Designs for the Quality of Service Support in Low-Energy Wireless Sensor Network Protocols

Abstract

A Wireless Sensor Network (WSN) consists of small, low cost, and low energy sensor nodes that cooperatively monitor physical quantities, control actuators, and perform data processing tasks. A network may consist of thousands of randomly deployed self-configurable nodes that operate autonomously to form a multihop topology. This Thesis focuses on Quality of Service (QoS) in low-energy WSNs that aim at several years operation time with small batteries. As a WSN may include both critical and non-critical control and monitoring applications, QoS is needed to make intelligent, content specific trade-offs between energy and network performance. The main research problem is defining and implementing QoS with constrained energy budget, processing power, communication bandwidth, and data and program memories. The problem is approached via protocol designs and algorithms. These are verified with simulations and with measurements in practical deployments. This Thesis defines QoS for WSNs with quantifiable metrics to allow measuring and managing the network performance. The definition is used as a basis for QoS routing protocol and Medium Access Control (MAC) schemes, comprising dynamic capacity allocation algorithm and QoS support layer. Dynamic capacity allocation is targeted at reservation based MACs, whereas the QoS support layer operates on contention based MACs. Instead of optimizing the protocols for a certain use case, the protocols allow configurable QoS based on application specific requirements. Finally, this Thesis designs sensor self-diagnostics and diagnostics analysis tool for verifying network performance. Compared to the related proposals on in-network sensor diagnostics, the diagnostics also detects performance problems and identifies reasons for the issues thus allowing the correction of problems. The results show that the developed protocols allow a clear trade-off between energy, latency, throughput, and reliability aspects of QoS while incurring a minimal overhead. The feasibility of results for extremely resource constrained WSNs is verified with the practical implementation with a prototype hardware platform having only few Million Instructions Per Second (MIPS) of processing power and less than a hundred kBs data and program memories. The results of this Thesis can be used in the WSN research, development, and implementation in general. The developed QoS definition, protocols, and diagnostics tools can be used separately or adapted to other applications and protocols.