Quality evaluation of the G.EV-VBR speech codec

Telephone networks normally transmit narrowband (NB) speech with a bandwidth restricted to 300 Hz to 3.4 kHz. While human speech comprises frequencies even further than the NB, there is a degradation of speech quality and intelligibility in such networks. This can be enhanced by recovering the speech components existing beyond the telephone bandwidth using Quadrature Mirror Filter (QMF) to give a perceptually superior wideband (WB) speech signal. QMF banks utilized as the fundamental mathematical tools to analyze and synthesize speech signals to coding of speech signal in sub-bands. Utilization of QMF's allows evading the aliasing effects due to samples decimation when signal is divide into sub-bands. In this paper, the performance of bandwidth extension (BWE) of speech through QMF was assessed by objective parameters like Log Spectral Distortion (LSD) and Itakura-Saito distance (ISD) and subjective parameters like Mean opinion score (MOS), Absolute Category Rating (ACR), Comparison Category Rating (CCR) and Perceptual Evaluation of Speech Quality (PESQ) which demonstrates that the recovered WB speech signal provides speech quality compared by original WB speech signal.

The paper presents a voice conversion framework that can be used in real-time applications. The conversion technique is based on hybrid (deterministic/stochastic) parametric speech representation. The conversion approach has been tested in two modifications for narrow-band and wide-band speech signals. Though the real-time requirement adds some significant limitations (frame by frame processing) the approach provides high quality of the reconstructed speech and recognizability of the target speaker's identity due to improved feature mapping. The proposed solution is embedded in a mobile communication system as an entertainment service.

We study superconductivity in the Hubbard model on various quasi-one-dimensional lattices with coexisting wide and narrow bands originating from multiple sites within a unit cell, where each site corresponds to a single orbital. The systems studied are the 2-leg and 3-leg ladders, the diamond chain, and the criss-cross ladder. These one-dimensional lattices are weakly coupled to form two-dimensional (quasi-one-dimensional) ones, and the fluctuation exchange approximation is adopted to study spin-fluctuation-mediated superconductivity. When one of the bands is perfectly flat, and the Fermi level, intersecting the wide band, is placed in the vicinity of the flat band, superconductivity arising from the interband scattering processes is found to be strongly enhanced owing to the combination of the light electron mass of the wide band and the strong pairing interaction due to the large density of states of the flat band. Even when the narrow band has finite band width, the pairing mechanism still works since the edge of the narrow band, due to its large density of states, plays the role of the flat band. The results indicate the generality of the high $T_c$ pairing mechanism due to coexisting wide and "incipient" narrow bands in quasi-one-dimensional systems.

This correspondence proposes a new technique for signal classification and jamming detection in wide-band (WB) radios. Theory of compressed sensing is exploited to recover the sparsely populated WB spectrum from sub-Nyquist samples, thus reducing the very high-rate sampling requirements at the receiver analog to digital converter. From the recovered WB, key spectral features of each narrow-band (NB) signal are extracted. These spectral features are then used to train a simple yet powerful classifier, the naive Bayes classifier (NBC). The trained NBC is then used not only to classify various NB signals into their respective modulations but also to detect the jamming on different NB signals, which are the main contributions of this letter. The proposed algorithm is then evaluated under different empirical setups and is shown to perform better when compared to a recently proposed feature-based jamming detection algorithm.

Successful management of the radio spectrum requires, as a first step, detailed information about spectrum occupancy. In this work, we present an end-to-end deep learning (DL) based framework to obtain information from wide spectrum bands through signal detection, localization, and modulation classification. By visually representing the radio signals in spectrograms, we formulate the wideband detection problem as an object detection task from the computer vision field. To this end, the proposed framework consists of two cascaded modules: an object detection network repurposed to detect and classify distinctive signals in wideband spectrograms, and a convolutional neural network (CNN) designed to extend the classification capabilities to support a wide range of analog and digital modulation schemes. To evaluate our framework, we use a public wideband recognition dataset, which we carefully analyze and curate through a series of preprocessing techniques. To tackle the challenges of insufficient training data and class imbalance observed in the dataset, we suggest a training strategy that includes data mixing and transfer learning. Our experimental results on a general test set demonstrate that the proposed approach can detect and classify a variety of narrowband signals with simultaneously high precision (77.1%), recall (81.8%), and localization accuracy, as indicated by an average Intersection over Union (IoU) of 86%.

A SEM cathodoluminescence study of quartz in mildly deformed granite from the region of the Atotsugawa fault, central Japan

In this paper, we study a bandwidth allocation strategy with state-dependent Bernoulli access (SDBA) and preemptive priority (PP) to serve wideband (WB) and narrowband (NB) traffics in wideband integrated networks. In this strategy, total bandwidth is divided into three regions, bandwidth for WB traffic, bandwidth for NB traffic and shared bandwidth. Under this environment, the minimum usable bandwidths for queueable NB traffic are controlled adaptively according to the system states. And the bandwidths used by the NB traffic are probabilistically preempted by a blockable WB traffic, when the number of NB messages waiting in the buffer is varying. The strategy is analyzed using the Neuts' matrix analytic approach. It shows better performance than other schemes which do not use SDBA and PP. In addition, the strategy with multiple buffer thresholds controls bandwidth allocation more dynamically and shows more performance build-up than the strategy with single buffer threshold.

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 paper proposes a fully backward compatible novel method for bandwidth extension of NB speech. The method uses magnitude spectrum data hiding technique to provide a perceptually better wideband speech signal. Code excited linear prediction (CELP) 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 low-amplitude high-frequency regions of the magnitude spectrum of NB speech signal. The embedded information is extracted at the receiving end to reconstruct the wideband speech signal. Theoretical and simulation analyses show that the proposed method is robust to quantization and channel noises. The log spectral distortion test clearly show that the reconstructed wideband signal gives a much better performance in terms of speech quality when compared to some of the existing speech bandwidth extension methods employing data hiding.

In this paper, the adaptive cancellation structure is first developed based on the LMS algorithm and FIR adaptive filtering. Then the novel practical noise and echo cancellation systems are built using the proposed adaptive technique and implemented using TX320TMS67C13 DSKs, which are Texas Instruments’ Digital Signal Processing (TI DSP) boards. Although adaptive filtering is an exciting topic in which many real-life applications can be explored [1]-[6], [9], building such a real-time system is often challenging due to the use of theoretical math, advanced DSP knowledge and practical industrial hands-on experience [1],[4]-[6],[9]. Therefore, this paper indicates that it is possible to apply traditional mathematics in adaptive filtering theory to real-time practical DSP systems. With the MATLAB software tool, we can simulate and verify various adaptive filtering designs first. Then, development and implementation of different noise or echo cancellation systems with adaptive filtering techniques can be successfully performed using the floating-point digital signal processor, TX320TMS67C13 DSK. Furthermore, it can be shown that TX320TMS67C13 DSKs with their stereo channels offer more effective and flexible tools for various noise cancellation applications.

In today's wireless communication system, quality of decoded speech at receiving end is found muffled and thin, mainly attributed to inherent band limitation (300-3400 Hz) and power constraints. In order to obtain toll quality of recovered speech in terms of intelligibility and naturalness in wireless systems, Narrowband (NB) speech coders should be upgraded to its counterpart Wideband (WB) coders (50-7000Hz). In the meantime, a novel and backward compatible solution is proposed that claims to artificially extend bandwidth of NB speech to WB at receiving end, popular as Artificial Bandwidth Extension (ABE). Out of many techniques which aim to mitigate the effect of the ever unpredictable channel conditions, Adaptive Multi Rate (AMR) NB coder is considered to be one of the potential candidates. Selection of particular bit rate mode (out of all eight bit rate modes between 4.75 kbps and 12.2 kbps) solely depends upon the channel condition. This paper discusses development of ABE algorithm for Code Excited Linear Prediction (CELP) based GSM AMR NB coder, and for the same, MATLAB based e-test bench is created for simulation. Such series of simulations are conducted to discover and gain insight about the overall performance of proposed ABE coder that includes subjective (Mean Opinion Score - MOS) and objective (Perceptual Evaluation of Speech Quality - PESQ) analyses. The evaluated results for both analyses clearly advocate that proposed ABE coder outperforms legacy GSM AMR 06.90 NB coder.

A static call access control scheme for a synchronous transfer mode (STM) multiplex system in a broadband network is investigated. The access resource, composed of N basic channels, is shared by narrowband (NB) calls that can be queued and wideband (WB) calls that can be blocked. NB and WB calls require 1 and m channels for 1/ mu /sub 1/ and 1/ mu /sub 2/ average holding times, respectively. The NB traffic has N/sub 1/ channels reserved for it, but it is allowed to compete with WB calls for the remaining resource if the queue exceeds a threshold h. Two approaches, one based on the matrix geometric method and the other on moment generating functions, are used to analyze the system. >

The number and proportion of neurons displaying GABA immunoreactivity were determined for 50-micron-wide columns through the thickness of 10 areas of monkey cerebral cortex, including the precentral motor area (area 4), 3 cytoarchitectonic fields of the first somatic sensory area (areas 3b, 1, and 2), 2 areas of parietal association cortex (areas 5 and 7), the first and second visual areas (areas 17 and 18), area 21 of the temporal lobe, and areas of the orbital and lateral frontal cortex. Methods of fixation and immunocytochemical processing were designed to maximize the number of stained cells in 15-micron-thick frozen sections and 1-micron-thick plastic sections. In 8 of the 10 areas the number and proportion of GABA-immunoreactive neurons per 50-micron-wide column were found to be the same (34-43 cells/column; 25% of the total neuronal population). Areas 17 and 3b differed. Area 17 contained 50% more GABA-immunoreactive neurons (52-66 cells/column) but more than twice the total number of neurons, so that the GABA cells made up less than 20% of the total. In 3 monkeys, the number and proportion of GABA-positive neurons per 50-micron-wide column in area 3b were smaller than in adjacent areas of sensorimotor cortex (26-42 cells/column; 19-22%). In 2 other monkeys, the number and proportion (34-43 cells/column; 24-26%) were the same as in adjacent areas. Despite the similarity among most areas of monkey cortex, within some areas, the number of GABA-positive neurons per 50-micron-wide column varied as much as 30%. These variations form a significant, repeating pattern only in area 18, where narrow bands (150-200 micron wide) of relatively few stained cells alternated with either narrow or wide bands (600-700 micron wide) in which columns contained more cells. The GABA-immunoreactive neurons were unevenly distributed across layers, with every area containing large numbers and proportions of stained cells in layer II, and every area but area 4 displaying a second concentration in the principal thalamocortical recipient layers. In area 4, the number of GABA-positive neurons declined sharply from layer II to layer III and remained low through layer VI. For areas displaying the greatest intra-areal variability, only 1 or 2 layers contributed significantly to that variability (layer IV in area 3b, layers III and V in area 18, and layers II and III in area 17).(ABSTRACT TRUNCATED AT 400 WORDS)

Real-time electromagnetic field distribution of devices under test (DUT) can be used to diagnose and analyze the system electromagnetic compatibility (EMC). An ultrawideband real-time monitoring system based on electro-optical under-sampling and data acquisition is proposed in this article and then can be used in small and complex measurement space like intelligent car, high-speed train, plane, and weapon. The monitoring system consists of a near-field magnetic probe (9 kHz-20 GHz), a near-field electric probe (9 kHz-43.5 GHz), an electro-optical under-sampling module, and a data acquisition module. First, the ultrawideband signal (≤50 GHz) is measured by the near-field probes. Then, the ultrawideband signal is moved into a narrow band (a few tens of MHz) by the electro-optical module. Finally, the narrow band signal is captured and analyzed by the data acquisition module. The field data measured under special working mode can be used to locate the weak noise source in many engineering applications.

Fatigue crack detection using dual laser induced nonlinear ultrasonic modulation

– An electrostatic in-plane overlap varying energy harvester is modeled and simulated using a circuit simulator. Both linear and nonlinear models are investigated. The nonli-near model includes mechanical stoppers at the displacement extremes. Large amplitude excitation signals, both narrow and wide band, are used to emulate environmental vibrations. Non-linear behavior is significant at large displacement due to the impact on mechanical stoppers. For a sinusoidal excitation the mechanical stoppers cause the output power to flatten and weakly decrease. For a wide band excitation, the output power first increases linearly with the power spectral density of the input signal, then grows slower than linearly. I.