DIT-Artículos

URI permanente para esta colecciónhttps://hdl.handle.net/10953/226

Examinar

Mostrando 1 - 20 de 25

Statistical Normalization for a Guided Clustering Type-2 Fuzzy System for WSN
(IEEE, 2022-03-15) Yuste Delgado, Antonio Jesús; Cuevas Martínez, Juan Carlos; Triviño Cabrera, Alicia
One of the main concerns in Wireless Sensor Networks is the efficient energy management of the nodes. Hierarchical techniques such as clustering have been developed in an effort to solve this problem. In this paper we present a smart evolution of a distributed clustering method that uses a turn-based scheduling cluster head selection process based on an interval Type-2 fuzzy system. The method we propose offers four main improvements. First, the setup process guided by the Base Station is adapted to tune the skip parameter during the network lifetime, which controls how many rounds the clusters are not updated. Second, the normalization of the fuzzy system input variables is carefully performed based on a statistical analysis to reduce the effect of fluctuations in edge values. Third, the value of the coefficient applied to the output of the inner Type-2 fuzzy system is updated to balance the number of cluster heads at early stages. Finally, only the strongest candidate nodes, those with the highest probability, are selected to become cluster heads. The proposed design and scheduling aim to achieve low-energy processing in the nodes. When our proposed techniques are applied, they give better results compared with other similar approaches.
Binaural lateral localization of multiple sources in real environments using a kurtosis-driven split-EM algorithm
(Elsevier, 2018-03) Reche-Lopez, Pedro; Pérez-Lorenzo, José Manuel; Rivas, Fernando; Viciana-Abad, Raquel
In this work a method for an unsupervised lateral localization of simultaneous sound sources is presented. Following a binaural approach, the kurtosis-driven split-EM algorithm (KDS-EM) implemented is able to estimate the direction of arrival of relevant sound sources without knowing a priori their number. Information about the localization is integrated within a period of observation time to serve as an auditory memory in the context of social robotics. Experiments have been conducted using two types of observation times, one shorter with the purpose of analyzing its performance in a reactive level, and other longer that allows the analysis of its contribution as an input of the building process of the sorroundings auditory models that serves to drive a more deliberative behavior. The system has been tested in real and reverberant environments, achieving a good performance based on an over-modeling process that is able to isolate the location of the relevant sources from adverse acoustic effects, such as reverberations.
Jump to the Next Level: A Four-Year Gamification Experiment in Information Technology Engineering
(IEEE - Institute of Electrical and Electronics Engineers, 2019-08-02) Cuevas-Martínez, Juan Carlos; Yuste-Delgado, Antonio Jesús; Pérez-Lorenzo, José Manuel; Triviño-Cabrera, Alicia
Higher education in Spain has to deal with constant troubles and uncertainties due to the economic crisis, high rates of unemployment in young people, lack for study habits in secondary school and legal fluctuations. This uncertain environment does not foster student effort and it is behind the important rates of abandon in higher education. The Bologna Process was thought to create a new paradigm in higher education in the European Union. However, the changes came from the top (governments) to the bottom (lectures and students) so they were not properly supported by specialized training oriented to lecturers. It did not include the appropriate changes in lower education stages (secondary education) to prepare student when facing University. Therefore, in the past decade several new teaching methodologies have appeared to deal with student demotivation and to fight against dropouts. Those methodologies try to keep the students engaged during the whole course paying more attention to their learning process, attitudes, motivations and expectations. Consequently, in this paper, we present a four year experiment whose main objective is to keep students engaged during the whole year and to foster their motivation in order to increase their learning outcomes. The experiment is based on the application of gamification to the assessment process emulating a traditional platform video-game, like Super Mario. The results show that this experiment was positive for most students who achieved good marks and good rates of task completion.
An audio enhancement system to improve intelligibility for social-awareness in HRI
(Springer, 2021-08-28) Martínez-Colón, Antonio; Viciana-Abad, Raquel; Pérez-Lorenzo, José Manuel; Evers, Christine; Naylor, Patrick A.
Improving the ability to interact through voice with a robot is still a challenge especially in real environments where multiple speakers coexist. This work has evaluated a proposal based on improving the intelligibility of the voice information that feeds an existing ASR service in the network and in conditions similar to those that could occur in a care centre for the elderly. The results indicate the feasibility and improvement of a proposal based on the use of an embedded microphone array and the use of a simple beamforming and masking technique. The system has been evaluated with 12 people and results obtained for time responsiveness indicate that the system would allow natural interaction with voice. It is shown to be necessary to incorporate a system to properly employ the masking algorithm, through the intelligent and stable estimation of the interfering signals. In addition, this approach allows to fix as sources of interest other speakers not located in the vicinity of the robot.
The Town Crier: A Use-Case Design and Implementation for a Socially Assistive Robot in Retirement Homes
(MDPI, 2024-04-09) Iglesias, Ana; Viciana, Raquel; Pérez-Lorenzo, José Manuel; Ting, Karine Lan Hing; Tudela, Alberto; Marfil, Rebeca; Qbilat, Malak; Hurtado, Antonio; Jerez, Antonio; Bandera, Juan Pedro
The use of new assistive technologies in general, and Socially Assistive Robots (SARs) in particular, is becoming increasingly common for supporting people’s health and well-being. However, it still faces many issues regarding long-term adherence, acceptability and utility. Most of these issues are due to design processes that insufficiently take into account the needs, preferences and values of intended users. Other issues are related to the currently very limited amount of long-term evaluations, performed in real-world settings, for SARs. This study presents the results of two regional projects that consider as a starting hypothesis that the assessment in controlled environments and/or with short exposures may not be enough in the design of an SAR deployed in a retirement home and the necessity of designing for and with users. Thus, the proposed methodology has focused on use-cases definitions that follow a human centred and participatory design approach. The main goals have been facilitating system acceptance and attachment by involving stakeholders in the robots design and evaluation, overcoming usage barriers and considering user’s needs integration. The implementation of the first use-case deployed and the two phase pilot test performed in a retirement home are presented. In particular, a detailed description of the interface redesign process based on improving a basic prototype with users’ feedback and recommendations is presented, together with the main results of a formal evaluation that has highlighted the impact of changes and improvements addressed in the first redesign loop of the system.
An User-Centered Evaluation of Two Socially Assistive Robots Integrated in a Retirement Home
(Springer, 2024-09-27) Jerez, Antonio; Iglesias, Ana; Pérez-Lorenzo, José Manuel; Tudela, Alberto; Cruces, Alejandro; Bandera, Juan Pedro
Socially assistive robots are receiving a growing interest in the health and social care sectors. They are considered a promising technology to add value to the work of caregivers, and relieve them of simple and repetitive tasks. However, these robots currently face significant difficulties when deployed in everyday scenarios due to a number of factors. Most of these factors are related to insufficient consideration of the user perspective and incorrect evaluation procedures. This paper aims to address these issues. Its objective is to analyze the long-term accessibility, usability, social acceptance and user experience for two different socially assistive robots performing the same tasks in a retirement home. The evaluation procedure is based on a framework specifically designed to consider these criteria. Collected results show that both robots received an overall positive feedback. Although the number of users participating in the evaluation was not very high, due to the chosen recruitment criteria and the period of activity of this research project, during the COVID19 pandemic, these results allow to extract relevant insights towards a meaningful use of social robots in shared social care contexts.
Sustainable expert virtual machine migration in dynamic clouds
(Elsevier, 2022-09) Seddiki, Doraid; García Galán, Sebastián; Muñoz Expósito, J. Enrique; Valverde Ibáñez, Manuel; Marciniak, Tomasz; Pérez de Prado, Rocío J.
Operation on demand flexibility in cloud computing services has resulted in great popularity and wide adoption. These services integrate thousands of computers, storage and communication networks, which implies a high consumption of electrical energy. Therefore, renewable energy-based cloud data centers are replacing traditional energy power grids. In this regard, the workload could be transferred to different nodes among different cloud data centers geographically distributed regarding renewable energy availability. In this sense, this paper presents a framework based on Cloudsim for virtual machine migrations among cloud data centers regarding sustainability optimization. Moreover, an approach for migrations among datacenters based on an expert system has been tested in several scenarios with renewable energy dynamically available. Experimental results show the improvements of the proposed framework and how expert systems can take advantage of renewable energy availability in terms of sustainability while preserving the QoS in terms of execution time.
An incremental algorithm based on multichannel non-negative matrix partial co-factorization for ambient denoising in auscultation
(Elsevier, 2021-11) De La Torre Cruz, Juan; Cañadas Quesada, Francisco Jesús; Martínez Muñoz, Damián; Ruiz Reyes, Nicolás; García Galán, Sebastián; Carabias Orti, Julio José
One of the major current limitations in the diagnosis derived from auscultation remains the ambient noise surrounding the subject, which prevents successful auscultation. Therefore, it is essential to develop robust signal processing algorithms that can extract relevant clinical information from auscultated recordings analyzing in depth the acoustic environment in order to help the decision-making process made by physicians. The aim of this study is to implement a method to remove ambient noise in biomedical sounds captured in auscultation. We propose an incremental approach based on multichannel non-negative matrix partial co-factorization (NMPCF) for ambient denoising focusing on high noisy environment with a Signal-to-Noise Ratio (SNR) 6 5 dB. The first contribution applies NMPCF assuming that ambient noise can be modelled as repetitive sound events simultaneously found in two single-channel inputs captured by means of different recording devices. The second contribution proposes an incremental algorithm, based on the previous multichannel NMPCF, that refines the estimated biomedical spectrogram throughout a set of incremental stages by eliminating most of the ambient noise that was not removed in the previous stage at the expense of preserving most of the biomedical spectral content. The ambient denoising performance of the proposed method, compared to some of the most relevant state-of-the-art methods, has been evaluated using a set of recordings composed of biomedical sounds mixed with ambient noise that typically surrounds a medical consultation room to simulate high noisy environments with a SNR from -20 dB to -5 dB. In order to analyse the drop in denoising performance of the evaluated methods when the effect of the propagation of the patient’s body material and the acoustics of the room is considered, results have been obtained with and without taking these effects into account. Experimental results report that: (i) the performance drop suffered by the proposed method is lower compared to MSS and NLMS when considering the effect of the propagation of the patient’s body material and the acoustics of the room active; (ii) unlike what happens with MSS and NLMS, the proposed method shows a stable trend of the average SDR and SIR results regardless of the type of ambient noise and the SNR level evaluated; and (iii) a remarkable advantage of the proposed method is the high robustness of the acoustic quality of the estimated biomedical sounds when the two single-channel inputs suffer from a delay between them.
Unsupervised detection and classification of heartbeats using the dissimilarity matrix in PCG signals
(Elsevier, 2022-06) De La Torre Cruz, Juan; Martínez Muñoz, Damián; Ruiz Reyes, Nicolás; Muñoz Montoro, Antonio Jesús; Puentes Chiachio, Miguel; Canadas Quesada, Francisco Jesús
Background and objective: Auscultation is the first technique applied to the early diagnose of any cardiovascular disease (CVD) in rural areas and poor-resources countries because of its low cost and noninvasiveness. However, it highly depends on the physician’s expertise to recognize specific heart sounds heard through the stethoscope. The analysis of phonocardiogram (PCG) signals attempts to segment each cardiac cycle into the four cardiac states (S1, systole, S2 and diastole) in order to develop automatic systems applied to an efficient and reliable detection and classification of heartbeats. In this work, we propose an unsupervised approach, based on time-frequency characteristics shown by cardiac sounds, to detect and classify heartbeats S1 and S2. Methods: The proposed system consists of a two-stage cascade. The first stage performs a rough heartbeat detection while the second stage refines the previous one, improving the temporal localization and also classifying the heartbeats into types S1 and S2. The first contribution is a novel approach that combines the dissimilarity matrix with the frame-level spectral divergence to locate heartbeats using the repetitiveness shown by the heart sounds and the temporal relationships between the intervals defined by the events S1/S2 and non-S1/S2 (systole and diastole). The second contribution is a verification-correctionclassification process based on a sliding window that allows the preservation of the temporal structure of the cardiac cycle in order to be applied in the heart sound classification. The proposed method has been assessed using the open access databases PASCAL, CirCor DigiScope Phonocardiogram and an additional sound mixing procedure considering both Additive White Gaussian Noise (AWGN) and different kinds of clinical ambient noises from a commercial database. Results: The proposed method outperforms the detection and classification performance of other recent state-of-the-art methods. Although our proposal achieves the best average accuracy for PCG signals without cardiac abnormalities, 99.4% in heartbeat detection and 97.2% in heartbeat classification, its worst average accuracy is always above 92% for PCG signals with cardiac abnormalities, signifying an improvement in heartbeat detection/classification above 10% compared to the other state-of-the-art methods evaluated. Conclusions: The proposed method provides the best detection/classification performance in realistic scenarios where the presence of cardiac anomalies as well as different types of clinical environmental noises are active in the PCG signal. Of note, the promising modelling of the temporal structures of the heart provided by the dissimilarity matrix together with the frame-level spectral divergence, as well as the removal of a significant number of spurious heart events and recovery of missing heart events, both corrected by the proposed verification-correction-classification algorithm, suggest that our proposal is a successful tool to be applied in heart segmentation.
An ambient denoising method based on multi‑channel non‑negative matrix factorization for wheezing detection
(Springer Netherlands, 2022-07-29) Muñoz Montoro, Antonio Jesús; Revuelta Sanz, Pablo; Martínez Muñoz, Damián; De La Torre Cruz, Juan; Ranilla, José
In this paper, a parallel computing method is proposed to perform the background denoising and wheezing detection from a multi-channel recording captured during the auscultation process. The proposed system is based on a non-negative matrix factorization (NMF) approach and a detection strategy. Moreover, the initialization of the proposed model is based on singular value decomposition to avoid dependence on the initial values of the NMF parameters. Additionally, novel update rules to simultaneously address the multichannel denoising while preserving an orthogonal constraint to maximize source separation have been designed. The proposed system has been evaluated for the task of wheezing detection showing a significant improvement over state-of-the-art algorithms when noisy sound sources are present. Moreover, parallel and high-performance techniques have been used to speedup the execution of the proposed system, showing that it is possible to achieve fast execution times, which enables its implementation in real-world scenarios.
Classification of Adventitious Sounds Combining Cochleogram and Vision Transformers
(Multidisciplinary Digital Publishing Institute (MDPI), 2024-01-21) Daria Mang, Loredana; González Martínez, Francisco David; Martinez Muñoz, Damián; García Galán, Sebastián; Cortina, Raquel
Early identification of respiratory irregularities is critical for improving lung health and reducing global mortality rates. The analysis of respiratory sounds plays a significant role in characterizing the respiratory system’s condition and identifying abnormalities. The main contribution of this study is to investigate the performance when the input data, represented by cochleogram, is used to feed the Vision Transformer (ViT) architecture, since this input–classifier combination is the first time it has been applied to adventitious sound classification to our knowledge. Although ViT has shown promising results in audio classification tasks by applying self-attention to spectrogram patches, we extend this approach by applying the cochleogram, which captures specific spectro-temporal features of adventitious sounds. The proposed methodology is evaluated on the ICBHI dataset. We compare the classification performance of ViT with other state-of-the-art CNN approaches using spectrogram, Mel frequency cepstral coefficients, constant-Q transform, and cochleogram as input data. Our results confirm the superior classification performance combining cochleogram and ViT, highlighting the potential of ViT for reliable respiratory sound classification. This study contributes to the ongoing efforts in developing automatic intelligent techniques with the aim to significantly augment the speed and effectiveness of respiratory disease detection, thereby addressing a critical need in the medical field.
Optimizing Rule Weights to Improve FRBS Clustering in Wireless Sensor Networks
(MDPI, Basel, Switzerland., 2024-08-27) Muñoz-Exposito, Jose-Enrique; Yuste-Delgado, Antonio-Jesús; Triviño-Cabrera, Alicia; Cuevas-Martínez, Juan-Carlos
Wireless sensor networks (WSNs) are usually composed of tens or hundreds of nodes powered by batteries that need efficient resource management to achieve the WSN’s goals. One of the techniques used to manage WSN resources is clustering, where nodes are grouped into clusters around a cluster head (CH), which must be chosen carefully. In this article, a new centralized clustering algorithm is presented based on a Type-1 fuzzy logic controller that infers the probability of each node becoming a CH. The main novelty presented is that the fuzzy logic controller employs three different knowledge bases (KBs) during the lifetime of the WSN. The first KB is used from the beginning to the instant when the first node depletes its battery, the second KB is then applied from that moment to the instant when half of the nodes are dead, and the last KB is loaded from that point until the last node runs out of power. These three KBs are obtained from the original KB designed by the authors after an optimization process. It is based on a particle swarm optimization algorithm that maximizes the lifetime of the WSN in the three periods by adjusting each rule in the KBs through the assignment of a weight value ranging from 0 to 1. This optimization process is used to obtain better results in complex systems where the number of variables or rules could make them unaffordable. The results of the presented optimized approach significantly improved upon those from other authors with similar methods. Finally, the paper presents an analysis of why some rule weights change more than others, in order to design more suitable controllers in the future.
The music demixing machine: toward real-time remixing of classical music
(Springer, 2023-04-06) Cabañas-Molero, Pablo Antonio; Muñoz-Montoro, Antonio Jesús; Vera-Candeas, Pedro; Ranilla, José
Classical music, unlike popular music, is usually recorded live with close microphone techniques. For this reason, isolated tracks are not available to create the final mixture/stream, and so the mixing process requires greater effort. Source separation methods are a potential solution to this problem. However, current algorithms are not fast enough to yield real-time separation in professional setups with dozens of microphones and sources. In this paper, we propose a fast approach consisting of a panning-based multichannel non-negative matrix factorization model to separate classical music. We tested the system on real professional recordings, where we were able to reach real-time with very low latency and promising quality.
A constrained tonal semi-supervised non-negative matrix factorization to classify presence/absence of wheezing in respiratory sounds
(Elsevier, 2020-04-01) Torre-Cruz, Juan; Cañadas-Quesada, Francisco Jesús; García-Galán, Sebastián; Ruiz-Reyes, Nicolás; Vera-Candeas, Pedro; Carabias-Orti, Julio José
From a clinical point of view, the detection of wheezing presence in respiratory sounds is a challenging task for early identification of pulmonary diseases since wheezing is the main manifestation associated to airway obstruction. In this article, we propose a novel method to detect the presence or absence of wheeze sounds in breath recordings in order to increase the reliability of the subjective diagnosis provided by the physician in the auscultation process. Specifically, it is assumed an unhealthy subject when wheeze sounds can be detected during breathing. The proposed method consists of three stages. The first stage attempts to estimate the spectral interval, band of interest (BOI), that shows the highest probability to find wheeze sounds. In the second stage, a constrained tonal semi-supervised non-negative matrix factorization (NMF) approach is applied to obtain spectral patterns that models the periodic or tonal nature typically shown by wheeze sounds. The third stage analyzes the estimated wheezing spectrogram based on the smoothness of the spectral trajectories from the most significant energy previously factorized in the BOI. Our system has been evaluated and compared to other state-of-the-art methods, yielding competitive results in the wheezing presence detection in respiratory sounds.
Monitoring the internal quality of ornamental stone using impact-echo testing
(Elsevier, 2019-12-01) Montiel-Zafra, María Violeta; Cañadas-Quesada, Francisco Jesús; Campos-Suñol, María José; Vera-Candeas, Pedro; Ruiz-Reyes, Nicolás
The decay and durability of stone materials is a natural response to the progressive adjustment to different environmental and harsh conditions. Usually stone building elements with no apparent sign of decay are affected by the loss of cohesion. Non-invasive, early and low-cost identification of the internal damage of stone materials would be a great step forward. This paper presents an impact-echo (IE) method to analyse the internal quality of ornamental stone. The proposed method attempts to estimate the P-wave velocity in the material applying a frequency estimator that best explains the energy distribution of the possible modes of vibration from the captured IE signals. The velocity estimation will be analysed along a set of freeze-thawing cycles in order to establish a correlation with the internal damage caused in the material confirmed by its porosity. This value has been measured after several freezing-thawing cycles at each stone specimen. Experimental results show that the proposed method can be considered as a valid and effective tool for determining the internal damage of ornamental stone materials. Besides, the proposed method could be easily adapted to analyse specimens of different sizes, shapes and types of rocks.
A novel wheezing detection approach based on constrained non-negative matrix factorization
(Elsevier, 2019-05-01) Torre-Cruz, Juan; Cañadas-Quesada, Francisco Jesús; Carabias-Orti, Julio José; Vera-Candeas, Pedro; Ruiz-Reyes, Nicolás
The early wheezing detection is still a challenging task in biomedical signal processing because the presence of wheeze sounds often indicate respiratory diseases from airway obstructions. Currently, most of the first clinical examinations to detect any airway obstructions are carried out using auscultation. However, a high percentage of diagnoses are misdiagnosed since they are highly dependent on the physician’s training in the wheezing detection, especially in noisy environments in which weak wheeze sounds can be masked by louder respiratory sounds. In this work, we propose a novel wheezing detection approach, based on Constrained Non-negative Matrix Factorization, that uses two-stage cascade: separation and detection. The novelty of the separation stage is to model wheeze and respiratory sounds as reliably as possible that they can be observed in the nature incorporating constraints (sparseness and smoothness) into the NMF factorization. Once the estimated wheezing and respiratory signal are obtained from the separation stage, the detection contribution is based on the use of the Kullback-Leibler divergence to discriminate between wheezing and respiratory areas. The experiments have been conducted using three different datasets composed of healthy or unhealthy patients. First, an optimization process is applied to obtain the optimal parameters of the separation stage. Finally, the performance of the wheezing detection of the proposed method is evaluated taking into account other state-of-the-art methods. Experimental results report that i) the proposed method outperforms recent state-of-the-art wheezing detection approaches showing a robust wheezing detection performance even evaluating noisy environments and ii) the ability of the proposal to reliably detect healthy patients.
Multimodal speaker diarization for meetings using volume-evaluated SRP-PHAT and video analysis
(Springer, 2018-04-11) Cabañas-Molero, Pablo Antonio; Lucena, Manuel; Fuertes, José Manuel; Vera-Candeas, Pedro; Ruiz-Reyes, Nicolás
Speaker diarization is traditionally defined as the problem of determining “who speaks when” given an audio or video stream. This is an important task in many applications for meeting rooms, including automatic transcription of conversations, camera steering or content summarization. When the room is equipped with microphone arrays and cameras, speakers can be distinguished according to their location and the problem can be addressed through localization techniques. This article proposes a multimodal speaker diarization system for meeting environments based on a modified SRP-PHAT function evaluated on space volumes rather than discrete points. In our system, this function is used in combination with a circular array, enabling audio-based localization based on the selection of local maxima. Voicing detection is used to detect speech frames, whereas video analysis is introduced to aid in the decision when users move or simultaneously speak. The approach is evaluated on the well-known AMI dataset with approximately 100 hours of realistic meeting recordings and shows an average diarization error rate of 21% – 25%.
Combining a recursive approach via non-negative matrix factorization and Gini index sparsity to improve reliable detection of wheezing sounds
(Elsevier, 2020-06-01) De La Torre Cruz, Juan; Cañadas Quesada, Francisco Jesús; Carabias Orti, Julio José; Vera Candeas, Pedro; Ruiz Reyes, Nicolás
Auscultation constitutes a fast, non-invasive and low-cost tool widely used to diagnose respiratory diseases in most of the health centres. However, the acoustic training and expertise acquired by the physician is still crucial to provide a reliable diagnosis of the status of the lung. Each wrong diagnosis increases the risk to the health of patients and the costs associated with the treatment of the disease detected. A wheezing detection system can be useful to the physician to minimize the subjectivity of the interpretation of the breathing sounds, misdiagnoses due to stress and elucidating complex acoustic scenes (such as louder background noises). Highlight that the presence of wheeze sounds is one of the main indicators of respiratory disorders from airway obstructions. This work presents an expert and intelligent system to detect wheeze sounds based on a recursive algorithm that combines orthogonal non-negative matrix factorization (ONMF) and the sparsity descriptor Gini index. The recursive algorithm is composed of four stages. The first stage is based on ONMF modelling to factorize the spectral bases as dissimilar as possible. The second stage clusters the ONMF bases into two categories: wheezing and normal breath. The third stage proposes a novel stopping criterion that controls the loss of wheezing spectral content at the expense of removing normal breath content in the recursive algorithm. Finally, the fourth stage determines the patient’s condition to locate the temporal intervals in which wheeze sounds are active for unhealthy patients. Experimental results report that the proposed method: (i) provides the best detection performance compared to the recent state-of-the-art wheezing detection approaches, achieving the highest robustness in noisy environments; and (ii) reliably distinguishes the patient’s condition (healthy/unhealthy). The strengths of the proposed method are the following: (i) its unsupervised nature since it does not depend on any training stage to learn in advanced the sounds of interest (wheezing). This fact could make this method attractive to be used in clinical settings because wheezing sound databases are often unavailable; and (ii) the modelling of the spectral behaviour by means of a common feature, the sparsity, that represents the typically energy distributions shown by most of the wheeze and normal breath sounds.
Multichannel Blind Sound Source Separation Using Spatial Covariance Model With Level and Time Differences and Nonnegative Matrix Factorization
(IEEE, 2018-04-27) Carabias-Orti, Julio José; Nikunen, Joonas; Virtanen, Tuomas; Vera-Candeas, Pedro
This paper presents an algorithm for multichannel sound source separation using explicit modeling of level and time differences in source spatial covariance matrices (SCM). We propose a novel SCM model in which the spatial properties are modeled by the weighted sum of direction of arrival (DOA) kernels. DOA kernels are obtained as the combination of phase and level difference covariance matrices representing both time and level differences between microphones for a grid of predefined source directions. The proposed SCM model is combined with the NMF model for the magnitude spectrograms. Opposite to other SCM models in the literature, in this work, source localization is implicitly defined in the model and estimated during the signal factorization. Therefore, no localization preprocessing is required. Parameters are estimated using complex-valued nonnegative matrix factorization with both Euclidean distance and Itakura-Saito divergence. Separation performance of the proposed system is evaluated using the two-channel SiSEC development dataset and four channels signals recorded in a regular room with moderate reverberation. Finally, a comparison to other state-of-the-art methods is performed, showing better achieved separation performance in terms of SIR and perceptual measures.
A Distributed Clustering Algorithm Guided by the Base Station to Extend the Lifetime ofWireless Sensor Networks
(MDPI, 2020-04-18) Yuste-Delgado, Antonio-Jesús; Cuevas-Martínez, Juan-Carlos; Triviño-Cabrera, Alicia
Clustering algorithms are necessary in Wireless Sensor Networks to reduce the energy consumption of the overall nodes. The decision of which nodes are the cluster heads (CHs) greatly affects the network performance. The centralized clustering algorithms rely on a sink or Base Station (BS) to select the CHs. To do so, the BS requires extensive data from the nodes, which sometimes need complex hardware inside each node or a significant number of control messages. Alternatively, the nodes in distributed clustering algorithms decide about which the CHs are by exchanging information among themselves. Both centralized and distributed clustering algorithms usually alternate the nodes playing the role of the CHs to dynamically balance the energy consumption among all the nodes in the network. This paper presents a distributed approach to form the clusters dynamically, but it is occasionally supported by the Base Station. In particular, the Base Station sends three messages during the network lifetime to reconfigure the skip value of the network. The skip, which stands out as the number of rounds in which the same CHs are kept, is adapted to the network status in this way. At the beginning of each group of rounds, the nodes decide about their convenience to become a CH according to a fuzzy-logic system. As a novelty, the fuzzy controller is as a Tagaki–Sugeno–Kang model and not a Mandami-one as other previous proposals. The clustering algorithm has been tested in a wide set of scenarios, and it has been compared with other representative centralized and distributed fuzzy-logic based algorithms. The simulation results demonstrate that the proposed clustering method is able to extend the network operability.

RUJA: Repositorio Institucional de Producción Científica

Examinar

Envíos recientes