This study is concerned with the problem of robust reliable dissipative filter design for networked control systems (NCSs) with sensor failures and random packet dropouts. The considered NCS model is subject to the sources of uncertainty in the system parameters. The sensor signals are modelled by sequences of a Bernoulli distributed white sequence and the packet dropouts may occur randomly during transmission. The main objective is to design a suitable reliable dissipative filter such that, for all network-induced imperfections, a resulting error system is robustly stochastically stable and strictly (𝒬, 𝒮, ℛ) dissipative. The results are obtained for known as well as unknown sensor failure rates, so the results are more general one because it can guarantee the dissipativity of system whether or not the sensor encounter failures. The sufficient conditions for existence of filters are derived in terms of linear matrix inequality (LMI) approach and the corresponding filter parameters can be obtained by solution to a set of LMIs, which can be easily solved by using some standard numerical packages. Finally, two numerical examples are given to illustrate the applicability and effectiveness of the proposed filter design.

The diagonal loading (DL) technique is the most widely used method to improve the robustness of the Capon beamformer in the presence of imprecise knowledge of the covariance matrix and the desired signal's steering vector. The selection of the DL level is challenging in practice and might depend on some user-defined parameters which are possibly hard to be determined. A fully automatic and training-free method for the DL level selection is herein presented to extract the desired signal with constant modulus, which is a common feature for communication signals. Simulated results of the beamforming performance have demonstrated the efficacy of the proposed method.

This study addresses the optimal spectrum sensing detection based on the complete or partial side-information on the signal and noise statistics. The use of the generalised-likelihood ratio test (GLRT) involves maximum-likelihood (ML) estimation of the nuisances. ML estimation of the unknowns is especially challenging for wideband cognitive radio because closed-form solutions are often not available. Based on the equivalence between the wideband regime and the low-signal-to-noise ratio regime, this study provides a general kernel framework for GLRT spectrum sensing. It is shown that any GLRT detector exclusively depends on the projection of the sample covariance matrix of the data onto a given underlying kernel that reflects the available side-information in the problem. The kernels in several scenarios of interest are derived, including the widespread single and multi-frequency channelisation cases. Theoretical interpretations and numerical results show the trade-off between detection performance and the degree of side-information on the most informative statistics for detection, that is, the modulation format and spectrum distribution of the primary users.

The authors propose a method of fast speaker clustering in which a distance (distance of feature matrix mean, DFMM) is first defined for characterising the similarities between any two clusters, and then an adaptive convergence threshold is introduced for terminating the procedure of speaker clustering. If the minimum of the DFMMs between any two clusters is smaller than the threshold, then they are merged. The above mergence of clusters is repeated until the minimum of the DFMMs between any two clusters is larger than the threshold. They conduct experiments on both shorter voice segments (≤ 3 s) and longer voice segments (> 3 s) to compare their method with state-of-the-art methods, agglomerative hierarchical clustering with Bayesian information criterion (AHC + BIC) and vector quantisation with spectral clustering. Experiments show that their method achieves the best results for clustering shorter voice segments, and also obtains satisfactory results for clustering longer voice segments in comparison with other two methods. What is more, their method is faster than other methods in all experimental cases. The initial results show that the hybrid methods by combining their method with the AHC + BIC obtain further improvement in terms of the F score.

This study presents a novel particle filtering approach for multiple sensor target tracking. In contrast to the standard form, each particle only uses the measurements received by a single selected sensor to estimate the mean and covariance of the target state. In order to do so, sensor selections are sampled using a particle filter and the hidden states are marginalised over. Finite number of sensors allows the exact calculation of the normalisation constant, thus the sampling can be done from the optimal importance distribution. In addition, an extension to the multiple sensor case of the probabilistic data association approach is also provided when clutter or false alarm is considered. Simulation examples that involve tracking a bearings-only target are provided to demonstrate the effectiveness of the proposed algorithms in critical situations where the single-sensor observability is lacking.

Gaussian mixture models (GMMs) are widely used in speech and speaker recognition. This study explores the idea that a mixture of skew Gaussians might capture better feature vectors that tend to have skew empirical distributions. It begins with deriving an expectation maximisation (EM) algorithm to train a mixture of two-piece skew Gaussians that turns out to be not much more complicated than the usual EM algorithm used to train symmetric GMMs. Next, the algorithm is used to compare skew and symmetric GMMs in some simple speaker recognition experiments that use Mel frequency cepstral coefficients (MFCC) and line spectral frequencies (LSF) as the feature vectors. MFCC are one of the most popular feature vectors in speech and speaker recognition applications. LSF were chosen because they exhibit significantly more skewed distribution than MFCC and because they are widely used [together with the related immittance spectral frequencies (ISF)] in speech transmission standards. In the reported experiments, models with skew Gaussians performed better than models with symmetric Gaussians and skew GMMs with LSF compared favourably with both skew symmetric and symmetric GMMs that used MFCC.

In this study, the authors investigate the problem of source localisation based on the time difference of arrival (TDOA) in a group of sensors. Aiming to minimise the squared range-difference errors, the problem leads to a quadratically constrained quadratic programme. It is well known that this approach results in a non-convex optimisation problem. By proposing a relaxation technique, they show that the optimisation problem would be transformed to a convex one which can be solved by semi-definite programming (SDP) and Lagrange multiplier methods. Moreover, these methods offer the exact solution of the original problem and the affirmation of its uniqueness. In contrast to other complicated state-of-the-art SDP algorithms presented in the TDOA localisation literature, the authors methods are derived in a few straightforward reformulations and insightful steps; thus, there are no confusing and unjustifiable changes in the main optimisation problem. Furthermore, complexity analysis and a new approach for performance analysis, which show the merit of their methods, are introduced. Simulations and numerical results demonstrate that the positioning estimators resulted from the proposed algorithms outperform existing SDP-based methods presented so far.

This study puts forward the problem of fast-rate fault detection based on multiple slow-rate sensors. A fast-rate residual generator with casuality constraint is established from the multi-sensor model. Parameters of the residual generator are determined via disturbance-decoupling based on left eigenvector assignment. It is found that the condition of disturbance-decoupling is related to the multi-rate sensor sampling nature. A numerical example is given to illustrate the effectiveness of the proposed residual generator.

This study is concerned with the problem of the elimination of overflow oscillations (EOOs) for two-dimensional (2D) digital filters with external interference. The main purpose is the presentation of a new unified criterion such that the underlying 2D digital filter is stable with a positive prescribed interference attenuation level. A performance index is proposed for the first time, which is referred to as generalised dissipativity property. By using this index and two harmonic slack matrices, a novel criterion is established, which can be used to solve ℋ_∞ EOOs, passive EOOs and l ₂–l _∞ EOOs for 2D digital filters with external interference in a unified framework, and reduce the conservatism of the existing results. The effectiveness of the criterion is demonstrated by a numerical example.

This study describes a new approach to enhance noisy speech signals using the discrete Tchebichef transform (DTT) and the discrete Krawtchouk transform (DKT). The DTT and DKT are based on well-known orthogonal moments: the Tchebichef and Krawtchouk moments, respectively. The representations of speech signals using a limited number of moment coefficients and their behaviour in the domain of orthogonal moments are shown. The method involves removing noise from the signal using a minimum-mean-square error in the domain of the DTT or DKT. According to comparisons with traditional methods, the initial experiments yield promising results and show that orthogonal moments are applicable in the field of speech signal enhancement. The application of orthogonal moments could be extended to speech analysis, compression and recognition.