Search

The time invariant (TI) widely linear (WL) minimum variance distortionless response (MVDR) beamformer for the optimal reception of an unknown signal corrupted by potentially noncircular interferences is discussed. Findings reveal the suitability of the beamformer for spectrum monitoring or passive listening in noncircular context.

Three polarization diversity (PD)-2q-MUSIC algorithms depending on the a priori knowledge about the polarization of the sources are discussed. Results suggest the increase in performance of PD-2q algorithms with q when resolution in both direction of arrival and polarization is required.

An extension of the Music method to an arbitrary even order 2q (q greater than equal to 1) is proposed, giving rise to the 2q-Music methods. The performance analysis of these methods show off important results for direction-finding applications and in particular the best performance with respect to 2-Music and 4-Music, of 2q-Music method with q > 2, despite their variance, when some resolution is required.

New enlightening results about the behavior, properties, and performance of the optimal widely linear (WL) receiver for demodulation of a binary phase-shift keying (BPSK) signal corrupted by rectilinear interferences are presented. By simple algebraic operation of demodulation on the baseband signal, the minimum shift keying and Gaussian minimum shift keying (GMSK) modulations can be made to approximately correspond to a BPSK modulation allowing the application of the simple antenna interference cancellation (SAIC) concept to the GMS cellular network at the mobile level.

The Independent Component Analysis using Redundancies in the quadricovariance (ICAR) algorithm, exploiting the information contained in the data statistics at fourth order only is proposed. This algorithm allows the processing of overdetermined mixtures of sources, provided the latter have marginal fourth-order (FO) cumulants with the same sign, which is generally the case in radio communications contexts.

A new blind identification (BI) method that exploits the fourth-order data statistics only (FOBIUM) is presented to process both overdetermined and underdetermined instantaneous mixtures of statistically independent sources. The FOBIUM method is shown to require a relatively weak number of snapshots to generate good output performances for currently used radiocommunication sources, such as quadrature phase shift keying (QPSK) sources.

The mechanisms, resolution and maximal processing capacity of numerous 2qth order array processing methods are discussed. The geometry, number of Virtual Sensors (VSs) of the Virtual Array (VA) and the number of sources that could be processed by a 2qth order method by exploiting the algebraic structure of C (sub 2q,x) are controlled by the arrangement of the 2qth order data statistics.

An analysis conducted to check the behavior, and to propose adaptations of the current second-order (SO) blind source separation (BSS) methods is presented. The extension for deterministic sources like frequency shift keying (FSK) and continuous phase frequency shift keying (CPFSK) sources is proposed.

The problem is addressed of estimating second-order statistics of a frequency-selective Rayleigh channel in the context of the wideband code division multiple access (CDMA) mode of the Universal Mobile Telephone Service (UMTS). Performances are evaluated, and it is shown that significant improvement of the performance of a RAKE receiver is achieved.

Fourth-order (FO) direction finding (DF) methods have been usually employed for non-Gaussian signals. It has been shown through the use of the virtual cross-correlation (VCC) concept that the application of FO cumulants for the DF problem increases the effective aperture of an arbitrary antenna array. This led to the virtual array (VA) concept. The VA concept is presented through an alternative manner that is easier and more direct to handle than the VCC tool.

Linear-quadratic (LQ) filters for signal detection and direction-of-arrival estimation in the real case were extended to the complex plane. Specifically, an optimal second-order complex LQ array receiver for random and non-random signals was designed for the non-Gaussian case using a criterion based on the deflection or output signal-to-noise ratio. Its performance in an environment corrupted by spatially fourth-order white noise was also evaluated.

The use of widely linear mean square estimation (WLMSE) in statistical signal processing of complex data is better than using linear systems. In normal complex data regression is widely linear, as it is linear to both x and x* axes, where x* is a complex conjugate of x. However WLMSE should not be used where circularity is assumed. The use of WLMSE and a comparison of the MSE using widely linear and linear systems is used.

Research into blind identification of single-input multiple-output systems using second-order statistics only is presented. It is concluded that the parametric subspace method gives consistent channel estimates when only an upper bound of the channel order is identified.

Research into the behaviour of the current second-order and fourth-order cumulant-based blind source separation methods for zero mean, (quasi)-cyclostationary and cyclo-ergodic sources is presented.

Orthogonal space-time block codes (STBC), and the Alamouti scheme in particular, are of particular interest in multiple-input multiple-output (MIMO) systems since they achieve full spatial diversity over fading channels and are decoded from linear processing at the receiver. Nevertheless, due to the expensive spectral resource, increasing network capacity requires the development of interference cancellation (IC) techniques allowing several users to share the same spectral resources without impacting the transmission quality. In this context several IC schemes have been developed during this last decade, where each user is equipped with multiple antennas and employs STBC at transmission. However, these IC techniques require multiple antennas at reception, which remains a challenge at the handset level due to cost and size limitations. For this reason, low complexity single antenna interference cancellation (SAIC) techniques, currently operational in GSM handsets, have been developed recently for single antenna users using real-valued modulations or complex filtering of real-valued modulations, by using a widely linear (WL) filtering at reception. Extension to multiple antennas at reception is called multiple antenna interference cancellation (MAIC) technique. The purpose of this paper is to extend the SAIC/MAIC technology to users using both real-valued constellations, such as amplitude shift keying (ASK) constellations, and the Alamouti scheme at transmission. A WL minimum mean square error (MMSE) receiver, completely new for IC purposes in the context of radio communications systems using the Alamouti scheme, is proposed and analyzed. This receiver, which corresponds to the maximum likelihood (ML) receiver for synchronous intranetwork interferences, is able to separate up to 2 N Alamouti users from N antennas at reception, hence SAIC capability for N = 1. [ABSTRACT FROM PUBLISHER]

The problem of blind separation of overdetermined mixtures of sources, that is, with fewer sources than (or as many sources as) sensors, is addressed in this paper. A new method, called Independent Component Analysis using Redundancies in the quadricovariance (ICAR), is proposed in order to process complex data. This method, without any whitening operation, only exploits some redundancies of a particular quadricovariance matrix of the data. Computer simulations demonstrate that ICAR offers in general good results and even outperforms classical methods in several situations: ICAR i) succeeds in separating sources with low signal-to-noise ratios, ii) does not require sources with different second-order or/and first-order spectral densities, iii) is asymptotically not affected by the presence of a Gaussian noise with unknown spatial correlation, iv) is not sensitive to an over estimation of the number of sources. [ABSTRACT FROM AUTHOR]