You are here

Fully Adaptive Radar Resource Allocation



OBJECTIVE: Develop techniques for radar resource allocation for closed-loop radar detection and tracking 

DESCRIPTION: Onerous challenges imposed by an A2AD environment call for closed loop radar operation for concurrent detection and tracking of targets from single and distributed radar systems. In the context of the sense-learn-adapt framework or perception-action framework, this necessitates the use of past data to determine future radar illumination and data collection. Current techniques do not optimally and automatically balance the need to detect, track, and identify targets. Recently, cognitive radar [1-3] approaches have been used to compute sensing actions that are expected to maximize the utility of the received data. Similarly, past efforts on information-theoretic sensor management [3] have produced a framework for managing the resources of an agile sensor, where the utility of the sensing action is judged by the expected amount of information flow. This effort solicits the development of smart sensor management approaches for optimized sensing in a dynamic, complicated environment characterized as containing many moving targets, performing maneuvers that are intermittently obscured to the sensor. While previous efforts have focused on portions of this problem, this topic seeks approaches that address using multiple sensors for detection, and tracking of multiple targets from single and distributed radar in a closed loop manner. Specifically, we seek approaches to capture the scene probabilistically and use this information to drive future sensing actions, and lead to quantitative improvements in performance over current approaches as measured by standard tracking benchmarks such as time until correct detection and identification, track mean square error, and optimal sub-pattern assignment (OSPA). Ideally, we seek radar resource allocation techniques that incur a weak dependence on the number of sensors and the number of targets. The approach must enable application of ideas from cognitive sensing to guide agile sensor action at the next time step and beyond, such as selection of pointing, mode, waveform [5], and PRF. 

PHASE I: Develop a closed loop sensor management framework for concurrent detection, and tracking of ground targets in a single sensor setting. A host of multi-objective optimization problems encountered in this context, need to be addressed. The approach should scale to large scenes with multiple targets exhibiting tradeoff between detection and tracking functions. Performance analysis and benchmarking of the approach are called for using standard measures. 

PHASE II: Extend the approach developed in Phase I to include distributed radars tracking multiple targets. The resulting optimization problem for resource allocation needs to be treated from an analytical standpoint to ensure that it incurs a weak dependence on the number of sensors and number of targets in a given scenario. Performance validation and comparison with other candidate methods needs to be undertaken with respect to standard metrics. Validation of the concepts developed in the effort need to be undertaken via simulation as well as experimental demonstrations. 

PHASE III: Techniques from this effort will be fundamental to the performance evaluation and benchmarking of closed loop radar detection and tracking. Transition opportunities for this effort include ongoing radar programs within AFRL. Technology insertion opportunities include platforms such as JSTARS and Global Hawk. 


1. S. Haykin, Y. Xue, and P. Setoodeh, “Cognitive Radar: Step Toward Bridging the Gap Between Neuroscience and Engineering”, The Proceedings of the IEEE, vol. 100, no. 11, pp. 3102-3130, Nov. 2012.; 2. N. Goodman, P. Venkata, and M. Neifeld, “Adaptive Waveform Design and Sequential Hypothesis Testing for Target Recognition With Active Sensors”, IEEE Journal of Selected Topics in Signal Processing, vol. 1, no. 1, pp. 105-113, June 2007.; 3. D. Fuhrmann, “Active-Testing Surveillance Systems, or, Playing Twenty Questions with Radar”, in Proc. 11th Annual Adaptive Sensor and Array Processing (ASAP) Workshop, MIT Lincoln Laboratory, Lexington, MA, Mar. 11-13, 2003.; 4. C. Kreucher, A. Hero, K. Kastella, and M. Morelande, “An Information-Based Approach to Sensor Management in Large Dynamic Networks”, The Proceedings of the IEEE, vol. 95, no. 5, pp. 978-999, 2007.

KEYWORDS: Closed Loop Radar, Resource Allocation, Distributed Radar 

US Flag An Official Website of the United States Government