Pub Date : 2017-05-22DOI: 10.1109/ATSIP.2017.8075585
Iman Zubeiri, Fatma Mrabti
The 5G network is a Heterogeneous network, which combines different types of infrastructures and different standards such as WLAN, WIMAX, 3G, 4G, operating in different frequencies, and having different architecture. The implementation of Heterogeneous architectures trend to enhance the coverage and the capacity of the network, and boost spectral efficiency. The integration of these architectures poses several challenges, the major one is Vertical handoff. It consists of transferring an ongoing session from a standard to another. Also this implementation requires a flexible architecture to incorporate multiple Radio Access Technologies (RATs) and facilitate the interworking between them. This work presents the 5G network architecture and some enabling key techniques it presents also a review of interworking methods, and introduces a new interworking technique among Long Term Evolution (LTE) and WiMax. Simulation results show the capacity of our methods to offload flows from different RATs.
{"title":"A comparative study of interworking methods among differents rats in 5G context","authors":"Iman Zubeiri, Fatma Mrabti","doi":"10.1109/ATSIP.2017.8075585","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075585","url":null,"abstract":"The 5G network is a Heterogeneous network, which combines different types of infrastructures and different standards such as WLAN, WIMAX, 3G, 4G, operating in different frequencies, and having different architecture. The implementation of Heterogeneous architectures trend to enhance the coverage and the capacity of the network, and boost spectral efficiency. The integration of these architectures poses several challenges, the major one is Vertical handoff. It consists of transferring an ongoing session from a standard to another. Also this implementation requires a flexible architecture to incorporate multiple Radio Access Technologies (RATs) and facilitate the interworking between them. This work presents the 5G network architecture and some enabling key techniques it presents also a review of interworking methods, and introduces a new interworking technique among Long Term Evolution (LTE) and WiMax. Simulation results show the capacity of our methods to offload flows from different RATs.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115152737","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 : 2017-05-22DOI: 10.1109/ATSIP.2017.8075537
Yassine Nasser, M. Hassouni, A. Brahim, H. Toumi, E. Lespessailles, R. Jennane
This paper focuses on the problem of osteoporosis disease diagnosis from bone X-ray images. The proposed approach takes advantage of the deep learning robustness to extract high-level features from low-level image (pixel intensities). However, the diagnosis of osteoporosis confronts two major challenges, the difficulty of distinguishing between osteoporosis and healthy subjects just from the visual inspection of bone X-ray images, and the need of a large-scale and small-size datasets for training the deep networks. In order to separate the Osteoporosis population (OP) from Control cases (CC) our proposed method performs a series of three consecutive steps, namely: 1) preprocessing to enhance the contrast of the image, 2) image subdivision with the sliding window operation and feature extraction with Staked Sparse Autoencoder (SSAE), 3) pooling operation followed by classification step using the SVM classifier. Experimental results indicate that a performance gain on classification of the two populations (OP and CC) was achieved.
{"title":"Diagnosis of osteoporosis disease from bone X-ray images with stacked sparse autoencoder and SVM classifier","authors":"Yassine Nasser, M. Hassouni, A. Brahim, H. Toumi, E. Lespessailles, R. Jennane","doi":"10.1109/ATSIP.2017.8075537","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075537","url":null,"abstract":"This paper focuses on the problem of osteoporosis disease diagnosis from bone X-ray images. The proposed approach takes advantage of the deep learning robustness to extract high-level features from low-level image (pixel intensities). However, the diagnosis of osteoporosis confronts two major challenges, the difficulty of distinguishing between osteoporosis and healthy subjects just from the visual inspection of bone X-ray images, and the need of a large-scale and small-size datasets for training the deep networks. In order to separate the Osteoporosis population (OP) from Control cases (CC) our proposed method performs a series of three consecutive steps, namely: 1) preprocessing to enhance the contrast of the image, 2) image subdivision with the sliding window operation and feature extraction with Staked Sparse Autoencoder (SSAE), 3) pooling operation followed by classification step using the SVM classifier. Experimental results indicate that a performance gain on classification of the two populations (OP and CC) was achieved.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115468639","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 : 2017-05-22DOI: 10.1109/ATSIP.2017.8075519
Noureddine Ait Ali, B. Cherradi, A. Abbassi, O. Bouattane, M. Youssfi
In order to save patients with cerebral tumor disease, analysis and time processing of MRI brain images must be efficient, fast and relevant. The implementation of BCFCM algorithm on parallel graphics cards (GPUs) is an adequate remedy for the problem of processing time which can be elevated in urgent pathological cases. In this paper we present two implementations of Bias Correction Fuzzy C-means Algorithm using GPU card. Indeed we have already parallelized this algorithm, but this time we have enhanced the implementation, first by using the released mode instead of debug mode which is slow in execution time compared to release mode. Also, we have included the image edge pixels which were not the case in the previous work. Moreover, we have introduced and applied another method that gives interesting results compared to the other one. In the rest of this paper we will give the main steps of each implementation and then compare the new results in term of execution time and speedups.
{"title":"New parallel hybrid implementation of bias correction fuzzy C-means algorithm","authors":"Noureddine Ait Ali, B. Cherradi, A. Abbassi, O. Bouattane, M. Youssfi","doi":"10.1109/ATSIP.2017.8075519","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075519","url":null,"abstract":"In order to save patients with cerebral tumor disease, analysis and time processing of MRI brain images must be efficient, fast and relevant. The implementation of BCFCM algorithm on parallel graphics cards (GPUs) is an adequate remedy for the problem of processing time which can be elevated in urgent pathological cases. In this paper we present two implementations of Bias Correction Fuzzy C-means Algorithm using GPU card. Indeed we have already parallelized this algorithm, but this time we have enhanced the implementation, first by using the released mode instead of debug mode which is slow in execution time compared to release mode. Also, we have included the image edge pixels which were not the case in the previous work. Moreover, we have introduced and applied another method that gives interesting results compared to the other one. In the rest of this paper we will give the main steps of each implementation and then compare the new results in term of execution time and speedups.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129839092","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 : 2017-05-22DOI: 10.1109/ATSIP.2017.8075522
H. Arahmane, R. Moursli, E. Hamzaoui
In this work, we apply Nonnegative Matrix Factorization (NMF) algorithms of the blind source separation methods to extract independent components from signals recorded at the output of fission chamber detector, which is used to perform the flux-mapping within the nuclear research reactors. The simulation of the recorded signals is based on using the python-based of Fission Chambers (pyFC) suite code, employs the TRIM code and the Bolzig software. The output signals of the simulated fission chamber will be processed through Nonnegative Matrix Factorization techniques in order to achieve blind source separation task. The selection of the most efficient NMF technique is carried out by computing the performance index of separability of each algorithm and the extracted independent components that will be characterized by using time-frequency representation.
{"title":"Fission chamber's neutron signal characterization using nonnegative matrix factorization","authors":"H. Arahmane, R. Moursli, E. Hamzaoui","doi":"10.1109/ATSIP.2017.8075522","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075522","url":null,"abstract":"In this work, we apply Nonnegative Matrix Factorization (NMF) algorithms of the blind source separation methods to extract independent components from signals recorded at the output of fission chamber detector, which is used to perform the flux-mapping within the nuclear research reactors. The simulation of the recorded signals is based on using the python-based of Fission Chambers (pyFC) suite code, employs the TRIM code and the Bolzig software. The output signals of the simulated fission chamber will be processed through Nonnegative Matrix Factorization techniques in order to achieve blind source separation task. The selection of the most efficient NMF technique is carried out by computing the performance index of separability of each algorithm and the extracted independent components that will be characterized by using time-frequency representation.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"255 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116807809","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 : 2017-05-22DOI: 10.1109/ATSIP.2017.8075520
N. Aharrane, A. Dahmouni, K. E. Moutaouakil, K. Satori
In this work, we present an end-to-end system devoted to automatic recognition of printed Amazigh script in complex document images containing different languages such as Web images and natural scene images. To this end, text extraction from images is performed; the extracted text serves as input for a trained convolutional neural network (CNN) to identify its language. Finally, we proceed to the recognition of the Amazigh text script using a developed optical character recognition (OCR) system. The CNN reaches 99,12% of accuracy while the OCR system gets 99,93%. The obtained results seem to be very satisfactory.
{"title":"End-to-end system for printed Amazigh script recognition in document images","authors":"N. Aharrane, A. Dahmouni, K. E. Moutaouakil, K. Satori","doi":"10.1109/ATSIP.2017.8075520","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075520","url":null,"abstract":"In this work, we present an end-to-end system devoted to automatic recognition of printed Amazigh script in complex document images containing different languages such as Web images and natural scene images. To this end, text extraction from images is performed; the extracted text serves as input for a trained convolutional neural network (CNN) to identify its language. Finally, we proceed to the recognition of the Amazigh text script using a developed optical character recognition (OCR) system. The CNN reaches 99,12% of accuracy while the OCR system gets 99,93%. The obtained results seem to be very satisfactory.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"03 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130455746","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 : 2017-05-22DOI: 10.1109/ATSIP.2017.8075573
Ouarda Lamrabet, N. Fezazi, F. E. Haoussi, E. Tissir
In this paper, input delay theory is used as a key tool to design an effective anti-windup compensator which is applied to active queue management (AQM) at router to avoid congestion. The desired anti-windup compensator can be computed by the solution of convex optimization problem. The simulation results suggest that our methodology is effective and is an improvement over previous ones.
{"title":"Using input delay approach for synthesizing an anti-windup compensator to AQM in TCP/IP networks","authors":"Ouarda Lamrabet, N. Fezazi, F. E. Haoussi, E. Tissir","doi":"10.1109/ATSIP.2017.8075573","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075573","url":null,"abstract":"In this paper, input delay theory is used as a key tool to design an effective anti-windup compensator which is applied to active queue management (AQM) at router to avoid congestion. The desired anti-windup compensator can be computed by the solution of convex optimization problem. The simulation results suggest that our methodology is effective and is an improvement over previous ones.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125575751","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 : 2017-05-22DOI: 10.1109/ATSIP.2017.8075580
S. Dahbi, A. Aziz, S. Zouggar, M. E. Hafyani, Oujda Morocco, A. Hanafi, M. Karim, I. Latachi, T. Rachidi
In April 2014, a nano-satellite project was initiated at the University of Sidi Mohamed Ben Abdellah, Fez, Morocco in order to give graduate students the rare opportunity to experience all the challenging aspects associated with satellite design, manufacture, launch and operation. The nano-satellite follows the design concept of a 1U CubeSat, which has the shape of a cube of 10cm3 edge, of 1.33kg mass and which consumes a power of 1W. The initial orbital parameters of the nano-satellite were set to a circular sun-synchronous Low Earth Orbit (LEO) with an inclination of 90°. The operation time was set to two years. To ensure a successful flight, it is necessary, within the scope of the project, to establish an adequate power budget to ensure that the consumed power on board does not exceed the produced power. In this work we analyze deeply the available energy required per single orbit, and summarize the estimated produced and consumed power for several operation modes of the nanosatellite in the execution of its mission.
{"title":"Power budget analysis for a LEO polar orbiting nano-satellite","authors":"S. Dahbi, A. Aziz, S. Zouggar, M. E. Hafyani, Oujda Morocco, A. Hanafi, M. Karim, I. Latachi, T. Rachidi","doi":"10.1109/ATSIP.2017.8075580","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075580","url":null,"abstract":"In April 2014, a nano-satellite project was initiated at the University of Sidi Mohamed Ben Abdellah, Fez, Morocco in order to give graduate students the rare opportunity to experience all the challenging aspects associated with satellite design, manufacture, launch and operation. The nano-satellite follows the design concept of a 1U CubeSat, which has the shape of a cube of 10cm3 edge, of 1.33kg mass and which consumes a power of 1W. The initial orbital parameters of the nano-satellite were set to a circular sun-synchronous Low Earth Orbit (LEO) with an inclination of 90°. The operation time was set to two years. To ensure a successful flight, it is necessary, within the scope of the project, to establish an adequate power budget to ensure that the consumed power on board does not exceed the produced power. In this work we analyze deeply the available energy required per single orbit, and summarize the estimated produced and consumed power for several operation modes of the nanosatellite in the execution of its mission.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125726151","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 : 2017-05-22DOI: 10.1109/ATSIP.2017.8075579
A. Tahraoui, R. Khedam, A. Bouakache, A. B. Aissa
The aim of this paper is to present a new unsupervised classification method for satellite multispectral images based on affinity propagation (AP) algorithm. Recently proposed, affinity propagation becomes the most widely methods for data clustering. This technique is essentially based on passing of messages between pixels to be automatically classified without any a priori knowledge about the number of classes. Its main advantage is that initially all pixels to classify are considered as centroids or “exemplars”. However, the AP method has two major drawbacks: 1) when it comes to partition multispectral images of high spatial size, complexity of computation becomes quadratic 2) it gives an overestimation of class number due to its great sensitivity to very small variations in the image. In this work, we present the AP algorithm in its original version and we have proposed an iterative AP-block procedure to address the two issues mentioned above. Both versions have been applied to classify a low spatial resolution image acquired by ETM+ sensor of american satellite LandSat-7. From obtained results, it is concluded that the proposed AP-block classifier is more appropriate and more efficient to unsupervised image classification than the classical AP algorithm.
{"title":"Affinity propagation for unsupervised classification of remotely sensed images","authors":"A. Tahraoui, R. Khedam, A. Bouakache, A. B. Aissa","doi":"10.1109/ATSIP.2017.8075579","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075579","url":null,"abstract":"The aim of this paper is to present a new unsupervised classification method for satellite multispectral images based on affinity propagation (AP) algorithm. Recently proposed, affinity propagation becomes the most widely methods for data clustering. This technique is essentially based on passing of messages between pixels to be automatically classified without any a priori knowledge about the number of classes. Its main advantage is that initially all pixels to classify are considered as centroids or “exemplars”. However, the AP method has two major drawbacks: 1) when it comes to partition multispectral images of high spatial size, complexity of computation becomes quadratic 2) it gives an overestimation of class number due to its great sensitivity to very small variations in the image. In this work, we present the AP algorithm in its original version and we have proposed an iterative AP-block procedure to address the two issues mentioned above. Both versions have been applied to classify a low spatial resolution image acquired by ETM+ sensor of american satellite LandSat-7. From obtained results, it is concluded that the proposed AP-block classifier is more appropriate and more efficient to unsupervised image classification than the classical AP algorithm.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121517311","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 : 2017-05-22DOI: 10.1109/ATSIP.2017.8075550
Oumaima Zine, A. Meziane
Text To Speech technologies are widely being used in applications to help users with special needs such as blind, deafened, individuals with severe speech impairments and dyslexics. In this context particular focus has been given by the Text To Speech (TTS) researchers to achieve a high level of intelligibility for many languages such as French and English. However, Arabic TTS is still in its early development stages and needs to be improved to reach high quality. In this paper, we describe a novel concatenative approach based on lemma and Arabic patterns. Moreover, an alternative method for synthesizing diacritized Arabic texts is proposed and adopted, using a set of sub-segments where the consonant is considered as the nucleus of the acoustic unit, and hence this latter is taken with its vocalic context. This speech unit consists of half vowel-Consonant-Half vowel, adapted to the different positions in the word (Initial, medial and final). A reduction process of the resulted combinations of the proposed acoustic units will also be described in this work, in order to reduce the theoretical number of the generated sub-segment models. Furthermore, a speech corpora design for Arabic Text To Speech (ATTS) based on pre-recorded Audiobooks from Masmoo3 Audiobooks website, is presented. The corpus contains more than 4 hours of a continuous speech of Modern Standard Arabic (MSA) recorded in high intelligibility and providing phonetically balanced sentences. The proposed approach was evaluated using the Diagnostic Rhyme Test (DRT) that measures the intelligibility of the synthesized speech on the word-level. The sentence-level test was also conducted. The results of the both tests are illustrated in the experiments and results section.
{"title":"Novel approach for quality enhancement of Arabic text to speech synthesis","authors":"Oumaima Zine, A. Meziane","doi":"10.1109/ATSIP.2017.8075550","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075550","url":null,"abstract":"Text To Speech technologies are widely being used in applications to help users with special needs such as blind, deafened, individuals with severe speech impairments and dyslexics. In this context particular focus has been given by the Text To Speech (TTS) researchers to achieve a high level of intelligibility for many languages such as French and English. However, Arabic TTS is still in its early development stages and needs to be improved to reach high quality. In this paper, we describe a novel concatenative approach based on lemma and Arabic patterns. Moreover, an alternative method for synthesizing diacritized Arabic texts is proposed and adopted, using a set of sub-segments where the consonant is considered as the nucleus of the acoustic unit, and hence this latter is taken with its vocalic context. This speech unit consists of half vowel-Consonant-Half vowel, adapted to the different positions in the word (Initial, medial and final). A reduction process of the resulted combinations of the proposed acoustic units will also be described in this work, in order to reduce the theoretical number of the generated sub-segment models. Furthermore, a speech corpora design for Arabic Text To Speech (ATTS) based on pre-recorded Audiobooks from Masmoo3 Audiobooks website, is presented. The corpus contains more than 4 hours of a continuous speech of Modern Standard Arabic (MSA) recorded in high intelligibility and providing phonetically balanced sentences. The proposed approach was evaluated using the Diagnostic Rhyme Test (DRT) that measures the intelligibility of the synthesized speech on the word-level. The sentence-level test was also conducted. The results of the both tests are illustrated in the experiments and results section.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"59 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116571445","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 : 2017-05-22DOI: 10.1109/ATSIP.2017.8075593
Sara Sekkate, Mohammed Khalil, A. Adib
This paper examines the development of a speaker identification system (SIS) for future aeronautical communication systems. SIS promises to improve flight safety by reducing the incidence of call-sign confusion events. However, the practical development of such a system faces many challenges, especially related to the signal corruption by the channel noise. Due to the dynamic motion of aircraft, the aeronautical channel experiences high Doppler shifts and fading due to multipath propagation. This means that the SIS is required to be robust against such perturbations. In the proposed system, aeronautical channel noise was generated and mixed with speech signals to get the testing data. Four spectral features including Linear Predictive Cepstral Coefficients (LPCC), Perceptual Linear Prediction (PLP), Mel Frequency Cepstral Coefficient (MFCC) and Gammatone Frequency Cepstral Coefficient (GFCC) were extracted and then Support Vector Machines (SVM) were used for classification. The performance of the system was evaluated using noiseless and noisy signals from the ATCOSIM speech corpus. The experimental results show that the better recognition rate is obtained for GFCC under noisy conditions as compared to PLP, LPCC and MFCC.
{"title":"Speaker identification: A way to reduce call-sign confusion events","authors":"Sara Sekkate, Mohammed Khalil, A. Adib","doi":"10.1109/ATSIP.2017.8075593","DOIUrl":"https://doi.org/10.1109/ATSIP.2017.8075593","url":null,"abstract":"This paper examines the development of a speaker identification system (SIS) for future aeronautical communication systems. SIS promises to improve flight safety by reducing the incidence of call-sign confusion events. However, the practical development of such a system faces many challenges, especially related to the signal corruption by the channel noise. Due to the dynamic motion of aircraft, the aeronautical channel experiences high Doppler shifts and fading due to multipath propagation. This means that the SIS is required to be robust against such perturbations. In the proposed system, aeronautical channel noise was generated and mixed with speech signals to get the testing data. Four spectral features including Linear Predictive Cepstral Coefficients (LPCC), Perceptual Linear Prediction (PLP), Mel Frequency Cepstral Coefficient (MFCC) and Gammatone Frequency Cepstral Coefficient (GFCC) were extracted and then Support Vector Machines (SVM) were used for classification. The performance of the system was evaluated using noiseless and noisy signals from the ATCOSIM speech corpus. The experimental results show that the better recognition rate is obtained for GFCC under noisy conditions as compared to PLP, LPCC and MFCC.","PeriodicalId":259951,"journal":{"name":"2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129719857","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：