Multi-sink mobile wireless sensor networks: dissemination protocols, design and evaluation

In pervasive systems, as they are getting smaller and smaller, computers can be found just about everywhere, but their presence is not noticed because the technologies are often embedded within items. One of the smallest and well known embedded computers is a wireless sensor node, which is a passive sensing device capable of communicating wirelessly with other devices. Early attempts to monitor the physical environment are primarily composed of these passive sensing devices which have been succeeded in many applications by the development of Stationary Wireless Sensor Networks. A wireless sensor network is typically composed of many tiny computers, often no bigger than a coin or a credit card, that feature a low frequency processor, some flash memory for storage, a radio for short-range wireless communication, on-chip sensors and an energy source such as AA batteries. Applications of stationary wireless sensor networks have emerged in many domains ranging from environmental monitoring to structural monitoring as well as industry manufacturing. In all these applications, the primary task of a wireless sensor network is to collect useful information by monitoring phenomena in the surrounding environment. Typically, in a wireless sensor network, sensor nodes generate data about a phenomenon and relay streams of data to a more resource rich device, namely a data sink, for analysis and processing. Early sensor networks have been modeled as having a single, predefined, stationary sink. However, as the size of the sensor network grows with the wide availability of economically viable embedded sensor nodes, the communication between the sensors and the single stationary sink can lead to high energy consumption, and consequently reduce the lifetime of the network. In recent years there has been renewal of interest in using multiple sinks for wireless sensor networks to achieve power saving. Although multi-sink partitioning of the sensed area enhances some performance metrics, such as network lifetime, of single sink sensor network, the development of multiple stationary sinks in an area of interest still creates an uneven energy depletion phenomenon around the sinks, since sensors near a data sink deplete their battery power faster than those far apart, due to their heavy overhead of relaying messages. The improvements of stationary wireless sensor networks in conjunction with the advances developed by the distributed robotics and low power embedded systems communities have led to a new class of Mobile Wireless Sensor Networks that can be utilized for a wide range of scenarios such as land, sea and air exploration and monitoring, habitat monitoring, vehicular applications, and emergency response, which require reliable and timely collection of data. Current studies have tried to utilize the advantages of mobile sensors to overcome the problems of stationary sensor networks. Mobile Wireless Sensor Networks have a similar architecture to their stationary counterparts, thus are governed by the same energy and processing limitations, but require the development of a new generation of algorithms targeting at constantly changing network topologies due to sink and/or sensor mobility. This thesis focuses on the efficient data extraction and dissemination in wireless sensor networks by making use of the multiple sinks and by handling mobility of sensors and sinks. We start with analyzing the characteristics of multi-sink wireless sensor networks. We propose a set of algorithms that enable multi-sink wireless sensor network to self-organize efficiently in the presence of mobility and adapt to dynamics in order to increase the functionality of the network. Our contributions include an algorithm for load balancing in multi-sink sensor networks, a protocol for query dissemination towards an area of interest combined with a set of algorithms that are used to handle mobility efficiently in a tree-based routing, and a data dissemination protocol that tackles sink mobility in a wireless sensor network. In short, the main contributions of the thesis are listed as follows: - Contribution 1 - Benefits and challenges of using multiple sinks in static wireless sensor networks: We review the state of the art multi-sink partitioning methods in wireless sensor networks. We present a multi-sink partitioning mechanism to achieve load balancing between sinks. - Contribution 2 - Design and evaluation of a query dissemination protocol for multi-sink wireless sensor networks: To enable sinks to efficiently route queries which are only valid in particular regions of the deployment, we propose a set of algorithms which combine coverage area reporting and geographical routing of queries that are injected by sinks. - Contribution 3 - Handling mobility of sensors in a tree-based dissemination protocol: To provide an up-to-date coverage area description to sinks, we focus on handling sensor node mobility in the network. We discuss what is the best method to handle mobility in tree-based routing of queries: (i) periodic global updates from every sink or (ii) local updates only from mobile sensors. We propose a method to achieve local updates which are needed to handle sensor mobility in a tree-based network. - Contribution 4 - Design and analysis of a data dissemination protocol for mobile multi-sink sensor networks: To achieve reliable data dissemination of events as well as the efficiency in handling the mobility of both multiple sinks and event sources, we propose a virtual infrastructure and a data dissemination protocol, namely HexDD (Hexagonal cell-based Data Dissemination) exploiting this infrastructure. We analytically compare the communication cost and hot region traffic cost of the proposed data dissemination with other approaches. - Contribution 5 - Evaluation of the data dissemination protocol for different wireless sensor network applications: We focus on the performance evaluation of the data dissemination protocol, HexDD, in two different classes of mobile wireless sensor networks: (i) mostly static, which contains scenarios in which most of the sensors are static and some sensors are attached to people or vehicles such as firefighters or unmanned aerial vehicles moving at low or medium velocities in an Emergency Response Application, (ii) highly mobile, which contains scenarios in which many sensors are attached to devices that move at high velocities such as cars in a Vehicular Sensor Network Application.