Pub Date : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840803
Rupender Singh, E. Yaacoub
The emerging fifth generation (5G) wireless communication systems are a promising solution to support the high data rate for high-speed trains (HSTs). In this article, we propose a cooperative mixed radio frequency (RF)-visible light communication (VLC) system to avoid the high penetration loss of the direct link between the end-users (UE) and the base station (BS). Our motivation is mainly based on the suitability of VLC technologies for indoor systems and their advantages to tackle forthcoming spectrum crunch, wide spectral availability, and easy bandwidth reuse. In the proposed system setup, the outdoor BS-relay is served by the backhaul RF links subject to double shadowing due to slow moving obstacles and pedestrian, while the data traffic inside the HST is conveyed by an indoor VLC system. Moreover, it is assumed that the relay node has imperfect channel state information (CSI) due to the mobility of the HST. We first statistically characterize the end-to-end signal-to-noise ratios (SNRs). Then, the performance is analyzed by deriving the exact closed-form expressions for key metrics such as outage probability and average bit error rate (BER).
{"title":"On the Performance of Mixed RF-VLC Relaying Systems for HST Communication","authors":"Rupender Singh, E. Yaacoub","doi":"10.1109/SPCOM55316.2022.9840803","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840803","url":null,"abstract":"The emerging fifth generation (5G) wireless communication systems are a promising solution to support the high data rate for high-speed trains (HSTs). In this article, we propose a cooperative mixed radio frequency (RF)-visible light communication (VLC) system to avoid the high penetration loss of the direct link between the end-users (UE) and the base station (BS). Our motivation is mainly based on the suitability of VLC technologies for indoor systems and their advantages to tackle forthcoming spectrum crunch, wide spectral availability, and easy bandwidth reuse. In the proposed system setup, the outdoor BS-relay is served by the backhaul RF links subject to double shadowing due to slow moving obstacles and pedestrian, while the data traffic inside the HST is conveyed by an indoor VLC system. Moreover, it is assumed that the relay node has imperfect channel state information (CSI) due to the mobility of the HST. We first statistically characterize the end-to-end signal-to-noise ratios (SNRs). Then, the performance is analyzed by deriving the exact closed-form expressions for key metrics such as outage probability and average bit error rate (BER).","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130706370","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 : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840767
S. Shahnawazuddin, Vinit Kumar, Avinash Kumar, Waquar Ahmad
Developing an automatic speech recognition (ASR) system for children’s speech is extremely challenging due to the unavailability of data from the child domain for the majority of the languages. Consequently, in such zero-resource scenarios, we are forced to develop an ASR system using adults’ speech for transcribing data from child speakers. However, differences in formant frequencies and speaking-rate between the two groups of speakers degrade recognition performance. To reduce the said mismatch, out-of-domain data augmentation approaches based on formant and duration modification are proposed in this work. For that purpose, formant frequencies of adults’ speech training data are up-scaled using warping of linear predictive coding coefficients. Next, the speaking-rate of adults’ data is also increased through time-scale modification. Due to simultaneous altering of formant frequencies and duration of adults’ speech and then pooling the modified data into training, the acoustic mismatch due to the aforementioned factors gets reduced. This, in turn, enhances the recognition performance significantly. Additional improvement is obtained by combining the recently reported voice-conversion-based data augmentation technique with the proposed ones. On combining the proposed and voice-conversion-based data augmentation techniques, a relative reduction of nearly 32.3% in word error rate over the baseline is obtained.
{"title":"Improving the Performance of Zero-Resource Children’s ASR System through Formant and Duration Modification based Data Augmentation","authors":"S. Shahnawazuddin, Vinit Kumar, Avinash Kumar, Waquar Ahmad","doi":"10.1109/SPCOM55316.2022.9840767","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840767","url":null,"abstract":"Developing an automatic speech recognition (ASR) system for children’s speech is extremely challenging due to the unavailability of data from the child domain for the majority of the languages. Consequently, in such zero-resource scenarios, we are forced to develop an ASR system using adults’ speech for transcribing data from child speakers. However, differences in formant frequencies and speaking-rate between the two groups of speakers degrade recognition performance. To reduce the said mismatch, out-of-domain data augmentation approaches based on formant and duration modification are proposed in this work. For that purpose, formant frequencies of adults’ speech training data are up-scaled using warping of linear predictive coding coefficients. Next, the speaking-rate of adults’ data is also increased through time-scale modification. Due to simultaneous altering of formant frequencies and duration of adults’ speech and then pooling the modified data into training, the acoustic mismatch due to the aforementioned factors gets reduced. This, in turn, enhances the recognition performance significantly. Additional improvement is obtained by combining the recently reported voice-conversion-based data augmentation technique with the proposed ones. On combining the proposed and voice-conversion-based data augmentation techniques, a relative reduction of nearly 32.3% in word error rate over the baseline is obtained.","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114701727","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 : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840789
Renuka Acharya, N. Puhan
Diabetic Retinopathy (DR) has emerged as one of the serious medical conditions over the years leading to blindness among patients. Microaneurysms (MAs) are generally the earliest objective evidence of DR captured in fundus imaging. This work proposes a novel methodology based on long short-term memory (LSTM) to exploit the sequence dependencies of 1-D feature signals extracted from MAs and aid in their classification in colour fundus images. The model is trained using 1-dimensional intensity based signals generated from various patches of preprocessed fundus images. The model is tested on e-ophtha & ROC datasets and sensitivity scores are computed against seven unique values of false positive per image. The average of these scores is utilized as performance measurement of the proposed model which shows 66.6% and 60.5% sensitivity for e-ophtha and ROC datasets, respectively.
{"title":"Long Short-Term Memory Model Based Microaneurysm Sequence Classification in Fundus Images","authors":"Renuka Acharya, N. Puhan","doi":"10.1109/SPCOM55316.2022.9840789","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840789","url":null,"abstract":"Diabetic Retinopathy (DR) has emerged as one of the serious medical conditions over the years leading to blindness among patients. Microaneurysms (MAs) are generally the earliest objective evidence of DR captured in fundus imaging. This work proposes a novel methodology based on long short-term memory (LSTM) to exploit the sequence dependencies of 1-D feature signals extracted from MAs and aid in their classification in colour fundus images. The model is trained using 1-dimensional intensity based signals generated from various patches of preprocessed fundus images. The model is tested on e-ophtha & ROC datasets and sensitivity scores are computed against seven unique values of false positive per image. The average of these scores is utilized as performance measurement of the proposed model which shows 66.6% and 60.5% sensitivity for e-ophtha and ROC datasets, respectively.","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115850597","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 : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840812
Renuka Acharya, Soumya P. Dash
Clinical depression is one of the crucial medical conditions affecting a substantial portion of today’s world population. This paper proposes a novel technology based on deep learning tools for efficient detection of clinical depression in potential patients. A novel algorithm is proposed to label a recent AFEW-VA dataset in terms of creating the depressed class and the non-depressed class based on the valence and arousal values for various individuals from their video frames. Furthermore, the full facial regions, the eye regions, and the mouth regions from the classified dataset are extracted as the regions of interest (ROIs) to be utilized to train three different pre-trained 2DCNN models, namely, ResNet50, VGG16, and InceptionV3 by using transfer learning. For each 2D-CNN architecture, a novel algorithm is proposed to merge the models trained on the three ROIs. It is observed that the merged model, combining all the three ROIs outperforms the individual models or a merged model merging only two of the three ROIs in terms of obtaining a higher accuracy of depression detection. It is also observed that the merged models based on the ResNet50 architecture results in the best accuracy value of 0.95 as compared to the VGG16 and InceptionV3 architectures.
{"title":"Automatic Depression Detection Based on Merged Convolutional Neural Networks using Facial Features","authors":"Renuka Acharya, Soumya P. Dash","doi":"10.1109/SPCOM55316.2022.9840812","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840812","url":null,"abstract":"Clinical depression is one of the crucial medical conditions affecting a substantial portion of today’s world population. This paper proposes a novel technology based on deep learning tools for efficient detection of clinical depression in potential patients. A novel algorithm is proposed to label a recent AFEW-VA dataset in terms of creating the depressed class and the non-depressed class based on the valence and arousal values for various individuals from their video frames. Furthermore, the full facial regions, the eye regions, and the mouth regions from the classified dataset are extracted as the regions of interest (ROIs) to be utilized to train three different pre-trained 2DCNN models, namely, ResNet50, VGG16, and InceptionV3 by using transfer learning. For each 2D-CNN architecture, a novel algorithm is proposed to merge the models trained on the three ROIs. It is observed that the merged model, combining all the three ROIs outperforms the individual models or a merged model merging only two of the three ROIs in terms of obtaining a higher accuracy of depression detection. It is also observed that the merged models based on the ResNet50 architecture results in the best accuracy value of 0.95 as compared to the VGG16 and InceptionV3 architectures.","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117067810","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 : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840818
Iman Junaid, S. Ari
Gait as a biometric has become a popular research topic in recent years as a result of its numerous applications in sectors such as surveillance, authentication, and so on. It is capable of achieving detection at a distance that few other technologies can equal. It is still a difficult problem to solve since real human gait is influenced by several variable elements such as alterations in clothing, speed, and carrying situation. Also, unknown covariate circumstances may impact the training and testing settings for a specific individual in gait recognition. Image sequences are typically used by computer-aided gait recognition systems without taking into account variables such as clothes and the contents of carrier bags while on the move. In this work, we provide a technique for selecting gait energy image-based (GEI) features, that is both effective and robust. The covariate factors have less impact on the given gait representation. A simple ten-layered convolutional neural network (CNN) is designed which intakes GEI as input. Several typical variations and occlusions that impact and worsen gait recognition ability are less susceptible to the suggested method. For both clothing and mobility variations, we used the CASIA datasets to assess our observations. The experimental findings reveal that in numerous circumstances, the deep neural network model created in this study achieved better results when compared with other existing algorithms.
{"title":"Gait Recognition under Different Covariate Conditions using Deep Learning Technique","authors":"Iman Junaid, S. Ari","doi":"10.1109/SPCOM55316.2022.9840818","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840818","url":null,"abstract":"Gait as a biometric has become a popular research topic in recent years as a result of its numerous applications in sectors such as surveillance, authentication, and so on. It is capable of achieving detection at a distance that few other technologies can equal. It is still a difficult problem to solve since real human gait is influenced by several variable elements such as alterations in clothing, speed, and carrying situation. Also, unknown covariate circumstances may impact the training and testing settings for a specific individual in gait recognition. Image sequences are typically used by computer-aided gait recognition systems without taking into account variables such as clothes and the contents of carrier bags while on the move. In this work, we provide a technique for selecting gait energy image-based (GEI) features, that is both effective and robust. The covariate factors have less impact on the given gait representation. A simple ten-layered convolutional neural network (CNN) is designed which intakes GEI as input. Several typical variations and occlusions that impact and worsen gait recognition ability are less susceptible to the suggested method. For both clothing and mobility variations, we used the CASIA datasets to assess our observations. The experimental findings reveal that in numerous circumstances, the deep neural network model created in this study achieved better results when compared with other existing algorithms.","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"174 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117194507","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 : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840782
S. Arthi, T. Sreenivas
In the case of ensemble like distributed presentation, inter-aural cross-correlation (IACC) is correlated with the measure for source width extension. We question the validity of this measure and develop an angular measure for ensemble source width. In this work, we distinguish the binaural correlation functions of localized source, source with reverberation and ensemble source. We also develop a novel phase-only spatial transform to localize as many sources as possible. The angular separation between the spatial extrema sources can be a physical measure for ensemble source width. We also observe that phase-only HRIR cross-correlation functions act as listener dependent functional bases for localizing multiple sources. We observe these functional bases are wavelet-like and their signature are listener dependent and direction dependent. We extend the spatial transform to time-varying short-time spatial transform and define “Spatio-gram” to understand the effect of time-varying nature of the signal.
{"title":"Binaural Spatial Transform for Multi-source Localization determining Angular Extent of Ensemble Source Width","authors":"S. Arthi, T. Sreenivas","doi":"10.1109/SPCOM55316.2022.9840782","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840782","url":null,"abstract":"In the case of ensemble like distributed presentation, inter-aural cross-correlation (IACC) is correlated with the measure for source width extension. We question the validity of this measure and develop an angular measure for ensemble source width. In this work, we distinguish the binaural correlation functions of localized source, source with reverberation and ensemble source. We also develop a novel phase-only spatial transform to localize as many sources as possible. The angular separation between the spatial extrema sources can be a physical measure for ensemble source width. We also observe that phase-only HRIR cross-correlation functions act as listener dependent functional bases for localizing multiple sources. We observe these functional bases are wavelet-like and their signature are listener dependent and direction dependent. We extend the spatial transform to time-varying short-time spatial transform and define “Spatio-gram” to understand the effect of time-varying nature of the signal.","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123567931","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 : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840772
S. Maity, K. Sinha, B. Sinha, Reema Kumari
This paper proposes a reinforcement learning (RL) based Q-learning to address the issues of joint spectrum prediction (SP) and device-to-device (D2D) data communication in cognitive radio (CR) framework. An optimization problem is formulated that addresses energy efficiency (EE) maximization of D2D communications under the constraints of its total transmit power and a certain data transmission rate while meeting an interference threshold and cooperation rate in primary user (PU) transmission. The high accuracy in SP offers reward as an improvement on EE while a compulsion of meeting an interference threshold and a penalty on PU data transmission are made based on the relative degree of wrong prediction. A large set of simulation results shows that the proposed method offers 30% gain in EE while 20% reduction in data collision with PU over the existing works.
本文提出了一种基于强化学习（RL）的 Q-learning，以解决认知无线电（CR）框架中联合频谱预测（SP）和设备到设备（D2D）数据通信的问题。本文提出了一个优化问题，即在满足干扰阈值和主用户（PU）传输合作率的同时，在总发射功率和一定数据传输速率的约束条件下实现 D2D 通信的能效（EE）最大化。SP 的高精确度可作为对 EE 的改进提供奖励，而满足干扰阈值的强制要求和对 PU 数据传输的惩罚则基于错误预测的相对程度。大量仿真结果表明，与现有方法相比，所提出的方法在 EE 方面提高了 30%，而在与 PU 的数据碰撞方面减少了 20%。
{"title":"Reinforcement Learning for Spectrum Prediction and EE Maximization in D2D Communication","authors":"S. Maity, K. Sinha, B. Sinha, Reema Kumari","doi":"10.1109/SPCOM55316.2022.9840772","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840772","url":null,"abstract":"This paper proposes a reinforcement learning (RL) based Q-learning to address the issues of joint spectrum prediction (SP) and device-to-device (D2D) data communication in cognitive radio (CR) framework. An optimization problem is formulated that addresses energy efficiency (EE) maximization of D2D communications under the constraints of its total transmit power and a certain data transmission rate while meeting an interference threshold and cooperation rate in primary user (PU) transmission. The high accuracy in SP offers reward as an improvement on EE while a compulsion of meeting an interference threshold and a penalty on PU data transmission are made based on the relative degree of wrong prediction. A large set of simulation results shows that the proposed method offers 30% gain in EE while 20% reduction in data collision with PU over the existing works.","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126784659","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 : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840845
Roopesh Ramesh, Sanjeev Gurugopinath
We present a non-orthogonal multiple access (NOMA)-based coooperative scheme for relay-assisted power line communication (PLC) systems. The network consists of a source (S) modem, a decode-and-forward relay (R) modem and a destination (D) modem. In the first time slot, S communicates to both R and D using NOMA, while S and R simultaneously communicate with D in the second time slot. We derive closed form expressions for the approximate average sum rate and overall outage probability of the network. Through Monte Carlo simulations and numerical techniques we show that the approximations used in our analysis are tight. Furthermore, we show that our scheme outperforms a single-stage cooperative relaying NOMA scheme for PLC proposed in the earlier literature, in terms of outage probability and sum rate.
{"title":"Sum Rate and Outage Performance of Relay-Aided NOMA Over Power Line Communication","authors":"Roopesh Ramesh, Sanjeev Gurugopinath","doi":"10.1109/SPCOM55316.2022.9840845","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840845","url":null,"abstract":"We present a non-orthogonal multiple access (NOMA)-based coooperative scheme for relay-assisted power line communication (PLC) systems. The network consists of a source (S) modem, a decode-and-forward relay (R) modem and a destination (D) modem. In the first time slot, S communicates to both R and D using NOMA, while S and R simultaneously communicate with D in the second time slot. We derive closed form expressions for the approximate average sum rate and overall outage probability of the network. Through Monte Carlo simulations and numerical techniques we show that the approximations used in our analysis are tight. Furthermore, we show that our scheme outperforms a single-stage cooperative relaying NOMA scheme for PLC proposed in the earlier literature, in terms of outage probability and sum rate.","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134322762","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 : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840806
Natha Harika, T. Kumar
According to the World Happiness Report 2021, created by the World Happiness Council, it has ranked 140 countries based on how satisfied their citizens are. Due to COVID-19 pandemic, they observed there was large decline in mental health because of unemployment and decrease inperson gatherings resulting in a decrease in the happiness index of countries. Therefore, in this work we targeted to improve the emotional state of the person and make him happy, by recognizing the persons emotion and plays corresponding music will help user in changing their mood. Music's magical power has been scientifically established and people enjoy listening to music that reflects their emotional feelings, it is a stress-relieving tool and has the ability to control a wide range of psychological states. We used Viola Jones algorithm, Data augmentation and CoAtNet algorithm to detect the emotion of a person. A high accuracy is achieved with proposed CoAtNet model when compared to other methods like Conventional CNN, Principal Component Analysis (PCA) and SVM etc. We have also deployed the model on the STM32H747I Board.
{"title":"Real Time Smart Music Player Using Facial Expression","authors":"Natha Harika, T. Kumar","doi":"10.1109/SPCOM55316.2022.9840806","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840806","url":null,"abstract":"According to the World Happiness Report 2021, created by the World Happiness Council, it has ranked 140 countries based on how satisfied their citizens are. Due to COVID-19 pandemic, they observed there was large decline in mental health because of unemployment and decrease inperson gatherings resulting in a decrease in the happiness index of countries. Therefore, in this work we targeted to improve the emotional state of the person and make him happy, by recognizing the persons emotion and plays corresponding music will help user in changing their mood. Music's magical power has been scientifically established and people enjoy listening to music that reflects their emotional feelings, it is a stress-relieving tool and has the ability to control a wide range of psychological states. We used Viola Jones algorithm, Data augmentation and CoAtNet algorithm to detect the emotion of a person. A high accuracy is achieved with proposed CoAtNet model when compared to other methods like Conventional CNN, Principal Component Analysis (PCA) and SVM etc. We have also deployed the model on the STM32H747I Board.","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132499381","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 : 2022-07-11DOI: 10.1109/SPCOM55316.2022.9840783
Tamoghno Nath, K. Benerjee, Adrish Banerjee
In a Molecular-Communication-via-Diffusion (MCvD) channel, the molecules follow a simple Brownian motion that leads to an irregular arrival of the molecules at the receiver and introduces Inter-Symbol-Interference (ISI) in the channel. In this work, we have used different sequence distributions to analyze the effect of ISI in an MCvD channel. It has been shown that the ISI strictly depends on the location of bit-1s in the sequence, and accordingly, the expected ISI has been computed for all the proposed sequences based on the bit-1 positions in the sequence. We have also derived an upper bound on the expected ISI for the proposed sequences. We have shown that One-at-Starting-Position (OSP) sequence shows the best performance among all the proposed sequence distributions, with the expected ISI converging to a constant value. Simulation results also corroborate that the OSP sequence provides the lowest ISI in an MCvD channel compared to other codes studied in the literature.
{"title":"On Effect of Different Sequence Distributions on ISI in an MCvD System","authors":"Tamoghno Nath, K. Benerjee, Adrish Banerjee","doi":"10.1109/SPCOM55316.2022.9840783","DOIUrl":"https://doi.org/10.1109/SPCOM55316.2022.9840783","url":null,"abstract":"In a Molecular-Communication-via-Diffusion (MCvD) channel, the molecules follow a simple Brownian motion that leads to an irregular arrival of the molecules at the receiver and introduces Inter-Symbol-Interference (ISI) in the channel. In this work, we have used different sequence distributions to analyze the effect of ISI in an MCvD channel. It has been shown that the ISI strictly depends on the location of bit-1s in the sequence, and accordingly, the expected ISI has been computed for all the proposed sequences based on the bit-1 positions in the sequence. We have also derived an upper bound on the expected ISI for the proposed sequences. We have shown that One-at-Starting-Position (OSP) sequence shows the best performance among all the proposed sequence distributions, with the expected ISI converging to a constant value. Simulation results also corroborate that the OSP sequence provides the lowest ISI in an MCvD channel compared to other codes studied in the literature.","PeriodicalId":246982,"journal":{"name":"2022 IEEE International Conference on Signal Processing and Communications (SPCOM)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116032464","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：