Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553457
Yingke Zhao, J. Nielsen, M. G. Christensen, Jinzdona Chen
One of the major challenges in wireless acoustic sensor networks (WASN) based speech enhancement is robust and accurate voice activity detection (VAD). VAD is widely used in speech enhancement, speech coding, speech recognition, etc. In speech enhancement applications, VAD plays an important role, since noise statistics can be updated during non-speech frames to ensure efficient noise reduction and tolerable speech distortion. Although significant efforts have been made in single channel VAD, few solutions can be found in the multichannel case, especially in WASN. In this paper, we introduce a distributed VAD by using model-based noise power spectral density (PSD) estimation. For each node in the network, the speech PSD and noise PSD are first estimated, then a distributed detection is made by applying the generalized likelihood ratio test (GLRT). The proposed global GLRT based VAD has a quite general form. Indeed, we can judge whether the speech is present or absent by using the current time frame and frequency band observation or by taking into account the neighbouring frames and bands. Finally, the distributed GLRT result is obtained by using a distributed consensus method, such as random gossip, i.e., the whole detection system does not need any fusion center. With the model-based noise estimation method, the proposed distributed VAD performs robustly under non-stationary noise conditions, such as babble noise. As shown in experiments, the proposed method outperforms traditional multichannel VAD methods in terms of detection accuracy.
{"title":"Model-Based Voice Activity Detection in Wireless Acoustic Sensor Networks","authors":"Yingke Zhao, J. Nielsen, M. G. Christensen, Jinzdona Chen","doi":"10.23919/EUSIPCO.2018.8553457","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553457","url":null,"abstract":"One of the major challenges in wireless acoustic sensor networks (WASN) based speech enhancement is robust and accurate voice activity detection (VAD). VAD is widely used in speech enhancement, speech coding, speech recognition, etc. In speech enhancement applications, VAD plays an important role, since noise statistics can be updated during non-speech frames to ensure efficient noise reduction and tolerable speech distortion. Although significant efforts have been made in single channel VAD, few solutions can be found in the multichannel case, especially in WASN. In this paper, we introduce a distributed VAD by using model-based noise power spectral density (PSD) estimation. For each node in the network, the speech PSD and noise PSD are first estimated, then a distributed detection is made by applying the generalized likelihood ratio test (GLRT). The proposed global GLRT based VAD has a quite general form. Indeed, we can judge whether the speech is present or absent by using the current time frame and frequency band observation or by taking into account the neighbouring frames and bands. Finally, the distributed GLRT result is obtained by using a distributed consensus method, such as random gossip, i.e., the whole detection system does not need any fusion center. With the model-based noise estimation method, the proposed distributed VAD performs robustly under non-stationary noise conditions, such as babble noise. As shown in experiments, the proposed method outperforms traditional multichannel VAD methods in terms of detection accuracy.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129467371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553524
Dong Kyoo Kim, You Jin Kim
De-offset estimation of quadrature continuous-wave (CW) radar has been studied for years. Studies have shown that the estimation error increases when target movement with respect to the radar is small. This paper presents a method that uses multiple simultaneous CW frequencies for the de-offset estimation, which makes the de-offset estimation easy in contrast to the conventional quadrature CW radar. A de-offset estimation method using the multiple CW frequencies is presented to demonstrate that the multiple CW frequencies provide sufficient information for the de-offset estimation.
{"title":"DC-offset Estimation of Multiple CW Micro Doppler Radar","authors":"Dong Kyoo Kim, You Jin Kim","doi":"10.23919/EUSIPCO.2018.8553524","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553524","url":null,"abstract":"De-offset estimation of quadrature continuous-wave (CW) radar has been studied for years. Studies have shown that the estimation error increases when target movement with respect to the radar is small. This paper presents a method that uses multiple simultaneous CW frequencies for the de-offset estimation, which makes the de-offset estimation easy in contrast to the conventional quadrature CW radar. A de-offset estimation method using the multiple CW frequencies is presented to demonstrate that the multiple CW frequencies provide sufficient information for the de-offset estimation.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"24 19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128458070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553478
Mhd Modar Halimeh, Andreas Brendel, Walter Kellermann
A resampling scheme is proposed for use with Sequential Monte Carlo (SMC)-based Probability Hypothesis Density (PHD) filters. It consists of two steps, first, regions of interest are identified, then an evolutionary resampling is applied for each region. Applying resampling locally corresponds to treating each target individually, while the evolutionary resampling introduces a memory to a group of particles, increasing the robustness of the estimation against noise outliers. The proposed approach is compared to the original SMC-PHD filter for tracking multiple targets in a deterministically moving targets scenario, and a noisy motion scenario. In both cases, the proposed approach provides more accurate estimates.
{"title":"Evolutionary Resampling for Multi-Target Tracking using Probability Hypothesis Density Filter","authors":"Mhd Modar Halimeh, Andreas Brendel, Walter Kellermann","doi":"10.23919/EUSIPCO.2018.8553478","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553478","url":null,"abstract":"A resampling scheme is proposed for use with Sequential Monte Carlo (SMC)-based Probability Hypothesis Density (PHD) filters. It consists of two steps, first, regions of interest are identified, then an evolutionary resampling is applied for each region. Applying resampling locally corresponds to treating each target individually, while the evolutionary resampling introduces a memory to a group of particles, increasing the robustness of the estimation against noise outliers. The proposed approach is compared to the original SMC-PHD filter for tracking multiple targets in a deterministically moving targets scenario, and a noisy motion scenario. In both cases, the proposed approach provides more accurate estimates.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128615349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553016
Roxana Alexandru, P. Malhotra, Stephanie Reynolds, P. Dragotti
We address the problem of estimating the effective connectivity of the brain network, using the input stimulus model proposed by Izhikevich in [1], which accurately reproduces the behaviour of spiking and bursting biological neurons, whilst ensuring computational simplicity. We first analyse the temporal dynamics of neural networks, showing that the spike propagation within the brain can be modelled as a diffusion process. This helps prove the suitability of NetRate algorithm proposed by Rodriguez in [2] to infer the structure of biological neural networks. Finally, we present simulation results using synthetic data to verify the performance of the topology estimation algorithm.
{"title":"Estimating the Topology of Neural Networks from Distributed Observations","authors":"Roxana Alexandru, P. Malhotra, Stephanie Reynolds, P. Dragotti","doi":"10.23919/EUSIPCO.2018.8553016","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553016","url":null,"abstract":"We address the problem of estimating the effective connectivity of the brain network, using the input stimulus model proposed by Izhikevich in [1], which accurately reproduces the behaviour of spiking and bursting biological neurons, whilst ensuring computational simplicity. We first analyse the temporal dynamics of neural networks, showing that the spike propagation within the brain can be modelled as a diffusion process. This helps prove the suitability of NetRate algorithm proposed by Rodriguez in [2] to infer the structure of biological neural networks. Finally, we present simulation results using synthetic data to verify the performance of the topology estimation algorithm.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"152 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127314346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553118
G. Wang, Liang Zhang, G. Giannakis, Jie Chen
Sparse phase retrieval (PR) aims at reconstructing a sparse signal vector from a few phaseless linear measurements. It emerges naturally in diverse applications, but it is NP-hard in general. Drawing from advances in nonconvex optimization, this paper presents a new algorithm that is termed compressive reweighted amplitude flow (CRAF) for sparse PR. CRAF operates in two stages: Stage one computes an initial guess by means of a new spectral procedure, and stage two implements a few hard thresholding based iteratively reweighted gradient iterations on the amplitude-based least-squares cost. When there are sufficient measurements, CRAF reconstructs the true signal vector exactly under suitable conditions. Furthermore, its sample complexity coincides with that of the state-of-the-art approaches. Numerical experiments showcase improved performance of the proposed approach relative to existing alternatives.
{"title":"Sparse Phase Retrieval Via Iteratively Reweighted Amplitude Flow","authors":"G. Wang, Liang Zhang, G. Giannakis, Jie Chen","doi":"10.23919/EUSIPCO.2018.8553118","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553118","url":null,"abstract":"Sparse phase retrieval (PR) aims at reconstructing a sparse signal vector from a few phaseless linear measurements. It emerges naturally in diverse applications, but it is NP-hard in general. Drawing from advances in nonconvex optimization, this paper presents a new algorithm that is termed compressive reweighted amplitude flow (CRAF) for sparse PR. CRAF operates in two stages: Stage one computes an initial guess by means of a new spectral procedure, and stage two implements a few hard thresholding based iteratively reweighted gradient iterations on the amplitude-based least-squares cost. When there are sufficient measurements, CRAF reconstructs the true signal vector exactly under suitable conditions. Furthermore, its sample complexity coincides with that of the state-of-the-art approaches. Numerical experiments showcase improved performance of the proposed approach relative to existing alternatives.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129096878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553326
Mingyang Cao, X. Mao, Xiaozhuan Long, Lei Huang
In this paper, elevation and azimuth estimation with uniform rectangular array (URA) is addressed. Since the temporal samples received by the URA could be written into a tensorial form, we introduce the multilinear projection for developing a direction-of-arrival (DOA) estimator. In the noiseless condition, the multilinear projector is orthogonal to the steering matrix of the URA. Thus the proposed DOA estimator is designed to find minimal points of the inner product of the steering vector and the multilinear projector. Based on the multilinear algebraic framework, the proposed approach provides a better subspace estimate than that of the matrix-based subspace. Simulation results are provided to demonstrate the effectiveness of the proposed method.
{"title":"Direction-of-Arrival Estimation for Uniform Rectangular Array: A Multilinear Projection Approach","authors":"Mingyang Cao, X. Mao, Xiaozhuan Long, Lei Huang","doi":"10.23919/EUSIPCO.2018.8553326","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553326","url":null,"abstract":"In this paper, elevation and azimuth estimation with uniform rectangular array (URA) is addressed. Since the temporal samples received by the URA could be written into a tensorial form, we introduce the multilinear projection for developing a direction-of-arrival (DOA) estimator. In the noiseless condition, the multilinear projector is orthogonal to the steering matrix of the URA. Thus the proposed DOA estimator is designed to find minimal points of the inner product of the steering vector and the multilinear projector. Based on the multilinear algebraic framework, the proposed approach provides a better subspace estimate than that of the matrix-based subspace. Simulation results are provided to demonstrate the effectiveness of the proposed method.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132182668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553239
M. Dhanaraj, Panos P. Markopoulos
L1-norm Principal-Component Analysis (L1-PCA) has been shown to exhibit sturdy resistance against outliers among the processed data. In this work, we propose L1-IPCA: an algorithm for incremental L1-PCA, appropriate for big-data and streaming-data applications. The proposed algorithm updates the calculated L1-norm principal components as new data points arrive, conducting a sequence of computationally efficient bit-flipping iterations. Our experimental studies on subspace estimation, image conditioning, and video foreground extraction illustrate that the proposed algorithm attains remarkable outlier resistance at low computational cost.
{"title":"Novel Algorithm for Incremental L1-Norm Principal-Component Analysis","authors":"M. Dhanaraj, Panos P. Markopoulos","doi":"10.23919/EUSIPCO.2018.8553239","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553239","url":null,"abstract":"L1-norm Principal-Component Analysis (L1-PCA) has been shown to exhibit sturdy resistance against outliers among the processed data. In this work, we propose L1-IPCA: an algorithm for incremental L1-PCA, appropriate for big-data and streaming-data applications. The proposed algorithm updates the calculated L1-norm principal components as new data points arrive, conducting a sequence of computationally efficient bit-flipping iterations. Our experimental studies on subspace estimation, image conditioning, and video foreground extraction illustrate that the proposed algorithm attains remarkable outlier resistance at low computational cost.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130895038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553083
Jacopo Lovatello, A. Bernardini, A. Sarti
An efficient continuous beam steering method, applicable to differential microphones of any order, has been recently developed. Given two identical reference beams, pointing in two different directions, the method allows to derive a beam of nearly constant shape continuously steerable between those two directions. In this paper, the steering method is applied to robust Differential Microphone Arrays (DMAs) characterized by uniform circular array geometries. In particular, a generalized filter performing the steering operation is defined. The definition of such a filter enables the derivation of closed-form formulas for computing the white noise gain and the directivity factor of the designed steerable differential beamformers for any frequency of interest. A study on the shape invariance of the steered beams is conducted. Applications of the steering approach to first-, second-and third-order robust circular DMAs are presented.
{"title":"Steerable Circular Differential Microphone Arrays","authors":"Jacopo Lovatello, A. Bernardini, A. Sarti","doi":"10.23919/EUSIPCO.2018.8553083","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553083","url":null,"abstract":"An efficient continuous beam steering method, applicable to differential microphones of any order, has been recently developed. Given two identical reference beams, pointing in two different directions, the method allows to derive a beam of nearly constant shape continuously steerable between those two directions. In this paper, the steering method is applied to robust Differential Microphone Arrays (DMAs) characterized by uniform circular array geometries. In particular, a generalized filter performing the steering operation is defined. The definition of such a filter enables the derivation of closed-form formulas for computing the white noise gain and the directivity factor of the designed steerable differential beamformers for any frequency of interest. A study on the shape invariance of the steered beams is conducted. Applications of the steering approach to first-, second-and third-order robust circular DMAs are presented.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125355678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553261
A. Janani, Tyler S. Grummett, Hanieh Bakhshayesh, T. Lewis, J. Willoughby, K. Pope
Scalp electrical recordings, or electroencephalograms (EEG), are heavily contaminated by cranial and cervical muscle activity from as low as 20 hertz, even in relaxed conditions. It is therefore necessary to reduce or remove this contamination to enable reliable exploration of brain neurophysiological responses. Scalp measurements record activity from many sources, including neural and muscular. Independent Component Analysis (ICA) produces components ideally corresponding to separate sources, but the number of components is limited by the number of EEG channels. In practice, at most 30% of components are cleanly separate sources. Increasing the number of channels results in more separate components, but with a significant increase in costs of data collection and computation. Here we present results to assist in selecting an appropriate number of channels. Our unique database of pharmacologically paralysed subjects provides a way to objectively compare different approaches to achieving an ideal, muscle free EEG recording. We evaluated an automatic muscle-removing approach, based on ICA, with different numbers of EEG channels: 21, 32, 64, and 115. Our results show that, for a fixed length of data, 21 channels is insufficient to reduce tonic muscle artefact, and that increasing the number of channels to 115 does result in better tonic muscle artefact reduction.
{"title":"How Many Channels are Enough? Evaluation of Tonic Cranial Muscle Artefact Reduction Using ICA with Different Numbers of EEG Channels","authors":"A. Janani, Tyler S. Grummett, Hanieh Bakhshayesh, T. Lewis, J. Willoughby, K. Pope","doi":"10.23919/EUSIPCO.2018.8553261","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553261","url":null,"abstract":"Scalp electrical recordings, or electroencephalograms (EEG), are heavily contaminated by cranial and cervical muscle activity from as low as 20 hertz, even in relaxed conditions. It is therefore necessary to reduce or remove this contamination to enable reliable exploration of brain neurophysiological responses. Scalp measurements record activity from many sources, including neural and muscular. Independent Component Analysis (ICA) produces components ideally corresponding to separate sources, but the number of components is limited by the number of EEG channels. In practice, at most 30% of components are cleanly separate sources. Increasing the number of channels results in more separate components, but with a significant increase in costs of data collection and computation. Here we present results to assist in selecting an appropriate number of channels. Our unique database of pharmacologically paralysed subjects provides a way to objectively compare different approaches to achieving an ideal, muscle free EEG recording. We evaluated an automatic muscle-removing approach, based on ICA, with different numbers of EEG channels: 21, 32, 64, and 115. Our results show that, for a fixed length of data, 21 channels is insufficient to reduce tonic muscle artefact, and that increasing the number of channels to 115 does result in better tonic muscle artefact reduction.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126689170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-09-01DOI: 10.23919/EUSIPCO.2018.8553231
Pablo Peso Parada, R. Saeidi
A novel approach for proximity detection on mobile handsets which does not require any additional transducers is presented. The method is based on transmitting a chirp and processing the received signal by applying Least Mean Square (LMS), where the desired signal is the transmitted chirp. The envelope of three signals (estimated filter taps, estimated output and error signal) are characterized with a set of 12 features which are used to classify a given frame into one of two classes: proximity active or proximity inactive. The classifier employed is based on Support Vector Machine (SVM) with linear kernel. The results show that over 13 minutes of recorded data, the accuracy achieved is 95.28% using 10-fold cross-validation. Furthermore, the feature importance analysis performed on the database indicates that the most relevant feature is based on the estimated filter taps.
{"title":"Ultrasonic Based Proximity Detection for Handsets","authors":"Pablo Peso Parada, R. Saeidi","doi":"10.23919/EUSIPCO.2018.8553231","DOIUrl":"https://doi.org/10.23919/EUSIPCO.2018.8553231","url":null,"abstract":"A novel approach for proximity detection on mobile handsets which does not require any additional transducers is presented. The method is based on transmitting a chirp and processing the received signal by applying Least Mean Square (LMS), where the desired signal is the transmitted chirp. The envelope of three signals (estimated filter taps, estimated output and error signal) are characterized with a set of 12 features which are used to classify a given frame into one of two classes: proximity active or proximity inactive. The classifier employed is based on Support Vector Machine (SVM) with linear kernel. The results show that over 13 minutes of recorded data, the accuracy achieved is 95.28% using 10-fold cross-validation. Furthermore, the feature importance analysis performed on the database indicates that the most relevant feature is based on the estimated filter taps.","PeriodicalId":303069,"journal":{"name":"2018 26th European Signal Processing Conference (EUSIPCO)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126744049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：