Cochannel interference and multipath propagation reduce the performance of mobile communication systems. Multi-input MLSE with whitening processing can mitigate the influence of the interference and provide path diversity gain. In conventional considerations, however, the required complexity rapidly rises with the number of array elements. In this paper, we propose multi-input MLSE that whitens error signals in the signal space by using a multibeam adaptive array. This scheme can reduce the computational load of multi-input MLSE than the conventional type when using a large-element array. The results of an analysis show that the proposed type is equivalent to conventional one in the sense of the metric and provides less computational complexity.

This paper presents a BER performance derivation considering imperfect channel estimation for a pilot-aided coherent forward link of W-CDMA system under multipath Rayleigh fading channels. In the forward link of the W-CDMA system, pilot signal is usually used for coherent demodulation. In this paper, the maximum likelihood estimator, Wiener filter, and moving average filter are applied to estimate the channel effect due to mobile speed and frequency offset. Then, we concentrate on determining optimal parameter values of the estimators such as the observation length, delay parameters for causal/non-causal filter, and filter resolution. Also it is verified that these parameters are closely associated with the performance, hardware complexity, and characteristics of OVSF code. In particular, effect of data rate and filter resolution on the BER performance is analyzed in more detail. In addition, we show the performance comparison between the estimators considering various imperfections. Finally, we verify the derived BER by using an extensive Monte-Carlo computer simulation.

This paper presents a maximum likelihood (ML) identification and restoration method for noisy blurred images. The unitary discrete sine transform (DST) is employed to decouple the large order spatial state-space representation of the noisy blurred image into a bank of one-dimensional real state-space scalar subsystems. By assuming that the noises are Gaussian distributed processes, the maximum likelihood estimation technique using the expectation-maximization (EM) algorithm is developed to jointly identify the blurring functions, the image model parameters and the noise variances. In order to improve the computational efficiency, the conventional Kalman smoother is incorporated to give the estimates. The identification process also yields the estimates of transformed image data, from which the original image is restored by the inverse DST. The experimental results show the effectiveness of the proposed method and its superiority over the recently proposed spatial domain DFT-based methods.

New algorithms for the soft-decision and the hard-decision maximum likelihood decoding (MLD) for binary linear block codes are proposed. It has been widely known that both MLD can be regarded as an integer programming with binary arithmetic conditions. Recently, Conti and Traverso have proposed an efficient algorithm which uses Grobner bases to solve integer programming with ordinary integer arithmetic conditions. In this paper, the Conti-Traverso algorithm is extended to solve integer programming with modulo arithmetic conditions. We also show how to transform the soft-decision and the hard-decision MLD to integer programming for which the extended Conti-Traverso algorithm is applicable.

In the last three decades of the 20th Century, research in speech recognition has been intensively carried out worldwide, spurred on by advances in signal processing, algorithms, architectures, and hardware. Recognition systems have been developed for a wide variety of applications, ranging from small vocabulary keyword recognition over dial-up telephone lines, to medium size vocabulary voice interactive command and control systems for business automation, to large vocabulary speech dictation, spontaneous speech understanding, and limited-domain speech translation. Although we have witnessed many new technological promises, we have also encountered a number of practical limitations that hinder a widespread deployment of applications and services. On one hand, fast progress was observed in statistical speech and language modeling. On the other hand only spotty successes have been reported in applying knowledge sources in acoustics, speech and language science to improving speech recognition performance and robustness to adverse conditions. In this paper we review some key advances in several areas of speech recognition. A bottom-up detection framework is also proposed to facilitate worldwide research collaboration for incorporating technology advances in both statistical modeling and knowledge integration into going beyond the current speech recognition limitations and benefiting the society in the 21st century.

In this paper, a new narrowband beamformer with derivative constraint is developed for wideband and coherent jammers suppression. The so-called IQML algorithm with linear constraint, which is used to estimate the unknown directions of the jammers in signal-free environment, is shown to be an inappropriate constraint estimator. In this paper, a new IQML algorithm with a norm constraint is considered, which is a consistent estimator and can be used to achieve desired performance. It can be also employed in the CDMA system for MAI suppression. We show that it outperforms the approach with the linear constraint used in the narrowband beamformer, in terms of directional pattern, output SINR and nulling capability for wideband and coherent jammers suppression.

This paper addresses a maximum likelihood method for source separation in the case of overdetermined mixtures corrupted by additive white Gaussian noise. We consider an approximate likelihood which is based on the Laplace approximation and develop a natural gradient adaptation algorithm to find a local maximum of the corresponding approximate likelihood. We present a detailed mathematical derivation of the algorithm using the Lie group invariance. Useful behavior of the algorithm is verified by numerical experiments.

In this paper, we present a new soft-decision iterative decoding algorithm using an efficient minimum distance search (MDS) algorithm. The proposed MDS algorithm is a top-down and recursive MDS algorithm, which finds a most likely codeword among the codewords at the minimum distance of the code from a given codeword. A search is made in each divided section by a "call by need" from the upper section. As a consequence, the search space and computational complexity are reduced significantly. The simulation results show that the proposed decoding algorithm achieves near error performance to the maximum likelihood decoding for any RM code of length 128 and suboptimal for the (256, 37), (256, 93) and (256, 163) RM codes.

A new modified maximum likelihood (ML) algorithm for frame synchronization in discrete multitone VDSL transmission system is presented. Computer simulation results are included to show its improvement in Et/N0 of each tone in the received data. This algorithm estimates the frame boundary at the initial transition edge rather than at the middle peak of a shortened twisted-pair channel response. The timing margin degradation caused by precursor intersymbol interference (ISI) can be reduced significantly, especially at the sub-channel loaded with more bits.

In this paper, a new maximum likelihood filter with finite impulse response (FIR) structures is proposed for state space signal models with both system and observation noises. This filter is called the maximum likelihood FIR (MLF) filter. The proposed MLF filter doesn't require a priori information of the window initial state and processes the finite observations on the most recent window linearly. The proposed MLF filter is first represented in a batch form, and then in an iterative form for computational advantage. The proposed MLF filter has good inherent properties such as time-invariance, unbiasedness, deadbeat, robustness. The validity of the proposed MLF filter is illustrated by a computer simulation on a sinusoidal signal.

In this semi-tutorial paper, the reliability-based decoding approaches using the reprocessing of the most reliable information set are investigated. This paper somehow homogenizes and compares former different studies, hopefully improving the overall transparency, and completing each one with tricks provided by the others. A couple of sensible improvements are also suggested. However, the main goal remains to integrate and compare recent works based on a similar general approach, which have unfortunately been performed in parallel without much efforts of comparison up to now. Their respective (dis)advantages, especially in terms of average or maximum complexity are elaborated. We focus on suboptimum decoding while some works to which we refer were developed for maximum likelihood decoding (MLD). No quantitative error performance analysis is provided, although we are in a position to benefit from some qualitative considerations, and to compare different strategies in terms of higher or lower expected error performances for a same complexity. With simulations, however, it turns out that all considered approaches perform very closely to each other, which was not especially obvious at first sight. The simplest strategy proves also the fastest in terms of CPU-time, but we indicate ways to implement the other ones so that they get very close to each other from this point of view also. On top of relying on the same intuitive principle, the studied algorithms are thus also unified from the point of view of their error performances and computational cost.

We propose a new solvable Markov random field model for Bayesian image processing and give the exact expressions of the marginal likelihood and the restored image by using the multi-dimensional Gaussian formula and the discrete Fourier transform. The proposed Markov random field model includes the conditional autoregressive model and the simultaneous autoregressive model as a special case. The estimates of hyperparameters are obtained by maximizing the marginal likelihood. We study some statistical properties of the solvable Markov random field model. In some numerical experiments for standard images, we show that the proposed Markov random field model is useful for practical applications in image restorations. The investigation of probabilistic information processing by means of a solvable probabilistic model is recently in progress not only for image processing but also for error correcting codes and so on. The solvable probabilistic model gives us some important aspects for the availability of probabilistic computational systems.

Y. S. Han et al. have proposed an efficient maximum likelihood decoding (MLD) algorithm using A* algorithm which is the graph search method. In this paper, we propose a new MLD algorithm for linear block codes. The MLD algorithm proposed in this paper improves that given by Han et al. utilizing codewords of dual codes. This scheme reduces the number of generated codewords in the MLD algorithm. We show that the complexity of the proposed decoding algorithm is reduced compared to that given by Han et al. without increasing the probability of decoding error.

We propose a blind source separation algorithm for the mixture of finite alphabet sources where sensors are less than sources. The algorithm consists of an update equation of an estimated mixing matrix and enumeration of the inferred sources. We present the bound of a step size for the stability of the algorithm and two methods of assignment of the initial point of the estimated mixing matrix. Simulation results verify the proposed algorithm.

The maximum likelihood estimate of a mixture model is usually found by using the EM algorithm. However, the EM algorithm suffers from a local optima problem and therefore we cannot obtain the potential performance of mixture models in practice. In the case of mixture models, local maxima often have too many components of a mixture model in one part of the space and too few in another, widely separated part of the space. To escape from such configurations we proposed a new variant of the EM algorithm in which simultaneous split and merge operations are repeatedly performed by using a new criterion for efficiently selecting the split and merge candidates. We apply the proposed algorithm to the training of Gaussian mixtures and the dimensionality reduction based on a mixture of factor analyzers using synthetic and real data and show that the proposed algorithm can markedly improve the ML estimates.

Noncoherent detectors for use in acquiring data-modulated direct-sequence spread-spectrum (DS/SS) signals are considered in this paper. Taking data modulation and timing uncertainty into account and using the generalized maximum likelihood (GML) or maximum likelihood (ML) detection approaches, we derive optimal detectors in the sense of Bayes or Neyman-Pearson and propose various suboptimal detectors. A simple systematic means for their realization is suggested and the numerical performance of these detectors is presented. We also compare their performance with that of the noncoherent combining (NC1) detector that had been proposed to serve the same need. Numerical results show that even the proposed suboptimal detectors can outperform the NC1 detector in most cases of interest.

In conventional methods for detecting vanishing points and vanishing lines, the observed feature points are clustered into collections that represent different lines. The multiple lines are then detected and the vanishing points are detected as points of intersection of the lines. The vanishing line is then detected based on the points of intersection. However, for the purpose of optimization, these processes should be integrated and be achieved simultaneously. In the present paper, we assume that the observed noise model for the feature points is a two-dimensional Gaussian mixture and define the likelihood function, including obvious vanishing points and a vanishing line parameters. As a result, the above described simultaneous detection can be formulated as a maximum likelihood estimation problem. In addition, an iterative computation method for achieving this estimation is proposed based on the EM (Expectation Maximization) algorithm. The proposed method involves new techniques by which stable convergence is achieved and computational cost is reduced. The effectiveness of the proposed method that includes these techniques can be confirmed by computer simulations and real images.

Tail-biting trellises of linear and nonlinear block codes are addressed. We refine the information-theoretic approach of a previous work on conventional trellis representation, and show that the same ideas carry over to tail-biting trellises. We present lower bounds on the state and branch complexity profiles of these representations. These bounds are expressed in terms of mutual information between different portions of the code, and they introduce the notions of superstates and superbranches. For linear block codes, our bounds imply that the total number of superstates, and respectively superbranches, of a tail-biting trellis of the code cannot be smaller than the total number of states, and respectively branches, of the corresponding minimal conventional trellis, though the total number of states and branches of a tail-biting trellis is usually smaller than that of the conventional trellis. We also develop some improved lower bounds on the state complexity of a tail-biting trellis for two classes of codes: the first-order Reed-Muller codes and cyclic codes. We show that the superstates and superbranches determine the Viterbi decoding complexity of a tail-biting trellis. Thus, the computational complexity of the maximum-likelihood decoding of linear block codes on a tail-biting trellis, using the Viterbi algorithm, is not smaller than that of the conventional trellis of the code. However, tail-biting trellises are beneficial for suboptimal and iterative decoding techniques.

A recursive maximum likelihood decoding (RMLD) algorithm is more efficient than the Viterbi algorithm. The decoding complexity of the RMLD algorithm depends on the recursive sectionalization. The recursive sectionalization which minimizes the decoding complexity is called the optimum sectionalization. In this paper, for a class of non-linear codes, called rectangular codes, it is shown that a near optimum sectionalization can be obtained with a dynamic programming approach. Furthermore, for a subclass of rectangular codes, called C-rectangular codes, it is shown that the exactly optimum sectionalization can be obtained with the same approach. Following these results, an efficient algorithm to obtain the optimum sectionalization is proposed. The optimum sectionalizations for the minimum weight subcode of some Reed-Muller codes and of a BCH code are obtained with the proposed algorithm.

Speech signals transmitted over telephone network often suffer from interference due to ambient noise and channel distortion. In this paper, a novel frame-dependent fuzzy channel compensation (FD-FCC) method employing two-stage bias subtraction is proposed to minimize the channel effect. First, through maximum likelihood (ML) estimation over the set of all word models, we choose the word model which is best matched with the input utterance. Then, based upon this word model, a set of mixture biases can be derived by averaging the cepstral differences between the input utterance and the chosen model. In the second stage, instead of using a single bias, a frame-dependent bias is calculated for each input frame to equalize the channel variations in the input utterance. This frame-dependent bias is achieved by the convex combination of those mixture biases which are weighted by a fuzzy membership function. Experimental results show that the channel effect can be effectively canceled even though the additive background noise is involved in a telephone speech recognition system.