Date of Award

21-10-2024

Document Type

Thesis

School

School of Electrical & Electroncis Engineering

Programme

Ph.D.-Doctoral of Philosophy

First Advisor

Dr.G.Rajkumar

Keywords

Wireless Body Area Networks, Optimization Algorithms, Bio-inspired Optimization Algorithms, Energy Aware Routing, Engineering and Technology

Abstract

In recent times, Wireless Body Area Networks (WBAN) a subsection of Wireless Sensor Networks (WSN) a promising technology for the future healthcare realm with cutting-edge technologies that can assist healthcare professionals like doctors, nurses and biomedical engineers which enables one-to-one care by implementing Tele-Medicine and Tele-Health solutions. Machine Learning (ML) and Internet of Things (IoT) enabled medical big data is the future of the healthcare sector and Medical Technology-based industries leading to applications in other sectors such as fitness tracking for commercial purposes, Sportsperson health monitoring to track their day-to-day activities and wearable devices for critical and emergency care.

Recent proliferation in miniaturized microelectronics and sensor-based wireless communication and networking industries paved the way for the emergence of WBAN. Mobility in WBAN has become a major challenge in framing the network topology and achieving better network performance. To address the dynamic nature of WBAN EADC-RP protocol is introduced to analyze the network performance with different mobility rates, where a network with the highest mobility rate of 1.0 m/s has less performance when compared to the mobility rates of 0.3 m/s and 0.6 m/s. WBAN setup with different postures and a multihop routing to select a relay node to collect data from far away sensor nodes and to CCN with different locations is analyzed.

ESTEEM is a novel approach for choosing the best-fitted AN by incorporating the HMM to identify the topology variations in the network. The outcome of the proposed ESTEEM achieves increased throughput, better stability with the first dead node and extended network lifetime with the last dead node at 71% and 95.1% of total simulation time respectively. A novel approach that requires dynamic clustering with different cluster members over the course of the data transmission process. The ANFIS integrated with MGWO to optimize the energy consumption in WBAN through sensor clustering. The key objective is to enhance energy efficiency by dynamically organizing sensors into clusters based on their contextual data.

ANFIS is employed to model the intricate relationship within the network, providing a flexible and adaptive system capable of learning and adjusting to the dynamic nature of the WBAN environment. MGWO inspired by social behavior in grey wolves, is utilized as a metaheuristic optimization algorithm to fine-tune the parameters of energy, distance and packet inter-arrival time. The proposed methodology involves formulating an objective function that encapsulates the energy efficiency goals of the WBAN.

An iterative MGWO optimizes the parameters and effectiveness of the hybrid approach is validated through simulations and achieves superior energy efficiency when compared to traditional optimization techniques. This research proposes Hybrid ANFIS-MGW Optimizer (HAMO) has achieved enhanced network performance and energy efficiency in WBAN. The proposed HAMO based on ANFIS and GWO for energy-efficient WBAN is 57.1%, 54.3% and 73.58% better than iM-SIMPLE, ACO-WBAN and GWO-WBAN respectively.

Download

Share

COinS