Network Management for Dynamic and Heterogeneous Wireless Sensor Networks

Sammanfattning: Wireless Sensor Networks (WSNs) are interconnections of spatially distributedsensor nodes with low-power wireless communication. In comparison withtraditional wireless networks, WSNs provide novel features in networkarchitecture, which are known as low-cost networking, rapid formation,widespread arrangement, self-organisation, and ease of deployment. For thesereasons, many WSNs are utilised and interconnected to form massiveInternet-of-Things (IoT) applications in areas such as e-healthcare, industrialautomation, and smart cities, to name a few. Nevertheless, WSN networking isaffected by dynamic network topology – nodes’ mobility and varying networkdensity, and heterogeneous networks (hetnets) – coexisting radio technologies.In more detail, mobile nodes create and break network connections as theymove around; varying network density causes alterations in the routing graph;and coexisting radios is a scenario when multiple WSNs utilising differentcommunication technologies are located within same physical space, wherenetwork interoperability is required for non-obtrusive operations between theco-located networks. These networking constraints must be properly managed,otherwise, they will lead to non-deterministic and erratic behavior in overallIoT applications, causing degraded Quality-of-Service (QoS). In recent literature, several proposals primarily address only one of theseaspects; either mobility, density, or coexisting radios. In this thesis, we proposea unified approach to manage dynamic networks and hetnets through the supportof network-wide decision-making in the management process. A unified methodresults in reduction of the overhead cost for system resources and computationcomplexity with respect to the constrained nature of WSNs, and improvementin decision-making efficiency in network management. The main contributionsare modelling of the network management process, proposing an architecture,simulating and implementing the proposed architecture, and evaluating thenetwork performance under network management for WSNs. Performancemetrics include latency, throughput, and packet loss rate. Software DefinedNetworking (SDN), a method for programmable network management, is asuitable solution. The challenges in design and implementation of network management forWSNs have been experimentally studied in the first paper in the thesis. Further,mobility management has been modelled in the second paper with the modulararchitecture in the third paper. In the third and fourth paper, we implement theproposed architecture and evaluate network performance. In the fifth paper, wehave analysed the network performance in different networking architectures.For standardised simulation and evaluations, We used the Contiki and Mininetsimulators, and Linux networking simulators, as well as Linux networkingconcepts on virtualisation. Empirically, simulation results show that the meanlatency has improved more than 100 times with network management comparedwith typical methods using distributed routing protocols, under varying densityin hetnets. Also, we observed a smooth-line performance in mobilitymanagement with close to zero packet losses under nodes’ mobility withsporadic communication patterns.