=====================================================================================
College of Computing and Mathematics
Computer Engineering Department
=====================================================================================
Presents Public Seminar
1. "Towards Trustworthy and Privacy-Preserving Federated Deep Learning Service Framework for Industrial Internet-of-Things"
Mr. Mumin Omer Adam
2. "Weather Prediction Using regression Neural Network"
Mr. Mohamed Osman Omar
3. "Trust Evaluation Mechanism for User Recruitment in Mobile Crowed-Sensing in the Internet of Things"
Mr. Ali Bello Imoukhuede
4. "LoRA Network Scalability Performance Evaluation"
Mr. Aliyu Habib Abubakar
Date: Wednesday, December 14, 2022,
Time: 02:30 PM – 04:30 PM
Location: Bldg. 22, Room 119
1. "Towards Trustworthy and Privacy-Preserving Federated Deep Learning Service Framework for Industrial Internet-of-Things"
Abstract:
This paper proposes a trustworthy privacy-preserving Federated Learning (FL) based Deep Learning (DL) service framework for Industrial Internet-of-Things (IIoT) enabled systems. FL mitigates the privacy issues of the traditional collaborative learning model by aggregating multiple locally trained models without sharing any datasets among the participants. Nevertheless, the FL-based DL (FDL) model cannot be trusted as it is susceptible to intermediate results and data structure leakage during the model aggregation process. The proposed framework introduces an edge and cloud-powered service-oriented architecture identifying the key components and a service model for Residual Networks based FDL with differential privacy for generating trustworthy locally trained models. The service model decomposes the functionality of the overall FDL process as services to ensure trustworthy execution through privacy preservation. Finally, we develop a privacy-preserving local model aggregation mechanism for FDL. We perform several experiments to assess the performance of the proposed framework.
Mr. Mumin Adam Biography:
Mr. Adam is currently a Ph.D. student. He received his master's degree from Computer Engineering Department, KFUPM University, and his bachelor's degree from Hodeidah University, Yemen. His research interests include the Internet of Things security, AI in cyber security, WSN, and Edge and Cloud Computing
2. "Weather Prediction Using Regression Neural Network"
Abstract:
Over the past few decades, weather forecasting has emerged as a significant area of study. The researcher often tried to create a linear link between the input weather data and the related target data. However, the emphasis has turned to the non-linear prediction of the weather data when non-linearity-like weather data was discovered. So in this project, we are going to analyze the weather data and extract the hypotheses to arrive at a prediction, using some of its parameters. The neural network is first trained and then testing is carried out to ascertain the accuracy of the predictions.
Mr. Mohamed Osman Omar Biography:
Mohamed Osman Omar is a master's student in Computer Engineering. He obtained his bachelor's degree in Telecommunication Engineering from Hormud University, Somalia. His research includes Network Security and IoT.
3. "Trust Evaluation Mechanism for User Recruitment in Mobile Crowd-Sensing in the Internet of Things"
Abstract:
In recent years, the world has experienced a large increase in the number of devices that are connected to the internet. Leveraging this has created room for large applications of the Internet of Things (IoT). This has in turn increased the need for sensing tasks. Environmental sensors are electronic devices that can sense and obtain a variety of data, including location, position, a person's motion, temperature, and others. One challenge with sensors has always been the issue of mobility. Deploying sensors to capture data from a large environment isn't cost-effective because many sensors would be needed. The cost of maintenance will also be high because sensors usually have limited resources. This is where mobile crowdsensing comes in. A sensing paradigm known as "mobile crowdsensing" allows regular people to participate by contributing data that is generated or detected by their mobile devices with added sensors. With mobile devices carried around by millions of users around the world, a huge amount of data can be collected and used for real-time analysis among other possible applications. In our work, we investigate how to recruit users for mobile crowd-sensing tasks using a trust-experience mechanism. This mechanism ensures that only users who will provide high-quality data are recruited.
Mr. Ali Bello Imoukhuede Biography:
Ali Bello Imoukhuede is a Computer Engineering master's student who earned his bachelor's degree from Bayero University Kano, Nigeria. His research interests include IoT and Network Automation.
4. "LoRa Network Scalability Performance Evaluation"
Abstract:
LoRa WAN network robustness in communication and ability to cover a wide range of areas leads to its deployment in many areas like smart buildings, health, and military, so the scalability of the network tends to be of concern. This project investigates the effect of the number of deployed LoRa nodes and that of gateways in an area on the efficiency of the network and the bit rate (Rb). It also investigates the effect of terrain size, that is the deployment size, on transmission range. Our result shows that the increase in the number of nodes leads to a decrease in the packet delivery ratio (PDR), the number of gateways has no significant effect on PDR, and both the number of nodes and gateways have no effect on the bit rate but spreading factor (SF) has. A terrain size of less than 2000x2000 m2 or a distance of less than 2km is recommended to be used with the settings described in this project for all nodes to be reachable. We then conclude that LoRa networks scale well up to 700 nodes with a PDR approximately equal to 100%, PDR is not affected by the number of gateways and terrain size can affect transmission range.
Mr. Aliyu Habib Abubakar Biography:
Aliyu Habib Abubakar is a master's student in Computer Networks, COE. He got his bachelor's degree in Computer Engineering from Bayero University, Kano, Nigeria. His research interests include is in Hardware security and logic locking.
All faculty, researchers and graduate students are invited to attend.
=====================================================================================
Computer Engineering Department, College of Computing and Mathematics
Telephone: +966 (13) 860 2110, Email: c-coe@kfupm.edu.sa, Website: www.kfupm.edu.sa/departments/coe/
Copyright © 2014 King Fahd University of Petroleum & Minerals
=====================================================================================
