Eleftherios Kofidis

A Novel Efficient Cluster-Based MLSE Equalizer for Satellite Communication Channels with

In satellites, nonlinear amplifiers used near saturation severely distort the transmitted signal and cause difficulties in its reception. Nevertheless, the nonlinearities introduced by memoryless bandpass amplifiers preserve the symmetries of the-ary quadrature amplitude modulation (-QAM) constellation. In this paper, a cluster-based sequence equalizer (CBSE) that takes advantage of these symmetries is presented. The proposed equalizer exhibits enhanced performance compared to other techniques, including the conventional linear transversal equalizer, Volterra equalizers, and RBF network equalizers. Moreover, this gain in performance is obtained at a substantially lower computational cost.

On the Best Rank-1 Approximation of Higher-Order Supersymmetric Tensors

Recently the problem of determining the best, in the least-squares sense, rank-1 approximation to a higher-order tensor was studied and an iterative method that extends the well-known power method for matrices was proposed for its solution. This higher-order power method is also proposed for the special but important class of supersymmetric tensors, with no change. A simplified version, adapted to the special structure of the supersymmetric problem, is deemed unreliable, as its convergence is not guaranteed. The aim of this paper is to show that a symmetric version of the above method converges under assumptions of convexity (or concavity) for the functional induced by the tensor in question, assumptions that are very often satisfied in practical applications. The use of this version entails significant savings in computational complexity as compared to the unconstrained higher-order power method. Furthermore, a novel method for initializing the iterative process is developed which has been observed to yield an estimate that lies closer to the global optimum than the initialization suggested before. Moreover, its proximity to the global optimum is a priori quantifiable. In the course of the analysis, some important properties that the supersymmetry of a tensor implies for its square matrix unfolding are also studied.

Review: Preamble-based channel estimation in OFDM/OQAM systems: A review

Filter bank-based multicarrier communications (FBMC) have recently attracted increased interest in both wired (e.g., xDSL, PLC) and wireless (e.g., cognitive radio) applications, due to their enhanced flexibility, higher spectral efficiency, and better spectral containment compared to conventional OFDM. A particular type of FBMC, the so-called FBMC/OQAM or OFDM/OQAM system, consisting of pulse shaped OFDM carrying offset QAM (OQAM) symbols, has received increasing attention due to, among other features, its higher spectral efficiency and implementation simplicity. It suffers, however, from an imaginary inter-carrier/inter-symbol interference that complicates signal processing tasks such as channel estimation. This paper focuses on channel estimation for OFDM/OQAM systems based on a known preamble. A review of the existing preamble structures and associated channel estimation methods is given, for both single- (SISO) and multiple-antenna (MIMO) systems. The various preambles are compared via simulations in both mildly and highly frequency selective channels.

Monotonic convergence of fixed-point algorithms for ICA

We re-examine a fixed-point algorithm proposed by Hyvarinen for independent component analysis, wherein local convergence is proved subject to an ideal signal model using a square invertible mixing matrix. Here, we derive step-size bounds which ensure monotonic convergence to a local extremum for any initial condition. Our analysis does not assume an ideal signal model but appeals rather to properties of the contrast function itself, and so applies even with noisy data and/or more sources than sensors. The results help alleviate the guesswork that often surrounds step-size selection when the observed signal does not fit an idealized model.

Preamble-Based Channel Estimation for CP-OFDM and OFDM/OQAM Systems: A Comparative Study

In this correspondence, preamble-based least squares (LS) channel estimation in orthogonal frequency division multiplexing (OFDM) systems of the QAM and offset QAM (OQAM) types is considered. The construction of optimal (in the mean squared error (MSE) sense) preambles is investigated, for sparse (a subset of pilot tones, surrounded by nulls) preambles. The two OFDM systems are compared for the same transmit power, which, for cyclic prefix (CP) based OFDM/QAM, also includes the power spent for CP transmission. OFDM/OQAM, with a sparse preamble consisting of equipowered and equispaced pilots embedded in zeros, turns out to perform at least as well as CP-OFDM. Simulations results are presented that verify the analysis.

Nonlinear adaptive filters for speckle suppression in ultrasonic images

Abstract A novel approach to suppression of ultrasonic speckle based on a combination of segmentation and optimum L-filtering is presented. With the aid of a suitable modification of the learning vector quantizer (LVQ) neural network, the image is segmented in regions of (approximately) homogeneous statistics. For each of the regions a minimum mean-squared-error (MMSE) L-filter is designed, by using the histogram of grey levels as an estimate of the parent distribution of the noisy observations and a suitable estimate of the (assumed constant) original signal in the corresponding region. Thus, a bank of L-filters results, with each of them corresponding to and operating on a different image region. Simulation results from both simulated and real B-mode ultrasonic images are presented, which verify the (qualitative and quantitative) superiority of our technique over a number of commonly used speckle filters.

Wavelet-based medical image compression

Abstract In view of the increasingly important role played by digital medical imaging in modern health care and the consequent blow up in the amount of image data that have to be economically stored and/or transmitted, the need for the development of image compression systems that combine high compression performance and preservation of critical information is ever growing. A powerful compression scheme that is based on the state-of-the-art in wavelet-based compression is presented in this paper. Compression is achieved via efficient encoding of wavelet zerotrees (with the embedded zerotree wavelet (EZW) algorithm) and subsequent entropy coding. The performance of the basic version of EZW is improved upon by a simple, yet effective, way of a more accurate estimation of the centroids of the quantization intervals, at a negligible cost in side information. Regarding the entropy coding stage, a novel RLE-based coder is proposed that proves to be much simpler and faster yet only slightly worse than context-dependent adaptive arithmetic coding. A useful and flexible compromise between the need for high compression and the requirement for preservation of selected regions of interest is provided through two intelligent, yet simple, ways of achieving the so-called selective compression. The use of the lifting scheme in achieving compression that is guaranteed to be lossless in the presence of numerical inaccuracies is being investigated with interesting preliminary results. Experimental results are presented that verify the superiority of our scheme over conventional block transform coding techniques (JPEG) with respect to both objective and subjective criteria. The high potential of our scheme for progressive transmission, where the regions of interest are given the highest priority, is also demonstrated.

On Training Optimization for Estimation of Correlated MIMO Channels in the Presence of Multiuser Interference

In this paper, the problem of estimating multiple-input multiple-output (MIMO) channels in a realistic environment involving correlated channel fading and multiuser interference is considered. Four estimation schemes are studied, including the linear minimum mean squared error (LMMSE), least squares (LS), and Gauss-Markov (GM) estimators, as well as a novel scheme which is derived here as an alternative to LMMSE estimation. The MSE-optimal training sequences for each of them are provided and their requirements for side information feedback are assessed. The new scheme is shown to exhibit a performance comparable to or even better than LMMSE, at a significantly lower feedback and computational cost. The analytical comparison of the estimation schemes is supported by numerous simulation results that cover a wide range of antenna configurations, relative interference power, and channel correlation strengths. The results of this paper provide a complete picture for a palette of estimation schemes, with their relative performance and costs of training.

The higher-order power method revisited: convergence proofs and effective initialization

We revisit the higher-order power method of De Lathauwer et al. (1995) for rank-one tensor approximation, and its relation to contrast maximization as used in blind deconvolution. We establish a simple convergence proof for the general nonsymmetric tensor case. We show also that a symmetric version of the algorithm, offering an order of magnitude reduction in computational complexity but discarded by De Lathauwer et al. as unpredictable, is likewise provably convergent. A new initialization scheme is also developed which, unlike the TSVD-based initialization, leads to a quantifiable proximity to the globally optimal solution.

On the perfect reconstruction problem in N-band multirate maximally decimated FIR filter banks

The problem of finding N-K filters of an N-band maximally decimated FIR analysis filter bank, given K filters, so that FIR perfect reconstruction can be achieved, is considered. The perfect reconstruction condition is expressed as a requirement of unimodularity of the polyphase analysis filter matrix. Based on the theory of Euclidean division for matrix polynomials, the conditions the given transfer functions must satisfy are given, and a complete parameterization of the solution is obtained. This approach provides an interesting alternative to the method of the complementary filter in the case of N>2,K<N-1, where the latter leads to a system of nonlinear equations. Moreover, it yields the polyphase synthesis filter matrix as a byproduct. As an application, a complete characterization of all paraunitary matrices with fixed first row is derived. The problem of appropriately choosing the parameters characterizing the complementary filters to lead to filters of practical useful frequency responses is studied, and analytical solutions for K=N-1 are given. It is demonstrated, through an example, that the complementary filters found via Euclids algorithm are not necessarily linear phase even if the given filters are. The problem of obtaining linear phase solutions of given orders is investigated for the general case (N/spl ges/2,K/spl les/N-1) and systematic ways are developed to compute such solutions in the K=N-1 and K=1 cases. It is also shown that the resulting synthesis filters are linear phase. Design examples illustrating the theory are presented.