Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987290
Fangqing Li
Aiming at the problem of the extension framework for complex data processing, this paper uses the CEP technology as a reference to propose a complex event big data generation method based on Bayesian networks. This method takes part of the real sample data as the research object, combines the experience of experts in related fields, gives the definition of complex event models, and uses algebraic expressions to describe the specific event information in the data set, such as event models such as cause and effect, sequence, selection, and coordination. Network communication relationship expansion based on multi-network integration uses multiple networks, analyzes the mapping between networks, and expands the connectivity of the network. The network communication relationship expansion based on named entity recognition extracts named entities that can expand the network from a single network. 11.2% reduction in complexity.
{"title":"Research on the Computer Complex Data Processing in the Big Data Era","authors":"Fangqing Li","doi":"10.1109/I-SMAC55078.2022.9987290","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987290","url":null,"abstract":"Aiming at the problem of the extension framework for complex data processing, this paper uses the CEP technology as a reference to propose a complex event big data generation method based on Bayesian networks. This method takes part of the real sample data as the research object, combines the experience of experts in related fields, gives the definition of complex event models, and uses algebraic expressions to describe the specific event information in the data set, such as event models such as cause and effect, sequence, selection, and coordination. Network communication relationship expansion based on multi-network integration uses multiple networks, analyzes the mapping between networks, and expands the connectivity of the network. The network communication relationship expansion based on named entity recognition extracts named entities that can expand the network from a single network. 11.2% reduction in complexity.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124295344","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987303
Simranjit Kaur, S. Vig
In this paper, an effective and efficient power hybrid power generation model is presented in which Maximum power point is tracked by using Fuzzy Logic Controller. The main objective of the proposed approach is to enhance the power capabilities of systems in order to fulfill the increasing load demand. To combat this task, a fuzzy based MPPT technique is implement in power generating system that takes two inputs. Furthermore, two optimization algorithms i.e. chaotic map and Differential Evolution (DE) are hybridized for optimizing the range of variables for two input functions of fuzzy model. The fitness value is calculated in terms of increase in power capabilities. Also, the proposed model utilized two energy sources i.e. Wind energy and solar energy for providing the necessary supply to customers during peak hours. A switching circuitry is also used in the proposed hybrid model for switching between two models when one is not able to generate electricity. The performance of the proposed fuzzy based approach is examined and validated by putting it in comparison with traditional ACO model in terms of their voltage, current and power generation abilities. In addition to this, analytical study is also conducted for wind and solar energy models to determine their abilities for generating power and satisfying load demands.
{"title":"Modeling of Hybrid Power Generation using FLC","authors":"Simranjit Kaur, S. Vig","doi":"10.1109/I-SMAC55078.2022.9987303","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987303","url":null,"abstract":"In this paper, an effective and efficient power hybrid power generation model is presented in which Maximum power point is tracked by using Fuzzy Logic Controller. The main objective of the proposed approach is to enhance the power capabilities of systems in order to fulfill the increasing load demand. To combat this task, a fuzzy based MPPT technique is implement in power generating system that takes two inputs. Furthermore, two optimization algorithms i.e. chaotic map and Differential Evolution (DE) are hybridized for optimizing the range of variables for two input functions of fuzzy model. The fitness value is calculated in terms of increase in power capabilities. Also, the proposed model utilized two energy sources i.e. Wind energy and solar energy for providing the necessary supply to customers during peak hours. A switching circuitry is also used in the proposed hybrid model for switching between two models when one is not able to generate electricity. The performance of the proposed fuzzy based approach is examined and validated by putting it in comparison with traditional ACO model in terms of their voltage, current and power generation abilities. In addition to this, analytical study is also conducted for wind and solar energy models to determine their abilities for generating power and satisfying load demands.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125175156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987297
Yee Jin Yeo, A. Balakrishnan, S. Selvaperumal, Illanur Muhaini Binti Mohd Nor
The main aim of this work is to develop a manually operated camera assisted firefighting robot with the capability to extinguish fire and controlled remotely by using an Android application. In this proposed work, a robot prototype was developed with the inclusion of camera module and relevant sensors. The robot was interfaced with Blynk IoT platform, which can be used by an Android device to control the robot. The performance of the developed robot is evaluated by testing the speed, water sprayer, sensors, fire extinguishment, and operating distance. The overall robot speed is lower than expected due to the condition of the test, which is 13.118 cm per second. The effective water sprayer area is 85 cm squared, that is considered as small due to the limited aiming angle. The overall sensors accuracy while considering several distances is 77.47%, which can be improved with omni-directional sensors. The fire extinguishment test proved that the robot is suitable for extinguishing spread type of fire. The optimal operating distance of the robot from the local server is from 0 to 26 meters, considering concrete walls as obstacles. Finally, the developed system has proved that the implementation of Android device and IoT platform is doable while retaining the core features such as live camera feed, fire detection, and fire extinguishment.
{"title":"Android Controlled Fire Fighter Robot Using IoT","authors":"Yee Jin Yeo, A. Balakrishnan, S. Selvaperumal, Illanur Muhaini Binti Mohd Nor","doi":"10.1109/I-SMAC55078.2022.9987297","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987297","url":null,"abstract":"The main aim of this work is to develop a manually operated camera assisted firefighting robot with the capability to extinguish fire and controlled remotely by using an Android application. In this proposed work, a robot prototype was developed with the inclusion of camera module and relevant sensors. The robot was interfaced with Blynk IoT platform, which can be used by an Android device to control the robot. The performance of the developed robot is evaluated by testing the speed, water sprayer, sensors, fire extinguishment, and operating distance. The overall robot speed is lower than expected due to the condition of the test, which is 13.118 cm per second. The effective water sprayer area is 85 cm squared, that is considered as small due to the limited aiming angle. The overall sensors accuracy while considering several distances is 77.47%, which can be improved with omni-directional sensors. The fire extinguishment test proved that the robot is suitable for extinguishing spread type of fire. The optimal operating distance of the robot from the local server is from 0 to 26 meters, considering concrete walls as obstacles. Finally, the developed system has proved that the implementation of Android device and IoT platform is doable while retaining the core features such as live camera feed, fire detection, and fire extinguishment.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114614152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987352
J. M. Sahayaraj, K. Gunasekaran, S. Verma, P. Ramesh, G. Murugesan
The purpose of this research is to develop protocols for the underutilized channels of primary user usage with scant transmission and minimal interference associating with the secondary users. This has been achieved with large scale fading channels using discrete time queues. Flow level analysis has been made by appropriate queuing model and packet level analysis has done with NS2 simulator. CRTTP uses the channel selection procedure based on utilization, throughput and minimal drop rate whereas the response time denotes whether to increment or decrement transmission power control. Cognitive Radio based Temporal Transmission Protocol Single Channel (CRTTP-SC) denies transmission if sustainable routing parameter does not by cognitive user. Cognitive Radio based Temporal Transmission Protocol Single Channel Receiver Capacity (CRTTP-SCRC). CRTTP-SCRC protocol calculates the channel utilization, drop rate and receiver capacity after which it determines whether to prolong transmission or to refrain from transmission. Cognitive Radio based Temporal Transmission Protocol Multiple Channel (CRTTP MC) uses exponential distribution with inter-arrival time of packets with appropriate transmission power assigned to each channel. Cognitive Radio based Temporal Transmission Protocol Multiple Channel Collision Avoidance (CRTTP-MCCA) assigning hyper exponential distribution with inter arrival time of packets for optimizing the usage of lesser utilized channel. Comparison has been done with simulations for single channel protocols of CRTTP and multiple channel protocols of CRTTP.
{"title":"Resource Allocation and Information Exchange of Cognitive user Connectivity with Minimal Interference using Simulation Analysis","authors":"J. M. Sahayaraj, K. Gunasekaran, S. Verma, P. Ramesh, G. Murugesan","doi":"10.1109/I-SMAC55078.2022.9987352","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987352","url":null,"abstract":"The purpose of this research is to develop protocols for the underutilized channels of primary user usage with scant transmission and minimal interference associating with the secondary users. This has been achieved with large scale fading channels using discrete time queues. Flow level analysis has been made by appropriate queuing model and packet level analysis has done with NS2 simulator. CRTTP uses the channel selection procedure based on utilization, throughput and minimal drop rate whereas the response time denotes whether to increment or decrement transmission power control. Cognitive Radio based Temporal Transmission Protocol Single Channel (CRTTP-SC) denies transmission if sustainable routing parameter does not by cognitive user. Cognitive Radio based Temporal Transmission Protocol Single Channel Receiver Capacity (CRTTP-SCRC). CRTTP-SCRC protocol calculates the channel utilization, drop rate and receiver capacity after which it determines whether to prolong transmission or to refrain from transmission. Cognitive Radio based Temporal Transmission Protocol Multiple Channel (CRTTP MC) uses exponential distribution with inter-arrival time of packets with appropriate transmission power assigned to each channel. Cognitive Radio based Temporal Transmission Protocol Multiple Channel Collision Avoidance (CRTTP-MCCA) assigning hyper exponential distribution with inter arrival time of packets for optimizing the usage of lesser utilized channel. Comparison has been done with simulations for single channel protocols of CRTTP and multiple channel protocols of CRTTP.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128002305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987254
R. Yugha, V. Vinodhini, J. Arunkumar, K. Varalakshmi, G. Karthikeyan, G. Ramkumar
A condition known as glaucoma, is an eye illness brought on by high intraocular pressure, may lead to total blindness. On the other hand, prompt glaucoma screening-based therapy may keep the individual from losing all vision. Professionals manually analyze retina to pinpoint the areas affected by glaucoma using precise testing procedures. However, because of complicated glaucoma testing methods and a lack of resources, delays in detection are often experienced that may raise the global rate of visual impairment. Moreover, the significant resemblance between the lesion and eye color also makes the manual categorization procedure more difficult. Hence, there exists an urgent need to develop an effective smart approach that can precisely detect the Optic Disc as well as Optic Cup lesions at the early stage in order to address the difficulties of manual methods. Therefore, a Deep Learning based strategy called EfficientDet-DO with EfficientNet-B0 serving as its foundation has been proposed in this paper. There are three phases in the conceptual methodology for the localization and categorization of glaucoma. First, the EfficientNet-B0 feature extractor computes the feature representations from the suspicious examples. Next, the top-down and bottom-up key points merging operations are repeatedly carried out by the Bi-Directional Feature Pyramid system modules of EfficientDet-DO using the calculated characteristics from EfficientNet-B0. The resulting localized areas of a glaucoma lesion and its accompanying classification are anticipated in the last stage.
{"title":"An Automated Glaucoma Detection from Fundus Images based on Deep Learning Network","authors":"R. Yugha, V. Vinodhini, J. Arunkumar, K. Varalakshmi, G. Karthikeyan, G. Ramkumar","doi":"10.1109/I-SMAC55078.2022.9987254","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987254","url":null,"abstract":"A condition known as glaucoma, is an eye illness brought on by high intraocular pressure, may lead to total blindness. On the other hand, prompt glaucoma screening-based therapy may keep the individual from losing all vision. Professionals manually analyze retina to pinpoint the areas affected by glaucoma using precise testing procedures. However, because of complicated glaucoma testing methods and a lack of resources, delays in detection are often experienced that may raise the global rate of visual impairment. Moreover, the significant resemblance between the lesion and eye color also makes the manual categorization procedure more difficult. Hence, there exists an urgent need to develop an effective smart approach that can precisely detect the Optic Disc as well as Optic Cup lesions at the early stage in order to address the difficulties of manual methods. Therefore, a Deep Learning based strategy called EfficientDet-DO with EfficientNet-B0 serving as its foundation has been proposed in this paper. There are three phases in the conceptual methodology for the localization and categorization of glaucoma. First, the EfficientNet-B0 feature extractor computes the feature representations from the suspicious examples. Next, the top-down and bottom-up key points merging operations are repeatedly carried out by the Bi-Directional Feature Pyramid system modules of EfficientDet-DO using the calculated characteristics from EfficientNet-B0. The resulting localized areas of a glaucoma lesion and its accompanying classification are anticipated in the last stage.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128373497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987341
Juanjuan Luo
Particle Array, Texture Expansion method, HDR high dynamic texture application, VRAY layered rendering setting, photon file application that can store radiosity information, application of dynamic system in some areas, etc. are some examples of wiring principle and 3D modeling. The module realizes the functions of human-computer interaction and image display, and the entity editing module realizes the editing function of each entity in the scene and transmits the real-time rendering and editing results through data transmission and displays them. This part is realized by the rendering engine, and this method avoids the complexity. It has very low requirements on hardware equipment and realizes automatic 3D reconstruction of virtual scenes based on sequence images.
{"title":"Software Design of 3D Animation Scene based on Virtual Image Modeling Algorithm","authors":"Juanjuan Luo","doi":"10.1109/I-SMAC55078.2022.9987341","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987341","url":null,"abstract":"Particle Array, Texture Expansion method, HDR high dynamic texture application, VRAY layered rendering setting, photon file application that can store radiosity information, application of dynamic system in some areas, etc. are some examples of wiring principle and 3D modeling. The module realizes the functions of human-computer interaction and image display, and the entity editing module realizes the editing function of each entity in the scene and transmits the real-time rendering and editing results through data transmission and displays them. This part is realized by the rendering engine, and this method avoids the complexity. It has very low requirements on hardware equipment and realizes automatic 3D reconstruction of virtual scenes based on sequence images.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130545232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987432
M. P. Suresh, V. R. Vedha Rhythesh, J. Dinesh, K. Deepak, J. Manikandan
A basic design of robot that can fight fires at an affordable cost could prove to be boon in fighting domestic fires, till help arrives. The robot developed consists of three elements which is the hardware, electronic interfacing circuits, and software program. The robot has four battery operated motor (BO motor). This firefighting robotic system is capable of detecting and extinguishing fire. These robots can be made to roll into places where it is not safe for humans to enter. Time is of essence when it comes to fighting fires as even a few minutes’ delay can turn small fires into raging inferno. This robot is designed as a first response unit so it can suppress the fire keeps it under control till help arrives. This firefighting robotic system is controlled by an Arduino Uno development board. It is also equipped with the fire flame sensor for detecting fires. It is equipped with a water tank and a pump. So, on detecting fires it sprays water extinguishing the fire. Water spraying nozzle is mounted on servo motor to cover maximum area. Although there is a lot of scope for improvement, this could be a first step in developing a complete fire-fighting robot that could also rescue victims. The main function of this robot is to become an unmanned support vehicle, developed to search and extinguish fire. By using such robots, fire identification and rescue activities can be done with greater accuracy and securely without exposing the fire fighters to dangerous conditions. In other words, robots can reduce the need to expose fire fighters to danger.
{"title":"An Arduino Uno Controlled Fire Fighting Robot for Fires in Enclosed Spaces","authors":"M. P. Suresh, V. R. Vedha Rhythesh, J. Dinesh, K. Deepak, J. Manikandan","doi":"10.1109/I-SMAC55078.2022.9987432","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987432","url":null,"abstract":"A basic design of robot that can fight fires at an affordable cost could prove to be boon in fighting domestic fires, till help arrives. The robot developed consists of three elements which is the hardware, electronic interfacing circuits, and software program. The robot has four battery operated motor (BO motor). This firefighting robotic system is capable of detecting and extinguishing fire. These robots can be made to roll into places where it is not safe for humans to enter. Time is of essence when it comes to fighting fires as even a few minutes’ delay can turn small fires into raging inferno. This robot is designed as a first response unit so it can suppress the fire keeps it under control till help arrives. This firefighting robotic system is controlled by an Arduino Uno development board. It is also equipped with the fire flame sensor for detecting fires. It is equipped with a water tank and a pump. So, on detecting fires it sprays water extinguishing the fire. Water spraying nozzle is mounted on servo motor to cover maximum area. Although there is a lot of scope for improvement, this could be a first step in developing a complete fire-fighting robot that could also rescue victims. The main function of this robot is to become an unmanned support vehicle, developed to search and extinguish fire. By using such robots, fire identification and rescue activities can be done with greater accuracy and securely without exposing the fire fighters to dangerous conditions. In other words, robots can reduce the need to expose fire fighters to danger.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130739807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987316
C. Padmavathy, V. Akshaya, R. Menaha, S. Raja
Node lifetime is an important factor in wireless sensor networks as the entire lifetime of the network depends on the individual nodes. Researchers pay more attention towards enhancement of node lifetime through various deployment models. Rather than concentrating over node deployment, efficient clustering, data aggregation in wireless sensor networks enhances the node and network lifetime, minimize the energy utilization, reduces network congestion and identifies an optimal route for better load balancing. Clustering approaches considers the parameters like residual energy of node, communication range, distance between node and sink. Specifically, cluster head selection and replacement is a crucial part in clustering which directly relates to energy management of network. Considering these facts, an energy efficient clustering approach to enhance node lifetime through hybrid adaptive neuro fuzzy inference system (ANFIS) is proposed in this research work. Conventional models are compared with proposed hybrid approach to demonstrate the superior performance.
{"title":"Hybrid Cluster Head Selection Approach for Node Lifetime Enhancement in Wireless Sensor Networks","authors":"C. Padmavathy, V. Akshaya, R. Menaha, S. Raja","doi":"10.1109/I-SMAC55078.2022.9987316","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987316","url":null,"abstract":"Node lifetime is an important factor in wireless sensor networks as the entire lifetime of the network depends on the individual nodes. Researchers pay more attention towards enhancement of node lifetime through various deployment models. Rather than concentrating over node deployment, efficient clustering, data aggregation in wireless sensor networks enhances the node and network lifetime, minimize the energy utilization, reduces network congestion and identifies an optimal route for better load balancing. Clustering approaches considers the parameters like residual energy of node, communication range, distance between node and sink. Specifically, cluster head selection and replacement is a crucial part in clustering which directly relates to energy management of network. Considering these facts, an energy efficient clustering approach to enhance node lifetime through hybrid adaptive neuro fuzzy inference system (ANFIS) is proposed in this research work. Conventional models are compared with proposed hybrid approach to demonstrate the superior performance.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128683785","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987393
Anshul Jindal, Jiby Mariya Jose, S. Benedict, M. Gerndt
The ongoing accomplishments in the decades-long realization of computer vision have infused new dimensions in various research areas such as smart mobility, smart healthcare, education, finance, and so forth. Research works relating to automated object detection, deep learning-assisted data pipelines, and energy-efficient end-to-end solutions have enabled newer perceptions among researchers, albeit the existence of challenges. This paper proposes an object detection system using energy-efficient Long Range (LoRA) communication media on edge nodes such as Raspberry Pi, Coral DevBoard, and Nvidia Jetson Nano. The proposed approach utilized energy-efficient methods to collaboratively offload object detection-related tasks such as capturing images, training images, and inferring objects across a compendium of computing nodes using LoRA. In addition, this research study has attempted to reveal the inference capabilities of images on three different edge nodes. The proposed work has achieved a power difference of at least 1.2 watts during the inference period of the deep learning models without challenging the prediction accuracy with respect to the base model.
{"title":"LoRa-Powered Energy-Effcient Object Detection Mechanism in Edge Computing Nodes","authors":"Anshul Jindal, Jiby Mariya Jose, S. Benedict, M. Gerndt","doi":"10.1109/I-SMAC55078.2022.9987393","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987393","url":null,"abstract":"The ongoing accomplishments in the decades-long realization of computer vision have infused new dimensions in various research areas such as smart mobility, smart healthcare, education, finance, and so forth. Research works relating to automated object detection, deep learning-assisted data pipelines, and energy-efficient end-to-end solutions have enabled newer perceptions among researchers, albeit the existence of challenges. This paper proposes an object detection system using energy-efficient Long Range (LoRA) communication media on edge nodes such as Raspberry Pi, Coral DevBoard, and Nvidia Jetson Nano. The proposed approach utilized energy-efficient methods to collaboratively offload object detection-related tasks such as capturing images, training images, and inferring objects across a compendium of computing nodes using LoRA. In addition, this research study has attempted to reveal the inference capabilities of images on three different edge nodes. The proposed work has achieved a power difference of at least 1.2 watts during the inference period of the deep learning models without challenging the prediction accuracy with respect to the base model.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126557822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-10DOI: 10.1109/I-SMAC55078.2022.9987327
Hui Wing Kuan, N. S. Lai
High electrical consumption in operating the factory has been a critical source of expense, especially for a frozen food warehouse. Hence, this project is proposing a solution by utilising Industrial IoT and Machine Learning to reduce the use of electricity. A simple prototype has been built by using ESPS266, DHT22 and Raspberry Pi, with the aid of NodeRed and TensorFlow for data collection and machine learning for prediction. The predicted temperature has obtained an accuracy of up to 98.24% for operating frozen food storage. Besides that, the efficiency of energy optimization forthe refrigeration compressor is up to 9 hours with the cost saved up to RM869.62 per year for 1HP.
{"title":"Condition Monitoring of Frozen Storage for Energy Optimization","authors":"Hui Wing Kuan, N. S. Lai","doi":"10.1109/I-SMAC55078.2022.9987327","DOIUrl":"https://doi.org/10.1109/I-SMAC55078.2022.9987327","url":null,"abstract":"High electrical consumption in operating the factory has been a critical source of expense, especially for a frozen food warehouse. Hence, this project is proposing a solution by utilising Industrial IoT and Machine Learning to reduce the use of electricity. A simple prototype has been built by using ESPS266, DHT22 and Raspberry Pi, with the aid of NodeRed and TensorFlow for data collection and machine learning for prediction. The predicted temperature has obtained an accuracy of up to 98.24% for operating frozen food storage. Besides that, the efficiency of energy optimization forthe refrigeration compressor is up to 9 hours with the cost saved up to RM869.62 per year for 1HP.","PeriodicalId":306129,"journal":{"name":"2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)","volume":"37 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113986262","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：