This study addresses a problem belonging to the domain of whale audio processing, more specifically the automatic classification of sounds produced by the Mysticete species. The specific task is quite challenging given the vast repertoire of the involved species, the adverse acoustic conditions and the nearly inexistent prior scientific work. Two feature sets coming from different domains (frequency and wavelet) were designed to tackle the problem. These are modelled by means of a hybrid technique taking advantage of the merits of a generative and a discriminative classifier. The dataset includes five species (Blue, Fin, Bowhead, Southern Right, and Humpback) and it is publicly available at http://www.mobysound.org/. The authors followed a thorough experimental procedure and achieved quite encouraging recognition rates.

This study deals with the issue of designing the sensing matrix for a compressed sensing (CS) system assuming that the dictionary is given. Traditionally, the measurement of small mutual coherence is considered to design the optimal sensing matrix so that the Gram of the equivalent dictionary is as close to the target Gram as possible, where the equivalent dictionary is not normalised. In other words, these algorithms are designed to solve the CS problem using an optimisation stage followed by normalisation. To achieve a global solution, a novel strategy of the sensing matrix design is proposed by using a gradient-based method, in which the measure of real mutual coherence for the equivalent dictionary is considered. According to this approach, a minimised objective function based on alternating minimisation is also developed through searching the target Gram within a set of relaxed equiangular tight frames. Some experiments are done to compare the performance of the newly designed sensing matrix with the existing ones under the condition that the dictionary is fixed. For the simulations of synthetic data and real image, the proposed approach provides better signal reconstruction accuracy.

Since it is very difficult to compare linear cutting tool marks quickly and quantitatively using existing image-processing and three-dimensional scanning methods, an adaptive matching algorithm for laser detection signals of linear cutting marks is proposed. Using locally weighted scatterplot smoothing regression, the proposed algorithm first performs noise reduction on the surface signals of linear cutting tool marks that are detected by a laser displacement sensor. Trends of the thick and consistent features of the signal data are then identified and the feature vectors are quantised via cosine vector curve fitting to individually calculate the spatial distances of the samples. Finally, the most similar samples are matched via batch similarity comparison using a dynamic programming (DP) algorithm. The accuracy and validity of the proposed algorithm are verified by similarity comparison tests of actual cutting marks from a variety of samples.

A digital filter is an essential structure in present day electronic devices. There is a variety of applications which demands tunability of digital filters in terms of bandwidth. It is desirable to have simple design process and minimum possible overhead in hardware implementation. A set of very low order subfilters derived from Farrow structure is proposed to function as a sample rate converter on the input signal. A fixed bandwidth low pass filter is placed in between these interpolator structures, resulting in a continuously variable bandwidth filter (CVF). This hardware efficient CVF architecture helps to achieve the tunability without altering the coefficients and the underlying structure. This model obtains a continuous variation in bandwidth employing a single sample rate change factor. Very low complexity and easy tunability of this technique are highly motivating factors to adopt CVF in various real-world applications.

This study addresses the coherent accumulation problem for detecting a manoeuvring target with complex motions, where range migration (RM) [i.e. range walk (RW) and range curvature (RC)] and Doppler frequency migration (DFM) occur during the coherent integration time. An efficient approach based on generalised keystone transform (GKT), radon transform (RT) and generalised dechirp process (GDP), i.e. GKT-RT-GDP, is presented to eliminate the RM and realise the coherent accumulation. More specifically, the GKT operation is first employed to remove the RC. Then, the RT is applied to estimate the trajectory slope for RW correction and velocity estimation. After that, GDP is introduced to obtain the estimations of target's acceleration and acceleration rate motion. Thereafter, the DFM caused by target's high-order motions can be compensated and then the coherent integration can be realised via Fourier transform. The advantage of the presented algorithm is that it can obtain a good balance between the computation cost and the detection performance, in comparison with the existing coherent integration algorithms.

The directional adaptive kernel distribution as a new time-frequency analysis method is proposed to analyse the vibration signal of rotor-bearing system. This method extends the adaptive kernel distribution to process the constructed complex-valued signal by defining directional ambiguity function (DAF). Based on the cyclic autocorrelation analysis and the complex-valued signal decomposition, the DAF is proposed, which is the product of the adaptive optimal kernel function and directional cyclic autocorrelation function. The kernel function taken as a two-dimension filter is optimised and used to suppress the cross-terms in the DAF. Then, the directional adaptive kernel distribution is obtained through the positive and inverse Fourier transform on the DAF. The new time-frequency transform is used to analyse the lateral vibration signals of the rotor and the bearing pedestal operating at oil whirling and whipping speeds. The experimental results verified that the proposed method is effective in the characterisation of the fault instantaneous characteristic frequency, rub-impact information, instantaneous planar motion and modulation information etc. in oil-film instability states of rotor-bearing system.

In this study, the authors propose and analyse a novel leak-detection method based on the ‘Haar’ continuous wavelet transform (CWT) and a double thresholding, i.e. CWTDT. Inspired by the idea of the binary integration technique in radar target detection, the algorithm analyses the non-stationary vibration signal issued from a water pipeline through which it decides whether or not there exists a leak in the water conveyance. To achieve this, the signal is first divided into several segments. Partial binary decisions within each segment are then obtained through the use of two preselected thresholds. The final binary decision is obtained by means of the ‘K out of L’ fusion rule. In doing this, the hardware leak system prototype is designed and a number of desirable leak positions in the water pipeline are first created to achieve the two best thresholds and ‘K out of L’ fusion rule. For comparison purposes, the performances of the proposed CWTDT method are assessed experimentally against those of the existing fast Fourier transform- and CWT-based methods under real operating conditions.

The temperature distribution in real-world industrial environments is often in a three-dimensional (3D) space, and developing an efficient algorithm to reconstruct such volume information may be beneficial for increasing the system efficiency, improving the energy saving and reducing the pollutant emission. A new algorithm is put forward for 3D temperature distribution reconstruction (3DTDR) tasks according to the given finite temperature observation data. Owing to the distinct advantages, including fast learning speed, good generalisation ability etc. a new robust regularised extreme learning machine (RRELM) algorithm is developed for the 3DTDR task. Unlike existing tomography-based measurement methods and local point measurement technologies, the proposed algorithm can reconstruct 3D temperature distributions on the basis of the given local temperature observation information. Furthermore, different from available inverse heat transfer problems, the proposed reconstruction method does not solve complicated partial differential equations. Numerical simulation results verify the effectiveness of the RRELM algorithm for 3DTDR problems. Furthermore, the proposed RRELM method has a satisfactory robustness for the noises in the measurement data and can achieve the high-quality reconstruction when the sampling ratio is low. As a result, an effective algorithm is introduced for the 3DTDR problem.

In this study, the authors propose reliable sequences of binary coded symbol (BCS) modulation, and their characteristics and performances for Global Navigation Satellite System (GNSS) application are described. A BCS sequence vector is formed by eight variable length sub-chips of alternated + 1 and −1 (or −1 and + 1) values. A judicious choice of the sub-chips lengths of the BCS sequence permitted to propose several BCS sequences that provide high performances in terms of multipath mitigation, resistance to the noise and interferences rejection. An overview of the essential characteristics and the resulting autocorrelation functions (ACFs) and power spectral densities of the proposed BCS sequences were introduced. The latter ACFs have a sharp main peak due to the increase in the number of transitions of the BCS sequences within a chip interval, which corresponds to a larger slope of the discrimination function, and consequently a reduced range of search in the delay locked loop with a minimum calculation load. The theoretical and simulation results indicate that the proposed BCS sequences are more consistent compared to the conventional signals adopted by the GNSS navigation systems.

In this study, the authors present an efficient time-dimension-reduced track-before-detect (TR-TBD) processor for slow-moving weak multi-targets detection in strong clutter environment. In their proposed framework, they elaborate observations from multiple frames (or scans) and resample them in time direction, then distinguish the slow-moving targets from the clutter in the radon parameter domain by exploiting the fact that different velocities of targets have different skewing angles corresponding to their tracks in the range–time (range–pulse) plane. To further enlarge the skewing angles differences between the slow-moving targets and the clutter, TR-TBD is proposed by incorporating the time-dimension reduction operator. This is very helpful to amplify the skewing angle of slow-moving targets, while the improvement is very small for the clutter. Therefore, it is much convenient to figure out the slow-moving weak targets from heavy clutter environment based on their amplified skewing angle differences by setting proper threshold. After detecting the targets, CLEAN-based track recovery method is proposed to eliminate the false tracks and recover the true tracks. Experimental results on real-data demonstrate that the proposed algorithm can detect the closely spaced targets and eliminate the false tracks under low signal-to-noise ratio and signal-to-clutter ratio.

Recasting the direction of arrival (DOA) estimation problem into sparse model is a subject of many researches, which has been carried out by different methods. Using the compressive sensing (CS) multiple measurement vector recovery algorithms have improved the resolution of DOA estimation in comparison with conventional methods such as multiple signal classification (MUSIC). In addition, a recently introduced hybrid method using CS and array processing named as ℓ_1,2-MUSIC exhibits high resolution as well as capability of resolving coherent sources. However, with recent advances in technology and employing higher order polynomial sources due to non-linear frequency characteristic, the performance of conventional methods is poor when the size of the array is small or the noise level is high. A novel hybrid method is presented based on matching pursuit (MP) signal decomposition and the ℓ_1,2-MUSIC algorithm. The new approach named Mℓ_1,2-MUSIC takes MP decomposition coefficients of the signal into account and applies the ℓ_1,2-MUSIC to achieve DOA estimation. Along with improving the resolvability of closely located non-stationary sources (high ordered polynomial phase) with small-sized sensor array, the main advantage of the proposed method is robustness to source coherency.

Public telephone systems transmit speech across a limited frequency range, about 300–3400 Hz, called narrowband (NB) which results in a significant reduction of quality and intelligibility of speech. This study proposes a fully backward compatible novel method for bandwidth extension (BWE) of NB speech. The method uses integer wavelet transform technique to provide a perceptually better wideband (WB) speech signal. The spectral envelope parameters are extracted from the down sampled frequency shifted version of the high-frequency components of speech signal existing above NB, which are spread by using pseudo-noise codes, and are embedded in the integer wavelet coefficients of NB speech signal. The hearing threshold is calculated in the integer wavelet domain and this threshold is employed as the embedding threshold. The embedded information is extracted at the receiving end to reconstruct the WB speech signal. Theoretical and simulation analyses show that the proposed method is robust to quantisation and channel noises. The comparison category rating listening and log spectral distortion tests clearly show that the reconstructed WB signal gives a much better performance in terms of speech quality when compared with some of the existing speech BWE methods employing data hiding.

Multipath (MP) interference in urban canyons is one of the major sources of the positioning error. Among different methods for MP mitigation, vectorised receiver (VR) is a promising one in which channels can share their information, and as a result the stronger channels can aid the weaker or affected ones to be tracked more accurately. This study proposes a weighted VR (WVR) with the strategy of giving more weights to the observations with lower vulnerability to MP. To increase the number of immune satellites to MP, global positioning system, and global navigation satellite system integration in WVR has been discussed. The performance is also compared with conventional VR. The experimental results show that the proposed method both in a static mode under an exaggerated MP condition and in a drive through an urban canyon could find the position, respectively, with about 10 and 13% improvements.

Empirical mode decomposition (EMD) is extensively realised in its potential of non-parametric signal denoising. Ensemble EMD (EEMD) is an improved self-adapting signal decomposition approach that can produce signal components with no frequency aliasing. In this study, the interval thresholding and iteration operation of EMD-based denoising techniques are applied to the EEMD and found not entirely feasible in the EEMD case. A modified interval thresholding is proposed, which can be adjustable for the intrinsic mode functions from EEMD. By taking advantage of the characteristics of EEMD, the internal and external iterations are compared and properly adopted in the EEMD-based denoising strategy. As a result, the EEMD-based denoising methods are proposed by combining the modified interval thresholding and the iterations. The denoising results on synthetic and real-life signals indicate that the presented methods exhibit better performance comparing with EMD-based methods, especially for signals with low signal-to-noise ratio. Based on the time complexities of the proposed methods, the acceptable sampling frequencies of the methods in real-time denoising are given.

This study presents a speech enhancement technique to improve noise corrupted speech via deep neural network (DNN)-based linear predictive (LP) parameter estimations of speech and noise. With regard to the LP coefficient estimation, an enhanced estimation method using a DNN with multiple layers was proposed. Excitation variances were then estimated via a maximum-likelihood scheme using observed noisy speech and estimated LP coefficients. A time-smoothed Wiener filter was further introduced to improve the enhanced speech quality. Performance was evaluated via log spectral distance, a composite multivariate adaptive regression splines modelling-based measure, and a segmental signal-to-noise ratio. The experimental results revealed that the proposed scheme outperformed competing methods.

In this study, the problem of filtering for sampled-data systems under unreliable communication links is investigated. The phenomena of data packet dropouts and signal transmission delays are addressed in a unified framework. The objective is to design an admissible filter to guarantee the asymptotic stability of the filtering error system and minimise the disturbance attenuation level. Thanks to the proposed novel Lyapunov–Krasovskii functional together with the improved Wirtinger's inequality and the reciprocally convex approach, novel sufficient linear-matrix-inequality-based conditions are obtained for the existence and design of admissible filters. Finally, two simulation examples are provided to illustrate the efficiency and less conservativeness of the proposed filter design methods.

A non-intrusive speech quality assessment method for complex environments was proposed. In the proposed approach, a new sparse representation-based speech reconstruction algorithm was presented to acquire the quasi-clean speech from the noisy degraded signal. Firstly, an over-complete dictionary of the clean speech power spectrum was learned by the K-singular value decomposition algorithm. Then in the sparse representation stage, the stopping residue error was adaptively achieved according to the estimated cross-correlation and the noise spectrum which was adjusted by a posteriori SNR-weighted factor, and the orthogonal matching pursuit approach was applied to reconstruct the clean speech spectrum from the noisy speech. The quasi-clean speech was considered as the reference to a modified PESQ perceptual model, and the mean opinion score of the noisy degraded speech was achieved via the distortions estimation between the quasi-clean speech and the degraded speech. Experimental results show that the proposed approach obtains a correlation coefficient of 0.925 on NOIZEUS complex environment database, which is 99% similar to the performance of the intrusive standard ITU-T PESQ, and 7.1% outperforms non-intrusive standard ITU-T P.563.