Waveform-Agile Sensing and Processing

IEEE SIGNAL PROCESSING MAGAZINE [10] JANUARY 2009 I magine being able to recognize objects under water while swimming, just like dolphins do, or being able to locate objects while actively homing in, just like bats do. Echolocation in these mammals has evolved over millions of years. Initial returns are processed to locate and identify objects, and then the mammal adapts the time-frequency signature of the emitting waveform to the environment and hunting mechanism. Instances of waveform adaptive sensing in nature have the potential to provide insight into man-made applications such as radar and sonar. News headlines, such as “Sensors for Bat-Inspired Spy Plane Under Development ...” (“Research News,” Michigan Today, University of Michigan, March 2008) may be portents for the value of waveform agility. The narrowband ambiguity function, introduced by P.M. Woodward in his 1953 text, provides a mathematical framework for reasoning about active sensing. Essentially it describes the effect of range and Doppler on the matched filter receiver response. Woodward captures the state of knowledge in 1953 as follows: “... the basic question of what to transmit remains substantially unanswered.” He may have failed to do adequate justice to his own contributions, but certainly many advances have been made since 1953, and many of these are enabled by the mathematical framework afforded by the ambiguity function. In the last decade, the emergence of prototype radar systems equipped with highly agile, software-driven waveform generators has provided the ability to change the transmit waveform at each time step to match environments and sensing objectives. In fact, the focus of our special issue is the development of signal processing techniques that are able to take full advantage of advances in radar systems. Phased array radars and space-time adaptive processing illuminate the value of waveform diversity with respect to time, space, frequency, and polarization. Articles in this special issue exploit spatial diversity on transmit and receive, use polarization diversity to resolve close targets and separate targets from clutter, and employ waveform diversity to increase sensing performance. Since there is an infinity of possible waveforms, it becomes critical to manage complexity by developing waveform libraries and optimization methods. The optimization is needed to select the transmit waveform at each time step to reduce estimation errors, increase collected information, or schedule sensors. Articles in this special issue show that adaptation of waveforms with highly localized correlation properties or time-frequency characteristics that match echo-locating signatures can improve sensing performance and reduce sensor usage leading to greater system efficiency. The special issue begins by considering the fundamentals of waveforms and waveform libraries; it then describes the resources represented by waveform agility, design, and diversity and the available approaches to adaptive detection, estimation, and tracking; it ends with lessons learned from the bio-sonar of bats. In fact, the articles were selected in part to tell a story, and in our selection process, we avoided overlaps between articles or digressing from the story. The editors would like to acknowledge the insights they have obtained from their participation in two Department of Defense (DoD) research programs: a current multidisciplinary university research initiative (MURI) on adaptive waveform design for full spectral dominance, directed by the Air Force Office of Scientific Research (AFOSR), and the Defense Advanced Research Projects Agency (DARPA) program on waveform-agile sensing (WAS). These programs are in collaboration with the Air Force Research Laboratory (AFRL) and the Naval Research Laboratory (NRL). We now briefly preview the six articles in this special issue. The importance of waveform libraries and information measures in waveformagile sensing is discussed by Cochran, Suvorova, Howard, and Moran, in their article “Waveform Libraries.” In the context of radar scheduling, the authors exploit transmitter agility by selecting the radar signal according to operational criteria. They also present closed-loop waveform scheduling for radar, and the design of small but effective waveform libraries, using information theoretic measures to assess their utility. Good libraries are designed offline, thereby replacing the real-time challenge of waveform design with the simpler task of selecting waveforms from a library. The role of the ambiguity function (AF) in radar and its relevance to waveform agility is reviewed by Benedetto, Konstantinidis, and Rangaswamy, in their article “Phase-Coded Waveforms and Their Design.” They consider constant amplitude zero autocorrelation (CAZAC) sequences that are important in waveform design due to their optimal transmission efficiency and tight time localization properties. Discrete periodic and aperiodic AFs are used to compare different CAZACs, and the AF behavior of phase-coded waveforms is analyzed. The processing techniques Digital Object Identifier 10.1109/MSP.2008.930413