Pub Date : 2021-12-01DOI: 10.1109/UEMCON53757.2021.9666506
Shekhar Karanwal
Literature reports several local descriptors where discriminativity is enhanced by updating or modifying the previous techniques. This work suggests such novel descriptor (in light variations) by updating the methodology of OC-LBP. In OC-LBP, the difference among orthogonal neighbors and center pixel (of respective group) is thresholded to 1 or 0 according to thresholding function, which leads to 2 OLBP codes by putting binomial weights and adding values. Then both codes are concatenated to form OC-LBP size. The major drawback of OC-LBP is the thresholding of only Sign (S) component. This limits discriminativity of OC-LBP. To build more productive descriptor so called Complete (C) OC-LBP (COC-LBP), the Magnitude (M) details are incorporated with the S component of OC-LBP. The M based OC-LBP is called as MOC-LBP and S based is called as SOC-LBP. In MOC-LBP, the absolute of differences among orthogonal neighbors and center pixel (of respective group) is thresholded to 1 or 0 according to thresholding function. In thresholding function this time mean is utilized for comparison rather than center pixel, which is used in SOC-LBP. The two evolved MOLBP codes are concatenated to form the MOC-LBP size. Finally SOC-LBP and MOC-LBP codes are concatenated to build the COC-LBP size. Moving constantly in all positions builds the SOC-LBP and MOC-LBP images. The region feature concatenation of SOC-LBP and MOC-LBP makes COC-LBP size. The huge feature size is compressed by FLDA and matching is assisted from SVMs. On 2 benchmark datasets YB and EYB, COC-LBP marks its influence by defeating the results of many descriptors.
{"title":"COC-LBP: Complete Orthogonally Combined Local Binary Pattern for Face Recognition","authors":"Shekhar Karanwal","doi":"10.1109/UEMCON53757.2021.9666506","DOIUrl":"https://doi.org/10.1109/UEMCON53757.2021.9666506","url":null,"abstract":"Literature reports several local descriptors where discriminativity is enhanced by updating or modifying the previous techniques. This work suggests such novel descriptor (in light variations) by updating the methodology of OC-LBP. In OC-LBP, the difference among orthogonal neighbors and center pixel (of respective group) is thresholded to 1 or 0 according to thresholding function, which leads to 2 OLBP codes by putting binomial weights and adding values. Then both codes are concatenated to form OC-LBP size. The major drawback of OC-LBP is the thresholding of only Sign (S) component. This limits discriminativity of OC-LBP. To build more productive descriptor so called Complete (C) OC-LBP (COC-LBP), the Magnitude (M) details are incorporated with the S component of OC-LBP. The M based OC-LBP is called as MOC-LBP and S based is called as SOC-LBP. In MOC-LBP, the absolute of differences among orthogonal neighbors and center pixel (of respective group) is thresholded to 1 or 0 according to thresholding function. In thresholding function this time mean is utilized for comparison rather than center pixel, which is used in SOC-LBP. The two evolved MOLBP codes are concatenated to form the MOC-LBP size. Finally SOC-LBP and MOC-LBP codes are concatenated to build the COC-LBP size. Moving constantly in all positions builds the SOC-LBP and MOC-LBP images. The region feature concatenation of SOC-LBP and MOC-LBP makes COC-LBP size. The huge feature size is compressed by FLDA and matching is assisted from SVMs. On 2 benchmark datasets YB and EYB, COC-LBP marks its influence by defeating the results of many descriptors.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134275950","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 : 2021-12-01DOI: 10.1109/UEMCON53757.2021.9666731
Kashav Piya, Quynh Anh Au, Srijal Shrestha, Apoorva Singh, Tauheed Khan Mohd
In the twenty-first century, Internet of Things (IoT) has boomed to be impossible to not to relate to. From computers, mobile phones, smart watches or your virtual assistants, everything is connected to the complex web of IoT. With the technological development in this century, healthcare sector has also had its fair share of innovations including innovation related to healthcare, diagnostics, sensors, or even data analytic along with development in Machine Learning and Artificial Intelligence. IoT in healthcare is expected to help improve the quality of life for the people due to the flexibility that it provides. We looked at the significant amount of work done in this sector, and found a trend that COVID-19 has only spurred the development of this field. We explored both the benefits and challenges that IoT in Healthcare might possess and have addressed challenges such as data security, cloud integration problems. Additionally, we looked at the issue of data security in depth and along with the current method of LSCSH being used, we have recommended that Zero-trust architecture being used as well as it caters to the needs of the evolving data management space. From 2013, IoT in healthcare has dominated the space as new devices are released and there has been a boom of devices that are being developed and being researched on. Most of the devices are affordable and some can even be accessed with a small subscription fee. Our results show us that IoT has started to pick up its popularity since 2015 as companies succeeded to win the trust of the market. Even though most of the development for these technologies look to come in the future, the progress made in the recent years look promising.
{"title":"IoT in Health Care Industry: A Promising Prospect","authors":"Kashav Piya, Quynh Anh Au, Srijal Shrestha, Apoorva Singh, Tauheed Khan Mohd","doi":"10.1109/UEMCON53757.2021.9666731","DOIUrl":"https://doi.org/10.1109/UEMCON53757.2021.9666731","url":null,"abstract":"In the twenty-first century, Internet of Things (IoT) has boomed to be impossible to not to relate to. From computers, mobile phones, smart watches or your virtual assistants, everything is connected to the complex web of IoT. With the technological development in this century, healthcare sector has also had its fair share of innovations including innovation related to healthcare, diagnostics, sensors, or even data analytic along with development in Machine Learning and Artificial Intelligence. IoT in healthcare is expected to help improve the quality of life for the people due to the flexibility that it provides. We looked at the significant amount of work done in this sector, and found a trend that COVID-19 has only spurred the development of this field. We explored both the benefits and challenges that IoT in Healthcare might possess and have addressed challenges such as data security, cloud integration problems. Additionally, we looked at the issue of data security in depth and along with the current method of LSCSH being used, we have recommended that Zero-trust architecture being used as well as it caters to the needs of the evolving data management space. From 2013, IoT in healthcare has dominated the space as new devices are released and there has been a boom of devices that are being developed and being researched on. Most of the devices are affordable and some can even be accessed with a small subscription fee. Our results show us that IoT has started to pick up its popularity since 2015 as companies succeeded to win the trust of the market. Even though most of the development for these technologies look to come in the future, the progress made in the recent years look promising.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"109 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132881709","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 : 2021-12-01DOI: 10.1109/UEMCON53757.2021.9666529
M. I. Nashiruddin, Pinasthika Aulia Fadhila, M. Nugraha
Spectrum scarcity is one of the challenges faced by operators in Indonesia. As a result, research needs to be conducted using 700 MHz. However, only a tiny number of research has been conducted on the 700 MHz band's performance in urban areas. The purpose of this study is to determine the feasibility of implementing the 700 MHz spectrum in an urban region, specifically North Jakarta, by analyzing capacity and coverage planning and simulating network deployment using Forsk Atoll software. According to the planned calculations, North Jakarta requires 179 eNodeB sites. Through simulation, the Reference Signal Received Power (RSRP) value is −40.41 dBm, the throughput is 11.11 Mbps, the radio bearer is 4.78 with a 1.48 efficiency, and the Signal Interference Noise Ratio (SINR) is 1.61 dB. These findings show that deploying 700 MHz for a 4G LTE-Advanced network in the North Jakarta region can relieve the frequency spectrum scarcity issue.
{"title":"4G LTE-Advanced Network Planning and Simulation Study in an Urban Region Utilizing the 700 MHz Frequency","authors":"M. I. Nashiruddin, Pinasthika Aulia Fadhila, M. Nugraha","doi":"10.1109/UEMCON53757.2021.9666529","DOIUrl":"https://doi.org/10.1109/UEMCON53757.2021.9666529","url":null,"abstract":"Spectrum scarcity is one of the challenges faced by operators in Indonesia. As a result, research needs to be conducted using 700 MHz. However, only a tiny number of research has been conducted on the 700 MHz band's performance in urban areas. The purpose of this study is to determine the feasibility of implementing the 700 MHz spectrum in an urban region, specifically North Jakarta, by analyzing capacity and coverage planning and simulating network deployment using Forsk Atoll software. According to the planned calculations, North Jakarta requires 179 eNodeB sites. Through simulation, the Reference Signal Received Power (RSRP) value is −40.41 dBm, the throughput is 11.11 Mbps, the radio bearer is 4.78 with a 1.48 efficiency, and the Signal Interference Noise Ratio (SINR) is 1.61 dB. These findings show that deploying 700 MHz for a 4G LTE-Advanced network in the North Jakarta region can relieve the frequency spectrum scarcity issue.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125945538","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 : 2021-12-01DOI: 10.1109/uemcon53757.2021.9666517
Deyby Huamanchahua, Luis Soto-Valqui
In the present research work, a Home Automation System is implemented for the geriatric sector because the rate of adults increases, and with it, the increase in geriatric problems that the patient could acquire becomes a problem that society must attend to. The automation system consists of control by voice commands, image monitoring, monitoring of environmental signals (temperature, humidity, and gas), and physiological signals (blood pressure, heart rate, and body temperature). The monitoring system in real-time and remotely by them, it was decided to incorporate the Internet of Things, Cloud Computing and opt for embedded controllers. The results show that the processes are carried out with minimal latency and response in milliseconds in each request on the internet network. So that the user (geriatric patient) can find himself in an environment of comfort and well-being that allows him to easily incorporate himself into his environment and carry out activities independently.
{"title":"Development and Implementation of a Home Automation-Type Real Estate Automation Control System-Oriented and Applied to the Elderly Using IoT","authors":"Deyby Huamanchahua, Luis Soto-Valqui","doi":"10.1109/uemcon53757.2021.9666517","DOIUrl":"https://doi.org/10.1109/uemcon53757.2021.9666517","url":null,"abstract":"In the present research work, a Home Automation System is implemented for the geriatric sector because the rate of adults increases, and with it, the increase in geriatric problems that the patient could acquire becomes a problem that society must attend to. The automation system consists of control by voice commands, image monitoring, monitoring of environmental signals (temperature, humidity, and gas), and physiological signals (blood pressure, heart rate, and body temperature). The monitoring system in real-time and remotely by them, it was decided to incorporate the Internet of Things, Cloud Computing and opt for embedded controllers. The results show that the processes are carried out with minimal latency and response in milliseconds in each request on the internet network. So that the user (geriatric patient) can find himself in an environment of comfort and well-being that allows him to easily incorporate himself into his environment and carry out activities independently.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130496491","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 : 2021-12-01DOI: 10.1109/uemcon53757.2021.9666729
Hosam Alamleh, Kevin Waters, Baker Al Smadi
In the two decades, there have been rapid advancements in the field of communication and information technology. Today, infrastructures include high-speed broadband networks that serve users at a fixed location which is referred to as fixed broadband networks. Another type serves users on the move which is referred to as mobile broadband networks. Fixed broadband networks offer fixed speed and reliable connections. Meanwhile, mobile broadband networks offer mobility, but they are less reliable. Therefore, some of the critical operations such as system update require users to be connected to a fixed broadband network. Different type of networks results in different type of traffic pattern. This paper utilizes a machine learning model at the server-side to help servers differentiate between data transmitted over fixed networks and data transmitted over mobile networks. Supervised training was used to build the model. The proposed system was tested and it showed an accuracy of 92.24 percent. This work is novel and the first of its kind since it is the first to attempt the detection of the nature of the network used for transmission based on the pattern of the arrival of packets.
{"title":"Server-Side Distinction of Incoming Traffic Transmission Medium Using Machine Learning","authors":"Hosam Alamleh, Kevin Waters, Baker Al Smadi","doi":"10.1109/uemcon53757.2021.9666729","DOIUrl":"https://doi.org/10.1109/uemcon53757.2021.9666729","url":null,"abstract":"In the two decades, there have been rapid advancements in the field of communication and information technology. Today, infrastructures include high-speed broadband networks that serve users at a fixed location which is referred to as fixed broadband networks. Another type serves users on the move which is referred to as mobile broadband networks. Fixed broadband networks offer fixed speed and reliable connections. Meanwhile, mobile broadband networks offer mobility, but they are less reliable. Therefore, some of the critical operations such as system update require users to be connected to a fixed broadband network. Different type of networks results in different type of traffic pattern. This paper utilizes a machine learning model at the server-side to help servers differentiate between data transmitted over fixed networks and data transmitted over mobile networks. Supervised training was used to build the model. The proposed system was tested and it showed an accuracy of 92.24 percent. This work is novel and the first of its kind since it is the first to attempt the detection of the nature of the network used for transmission based on the pattern of the arrival of packets.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129449225","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 : 2021-12-01DOI: 10.1109/UEMCON53757.2021.9666695
Tanjir Arafat, Md. Anis Uz Zaman, Mohammad Monirujjaman Khan, Rajesh Dey, Ipseeta Nanda, Debasis Mukherjee
In this paper the development of a voltage producing smart wheelchair is presented. This wheel chair can use the rotational kinetic energy generated from wheel spin to charge a power bank. In this wheelchair a motor-controller is connected with an Arduino. It receives power from source and controls the motors connected with the wheels according to the signals given by Arduino mega. A dynamo is connected with the wheels to convert the rotational kinetic energy generated from wheel spins. This wheelchair will help us to save the kinetic energy generated from wheel spin and use this energy to recharge a power bank. Features like heart beat monitoring system and emergency calling system are also added in this wheel chair.
{"title":"A Voltage Producing Smart Wheelchair Development with Heartbeat Monitoring System","authors":"Tanjir Arafat, Md. Anis Uz Zaman, Mohammad Monirujjaman Khan, Rajesh Dey, Ipseeta Nanda, Debasis Mukherjee","doi":"10.1109/UEMCON53757.2021.9666695","DOIUrl":"https://doi.org/10.1109/UEMCON53757.2021.9666695","url":null,"abstract":"In this paper the development of a voltage producing smart wheelchair is presented. This wheel chair can use the rotational kinetic energy generated from wheel spin to charge a power bank. In this wheelchair a motor-controller is connected with an Arduino. It receives power from source and controls the motors connected with the wheels according to the signals given by Arduino mega. A dynamo is connected with the wheels to convert the rotational kinetic energy generated from wheel spins. This wheelchair will help us to save the kinetic energy generated from wheel spin and use this energy to recharge a power bank. Features like heart beat monitoring system and emergency calling system are also added in this wheel chair.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129464148","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 : 2021-12-01DOI: 10.1109/UEMCON53757.2021.9666614
Rahat Bin Aziz Tonoy, Mahmudunnabi, Hasan Mohiuddin Zilany, Rashedur M. Rahman
The use of IoT devices is incredibly popular nowadays. IoT technology helps us to innovate more secure and smart devices. For the past few years, the concept of a safe and smart home has gotten a lot of attention. A home automation system is a collection of smart devices and sensors that enable various functions within a house or building, such as light, fan, and plug control, energy monitoring, temperature metering, air conditioning, heating, etc. These devices are also implemented with low power communication as well as better quality at a lower price using the internet or Bluetooth connection. The main goal of this research is to create a clever, safe, and intelligent system using IoT devices that will create an automatic home environment. This paper gives a reliable, secure, and decentralized way for recognizing and verifying IoT gadgets in the proposed home automation framework, where the SVM classifier is utilized to separate the situation with the gadgets into two classifications, e.g., "ON" and "OFF". The proposed system design is tailored to the needs of the user and offers great reliability.
{"title":"A Smart and Intelligent Home Automation System","authors":"Rahat Bin Aziz Tonoy, Mahmudunnabi, Hasan Mohiuddin Zilany, Rashedur M. Rahman","doi":"10.1109/UEMCON53757.2021.9666614","DOIUrl":"https://doi.org/10.1109/UEMCON53757.2021.9666614","url":null,"abstract":"The use of IoT devices is incredibly popular nowadays. IoT technology helps us to innovate more secure and smart devices. For the past few years, the concept of a safe and smart home has gotten a lot of attention. A home automation system is a collection of smart devices and sensors that enable various functions within a house or building, such as light, fan, and plug control, energy monitoring, temperature metering, air conditioning, heating, etc. These devices are also implemented with low power communication as well as better quality at a lower price using the internet or Bluetooth connection. The main goal of this research is to create a clever, safe, and intelligent system using IoT devices that will create an automatic home environment. This paper gives a reliable, secure, and decentralized way for recognizing and verifying IoT gadgets in the proposed home automation framework, where the SVM classifier is utilized to separate the situation with the gadgets into two classifications, e.g., \"ON\" and \"OFF\". The proposed system design is tailored to the needs of the user and offers great reliability.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"9 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120917437","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 : 2021-12-01DOI: 10.1109/UEMCON53757.2021.9666675
Sofian Abbasi, Soleil Gignac, Hakduran Koc
Monitoring and maintaining the air quality in confined spaces have become crucial in recent years due to various reasons. This paper presents the design and implementation of an atmospheric testing device for confined spaces using FPGAs. Specifically, the device connects multiple gas sensors such as oxygen, methane, and nitrogen dioxide to an FPGA architecture that outputs a warning signal to the user in the field. Field usage warnings include a camera for image capturing, a buzzer for audio feedback, an LED for visual feedback, and a display for the gas level in the space. The device also outputs to Matlab for graphing to illustrate the accuracy of the sensors. The project seeks to design, simulate, and implement an expandable device that monitors different gases within a confined space with various safety levels and respective warning systems.
{"title":"An FPGA Based Multi-Sensor Atmospheric Testing Device for Confined Spaces","authors":"Sofian Abbasi, Soleil Gignac, Hakduran Koc","doi":"10.1109/UEMCON53757.2021.9666675","DOIUrl":"https://doi.org/10.1109/UEMCON53757.2021.9666675","url":null,"abstract":"Monitoring and maintaining the air quality in confined spaces have become crucial in recent years due to various reasons. This paper presents the design and implementation of an atmospheric testing device for confined spaces using FPGAs. Specifically, the device connects multiple gas sensors such as oxygen, methane, and nitrogen dioxide to an FPGA architecture that outputs a warning signal to the user in the field. Field usage warnings include a camera for image capturing, a buzzer for audio feedback, an LED for visual feedback, and a display for the gas level in the space. The device also outputs to Matlab for graphing to illustrate the accuracy of the sensors. The project seeks to design, simulate, and implement an expandable device that monitors different gases within a confined space with various safety levels and respective warning systems.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"218 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124296991","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 : 2021-12-01DOI: 10.1109/uemcon53757.2021.9666725
Joseph M. Carmack, Amit Bhatia, Josh Robinson, J. Majewski, Scott Kuzdeba
Neural Networks (NN) provide great flexibility in modeling non-linear relationships and have recently proved to be very valuable in Radio Frequency (RF) domain, e.g. NN based Digital Signal Processing (DSP) provide equal or better performance than traditional DSP function blocks. One key requirement of NN models is their need for large amounts of training data to mitigate model overfitting. If this data requirement is satisfied it leads to models that generalize well, but does not hold in cases of limited training data. Generating high fidelity data can help to mitigate data scarcity. In this paper we demonstrate the application of Generative Adversarial Network (GAN) based RF data generation capability for training interference removing NN models. We developed two GAN models that learn to generate RF data with tone jammers. The first is a "Sequence to Sequence GAN model", which learns the jammer data distribution by learning to map clean RF signals to impaired RF signals with a jammer present. The second is a physics driven "Tone GAN" model, which learns the frequency, phase, and magnitude of the interferer from a limited data distribution and applies this impairment to the clean signal to generate impaired waveforms similar to the GAN training data. Experimental results show that the new data generated from the trained GAN models can be used to augment the original NN model, by fine-tuning them via the GAN generated data, leading to improved performance and generalization.
{"title":"Neural Network Generative Models for Radio Frequency Data","authors":"Joseph M. Carmack, Amit Bhatia, Josh Robinson, J. Majewski, Scott Kuzdeba","doi":"10.1109/uemcon53757.2021.9666725","DOIUrl":"https://doi.org/10.1109/uemcon53757.2021.9666725","url":null,"abstract":"Neural Networks (NN) provide great flexibility in modeling non-linear relationships and have recently proved to be very valuable in Radio Frequency (RF) domain, e.g. NN based Digital Signal Processing (DSP) provide equal or better performance than traditional DSP function blocks. One key requirement of NN models is their need for large amounts of training data to mitigate model overfitting. If this data requirement is satisfied it leads to models that generalize well, but does not hold in cases of limited training data. Generating high fidelity data can help to mitigate data scarcity. In this paper we demonstrate the application of Generative Adversarial Network (GAN) based RF data generation capability for training interference removing NN models. We developed two GAN models that learn to generate RF data with tone jammers. The first is a \"Sequence to Sequence GAN model\", which learns the jammer data distribution by learning to map clean RF signals to impaired RF signals with a jammer present. The second is a physics driven \"Tone GAN\" model, which learns the frequency, phase, and magnitude of the interferer from a limited data distribution and applies this impairment to the clean signal to generate impaired waveforms similar to the GAN training data. Experimental results show that the new data generated from the trained GAN models can be used to augment the original NN model, by fine-tuning them via the GAN generated data, leading to improved performance and generalization.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115616239","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 : 2021-12-01DOI: 10.1109/UEMCON53757.2021.9666535
Shreehar Joshi, Eman Abdelfattah, Shreekar Tiwari
Police killings is one of the most controversial topics of discussion in the United States. Every death that occurs in police shootings is put under the microscope. Almost always, these deaths are categorized as the result of police officers being biased against a certain community or they are acting out of desperation for their own protection, depending on which side of the argument is favored. As will be evident from this work, the answer, and the possible solutions, are not straightforward enough to side with one reason. This research attempts at breaking down the data related to police killings obtained through the Washington Post from the beginning of 2015 to the second quarter of 2020. The motive of this research is to have a better insight into the patterns of police killings. The data have been interpreted with respect to sex, age and race and also the number of police killings in every state. The paper also reveals the weapons that are the most favored used by the suspects involved in the police killings. A graph showing the fleeing status of the suspects and whether they were captured during their killings concludes our findings.
{"title":"Analyzing Police Shootings in the US","authors":"Shreehar Joshi, Eman Abdelfattah, Shreekar Tiwari","doi":"10.1109/UEMCON53757.2021.9666535","DOIUrl":"https://doi.org/10.1109/UEMCON53757.2021.9666535","url":null,"abstract":"Police killings is one of the most controversial topics of discussion in the United States. Every death that occurs in police shootings is put under the microscope. Almost always, these deaths are categorized as the result of police officers being biased against a certain community or they are acting out of desperation for their own protection, depending on which side of the argument is favored. As will be evident from this work, the answer, and the possible solutions, are not straightforward enough to side with one reason. This research attempts at breaking down the data related to police killings obtained through the Washington Post from the beginning of 2015 to the second quarter of 2020. The motive of this research is to have a better insight into the patterns of police killings. The data have been interpreted with respect to sex, age and race and also the number of police killings in every state. The paper also reveals the weapons that are the most favored used by the suspects involved in the police killings. A graph showing the fleeing status of the suspects and whether they were captured during their killings concludes our findings.","PeriodicalId":127072,"journal":{"name":"2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115205395","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：