9:30, SAM-L2.1
PRE-STEERING DERIVATIVE CONSTRAINTS FOR ROBUST BROADBAND ANTENNA ARRAYS
S. ZHANG, I. THNG
The weights of an optimum PB (Pre-steered broadband) antenna array processor are often obtained by solving a LCMV (linearly constrained minimum variance) problem. The objective function is the mean output power (variance) and the constraints space is a set of linear equations which ensure a constant gain in a fixed direction known as the look direction. However, errors in a practical scenario could degrade the performance of the LCMV processor significantly, namely, mismatches between the look direction and the actual DOA (direction of arrival) of the desired signal, positional errors in the sensors and quantization errors in the pre-steered front end of the broadband processor. The main contribution of this paper is the derivation of a new set of constraints, referred to as the Pre-steering derivative constraints, which is able to maintain the processor robustness in the general 3D (three dimensional) space scenario with all the errors mentioned above.

9:50, SAM-L2.2
FILTER-AND-SUM BEAMFORMER WITH ADJUSTABLE FILTER CHARACTERISTICS
M. KAJALA, M. HÄMÄLÄINEN
In this paper we introduce a polynomial filter structure for filter-and-sum beamforming applied to microphone array application. The structure is a multi-dimensional extension of well-known Farrow structure, which has mainly been used for fractional delay filtering and interpolation of 1-D signals. The proposed method enables an easy, smooth, and efficient control of beamforming filter characteristic by adjusting only a single control variable e.g. for dynamic beam steering. The optimization method for polynomial beamforming filter design is presented and illustrated with simulations of beamforming filter characteristics. The design example is given for a linear array of four omni-directional microphones and a polynomial FIR filter with 20-tap delay lines.

10:10, SAM-L2.3
MATRIX FILTERS FOR PASSIVE SONAR
R. VACCARO, B. HARRISON
This paper introduces matrix filters as a tool for localization and detection problems in passive sonar. The outputs of an array of sensors, at some given frequency, can be represented by a vector of complex numbers. A linear filtering operation on the sensor outputs can be expressed as the multiplication of a matrix (called a matrix filter) times this vector. The purpose of a matrix filter is to attenuate unwanted components in the measured sensor data while passing desired components with minimal distortion. Matrix filters are designed by defining an appropriate pass band and stop band and solving a convex optimization problem. This paper formulates the design of matrix filters for passive sonar and gives two examples.

10:30, SAM-L2.4
OPTIMAL ARRAY PATTERN SYNTHESIS USING SEMIDEFINITE PROGRAMMING
F. WANG, V. BALAKRISHNAN, P. ZHOU, J. CHEN, R. YANG, C. FRANK
In this paper, we present a new technique to solve array pattern synthesis problems by using semidefinite programming. We first formulate (or reformulate) the array design problems into semidefinite programming problems, and then use the recently developed efficient numerical algorithms and software to compute the numerical solution of antenna array weights. Using this approach, we can directly solve not only the standard synthesis problems for nonuniform arrays, but also the synthesis problems for arrays having power restrictions and uncertainties. Numerical examples are presented to illustrate our approach.

10:50, SAM-L2.5
VANDERMONDE INVARIANCE TRANSFORMATION
T. KURPJUHN, M. IVRLAC, J. NOSSEK
In this article we introduce a novel multilpe-input-multiple-output (MIMO) spatial filter (SF) which can be applied as a preprocessing scheme to uniform linear arrays, preserving the Vandermonde structure of the steering vectors while changing the amplitude and the phase gradient of the steering vector in a nonlinear fashion. The new scheme is therefore titled Vandermonde Invariance Transformation. The introduced degrees of freedom due to this preprocessing transformation can be used to beneficially influence the properties of the channel to achieve an enhanced performance of the subsequent signal processing algorithm.

11:10, SAM-L2.6
BLIND WIDEBAND SPATIAL FILTERING BASED ON HIGHER-ORDER CYCLOSTATIONARITY PROPERTIES
G. GELLI, D. MATTERA, L. PAURA
In this paper, a new method for blind spatial signal filtering is proposed. It utilizes the selectivity property of the higher-order cyclostationary statistics exhibited by the signal of interest to accurately estimate its unknown steering vector, also in the presence of strong frequency-overlapped interfering signals, which would render the standard stationarity-based techniques ineffective. The method is useful when the desired signal cannot be accurately extracted by exploiting its second-order cyclostationarity properties, as proposed in [1]. The performance analysis, carried out by computer simulations in the case of QAM-modulated signals with partially overlapping bands, substantiates the effectiveness of the proposed method.