1:00, SAM-P5.1
CRAMER-RAO BOUNDS FOR CONTINUOUS-TIME AR PARAMETER ESTIMATION WITH IRREGULAR SAMPLING
E. LARSSON, E. LARSSON
Consider the problem of estimating the parameters in a continuous-time autoregressive (AR) model given measurements taken at arbitrary time instants. In this paper the Cramer-Rao bound for this problem is derived by using a technique based on the Slepian-Bang's formula and residue calculus. Furthermore, we investigate by means of numerical experiments how different sampling schemes can affect accuracy. Interestingly enough, however, for the examples studied, the estimation accuracy is relatively insensitive to the choice of sampling strategy.

1:00, SAM-P5.2
THE CRAMER-RAO BOUND FOR THE ESTIMATION OF NOISY PHASE SIGNALS
A. ZOUBIR, A. TALEB
This paper deals with noisy phase monocomponent signals in additive noise. This model is more appropriate for real world applications, in particular for radar and communications. The problem is introduced and a maximum likelihood solution is proposed. Specifically, the Cramer-Rao bound is explicitly derived and compared to the case of noise free phase.

1:00, SAM-P5.3
BROADBAND MAXIMUM LIKELIHOOD ESTIMATION OF SHALLOW OCEAN PARAMETERS USING SHIPPING NOISE
C. MECKLENBRÄUKER, A. GERSHMAN
Broadband Maximum Likelihood Estimation of Shallow Ocean Parameters Using Shipping Noise C.F. Mecklenbraeuker and A.B. Gershman In this paper, environmental parameter estimation for a shallow ocean is addressed by using wideband shipping noise as a source of acoustic energy. Unknown locations of the broadband acoustic sources are estimated simultaneously with the ocean depth using the approximate Conditional Maximum Likelihood Estimator (CMLE). This procedure is tested via computer simulations and applied to the experimental hydrophone towed array data.

1:00, SAM-P5.4
A NONITERATIVE MAXIMUM LIKELIHOOD PARAMETER ESTIMATOR OF SUPERIMPOSED CHIRP SIGNALS
S. SAHA, S. KAY
We address the problem of parameter estimation of superimposed chirp signals in noise. The approach used here is a computationally modest implementation of a maximum likelihood (ML) technique. The ML technique for estimating the complex amplitudes, chirping rates and frequencies reduces to a separable optimization problem where the chirping rates and frequencies are determined by maximizing a compressed likelihood function which is a function of only the chirping rates and frequencies. Since the compressed likelihood function is multidimensional, its maximization via grid search is impractical. We propose a non-iterative maximization of the compressed likelihood function using importance sampling. Simulation results are presented for a scenario involving closely spaced parameters for the individual signals.

1:00, SAM-P5.5
LOCALLY OPTIMAL MAXIMUM-LIKELIHOOD ESTIMATE OF A TOEPLITZ MATRIX OF GIVEN RANK
Y. ABRAMOVICH, N. SPENCER
We derive an algorithm to compute a maximum-likelihood (ML) estimate of a Toeplitz covariance matrix T whose rank is known a priori that is "locally optimal", by maximising the likelihood ratio in the neighborhood of T. This problem arises, for example, in the detection and/or direction-of-arrival (DOA) estimation of m uncorrelated plane-wave sources using a nonuniform (sparse) M-sensor linear antenna array, where m >= M (the "superior" case), but the problem is important in its own right, and has application to other areas of signal processing and communications. The algorithm relies upon the solution of a convex linear programming (LP) problem, whose feasibility is guaranteed.

1:00, SAM-P5.6
EFFICIENT EXTRACTION OF EVOKED POTENTIALS BY COMBINATION OF WIENER FILTERING AND SUBSPACE METHODS
A. CICHOCKI, R. GHARIEB, T. HOYA
A novel approach is proposed in order to reduce the number of sweeps (trials) required for the efficient extraction of the brain evoked potentials (EP). This approach is developed by combining both the Wiener filtering and the subspace methods. First, the signal subspace is estimated by applying the singular-value decomposition (SVD) to an enhanced version of the raw data obtained by Wiener filtering. Next, estimation of the EP data is achieved by orthonormal projecting the raw data onto the estimated signal subspace. Simulation results show that combination of both two methods provides much better capability than each of them separately.

1:00, SAM-P5.7
JOINT TIME DELAY AND FREQUENCY ESTIMATION OF MULTIPLE SINUSOIDS
G. LIAO, H. SO, P. CHING
In this paper, we devise a new subspace method for estimating the differential time delay of a signal received at two separated sensors as well as the frequencies of the source signal, assuming that it consists of multiple sinusoids. The time delay and frequency estimates are related to the eigenvalues and eigenvectors of a matrix obtained from the covariances of the received signals. The effectiveness of the proposed algorithm is demonstrated via computer simulations using sinusoidal signals as well as real speech data.

1:00, SAM-P5.8
OCEAN ACOUSTIC TOMOGRAPHY STRUCTURED COVARIANCE ESTIMATION
S. BAUSSON, J. MOURA, D. MAUUARY
Classic Ocean Acoustic Tomography by Wiener inversion needs good estimates of the noises power affecting the errors between the in situ measurements of the travel times and their estimates obtained by reliable simulations. We investigate the maximum likelihood estimation of a structured covariance matrix, whose subspaces of interest are known, but whose associated powers are unknown. Using the Ocean Acoustic Tomography constraints, we assume that the covariance is the sum of a full rank known matrix and an unknown component. We derive the maximum likelihood estimates for these noise powers and compute the Fisher information matrix to get insight into the geometric properties of the estimators. We verify with a realistic classic Ocean Acoustic Tomography simulation the good quality of our noise power estimates.

1:00, SAM-P5.9
BAYESIAN ESTIMATION OF CHIRPLET SIGNALS BY MCMC SAMPLING
C. LIN, P. DJURIC
We address the problem of parameter estimation of chirplets, which are chirp signals with Gaussian shaped envelopes. The procedure we propose is an extension of our previous work on estimation of chirp signals, and it is based on MCMC sampling. For fast convergence of the MCMC sampling based method, a critical step is the initialization of the method. Since the chirplets have finite durations and may or may not overlap in time, we propose initialization procedures for each of these cases. We have tested the method by extensive simulations and compared it with Cramer-Rao bounds. The obtained results have been excellent.

1:00, SAM-P5.10
ESTIMATION OF CAR PROCESSES OBSERVED IN NOISE USING BAYESIAN INFERENCE
P. GIANNOPOULOS, S. GODSILL
We consider the problem of estimating continuous-time autoregressive (CAR) processes from discrete-time noisy observations. This can be done within a Bayesian framework using Markov chain Monte Carlo (MCMC) methods. Existing methods include the standard random walk Metropolis algorithm. On the other hand, least-squares (LS) algorithms exist where derivatives are approximated by differences and parameter estimation is done in a least-squares manner. In this paper, we incorporate the LS estimation into the MCMC framework to develop a new MCMC algorithm. This new algorithm is combined with the standard Metropolis algorithm and it is found that the combined algorithm improves performance compared to the standard Metropolis algorithm or the Metropolis algorithm using a LS proposal. Simulation results are presented to support our findings.