QoS Support in Wireless Sensor Networks: A Survey

1
QoS Support in Wireless Sensor Networks A Survey

• Dazhi Chen Pramod K. Varshney
• Presented by So Ra Park

2
Wireless Sensor Network (WSN)

• A new technological vision.
• Provide information regarding the physical
  phenomena of interest
• Possibly detect and control phenomena
• Enable to construct more accurate models of the
  physical world.

3
Quality of Service (QoS)

• Different perspectives on QoS
• User/application perspective
• Network perspective
• Qos in traditional data network

4
QoS Reqs from Applications in the Traditional
data Networks

• Rising multimedia applications
• Different multimedia applications different QoS
  requirements
• Guaranteed services differentiated services.
• Different types of networks
• Dynamic topology of mobile ad hoc networks.

5
QoS Support in Traditional Data Networks (1)

• Over-provisioning of resources
• Abundant resources
• All users same class
• Traffic engineering
• Reservation-based
• Asynchronous Transfer Mode, InterServ model
• Reservationless
• Admission control strategy
• Policy manager
• Traffic classes
• Queuing mechanism
• Diffserv model

6
QoS Support in Traditional Data Networks (2)

• Infrastructure-based wireless networks
• Ex. Wireless Local Area Network, Broadband
  Wireless Access Networks
• All mobile hosts one hop away from the
  communication cell
• Scarce bandwidth
• Complexity of user mobility during the last
  wireless hop.
• Integration of QoS architecture in wired with
  wireless MAC protocols.
• Wireless MAC priorities.

7
QoS Support in Traditional Data Networks (3)

• Wireless ad hoc networks
• Autonomous system
• Individual routing protocols
• Multi-hop wireless extension to the Internet
• Need to provide seamless access to the Net.
• Bandwidth constraint dynamic network topology
• QoS parameters
• QoS model
• QoS resource reservation signaling
• QoS routing
• QoS Medium Access Control (MAC)

8
QoS Support in Traditional Data Networks (4)

• Conclusion about QoS in Trad. Data Network
• They have common QoS requirements
• Same end-to-end parameters are used
• Specific techniques to realize QoS support are
  diverse due to unique properties of underlying
  networks.

9
QoS Requirements in WSN (1)

• WSN is composed of a large number of sensor nodes
  scattered in a terrain of interest.
• Each sensor note has the capability of collecting
  data about an ambient condition, i.e.,
  temperature, pressure, humidity, noise, lighting
  condition etc., and sending data reports to a
  sink node.
• More application we might envision now
  impossible to analyze QoS requirements by each
  application.

10
QoS Requirements in WSN (2)

• Application specific Qos
• Coverage
• Exposure
• Measurement errors
• Optimum number of active sensors

11
QoS Requirements in WSN (3)

• Network Qos
• How the underlying communication network can
  deliver the QoS-constrained sensor data while
  efficiently utilizing network resources.
• Analyze each class of applications classified by
  data delivery models.
• Data delivery models (event-driven, query-driven,
  and continuous delivery models)

12
QoS Requirements in WSN (4)

• Event-driven
• Interactive
• Delay intolerant (real-time)
• Mission critical
• Non-end-to-end applications
• Important points
• The application itself is not end-to-end
• The data flows from these sensors are likely to
  be highly correlated and contains much redundancy
• Data traffic generated by a single sensor may be
  of low intensity event showers
• Action response from the sink to sensors or
  actuators should be distributed quickly and
  reliably.

13
QoS Requirements in WSN (5)

• Query driven
• Interactive
• Query-specific delay tolerant
• Mission critical
• Non-end-to-end applications
• Queries could be sent on demand
• Data is pulled by the sink (differ from event
  driven model)
• A query may be used to manage and reconfigure the
  sensor nodes
• Commands from the sink constitute one-way traffic
  and require high reliability

14
QoS Requirements in WSN (6)

• Continuous
• Sensors send their data continuously to the sink
  at a pre-specified rate
• Real-time video, image, audio or video data
• Delay constrained
• Certain bandwidth requirement
• Packet loss can be tolerated to a certain extent
• Not end-to-end application
• Non-real-time data
• Delay tolerated
• Packet loss tolerated

15
QoS Requirements in WSN (7)
16
Application Requirements the Differences between
WSNs and Trad. Networks

• Applications in WSN are no longer end-to-end
  applications
• Bandwidth is not the main concern for a single
  sensor node for a group of sensors might.
• Packet losses in traffic generated by a single
  sensor node can be tolerated to a certain extent.
• Most applications in WSNs are mission-critical

17
Non-End-to-End Parameters (Collective Parameters)
for Meeting Requirements of WSNs

• Collective latency
• Collective Packet loss
• Collective bandwidth
• Information throughput

18
Unique Challenges in WSNs Support

• Severe resource constraints
• Unbalanced traffic
• Data redundancy
• Network dynamics
• Energy balance
• Scalability
• Multiple sinks
• Multiple traffic types
• Packet criticality

19
Current Research Efforts

• Traditional end-to-end QoS
• Reliability assurance
• Application-specific QoS
• None of these definitions is from the network
  perspective, and the QoS support in their methods
  is not directly related to the QoS support from
  the underlying network

20
Goals of WSNs QoS

• Efficient bandwidth utilization
• Minimal usage of energy
• QoS support in WSNs should also include QoS
  control besides QoS assurance mechanism

21
Research Issues

• Simpler QoS models
• QoS-aware data dissemination protocols
• Services
• QoS support based on collective QoS parameters
• Traditional end-to-end energy-aware QoS support
• Trade-offs
• Adaptive QoS assurance algorithms
• Service differentiation
• QoS support via a middleware layer
• QoS control mechanisms
• The integration of QoS support