Abstract
This chapter presents a survey of the main aspects of QoS that are being used on the Wireless Sensor Networks technology world. We describe approaches based on traditional networks, as well as new approaches that face with resources limitation, which is one of the main characteristics of WSN nodes. This chapter also describes how QoS management creates a series of challenges to deal with, and how some QoS features could be used to enable users to make a better profit of their resource limitations. In our opinion, an exposition of these topics will improve the set of techniques and strategies that designers and programmers could use to develop and implement satisfactory services level to their applications in WSNs.
TopQos Requirements In Wsns
QoS can be defined as the collective effect of service performance which determines the degree of satisfaction of a user of the service (ITU-T, 1994). Different applications require different QoS and if they do not meet the requirements, the resulting behavior will be unsatisfactory (Chen & Varshney, 2004).
Key Terms in this Chapter
Reliability: It is a measure of how much everything is working properly on the system. A fully reliable network would be that where: there is no one packet lost, each network component is working without appreciable faults, and each system component would have enough time and resources to do its work.
QoS: It is the establishment of some goals or requirements to application performance metrics, in order to reach some user’s degree of satisfaction, about the services provided by a system. Different applications and users could require different QoS metrics and levels, and if they do not meet the requirements, the system behavior will be unsatisfactory.
Differentiated Services: (DiffServ) model: It is a model that classifies information packets flows into different service levels. The idea is to give the same treatment to the flows associated to an specific service level, but offering special treatments to some services depending on the system objectives.
Latency: It is the amount of time since an event occurs, in the physical world, and its detection in the WSNs Sink node, in order to be registered and processed by system users.
Network Lifetime: It is the amount of time that a Wireless Sensor Network would be fully operative. One of the most used definitions of network lifetime is the time at which the first network node runs out of energy to send a packet, because to lose a node could mean that the network could lose some functionalities. But, is also possible to use a different definition, in which some nodes could die or run out of battery power, whenever other network nodes could be used to capture desired information or to route information messages to their destination.
Throughput: It is a measure of how much work is being done by a system in each unit of time.
Wireless Sensor Network: It is a set of related devices with small size and some resource limitations that are used to collect data from the physical world, and gather them through a wireless interface to a place were this information would be used. In some cases a wireless sensor node could be used in automation closed loops to make some control activities, very close to the place where the physical process is located.
Energy aware routing protocol: It is a routing protocol that uses some information about power energy characteristics to take its routing decisions. Some of the more commons energy power features used by these protocols are: the amount of battery power spent by sending a message to a node, the amount of battery power spent by sending a message through a specific path, or the remaining battery power of a node.