Adaptive dynamic cluster deinterleaving

• US 9,229,095 B2
• Filed: 10/28/2013
• Issued: 01/05/2016
• Est. Priority Date: 10/28/2013
• Status: Active Grant

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Hammack, George D.
• Primary Examiner(s): Gregory, Bernarr
• Application Number:
  US14/064,566
• Publication Number:
  US 20150116141A1
• Time in Patent Office: 799 Days
• Field of Search: 342 89-103, 342/118, 342/134, 342/135, 342/165, 342173-175, 342/195, 342 13- 20, 342/42, 342/46, 342/47, 342145-147, 342/350, 342378-380, 708/100, 708/200, 708/212, 706/15, 706/16, 706/20, 702/127, 702/189, 702/190
• US Class Current: 1/1
• CPC Class Codes: G01S 7/021 Auxiliary means for detecting or identifying radar signals or the like, e.g. radar jamming signals
• CPC Class Codes: G01S 7/285 Receivers
• CPC Class Codes: G01S 7/2923 based on data belonging to a number of consecutive radar periods

Abstract

Described herein are methods and systems capable of dynamically adapting weights in response to a received stream of pulses by deinterleaving a stream of pulses according to an initial weighted distance, adjusting the weighted distance, and deinterleaving the stream of pulses according to the adjusted weighted distance.

Claims

1. A method for adaptive dynamic cluster deinterleaving of received radar pulses in a radar system, the method comprising:

deinterleaving a stream of pulses received by a radar receiver by determining whether each of a plurality of received pulses is a member of one or more clusters, each cluster having an anchor point, each received pulse and the anchor point having a respective feature vector, each feature vector having at least two features, wherein the at least two features are selected from at least two measured features of each received pulse, the determining performed by;

generating at least one weight corresponding to each of the at least two features;

calculating, using the at least one weight, a weighted distance between the feature vector of each received pulse and the feature vector of the anchor point;

comparing the weighted distance to a distance threshold, and;

identifying the received pulse as a member of one of the one or more clusters if the weighted distance is less than the distance threshold;

adjusting the at least one weight corresponding to one or more of the at least two features by;

selecting one or more reference features from the at least two measured features;

acquiring, for each cluster, a measured value of the one or more reference features for each member of the cluster;

calculating, for each cluster, a variance of the measured values of each member of the cluster;

comparing the variance of the measured values to a threshold variance, and;

increasing at least one of the at least one weights if the variance is less than the threshold variance or decreasing at least one of the at least one weights if the variance is greater than the threshold variance to generate at least one adjusted weight; and

deinterleaving the stream of pulses using the at least one adjusted weight by determining whether each of the plurality of received pulses is a member of one or more adjusted clusters, each adjusted cluster having an adjusted anchor point, each received pulse and the adjusted anchor point having a respective adjusted feature vector, each adjusted feature vector having at least two adjusted features correlating to the at least one adjusted weight, the determining performed by;

calculating, using the at least one adjusted weight, an adjusted weighted distance between the adjusted feature vector of the received pulse and the adjusted feature vector of the adjusted anchor point;

comparing the adjusted weighted distance to an adjusted distance threshold, and;

identifying the received pulse as a member of the one or more adjusted clusters if the adjusted weighted distance is less than the adjusted distance threshold.

2. The method of claim 1, further comprising iterating, at least once, the step of adjusting the at least one weight.

3. The method of claim 2, further comprising iterating, at least once, the step of deinterleaving the stream of pulses using the at least one adjusted weight.

4. The method of claim 1, further comprising identifying, using a cognitive threat matching system, one or more radar transmitters associated with one or more values of at least one of the measured features of the members of at least one of the adjusted clusters.

5. The method of claim 1, wherein the step of generating at least one weight further comprises providing weights for the at least two features according to known radar transmitter operating parameters.

6. The method of claim 1, wherein the step of generating at least one weight further comprises randomly selecting weights for each of at least two features.

7. The method of claim 1, wherein the weighted distance is calculated in accordance with d(x,y)=√

{square root over ((x−

y)^TB(x−

y))}{square root over ((x−

y)^TB(x−

y))}, wherein x is the feature vector of a particular received pulse, y is the feature vector of the anchor point, and B is a weighting matrix.

8. A method for adaptive dynamic cluster deinterleaving of received radar pulses in a radar system, the method comprising:

deinterleaving a stream of pulses received by a radar receiver by determining whether each of a plurality of received pulses is a member of one or more clusters, each cluster having an anchor point, each received pulse and the anchor point having a respective feature vector, each feature vector having at least two features, wherein the at least two features are selected from at least two measured features of each received pulse, the determining performed by;

generating at least one weight corresponding to each of the at least two features;

calculating, using the at least one weight, a weighted distance between the feature vector of each received pulse and the feature vector of the anchor point;

comparing the weighted distance to a distance threshold, and;

identifying the received pulse as a member of one of the one or more clusters if the weighted distance is less than the distance threshold;

adjusting the at least one weight corresponding to one or more of the at least two features by;

selecting one or more reference features from the at least two measured features;

acquiring, for each cluster, a measured value of the one or more reference features for each member of the cluster;

calculating, for each cluster, a variance of the measured values of each member of the cluster;

comparing the variance of the measured values to a threshold variance, and;

decreasing at least one of the at least one weights if the variance is greater than the threshold variance to generate at least one adjusted weight;

removing, for each cluster having a variance less than the threshold variance, each member of the cluster from the stream of pulses to generate an adjusted stream of pulses; and

deinterleaving the adjusted stream of pulses using the at least one adjusted weight by determining whether each of a plurality of adjusted pulses is a member of one or more adjusted clusters, each adjusted cluster having an adjusted anchor point, each adjusted pulse and the adjusted anchor point having a respective adjusted feature vector, each adjusted feature vector having at least two adjusted features correlating to the at least one adjusted weight, the determining performed by;

calculating, using the at least one adjusted weight, an adjusted weighted distance between the adjusted feature vector of the adjusted pulse and the adjusted feature vector of the adjusted anchor point;

comparing the adjusted weighted distance to an adjusted distance threshold, and;

identifying the adjusted pulse as a member of the one or more adjusted clusters if the adjusted weighted distance is less than the adjusted distance threshold.

9. The method of claim 8, further comprising iterating, at least once, the step of adjusting the at least one weight.

10. The method of claim 9, further comprising iterating, at least once, the step of deinterleaving the adjusted stream of pulses using the at least one adjusted weight.

11. A system for adaptive dynamic cluster deinterleaving of received radar pulses in a radar system, comprising:

one or more processors; and

a memory, the memory including executable code representing instructions that when executed cause the one or more processors to;

deinterleave a stream of pulses received by a radar receiver by determining whether each of a plurality of received pulses is a member of one or more clusters, each cluster having an anchor point, each received pulse and the anchor point having a respective feature vector, each feature vector having at least two features, wherein the at least two features are selected from at least two measured features of each received pulse, the determining performed by;

generating at least one weight corresponding to each of the at least two features;

calculating, using the at least one weight, a weighted distance between the feature vector of each received pulse and the feature vector of the anchor point;

comparing the weighted distance to a distance threshold, and;

identifying the received pulse as a member of one of the one or more clusters if the weighted distance is less than the distance threshold;

adjust the at least one weight corresponding to one or more of the at least two features by;

selecting one or more reference features from the at least two measured features;

acquiring for each cluster, a measured value of the one or more reference features for each member of the cluster;

calculating for each cluster, a variance of the measured values of each member of the cluster;

comparing the variance of the measured values to a threshold variance, and;

increasing at least one of the at least one weights if the variance is less than the threshold variance or decreasing at least one of the at least one weights if the variance is greater than the threshold variance to generate at least one adjusted weight; and

deinterleave the stream of pulses using the at least one adjusted weight by determining whether each of the plurality of received pulses is a member of one or more adjusted clusters, each adjusted cluster having an adjusted anchor point, each received pulse and the adjusted anchor point having a respective adjusted feature vector, each adjusted feature vector having at least two adjusted features correlating to the at least one adjusted weight, the determining performed by;

calculating, using the at least one adjusted weight, an adjusted weighted distance between the adjusted feature vector of the received pulse and the adjusted feature vector of the adjusted anchor point;

comparing the adjusted weighted distance to an adjusted distance threshold, and;

identifying the received pulse as a member of the one or more adjusted clusters if the adjusted weighted distance is less than the adjusted distance threshold.

12. The system of claim 11, further comprising a radar receiver for receiving the stream of pulses.

13. The system of claim 11, further comprising a transmitter for transmitting one or more measured values of the at least two measured features of each member of each cluster and each adjusted cluster to a cognitive threat matching system.

14. The system of claim 13, wherein the cognitive threat matching system identifies one or more radar transmitters associated with one or more values of at least one of the measured features of the members of at least one of the adjusted clusters.