;
wherein forming clusters includes sorting the PDWs based on each PDW's Euclidean distance from the center of a respective cluster.
2. The method of claim 1 wherein
step (d) includes de-interleaving pulses from the cluster using a differential time of arrival (DTOA) histogram technique.
3. The method of claim 2 including
subtracting a TOA of each pulse from every subsequent time of arrival (TOA) of other pulses to form differential TOAs, and
accumulating the differential TOAs which are substantially similar to each other into respective bins to form the DTOA histogram.
4. The method of claim 3 including
correcting the DTOA histogram by removing an accumulated DTOA formed in a second bin that is a harmonic of an accumulated DTOA formed in a first bin, the first bin having a lower PRI than the second bin.
5. The method of claim 1 wherein
step (a) includes receiving the pulses during a predetermined time interval and storing the received pulses as a snapshot representing the pulses present during the predetermined time interval.
6. The method of claim 1 wherein
step (d) includes de-interleaving pulses from the cluster using a spectrum estimation technique.
7. The method of claim 6 including
denoting a plurality of pulse trains in the cluster as a complex signal;
forming a discrete Fourier transform (DFT) of the complex signal; and
determining a number for the plurality of pulse trains and a frequency of each pulse train.
8. The method of claim 7 including
removing a pulse train from the plurality of pulse trains in the cluster, when a frequency of the pulse train is a harmonic of another frequency of another pulse train.
9. The method of claim 1 wherein
step (d) includes de-interleaving pulses from the cluster using a Hough transform technique.
10. The method of claim 9 including
calculating a straight line, based on TOA of pulses from the cluster versus TOA modulo by τ, where τ is an arbitrary window width;
determining a slope of the straight line; and
calculating a pulse repetition interval (PRI) of a pulse train in the cluster based on the slope of the straight line.
11. A method of classifying radar emitters comprising the steps of:
(a) receiving pulses from multiple radar emitters;
(b) categorizing received pulses based on pulse data descriptors (PDWs);
(c) forming clusters of received pulses based on the PDWs; and
(d) de-interleaving pulses from the cluster using one of a differential time of arrival (DTOA) histogram technique, a spectrum estimation technique, or a Hough transform technique;
wherein forming clusters includes sorting the PDWs based on each PDW's Euclidean distance from the center of a respective cluster.
12. The method of claim 11 wherein
step (a) includes receiving the pulses during a predetermined time interval and storing the received pulses as a snapshot representing the pulses present during the predetermined time interval.
13. The method of claim 12 wherein
the predetermined time interval varies between 4 msec. and 800 msec.
14. The method of claim 12 wherein
the PDWs include at least radio frequency (RF) data and pulse width (PW) data for each received pulse.
15. The method of claim 14 wherein
the PDWs includes pulse amplitude (PA).
16. The method of claim 11 wherein
step (b) includes categorizing the received pulses based on vectors, each vector comprised of a sum of weighted PDWs.
17. A system for classifying radar emitters comprising
a receiver for receiving a plurality of signals from radar emitters,
a processor coupled to the receiver for
(a) generating data components for each signal received from the radar emitters,
(b) forming multi-dimensional samples from the generated data components,
(c) sorting the multi-dimensional samples into a plurality of data clusters, based on their respective proximity to a center of each data cluster, and
(d) extracting pulses from at least one of the formed clusters, based on pulse repetition intervals (PRIs) between the extracted pulses, to form a pulse train belonging to a radar emitter,
wherein the processor is configured to sort the multi-dimensional samples based on each sample's Euclidean distance from the center of a respective cluster.
18. The system of claim 17 wherein the processor is configured for de-interleaving pulses from the at least one cluster using one of a differential time of arrival (DTOA) histogram technique, a spectrum estimation technique, or a Hough transform technique.
19. The system of claim 17 including
a storage device for storing the received signals as a snapshot representing received signals present during a predetermined time interval,
wherein the processor is configured to generate data components for each of the received signal stored as a snapshot.