عنوان

Effective Wireless Sensor Networks Utilizing Residue Number Systems and Intelligent Multi-Agent Technologies

Number

TL50963

.Language of Text, Soundtrack etc

انگلیسی

Title Proper

Effective Wireless Sensor Networks Utilizing Residue Number Systems and Intelligent Multi-Agent Technologies

General Material Designation

[Thesis]

First Statement of Responsibility

Raji, Ayodele Kamaldeen

Subsequent Statement of Responsibility

Gbolagade, Kazeem Alagbe

Name of Publisher, Distributor, etc.

Kwara State University (Nigeria)

Date of Publication, Distribution, etc.

2019

Text of Note

226 p.

Dissertation or thesis details and type of degree

Ph.D.

Body granting the degree

Kwara State University (Nigeria)

Text preceding or following the note

2019

Text of Note

Wireless Sensor Networks (WSNs) consist of enormous number of sensor hubs scattered in a territory of enthusiasm with at least one base station for observing environment and physical conditions. These sensors nodes are disseminated in hostile environment and are defenceless to shortcomings such as power dissipation, hardware glitches, communication link errors and malicious attacks, among others. It has been established that vitality is the foremost challenge in the functionality of WSNs as they are controlled with constrained vitality and replacement or revitalize of vitality resources might be difficult. Thus, it is important to plan WSNs in an energy-efficient manner because it affects the network lifetime. Additionally, speed, reliability and fault tolerant are other major challenges affecting many WSNs applications. This is because WSNs are used to collect vital information from the environment and need to be fast, reliable and continue to operate even if some failures occur. This thesis therefore designs energy efficient, high speed, reliable and fault tolerant WSNs using Residue Number System (RNS) and intelligent multi-agent technologies. First, in order to reduce energy consumption, Chinese Remainder Theorem (CRT)-based packet splitting incorporated with Low-Energy Adaptive Clustering Hierarchy (LEACH) algorithm was proposed. The design integrates Local Closet First (LCF), in order to select closest cluster members and utilized the moduli set {2n+1 − 1, 2n, 2n − 1} so as to break the sensed data by each hub and send equivalent residues to the Cluster Head (CH). This reduces communication congestion and eliminate data aggregation. Consequently, the CH correspondingly uses LCF to send the residues to neighbouring CH until it gets to the base station. When the base station receives the residues of the sent message from a CH, it rearranges the message by utilizing the moduli set which serves as the secret key. A prototype experiment was conducted on Probabilistic Wireless Network Simulator (Prowler) running under MATLAB in order to simulate the proposed algorithm and other existing similar state of the art algorithms. The obtained results indicate that the proposed scheme outperforms traditional methodologies with respect to vitality utilization, network lifespan, speed, congestion reduction, packet delivery, and decrease in delay of the sent data. Second, High SPeed REverse Converter (HISPREC) was also proposed for the moduli set {2n+1 − 1, 2n, 2n − 1} and implemented based on Mixed Radix Conversion in order to convert the received residues from the CH to the original message at high speed. The results of theoretical and experimental simulations show that the proposed HISPREC outperforms existing reverse converters in terms of speed, hardware requirements, conversion delay and reliability of the received data. Additionally, an improved CRT-based error control scheme utilizing Redundant Residue Number System (RRNS) was proposed to boost the correctness of received message for a reliable WSNs. The system utilized the moduli set {2n+1 −1, 2n, 2n −1} with additional redundant moduli set {2n+1 + 3, 22n − 3} in order to achieve higher error controllability. The obtained results show that the proposed scheme provides low vitality utilization and consistency of the received message in real-time WSNs. Lastly, the research proposed fault tolerant WSNs using intelligent multi-agent technology to monitor, control and allow the network to continuously operate in the presence of faults, as well as eliminate any faulty node due to power loss. Netlogo and C# were employed as the software framework to facilitate development of intelligent multi-agent systems. The results show that agent-based fault tolerant allows the fault tolerance level of each node leading to a better fault recovery process, and therefore, a longer lifetime of the sensor network. The overall design was an effective WSNs using RNS and intelligent multi-agent technologies. This approach to WSNs design is one of the few designs that address the integration of RNS and intelligent multi-agent technologies in real-time WSNs. This research makes several unique contributions to the previously developed and documented energy efficient and fault tolerant models.

Subject Term

Computer science

Raji, Ayodele Kamaldeen

Gbolagade, Kazeem Alagbe

Kwara State University (Nigeria)

Electronic name

p

[Thesis]

276903

a

Y