Pub Date : 2014-04-21DOI: 10.1109/ISSNIP.2014.6827661
Matti Mononen, Jukka Saarenpaa, Markus Johansson, Harri Niska
The building sector is a major energy consumer and CO2 emitter, being responsible for approximately 40% of the total consumption in the EU. Active demand side participation of electricity customers is seen as crucial in the management and reduction of the building sector's CO2 emissions. However, today's electricity markets are often lacking strong incentives for active demand side participation. Understandable customer specific comparison information and easy-to-use energy displays can be used to influence customer behaviour and encourage customer participation. This paper presents a data-driven method for producing household level comparison information, based on hourly interval smart meter data and additional household information. Firstly, the customers are segmented by the heating system and the type of housing, followed by weighted clustering that is used to refine the comparison group. In the weighted clustering, normalized load profiles together with properties of the dwelling and the residents are considered, and weights are assigned to the properties according to how much they contribute to the electricity consumption. In this paper, the initial experimental results are presented and discussed, and future development ideas are laid out. The method is under development and testing as a part of the Finnish SGEM-project.
{"title":"Data-driven method for providing feedback to households on electricity consumption","authors":"Matti Mononen, Jukka Saarenpaa, Markus Johansson, Harri Niska","doi":"10.1109/ISSNIP.2014.6827661","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827661","url":null,"abstract":"The building sector is a major energy consumer and CO2 emitter, being responsible for approximately 40% of the total consumption in the EU. Active demand side participation of electricity customers is seen as crucial in the management and reduction of the building sector's CO2 emissions. However, today's electricity markets are often lacking strong incentives for active demand side participation. Understandable customer specific comparison information and easy-to-use energy displays can be used to influence customer behaviour and encourage customer participation. This paper presents a data-driven method for producing household level comparison information, based on hourly interval smart meter data and additional household information. Firstly, the customers are segmented by the heating system and the type of housing, followed by weighted clustering that is used to refine the comparison group. In the weighted clustering, normalized load profiles together with properties of the dwelling and the residents are considered, and weights are assigned to the properties according to how much they contribute to the electricity consumption. In this paper, the initial experimental results are presented and discussed, and future development ideas are laid out. The method is under development and testing as a part of the Finnish SGEM-project.","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":"132738806","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.2014.6827666
Yoshimasa Ishi, Tomoya Kawakami, T. Yoshihisa, Y. Teranishi
Due to the increasing use of sensors, such as security cameras and environmental sensors, sensor data stream delivery, the delivering of sensor data through cyclic collection, is attracting considerable attention. Various methods for distributing communication loads, when delivering the same sensor data streams to multiple clients, have been investigated. Our research team developed a peer-to-peer streaming system for distributing communication loads when delivering sensor data streams with different data collection cycles. In this study, we performed a comparative system evaluation utilizing the JGN-X PIAX testbed provided by the NICT.
{"title":"A P2P streaming system for delivering sensor data streams with different collection cycles","authors":"Yoshimasa Ishi, Tomoya Kawakami, T. Yoshihisa, Y. Teranishi","doi":"10.1109/ISSNIP.2014.6827666","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827666","url":null,"abstract":"Due to the increasing use of sensors, such as security cameras and environmental sensors, sensor data stream delivery, the delivering of sensor data through cyclic collection, is attracting considerable attention. Various methods for distributing communication loads, when delivering the same sensor data streams to multiple clients, have been investigated. Our research team developed a peer-to-peer streaming system for distributing communication loads when delivering sensor data streams with different data collection cycles. In this study, we performed a comparative system evaluation utilizing the JGN-X PIAX testbed provided by the NICT.","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":"121830374","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.2014.6827617
Ke Hu, Yan Wang, Ashfaqur Rahman, V. Sivaraman
In recent years several research groups, including ours, have demonstrated participatory systems that use wearable or vehicle-mounted portable units coupled with smartphones to crowdsource urban air pollution data from lay users. These systems have shown remarkable improvement in spatial granularity over government-operated monitoring systems, leading to better mapping and understanding of urban air pollution, at relatively low cost. In this paper we extend the paradigm to personalize the consumption of data by individuals. Specifically, we combine the pollution concentrations obtained from participatory systems with the individual's on-body activity monitors to estimate the personal inhalation dosage of air pollution. We show that the individual's activity, such as jogging, cycling, or driving, impacts their dosage, and develop an app that gives them this personalised information. Our system is a step towards enabling medical inferencing of the impact of air pollution on individual health.
{"title":"Personalising pollution exposure estimates using wearable activity sensors","authors":"Ke Hu, Yan Wang, Ashfaqur Rahman, V. Sivaraman","doi":"10.1109/ISSNIP.2014.6827617","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827617","url":null,"abstract":"In recent years several research groups, including ours, have demonstrated participatory systems that use wearable or vehicle-mounted portable units coupled with smartphones to crowdsource urban air pollution data from lay users. These systems have shown remarkable improvement in spatial granularity over government-operated monitoring systems, leading to better mapping and understanding of urban air pollution, at relatively low cost. In this paper we extend the paradigm to personalize the consumption of data by individuals. Specifically, we combine the pollution concentrations obtained from participatory systems with the individual's on-body activity monitors to estimate the personal inhalation dosage of air pollution. We show that the individual's activity, such as jogging, cycling, or driving, impacts their dosage, and develop an app that gives them this personalised information. Our system is a step towards enabling medical inferencing of the impact of air pollution on individual health.","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":"122173393","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.2014.6827605
Ido Nevat, Ory Eger, G. Peters, F. Septier
We present a new architecture to perform localization (position estimation) in GNSS systems, termed NEPS (Narrowband Efficient Positioning System). The NEPS architecture is composed of three components: a low powered cheap receiver; a communication system which transmits the measurements; and a processing unit which receives the distorted observations (due to quantisation and imperfect transmission medium) and performs the position estimation algorithm. The NEPS is a stand-alone system which is designed to incorporate the quantised measurements as well as the imperfect communication channels between receiver and the backend in order to perform inference on the user's position. Compared with a conventional system, the NEPS consumes less bandwidth, requires lower power consumption and provides faster reporting rates. We derive the joint Maximum Likelihood (ML) for the position and the receiver's clock offset. We then develop an efficient algorithm to solve the resulting non-convex oinferenceptimisation problem. Furthermore, we derive a theoretical performance lower bound on the achievable accuracy via Cramér-Rao lower bound (CRLB). Simulation results show that the performance of the NEPS ML position estimator is close to the theoretical performance bound.
{"title":"NEPS: “Narrowband Efficient Positioning System” for delivering resource efficient GNSS receivers","authors":"Ido Nevat, Ory Eger, G. Peters, F. Septier","doi":"10.1109/ISSNIP.2014.6827605","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827605","url":null,"abstract":"We present a new architecture to perform localization (position estimation) in GNSS systems, termed NEPS (Narrowband Efficient Positioning System). The NEPS architecture is composed of three components: a low powered cheap receiver; a communication system which transmits the measurements; and a processing unit which receives the distorted observations (due to quantisation and imperfect transmission medium) and performs the position estimation algorithm. The NEPS is a stand-alone system which is designed to incorporate the quantised measurements as well as the imperfect communication channels between receiver and the backend in order to perform inference on the user's position. Compared with a conventional system, the NEPS consumes less bandwidth, requires lower power consumption and provides faster reporting rates. We derive the joint Maximum Likelihood (ML) for the position and the receiver's clock offset. We then develop an efficient algorithm to solve the resulting non-convex oinferenceptimisation problem. Furthermore, we derive a theoretical performance lower bound on the achievable accuracy via Cramér-Rao lower bound (CRLB). Simulation results show that the performance of the NEPS ML position estimator is close to the theoretical performance bound.","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":"121697622","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.2014.6827692
Jinseok Yang, T. Simunic, S. Tilak
Today's platforms for long-term environmental monitoring (e.g. buoys or towers) typically host large solar panels and batteries. Ideally, miniaturized platforms could be used instead, so state of the art power management technique that takes into account battery levels and harvested energy to provide uniform sampling rate. However, the fixed pre-defined intervals is not desirable. The state-of-art adaptive sampling mechanism, optimal adaptive sampling algorithm (OSA) uses data uncertainty and past measurements to determine the optimal sampling rate at the cost of high computational complexity O(n3), thus draining the batteries even further. Even if the sampling were done optimally, there are still significant challenges with data transmission. The state of the art approach for determining optimal transmission policy offers limited control over the energy-delay tradeoff and is not suitable to support wide range of applications ranging from real-time and delay-tolerant. To address these challenges, we have developed a novel power management framework that adapts sampling and transmission rates based on battery level, energy harvesting level and application-context (e.g. characteristics of the gathered data). Our framework is optimal in terms of energy efficiency with low computational complexity. We evaluate the performance of the proposed framework using datasets from two real-world deployments. Our results show that our approach saves significant amounts of energy (between 20% to 60%) by avoiding oversampling when the application does not need it and uses this saved energy to support sampling at high rates to capture event with necessary fidelity when needed.
{"title":"Leveraging application context for efficient sensing","authors":"Jinseok Yang, T. Simunic, S. Tilak","doi":"10.1109/ISSNIP.2014.6827692","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827692","url":null,"abstract":"Today's platforms for long-term environmental monitoring (e.g. buoys or towers) typically host large solar panels and batteries. Ideally, miniaturized platforms could be used instead, so state of the art power management technique that takes into account battery levels and harvested energy to provide uniform sampling rate. However, the fixed pre-defined intervals is not desirable. The state-of-art adaptive sampling mechanism, optimal adaptive sampling algorithm (OSA) uses data uncertainty and past measurements to determine the optimal sampling rate at the cost of high computational complexity O(n3), thus draining the batteries even further. Even if the sampling were done optimally, there are still significant challenges with data transmission. The state of the art approach for determining optimal transmission policy offers limited control over the energy-delay tradeoff and is not suitable to support wide range of applications ranging from real-time and delay-tolerant. To address these challenges, we have developed a novel power management framework that adapts sampling and transmission rates based on battery level, energy harvesting level and application-context (e.g. characteristics of the gathered data). Our framework is optimal in terms of energy efficiency with low computational complexity. We evaluate the performance of the proposed framework using datasets from two real-world deployments. Our results show that our approach saves significant amounts of energy (between 20% to 60%) by avoiding oversampling when the application does not need it and uses this saved energy to support sampling at high rates to capture event with necessary fidelity when needed.","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":"122403159","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.2014.6827630
Filip Barac, M. Gidlund, Tingting Zhang
Three major obstacles to wireless communication are electromagnetic interference, multipath fading and signal attenuation. The former stems mainly from collocated wireless systems operating in the same frequency band, while the latter two originate from physical properties of the environment. Identifying the source of packet corruption and loss is crucial, since the adequate countermeasures for different types of threats are essentially different. This problem is especially pronounced in industrial monitoring and control applications, where IEEE 802.15.4 communication is expected to deliver data within tight deadlines, with minimal packet loss. This work presents the Lightweight Packet Error Discriminator (LPED) that distinguishes between errors caused by multipath fading and attenuation, and those inflicted by IEEE 802.11 interference. LPED uses Forward Error Correction to determine the symbol error positions inside erroneously received packets and calculates the error density, which is then fed to a discriminator for error source classification. The statistical constituents of LPED are obtained from an extensive measurement campaign in two different types of industrial environments. The classifier incurs no overhead and in ≥90% of cases a single packet is sufficient for a correct channel diagnosis. Experiments show that LPED accelerates link diagnostics by at least 190%, compared to the relevant state-of-the-art approaches.
{"title":"LPED: Channel diagnostics in WSN through channel coding and symbol error statistics","authors":"Filip Barac, M. Gidlund, Tingting Zhang","doi":"10.1109/ISSNIP.2014.6827630","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827630","url":null,"abstract":"Three major obstacles to wireless communication are electromagnetic interference, multipath fading and signal attenuation. The former stems mainly from collocated wireless systems operating in the same frequency band, while the latter two originate from physical properties of the environment. Identifying the source of packet corruption and loss is crucial, since the adequate countermeasures for different types of threats are essentially different. This problem is especially pronounced in industrial monitoring and control applications, where IEEE 802.15.4 communication is expected to deliver data within tight deadlines, with minimal packet loss. This work presents the Lightweight Packet Error Discriminator (LPED) that distinguishes between errors caused by multipath fading and attenuation, and those inflicted by IEEE 802.11 interference. LPED uses Forward Error Correction to determine the symbol error positions inside erroneously received packets and calculates the error density, which is then fed to a discriminator for error source classification. The statistical constituents of LPED are obtained from an extensive measurement campaign in two different types of industrial environments. The classifier incurs no overhead and in ≥90% of cases a single packet is sufficient for a correct channel diagnosis. Experiments show that LPED accelerates link diagnostics by at least 190%, compared to the relevant state-of-the-art approaches.","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":"124749102","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.2014.6827683
V. Balasubramanian
In recent years, the drive for the Healthcare Monitoring Application (HMA) aims to provide continuous remote monitoring of a patient's health. For this to happen, the sensors in the monitoring component of the Body Area Wireless Sensor Networks (BAWSN) need to continuously send data to a Healthcare Application. We show that to provide continuous health data, the BAWSN depends on the collective data delivered by all the sensor nodes and not on a single sensor because medical diagnosis is rarely performed from a single data point. In addition, the arrival time of data should occur within the expected time to be indicative of the actual health of the patient. In this paper, we characterize the HMA as a time-critical application because the BAWSN has stringent timing requirements concerning the arrival of data from the sensor nodes within the defined critical time. Thereby, we formulate the critical time parameters to evaluate the BAWSN operations.
{"title":"Critical time parameters for evaluation of body area Wireless Sensor Networks in a Healthcare Monitoring Application","authors":"V. Balasubramanian","doi":"10.1109/ISSNIP.2014.6827683","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827683","url":null,"abstract":"In recent years, the drive for the Healthcare Monitoring Application (HMA) aims to provide continuous remote monitoring of a patient's health. For this to happen, the sensors in the monitoring component of the Body Area Wireless Sensor Networks (BAWSN) need to continuously send data to a Healthcare Application. We show that to provide continuous health data, the BAWSN depends on the collective data delivered by all the sensor nodes and not on a single sensor because medical diagnosis is rarely performed from a single data point. In addition, the arrival time of data should occur within the expected time to be indicative of the actual health of the patient. In this paper, we characterize the HMA as a time-critical application because the BAWSN has stringent timing requirements concerning the arrival of data from the sensor nodes within the defined critical time. Thereby, we formulate the critical time parameters to evaluate the BAWSN operations.","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":"130423200","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.2014.6827693
Håkan Jonsson, P. Nugues
We present a comparative exploratory analysis of two proximity networks of mobile phone users, the Proximates network and the Reality Mining network. Data for both networks were collected from mobile phones carried by two groups of users. Periodic Bluetooth scans were performed to detect the proximity of other mobile phones. The Reality Mining project took place in 2004-2005 at MIT, while Proximates took place in Sweden in 2012-2013. We show that the differences in sampling strategy between the two networks has effects on both static and dynamic metrics. We also find that fundamental metrics of the static Proximates network capture social interactions characteristics better than in the static Reality Mining network.
{"title":"A comparison of two proximity networks","authors":"Håkan Jonsson, P. Nugues","doi":"10.1109/ISSNIP.2014.6827693","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827693","url":null,"abstract":"We present a comparative exploratory analysis of two proximity networks of mobile phone users, the Proximates network and the Reality Mining network. Data for both networks were collected from mobile phones carried by two groups of users. Periodic Bluetooth scans were performed to detect the proximity of other mobile phones. The Reality Mining project took place in 2004-2005 at MIT, while Proximates took place in Sweden in 2012-2013. We show that the differences in sampling strategy between the two networks has effects on both static and dynamic metrics. We also find that fundamental metrics of the static Proximates network capture social interactions characteristics better than in the static Reality Mining network.","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":"129579373","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.2014.6827602
M. Ruhnow, Julia Kohser, T. Bley, E. Boschke, M. Bulst, S. Wegner
Biofilm formation can cause serious health hazards, mostly due to the uncontrolled release of pathogens. This can generate several problems in industrial facilities, e.g., in the food industry. The aim of the present study was to develop and implement a multi-parametric sensor system to monitor biofilm formation in laboratory as well as industrial set-ups. To minimize cross sensitivity or interference, the device was based on a combination of different measurement principles. Micro-organisms were initially cultivated in a laboratory scale reactor. Afterwards, biofilm formation will be studied with each prototype of the multi-parametric sensor followed by final tests on an industrial scale.
{"title":"Robust multi-parametric sensor system for the online detection of microbial biofilms in industrial applications — Preliminary examinations","authors":"M. Ruhnow, Julia Kohser, T. Bley, E. Boschke, M. Bulst, S. Wegner","doi":"10.1109/ISSNIP.2014.6827602","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827602","url":null,"abstract":"Biofilm formation can cause serious health hazards, mostly due to the uncontrolled release of pathogens. This can generate several problems in industrial facilities, e.g., in the food industry. The aim of the present study was to develop and implement a multi-parametric sensor system to monitor biofilm formation in laboratory as well as industrial set-ups. To minimize cross sensitivity or interference, the device was based on a combination of different measurement principles. Micro-organisms were initially cultivated in a laboratory scale reactor. Afterwards, biofilm formation will be studied with each prototype of the multi-parametric sensor followed by final tests on an industrial scale.","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":"126532409","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}
This paper presents a novel procedure involving waveatom transform and Circular Complex-valued Extreme Learning Machine (CC-ELM) for automatic characterization of mammographic microcalcifications into benign or malignant. Waveatom transform is used to transform the mammogram image into multi-frequency domain features. The best feature set is obtained by feature reduction through Principal Component Analysis. The reduced feature set is then used to perform classification through a CC-ELM classifier. CC-ELM is a fast learning fully complex-valued classifier to perform real-valued classification tasks efficiently. Mammographic images obtained from Digital Database for Screening Mammography have been used in the study. About 400 Region of Interests extracted from mammograms are used. The performance of the proposed method is about 96.19%, which is significantly higher than the existing methods.
{"title":"A novel method for benign and malignant characterization of mammographic microcalcifications employing waveatom features and circular complex valued — Extreme Learning Machine","authors":"Malar Elangeeran, Savitha Ramasamy, Kandaswamy Arumugam","doi":"10.1109/ISSNIP.2014.6827660","DOIUrl":"https://doi.org/10.1109/ISSNIP.2014.6827660","url":null,"abstract":"This paper presents a novel procedure involving waveatom transform and Circular Complex-valued Extreme Learning Machine (CC-ELM) for automatic characterization of mammographic microcalcifications into benign or malignant. Waveatom transform is used to transform the mammogram image into multi-frequency domain features. The best feature set is obtained by feature reduction through Principal Component Analysis. The reduced feature set is then used to perform classification through a CC-ELM classifier. CC-ELM is a fast learning fully complex-valued classifier to perform real-valued classification tasks efficiently. Mammographic images obtained from Digital Database for Screening Mammography have been used in the study. About 400 Region of Interests extracted from mammograms are used. The performance of the proposed method is about 96.19%, which is significantly higher than the existing methods.","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":"128176737","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}