Adewale Sunday

Followers

Following

Public Views

Interests

Uploads

Papers by Adewale Sunday

Threshold-based bandwidth allocation scheme for multiservice wireless networks

Journal of Computer Science and Its Application, 2015

Priority Queuing with Buffer Management (PQBM) has been recently proposed in the literature. The ... more Priority Queuing with Buffer Management (PQBM) has been recently proposed in the literature. The protocol is acclaimed to be a very attractive candidate in current (2.5G and 3G) and next (4G) multiservice wireless networks. However, it suffers bandwidth monopolization by higher priority data packets. This paper proposes a new scheme namely Improved Priority Queuing with Buffer Management (IPQBM) with a view to redistributing service provision to the data classes optimally. In particular, the scheduling discipline of PQBM is modified so as to correct the monopoly problem. IPQBM is compared with PQBM vis-a-vis throughput and mean delay of data packets. Numerical results reveal that IPQBM offers improved throughput for lower priority data packets by ensuring that higher priority data class does not have a monopoly of bandwidth resources. Keywords : Priority Queuing, Buffer Management, Protocol, Bandwidth, multi-service wireless networks.

A framework for detecting credit card fraud with cost-sensitive meta-learning ensemble approach

Scientific African, Jul 1, 2020

Electronic payment systems continue to seamlessly aid business transactions across the world, and... more Electronic payment systems continue to seamlessly aid business transactions across the world, and credit cards have emerged as a means of making payments in E-payment systems. Fraud due to credit card usage has, however, remained a major global threat to financial institutions with several reports and statistics laying bare the extent of this challenge. Several machine learning techniques and approaches have been established to mitigate this prevailing menace in payment systems, effective amongst which are ensemble methods and cost-sensitive learning techniques. This paper proposes a framework that combines the potentials of meta-learning ensemble techniques and cost-sensitive learning paradigm for fraud detection. The approach of the proposed framework is to allow base-classifiers to fit traditionally while the cost-sensitive learning is incorporated in the ensemble learning process to fit the cost-sensitive meta-classifier without having to enforce cost-sensitive learning on each of the base-classifiers. The predictive accuracy of the trained cost-sensitive meta-classifier and base classifiers were evaluated using Area Under the Receiver Operating Characteristic curve (AUC). Results obtained from classifying unseen data show that the cost-sensitive ensemble classifier maintains an excellent AUC value indicating consistent performance across different fraud rates in the dataset. These results indicate that the cost-sensitive ensemble framework is efficient in producing cost-sensitive ensemble classifiers that are capable of effectively detecting fraudulent transactions in different databases of payment systems irrespective of the proportion of fraud cases as compared to the performances of ordinary ensemble classifiers.

Download

An Internet-Based Telemedicine System in Nigeria

Online Journal of Space Communication, 2021

Download

Towards a Sustainable Agricultural Development in Nigeria: Taking advantage of distance education and eLearning, Anchored on Information and Communication Technology

CSREA EEE, 2008

The generic architecture for data/internet telephony in Nigeria

Telematics and Informatics, Nov 1, 2003

Nigerians were yet to get value for their money in telephony services through NITEL and GSM opera... more Nigerians were yet to get value for their money in telephony services through NITEL and GSM operators in Nigeria. After several years or months of commencement of operations, the service is not only too expensive, limited but inefficient. People because of the inefficiency in the sectors, got to a level where they have to procure sets connected to MTN, Econet and NITEL among others. In spite of this, there is problem of connectivity and reliability. Tariffs for the service are rather too expensive for both local and international calls. This paper proposes an innovative generic model for delivering voice and data traffics over IP to residential, small offices, medium and large enterprises in Nigeria at moderately reduced price, which cannot break oneÕs backbone while eliminating the current problems associated with NITEL telephony services. The paper also presents the benefits, the barriers of implementing Internet telephony as well as the Internet telephony implementation options. The cost for implementing the model is outlined in the paper.

Download

An Ontology-Based Diabetes Prediction Algorithm Using Naïve Bayes Classifier and Decision Tree

2023 International Conference on Science, Engineering and Business for Sustainable Development Goals (SEB-SDG)

A Diabetes Prediction Classifier Model Using Naive Bayes Algorithm

Fudma Journal of Sciences, Feb 28, 2023

Financial accounts reconciliation system using enhanced mapping algorithm

International Journal of Business Intelligence and Data Mining

Design of a personalised adaptive ubiquitous learning system

Interactive Learning Environments

Treatment strategies for the effluents of recycled waste paper mills of Punjab

Journal of Emerging Trends in Engineering and Applied Sciences, 2013

Recycled paper industry is a globally growing industry that consumes a significant amount of reso... more Recycled paper industry is a globally growing industry that consumes a significant amount of resources, raw materials and energy. Due to this fact, minimizing the environmental impact of this sector is important. The characteristics of the waste waters generated from various processes of the recycled pulp and paper industry depend upon the types of processes and raw materials, process technology applied, management practices,internal recirculation of the effluent for recovery and the amount of water being used in a particular process. One of the main causes of surface water and ground water contamination is industrial discharge from various industries viz. pulp and paper, textile, pharmaceutical and other chemical industries. These compounds are quite toxic and degrade slowly in the environment. Therefore it becomes imperative to completely degrade these organic compounds. The incapability of conventional waste water treatment methods to effectively remove many bio-recalcitrant poll...

A Fuzzy-Based Congestion Control Scheme for Vehicular Adhoc Network Communication

International Journal of Smart Vehicles and Smart Transportation, 2021

Vehicular ad hoc network (VANET) is a self-organized, multi-purpose, service-oriented communicati... more Vehicular ad hoc network (VANET) is a self-organized, multi-purpose, service-oriented communication network that enables communication between vehicles and between vehicles and roadside infrastructures for the purpose of exchanging messages. In a dense traffic scenario, the message traffic may generate a load higher than the available capacity of the transmission medium leading to channel congestion problem. This situation leads to a rise in packet loss rates and transmission delay. Some existing congestion control schemes adapt the transmission power, transmission rate, and contention window parameters by making comparison with neighboring values through classical logic. However, the approach does not consider points between two close parameter values. This work uses fuzzy logic to improve the adaptation process of the network contention window parameter. The proposed scheme achieved a 15% higher in-packet delivery ratio and 10ms faster transmission compared with related work in te...

Enhancing Network Life-Time of Wireless Sensor Networks through Itinerary Definition and Mobile Agents for Routing among Sensor Nodes

2020 International Conference in Mathematics, Computer Engineering and Computer Science (ICMCECS), 2020

Mobile Agents-Based Wireless Sensor Networks (WSNs) offer a good alternative to the conventional ... more Mobile Agents-Based Wireless Sensor Networks (WSNs) offer a good alternative to the conventional client/server network model. Mobile Agents (MA) is a computational code that is injected inside a packet that migrates among sensor nodes for data collection. The itinerary planning for MA is the most challenging task during mobile agents migration in wireless sensor networks. In this study, a routing itinerary scheme for multi-mobile agent itinerary planning for energy efficiency in wireless sensor networks by constructing the spanning tree of wireless sensor nodes is formulated. In the scheme, the sensor network is divided into several clusters. The cluster formation of the network consist of n sensor nodes divided into m disjoint clusters and the sequence of the collected data is combined with a fusion factor. The itinerary paths of the MA is defined using minimum spanning tree. It is assumed that the sink node is equipped with all the information about the SNs so that the sink can calculate the weights between cluster heads. The proposed Energy-Aware Mobile Agent Based (EAMB) model performed better than Least Energy Adaptive Clustering Hierarchy (LEACH), Local Closest First (LCF) and Global Closest First (GCF).

A Review of Congestion Control Schemes for Vehicular Ad-Hoc Network (VANET) Communication

International Journal of Vehicular Telematics and Infotainment Systems, 2017

Intelligent Transportation Systems (ITS) are built on top of self-organizing networks, known as V... more Intelligent Transportation Systems (ITS) are built on top of self-organizing networks, known as Vehicular Ad hoc Networks (VANET). In VANET, each vehicle equipped with communication devices represents a node and is allowed to send and receive safety messages through wireless communication channels. These messages are either periodic (beacons) or event-driven. Beacons are transmitted periodically while the event-driven messages are generated when an abnormal condition or an imminent danger is detected. The event-driven messages should be delivered to neighbouring nodes with high reliability and limit time as a single delayed or lost message could result in loss of lives. In dense network, the periodic messages broadcast lead to broadcast storm/blind flooding problem in VANETs. It is very important to keep the communication channel free from congestion in order to ensure timely and reliable delivery of event-driven safety messages. This study presents a review of existing congestion c...

Download

The generic architecture for data/internet telephony in Nigeria

Telematics and Informatics, 2003

Download

Development of an Enhanced Congestion Control Protocol in Transmission Control Protocol

International Journal of Computer Applications, 2020

The common problems that motivate most of the researchers in the field of congestion control revo... more The common problems that motivate most of the researchers in the field of congestion control revolves around how to have a congestion free network while maintaining efficient links utilization, fairness, elimination of packet loss/drop as well as achieving throughput in an heterogeneous network environments. This work adopted internet prey-predator model protocol and enhanced it. The network topology consist of a router, an access point (AP), four wired and wireless nodes were used as test bed. Network Simulator 3 (ns-3) and C++ was used to simulate the work. The work examined the basis of congestion variant and develop a new enhanced congestion control protocol. The numerical results obtained under different simulation scenarios show that the new protocol manifest high bandwidth utilization, reduced loss rate, fairness among the flows and improved throughput. The new protocol is recommended to be deployed on the public transmission control protocol stack, this will not only improve the network performance but also better subscribers experience and give them value for their money. It will also afford the Internet Service Providers the opportunity to provide uninterruptible internet access. Future work can focus on improving the performance of TCP with chaotic control theory using machine learning to better explore the model.

Download

Schematic Structure of National Data Harmonization System for Identity Management

European Scientific Journal, 2017

Imagine a web-application in which a user can query varieties of information about an individual ... more Imagine a web-application in which a user can query varieties of information about an individual (like name, age, state of origin, driver's license number, national identification number, etc.). This information would be stored on different databases, each having its own schema. Often times, these resources are replicated in different locations and on different platforms. Hence, the need for data harmonization or integration. Data harmonization addresses this issue by considering these external resources as materialized views over a virtual mediated schema resulting in "virtual data integration". When mediated schema is queried, the solution transforms the result into appropriate queries over the various and existing data sources. This paper focuses on data harmonization that cuts across different governmental database and incorporating them all into a centralized view and this is possible as a result of development and implementation of a webbased databank application.

Download

Developing a secured Mobile-Agent-Based Electronic Commerce Using Crypto-Steganography

International journal of sciences, 2018

Mobile agent is gaining attention in the research community because of its potential to replace t... more Mobile agent is gaining attention in the research community because of its potential to replace traditional client-Server application in computer network system. Mobile agent has the ability to migrate autonomously from one server to another server as specified in its itinerary in computer network. Areas where mobile agent profit includes network management, information retrieval and electronic commerce. Mobile agent can be delegated to carry out all stages that are involved in electronic commerce such as Product brokering, Merchant brokering, Negotiation and Purchase. However there are security vulnerability issues with mobile agent system such as the need to ensure confidentiality of information of the agent from malicious parties, integrity of the data of the agent and non-repudiation from parties involves in the transactions. Security measure by agent developers has been mobile cryptography. This research combines Elliptic curve cryptography (ECC) and Steganography known as cryptosteganography to secure its agent. The agent was developed using PHP 5.0, Javascript, HTML, Wampserver and MySQL. The system was tested on a local area network with six computer systems, one of the computers represent client and the other five as servers. Performance evaluation was carried out to test for execution time and workload effect on the system. The result was compared with the existing secured agent-based electronic commerce systems in term of response time. The result shows that the proposed security is faster than RSA but a little slower than ECC only. A trade off was observed between a more secured system and response time of the system.

Download

Web-Based Neural Network Model for University Undergraduate Admission Selection and Placement

The number of applicants who apply for University admission through the University Matriculation ... more The number of applicants who apply for University admission through the University Matriculation Examination (UME) organized by the Joint Admission and Matriculation Board (JAMB) in Nigeria is increasing yearly. Unfortunately, the facilities in the Nigerian Universities have not been expanding commensurately with the increasing number of applicants. Consequently, large numbers of applicants have to be screened yearly by University Admission Officers in order to select the few that their universities' facilities can support. Under this scenario, Admission Officers have to manually evaluate every candidate's data against the set admission requirements before selecting the few successful ones. This manual admission processing system exhibits three major problems, namely: it is laborious to manually evaluate the data of a very large number of candidates only to end up selecting less than 15% of them; most candidates are dropped off immediately if they fail to meet the admission requirements for their chosen course whereas, they could have met admission requirements for some other courses for which they could have been considered for placement; and the admission screening procedure is not transparent to the applicants and as such, many unsuccessful applicants find it difficult to believe that they are not offered admission for objective reasons. An effort towards developing a system that can take care of these problems is therefore a step in the right direction. This model is therefore being presented to provide a time-efficient, detailed, and unbiased automated procedure for selecting the most qualified candidates for admission into universities, ensuring that candidates who fail to meet some of the admission requirements for their chosen course are automatically considered for placement into another course for which they may meet the admission requirements, provided there are vacant positions, and providing an avenue for a candidate self-screening admission system.

Download

Understanding the Transmission Dynamics and Control of HIV Infection: A Mathematical Model Approach

Journal of Nigerian Society of Physical Sciences, Apr 19, 2023

Security Intelligence Framework for Suicide Bombers Identification in a Crowd

International Journal of Academic Research in Business and Social Sciences

Terrorism is the aggression unjustly carried out by individuals, groups, or states against human ... more Terrorism is the aggression unjustly carried out by individuals, groups, or states against human beings. It includes the form of unjustly terrifying, harming, threatening, and killing of people in the target area. In effect, this evil is perpetrated by some groups that believe in the use of such tactics as a means of advancing their cause. These acts are always caused by foreign terrorists whose identities and where about are not always known. Therefore there is need for an individual to identify these terrorists while in a crowd. This will be achieved by developing a security intelligent model for easy identification of a suicide bomber. The research involved document analysis and interview protocol with the security experts within Ondo and Ekiti State, Nigeria to determine the constructs for identifying the suicide bombers. Their responses were documented and audio recorded for transcription purposes. The sample for this phase of the study was obtained from among the senior security personnel until a saturation stage were reached. And the findings from them were analyzed thematically. The questionnaires were developed from the output of the thematic analysis and the result was analyzed using Rasch model which were used to determine the validity of the responses concerning the constructs. The results from the findings will form the model that can guide an individual to identify suicide bombers while in a crowd. It is hoped that the outcome of the study will increase the body of knowledge on security and contribute significantly to safety measures in avoiding suicide bombers' menace.

Download

Adewale Sunday

Uploads

Papers by Adewale Sunday

Log In