**2. Mobile entities**

50 Wireless Sensor Networks – Technology and Protocols

Power constraints result in small message sizes

resulting in their energy being consumed earlier.

memory and power.

determined.

network lifetime.

 Most network applications require dense deployment and physical collocation of nodes. Individual sensor nodes have limited resources in terms of processing capability,

The placement of nodes in a WSN is application dependent and may not be pre-

A WSN also differs from other wireless networks, such as cellular networks and mobile ad hoc networks (MANETS) because these networks are linked to a wired or renewable energy supply. In cellular networks and MANETS, the organising, routing and mobility management tasks focus on optimizing quality of service (QoS) and ensuring high bandwidth efficiency. There is a large amount of network traffic and the data rate is high to cater for the demand for multimedia rich data. These networks are designed to provide good throughput/delay characteristics under high mobility conditions [2]. Energy consumption is of secondary

As the term ``wireless'' implies, there is no fixed physical connection between sensors to provide continuous energy and an enclosed communication medium. This creates two problems, firstly, the sensor has a finite amount of energy, which once depleted, disables the sensor and hence reduces network lifetime. Secondly, all transmitted messages will be detected by any listening device within receiving range, which then has to decide whether to accept, forward or ignore the message. This signal transmission and reception has a power cost. In addition, many WSN applications do not have a pre-planned network topology and nodes are only aware of their immediate neighbours. When routing a message to a sink, the nodes closest to the sink receive a disproportionate amount of messages,

Initial message routing protocols assumed the sink or destination node was in a fixed location, and that network nodes had no or limited knowledge of the network topology [5]. An area of active research for a number of years has been how to notify the central sink (or monitoring hub) about an event in real-time by utilising the minimum amount of power of sensor nodes. Strategies to improve node energy efficiency include using multiple sinks in the application area and the use of mobile sinks to collect data from stationary sensor nodes to prevent nodes close to a sink from having their energy depleted and hence decreasing

A model for optimum path movement of mobile sinks to reduce the number of messages transmitted and received by an individual sensor node is proposed. An investigation is conducted into the optimum route a mobile sink can travel that will reduce the number of messages transmitted within a network, allow equitable usage of all nodes to transfer an

In the following sections a brief discussion of the use of mobile elements in WSNs as well as current research using mobile sinks and/or nodes to improve the energy efficiency of routing protocols is provided. The algorithm to transmit data from a sensor node to a

event message and still allow an event to be reported in real-time.

mobile sink is discussed and the results analysed.

importance as the battery packs can be replaced or re-charged as needed.

The application and routing challenges presented by static nodes in a dense, multi-hop WSN has led to the investigation of the use of mobile elements in WSNs for data collection and/or dissemination. The advantages of using mobile entities in WSNs include [6, 7]:


The use of mobility in WSNs introduces complications not found in static WSN applications, such as detecting when nodes are within transmission range of a mobile sink, ensuring reliable data transfer as nodes may move as messages are exchanged, tracking sink location and design of a virtual backbone to store data reports so that the mobile sink can easily collect them, and managing sensor nodes to support sink mobility [6, 7].

Current strategies for data collection and dissemination using mobile elements include a rendezvous-based virtual infrastructure which uses limited and unlimited multi-hop relays to route data messages, or a backbone-based approach where mobile sinks only communicate with pre-defined cluster heads or gateways, or passive data collection where there is direct communication between the source and sink [7, 8].

The mobility patterns of mobile elements (sinks and relays) are dependent on the type of WSN application, its data collection requirements and the controllability of the mobile elements. Current mobility patterns can be classified into the following categories [6, 8]:

