Pub Date : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827689
K. Subramanian, R. Venkatesh Babu, S. Suresh
We propose to develop a 3-D optical flow features based human action recognition system. Optical flow based features are employed here since they can capture the apparent movement in object, by design. Moreover, they can represent information hierarchically from local pixel level to global object level. In this work, 3-D optical flow based features are extracted by combining the 2-D optical flow based features with the depth flow features o btained from d epth camera. In order to develop an action recognition system, we employ a Meta-Cognitive Neuro-Fuzzy Inference System (McFIS). The aim of McFIS is to find the decision boundary separating different classes based on their respective optical flow based features. McFIS consists of a neuro-fuzzy inference system (cognitive component) and a self-regulatory learning mechanism (meta-cognitive component). During the supervised learning, self-regulatory learning mechanism monitors the knowledge of the current sample with respect to the existing knowledge in the network and controls the learning by deciding on sample deletion, sample learning or sample reserve strategies. The performance of the proposed action recognition system was evaluated on a proprietary data set consisting of eight subjects. The performance evaluation with standard support vector machine classifier and extreme learning machine indicates improved performance of McFIS is recognizing actions based of 3-D optical flow based features.
{"title":"An optical flow feature and McFIS based approach for 3-dimensional human action recognition","authors":"K. Subramanian, R. Venkatesh Babu, S. Suresh","doi":"10.1109/ISSNIP.2014.6827689","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827689","url":null,"abstract":"We propose to develop a 3-D optical flow features based human action recognition system. Optical flow based features are employed here since they can capture the apparent movement in object, by design. Moreover, they can represent information hierarchically from local pixel level to global object level. In this work, 3-D optical flow based features are extracted by combining the 2-D optical flow based features with the depth flow features o btained from d epth camera. In order to develop an action recognition system, we employ a Meta-Cognitive Neuro-Fuzzy Inference System (McFIS). The aim of McFIS is to find the decision boundary separating different classes based on their respective optical flow based features. McFIS consists of a neuro-fuzzy inference system (cognitive component) and a self-regulatory learning mechanism (meta-cognitive component). During the supervised learning, self-regulatory learning mechanism monitors the knowledge of the current sample with respect to the existing knowledge in the network and controls the learning by deciding on sample deletion, sample learning or sample reserve strategies. The performance of the proposed action recognition system was evaluated on a proprietary data set consisting of eight subjects. The performance evaluation with standard support vector machine classifier and extreme learning machine indicates improved performance of McFIS is recognizing actions based of 3-D optical flow based features.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115814291","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 : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827695
S. Blayac, Arnaud Schreiner, Maxime Nouaille, Beatrice Dubois, F. Depoutot
A flexible touch panel touch panel based on the principle of interaction between a coplanar waveguide with the finger is presented. A concept of location based on harmonic analysis is detailed. The interaction is precisely modelled; it is shown that it is possible to have at the same time a precise location and measurement of touch force on a single port, independently. Measurements are carried out in order to assess the sensing concepts developed and evaluate the precision of location process. The combination of the printed sensor and sensing algorithm opens the way for potential very large flexible device, more than meter square, with ultra-simple connections. The combination of location and force sensing is an additional feature which has the potential to enrich strongly the degree of interaction in the case of application to human to machine interface.
{"title":"Single port, large area touch and force sensing: Towards low cost sensitive printed surfaces","authors":"S. Blayac, Arnaud Schreiner, Maxime Nouaille, Beatrice Dubois, F. Depoutot","doi":"10.1109/ISSNIP.2014.6827695","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827695","url":null,"abstract":"A flexible touch panel touch panel based on the principle of interaction between a coplanar waveguide with the finger is presented. A concept of location based on harmonic analysis is detailed. The interaction is precisely modelled; it is shown that it is possible to have at the same time a precise location and measurement of touch force on a single port, independently. Measurements are carried out in order to assess the sensing concepts developed and evaluate the precision of location process. The combination of the printed sensor and sensing algorithm opens the way for potential very large flexible device, more than meter square, with ultra-simple connections. The combination of location and force sensing is an additional feature which has the potential to enrich strongly the degree of interaction in the case of application to human to machine interface.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128937491","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 : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827674
César Ortega-Corral, Luis E. Palafox, J. A. García-Macías, J. Sánchez-García, Leocundo Aguilar, J. Nieto-Hipólito
In this paper we present a novel Wireless Sensor Network (WSN) Transmission Power Control (TPC) scheme based on prevailing weather conditions: temperature and relative humidity (TRH). After the analysis of several days worth of experimental data containing received signal strength indicator (RSSI) values, a strong correspondence became evident between signal loss and wireless channel weather conditions. This led us to characterize a TRH normalized gradient that describes a relative amount of power loss that is incurred when using the wireless channel. After review, we deployed this gradient with certain modifications as a gain coefficient in a novel way of compensating wireless transmission power. Meaning that a TRH TPC scheme is proposed, for more dependable wireless communications, which determines power gain using RSSI feedback. Performance metrics of a prototype TRH TPC algorithm, with differing parametric values, are presented and compared to another node with no TPC, transmitting at the same time with a fixed power level. Finally, after weighing in the full range of results, we establish a set of conditions where this TRH compensator is effective in improving the RSSI and the overall packet received rate (PRR), while operating in harsh daylight and humid conditions.
{"title":"Transmission Power Control based on temperature and relative humidity","authors":"César Ortega-Corral, Luis E. Palafox, J. A. García-Macías, J. Sánchez-García, Leocundo Aguilar, J. Nieto-Hipólito","doi":"10.1109/ISSNIP.2014.6827674","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827674","url":null,"abstract":"In this paper we present a novel Wireless Sensor Network (WSN) Transmission Power Control (TPC) scheme based on prevailing weather conditions: temperature and relative humidity (TRH). After the analysis of several days worth of experimental data containing received signal strength indicator (RSSI) values, a strong correspondence became evident between signal loss and wireless channel weather conditions. This led us to characterize a TRH normalized gradient that describes a relative amount of power loss that is incurred when using the wireless channel. After review, we deployed this gradient with certain modifications as a gain coefficient in a novel way of compensating wireless transmission power. Meaning that a TRH TPC scheme is proposed, for more dependable wireless communications, which determines power gain using RSSI feedback. Performance metrics of a prototype TRH TPC algorithm, with differing parametric values, are presented and compared to another node with no TPC, transmitting at the same time with a fixed power level. Finally, after weighing in the full range of results, we establish a set of conditions where this TRH compensator is effective in improving the RSSI and the overall packet received rate (PRR), while operating in harsh daylight and humid conditions.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114688741","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 : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827646
Dong Chen, Jason Brown, J. Khan
The Smart Grid communication networks need to serve a large number of devices distributed over a wide geographical area. The Smart Grid devices will generate different classes of traffic with varying Quality of Service (QoS) requirements. One of the key problems in a wide area network is the cost-effective connectivity to all devices which may be operating in an energy-constrained environment. In this paper, we propose a cluster-tree based 6LoWPAN (IPv6 over Low power Wireless Personal Area Networks) network for the Neighborhood Area Network (NAN) applications. To maximize the throughput and to minimize the packet latency in the NAN, we propose a staggered link design approach to support a beacon based 6LoWPAN distributed network architecture. An OPNET simulation model has been developed to analyze the performance of a NAN for smart meter and demand management data communications. Initial simulation results show that this staggered link design approach combined with a packet aggregation technique can significantly enhance the performance of a NAN.
智能电网通信网络需要为分布在广泛地理区域的大量设备提供服务。智能电网设备将产生具有不同服务质量(QoS)要求的不同类别的流量。广域网的关键问题之一是经济有效地连接所有可能在能源受限环境中运行的设备。在本文中，我们提出了一个基于簇树的6LoWPAN (IPv6 over Low power Wireless Personal Area Networks)网络，用于邻域网络(NAN)的应用。为了最大限度地提高吞吐量和最小化NAN中的数据包延迟，我们提出了一种交错链路设计方法来支持基于信标的6LoWPAN分布式网络架构。建立了一个OPNET仿真模型，分析了智能电表和需求管理数据通信中NAN的性能。初步仿真结果表明，这种交错链路设计方法与分组聚合技术相结合，可以显著提高NAN的性能。
{"title":"Performance analysis of a distributed 6LoWPAN network for the Smart Grid applications","authors":"Dong Chen, Jason Brown, J. Khan","doi":"10.1109/ISSNIP.2014.6827646","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827646","url":null,"abstract":"The Smart Grid communication networks need to serve a large number of devices distributed over a wide geographical area. The Smart Grid devices will generate different classes of traffic with varying Quality of Service (QoS) requirements. One of the key problems in a wide area network is the cost-effective connectivity to all devices which may be operating in an energy-constrained environment. In this paper, we propose a cluster-tree based 6LoWPAN (IPv6 over Low power Wireless Personal Area Networks) network for the Neighborhood Area Network (NAN) applications. To maximize the throughput and to minimize the packet latency in the NAN, we propose a staggered link design approach to support a beacon based 6LoWPAN distributed network architecture. An OPNET simulation model has been developed to analyze the performance of a NAN for smart meter and demand management data communications. Initial simulation results show that this staggered link design approach combined with a packet aggregation technique can significantly enhance the performance of a NAN.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129530168","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 : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827644
Rotthaporn Palasuek, S. Kladsomboon, T. Thepudom, T. Kerdcharoen
In this paper, we have developed an optoelectronic nose for rice classification based on measurement of changes in the optical absorption of organic thin films upon interactions with volatile analytes. The organic thin films based on metalloporphyrin (ZnTPP) and metallophthalocyanine (ZnOBPc) for gas sensing application were fabricated by spin coating technique. The optical properties of these thin films were tested with UV-VIS spectrophotometry. The surface topology of the films was observed by atomic force microscopy (AFM) technique which reveals that the film's topology is homogeneous with nano-scale thickness. The sensing behavior of the films toward various volatile compounds representing the simulated aroma rice, such as 5% v/v hexanal, 5% v/v octanal and deionized water, was investigated under N2 atmosphere at room temperature. The results have shown that ZnTPP and ZnOBPc thin film gas sensors can discriminate such volatile compounds. The optoelectronic nose was then applied to examine the odors of Jasmine rice, normal white rice and various of their mixtures. It was shown that ZnOBPc thin film can clearly classify the type of rice better than ZnTPP thin film. These results were confirmed by computational quantum mechanics based on density functional theory, which reveals that ZnOBPc molecule binds to the simulated rice aroma molecules better than ZnTPP molecule.
{"title":"Optical electronic nose based on porphyrin and phthalocyanine thin films for rice flavour classification","authors":"Rotthaporn Palasuek, S. Kladsomboon, T. Thepudom, T. Kerdcharoen","doi":"10.1109/ISSNIP.2014.6827644","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827644","url":null,"abstract":"In this paper, we have developed an optoelectronic nose for rice classification based on measurement of changes in the optical absorption of organic thin films upon interactions with volatile analytes. The organic thin films based on metalloporphyrin (ZnTPP) and metallophthalocyanine (ZnOBPc) for gas sensing application were fabricated by spin coating technique. The optical properties of these thin films were tested with UV-VIS spectrophotometry. The surface topology of the films was observed by atomic force microscopy (AFM) technique which reveals that the film's topology is homogeneous with nano-scale thickness. The sensing behavior of the films toward various volatile compounds representing the simulated aroma rice, such as 5% v/v hexanal, 5% v/v octanal and deionized water, was investigated under N2 atmosphere at room temperature. The results have shown that ZnTPP and ZnOBPc thin film gas sensors can discriminate such volatile compounds. The optoelectronic nose was then applied to examine the odors of Jasmine rice, normal white rice and various of their mixtures. It was shown that ZnOBPc thin film can clearly classify the type of rice better than ZnTPP thin film. These results were confirmed by computational quantum mechanics based on density functional theory, which reveals that ZnOBPc molecule binds to the simulated rice aroma molecules better than ZnTPP molecule.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127206428","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 : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827634
T. Seesaard, Sasiprapa Seaon, P. Lorwongtragool, T. Kerdcharoen
In this paper, we have developed a wearable electronic nose (e-nose) embedded on cloth for detecting the body odor of wearer, aiming to be a self-monitoring healthcare consumer electronics. This "smelling shirt" consists of eight on-fabric chemical gas sensors based on polymer/functionalized-SWCNTs simply fabricated by embroidering and drop coating process. Sew-able electronic modules based on LilyPad Arduino together with wireless ZigBee module were used for data acquisition and communication. The performance of the sensors has been tested by a selected set of volatile organic compounds as presented in the body odor, such as dodecane, 2-ethylhexanol, butyric acid, 6-Methyl-5-heptenone, nonanal and water. Discrimination of individual health status and analysis of the odor-print (or smell-print) of specific persons were confirmed by principal component analysis (PCA), which was found to be able to track change in the human body odor, thereby showing its potential to be applied for real-time point-of-care health monitoring.
{"title":"On-cloth wearable E-nose for monitoring and discrimination of body odor signature","authors":"T. Seesaard, Sasiprapa Seaon, P. Lorwongtragool, T. Kerdcharoen","doi":"10.1109/ISSNIP.2014.6827634","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827634","url":null,"abstract":"In this paper, we have developed a wearable electronic nose (e-nose) embedded on cloth for detecting the body odor of wearer, aiming to be a self-monitoring healthcare consumer electronics. This \"smelling shirt\" consists of eight on-fabric chemical gas sensors based on polymer/functionalized-SWCNTs simply fabricated by embroidering and drop coating process. Sew-able electronic modules based on LilyPad Arduino together with wireless ZigBee module were used for data acquisition and communication. The performance of the sensors has been tested by a selected set of volatile organic compounds as presented in the body odor, such as dodecane, 2-ethylhexanol, butyric acid, 6-Methyl-5-heptenone, nonanal and water. Discrimination of individual health status and analysis of the odor-print (or smell-print) of specific persons were confirmed by principal component analysis (PCA), which was found to be able to track change in the human body odor, thereby showing its potential to be applied for real-time point-of-care health monitoring.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132061815","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 : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827610
M. Tanweer, S. Sundaram
This paper presents a human cognition inspired particle swarm optimization algorithm, and is referred as Cognition Inspired Particle Swarm Optimization (CIPSO). As suggested by the human learning psychology, the particles control the cognition based on their global performance and also the social cognition does not influence one-self directly based on his current knowledge. Hence, in the proposed CIPSO, the particle with global best explores more by only using cognitive component with increasing inertia and self-cognition, where as other particles use explore and exploit using self with entire dimension selection and random social cognition with randomly selected dimensions for updating velocities. The performance of the proposed CIPSO is evaluated using 10 benchmark test functions as suggested in CEC2005 [3]. The performance is also compared with different variants of PSO algorithms reported in the literature. The results clearly indicate that human cognition inspired PSO performs better for most functions than other PSO algorithms reported in the literature.
{"title":"Human cognition inspired particle swarm optimization algorithm","authors":"M. Tanweer, S. Sundaram","doi":"10.1109/ISSNIP.2014.6827610","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827610","url":null,"abstract":"This paper presents a human cognition inspired particle swarm optimization algorithm, and is referred as Cognition Inspired Particle Swarm Optimization (CIPSO). As suggested by the human learning psychology, the particles control the cognition based on their global performance and also the social cognition does not influence one-self directly based on his current knowledge. Hence, in the proposed CIPSO, the particle with global best explores more by only using cognitive component with increasing inertia and self-cognition, where as other particles use explore and exploit using self with entire dimension selection and random social cognition with randomly selected dimensions for updating velocities. The performance of the proposed CIPSO is evaluated using 10 benchmark test functions as suggested in CEC2005 [3]. The performance is also compared with different variants of PSO algorithms reported in the literature. The results clearly indicate that human cognition inspired PSO performs better for most functions than other PSO algorithms reported in the literature.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132166263","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 : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827678
Dinesh Thangavel, Xiaoping Ma, A. Valera, H. Tan, C. Tan
Wireless sensor networks (WSNs) typically consist of sensor nodes and gateways that operate on devices with limited resources. As a result, WSNs require bandwidth-efficient and energy-efficient application protocols for data transmission. Message Queue Telemetry Transport (MQTT) and Constrained Application Protocol (CoAP) are two such protocols proposed for resource-constrained devices. In this paper, we design and implement a common middleware that supports MQTT and CoAP and provides a common programming interface. We design the middleware to be extensible to support future protocols. Using the common middleware, we conducted experiments to study the performance of MQTT and CoAP in terms of end-to-end delay and bandwidth consumption. Experimental results reveal that MQTT messages have lower delay than CoAP messages at lower packet loss rates and higher delay than CoAP messages at higher loss rates. Moreover, when the message size is small and the loss rate is equal to or less than 25%, CoAP generates lower additional traffic than MQTT to ensure message reliability.
{"title":"Performance evaluation of MQTT and CoAP via a common middleware","authors":"Dinesh Thangavel, Xiaoping Ma, A. Valera, H. Tan, C. Tan","doi":"10.1109/ISSNIP.2014.6827678","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827678","url":null,"abstract":"Wireless sensor networks (WSNs) typically consist of sensor nodes and gateways that operate on devices with limited resources. As a result, WSNs require bandwidth-efficient and energy-efficient application protocols for data transmission. Message Queue Telemetry Transport (MQTT) and Constrained Application Protocol (CoAP) are two such protocols proposed for resource-constrained devices. In this paper, we design and implement a common middleware that supports MQTT and CoAP and provides a common programming interface. We design the middleware to be extensible to support future protocols. Using the common middleware, we conducted experiments to study the performance of MQTT and CoAP in terms of end-to-end delay and bandwidth consumption. Experimental results reveal that MQTT messages have lower delay than CoAP messages at lower packet loss rates and higher delay than CoAP messages at higher loss rates. Moreover, when the message size is small and the loss rate is equal to or less than 25%, CoAP generates lower additional traffic than MQTT to ensure message reliability.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130970808","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 : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827701
Xiangfa Guo, M. Chan
Limited contact bandwidth is one of the main resource constraints in delay/disruption tolerant network (DTN). In this paper, we investigate the use of overhearing to relieve the bandwidth constraint. We ask two questions in this work. First, how many receivers are within the transmission range when a sender transmits? Second, how does the larger bandwidth achieved through overhearing improve routing performance? For the first question, we analyze available mobility traces and empirically quantify the amount of potential transmission gains through overhearing. We find that the increase can be significant. For one of the traces, the gain is 10 times that of unicast. For the second question, we experiment with different approaches to enhance existing DTN routing to exploit overhearing. We find that the improvement in delivery ratio can be up to 50%.
{"title":"On the utility of overhearing in DTN","authors":"Xiangfa Guo, M. Chan","doi":"10.1109/ISSNIP.2014.6827701","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827701","url":null,"abstract":"Limited contact bandwidth is one of the main resource constraints in delay/disruption tolerant network (DTN). In this paper, we investigate the use of overhearing to relieve the bandwidth constraint. We ask two questions in this work. First, how many receivers are within the transmission range when a sender transmits? Second, how does the larger bandwidth achieved through overhearing improve routing performance? For the first question, we analyze available mobility traces and empirically quantify the amount of potential transmission gains through overhearing. We find that the increase can be significant. For one of the traces, the gain is 10 times that of unicast. For the second question, we experiment with different approaches to enhance existing DTN routing to exploit overhearing. We find that the improvement in delivery ratio can be up to 50%.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126516573","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 : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827690
K. Subramanian, R. Venkatesh Babu, Savitha Ramasamy
Facial emotions are the most expressive way to display emotions. Many algorithms have been proposed which employ a particular set of people (usually a database) to both train and test their model. This paper focuses on the challenging task of database independent emotion recognition, which is a generalized case of subject-independent emotion recognition. The emotion recognition system employed in this work is a Meta-Cognitive Neuro-Fuzzy Inference System (McFIS). McFIS has two components, a neuro-fuzzy inference system, which is the cognitive component and a self-regulatory learning mechanism, which is the meta-cognitive component. The meta-cognitive component, monitors the knowledge in the neuro-fuzzy inference system and decides on what-to-learn, when-to-learn and how-to-learn the training samples, efficiently. For each sample, the McFIS decides whether to delete the sample without being learnt, use it to add/ prune or update the network parameter or reserve it for future use. This helps the network avoid over-training and as a result improve its generalization performance over untrained databases. In this study, we extract pixel based emotion features from well-known (Japanese Female Facial Expression) JAFFE and (Taiwanese Female Expression Image) TFEID database. Two sets of experiment are conducted. First, we study the individual performance of both databases on McFIS based on 5-fold cross validation study. Next, in order to study the generalization performance, McFIS trained on JAFFE database is tested on TFEID and vice-versa. The performance The performance comparison in both experiments against SVM classifier gives promising results.
{"title":"Database independent human emotion recognition with Meta-Cognitive Neuro-Fuzzy Inference System","authors":"K. Subramanian, R. Venkatesh Babu, Savitha Ramasamy","doi":"10.1109/ISSNIP.2014.6827690","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827690","url":null,"abstract":"Facial emotions are the most expressive way to display emotions. Many algorithms have been proposed which employ a particular set of people (usually a database) to both train and test their model. This paper focuses on the challenging task of database independent emotion recognition, which is a generalized case of subject-independent emotion recognition. The emotion recognition system employed in this work is a Meta-Cognitive Neuro-Fuzzy Inference System (McFIS). McFIS has two components, a neuro-fuzzy inference system, which is the cognitive component and a self-regulatory learning mechanism, which is the meta-cognitive component. The meta-cognitive component, monitors the knowledge in the neuro-fuzzy inference system and decides on what-to-learn, when-to-learn and how-to-learn the training samples, efficiently. For each sample, the McFIS decides whether to delete the sample without being learnt, use it to add/ prune or update the network parameter or reserve it for future use. This helps the network avoid over-training and as a result improve its generalization performance over untrained databases. In this study, we extract pixel based emotion features from well-known (Japanese Female Facial Expression) JAFFE and (Taiwanese Female Expression Image) TFEID database. Two sets of experiment are conducted. First, we study the individual performance of both databases on McFIS based on 5-fold cross validation study. Next, in order to study the generalization performance, McFIS trained on JAFFE database is tested on TFEID and vice-versa. The performance The performance comparison in both experiments against SVM classifier gives promising results.","PeriodicalId":269784,"journal":{"name":"2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123134233","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：