Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.778523
Title: Cross-layer optimization for hybrid wireless-power line sensor networks
Author: Zhu, K.
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Wireless sensor networks (WSNs) have a wide range of applications in the realization of the Internet of Things (IoT). However, WSNs are confronted with dynamic wireless medium, limited bandwidth, limited energy supply (in battery powered WSNs) and the often present blind spot problems. In particular, battery powered WSNs used for the in-door applications, such as impromptu surveillance installation or industrial chemical process monitoring (mains powered WSNs are prohibited due to safety concerns), have stringent energy budgets. Power line communication (PLC) is considered as a promising technology for data transmission that utilises the widespread presence of power line (PL) cables as a communication medium. Owing to this advantage, PLC networking can be considered as a practical supplement to existing in-door WSNs. In this thesis, it is aimed to investigate the performance improvement (i.e., network lifetime) in the hybrid wireless-PL sensor networks for in-door applications through cross-layer design. In the first contribution, a hybrid sensor network for industrial sensor network applications, which consists of both wireless and PL sensor nodes is proposed. Since the data rate requirement for such applications is typically low, and for the ease of derivation, the power consumption model takes into account the transmission signal power and the power consumption of the power amplifier. To the best of our knowledge, it is the first reported work in the literature that focuses on the cross-layer design of such a heterogeneous network. The hybrid sensor network takes the advantage of the flexibility of WSNs while the PL sensors are deployed to prolong the lifetime of the network. This work studies the joint design of the PHY, MAC and network layers to maximize the hybrid network lifetime, which is limited by the battery capacity of wireless sensors. Second, closed-form expressions of the globally optimal solution for lifetime maximization of the hybrid sensor network are derived for two different network topologies, namely string topology and linear topology. Such closed-form solutions give insights in factors that are significant to the network lifetime when designing the hybrid sensor network. Third, the impacts of different network configurations such as source rate, sensor node densities, etc., on the hybrid network lifetime are investigated. The impact of different transmission strategies of PL nodes on the effectiveness of the network is studied. In the second contribution, a hybrid video sensor network (HVSN) which comprises both battery-powered wireless sensor nodes and PL sensor nodes is proposed to maximize the network lifetime. Since HVSN have a high data rate requirement, the power consumption model includes the power consumption due to video encoding, data transmission and reception. To the best of our knowledge, it is the first reported work to investigate video sensor networks with hybrid power sources and hybrid communication schemes. The proposed HVSN utilizes the flexibility of wireless nodes while PL nodes are used to extend the network lifetime. Second, the joint design of video encoding rate, aggregate power consumption, channel access control, along with link rate allocation is studied for maximizing the hybrid network lifetime. The joint design achieves much better performance than separate optimization. Third, a distributed algorithm is proposed for the network lifetime maximization problem. The distributed algorithm divides the computational burden among all nodes with much lower communication overhead. Fourth, the impact of dynamic network change and network scalability is studied. The effectiveness of the proposed algorithm is validated through extensive simulation results.
Supervisor: Zhu, X. ; Huang, Y. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.778523  DOI:
Share: