DEVICE & METHOD FOR COGNITIVE RADAR INFORMATION NETWORK

• US 20140097979A1
• Filed: 10/09/2012
• Published: 04/10/2014
• Est. Priority Date: 10/09/2012
• Status: Active Grant

• Current Assignee: Accipter Radar Technologies Inc.
• Original Assignee: Accipiter Radar Technologies Inc.
• Inventors: Nohara, Timothy J., Haykin, Simon
• Granted Patent: US 8,860,602 B2
• Time in Patent Office: Days
• US Class Current: 342/90
• CPC Class Codes: G01S 13/538 eliminating objects that have not moved between successive antenna scans, e.g. area MTi
• CPC Class Codes: G01S 13/87 Combinations of radar systems, e.g. primary radar and secondary radar
• CPC Class Codes: G01S 13/95 for meteorological use
• CPC Class Codes: G01S 7/003 Transmission of data between radar, sonar or lidar systems and remote stations
• CPC Class Codes: G01S 7/295 Means for transforming co-ordinates or for evaluating data, e.g. using computers
• CPC Class Codes: Y02A 90/10 Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Abstract

In cognitive radar information networks (CRINs) human-like cognitive abilities of attention and intelligence are built into radar systems and radar information networks (RINS) to assist operators with information overload. A CRIN comprises a plurality of radar sensing nodes monitoring an environment, a repository or memory, and a cognitive radar controller. Each radar sensing node includes a radio frequency transmitter, a transmitting antenna, and a receiver and receiving antenna. The receiver includes a digital radar processor for generating receiver information from the received echoes about the environment. The repository is configured for receiving and storing the receiver information generated by the digital radar processor. The cognitive controller is configured to automatically focus the system'"'"'s attention on a region of interest within the surveillance volume in response to an attention request, by selecting the transmitter'"'"'s waveform, selecting the receiver'"'"'s processing mode, and controlling the transmitter'"'"'s antenna. The cognitive controller learns from the environment by exploiting the repository'"'"'s historical receiver information and further learns from the consequences of its past decision.

Claims

1. A cognitive radar information network surveillance system monitoring a multi-target environment, said system comprising at least one said radar sensing node, a repository, and a cognitive radar controller, said radar sensing node comprising:

a radar transmitter for illuminating a respective surveillance volume of said environment with a radio frequency waveform emitted through a transmitting antenna operatively connected to said transmitter; and

a receiver operatively connected to a receiving antenna for receiving echoes from multiple targets and clutter in said environment in response to said emitted waveform, said receiver comprising a digital radar processor for generating receiver information about said environment;

said repository being operatively connected to said receiver for receiving and storing said receiver information, the stored receiver information being accessible by said cognitive controller as historical receiver information;

to automatically focus system attention on a region contained within a combined surveillance volume in response to an attention request for improved surveillance performance in said region, said cognitive controller being configured to execute at least one action taken from the group consisting of operating or energizing said transmitter in accordance with a set of selected illumination or transmit-waveform parameters, operating said receiver under a selected processing mode, and controlling said transmitting antenna;

wherein said cognitive controller is further configured to make selections of radar control parameters including illumination or transmit-waveform parameters, receiver processing modes, and antenna control parameters in accordance with consequences of past decisions as to radar control parameters, stored as additional said historical receiver information in said repository, said consequences comprising effects on surveillance performance figures of merit in said region.

2. The system defined in claim 1 wherein said repository is configured to provide some of said receiver information to an operator in real-time.

3. The system defined in claim 1 wherein said region is one or more cells selected from an arrangement of predefined attention cells, said predefined attention cells covering said combined surveillance volume.

4. The system defined in claim 1 wherein said region is defined in relation to a target of interest.

5. The system defined in claim 4 wherein said target of interest is dynamic and said region moves with said target of interest.

6. The system defined in claim 1 wherein said region is defined in relation to an environmental disturbance such as weather.

7. The system defined in claim 1 wherein said region is defined in relation to geometries of interest taken from the set including borders and exclusion zones.

8. The system defined in claim 1, further comprising an automatic behavior and analysis detection processor operatively connected to said cognitive controller, said attention request being initiated by an event taken from the group consisting of an operator request or an automatic request from said automatic behavior and analysis detection processor in response to the detection of suspicious target behavior.

9. The system defined in claim 8 wherein said automatic behavior and analysis detection processor includes generic digital processing circuits modified by software computing algorithms taken from the group consisting of fuzzy logic, neural networks and probabilistic reasoning to detect suspicious target behavior.

10. The system defined in claim 1 wherein said improved surveillance performance figures of merit includes increased target detection sensitivity over said region.

11. The system defined in claim 10 wherein said cognitive controller, said transmitter, said receiver and said transmitting antenna are configured to maintain overall surveillance performance across said combined surveillance volume within acceptable limits while said cognitive controller operates to improve performance in said region.

12. The system defined in claim 1 wherein said use of historical receiver information includes the use of adaptive target maps based on targets of opportunity observed by said system in said environment.

13. The system defined in claim 1 wherein said cognitive controller is configured to execute a perception-action cycle at a clock rate determined by a master clock of said cognitive radar information network, said master clock being operatively connected to said receiver and said repository for aligning in time said receiver information with receiver information from other radar sensing nodes in said network, further comprising a feedback path between said receiver and said cognitive controller effected by an operative connection between said cognitive controller and said repository, thereby enabling an execution of said perception-action cycle.

14. The system defined in claim 13 wherein said feedback path is based on the entropic state derived from the multiple tracked targets tracked by said system.

15. The system in claim 1 wherein said at least one radar sensing node is one of a plurality of radar sensing nodes each taken from the group consisting of fixed, surface-mounted marine radars, agile radars, ship-based radars, mobile radars, aerostat radars, air traffic control radars, and military radars.

16. A method for intelligently focusing the attention of a radar surveillance system on a region of interest contained in a surveillance volume, comprising:

operating a cognitive radar controller to receive and process an attention request, the processing of said attention request including selecting at least one set of radar operating parameters taken from the group consisting of illumination or transmit-waveform parameters, receiver processing mode parameters, and transmitting antenna control parameters; and

providing the selected set of operating parameters to at least one respective selected radar sensing node of a plurality of radar sensing nodes in said surveillance system so that said selected radar sensing node operates in accordance with said selected set of operating parameters, the selecting of said at least one set of radar operating parameters including operating said cognitive radar controller to consult or access a repository containing stored historical target information entailing results of radar scans in accordance with previous selected sets of radar operating parameters, thus enabling a learning from the environment and consequences of past radar operating decisions.

17. The method defined in claim 16, further comprising accessing said repository to obtain some of said receiver information in real-time.

18. The method defined in claim 16 wherein said region of interest is one or more attention cells selected from an arrangement of predefined cells in said surveillance volume, said predefined cells collectively covering said combined surveillance volume.

19. The method defined in claim 16 wherein said region of interest is defined in relation to a target of interest identified as such by predetermined behavioral indicia.

20. The method defined in claim 19 wherein said target of interest is dynamic and said region moves with said target of interest.

21. The method defined in claim 16 wherein said region is defined in relation to an environmental disturbance such as weather.

22. The method defined in claim 16 wherein said region is defined in relation to geometries of interest taken from the set including borders and exclusion zones.

23. The method defined in claim 16, further comprising automatically analyzing behaviors of detected targets and thereby detecting suspicious behaviors in accordance with predetermined indicia.

24. The method defined in claim 16 wherein said attention request is initiated by an event taken from the group consisting of an operator request or an automatic request in response to a detection of suspicious target behavior.

25. The method defined in claim 24 wherein the automatic analyzing of behaviors of detected targets includes generic operating digital processing circuits modified by software computing algorithms taken from the group consisting of fuzzy logic, neural networks and probabilistic reasoning to detect suspicious target behavior.

26. The method defined in claim 16 wherein said improved surveillance performance includes increased target detection sensitivity over said region of interest.

27. The method defined in claim 26, further comprising maintaining overall performance across said combined surveillance volume within acceptable limits while improving performance in said region of interest.

28. The method defined in claim 16 wherein the operating of said cognitive radar controller to consult or access said repository for stored historical target information includes using adaptive target maps based on targets of opportunity observed by said method in said environment.

29. The method defined in claim 16, further comprising operating said cognitive controller to execute a perception-action cycle at a clock rate determined by a master clock of a cognitive radar information network, said radar sensing node being part of said cognitive radar information network, said master clock being operatively connected to a receiver of said radar sensing node and to said repository for aligning in time said receiver information with receiver information from other radar sensing nodes in said cognitive radar information network, operating said cognitive controller to execute a perception-action cycle including transmitting radar information along a feedback path between said receiver and said cognitive controller effected by an operative connection between said cognitive controller and said repository, thereby enabling an execution of said perception-action cycle.

30. The method defined in claim 29 wherein said feedback path is based on the entropic state derived from the multiple tracked targets tracked by said method.

31. The method in claim 16 wherein said at least one radar sensing node is one of a plurality of radar sensing nodes each taken from the group consisting of fixed, surface-mounted marine radars, agile radars, ship-based radars, mobile radars, aerostat radars, air traffic control radars, and military radars.

32. A method for controlling a surveillance radar system including at least one radar node, comprising:

storing, in a repository or database, radar decisions previously taken in response to respective sets of target attributes, each of which characterizes the nature of targets of interest requiring attention for a particular region contained within the combined surveillance volume of said radar system, each said radar decision taking the form of selection of radar control parameters including illumination parameter vectors and receiver processing mode parameters;

storing in said repository or database said sets of target attributes;

storing, in said repository or database, respective consequences of said decisions as to radar control parameters used in monitoring said targets of interest, said consequences comprising effects on surveillance performance figures of merit for said targets of interest by said system in said region;

receiving radar data from said at least one radar node and generating target information from said radar data and storing said target information in said repository or database;

processing said stored target information to generate adaptive target maps, said adaptive target maps organized as to said radar control parameters;

storing said adaptive target maps in said repository or database;

accessing said adaptive target maps in said repository or database;

based on said adaptive target maps, said decisions, and said consequences in said repository or database, and in response to a current set of target attributes for said region, selecting an illumination parameter vector and a set of receiver processing mode parameters for at least one selected radar node;

sending the selected illumination parameter vector and the selected set of receiver processing mode parameters to said at least one selected radar node for modifying the operation thereof.

33. The method defined in claim 32, further comprising filtering the adaptive target maps into subgroups in accordance with said sets of target attributes.

34. The method defined in claim 32 wherein said selected target attributes are taken from the group consisting of speed, direction, acceleration, and radar cross section.

35. The method defined in claim 32 wherein the processing of said stored target information to generate said adaptive target maps includes extracting, from said repository or database, recent targets of opportunity associated with current transmitter and receiver modes to update the adaptive target maps.

36. The method defined in claim 32 wherein the update rate(s) of adaptive target maps is selected in accordance with use requirements.

37. A method for controlling a radar network, comprising:

receiving radar data over a network from multiple radar nodes distributed throughout a region;

processing the radar data to detect a suspicious target or an environmental disturbance and to characterize said target or environmental disturbance with a set of attributes;

defining one or more volumetric attention cells in said region;

determining which of said attention cells are associated with said suspicious target or environmental disturbance;

accessing a database or repository of past radar control parameter vectors, each previously selected and used in response to past said respective set of attributes, each of said radar control parameter vectors including an illumination parameter vector and receiver processing mode parameter vector for at least one radar node;

selecting from said database or repository an illumination parameter vector and a receiver processing mode parameter vector for at least one radar node from among said multiple radar nodes in said region, wherein said attention cells determined to be associated with said suspicious target or environmental disturbance are located within a surveillance volume of said at least one radar node; and

sending the selected illumination parameter vector and the selected receiver processing mode parameter vector to said at least one radar node for modifying the operation thereof so as to ultimately improve radar surveillance performance in said attention cells pertaining to said suspicious target or environmental disturbance.

38. The method defined in claim 37 wherein said attention cells are defined after identifying the attributes of said suspicious target or environmental disturbance.

39. The method defined in claim 37 wherein said attention cells are predefined, prior to processing the radar data to identify attributes of said suspicious target or said environmental disturbance.

40. A method for controlling a radar operation including successive perception-action cycles, said method comprising:

during each said perception-action cycle;

energizing a radar transmitter to transmit an electromagnetic waveform to illuminate an environment;

activating an associated receiver to (a) make measurements of the environment from incoming electromagnetic echoes generated by the environment in response to the transmitted waveform and (b) provide feedback information about the environment at least indirectly back to said transmitter, said feedback information based on the multi-target entropic state calculated from the track state-estimation error information associated with the targets of opportunity tracked in the surveillance volume of said radar; and

(c) adaptively selecting in accordance with prior radar scan results, a new and potentially different illumination to energize said transmitter during a succeeding perception-action cycle.

41. The method defined in claim 40, further comprising automatically storing, in a repository or database, radar scan results including said measurements and data characterizing the transmitted electromagnetic waveform, the adaptive selecting of said new and potentially different illumination including operating a cognitive controller operatively connected to said repository or database.

42. The method defined in claim 40 wherein the activating of said associated receiver includes operating said associated receiver in accordance with a set of receiver processing mode parameters, further comprising adaptively selecting, in accordance with said prior radar scan results, a new and potentially different set of receiver processing mode parameters with which to activate said receiver during said succeeding perception-action cycle.

43. A method for controlling a radar surveillance network consisting of multiple radar nodes, comprising:

dividing up a combined surveillance volume of said network into a number of predefined attention cells or regions for applying attention;

detecting a situation of interest such as a suspicious target, environmental disturbance or area of interest, said situation of interest being associated with at least one of said attention cells;

determining a subset of radar nodes whose respective surveillance volumes intersect with said at least one of said attention cells;

deciding to focus said network'"'"'s attention on said at least one of said attention cells;

focusing attention on said at least one of said attention cells by selecting a new radar parameter vector for use by at least one of said subset of radar nodes;

using said selected new radar parameter vector to reconfigure a transmitter and a receiver of said at least one of said subset of radar nodes so as to improve performance in said at least one of said attention cells;

said selecting of said new radar parameter vector determined adaptively by learning from past selections made in response to similar, previous situations of interest and their respective impact on surveillance performance in the vicinity of said at least one of said attention cells.