3:30, SAM-P2.1
MARKOV MODELING OF TRANSIENT SCATTERING AND ITS APPLICATION IN MULTI-ASPECT TARGET CLASSIFICATION
Y. DONG, P. RUNKLE, L. CARIN
Transient scattered fields from a general target are composed of wavefronts, resonances and time delays, with these constituents linked to the target geometry. A classifier applied to transient scattering data requires a statistical model for such fundamental constituents. A Markov model is employed to characterize the transient scattered fields - for a set of target-sensor orientation over which the transient scattering is stationary - utilizing a wavefront, resonance, time-delay "alphabet". The Markov model is utilized in a classifier developed for multi-aspect transient scattering data, with a hidden Markov model (HMM) employed to address the generally non-stationary nature of the multi-aspect waveforms. Each state of the HMM is characteristic of a set of target-sensor orientations for which the scattering statistics are stationary, the statistics of which are characterized via the aforementioned Markov model. The wavefront, resonance and time-delay features are extracted via a modified matching-pursuits algorithm.

3:30, SAM-P2.2
ORTHOGONAL MATCHED FILTER DETECTION
Y. ELDAR, A. OPPENHEIM
In this paper we consider the generic problem of detecting a transmitted signal when one of $M$ known signals is transmitted. Instead of using a classical matched filter (MF) detector, matched to the transmitted signals, we propose using an orthogonal matched filter (OMF) detector, which is matched to a set of orthogonal signals that are closest in a least-squares sense to the transmitted signals. We show that this approach is equivalent to optimally whitening the output of the MF demodulator, and then basing the detection on the whitened output. We provide simulation results that suggest that in many cases the OMF detector outperforms the MF detector.

3:30, SAM-P2.3
ADAPTIVE RESOLUTION OF MAINLOBE AND SIDELOBE DETECTIONS USING MODEL ORDER DETERMINATION
A. JAFFER, J. CHEN, T. MILLER
This paper presents the development and performance evaluation of a methodology for distinguishing between mainlobe and sidelobe detections that arise in adaptive radar systems operating in adverse environments. Various adaptive detection test statistics such as the adaptive matched filter (AMF), the generalized likelihood ratio test (GLRT) and adaptive coherence estimate (ACE), and combinations of these, have been previously analyzed with respect to their sidelobe rejection capabilities. In contrast to these methods which are based on detecting a single target with known direction and Doppler, the present method uses model order determination techniques applied to the AMF or GLRT data observed over the range of unknown angle and Doppler parameters. The determination of model order, i.e. the number of signals present in the data, is made by using least-squares model fit error residuals and applying the Akaike Information Criterion. Comprehensive computer simulation results are presented which demonstrate substantial improvement in sidelobe rejection performance compared to previous methods.

3:30, SAM-P2.4
APPROXIMATE CFAR SIGNAL DETECTION IN STRONG LOW RANK NON-GAUSSIAN INTERFERENCE
I. KIRSTEINS, M. RANGASWAMY
Recent work suggests that the performance of conventional Gaussian-based adaptive methods can degrade severely in correlated non-Gaussian interference. We have addressed this problem by developing a new generalized likelihood ratio test (GLRT) for detecting a signal in unknown, strong non-Gaussian low rank interference plus white Gaussian noise which does not need detailed knowledge of the non-Gaussian distribution. The optimality of the proposed GLRT detector is established using perturbation expansions of the test statistic to show that it is closely related to the UMPI test for this problem. Computer simulations indicate that the new detector significantly outperforms standard adaptive methods in non-Gaussian interference and is robust.

3:30, SAM-P2.5
ADAPTIVE CFAR DETECTION OF MULTIDIMENSIONAL SIGNALS
E. CONTE, A. DE MAIO, C. GALDI, G. RICCI
Adaptive detection of multidimensional signals in the presence of interference with unknown covariance matrix is an expanding topic in a variety of scenarios ranging from radar/sonar to digital communication systems. In this paper we attack the problem of detecting a multidimensional radar signal, modeled as an unknown NxH matrix, embedded in Gaussian noise with unknown covariance matrix, with the ambition of devising receivers which yield the Constant False Alarm Rate (CFAR) property. We show that this aim can be achieved resorting to the principle of invariance, namely restricting our attention to hypotheses testing problems which remain unaltered under a proper group of transformations. Several detectors based on the maximal invariant statistic are studied and, in particular, the Generalized Likelihood Ratio Test (GLRT) is shown to belong to the class of invariant tests.

3:30, SAM-P2.6
A MULTI-HYPOTHESIS GLRT APPROACH TO THE COMBINED SOURCE DETECTION AND DIRECTION OF ARRIVAL ESTIMATION PROBLEM
R. BETHEL, K. BELL
The problem of detecting the number of uncorrelated narrowband signals received by an array of sensors when the direction of arrival (DOA) and power level of each source is unknown is investigated. A multi-hypothesis generalized likelihood ratio test (GLRT) approach is used, resulting in a procedure which maximizes the likelihood function with respect to the number of signals and their DOAs and powers. A tuning mechanism for controlling the trade-off between the probability of correct detection and the probability of false alarm is obtained by imposing a constraint on the minimum allowable value for the power level estimate. A sequential search over the number of sources is used for a computationally feasible solution. Performance comparisons are made to the Minimum Description Length (MDL) and Minimum Variance Distortionless Response (MVDR) signal detection approaches.

3:30, SAM-P2.7
DYNAMIC ENUMERATION ALGORITHM USING ARRAY ANTENNAS
P. GREEN, D. TAYLOR
This paper develops a robust method to enumerate the incident signals impinging on a uniform but variable size linear array independant of their extent of correlation in a Rayleigh flat fading channel environment. The method also optimizes by minimizing the number of antennas to the number of signals and adapts continously to maintain optimum performance in a mobile environment where users (signals) come and go.The technique is a modification of the matrix decomposition method of Cozzens and Sousa. A new set of stability, stopping and adaptive control criteria is presented. An algorithm is formulated with simulation and field results presented.

3:30, SAM-P2.8
IRREDUCIBLE FORM FOR AP ALGORITHM FOR DETECTING THE NUMBER OF COHERENT SIGNALS BASED ON THE MDL PRINCIPLE
M. SUZUKI, H. SANADA, N. NAGAI
This paper presents an improvement of the Alternating Projection (AP) algorithm for detecting the number of coherent signals based on the Minimum Description Length (MDL) principle using a uniform linear array of sensors. The criterion of the AP algorithm for the detection becomes indefinite, when estimated bearings more than one approach to the identical value. This paper derives an irreducible form of the AP criterion for the detection, which never get indefinite. The irreducible form is represented as a rational function and real-valued version of FFT can be exploited efficiently. The proposed algorithm reduces the order of the amount of arithmetic operations. Finally, simulation results are shown to demonstrate the validity of the proposed algorithm.

3:30, SAM-P2.9
MULTIPLE SIGNAL DETECTION UNDER A FALSE ALARM CONSTRAINT
J. ERIKSSON
This paper presents a sequential detection scheme for sinusoidal signals observed in spatially colored noise which controls the probability of overestimating the number of signals. The scheme is based on the Sequentially Rejective Bonferroni Test together with the nonlinear weighted least squares approach for estimation of the signal parameters under the hypotheses and alternatives respectively. The power of the method is compared via computer simulations to the commonly used minimum description length method, MDL. The proposed scheme performs well in comparison to the MDL method and gives at the same time control of the false alarm probability.

3:30, SAM-P2.10
COMPARISON OF GLR AND MAXIMAL INVARIANT DETECTORS UNDER STRUCTURED CLUTTER COVARIANCE
H. KIM, A. HERO
There has been considerable recent interest in applying maximal invariant (MI) hypothesis testing as an alternative to the generalized likelihood ratio (GLR) test. This interest has been motivated by several attractive theoretical properties of MI tests including: exact robustness to variation of nuisance parameters, finite-sample min-max optimality (in some cases), and distributional robustness. However, in the deep hide target detection problem, there are regimes for which either of the MI and the GLR tests can outperform the other. We will discuss conditions under which the MI tests can be expected to outperform the GLR tests in the context of a radar imaging and target detection application. We will also show that the relative advantage of the MI tests is robust to boundary estimation errors.