Mohammadnavid Ghader, Bahar Farahani, Z. Rezvani, Mahyar Shahsavari, M. Fazlali
{"title":"Exploiting Federated Learning for EEG-based Brain-Computer Interface System","authors":"Mohammadnavid Ghader, Bahar Farahani, Z. Rezvani, Mahyar Shahsavari, M. Fazlali","doi":"10.1109/COINS57856.2023.10189325","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189325","url":null,"abstract":"Motor imagery (MI) is a widely used technique in brain-computer interface (BCI) systems, which allows users to control external devices using their brain signals. Electroen-cephalogram (EEG) signals are commonly used to detect and classify MI tasks. However, the lack of annotated data hampers the performance of machine learning (ML) algorithms. Federated learning (FL) is a promising approach to address these challenges, as it allows models to be trained on decentralized datasets without exchanging data. This study proposes an FL approach for MI-EEG signal classification using a convolutional neural network (CNN) on the PhysioNet dataset containing EEG recordings of left and right-hand imagery movements. We evaluate the performance of the FL approach using two different aggregation methods, namely FedAvg and FedProx, and compare it to the centralized ML approach. Furthermore, we explore the effect of increasing the number of clients who participate in the learning process on the performance of the model. Our findings demonstrate that FL maintains consistent classification accuracy comparable to the centralized ML approach while reducing data leakage risks. Thus, FL shows great promise as an essential instrument for MI-EEG signal classification and BCI systems, enabling distributed training of a comprehensive model when privacy-sensitive data is dispersed across multiple clients.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133424603","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}
Artem Barger, V. Gorgadze, A. Belenov, Ekaterina Frolovicheva
{"title":"Atomyze - The Scalable Hyperledger Fabric-Powered Blockchain Platform Revolutionizing Asset-Backed Tokenization","authors":"Artem Barger, V. Gorgadze, A. Belenov, Ekaterina Frolovicheva","doi":"10.1109/COINS57856.2023.10189252","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189252","url":null,"abstract":"In recent years tokenization of different assets has become widespread in the crypto industry. There are many examples of tokenization platforms that are increasingly attracting new participants. However, the architecture of these platforms is often concealed from a wide range of possible users, which does little to contribute to the idea of transparency and discoverability. The Atomyze platform offers a working solution for tangible assets tokenization based on the Hyperledger Fabric permissioned blockchain technology with the ability to connect validators. Atomyze provides a flexible architecture and easily scalable multichannel approach in token development as well as solves complicated problems of improving the quality and speed of transaction processing using batching and randomizer to deal with the endorsement selection by adding randomness to endorsement rules as well as the BFT consensus for high throughput, low transaction costs, and more scalability.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"58 11","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132575213","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}
Amani Ibraheem, Zhengguo Sheng, G. Parisis, Daxin Tian
{"title":"Network Tomography-based Anomaly Detection and Localisation in Centralised In-Vehicle Network","authors":"Amani Ibraheem, Zhengguo Sheng, G. Parisis, Daxin Tian","doi":"10.1109/COINS57856.2023.10189258","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189258","url":null,"abstract":"The new automotive Electrical/Electronic (E/E) architecture is shifting towards a new design of in-vehicle network that is based on a centralised, cross-domain architecture. Such architecture implies communication between different domains of the vehicle network. From security standpoint, such cross-traffic can easily be exploited by adversaries to gain access to different system domains, including the safety-critical ones, and perform attacks that may result in serious consequences. Accurate detection and localisation of these anomalies is important in such critical systems where false alarms cannot be tolerated. To this end, in this work, we propose an anomaly detection and localisation approach using network tomography-based monitoring solution. Compared to existing solutions, network tomography approaches require only limited number of probes and do not necessitate direct access to the vehicle's networking devices. In this work, we evaluate three types of network tomography (binary tomography, delay tomography, and deep learning-based tomography) to detect and locate anomalies in in-vehicle networks. The results show that binary tomography can accurately detect and locate Denial-of-Service (DoS) attacks in centralised in-vehicle networks.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114703284","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}
{"title":"Performance Comparison of Post-Quantum Signature Algorithms Through An Android Email Application Plug-in","authors":"Radoslav Mandev, E. Kavun","doi":"10.1109/COINS57856.2023.10189254","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189254","url":null,"abstract":"Public key cryptography and signature schemes are essential for reliable digital communication. Asymmetric encryption has been achieved through complex algorithms, such as RSA and Diffie-Hellman, but recent advancements in quantum computing pose threats to the security of these algorithms in the near future as a quantum computer may solve the problems these algorithms are based on. Therefore, Post-Quantum Cryptography (PQC) algorithms have been proposed and investigated by an international community of scientists, especially through the NIST PQC competition ongoing since 2016. They have gathered and classified four different key exchange algorithms and three signature schemes. This paper focuses on implementing the signature schemes inside a popular open-source Android email application as a plug-in and comparing their performances based on this app plug-in utilization. We have used a library, liboqs, from the Open Quantum Safe project as the starting point, which ensures regularity in our results. We also compare the current standard candidate signature schemes with the currently used standard RSA-4096. Our results show that Dilithium has noticeably faster key generation, signing, and verification with the relatively affordable key and signature size. Also, to the best of our knowledge, our study provides the first practical Android email application plug-in supporting PQ-based signatures.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114896736","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}
{"title":"Quantity Analysis of Solar Modules Based on Aerial Images","authors":"Chao-Wei Ou, Cheng-Yu Peng, Han-Chang Liu, Guo-Shing Huang, Chi-Chun Chen, Meng-Hua Yen","doi":"10.1109/COINS57856.2023.10189306","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189306","url":null,"abstract":"This paper proposes a system evaluation to calculate the number of photovoltaic(PV) modules in an aerial image to help the acceptance inspection of large-area PV module installation and PV module defects. First, the aerial image of the PV module must be converted to a grayscale image and then binarized. Extract the PV module from the background. Next, morphological erosion operations are used to remove noise and enhance the edge profile of the PV module. Multiple PV module arrays are then segmented from the image by pattern-matching. Finally, the edges of the PV modules are identified using Hough transformation. The total number of PV modules is calculated by finding the relationship between the edge number and the module number The study method determines the X-axis direction by self-defining two points in the aerial image. And use one point as the positioning origin. Calculate the coordinate information of each module. In the experiment, we rotate the image at several angles for detection. We discussed the discrepancy between the number of PV modules in the image and the number of PV modules calculated by the system.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117255261","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}
{"title":"Chatbot Integration for Metaverse - A University Platform Prototype","authors":"Qitao Xie, Wenxi Lu, Qingquan Zhang, Lingxiu Zhang, Ting Zhu, Jianwu Wang","doi":"10.1109/COINS57856.2023.10189232","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189232","url":null,"abstract":"The integration of chatbots in a university platform prototype for a metaverse has the potential to significantly enhance the user experience by providing instant, interactive, and personalized support. Chatbots can assist both students and faculty with a variety of tasks, including course navigation, assignment submissions, and answering frequently asked questions. By utilizing natural language processing (NLP) and machine learning algorithms, the chatbot is able to understand and respond to user inquiries with a high degree of accuracy. This integration has the added benefit of increasing the efficiency of information dissemination, reducing the workload of administrative staff. The use of a metaverse in the university platform creates an immersive and interactive virtual environment, making education more engaging and accessible. In this paper, we will highlight the representative applications for chatbot integration for metaverse on a university platform. To achieve this, we propose a three-layer metaverse architecture from a macro perspective, which includes a chatbot platform, a university platform, and a metaverse platform. We will illustrate our implemented metaverse prototype of a university campus and discuss the prototype design and insights for educational purposes. Overall, the integration of chatbots in a university platform prototype for a metaverse provides a promising avenue for enhancing the user experience and facilitating education in the digital age.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116446064","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}
{"title":"Secure Decentralized Learning with Incentives on UTXO-based Blockchains","authors":"Luigi Lunardon","doi":"10.1109/COINS57856.2023.10189235","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189235","url":null,"abstract":"Increasing computational and economic costs in training artificial intelligence motivate the development of alternative decentralized training paradigms. Federated learning achieves decentralization by creating the final model combining local models trained on edge devices. This process does not require sharing local private data with the central service. The blockchain can be used to distribute and collect edge models. The benefits of relying on the blockchain are a native way to provide economic incentives to the local devices and the ability to log events and track training steps. The former incentivizes honest behavior, while the latter enables auditing and testing of the model. We present a peer-to-peer cloud protocol to outsource computations using an unspent transaction output blockchain relying on proof of work. In designing this protocol, we face three main challenges: how to communicate the computation to the computing party, how to make its execution safe for them, and how to ensure the computing party executed it faithfully. Our solution introduces a verification script in the outsourced computation. The computing party receives the computation as an obfuscated executable and provides the verification output to show they faithfully executed it. Data integrity and safety in the execution rely on proof of existence.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129987202","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}
{"title":"Lightweight Micro Aerial Vehicles (MAVs) Application for Indoor Environment Mapping using Ultrasonic Sensors","authors":"M. Conceição, A. Grilo, M. Basiri","doi":"10.1109/COINS57856.2023.10189202","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189202","url":null,"abstract":"Indoor localisation and mapping missions usually require several sensors to correctly calculate the position of the Unmanned Aerial Vehicle (UAV) within the space and detect and localize obstacles. This research paper presents the development of a mapping system for UAVs in unknown indoor environments resorting to ultrasonic sensors. The core of the system is the DJI Ryze Tello UAV, a small 80g Micro Air Vehicle (MAV) with very limited capabilities such as payload and available flight time. A distributed system was developed to ensure the successful execution of the mission without exceeding the capabilities of the MAV. A prototype was developed that consists of two hardware components, one integrated into the MAV and another in a teth-ered configuration. The operation of the system is independent of the environment and supports multiple simultaneous missions. An original autonomous flight and mapping algorithm was created, based on vertical motion patterns, to provide an efficient and effective mapping solution. Data processing procedures were also implemented to complement the initial framing of detected obstacles. The obtained system allows the digital reconstruction of the space. The mapped obstacle distances in the test environment have resulted in more than 90% of accuracy.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"135 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131714850","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}
E. Ogliari, Maciej Sakwa, Jianguo Wei, B. Schubert, Weilin Liu, M. Palo
{"title":"Partial Discharge UHF detection and classification with Machine Learning: a real case study","authors":"E. Ogliari, Maciej Sakwa, Jianguo Wei, B. Schubert, Weilin Liu, M. Palo","doi":"10.1109/COINS57856.2023.10189203","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189203","url":null,"abstract":"This paper discusses the results of applying a robust Machine Learning methodology in a classification procedure of Ultra High-Frequency radiation pulses emitted by Partial Dis-charge events. The proposed method employs Machine Learning algorithms in order to achieve high accuracy in filtering the incoming interference and low compilation times in order to potentially exploit the procedure on edge sensors. In the studied case two different classification models have been compared an Artificial Neural Network (as the principal model) and a Random Forest ensemble (as a benchmark for comparison). The results of the tests show a high potential for utilization of the neural networks for PD classification problems as the achieved global accuracy exceeds 90% for both tested models. The achieved time complexity on modern hardware shows potential for implementation of the procedure on edge sensors as filtering software.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126733665","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}
{"title":"From Centralized to Decentralized: A DID-based Authentication Concept in Forestry 4.0","authors":"Jiahang Chen, Anil Riza Bektas, J. Roßmann","doi":"10.1109/COINS57856.2023.10189199","DOIUrl":"https://doi.org/10.1109/COINS57856.2023.10189199","url":null,"abstract":"The ability of Internet of Things (IoT) systems to authenticate users and devices is regarded as a crucial defense mechanism in today's IoT cybersecurity. It serves as a basis for the confidentiality, integrity, and availability of data and services in the IoT world. In European forest-based sector, a number of cutting-edge digital solutions from Industry 4.0 are being integrated to establish new concepts for digitized and value-added chains in various fields. In this context, the term Forestry 4.0 has been derived. There, authentication is a keyword to ensure IoT security. Decentralized authentication solutions provide several advantages over a centralized one, such as avoiding a single point of failure, enabling cross-domain usage, and ensuring data sovereignty. Nevertheless, the forest-based sector has not adopted any decentralized authentication techniques due to the particular nature of this sector until now. Hence, in this paper, we discuss the potential of using Decentralized Identifiers (DIDs) in Forestry 4.0. In addition, we propose our decentralized authentication architecture using the W3C DID core model. We also introduce the concept of self-sovereign Things, which are distinct classes with varying levels of autonomy over their identity and data-access policies.","PeriodicalId":328385,"journal":{"name":"2023 IEEE International Conference on Omni-layer Intelligent Systems (COINS)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121426008","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}