top of page

 

NARRATIVE ON RESEARCH/CREATIVE ACTIVITIES

 

The realization of IoT wireless connectivity is bringing fundamental changes to communications and computing and profoundly affects the way we compute, communicate, and interact. It provides fully distributed and ubiquitous computing and communications, thus bringing an end to the tyranny of geography. The IoT wireless and mobile networks are quickly becoming the networks of choice, not only because of the large bandwidth but due to the flexibility and freedom they offer.

 

My interdisciplinary research interest includes the design, analysis, and deployment of the following.

  1. Network and Communication Sciences

  2. Health Informatics

  3. Data/Customer Analytics and Information Systems

  4. Churn Prediction

  5. Network Security

  6. Machine and Deep Learning

  7. IoT & Social Internet

​

​

Current and Future Research Agenda

 

Health Informatics

With the rapid advancements of wireless communication and semiconductor technologies such as wearable, health monitoring, etc., the area of sensor networks has grown significantly by altering the future of healthcare services and enabling ubiquitous monitoring of patients. A Wireless Body Area Network (WBAN) is a special-purpose sensor network designed to operate autonomously by connecting various independent medical nodes (e.g., sensors and actuators) situated in the clothes, stick on the human body, or under the skin. These nodes are spanned and scattered all over the human body and connected via a wireless communication channel using a star or multi-hop topology.

In many developed countries, the healthcare systems are currently confronting an increase in the number of people diagnosed with chronic diseases such as obesity, diabetes, cardiovascular diseases, and cancer. These chronic illnesses are not simply a result of an aging population but are due to inappropriate diet, sedentary lifestyle and insufficient physical activity. Consequently, the need to provide quality care and service in these countries for a rapidly growing population of elderly people, while reducing the health care costs, is of paramount importance for governments and health service providers. Wearable and implantable body sensor network systems are apparatus to achieve this objective, (i.e., a prominent application in this area is the integration of sensing and consumer electronics technologies, in order to continuously monitor people during their daily activities). Standardized hardware and software architectures can support compatible devices, which are expected to significantly affect the next generation of healthcare systems. Some of these devices can then be incorporated into the WBAN, providing new opportunities for technology to monitor health status.

In our second Health Informatics project, we are implementing a more medically intuitive clinical decision support system that helps physicians, patients, and caregivers to manage Alzheimer’s disease. The proposed system will save lots of resources spent annually to diagnose, treat, and monitor Alzheimer patients globally and especially in the UAE. Furthermore, the proposed system will be medically oriented in the system design and extend the state-of-art techniques of machine learning and deep learning.

To sum up, the criticality of this project can be summarized in the following points:

  1. Building a patient-centric and web-based recommendation system to help Alzheimer’s patients manage their disease and predict their future conditions.

  2. Designing and implementing medically trustful and personalized clinical decision support systems to help physicians automatically manage and monitor Alzheimer’s disease patients.

  3. Designing and implementing a real-time patient monitoring system to continuously provide assistance to Alzheimer’s patients.

​

Anchor 1

Sensor and Communication Technologies for IoT

I am conducting research across the key technological areas of the IoT including 1) the design of sensor platforms and tools for heterogeneous sensor technologies and 2) context-aware and user-centric resource management schemes for future IoT systems. For sensor platform design, the aim is to provide intelligence and flexibility to sensor nodes in order to adapt them to various environments in a controlled or autonomous fashion. This also means that we need to make sensors smart enough by equipping them with software-definable capabilities in an IoT system. Software-defined sensor nodes can be equipped with several different types of sensors and is able to perform various sensing tasks according to the deployed program. The integration of context information and user experience in the IoT system enables easy adaptation and deployment of the technology.

Anchor 5
Anchor 7

Machine Learning / Deep Learning

I am also focusing on modern research topics and continued work in the machine and deep learning. Our recent research on the machine and deep learning is focusing on the Airline industry which has witnessed tremendous growth in the recent past, Agriculture and Health. Highly demanding and congested air routes are resulting in inadvertent delays, additional fuel consumption, and high emission of greenhouse gases. Trajectory planning involves the creation and identification of cost-effective flight plans for optimal utilization of fuel and time. Therefore, we proposed an algorithm for the dynamic planning of optimized flight trajectories. In machine learning, I also focused and continued research on Edge computing which springs up a modern computing platform for the Internet of Things (IoT), smart systems, and multimedia applications. These technologies are built using resource-constrained devices, which are incapable of executing complex tasks. Our research focuses on addressing scalability issues by proposing a state-of-the-art cross-entropy based scalable edge computing framework. The framework comprises IoT devices, edge servers, and the cloud.

 

Load forecasting has received crucial research attention to reduce peak load and contribute to the stability of the power grid using machine learning or deep learning models. Therefore, we started research on hybrid deep learning architecture to forecast maximum load duration based on time of use.

 

A lot of different methods are being opted for improving educational standards through monitoring of online classrooms in this pandemic. The developed world uses Smart classrooms to enhance faculty efficiency based on accumulated learning outcomes and interests. Along with the modern facilities, more effort is required to monitor and analyze students’ outcomes, teachers’ performance, attendance records, and content delivery in on-campus classrooms. One can achieve more improvement in quality teaching and learning outcomes by developing digital twins in on-campus classrooms. Therefore, we are proposing Deep Class-Rooms, a deep learning-based digital twin framework for attendance and course content monitoring for public sector schools.

Anchor 4
Anchor 2

Medical Smart Homes

This project studies a soft robotics system for health monitoring at a home setting. It is a multi-disciplinary effort with competencies from prominent healthcare and medical professionals, IT, IoT and bio-sensors, signal processing, and machine learning. The idea is to use smart sensors to analyse activity using environmental and physical sensors. The activity patterns will be used in combination with physiological signals to characterize cardiovascular diseases as a case study. In addition, the project studies ways of controlling the home environment to implement deconditioning tactics. This is done through an adaptive recommendation system that derives control actions depending on the resident’s profile. Actions can be directed to the physical environment such as light and temperature. But can also be related to social interaction or psycho-cognitive activities such as listening to specific sounds (music, audibles). The overall scope of the research program is health monitoring using smart homes. The project will mainly focus on data acquisition protocols and data analysis methods for healthcare monitoring using smart homes.

Customer Churn

Customer churn is a complex problem for rapidly growing UAE competitive organizations in general, particularly in the telecommunications industry. It refers to customers who abandon the services or even the company more quickly and shift to the next competitor. Traditionally, customer churn prediction (CCP) models are applied to aid in analyzing customer behavior and achieving prediction accuracy, which allows the telecommunication industry to target prior retention efforts toward them. Therefore, it is essential to design an approach that easily adapts to learn from new decision scenarios and provides instant insights. We are proposing a novel technique for this perplexing problem of CCP using an adaptive Naïve Bayes algorithm with a Genetic Algorithm-based feature weighting approach.
 

Anchor 3
Anchor 6

RESEARCH GRANTS

​

Project as PI:

 

  1. [2021-2023]: “Machine Learning-Based Human Body Characteristics-Aware Communication Protocol in UAE”. This project is funded by Research Cluster Fund of Zayed University Abu Dhabi (UAE). 

  2. [2020-2022]: “Countering Malicious URLs in Open Social Network APIs”. This project is funded by Research Incentive Fund of Zayed University Abu Dhabi (UAE). 

  3. [2019-2020]: “Real-Time Stream for Mobile Object Tracking in Wireless Multimedia Sensor Networks”. This project is funded by Research Incentive Fund of Zayed University Abu Dhabi (UAE). 

  4. [2017-2019]: “Communication Protocol for Health Monitoring Over Wireless Body Area Networks in UAE”. This project is funded by Research Cluster Fund of Zayed University Abu Dhabi (UAE). 

  5. [2018-2019]: “An incremental learning approach for countering Cross-Site Scripting attacks in UAE”. This project is funded by RIF Project of Zayed University Abu Dhabi (UAE). 

  6. [2017-2019]: “Securing Critical Cyber Infrastructures for Smart Cities in the UAE”.  This project is funded by Research Cluster Fund of Zayed University Abu Dhabi (UAE). 

  7. [2016-2017]: “A Study on Broadcast and Multicast Communications for (m, k)-firm Streams in Wireless Sensor Networks”. This project is funded by RIF Project of Zayed University Abu Dhabi (UAE). 

  8. [2015-2016]: “Searching architecture for unstructured mobile peer-to-peer networks”. The project is funded by Zayed University, Abu Dhabi (UAE) and is executed in collaboration with Gyeongsang National University, South Korea.  

  9. [2014-2015]: “Fuzzy Logic Model for QoS Real-time in Guaranteed WSN Lifetime”.  The project is funded by Zayed University, Abu Dhabi (UAE). 

​

Project as Co-PI:

 

 

  1. [2020-2021]: “Accident and Crime Detection in Surveillance Videos”. This project is funded by RIF Project of Zayed University Abu Dhabi (UAE). 

  2.  [2020-2021]: “Children’s Privacy Protection Framework for Smart Toys in the UAE”. This project is funded by RIF Project of Zayed University Abu Dhabi (UAE). 

  3. [2018-2020]: “Cluster Proposal: Building Cyber Resilience for Large Scale Critical Network Infrastructures with Data Science”. This project is funded by Research Cluster Fund of Zayed University Abu Dhabi (UAE). 

  4. [2018-2019]: “Energy Management for Internet of Things in Smart Cities”. This project is funded by RIF Project of Zayed University Abu Dhabi (UAE). 

  5. [2017-2018]: “Temperature Models for Energy Efficiency in MENA Buildings”. This project is funded by RIF Project of Zayed University Abu Dhabi (UAE). 

  6.  [2016-2017]: “Data analytics for improving buyer-seller relationships for small and medium enterprises in UAE”. This project is funded by RIF Project of Zayed University Abu Dhabi (UAE). 

  7. [2016-2017]: “Impersonal Models for Recognizing a Wide-Range of Daily Life Activities on Smart Devices”. This project is funded by RIF Project of Zayed University Abu Dhabi (UAE). 

  8. [2016-2018]: “Smart data curation and privacy for IoT enabled devices”. This project is funded by Research Cluster Fund of Zayed University Abu Dhabi (UAE). 

  9. [2015-2016]: “Multi-Model Data Fusion for Healthcare”. The project is funded by Zayed University, Abu Dhabi (UAE). 

Anchor 8
bottom of page