Home

Research @ Computer Engineering

Main block

1. Hardware Module Design and Software Implementation of Multi-Sensor Fire Detection and Notification System using Fuzzy Logic and Convolutional Neural Networks (CNN)
This research project proposed the design and development of a fuzzy logic-based multi-sensor fire detection and a web-based notification system with trained Convolutional Neural Networks for both proximity and wide-area fire detection. Until recently, most consumer-grade fire detection systems relied solely on smoke detectors. These offer limited protection due to the type of fire present and the detection technology at use. To solve this problem, we present a multi-sensor data fusion with Convolutional Neural Networks (CNN) fire detection and notification technology. Convolutional Neural Networks are mainstream methods of deep learning due to their ability to perform feature extraction and classification in the same architecture. The system is designed to enable early detection of fire in residential, commercial, and industrial environments by using multiple fire signatures such as flames, smoke, and heat. The incorporation of the Convolutional Neural Networks enables broader coverage of the area of interest, using visuals from surveillance cameras. With access granted to the web-based system, the Fire and Rescue Crew get notified in real-time with location information. The efficiency of the fire detection and notification system employed by standard fire detectors and the multi-sensor remote-based notification approach adopted in this paper showed significant improvements with timely fire detection, alerting, and response time for firefighting. The final experimental and performance evaluation results showed that the accuracy rate of CNN was 94%, and that of the fuzzy logic unit is 90%.

2. Decision Support System (DSS) for Fraud Detection in Health Insurance Claims Using Genetic Support Vector Machines (GSVMs)
Fraud in health insurance claims has become a significant problem whose rampant growth has deeply affected the global delivery of health services. In addition to financial losses incurred, patients who genuinely need medical care suffer because service providers are not paid on time as a result of delays in the manual vetting of their claims and are therefore unwilling to continue offering their services. Health insurance claims fraud is committed through service providers, insurance subscribers, and insurance companies. The need for the development of a decision support system (DSS) for accurate, automated claim processing to offset the attendant challenges faced by the National Health Insurance Scheme cannot be overstated. This paper utilized the National Health Insurance Scheme claims dataset obtained from hospitals in Ghana for detecting health insurance fraud and other anomalies. Genetic support vector machines (GSVMs), a novel hybridized data mining and statistical machine learning tool, which provides a set of sophisticated algorithms for the automatic detection of fraudulent claims in these health insurance databases, are used. The experimental results have proven that the GSVM possessed better detection and classification performance when applied using SVM kernel classifiers. Three GSVM classifiers were evaluated, and their results compared. Experimental results show a significant reduction in computational time on claims processing while increasing classification accuracy via the various SVM classifiers (linear (80.67%), polynomial (81.22%), and radial basis function (RBF) kernel (87.91%).

3. Ghana TVWS Project 
The continuous growth in wireless data demand and transmission has called for research into alternative technologies to augment the current spectrum schemes. Fortunately, the currently existing vacant spectrum located within the analogue TV broadcasts in the UHF frequency band of 470 MHz to 790 MHz, is available to alleviate the spectrum challenges. Moreover, as TV broadcasting shifts from the analogue to digital broadcasting, this will free up more unused spectrum and open the path for dynamic spectrum access to support data services. These vacant unused frequency spectrums within the analogue TV broadcasting spectrum are referred to as the Television white space (TVWS).
The Department of Computer Engineering at the School of Engineering Sciences, is currently collaborating with the University of Strathclyde, Glasgow and three other partner institutions in Africa on a TVWS project that seeks to provide affordable internet access with dynamic spectrum management and software defined radio. The African partner institutions on the project are the Strathmore University (Kenya), University of Malawi (Malawi), and Copperbelt University (Zambia). The project is being sponsored by the EPSRC (Engineering and Physical Sciences Research Council). This project follows a successful pilot TVWS project that was carried out (2015) by the Department in collaboration with the UGCS of the University of Ghana with equipment support (one base station and 6 customer equipment) from the NRCS and free spectrum allocation from the NCA of Ghana. The focus of the UG TVWS project on campus was to explore the possibility of using the technology to extend data services to constrained service areas in the university for students and the staff residential areas. In this current TVWS project however, the focus is to investigate how the use of dynamic spectrum access (DSA) management and geo-location database technology, combined with the software defined radio (SDR) implementations could be used to enable effective and efficient wireless networks to be built in Ghana to support affordable Internet access using the shared spectrum resource of the TV broadcasting spectrum.

Impact on society
On successful completion, the project will provide useful technical information to the NCA in shaping their regulatory framework for TVWS deployment in Ghana. Currently, licenses in Ghana are statically applied for and released by the NCA, but as more networks come online, there will be the need for methods that will enable sharing of spectrum resources, including RF sensing and software-defined radios and accessible cloud services to manage the data. This project will provide the NCA benefits that can be derived from dynamic spectrum access and dynamic spectrum management using the geo-location spectrum database method. Further, it will facilitate the extension of data services to Ghanaians in service constrained areas of the country with difficult terrain using the TV broadcasting spectrum. Through the project, a geo-location spectrum database for the country will be developed in collaboration with the NCA to enhance and regulate TVWS operations.  

4. Geospatial distributed systems:

This project seeks to integrate emerging technology such as the Internet of Things (IoT), Artificial Intelligence (AI), and Robotics into geo-distributed systems. The resulting Geo-IoT, Geo-AI and Geo for Robotics can be applied in solving engineering, socio-economic, health and land administration related issues.

5. Fraud Detection and Alert Systems on Mobile Money platforms:

This research is exploring innovative ways of detecting fraudulent activities on the mobile money platforms in Ghana. Current approaches attempt to combine Artificial Intelligence with Web and Cloud Systems to alert mobile users of incoming calls from potential fraudsters.

List of Major equipment
1. Robotic arm in the Control and Automation Lab
2. Telecommunication and Networks Trainer
3. Signal Processing Lab Trainer
4. Huawei Lab with over 40 computer and networking