Session S7

Session S7: Waveform-Agile Sensing

Time:                 Tuesday, May 13, 17:00-19:00
Chair:                 Arye Nehorai, Washington University in St. Louis
Co-Chair:          Y. Rosa Zheng, University of Missouri-Rolla

S7-1: The Ambiguity Function of Signal Coded by Finite Gabor Systems
Abdelkrim Bourouihiya Norbert Wiener Center, University of Maryland, College Park MD
In this paper, we extend the class of phase-coded waveforms to a new class of signals coded by finite Gabor systems. Our motivation is based on two facts. First, it is not difficult to compute the ambiguity function of the new signals. Second, these signals lead to some ambiguity function performances which cannot be reached by considering the class of phase coded waveforms alone. To prove this, we derive formulas allowing a global analysis of the ambiguity function of phase-coded waveforms. Then we compare ambiguity function performances for signals coded by finite Gabor systems and waveforms coded by some families of CAZACs. In addition, our analysis reveals differences between the ambiguity function behavior for these families.

S7-2: Performance of Beampattern Synthesis using High-Dimensional Vector Antenna Arrays
Jin-Jun Xiao and Arye Nehorai Washington University in St. Louis, St. Louis MO
We consider the synthesis of a spatially directional beam with full polarization control using an array of electromagnetic vector antennas (EMVA), among which each antenna consists of p orthogonal electric or magnetic dipole elements. We have shown previously that when p=2, the vector array enables polarization control of the synthesized beam while the spatial power pattern remains the same as that achieved by the scalar array. In this paper, we study the high-dimensional case when p>=3. Our results indicate that the vector antenna array with p>=3 improves the power gain of the main beam (over the sidelobes), namely the gain is shown to be linearly proportional to the vector-antenna dimension p, in addition to enabling full polarization control of the beampattern. This implies that EMVA virtually increases the array size by exploiting the full electromagnetic (EM) field components, in addition to offering the freedom to control the beampattern polarization.

S7-3: Particle Filtering Based Radar Tracking Using CAZAC Sequences and Linear Frequency Modulated Waveforms
Ioannis Kyriakides Arizona State University, Tempe AZ, Ioannis Konstantinidis Norbert Wiener Center, University of Maryland, College Park MD, Thomas Trueblood, Darryl Morrell Arizona State University, John J. Benedetto Norbert Wiener Center and Antonia Papandreou-Suppappola Arizona State University
In this paper, we apply sequential Monte Carlo methods to estimate locations in the delay-Doppler plane from where useful measurements can be extracted, based on the target position. Moreover, we enable the use of resolution cells that have the exact shape of the probability of detection contour. These methods offer an advantage over traditional radar tracking methods that require tessellating shaped resolution cells placed on a fixed grid. Additionally, we design a likelihood based tracking algorithm that is able to make effective use of the high measurement resolution Bjorck constant amplitude zero-autocorrelation (CAZAC) sequences. We demonstrate improved tracking performance when using Bjorck CAZACs over linearly modulated chirps in a single target tracking scenario.

S7-4: Coherent Clutter Suppression using Generalized Sidelobe Cancellers with Adaptive Spatial Blocking Filters
Y. Rosa Zheng University of Missouri-Rolla, Rolla MO, Robert Lynch Naval Undersea Warfare Center, Newport RI and Genshe Chen Intelligent Automation Inc, Rockville MD
This paper investigates radar clutter scenarios where strong reflections of the target result in highly correlated clutters. Conventional adaptive beamformers suffer from the desired signal cancellation problem which results in low detection rates. The proposed work tackles this problem by adding an adaptive blocking filter in the conventional GSC (Generalized Sidelobe Canceller). Results using simulation and measured data show that the method improves the Signal-to-Interference-and-Noise-Ratio (SINR) and CFAR (constant false alarm rate) detection in such clutter scenarios as mountainous terrain, dense urban, and concrete ground, etc.

S7-5: Feature Vectors for Silicon Ion-Channel Sensors
Homin Kwon, Peter Knee and Andreas Spanias Arizona State University, Tempe AZ
A silicon ion-channel sensor was developed at Arizona State University. Experiments with this sensor have shown that it is feasible to use transform-domain feature extraction and classification techniques for the characterization of its current responses. In this paper, we examine unitary transform representations of the ion-channel current. In particular, we study constrained transform representations and present preliminary classification results using different types of classifiers.

S7-6: Performance Analysis of Different Pulse Compression Sequences
Pathipati Srihari Dadi Institute of Engineering and Technology, Anakapalle, India, Dasari Tiirumal Rao GMR Insttute of Science and Technology, Rajam, India, M. Murali Sankethika Vidya Parishad Engineering College, Visakhapatnam, India, Ch. Srinivas Anil Neerukonda Institute of Technology, Visakhapatnam, India, B. Leela Ram Prakash Vignan Institute of Information Technolgy, India and Konduri Raja Rajeswari Andhra University, Visakhapatnam, India
Pulse compression sequences with optimal merit factor and discrimination are very useful in synchronization, radar, sonar and spread spectrum communications. Progressive bounds on merit factor are drawn for binary, ternary and non-binary quinqenary sequences to compare their performance. Merit factors for sequences are found at a larger length using Kronecker product. A measure known as noise enhancement factor is defined to compare their performance in noisy environment. Sequences noise performance. The performance analysis gives good results in the search for pulse compression sequences with optimal merit factors.

S7-7: Characterization of Sea Clutter Based on Estimating the Space-Time Covariance Matrix From Real Data
Ying Li Arizona State University, Tempe AZ, Sandeep Sira Zounds Inc., Mesa AZ, William Moran, Sofia Suvorova The University of Melbourne, Victoria, Australia, Douglas Cochran Arizona State University, Darryl Morrell Arizona State University Polytechnic Campus, Mesa AZ and Antonia Papandreou-Suppappola Arizona State University
We propose an estimation method for the space-time covariance matrix of sea clutter to support the application of waveform-agile sensing procedures that rely on accurate estimation of this matrix. The method exploits the special structure of the vectorized states of the scattering function for the dynamical system model governing the temporal evolution of the clutter matrix followed by a multiple particle filtering approach to estimate the covariance matrix and deal with the high dimensionality on the formulation. The effectiveness of the method is demonstrated by estimating the clutter scattering function covariance matrix and detecting a small moving target embedded in the clutter. We use both simulated sea clutter and real sea clutter from the DSTO INGARA radar.

S7-8: Dynamic Waveform Selection for Target Tracking in Low Signal-to-Noise Ratio Environments
Shwetha Edla, Antonia Papandreou-Suppapola Arizona State University, Tempe AZ and Darryl Morrell Arizona State University, Polytechnic Campus, Mesa AZ
Dynamic waveform configuration is a fast emerging technique in radar that enables the design of radar waveforms at each time instant in order to improve tracking. Most adaptive waveform configuration algorithms have been designed based on the use of the Cramer-Rao lower bound for obtaining the measurement noise covariance that considers only the mainlobe of the ambiguity function. As a result, it is only applicable for tracking under low noise conditions. We propose a method for waveform-agile tracking under high noise conditions by using the concept of resolution cells so that it can include any number of ambiguity function sidelobes and thus perform well in realistic scenarios of low signal-to-noise ratios. A waveform selection algorithm is developed for an active sensor that tracks a target moving in a single dimension in white Gaussian noise. The algorithm chooses the time duration and chirp rate of the next transmitted linear frequency-modulated waveform to minimize the mean squared tracking error. Simulations demonstrate the increased tracking performance when the waveform parameters are selected in comparison to using a fixed waveform. The performance of the dynamically selected waveforms was also shown to improve as the number of ambiguity function sidelobes increases.