You are here

Fully Adaptive Radar


OBJECTIVE: Develop closed-loop radar operation concepts and analyze performance of optimal and adaptive radar from a single and distributed sensing perspective. DESCRIPTION: The concept of fully adaptive radar (FAR) seeks to exploit all available degrees-of-freedom on transmit and receive in order to maximize target detection performance. This area has received increased interest in recent times and builds upon a rich history outlined in the references and citations therein. Of key importance is the concept of closed loop radar operation via feedback from the receiver to transmitter for guiding the next illumination. This enables enhanced adaptive target detection, in computationally demanding and training data starved scenarios. The most significant Air Force benefit afforded by FAR is the ubiquitous ability of the radar to adapt to its environment as opposed to pointing the radar in a given direction and interrogating for presence of targets. Closed loop radar operation is replete with open problems spanning a broad gamut of research areas-detection, tracking, and classification from single as well as a multi-sensor perspectives. The focus of this effort is the radar target detection problem in environments with unknown spectral properties. Estimating the unknown spectral properties has been the focus of a large body of research spanning 5 decades. The underlying challenges in this context are documented in the references and citations therein. Two important issues that arise in the context of FAR for adaptive target detection include (i) Deriving the performance limit (optimal performance) afforded by FAR? (ii) Developing the criteria for adaptive processor performance to lie within a prescribed level of the optimal. A principled investigation of these issues includes the following items. A first step is the development of the feedback signal from the receiver to the transmitter via prescribed metrics such as mean squared error, entropy, or mutual information. The next step is to develop analytical as well as computer simulation methods for determining the false alarm and detection probability for the optimal processor with respect to single and multiple radar waveforms. Furthermore, due the large number of degrees of freedom, the number of unknown nuisance parameters incurs a substantial increase. Since these parameters need to be estimated from training data, the amount of training data needed to attain adaptive performance within 3 dB of the optimal processor is required. Performance analysis and validation of the adaptive technique with respect to the training data support, false alarm probability, robustness of detection performance to parameter mismatch, and computational cost is sought. Extension of this approach to handle distributed and multiple input/multiple output (MIMO) radar performance must be undertaken. Performance validation for both single and distributed radar needs to be analyzed using simulated and measured data sets. Accordingly a three phase campaign as outlined below is in order. PHASE I: Develop criteria for characterizing feedback from the receiver to transmitter. Analyze optimal closed loop radar processor detection probability performance for a fixed false alarm probability. Develop and analyze adaptive processor performance using simulated data. Key considerations are training data support and computational cost. PHASE II: Techniques developed in Phase I will be extended to MIMO radar and distributed radar configurations employing fixed and varying waveforms. A key challenge in this regard is the dependence of the training data on the geometry of the MIMO/distributed radar configuration. Compensation techniques for training data heterogeneity are needed. Adaptive processor performance must be extensively analyzed with simulated and measured data. PHASE III: Transition opportunities for this effort include RLSTAP and RAST-W upgrades within AFRL, ongoing MIMO radar programs at ESC/EN as well as JSTARS and AWACS. Commercial applications include law-enforcement and toxic waste site detection using distributed sensors. REFERENCES: 1. J.R. Guerci,"Cognitive Radar: The Knowledge-Aided Fully Adaptive Approach,"ARTECH HOUSE INC., Norwood, MA, 2010. 2. Editors: F. Gini and M. Rangaswamy,"Knowledge-Based Radar Detection, Tracking, and Classification,"Wiley INTERSCIENCE Series, May 2008. 3. M.L. Pugh and P. A. Zulch,"RLSTAP algorithm development tool for analysis of advanced signal processing techniques,"Proceedings of the 29th Asilomar conference on signals, systems, and computers, Pacific Grove, CA, November 1995.
US Flag An Official Website of the United States Government