Search

The paper discusses computationally efficient NLMS and RLS algorithms for perfect and imperfect periodic excitation sequences. The most interesting aspect of these algorithms is that they are exact LMS and RLS algorithms suitable for identification and tracking of every linear system and they require a real-time computational effort of just a multiplication, an addition and a subtraction per sample time. Moreover, the algorithms have convergence and tracking properties that can be better than or comparable with the NLMS algorithm for white noise input. The transient and steady state behavior of the algorithms and their tracking properties are also studied in the paper. [ABSTRACT FROM AUTHOR]

In a companion paper (see "Resource Allocation for Downlink Cellular OFDMA Systems-Part I: Optimal Allocation," IEEE Trans. Signal Process., vol. 58, no. 2, pp. 720-734, Feb. 2010), we characterized the optimal resource allocation in terms of power control and subcarrier assignment, for a downlink sectorized OFDMA system impaired by multicell interference. In our model, the network is assumed to be one dimensional (linear) for the sake of analysis. We also assume that a certain part of the available bandwidth is likely to be reused by different base stations while that the other part of the bandwidth is shared in an orthogonal way between these base stations. The optimal resource allocation characterized in Part I is obtained by minimizing the total power spent by the network under the constraint that all users' rate requirements are satisfied. It is worth noting that when optimal resource allocation is used, any user receives data either in the reused bandwidth or in the protected bandwidth, but not in both (except for at most one pivot-user in each cell). We also proposed an algorithm that determines the optimal values of users' resource allocation parameters. As a matter of fact, the optimal allocation algorithm proposed in Part! requires a large number of operations. In the present paper, we propose a distributed practical resource allocation algorithm with low complexity. We study the asymptotic behavior of both this simplified resource allocation algorithm and the optimal resource allocation algorithm of Part I as the number of users in each cell tends to infinity. Our analysis allows to prove that the proposed simplified algorithm is asymptotically optimal, i.e., it achieves the same asymptotic transmit power as the optimal algorithm as the number of users in each cell tends to infinity. As a byproduct of our analysis, we characterize the optimal value of the frequency reuse factor. Simulations sustain our claims and show that substantial performance improvements are obtained when the optimal value of the frequency reuse factor is used. [ABSTRACT FROM AUTHOR]

The decoding of a class of discrete cosine transform (DCT) and discrete sine transform (DST) codes that are maximum distance separable codes (MDS) is considered in this paper. These class of codes are considered for error correction over real fields. All the existing algebraic decoding algorithms are capable of decoding only a subclass of these codes [which can be characterized into the Bose-Chaudhuri-Hocquenghem (BCH) form], and fails to decode the remaining even though they are MDS. In this paper, we propose a new generic algorithm along the lines of coding theoretic and subspace methods to decode the entire class of MDS DCT and DST codes. The proposed subspace approaches are similar to popular ESPRIT and MUSIC algorithms. The proposed algorithms also perform significantly better than the existing algorithms on the BCH-like subclass. A perturbation analysis is also presented to study the effect of various parameters on the error localization due to the quantization noise. Simulation results are presented to demonstrate the capability of proposed algorithms to decode the entire class and to perform significantly better on the BCH-like subclass than the existing algorithm under the influence of quantization noise. [ABSTRACT FROM AUTHOR]

Passive source localization is one of the issues in array signal processing fields. In some practical applications, the signals received by an array are the mixture of near-field and far-field sources, such as speaker localization using microphone arrays and guidance (homing) systems. To localize mixed near-field and far-field sources, this paper develops a two-stage MUSIC algorithm using cumulant. The key points of this paper are: i) in the first stage, this paper derives one special cumulant matrix, in which the virtual "steering vector" is the function of the common electric angle in both near-field and far-field signal models so that source direction-of-arrival (DOA) (near-field or far-field one) can be obtained from this electric angle using the conventional high-resolution MUSIC algorithm; ii) in the second stage, this paper derives another particular cumulant matrix, in which the virtual "steering matrix" has full column rank no matter whether the received signals are multiple near-field sources or multiple far-field ones or their mixture. What is more important, the virtual "steering vector" can be separated into two parts, in which the first one is the function of the common electric angle in both signal models, whereas the second part is the function of the electric angle that exists only in near-field signal model. Furthermore, by substituting the common electric angle estimated in the first stage into one special Hermitian matrix formed from another MUSIC spectral function, the range of near-field sources can be obtained from the eigenvector of the Hermitian matrix. The resultant algorithm avoids twodimensional search and pairing parameters; in addition, it avoids the estimation failure problem and alleviates aperture loss. Simulation results are presented to validate the performance of the proposed method. [ABSTRACT FROM AUTHOR]

This paper introduces Bayes risk (expected loss) as a criterion for linear equalization. Since the probability of error is equal to the Bayes risk (BR) for a particular binary loss function, this work is a natural generalization of previous works on minimum probability of error (PE) equalizers. Adaptive equalization algorithms are developed that minimize the BR. Like the minimum PE equalizers, the BR algorithms have low computational complexity which is comparable to that of the LMS algorithm. The advantage of the BR criterion is that the loss function can be specified in a manner that accelerates adaptive equalizer convergence relative to the minimum PE adaptive algorithm as illustrated in simulation examples. Besides introducing a new criterion, this paper provides another independent contribution to the field of PE minimizing equalization. While most prior works focus on M-ary QAM type modulations with rectangular decision regions, this paper uses upper bounds on the probabilities of certain events to yield tractable mathematics that apply to two-dimensional constellations with arbitrarily shaped decision regions. The resulting adaptive algorithm use the full information available in the phase of the error signal, whereas previous algorithms use a quantized version of this error phase. [ABSTRACT FROM AUTHOR]

This paper considers the problem of recovering a sparse signal representation according to a signal dictionary. This problem could be formalized as a penalized least-squares problem in which sparsity is usually induced by a ℓ1-norm penalty on the coefficients. Such an approach known as the Lasso or Basis Pursuit Denoising has been shown to perform reasonably well in some situations. However, it was also proved that nonconvex penalties like the pseudo ℓq-norm with q < 1 or smoothly clipped absolute deviation (SCAD) penalty are able to recover sparsity in a more efficient way than the Lasso. Several algorithms have been proposed for solving the resulting nonconvex least-squares problem. This paper proposes a generic algorithm to address such a sparsity recovery problem for some class of nonconvex penalties. Our main contribution is that the proposed methodology is based on an iterative algorithm which solves at each iteration a convex weighted Lasso problem. It relies on the family of nonconvex penalties which can be decomposed as a difference of convex functions (DC). This allows us to apply DC programming which is a generic and principled way for solving nonsmooth and nonconvex optimization problem. We also show that several algorithms in the literature dealing with nonconvex penalties are particular instances of our algorithm. Experimental results demonstrate the effectiveness of the proposed generic framework compared to existing algorithms, including iterative reweighted least-squares methods. [ABSTRACT FROM AUTHOR]

This paper considers the problem of time difference-of-arrival (TDOA) source localization when the TDOA measurements from multiple disjoint sources are subject to the same sensor position displacements from the available sensor positions. This is a challenging problem and closed-form solution with good localization accuracy has yet to be found. This paper proposes an estimator that can achieve this purpose. The pro- posed algorithm jointly estimates the unknown source and sensor positions to take the advantage that the TDOAs from different sources have the same sensor position displacements. The joint estimation is a highly nonlinear problem due to the coupling of source and sensor positions in the measurement equations. We introduce the novel idea of hypothesized source locations in the algorithm development to enable the formulation of psuedolinear equations, thereby leading to the establishment of closed-form solution for source location estimates. Besides the advantage of closed-form, the newly developed algorithm is shown analytically, under the condition that the TDOA measurement noise and the sensor position errors are sufficiently small, to reach the CRLB accuracy. For clarity, the localization of two disjoint sources is used in the algorithm development. The developed algorithm is then examined under the special case of a single source and extended to the more general case of more than two unknown sources. The theoretical developments are supported by simulations. [ABSTRACT FROM AUTHOR]

Signal processing applications have high instruction level parallelism (ILP) and real-time performance requirements. Embedded and application specific multicluster architecture is desirable to provide the large computation power that these applications need. As technology moves to deep submicron level, it becomes more important and challenging to design an efficient intercluster connection network to satisfy the rapid growing intercluster data transfer needs under the power and cost constraints. This paper addresses the automatic generation of intercluster connection network with partially connected buses. An application specific approach is proposed in this paper to determine the minimum number of required partially connected buses without performance degradation for a given schedule in polynomial time. The intercluster connection topology is then generated with the determined minimum number of partially connected buses to minimize the connection bus segments. Further, a scheduling algorithm is presented in this paper to minimize the intercluster communication needs for the given application and to reduce the minimum number of partially connected buses required in the interciuster connection network under schedule length constraint. Experimental results indicate that an average reduction up to 50.6% in the number of minimum required buses and an average reduction of 64.5% in bus segments can be achieved compared to commonly used intercluster communication aware scheduling techniques and as soon as possible (ASAP) data transfer scheme. [ABSTRACT FROM AUTHOR]

In this paper, we study the invertibility of M-variate Laurent polynomial N x P matrices. Such matrices represent multidimensional systems in various settings such as filter banks, multiple-input multiple-output systems, and multirate systems. Given an N x P Laurent polynomial matrix H( z1,.. . , ZM) of degree at most k, we want to find a P x N Laurent polynomial left inverse matrix G(z) of H(z) such that G(z)H(z) = I. We provide computable conditions to test the invertibility and propose algorithms to find a particular inverse. The main result of this paper is to prove that H(z) is generically invertible when N P ⩾ M; whereas when N P < M, then H(z) is generically noninvertible. As a result, we propose an algorithm to find a particular inverse of a Laurent polynomial matrix that is faster than current algorithms known to us. [ABSTRACT FROM AUTHOR]

An algorithm that is capable not only of tracking multiple targets but also of "track management"-meaning that it does not need to know the number of targets as a user input-is of considerable interest. In this paper we devise a recursive track-managed filter via a quantized state-space ("bin") model. In the limit, as the discretization implied by the bins becomes as refined as possible (infinitesimal bins) we find that the filter equations are identical to Mahier's probability hypothesis density (PHD) filter, a novel trackmanaged filtering scheme that is attracting increasing attention. Thus, one contribution of this paper is an interpretation of, if not the PHD itself, at least what the PHD is doing. This does offer some intuitive appeal, but has some practical use as well: with this model it is possible to identify the PHD's "target-death" problem, and also the statistical inference structures of the PHD filters. To obviate the target death problem, PHD originator Mahler developed a new "cardinalized" version of PHD (CPHD). The second contribution of this paper is to extend the "bin-occupancy" model such that the resulting recursive filter is identical to the cardinalized PHD filter. [ABSTRACT FROM AUTHOR]

This paper considers a noncooperative game in which competing users sharing a frequency-selective interference channel selfishly optimize their power allocation in order to improve their achievable rates. Previously, it was shown that a user having the knowledge of its opponents' channel state information can make foresighted decisions and substantially improve its performance compared with the case in which it deploys the conventional iterative water-filling algorithm, which does not exploit such knowledge. This paper discusses how a foresighted user can acquire this knowledge by modeling its experienced interference as a function of its own power allocation. To characterize the outcome of the multiuser interaction, the conjectural equilibrium is introduced, and the existence of this equilibrium for the investigated water-filling game is proven. Importantly, we show that both the Nash equilibrium and the Stackelberg equilibrium are special cases of the conjectural equilibrium. We also develop practical algorithms to form accurate beliefs and select desirable power allocation strategies. Numerical simulations indicate that a foresighted user without any a priori knowledge of its competitors' private information can effectively learn how the other users will respond to its actions, and induce the entire system to an operating point that improves both its own achievable rate as well as the rates of the other participants in the water-filling game. [ABSTRACT FROM AUTHOR]

Constrained least-squares design and constrained Chebyshev design of one- and two-dimensional nonlinear-phase FIR filters with prescribed phase error are considered in this paper by a unified semi-infinite positive-definite quadratic programming approach. In order to obtain unique optimal solutions, we propose to impose constraints on the complex approximation error and the phase error. By introducing a sigmoid phase-error constraint bound function, the group-delay error can be greatly reduced. A Goldfarb-Idnani based algorithm is presented to solve the semi-infinite positive-definite quadratic program resulting from the constrained least-squares design problem, and then applied after some modifications to the constrained Chebyshev design problem, which is proved in this paper to be equivalent also to a semi-infinite positive-definite quadratic program. Through design examples, the proposed method is compared with several existing methods. Simulation results demonstrate the effectiveness and efficiency of the proposed method. [ABSTRACT FROM AUTHOR]

This paper presents a new approach to rejection of complex-valued sinusoidal disturbances acting at the output of a discrete-time stable linear plant with unknown and possibly time-varying dynamics. It is assumed that both the instantaneous frequency of the sinusoidal disturbance and its amplitude may be slowly varying with time and that the output signal is contaminated with wideband measurement noise. It is not assumed that a reference signal, correlated with the disturbance, is available. The proposed disturbance rejection algorithm is an extension of the algorithm derived for the constant-known-frequency case, described in Part I of this paper. [ABSTRACT FROM AUTHOR]

This paper presents an analysis of the steady-state mean-square error (MSE) of the set-membership normalized least-mean square (SM-NLMS) algorithm with relaxation and regularization parameters. These parameters are introduced for the purpose of deriving in a unified way the steady-state MSE performances of the ϵ-normalized least mean square (ϵ-NLMS) algorithm and a special case of the adaptive parallel subgradient projection (PSP) algorithm. The approach of the paper is to employ the energy conservation relation as a starting point of our analysis. This relation enables us to avoid the transient analysis of the SM-NLMS algorithm, which is in general hard due to the nonlinearity of the SM-NLMS algorithm. As a result, a few nonlinear equations whose solutions are theoretical steady-state MSEs are derived, where two types of reasonable assumptions are introduced to overcome the nonlinearity of the SM-NLMS algorithm. Our results are generalizations of well-known results of the steady-state MSE of the ϵ-NLMS. Extensive simulations show the close agreement between our theories and experiments. [ABSTRACT FROM AUTHOR]

This paper proposes an approach to deducing the floor plan of a building using only a set of noisy room dimension measurements that have been obtained indirectly through radar probes. It is not assumed that the set of measurements is complete. The key feature of this approach is a method of relating, via a graph-theoretical construction, a given floor plan topology to a model that depends linearly on a minimal set of parameters. Given a list of floor-plan topologies and a system of realizability constraints, it is possible to formulate this problem as a con- strained least-squares optimization problem. An algorithm for generating all floor plan topologies with a specified number of rooms is presented, as well as an algorithm for deciding if two floor plan representations are isomorphic. The techniques in this paper are intended to serve as an exploration into the theoretical performance bounds on the floor plan estimation problem. [ABSTRACT FROM AUTHOR]

This paper considers the noncooperative maximization of mutual information in the vector Gaussian interference channel in a fully distributed fashion via game theory. This problem has been widely studied in a number of works during the past decade for frequency-selective channels, and recently for the more general multiple-input multiple-output (MIMO) case, for which the state-of-the art results are valid only for nonsingular square channel matrices. Surprisingly, these results do not hold true when the channel matrices are rectangular and/or rank deficient matrices. The goal of this paper is to provide a complete characterization of the MIMO game for arbitrary channel matrices, in terms of conditions guaranteeing both the uniqueness of the Nash equilibrium and the convergence of asynchronous distributed iterative waterfilling algorithms. Our analysis hinges on new technical intermediate results, such as a new expression for the MIMO waterfilling projection valid (also) for singular matrices, a mean-value theorem for complex matrix-valued functions, and a general contraction theorem for the multiuser MIMO watefilling mapping valid for arbitrary channel matrices. The quite surprising result is that uniqueness/convergence conditions in the case of tall (possibly singular) channel matrices are more restrictive than those required in the case of (full rank) fat channel matrices. We also propose a modified game and algorithm with milder conditions for the uniqueness of the equilibrium and con- vergence, and virtually the same performance (in terms of Nash equilibria) of the original game. [ABSTRACT FROM AUTHOR]

Recently, the duality between multiple-input multiple-output (MIMO) multiple-access channels (MAC) and MIMO broadcast channels (BCs) has been established under a total power constraint. The same set of rates for MAC can be achieved in BC exploiting the MAC-BC duality formulas while preserving the total power constraint. In this paper, we describe the BC optimal power allocation applying this duality in a downstream x-digital subscriber lines (xDSLs) context under a total power constraint for all modems over all tones. Then, a new algorithm called BC-optimal spectrum balancing (BC-OSB) is devised for a more realistic power allocation under per-modem total power constraints. The capacity region of the primal BC problem under per-modem total power constraints is found by the dual optimization problem for the BC under per-modem total power constraints which can be rewritten as a dual optimization problem in the MAC by means of a precoder matrix based on the Lagrange multipliers. We show that the duality gap between the two problems is zero. The multiuser power allocation problem has been solved for interference channels and MAC using the OSB algorithm. In this paper we solve the problem of multiuser power allocation for the BC case using the OSB algorithm as well and we derive a computational efficient algorithm that will be referred to as BC-OSB. Simulation results are provided for two VDSL2 scenarios: the first one with differential-mode (DM) transmission only and the second one with both DM and phantom-mode (PM) transmissions. [ABSTRACT FROM AUTHOR]

Very often, random signals are modeled as autoregressive moving-average (ARMA) processes in engineering and scientific applications. The acquisition of the second-order statistics or the calculation of the autocorrelation function for an ARMA process is very important in those applications. In this paper, we discuss two kinds of ARMA-autocorrelation computation approaches, namely the recursive approach and the direct approach. To overcome the problems of memory usage and computational burden, we design a new ARMA-autocorrelation calculation algorithm which belongs to the direct approach. Our novel algorithm originates from the partial fractional decomposition for the rational trigonometric functions and the associated definite integrals. For comparison, we also provide the theoretical analysis of computational complexity for our new scheme and the conventional recursion-based algorithm. From our studies, it can be shown that our new ARMA-autocorrelation calculation method is more efficient than the conventional algorithm when the underlying autocorrelation function is long or the corresponding power spectrum possesses the narrowband characteristics. On the other hand, the conventional algorithm is more efficient than our new algorithm when the model orders of the ARMA process are large' and the underlying autocorrelation function is short (the corresponding power spectrum is wideband). Since our proposed new scheme involves the root-finding of a polynomial, the effect, of the error-tolerance in solving polynomials on the induced complexity is also discussed in this paper. [ABSTRACT FROM AUTHOR]

Abstract-This paper studies the blind block synchronization problem in block transmission systems using linear redundant precoders (LRP). Two commonly used LRP systems, namely, zero padding (ZP) and cyclic prelix (CP) systems, are considered in this paper. In particular, the block synchronization problem in CP systems is a broader version of timing synchronization problem in the popular orthogonal frequency division multiplexing (OFDM) systems. The proposed algorithms exploit the rank deficiency property of the matrix composed of received blocks when the block synchronization is perfect and use a parameter called repetition index which can be chosen as any positive integer. Theoretical results suggest advantages in blind block synchronization performances when using a large repetition index. Furthermore, unlike previously reported algorithms, which require a large amount of received data, the proposed methods', with properly chosen repetition indices, guarantee correct block synchronization in absence of noise using only two received blocks in ZP systems and three in CP systems. Computer simulations are conducted to evaluate the performances of the proposed algorithms and compare them with previously reported algorithms. Simulation results not only verify the capability of the proposed algorithms to work with limited received data but also shown significant improvements in the block synchronization error rate performance of the proposed algorithms over previously reported algorithms. [ABSTRACT FROM AUTHOR]

Abstract-This paper compares the complexity of the sphere decoder (SD) and a previously proposed detection scheme, denoted here as block SD (BSD), when they are applied to the detection of multiple-input multiple-output (MIMO) systems `in frequency-selective channels. The complexity of both algorithms depends on their preprocessing and tree search stages. Although the BSD was proposed as a means of greatly reducing the complexity of the preprocessing stage of the SD, no study was done on how the complexity of the tree search stage could be affected by that reduced preprocessing stage. This paper shows, both analytically and through simulation, that the reduction iü preprocessing complexity provided by the BSD has the side effect of increasing the complexity of its tree search stage compared to that of the SD, independent of the signal-to-noise ratio (SNR). In addition, this paper shows how sorting the columns of the frequency-selective channel matrix in the SD does not reduce the complexity of the tree search stage, contrary to what occurs in frequency-flat channels. [ABSTRACT FROM AUTHOR]

Abstract-Sensor position uncertainty is known to degrade significantly the source localization accuracy. This paper investigates the use of a single calibration emitter, whose position is known to the sensor array, to reduce the loss in localization accuracy due to sensor position errors that are random. Using a Gaussian noise model, we first derive the Cramér-Rao lower bound (CRLB) for a time difference of arrival (TDOA)-based source location estimate with the use of a calibration source. The differential calibration technique that is commonly used in Global Positioning System through the use of a calibration source to mitigate the inaccuracy in satellite ephemeris data is analyzed. The analysis indicates that differential calibration in most cases cannot reach the CRLB accuracy. The paper then proceeds to propose an algebraic closed-form solution for the source location estimate using both TDOA measurements from the unknown and the calibration source. The proposed algorithm is shown analytically, under high signal-to-noise ratio (SNR) and small sensor position noise, or under moderate level of SNR and sensor position noise together with distant unknown and calibration sources, to reach the CRLB accuracy. Simulations are used to corroborate and support the theoretical development. [ABSTRACT FROM AUTHOR]

Stochastic decoding is a new approach to iterative decoding on graphs. This paper presents a hardware architecture for fully parallel stochastic low-density parity-check (LDPC) de- coders. To obtain the characteristics of the proposed architecture, we apply this architecture to decode an irregular state-of-the-art (1056,528) LDPC code on a Xilinx Virtex-4 LX200 field-pro- grammable gate-array (FPGA) device. The implemented decoder achieves a clock frequency of 222 MHz and a throughput of about 1.66 Gb/s at Eb/No = 4.25 dB (a bit error rate of 10-8). It provides decoding performance within 0.5 and 0.25 dB of the floating-point sum-product algorithm with 32 and 16 iterations, respectively, and similar error-floor behavior. The decoder uses less than 40% of the lookup tables, flip-flops, and 10 ports available on the FPGA device. The results provided in this paper validate the potential of stochastic LDPC decoding as a practical and competitive fully parallel decoding approach. [ABSTRACT FROM AUTHOR]

The paper presents two formulations of causal cubic splines with equidistant knots. Both are based on a causal direct B-spline filter with parallel or cascade implementation. In either implementation, the causal part of the impulse response is realized with an efficient infinite-impulse-response (HR) structure, while only the anticausal part is approximated with a finite-order finite-impulse-response (FIR) filter. Resulting cubic coefficients are computed from the causal B-spline coefficients by using a third-order output FIR filter with either single-input multiple-output (SIMO) or multiple-input multiple-output (MIMO) structure, depending on the chosen formulation of the cubic spline. The paper demonstrates and proves that the properties of the resulting causal splines are quite different, whether they are based on a more popular B-spline formulation, or a bit neglected tridiagonal matrix formulation. It is shown that the proposed low-complexity but accurate causal interpolators can be realized for many practical applications with the delay of only a few samples. [ABSTRACT FROM AUTHOR]

This paper presents a method to obtain a trigonometric polynomial that accurately interpolates a given band-limited signal from a finite sequence of samples. The polynomial delivers accurate approximations in the range covered by the sequence, except for a short frame close to the range limits. Besides, its accuracy increases exponentially with the frame width. The method is based on using a band-limited window in order to reduce the truncation error of a convolution series. It is shown that the polynomial can be efficiently constructed and evaluated using algorithms designed for the discrete Fourier transform (DFT). Specifically, two basic procedures are presented, one based on the fast Fourier transform (FFT), and another based on a recursive update algorithm for the short-time FFT. The paper contains three applications. The first is a variable fractional delay (VFD) filter, which consists of a short-time FFT combined with the evaluation of a trigonometric polynomial. This filter has low complexity and can be implemented using CORDIC rotations. The second is the interpolation of nonuniform Fourier summations, where the proposed method eliminates the need to interpolate any kernel sample. Finally, the third can be viewed as a generalization of the FFT convolution algorithm and makes it possible to interpolate the output of an finite-impulse-response (FIR) filter efficiently. [ABSTRACT FROM AUTHOR]

As an example of the recently introduced concept of rate of innovation, signals that are linear combinations of a finite number of Diracs per unit time can be acquired by linear filtering followed by uniform sampling. However, in reality, samples are rarely noiseless. In this paper, we introduce a novel stochastic algorithm to reconstruct a signal with finite rate of innovation from its noisy samples. Even though variants of this problem have been approached previously, satisfactory solutions are only available for certain classes of sampling kernels, for example, kernels that satisfy the Strang-Fix condition. In this paper, we consider the infinite-support Gaussian kernel, which does not satisfy the Strang-Fix condition. Other classes of kernels can be employed. Our algorithm is based on Gibbs sampling, a Markov chain Monte Carlo method. Extensive numerical simulations demonstrate the accuracy and robustness of our algorithm. [ABSTRACT FROM AUTHOR]

This paper presents a way to access both the multiple-order and parameters of a multidimensional complex number autoregressive (AR) model through matrix factorization. The principle of this technique consists of the transformation of the multidimensional model to a pseudo simple-input simple-output AR model, then performing factorization of the covariance matrix of the data. This factorization then leads- to a recursive form of the parameter and order estimation. This paper makes two principal contributions. The first is a generalization of one dimensional factored form algorithm, and the second is that it makes it possible to access all the possible different orders and parameters of a multidimensional complex number AR model of any dimension, whereas classical approaches are limited to at most four-dimensional models. Computer simulation results are provided to illustrate the behavior of this method. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the existence and the algorithm analysis of an adaptive scheme that has been introduced for covariance structure matrix estimation in the context of adaptive radar detection under non-Gaussian noise. This latter has been modeled by spherically invariant random vector (SIRV), which is the product c of the square root of a positive unknown random variable T and an independent Gaussian vector x, c = √T x. A similar line of work was undertaken in the context of compound Gaussian noise, and this paper extends the previous results in the case of SIRV modeled noise. More precisely, the fixed-point estimate to be studied verifies a nonlinear algebraic equation (E) x = ƒ (x). The aim of this paper is twofold. First, we prove that (E) admits a unique solution ¯x; secondly, we show that the corresponding iterative algorithm xn+1 = ƒ(xn) converges to ¯x for every admissible initial condition. [ABSTRACT FROM AUTHOR]

This paper considers localization of a source or a sensor from distance measurements. We argue that linear algorithms proposed for this purpose are susceptible to poor noise performance. Instead given a set of sensors/anchors of known positions and measured distances of the source/sensor to be localized from them we propose a potentially nonconvex weighted cost function whose global minimum estimates the location of the source/sensor one seeks. The contribution of this paper is to provide nontrivial ellipsoidal and polytopic regions surrounding these sensors/anchors of known positions, such that if the object to be localized is in this region, localization occurs by globally exponentially convergent gradient descent in the noise free case. Exponential convergence in the noise free case represents practical convergence as it ensures graceful performance degradation in the presence of noise. These results guide the deployment of sensors/anchors so that small subsets can be made responsible for practical localization in geographical areas determined by our approach. [ABSTRACT FROM AUTHOR]

In this paper, we propose a generalized greedy (G-greedy) algorithm based on zero-forcing beamforming (ZFBF) for the multiple-input multiple-output (MIMO) broadcast channel. This algorithm serves as a general mathematical framework that includes a number of existing greedy user selection methods as its realizations. As previous results only give the scaling law of the sum rate of dirty paper coding (DPC), with the help of the G-greedy structure, we are able to obtain the exact limit of the DPC sum rate for a large number of users. We also prove that the difference between the sum rates obtained by G-greedy user selection and by DPC goes to zero as the number number of users increases. In addition to this, we investigate one particular greedy user selection scheme called sequential water-filling (SWF). For this algorithm, a complexity reduction is achieved byan iterative procedure based on an LQ decomposition, which converts the calculation of the Moore-Penrose matrix inverse to one vector-matrix multiplication. A sufficient condition is given to prune the search space of this algorithm that results in further complexity reduction. With the help of the G-greedy algorithm, we prove that SWF achieves the full DPC sum rate for a large number of users. For a moderate number of users, simulation demonstrates that, compared with other user selection algorithms, SWF achieves a higher sum rate that is close to the maximal sum rate achievable by ZFBF with the same order of complexity. [ABSTRACT FROM AUTHOR]

This paper presents a new implementation of the VAST algorithm for principal and minor subspace tracking. VAST was initially derived from the Subspace Projection (SP) algorithm by Davila, which was known for its exceptional convergence rate, compared with other classical principal subspace trackers. The novelty in the VAST algorithm was the lower computational cost (linear if the data correlation matrix satisfies a so-called shift-invariance property), and the extension to minor subspace tracking. However, the original implementation of the YAST algorithm suffered from a numerical stability problem (the subspace weighting matrix slowly loses its orthonormality). We thus propose in this paper a new implementation of VAST, whose stability is established theoretically and tested via numerical simulations. This algorithm combines all the desired properties for a subspace tracker: remarkably high convergence rate, lowest steady-state error, linear complexity, and numerical stability regarding the orthonormality of the subspace weighting matrix. [ABSTRACT FROM AUTHOR]

This paper proves that the mathematical complexity of running filters on semi-groups is C(p) = 3 - (6/(p + 1)) operations per sample, where p is the filter window. On other algebraic structures, the mathematical complexity of the filtering can be lower. Two such cases are investigated: the running filtering on max/mm lattice and on additions group. They correspond to max/mm and moving average filters. It is shown that the algorithms developed for semi-groups are of interest in such cases, too. Thus, the fastest deterministic algorithm for max/mm filtering known so far is based on an algorithm for running filtering on semi-groups. First, an optimized version is proposed in this paper. Then, further improvement for data-dependent max/mm algorithms is proposed. Finally, the case of moving average and exponential smoothing is investigated. Their implementation by using the algorithms for semi-groups is of lower complexity than the classical recursive implementation for the particular case of small size windows (p = 3 and p = 4). [ABSTRACT FROM AUTHOR]

The goal of this paper is to derive a novel online algorithm for classification in reproducing kernel hilbert spaces (RKHS) by exploiting projection-based adaptive filtering tools. The paper brings powerful convex analytic and set theoretic estimation arguments in machine learning by revisiting the standard kernel-based classification as the problem of finding a point which belongs to a closed halfspace (a special closed convex set) in an RKHS. In this way, classification in an online setting, where data arrive sequentially, is viewed as the problem of finding a point (classifier) in the nonempty intersection of an infinite sequence of closed halfspaces in the RKHS. Convex analysis is also used to introduce sparsification arguments in the design by imposing an additional simple convex constraint on the norm of the classifier. An algorithmic solution to the resulting optimization problem, where new convex constraints are added every time instant, is given by the recently introduced adaptive projected subgradient method (APSM), which generalizes a number of well-known projection-based adaptive filtering algorithms such as the classical normalized least mean squares (NLMS) and the affine projection algorithm (APA). Under mild conditions, the generated sequence of estimates enjoys monotone approximation, strong convergence, asymptotic optimality, and a characterization of the limit point. Further, we show that the additional convex constraint on the norm of the classifier naturally leads to an online sparsification of the resulting kernel series expansion. We validate the proposed design by considering the adaptive equalization problem of a nonlinear channel, and by comparing it with classical as well as with recently developed stochastic gradient descent techniques. [ABSTRACT FROM PUBLISHER]

The data of interest are assumed to be represented as N-dimensional real vectors, and these vectors are compressible in some linear basis B, implying that the signal can be reconstructed accurately using only a small number M ⪡ N of basis-function coefficients associated with B. Compressive sensing is a framework whereby one does not measure one of the aforementioned N-dimensional signals directly, but rather a set of related measurements, with the new measurements a linear combination of the original underlying N-dimensional signal. The number of required compressive-sensing measurements is typically much smaller than N, offering the potential to simplify the sensing system. Let f denote the unknown underlying N-dimensional signal, and g a vector of compressive-sensing measurements, then one may approximate f accurately by utilizing knowledge of the (under-determined) linear relationship between f and g, in addition to knowledge of the fact that f is compressible in B. In this paper we employ a Bayesian formalism for estimating the underlying signal f based on compressive-sensing measurements g. The proposed framework has the following properties: i) in addition to estimating the underlying signal f, "error bars" are also estimated, these giving a measure of confidence in the inverted signal; ii) using knowledge of the error bars, a principled means is provided for determining when a sufficient number of compressive-sensing measurements have been performed; iii) this setting lends itself naturally to a framework whereby the compressive sensing measurements are optimized adaptively and hence not determined randomly; and iv) the framework accounts for additive noise in the compressive-sensing measurements and provides an estimate of the noise variance. In this paper we present the underlying theory, an associated algorithm, example results, and provide comparisons to other compressive-sensing inversion algorithms in the literature. [ABSTRACT FROM AUTHOR]

One of the fundamental assumptions in traditional sampling theorems is that the signals to be sampled come from a single vector space (e.g., bandlimited functions). However, in many cases of practical interest the sampled signals actually live in a union of subspaces. Examples include piecewise polynomials, sparse representations, nonuniform splines, signals with unknown spectral support, overlapping echoes with unknown delay and amplitude, and so on. For these signals, traditional sampling schemes based on the single subspace assumption can be either inapplicable or highly inefficient. In this paper, we study a general sampling framework where sampled signals come from a known union of subspaces and the sampling operator is linear. Geometrically, the sampling operator can be viewed as projecting sampled signals into a lower dimensional space, while still preserving all the information. We derive necessary and sufficient conditions for invertible and stable sampling operators in this framework and show that these conditions are applicable in many cases. Furthermore, we find the minimum sampling requirements for several classes of signals, which indicates the power of the framework. The results in this paper can serve as a guideline for designing new algorithms for various applications in signal processing and inverse problems. [ABSTRACT FROM AUTHOR]

Low-rank matrix approximation has applications in many fields, such as 2-D filter design and 3-D reconstruction from an image sequence. In this paper, one issue with low-rank matrix approximation is investigated: heteroscedastic noise. In most of previous research, the covariance matrix of the heteroscedastic noise is assumed to be positive definite. This requirement restricts the usefulness of results derived from such research. In this paper, we extend the Wiberg algorithm, which originally deals with the missing data problem with low-rank approximation, to the cases, where the heteroscedastic noise has a singular covariance matrix. Experiments show that the proposed Wiberg algorithm converges much faster than the bilinear approach, and consequently avoids many nonconvergent cases in the bilinear approach. Experiments also show that, to some extent, the Wiberg algorithm can tolerate outliers and is not sensitive to parameter variation. [ABSTRACT FROM AUTHOR]

This paper analyzes the effects of quantization or other low-level noise on the error correcting capability of a popular class of real-number Bose-Chaudhuri-Hocquenghem (BCH) codes known as discrete Fourier transform (DFT) codes. In the absence of low-level noise, a modified version of the Peterson-Gorenstein-Zierler (PGZ) algorithm allows the correction of up to⌊(N - K)/2⌋ corrupted entries in the real-valued code vector of an (N, K) DFT code. In this paper, we analyze the performance of this modified PGZ algorithm in the presence of low-level (quantization or other) noise that might affect each entry of the code vector (and not simply ⌊(N - K)/2⌋ of them). We focus on the part of the algorithm that determines the number of errors that have corrupted the real-number codeword. Our approach for determining the number of errors is more effective than existing systematic approaches in the literature and results in an explicit lower bound on the precision needed to guarantee the correct determination of the number of errors; our simulations suggest that this bound can be tight. Finally, we prove that the optimal bit allocation for DFT codes (in terms of correctly determining the number of errors) is the uniform one. [ABSTRACT FROM AUTHOR]

The combination of the famed kernel trick and the least-mean-square (LMS) algorithm provides an interesting sample-by-sample update for an adaptive filter in reproducing kernel Hubert spaces (RKHS), which is named in this paper the KLMS. Unlike the accepted view in kernel methods, this paper shows that in the finite training data case, the KLMS algorithm is well posed in RKHS without the addition of an extra regularization term to penalize solution norms as was suggested by Kivinen [Kivinen, Smola and Williamson, "Online Learning With Kernels," IEEE Transactions on Signal Processing, vol. 52, no. 8, pp. 2165-2176, Aug. 2004] and Smale [Smale and Yao, "Online Learning Algorithms," Foundations in computational Mathematics, vol. 6, no. 2, pp. 145-176, 20061. This result is the main contribution of the paper and enhances the present understanding of the LMS algorithm with a machine learning perspective. The effect of the KLMS step size is also studied from the viewpoint of regularization. Two experiments are presented to support our conclusion that with finite data the KLMS algorithm can be readily used in high dimensional spaces and particularly in RKHS to derive nonlinear, stable algorithms with comparable performance to batch, regularized solutions. [ABSTRACT FROM AUTHOR]

In this paper, we consider the problem of blind identification of a convolutive multiple-input-multiple-output (MIMO) system with No outputs and Ni inputs. While many methods have been proposed to blindly identify convolutive MIMO systems with No ≥ Ni (overdetermined), very scarce results exist for the case of No < Ni (underdetermined), all of which refer to systems that either have some special structure or special No and Ni values. In this paper, we show that, as long as min(No, Ni) ≥ 2, independent of whether the system is overdetermined or underdetermined, we can always find the appropriate order of statistics that guarantees identifiability of the system response within trivial ambiguities. We also propose an algorithm to reach the solution, that consists of parallel factorization (PARAFAC) of a K-way tensor containing K th-order statistics of the system outputs, followed by an iterative scheme. For a certain order of statistics K, we provide the description of the class of identifiable MIMO systems. We also show that this class can be expanded by applying PARAFAC decomposition to a pair of tensors instead of one tensor. The proposed approach constitutes a novel scheme for estimation of underdetermined systems, and improves over existing approaches for overdetermined systems. [ABSTRACT FROM AUTHOR]

In this paper, a new deterministic technique for blind equalization of multiple-input multiple-output (MIMO) channels is presented. The proposed method relies on the channel order diversity of the finite impulse response-MIMO channel, and it does not make any assumption about the spectra or the finite alphabet property of the sources. Assuming that the lengths of the single-input multiple-output channels are known or estimated a priori, it is able to extract the sources up to a scale and rotation matrix ambiguity, which only affects to those sources associated to single-input multiple-output (SIMO) channels with equal lengths. Unlike previously proposed techniques, the method described in this paper obtains the equalizers and the best linear combination of their outputs in a single step and, thus, optimally. The reformulation of the proposed method as a set of nested canonical correlation analysis problems is exploited to obtain efficient batch and adaptive equalization algorithms. Finally, the performance of the proposed algorithms is evaluated by means of some simulation examples. [ABSTRACT FROM AUTHOR]

This paper considers the problem of piecewise linear prediction from a competitive algorithm approach. In prior work, prediction algorithms have been developed that are "universal" with respect to the class of all linear predictors, such that they perform nearly as well, in terms of total squared prediction error, as the best linear predictor that is able to observe the entire sequence in advance. In this paper, we introduce the use of a "context tree, to compete against a doubly exponential number of piecewise linear (affine) models. We use the context tree to achieve the total squared prediction error performance of the best piecewise linear model that can choose both its partitioning of the regressor space and its real-valued prediction parameters within each region of the partition, based on observing the entire sequence in advance, uniformly, for every bounded individual sequence. This performance is achieved with a prediction algorithm whose complexity is only linear in the depth of the context tree per prediction. Upper bounds on the regret with respect to the best piecewise linear predictor are given for both the scalar and higher order case, and lower bounds on the regret are given for the scalar case. An explicit algorithmic description and examples demonstrating the performance of the algorithm are given. [ABSTRACT FROM AUTHOR]

In this paper, we consider the problem of sampling signals that are nonband-limited but have finite number of degrees of freedom per unit of time and call this number the rate of innovation. Streams of Diracs and piecewise polynomials are the examples of such signals, and thus are known as signals with finite rate of innovation (FR!). We know that the classical ("band-limited sinc") sampling theory does not enable perfect reconstruction of such signals from their samples since they are not band-limited. However, the recent results on FR! sampling suggest that it is possible to sample and perfectly reconstruct such nonband-limited signals using a rich class of kernels. In this paper, we extend those results in higher dimensions using compactly supported kernels that reproduce polynomials (satisfy Strang-Fix conditions). In fact, the polynomial reproduction property of the kernel makes it possible to obtain the continuous moments of the signal from its samples. Using these moments and the annihilating filter method (Prony's method), the innovative part of the signal, and therefore, the signal itself is perfectly reconstructed. In particular, we present local (directional-derivatives-based) and global (complex-moments-based, Radon-transform-based) sampling schemes for classes of FRI signals such as sets of Diracs, bilevel, and planar polygons, quadrature domains (e.g., circles, ellipses, and cardioids), 2-fl polynomials with polygonal boundaries, and n-dimensional Diracs and convex poly- topes. This work has been explored in a promising way in super- resolution algorithms and distributed compression, and might find its applications in photogrammetry, computer graphics, and ma- chine vision. [ABSTRACT FROM AUTHOR]

This paper concerns underdetermined linear instantaneous and convolutive blind source separation (BSS), i.e., the case when the number P of observed mixed signals is lower than the number N of sources. We propose partial BSS methods, which separate P supposedly nonstationary sources of interest (while keeping residual components for the other N - P, supposedly stationary, "noise" sources). These methods are based on the general differential BSS concept that we introduced before. In the instantaneous case, the approach proposed in this paper consists of a differential extension of the FastICA method (which does not apply to underdetermined mixtures). In the convolutive case, we extend our recent time-domain fast fixed-point C-FICA algorithm to underdetermined mixtures. Both proposed approaches thus keep the attractive features of the FastICA and C-FICA methods. Our approaches are based on differential sphering processes, followed by the optimization of the differential nonnormalized kurtosis that we introduce in this paper. Experimental tests show that these differential algorithms are much more robust to noise sources than the standard FastICA and C-FICA algorithms. [ABSTRACT FROM AUTHOR]

In this paper, we present a new parametric modeling of three-dimensional homogeneous volumes of textured images (3-D texture) based on 3-D Wold decomposition for discrete processes. This decomposition additively separates any 3-D discrete homogeneous process into three mutually orthogonal components: a purely indeterministic, a "remote past"—type deterministic and an evanescent component. The evanescent component can be further decomposed into two orthogonal components called type 1 evanescent and type 2 evanescent. Assuming that the 3-D image volume is a finite realization of a discrete homogeneous stochastic process, the 3-D texture field can be uniquely represented as a sum of four mutually orthogonal components with different spectral properties. The aim of this paper is to present new parametric models able to describe both the spectral support and spatial characteristics of each component. A spectral decomposition algorithm to separate a 3-D texture into structured and unstructured parts is also proposed. [ABSTRACT FROM AUTHOR]

This paper presents the results of a series of four psychophysical experiments carried out to study the appearance, annoyance, and detectability of common digital video compression artifacts. The approach chosen in this paper was to use synthetic artifacts that look like "real" artifacts, yet are simpler, purer, and easier to describe. This approach allowed the control of the amplitude, distribution, and mixture of different types of artifacts. The algorithms for generating four of the most common types of artifacts: blockiness, blurriness, ringing, and noisiness, are described. The psychophysical experiments performed used video sequences containing different combinations of the generated synthetic artifacts. In these experiments, subjects were asked to detect any impairment and rate its annoyance. With the data gathered, the probability of detection and annoyance values were determined as a function of the total squared error. The results showed that "original video" (content) has a significant effect on both the detection threshold and the mid-annoyance parameter, while the "artifact signal type" does not. It was also found that these two parameters are highly correlated and linearly related. [ABSTRACT FROM AUTHOR]

This paper presents novel Q-learning based stochastic control algorithms for rate and power control in V-BLAST transmission systems. The algorithms exploit the supermodularity and monotonic structure results derived in the companion paper. Rate and power control problem is posed as a stochastic optimization problem with the goal of minimizing the average transmission power under the constraint on the average delay that can be interpreted as the quality of service requirement of a given application. Standard Q-learning algorithm is modified to handle the constraints so that it can adaptively learn structured optimal policy for unknown channel/traffic statistics. We discuss the convergence of the proposed algorithms and explore their properties in simulations. To address the issue of unknown transmission costs in an unknown time-varying environment, we propose the variant of Q-learning algorithm in which power costs are estimated in online fashion, and we show that this algorithm converges to the optimal solution as long as the power cost estimates are asymptotically unbiased [ABSTRACT FROM PUBLISHER]

Characterized by sudden structural changes, fault-prone systems are modeled using the framework of systems with switching parameters or hybrid systems. Since a closed-form mean-square optimal filtering algorithm for this class of systems does not exist, it is of particular interest to derive a lower bound on the state estimation error covariance. The well known Crameacuter-Rao bound is not applicable to fault-prone systems because of the discrete distribution of the fault indicators, which violates the regularity conditions associated with this bound. On the other hand, the Weiss-Weinstein lower bound is essentially free from regularity conditions. Moreover, a sequential version of the Weiss-Weinstein bound, suitable for Markovian dynamic systems, is presented by the authors in a companion paper. In the present paper, this sequential version is applied to several classes of fault-prone dynamic systems. The resulting bounds can be used to examine fault detectability and identifiability in these systems. Moreover, it is shown that several recently reported lower bounds for fault-prone systems are special cases of, or closely related to, the sequential version of the Weiss-Weinstein lower bound [ABSTRACT FROM PUBLISHER]

Stepping-stone attacks are often used by network intruders to hide their identities. In a stepping-stone attack, attacking commands are sent indirectly to the victim through a chain of compromised hosts acting as ‘stepping stones.’ In defending against such attacks, it is necessary to detect stepping-stone connections at the compromised hosts. The use of encrypted connections by the attacker complicates the detection problem and the attacker's active timing perturbation and insertion of chaff make it even more challenging. This paper considers strategies to identify stepping-stone connections when the attacker is able to encrypt the attacking packets and perturb their timing. Furthermore, the attacker can also add chaff packets in the attacking stream. The paper first considers stepping-stone connections subject to packet-conserving transformations by the attacker. Two activity-based algorithms are proposed to detect stepping-stone connections with bounded memory or bounded delay perturbation, respectively. These algorithms are proven to have exponentially decaying false alarm probabilities if normal traffic can be modelled as Poisson processes. It is shown that the proposed algorithms improve the performance of an existing stepping-stone detection algorithm. This paper then addresses the detection of stepping-stone connections with both timing perturbation and chaff. Robust algorithms are developed to deal with chaff evasion. It is proven that the proposed robust algorithms can tolerate a number of chaff packets proportional to the size of the attacking traffic, and have vanishing false alarm probabilities for Poisson traffic. Simulations using synthetic data are used to validate the theoretical analysis. Further results using actual Internet traces are shown to demonstrate the performance of the proposed algorithms. [ABSTRACT FROM AUTHOR]

-This paper proposes and examines the performance of an algorithm used to locate a finite time-duration acoustic source that may have a small time-bandwidth product. The localization algorithm hypothesizes a source location and aligns the sensor signals by removing the propagation delay for that hypothesized location. The energy of the sum of the aligned signals is calculated and the hypothesized source location with the maximum energy is the estimate of the source position. The primary contribution of this paper is the development of the expression for the variance of localization error in any specified direction. The variance expression is validated through a comparison with Monte Carlo simulations. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of blind separation of multiple independent sources from observed array output signals. The main contributions in this paper include an improved whitening scheme for estimation of signal subspace, a novel biquadratic contrast function for extraction of independent sources, and an efficient alterative method for joint implementation of a set of approximate diagonalization-structural matrices. Specifically, an improved whitening scheme is first developed by estimating the signal subspace jointly from a set of diagonalization-structural matrices based on the proposed cyclic maximizer of an interesting cost function. Moreover, the globally asymptotical convergence of the proposed cyclic maximizer is analyzed and proved. Next, a novel biquadratic contrast function is proposed for extracting one single independent component from a slice matrix group of any order cumulant of the array signals in the presence of temporally white noise. A fast fixed-point algorithm that is a cyclic minimizer is constructed for searching a minimum point of the proposed contrast function. The globally asymptotical convergence of the pro- posed fixed-point algorithm is analyzed. Then, multiple independent components are obtained by using repeatedly the proposed fixed-point algorithm for extracting one single independent component, and the orthogonality among them is achieved by the well- known QR factorization. The performance of the proposed algorithms is illustrated by simulation results and is compared with three related blind source separation algorithms. [ABSTRACT FROM AUTHOR]

Most of the signal processing operations involve some kind of a transformation or approximation of the signal. The transform coefficients or the approximation parameters reveal many hidden characteristics of a signal that can be appropriately processed to extract useful information. In recent years, adaptive time-frequency (TF) transformations have significantly contributed to this area. The TF transformation can be classified into two main categories based on 1) signal decomposition approaches and 2) bilinear TF distributions (also known as Cohen's class). TF distributions are nonparametric in nature and mainly used for visualization purposes. On the other hand, decomposition approaches are parametric in nature and highly suitable for objective feature extraction. This paper focuses on a particular TF decomposition approach (adaptive TF transformation) based on matching pursuit-type algorithm. Using this decomposition approach, a novel time-width versus frequency band (TWFB) energy mapping is proposed that possesses both parameterization benefits and meaningful visual patterns with favorable properties for pat- tern recognition. This organized mapping of the TF decomposition parameters allows the application of pruning algorithms such as local discriminant bases (LDB) to identify application specific TF subspaces. The identification of these subspaces enables efficient processing of information and reduces the computational effort considerably. The visual patterns of the TWFB mappings exhibit high potential of becoming a powerful pattern analysis tool. The paper covers in detail the formulation of the TWFB mapping and some of its properties. Experiments performed with speech and synthetic signals produced desirable results demonstrating the benefits of TWFB for pattern recognition related applications. [ABSTRACT FROM AUTHOR]