Pub Date : 2020-09-14DOI: 10.1504/ijguc.2020.10032064
Bruno Andriamanalimanana, Chen-Fu Chiang, Jorge E. Novillo, Sam Sengupta, Ali Tekeoglu
For decentralised cryptocurrency systems, it is important to provide users an efficient network. One performance bottleneck is the latency issue. To address this issue, we provide four protocols to utilise the resources based on the traffic in the network to alleviate the latency in the network. To facilitate the verification process, we discuss three variant injection protocols: Periodic Injection of Transaction via Evaluation Corridor (PITEC), Probabilistic Injection of Transactions (PIT) and Adaptive Semi-synchronous Transaction Inject (ASTI). The injection protocols are variants based on the given assumptions of the network. The goal is to provide dynamic injection of unverified transactions to enhance the performance of the network. The Adaptive Policy Optimisation (APO) protocols aim at optimising a cryptocurrency system's own house policy. The house policy optimisation is translated into a 0/1 knapsack problem. The APO protocol is a fully polynomial time approximation scheme for the decentralised ledger system.
{"title":"Efficient variant transaction injection protocols and adaptive policy optimisation for decentralised ledger systems","authors":"Bruno Andriamanalimanana, Chen-Fu Chiang, Jorge E. Novillo, Sam Sengupta, Ali Tekeoglu","doi":"10.1504/ijguc.2020.10032064","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10032064","url":null,"abstract":"For decentralised cryptocurrency systems, it is important to provide users an efficient network. One performance bottleneck is the latency issue. To address this issue, we provide four protocols to utilise the resources based on the traffic in the network to alleviate the latency in the network. To facilitate the verification process, we discuss three variant injection protocols: Periodic Injection of Transaction via Evaluation Corridor (PITEC), Probabilistic Injection of Transactions (PIT) and Adaptive Semi-synchronous Transaction Inject (ASTI). The injection protocols are variants based on the given assumptions of the network. The goal is to provide dynamic injection of unverified transactions to enhance the performance of the network. The Adaptive Policy Optimisation (APO) protocols aim at optimising a cryptocurrency system's own house policy. The house policy optimisation is translated into a 0/1 knapsack problem. The APO protocol is a fully polynomial time approximation scheme for the decentralised ledger system.","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125244092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-09-14DOI: 10.1504/ijguc.2020.10032062
M. A. Hamid, Y. Hafeez, Bushra Hamid, M. Humayun, Noor Zaman Jhanjhi
Architectural design is expected to provide virtuous outcomes of quality software products by satisfying customer requirements. A foremost apprehension of the customer is to have a better quality product within a minimal time span. The evaporation of architectural knowledge causes snags for the quality of a system being developed. The research study aims to propose and validate a framework bridging the gaps in architectural knowledge management. A mixed research approach has been employed. In order to align closely with industry practices, action research has been considered as research methodology, while the evaluation has been performed using a multiple case study approach. The results show that the framework enables the architects to cope with complex designs in distributed software development environments. The developed tool enabled to shift the theory into practice by assisting in creation of system architecture, knowledge survival and support architectural evolution with changing requirements.
{"title":"Towards an effective approach for architectural knowledge management considering global software development","authors":"M. A. Hamid, Y. Hafeez, Bushra Hamid, M. Humayun, Noor Zaman Jhanjhi","doi":"10.1504/ijguc.2020.10032062","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10032062","url":null,"abstract":"Architectural design is expected to provide virtuous outcomes of quality software products by satisfying customer requirements. A foremost apprehension of the customer is to have a better quality product within a minimal time span. The evaporation of architectural knowledge causes snags for the quality of a system being developed. The research study aims to propose and validate a framework bridging the gaps in architectural knowledge management. A mixed research approach has been employed. In order to align closely with industry practices, action research has been considered as research methodology, while the evaluation has been performed using a multiple case study approach. The results show that the framework enables the architects to cope with complex designs in distributed software development environments. The developed tool enabled to shift the theory into practice by assisting in creation of system architecture, knowledge survival and support architectural evolution with changing requirements.","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122434030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-09-14DOI: 10.1504/ijguc.2020.10032055
Kshira Sagar Sahoo, S. Ramasubbareddy, B. Balusamy, B. Deep
In Software Defined Networks (SDN), the decoupling of the control logic from the data plane provides numerous advantages. Since its inception, SDN is a subject of a wide range of criticism mainly related to the scalability issues of the control plane. To address these limitations, recent architectures have supported the implementation of multiple controllers. Usage of multiple controllers leads to Controller Placement Problems (CPP) particularly in wide area networks. In most of the placement strategies, authors focused on propagation latency, because it is a critical factor in real networks. In this paper, the placement problem has been formulated on the basis of propagation latency as an optimisation problem, and Simulated Annealing (SA) technique has been used to analyse the problem. Further, we investigate the behaviour of SA with four different neighbouring solutions technique. The effectiveness of the algorithms is carried out on TataNld topology and implemented using MATLAB simulator.
{"title":"Analysing control plane scalability issue of software defined wide area network using simulated annealing technique","authors":"Kshira Sagar Sahoo, S. Ramasubbareddy, B. Balusamy, B. Deep","doi":"10.1504/ijguc.2020.10032055","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10032055","url":null,"abstract":"In Software Defined Networks (SDN), the decoupling of the control logic from the data plane provides numerous advantages. Since its inception, SDN is a subject of a wide range of criticism mainly related to the scalability issues of the control plane. To address these limitations, recent architectures have supported the implementation of multiple controllers. Usage of multiple controllers leads to Controller Placement Problems (CPP) particularly in wide area networks. In most of the placement strategies, authors focused on propagation latency, because it is a critical factor in real networks. In this paper, the placement problem has been formulated on the basis of propagation latency as an optimisation problem, and Simulated Annealing (SA) technique has been used to analyse the problem. Further, we investigate the behaviour of SA with four different neighbouring solutions technique. The effectiveness of the algorithms is carried out on TataNld topology and implemented using MATLAB simulator.","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130266932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-09-14DOI: 10.1504/ijguc.2020.10029811
Ankita Sharma, Puja Munjal, H. Banati
Cloud computing has now been introduced in organisations all around the globe. With the developing prevalence of grid and distributed computing, it has become incredibly important to maintain security and trust. Researchers have now begun concentrating on mining information in cloud computing and have begun distinguishing the basic factor of moral trust. Moral angles in the cloud rely upon the application and the present conditions. Data mining is a procedure for distinguishing the most significant data from a lot of irregular information. In this paper, a three phased methodology is adopted, involving machine learning techniques to discover the most important parameter on which trust is based in the cloud environment. The methodology was then implemented on data sets, proving privacy is the most important factor to calculate ethical trust in cloud computing. The results can be employed in real cloud environments to establish trust as service providers can now consider privacy as the main issue in this relatively new distributed computing environment.
{"title":"Entropy-based classification of trust factors for cloud computing","authors":"Ankita Sharma, Puja Munjal, H. Banati","doi":"10.1504/ijguc.2020.10029811","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10029811","url":null,"abstract":"Cloud computing has now been introduced in organisations all around the globe. With the developing prevalence of grid and distributed computing, it has become incredibly important to maintain security and trust. Researchers have now begun concentrating on mining information in cloud computing and have begun distinguishing the basic factor of moral trust. Moral angles in the cloud rely upon the application and the present conditions. Data mining is a procedure for distinguishing the most significant data from a lot of irregular information. In this paper, a three phased methodology is adopted, involving machine learning techniques to discover the most important parameter on which trust is based in the cloud environment. The methodology was then implemented on data sets, proving privacy is the most important factor to calculate ethical trust in cloud computing. The results can be employed in real cloud environments to establish trust as service providers can now consider privacy as the main issue in this relatively new distributed computing environment.","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128072058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-09-14DOI: 10.1504/ijguc.2020.10032061
Chinwe Peace Igiri, Yudhveer Singh, D. Bhargava, S. Shikaa
Real-world supply chain network is complex due to large problem size and constraints. An optimum petroleum products scheduling would not only influence the distribution cost but also result in optimal product scheduling. The bio-inspired method is preferred alternative to exact algorithms because it does not require prior knowledge of the initial solution unlike the latter. The study proposes an improved African Buffalo Optimisation (ABO) algorithm for petroleum supply chain distribution. The ABO is a swarm intelligence-based bio-inspired algorithm with significant performance track record. It models the grazing and defending lifestyle of the African buffaloes in the savannah. The chaotic ABO and chaotic-Levy ABO are the ABO's improved variants with outstanding performance in recent studies. The present study applies the standard ABO and its improved variants to obtain a near optimum petroleum distribution scheduling solution. The comparative result shows that the proposed approach outperformed existing exact algorithms.
{"title":"Improved African buffalo optimisation algorithm for petroleum product supply chain management","authors":"Chinwe Peace Igiri, Yudhveer Singh, D. Bhargava, S. Shikaa","doi":"10.1504/ijguc.2020.10032061","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10032061","url":null,"abstract":"Real-world supply chain network is complex due to large problem size and constraints. An optimum petroleum products scheduling would not only influence the distribution cost but also result in optimal product scheduling. The bio-inspired method is preferred alternative to exact algorithms because it does not require prior knowledge of the initial solution unlike the latter. The study proposes an improved African Buffalo Optimisation (ABO) algorithm for petroleum supply chain distribution. The ABO is a swarm intelligence-based bio-inspired algorithm with significant performance track record. It models the grazing and defending lifestyle of the African buffaloes in the savannah. The chaotic ABO and chaotic-Levy ABO are the ABO's improved variants with outstanding performance in recent studies. The present study applies the standard ABO and its improved variants to obtain a near optimum petroleum distribution scheduling solution. The comparative result shows that the proposed approach outperformed existing exact algorithms.","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125054409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-09-14DOI: 10.1504/ijguc.2020.10032054
Victor Chang, G. Sunitha, S. D. Dilip Kumar, S. Raghavendra, N. Srinidhi
The Internet of Things (IoT) consists of large amount of energy consuming devices which are pre-figured to progress the effective competence of several industrial applications. It is very much essential to bring down the energy use of every device deployed in the IoT network without compromising the Quality of Service (QoS) for intelligent transportation system. To achieve this objective, a multiobjective optimisation problem to accomplish the aim of estimating the outage performance of clustering process and the network lifetime is devised. Subsequently, a Hybrid Energy Efficient and QoS Aware (HEEQA) algorithm that is a combination of Quantum Particle Swarm Optimisation (QPSO) along with improved Non-dominated Sorting Genetic Algorithm (NSGA) to achieve energy balance among the devices is proposed. NSGA is applied to solve the problem of multiobjective optimisation and the QPSO algorithm is used to find the optima cooperative nodes and cluster head in the clusters.
{"title":"Hybrid energy-efficient and QoS-aware algorithm for intelligent transportation system in IoT","authors":"Victor Chang, G. Sunitha, S. D. Dilip Kumar, S. Raghavendra, N. Srinidhi","doi":"10.1504/ijguc.2020.10032054","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10032054","url":null,"abstract":"The Internet of Things (IoT) consists of large amount of energy consuming devices which are pre-figured to progress the effective competence of several industrial applications. It is very much essential to bring down the energy use of every device deployed in the IoT network without compromising the Quality of Service (QoS) for intelligent transportation system. To achieve this objective, a multiobjective optimisation problem to accomplish the aim of estimating the outage performance of clustering process and the network lifetime is devised. Subsequently, a Hybrid Energy Efficient and QoS Aware (HEEQA) algorithm that is a combination of Quantum Particle Swarm Optimisation (QPSO) along with improved Non-dominated Sorting Genetic Algorithm (NSGA) to achieve energy balance among the devices is proposed. NSGA is applied to solve the problem of multiobjective optimisation and the QPSO algorithm is used to find the optima cooperative nodes and cluster head in the clusters.","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114532615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-09-14DOI: 10.1504/ijguc.2020.10029963
R. VijayaSaraswathi, L. Sree, K. Anuradha
A Group Key Management is a technique to establish shared group key, between the cluster head and sensor nodes for multiple sessions in a clustered network environment. The common use of this established group key (also termed as conference key) is to permit users to encrypt and decrypt a particular broadcast message that is meant for the total user group. In this work, we propose a cluster based dynamic group key management protocol that is based on public key cryptography. Cluster head initiates establishment of group key to the sensor nodes efficiently and achieves secure communication. Later, the computation of the common group key is performed by each sensor node. Group members have functionality to join and leave from particular communication along with this, other nodes, equal to threshold compute new conference key without involvement of cluster head. The proposed protocol is investigated in terms of security and complexity analysis using network simulator (NS-2).
{"title":"Dynamic group key management scheme for clustered wireless sensor networks","authors":"R. VijayaSaraswathi, L. Sree, K. Anuradha","doi":"10.1504/ijguc.2020.10029963","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10029963","url":null,"abstract":"A Group Key Management is a technique to establish shared group key, between the cluster head and sensor nodes for multiple sessions in a clustered network environment. The common use of this established group key (also termed as conference key) is to permit users to encrypt and decrypt a particular broadcast message that is meant for the total user group. In this work, we propose a cluster based dynamic group key management protocol that is based on public key cryptography. Cluster head initiates establishment of group key to the sensor nodes efficiently and achieves secure communication. Later, the computation of the common group key is performed by each sensor node. Group members have functionality to join and leave from particular communication along with this, other nodes, equal to threshold compute new conference key without involvement of cluster head. The proposed protocol is investigated in terms of security and complexity analysis using network simulator (NS-2).","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132484850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-09-14DOI: 10.1504/ijguc.2020.10032068
Pardeep Kumar, S. Verma
Airborne Ad-hoc Networks (AANETs) are a subclass of Vehicular Ad-hoc Networks (VANETs). The major challenge in AANET implementation is the regular route breaks due to very high mobility of aircraft. To deal with this routing challenge in AANETs, we have designed a new protocol named Airborne OLSR (AOLSR) which is based on more optimisation of the MPR selection technique used in existing OLSR protocol. A hardware implementation of any protocol is always crucial for the validation of its design. So, this manuscript also provides a hardware implementation of both OLSR and AOLSR protocols using Verilog. The node architectures for both the protocols have been implemented in Vivado. The node's sub-component implementation area, execution time and total power consumption have been used as parameters to compare their performance. The simulation analysis shows that the proposed AOLSR performs better as compared to OLSR in terms of total power consumption and execution time.
{"title":"Hardware implementation of OLSR and improved OLSR (AOLSR) for AANETs","authors":"Pardeep Kumar, S. Verma","doi":"10.1504/ijguc.2020.10032068","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10032068","url":null,"abstract":"Airborne Ad-hoc Networks (AANETs) are a subclass of Vehicular Ad-hoc Networks (VANETs). The major challenge in AANET implementation is the regular route breaks due to very high mobility of aircraft. To deal with this routing challenge in AANETs, we have designed a new protocol named Airborne OLSR (AOLSR) which is based on more optimisation of the MPR selection technique used in existing OLSR protocol. A hardware implementation of any protocol is always crucial for the validation of its design. So, this manuscript also provides a hardware implementation of both OLSR and AOLSR protocols using Verilog. The node architectures for both the protocols have been implemented in Vivado. The node's sub-component implementation area, execution time and total power consumption have been used as parameters to compare their performance. The simulation analysis shows that the proposed AOLSR performs better as compared to OLSR in terms of total power consumption and execution time.","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124809281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-08-01DOI: 10.1504/ijguc.2020.10030947
Júlio Mendonça, R. Lima, E. Andrade
Systems outages can have disastrous effects on businesses such as data loss, customer dissatisfaction, and subsequent revenue loss. Disaster recovery (DR) solutions have been adopted by companies to minimise the effects of these outages. However, the selection of an optimal DR solution is difficult since there does not exist a single solution that suits the requirement of every company (e.g., availability and costs). In this paper, we propose an integrated model-experiment approach to evaluate DR solutions. We perform experiments in different real-world DR solutions and propose analytic models to evaluate these solutions regarding DR key-metrics: steady-state availability, recovery time objective (RTO), recovery point objective (RPO), downtime, and costs. The results reveal that DR solutions can significantly improve availability and minimise costs. Also, a sensitivity analysis identifies the parameters that most affect the RPO and RTO of the DR adopted solutions.
{"title":"Evaluating and modelling solutions for disaster recovery","authors":"Júlio Mendonça, R. Lima, E. Andrade","doi":"10.1504/ijguc.2020.10030947","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10030947","url":null,"abstract":"Systems outages can have disastrous effects on businesses such as data loss, customer dissatisfaction, and subsequent revenue loss. Disaster recovery (DR) solutions have been adopted by companies to minimise the effects of these outages. However, the selection of an optimal DR solution is difficult since there does not exist a single solution that suits the requirement of every company (e.g., availability and costs). In this paper, we propose an integrated model-experiment approach to evaluate DR solutions. We perform experiments in different real-world DR solutions and propose analytic models to evaluate these solutions regarding DR key-metrics: steady-state availability, recovery time objective (RTO), recovery point objective (RPO), downtime, and costs. The results reveal that DR solutions can significantly improve availability and minimise costs. Also, a sensitivity analysis identifies the parameters that most affect the RPO and RTO of the DR adopted solutions.","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122923088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-23DOI: 10.1504/ijguc.2020.10028761
Dante Arias Torres, J. Hernández-Nolasco, Miguel A. Wister, Pablo Pancardo
In this paper, we aim to detect the fatigue based on accelerometer data from human gait using traditional classifiers from machine learning. First, we compare widely used machine learning classifiers to know which classifier can detect fatigue with the fewest errors. We observe that the best results were obtained with a Support Vector Machine (SVM) classifier. Later, we propose a new approach to solve the feature selection problem to know which features are more relevant to detect fatigue in healthy people based on their gait patterns. Finally, we used relevant gait features discovered in a previous step as input in classifiers used previously to know its impact on the classification process. Our results indicate that using only some gait features selected by our proposed feature selection method it is possible to improve fatigue detection based on data from human gait. We conclude that it is possible to distinguish between a normal gait person and a fatigued gait person with high accuracy.
{"title":"Detection of fatigue on gait using accelerometer data and supervised machine learning","authors":"Dante Arias Torres, J. Hernández-Nolasco, Miguel A. Wister, Pablo Pancardo","doi":"10.1504/ijguc.2020.10028761","DOIUrl":"https://doi.org/10.1504/ijguc.2020.10028761","url":null,"abstract":"In this paper, we aim to detect the fatigue based on accelerometer data from human gait using traditional classifiers from machine learning. First, we compare widely used machine learning classifiers to know which classifier can detect fatigue with the fewest errors. We observe that the best results were obtained with a Support Vector Machine (SVM) classifier. Later, we propose a new approach to solve the feature selection problem to know which features are more relevant to detect fatigue in healthy people based on their gait patterns. Finally, we used relevant gait features discovered in a previous step as input in classifiers used previously to know its impact on the classification process. Our results indicate that using only some gait features selected by our proposed feature selection method it is possible to improve fatigue detection based on data from human gait. We conclude that it is possible to distinguish between a normal gait person and a fatigued gait person with high accuracy.","PeriodicalId":375871,"journal":{"name":"Int. J. Grid Util. Comput.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124978423","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：