Pub Date : 2020-02-01DOI: 10.1109/NCC48643.2020.9055998
T. Pratap, Priyanka Kokil
In this paper, a methodology to improve the performance of existing automatic cataract detection systems (ACDS) in noisy/blur environment is proposed. The presented approach consists of dual-threshold based image quality evaluation module to enhance the performance diminution of ACDS in noisy/blur environment. Initially the first threshold is obtained from naturalness image quality evaluator (NIQE) and then second threshold is achieved through noise level estimation (NLE). In order to ensure robustness, the proposed method is evaluated with artificially created noise and blur datasets in association with existing pre-trained convolution neural network based ACDS. The experiments results show superiority in performance over existing methods in literature.
{"title":"Correcting Automatic Cataract Diagnosis Systems Against Noisy/Blur Environment","authors":"T. Pratap, Priyanka Kokil","doi":"10.1109/NCC48643.2020.9055998","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9055998","url":null,"abstract":"In this paper, a methodology to improve the performance of existing automatic cataract detection systems (ACDS) in noisy/blur environment is proposed. The presented approach consists of dual-threshold based image quality evaluation module to enhance the performance diminution of ACDS in noisy/blur environment. Initially the first threshold is obtained from naturalness image quality evaluator (NIQE) and then second threshold is achieved through noise level estimation (NLE). In order to ensure robustness, the proposed method is evaluated with artificially created noise and blur datasets in association with existing pre-trained convolution neural network based ACDS. The experiments results show superiority in performance over existing methods in literature.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124180247","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 : 2020-02-01DOI: 10.1109/NCC48643.2020.9056019
N. Joshi, P. Peshwe, A. Kothari
In the present work, a dual feed, compact circularly polarized antenna is designed for activity classification purpose. It consists of a circular ring on the top of the substrate which acts as the radiating patch. The ground plane exactly compliments the patch and thus is a circle with radius equal to inner radius of the ring. A rectangular stub is added to the ground plane for impedance matching. The antenna has been fabricated and the measured results are in very good agreement with the simulations. Excellent circular polarization performance is observed in the antenna which is highly desirable for the intended application. The transmission and reflection co-efficient of the antenna are a function of motion activities. This is due to specific obstruction of EM waves by the antenna when involved in performing daily human activities. Datasets have been collected by actual activity performance involving the fabricated antenna. Specific and distinct signatures of S11 parameter have been obtained for different activities. Thus, the antenna can be used for activity classification purpose.
{"title":"A Compact, Circularly Polarized Antenna for Human Activity Classification","authors":"N. Joshi, P. Peshwe, A. Kothari","doi":"10.1109/NCC48643.2020.9056019","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9056019","url":null,"abstract":"In the present work, a dual feed, compact circularly polarized antenna is designed for activity classification purpose. It consists of a circular ring on the top of the substrate which acts as the radiating patch. The ground plane exactly compliments the patch and thus is a circle with radius equal to inner radius of the ring. A rectangular stub is added to the ground plane for impedance matching. The antenna has been fabricated and the measured results are in very good agreement with the simulations. Excellent circular polarization performance is observed in the antenna which is highly desirable for the intended application. The transmission and reflection co-efficient of the antenna are a function of motion activities. This is due to specific obstruction of EM waves by the antenna when involved in performing daily human activities. Datasets have been collected by actual activity performance involving the fabricated antenna. Specific and distinct signatures of S11 parameter have been obtained for different activities. Thus, the antenna can be used for activity classification purpose.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130065741","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 : 2020-02-01DOI: 10.1109/NCC48643.2020.9056084
Abhijit Mohanta, V. K. Mittal
Children affected with autism spectrum disorder (ASD) produce speech that consists of distinctive acoustic patterns, as compared to normal children. Hence, acoustic analyses can help classifying speech of ASD affected children from that of normal children. In this study, the aim is to identify those discriminating characteristics of speech production that help classification between speech of children with ASD and normal children. Two separate datasets were recorded for this study: the English speech of children affected with ASD and the English speech of normal children. Comparative analyses of acoustic features derived for both datasets are carried out. Changes in the speech production characteristics are examined in three parts. Firstly, changes in the excitation source features F0 and strength of excitation (SoE) are analyzed. Secondly, changes in the vocal tract filter features the formants (F1 to F5) and dominant frequencies (FD1, FD2) are analyzed. Thirdly, changes in the combined source-filter features signal energy and zero-crossing rate are analyzed. Different combinations of the feature sets are then classified using three different classifiers for validation of results: SVM, KNN and ensemble classifiers. Performance evaluation is carried using different combinations of features sets and classifiers. Results up to 97.1% are obtained for classification accuracy between speech of ASD affected children and normal children, using a combination of feature set with SVM classifier. The results are better than other similar few studies. This study should be helpful in developing an automated system for identffying ASD speech, in future.
{"title":"Classifying Speech of ASD Affected and Normal Children Using Acoustic Features","authors":"Abhijit Mohanta, V. K. Mittal","doi":"10.1109/NCC48643.2020.9056084","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9056084","url":null,"abstract":"Children affected with autism spectrum disorder (ASD) produce speech that consists of distinctive acoustic patterns, as compared to normal children. Hence, acoustic analyses can help classifying speech of ASD affected children from that of normal children. In this study, the aim is to identify those discriminating characteristics of speech production that help classification between speech of children with ASD and normal children. Two separate datasets were recorded for this study: the English speech of children affected with ASD and the English speech of normal children. Comparative analyses of acoustic features derived for both datasets are carried out. Changes in the speech production characteristics are examined in three parts. Firstly, changes in the excitation source features F0 and strength of excitation (SoE) are analyzed. Secondly, changes in the vocal tract filter features the formants (F1 to F5) and dominant frequencies (FD1, FD2) are analyzed. Thirdly, changes in the combined source-filter features signal energy and zero-crossing rate are analyzed. Different combinations of the feature sets are then classified using three different classifiers for validation of results: SVM, KNN and ensemble classifiers. Performance evaluation is carried using different combinations of features sets and classifiers. Results up to 97.1% are obtained for classification accuracy between speech of ASD affected children and normal children, using a combination of feature set with SVM classifier. The results are better than other similar few studies. This study should be helpful in developing an automated system for identffying ASD speech, in future.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131054077","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 : 2020-02-01DOI: 10.1109/NCC48643.2020.9056078
Rakesh Munagala, Rohit Budhiraja
The 5G new radio (NR) cellular system transmits demodulation reference signals (DM-RS) for a user to estimate precoded channel. The DM-RS can be optionally transmitted at various orthogonal frequency division multiplexing (OFDM) symbols. The 5G NR specifications does not explicitly tell how many DM-RS should be used, and needs to be decided by system designers. This work designs various channel estimators for the DM-RS, and based on the MSE performance, proposes an algorithm to decide the number of DM-RS required at a particular user speed.
{"title":"Channel Estimator Designs For Emerging 5G New Radio Cellular Systems","authors":"Rakesh Munagala, Rohit Budhiraja","doi":"10.1109/NCC48643.2020.9056078","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9056078","url":null,"abstract":"The 5G new radio (NR) cellular system transmits demodulation reference signals (DM-RS) for a user to estimate precoded channel. The DM-RS can be optionally transmitted at various orthogonal frequency division multiplexing (OFDM) symbols. The 5G NR specifications does not explicitly tell how many DM-RS should be used, and needs to be decided by system designers. This work designs various channel estimators for the DM-RS, and based on the MSE performance, proposes an algorithm to decide the number of DM-RS required at a particular user speed.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126609907","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 : 2020-02-01DOI: 10.1109/NCC48643.2020.9056065
Deepti Agarwal, Ankur Bansal
This paper analyzes the performance of serial decode-and-forward (DF) relay assisted free-space-optical (FSO) network with pointing errors and finite-sized receivers employing differential M-ary phase shift keying (DMPSK) data. The atmospheric fading optical links are modeled by unified Gamma-Gamma (ΓΓ) distribution subject to both heterodyne and intensity modulation/direct detection (IM/DD) techniques. In particular, we derive the average symbol error rate (SER) by utilizing symbol transition probability matrix (STPM) whose entries are the average symbol transition probabilities (ASTPs) of a relay. The ASTPs of single link STPM are then utilized to calculate the SER of serial DF relaying network. Further, the unified outage probability of considered network is obtained. The results indicate that the point receiver performance is more affected with pointing error as compared to the finite-sized receiver. Further, it is showcased through results that when the number of serial relays increases, the improvement in error performance is more in case of heterodyne as compared to that of IM/DD.
{"title":"STPM Based Performance Analysis of Finite-Sized Differential Serial FSO Network","authors":"Deepti Agarwal, Ankur Bansal","doi":"10.1109/NCC48643.2020.9056065","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9056065","url":null,"abstract":"This paper analyzes the performance of serial decode-and-forward (DF) relay assisted free-space-optical (FSO) network with pointing errors and finite-sized receivers employing differential M-ary phase shift keying (DMPSK) data. The atmospheric fading optical links are modeled by unified Gamma-Gamma (ΓΓ) distribution subject to both heterodyne and intensity modulation/direct detection (IM/DD) techniques. In particular, we derive the average symbol error rate (SER) by utilizing symbol transition probability matrix (STPM) whose entries are the average symbol transition probabilities (ASTPs) of a relay. The ASTPs of single link STPM are then utilized to calculate the SER of serial DF relaying network. Further, the unified outage probability of considered network is obtained. The results indicate that the point receiver performance is more affected with pointing error as compared to the finite-sized receiver. Further, it is showcased through results that when the number of serial relays increases, the improvement in error performance is more in case of heterodyne as compared to that of IM/DD.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114865067","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 : 2020-02-01DOI: 10.1109/NCC48643.2020.9056056
Sukla Satapathy, R. R. Sahay
Occurrence of regions with missing data in depth maps either captured by active sensors or estimated by different passive computer vision algorithms, is unavoidable due to several reasons. The task of depth inpainting from a single degraded depth map is more challenging as compared to using multiple depth observations or RGB-D data. Recently, low rank techniques have become popular and shown supremacy over several state-of-the-art techniques for image deblurring, denoising, upsampling, etc. Since completion of missing regions in a given degraded depth observation is a severely ill-posed problem, low rank property of the inpainted depth map can be posed as the regularization constraint. We perform several experiments to show the superiority of the proposed method over the state-of-the-art depth inpainting techniques.
{"title":"Exploiting Low Rank Prior for Depth Map Completion","authors":"Sukla Satapathy, R. R. Sahay","doi":"10.1109/NCC48643.2020.9056056","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9056056","url":null,"abstract":"Occurrence of regions with missing data in depth maps either captured by active sensors or estimated by different passive computer vision algorithms, is unavoidable due to several reasons. The task of depth inpainting from a single degraded depth map is more challenging as compared to using multiple depth observations or RGB-D data. Recently, low rank techniques have become popular and shown supremacy over several state-of-the-art techniques for image deblurring, denoising, upsampling, etc. Since completion of missing regions in a given degraded depth observation is a severely ill-posed problem, low rank property of the inpainted depth map can be posed as the regularization constraint. We perform several experiments to show the superiority of the proposed method over the state-of-the-art depth inpainting techniques.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123926383","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 : 2020-02-01DOI: 10.1109/NCC48643.2020.9056075
A. Punnoose
This paper discusses an approach for confidence scoring at the phoneme level. Various features derived from multi layer perceptron (MLP) posteriors that indicates the strength of a phoneme detection are introduced. The capability of these features to discriminate between true positive and false positive phoneme detection is demonstrated. Appropriate distributions are fit on these features. These distributions are combined to derive the posterior odds ratio, which signals the confidence of a phoneme detection. Finally, simple thresholding on the posterior odds ratio is used to classify a detected phoneme as true/false positive. Relevant real world datasets are used to benchmark the proposed approach.
{"title":"Substate Detection Based Confidence Scoring in Speech Recognition","authors":"A. Punnoose","doi":"10.1109/NCC48643.2020.9056075","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9056075","url":null,"abstract":"This paper discusses an approach for confidence scoring at the phoneme level. Various features derived from multi layer perceptron (MLP) posteriors that indicates the strength of a phoneme detection are introduced. The capability of these features to discriminate between true positive and false positive phoneme detection is demonstrated. Appropriate distributions are fit on these features. These distributions are combined to derive the posterior odds ratio, which signals the confidence of a phoneme detection. Finally, simple thresholding on the posterior odds ratio is used to classify a detected phoneme as true/false positive. Relevant real world datasets are used to benchmark the proposed approach.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124056174","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 : 2020-02-01DOI: 10.1109/NCC48643.2020.9055999
Udayeni Anumala, M. Okade
This paper presents a novel application of local tetra patterns to the median filtering detection problem. The premise of the proposed method is based on the ability of the local tetra patterns in identifying the streaking fingerprints left over by the application of a median filter on an image. These streaking fingerprints serve as a clue in determining the authenticity of an image towards the application of a median filter. The streaking pixels are identified by establishing the relationship of every pixel with respect to its neighboring pixels. The relationship is in the form of horizontal and vertical derivative directions and magnitudes followed by the tetra pattern and magnitude assignment. The feature vector generated utilizing the local tetra patterns is reduced by using the J-divergence in-order to keep the computational complexity low. Experimental testing for the proposed method along with comparative analysis carried out with existing state-of-the-art methods shows good performance at reduced computational complexity for the proposed method.
{"title":"Forensic detection of Median filtering in Images using Local Tetra Patterns and J-Divergence","authors":"Udayeni Anumala, M. Okade","doi":"10.1109/NCC48643.2020.9055999","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9055999","url":null,"abstract":"This paper presents a novel application of local tetra patterns to the median filtering detection problem. The premise of the proposed method is based on the ability of the local tetra patterns in identifying the streaking fingerprints left over by the application of a median filter on an image. These streaking fingerprints serve as a clue in determining the authenticity of an image towards the application of a median filter. The streaking pixels are identified by establishing the relationship of every pixel with respect to its neighboring pixels. The relationship is in the form of horizontal and vertical derivative directions and magnitudes followed by the tetra pattern and magnitude assignment. The feature vector generated utilizing the local tetra patterns is reduced by using the J-divergence in-order to keep the computational complexity low. Experimental testing for the proposed method along with comparative analysis carried out with existing state-of-the-art methods shows good performance at reduced computational complexity for the proposed method.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122904633","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 : 2020-02-01DOI: 10.1109/NCC48643.2020.9056052
Sidharth Aggarwal, Rini A. Sharon, H. Murthy
P300 is widely used for developing Brain-Computer Interfaces (BCIs) and also in clinical applications for research and diagnosis. In this study, a novel way of performing oddball paradigm by stereo-localization of single frequency audio stimulus is proposed. In the proposed stereo oddball technique, a single frequency audio stimulus is presented to the subject in alternating ears with one ear being the target and the other non-target. Non-target is presented more often than target. The experiments are conducted for two configurations, left (target) - right (non-target) and right (target) - left (non-target). Noninvasive Electroencephalogram (EEG) signals are collected for the above mentioned protocol and the P300 component is detected using event-related potentials (ERPs) and analyzed. The proposed Stereo oddball technique is also compared with classical (target and non-target are beeps of different frequency) oddball technique, where the stimulus is presented simultaneously to both ears. The P300 responses are also analyzed using both temporal regions individually. Despite differing inputs(single frequency and dual frequency), similar P300 responses are observed for stereo localized and binaural stimuli presentations.
{"title":"P300 based Stereo localization of single frequency audio stimulus","authors":"Sidharth Aggarwal, Rini A. Sharon, H. Murthy","doi":"10.1109/NCC48643.2020.9056052","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9056052","url":null,"abstract":"P300 is widely used for developing Brain-Computer Interfaces (BCIs) and also in clinical applications for research and diagnosis. In this study, a novel way of performing oddball paradigm by stereo-localization of single frequency audio stimulus is proposed. In the proposed stereo oddball technique, a single frequency audio stimulus is presented to the subject in alternating ears with one ear being the target and the other non-target. Non-target is presented more often than target. The experiments are conducted for two configurations, left (target) - right (non-target) and right (target) - left (non-target). Noninvasive Electroencephalogram (EEG) signals are collected for the above mentioned protocol and the P300 component is detected using event-related potentials (ERPs) and analyzed. The proposed Stereo oddball technique is also compared with classical (target and non-target are beeps of different frequency) oddball technique, where the stimulus is presented simultaneously to both ears. The P300 responses are also analyzed using both temporal regions individually. Despite differing inputs(single frequency and dual frequency), similar P300 responses are observed for stereo localized and binaural stimuli presentations.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129754151","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 : 2020-02-01DOI: 10.1109/NCC48643.2020.9055989
P. Dileep, Dibyajyoti Das, P. Bora
Automatic modulation classification (AMC) is an important part of signal identification for cognitive radio as well as military communication. The problem has been approached traditionally using either likelihood-based or feature-based methods. Since the problem is a classification task, a deep learning (DL) based approach can be an attractive solution. A number of convolutional neural network (CNN) based DL algorithms were introduced for AMC recently. The complex baseband signals that are represented as In-phase and Quadrature (IQ) samples are applied to train the CNN. We propose a new CNN architecture that significantly improves the classification accuracy over existing results in the literature while keeping the number of trainable parameters low. In this architecture, dropouts are applied only in the dense layers.
{"title":"Dense Layer Dropout Based CNN Architecture for Automatic Modulation Classification","authors":"P. Dileep, Dibyajyoti Das, P. Bora","doi":"10.1109/NCC48643.2020.9055989","DOIUrl":"https://doi.org/10.1109/NCC48643.2020.9055989","url":null,"abstract":"Automatic modulation classification (AMC) is an important part of signal identification for cognitive radio as well as military communication. The problem has been approached traditionally using either likelihood-based or feature-based methods. Since the problem is a classification task, a deep learning (DL) based approach can be an attractive solution. A number of convolutional neural network (CNN) based DL algorithms were introduced for AMC recently. The complex baseband signals that are represented as In-phase and Quadrature (IQ) samples are applied to train the CNN. We propose a new CNN architecture that significantly improves the classification accuracy over existing results in the literature while keeping the number of trainable parameters low. In this architecture, dropouts are applied only in the dense layers.","PeriodicalId":183772,"journal":{"name":"2020 National Conference on Communications (NCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128719372","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：