Ad hoc network notes MAC Protocols for Ad-Hoc Wireless Networks
Lecture 23: Mobile Ad-Hoc Networks · Mobile Ad-Hoc Networks: Quality of Service Reading: •...
Transcript of Lecture 23: Mobile Ad-Hoc Networks · Mobile Ad-Hoc Networks: Quality of Service Reading: •...
Lecture 7Mobile Ad-Hoc Networks:
Quality of Service
Reading: • “Quality of Service in Ad Hoc Wireless Networks,” in Ad Hoc Wireless
Networks: Architectures and Protocols, Chapter 10.• K. Wu and J. Harms, “QoS Support in Mobile Ad Hoc Networks,” Crossing
Boundaries– an interdisciplinary journal, Vol. 1, No. 1, Fall 2001.• H. Zhu, M. Li, I. Chlamtac and B. Prabhakaran, “A Survey of Quality of
Service in IEEE 802.11 Networks,” IEEE Wireless Communications, August 2004.
2
Quality of Service Challenge“Providing complex functionality with limited available resources in a dynamic environment”Supporting QoS requires knowledge of
Link delaysBandwidthLoss ratesError rates
Problem with ad hoc networksHard to obtain this informationLinks constantly changing: node mobility, environmental affects, etc.
3
QoS Services Hard QoS
Guarantee parameters such as delay, jitter, bandwidthRequired for mission-critical applicationsE.g., air traffic control, nuclear reactor controlNot feasible in MANETs
Soft QoSAim to meet QoS goalsLoss in QoS degrades application but does not have disastrous consequencesE.g., voice, videoMost research focuses on providing soft QoS
4
QoS ParametersBandwidthDelay jitter DelaySecurityNetwork availabilityBattery life…
5
Why is QoS Hard in MANETs?Dynamic network topology
Flow stops receiving QoS provisions due to path breaksNew paths must be established, causing data loss and delays
Imprecise state informationLink state changes continuouslyFlow states change over time
No central controlError-prone shared mediumHidden terminal problemLimited resource availability
Bandwidth, battery life, storage, processing capabilities
Insecure medium
6
Design Choices for QoSHard state vs. soft stateHard state
Resources reserved at all intermediate nodes in path for duration of flowIf path broken, resources must be explicitly releasedRequires control overheadMay fail to release resources if nodes on path unreachable
Soft stateResources reserved for small amount of timeReservations automatically renewed as long as flow continuesResources deallocated after timeout period if no new dataNo explicit tear-down neededLow overhead
7
Design Choices for QoS (cont.)Stateful vs. statelessStateful
Nodes keep either global or local stateState includes topology information and flow informationGlobal state not scalable
StatelessNo flow or topology information maintained at each nodeScalableDifficult to provide QoS without knowing any state information
8
Aspects of QoS in MANETSQoS models
What type of services can be provided? Defines the types of service differentiation
QoS resource reservation signalingCoordinates routing, MAC, admission control and scheduling
QoS routingFinds path with requested resources
QoS MACProvide support for QoS services
9
QoS Models for the InternetIntegrated Service (IntServ)
Routers keep flow-specific stateBandwidth requirementDelay boundFlow cost
Service modelsBest effortGuaranteed service: fixed delay boundControlled load service: better then best effort
RSVP protocol used to reserve resources in routersAdmission control used to accept/decline reservations at hostsPriority queues implemented to provide service guarantees to flows with accepted reservations
10
IntServ Model for MANETsIntServ not feasible in MANETs
ScalabilityState information increases with number of flowsStorage and processing overhead
RSVP signaling packets use bandwidth needed to send data packetsBurden on hosts
Mobile hosts must perform admission control, classification of all incoming data packets, and priority scheduling
11
DiffServ ModelDifferentiated Service (DiffServ) model
Traffic separated into small number of classesRouting decisions based on class of packetNo per-flow state
Scalable modelLimited processing for routers
Example servicesPremium service: low loss, low delay, low jitter, end-to-end bandwidth guaranteeAssured Service: better than best effort serviceOlympic Service: three tiers of services
Service Level Agreement (SLA) used to receive DiffServAgreement between customer and ISP
12
DiffServ Model for MANETsNo per-flow state ensures scalability DiffServ may be feasible for MANETs
Premium service impossible to supportAssured service possible
SLA difficult to implement in MANETs
13
New QoS Model for MANETsFlexible QoS Model for MANET (FQMM)
Hybrid approach: combines features from IntServ and DiffServmodelsPer-flow QoS for high priority flowsAggregate QoS for lower priority flowsSource node responsible for traffic shaping
Delaying packets belonging to flow to meet traffic profileMeet criteria such as mean rate, burst size
Issues with FQMMHow many per-flow sessions possible?How do intermediate nodes determine packet information?How should scheduling be performed at intermediate nodes?
New QoS models still needed for MANETs
14
QoS SignalingUsed to reserve and release resources when flows created, removed or changedInform application of success/failure of resource reservationTwo issues
Reliable exchange of QoS signaling informationIn-band signaling: control information with dataOut-of-band signaling: separate control packets
Interpretation of QoS signaling informationProtocols
RSVP, Insignia
15
In-band vs. Out-of-band Signaling
In-band signalingLow overheadCannot implement complex functionality
Out-of-band signalingAdds overheadHigher priority for signaling messages
Reduces effective bandwidth for data transmissionEasier to implement signaling protocol
16
RSVPQoS signaling for the InternetOut-of-band signaling systemRequest message sent via routing protocol to receiver
Request includes traffic specifications (rate, burst size)Receiver sends back a reservation message to sender
Intermediate routers check if they can support requested servicesIf so, allocate resourcesIf not, send error message to receiver
Receiver initiates resource requestFlow information periodically refreshedProblems for MANETS
Too much overhead to apply RSVP to MANETsNot adaptive to dynamic networks
17
MRSVPExtension of RSVP for cellular network with mobile hostsPredicts future locations and reserves resourcesActive and passive reservations
Other flows can use resources from passive reservationsNot suitable for MANETs
Unpredictability of mobile hosts’ future locationsCurrent topology different than future topology so making passive reservations does not make sense
18
InsigniaDesigned specifically for MANETs
In-band signalingBase and enhanced QoS levels
Per-flow managementResource management adapted as topology changesIntelligent packet scheduling Flow reservation, restoration and adaptation
QoS reports periodically sent to source nodeSource node takes action to adapt flows to observed network conditions
19
Insignia (cont.)Routing
Any routing protocol can be usedRoute maintenance procedure will affect QoS
In-band signalingEstablish, adapt, tear down reservationsControl information embedded in data packets
Admission controlDetermine whether or not to accept reservationRefresh reservation periodically based on current state
Packet schedulingWeighted round-robin for different flows
MACAny MAC protocol can be used
Automatic reconfirmation or de-allocation of reservation based on data packets received and timeouts
20
Insignia (cont.)Integrated in-band signaling, admission control and packet schedulingUseful for multimedia applications
Support multiple operation modes (max and min bandwidth modes)Some loss acceptable
MAC and routing protocols affect ability to support QoSCannot provide absolute guarantees
21
QoS Routing ProtocolsGoal: search for a path through the network that provides sufficient resources to meet QoS goals
E.g., delay, delay jitter, bandwidthConcave or additive metrics for paths
E.g., bandwidth is concave, whereby each link must satisfy minimum bandwidth constraintsE.g., delay is additive, whereby route delay is sum of individual link delays
NP-complete problem to find paths with two or more metrics
E.g., finding delay-constrained least-cost pathUse heuristics
22
QoS Routing ProtocolsDifficulties with QoS routing
Overhead highMaintaining link state information difficultCannot guarantee QoS as in wired networks
Route breaksNode failuresMust update paths with new paths that have enough resources– may not be possible
23
CEDAR ProtocolCore-Extraction Distributed Ad-hoc RoutingCore: approximation of minimum dominating set
Every node in core or neighbor of core nodeDominator of node: one core node neighborMaintenance of core as nodes move
Core nodes keep link state for high bandwidth, stable linksIncrease and decrease waves to inform core nodes of current state of links
Routes created using core path between source and destination as guidelineRoutes maintained
LocallyRe-initiate route set-up
24
Ticket-based QoS RoutingProbe packets (PKT) issued with certain number of “tickets”
Tickets determine maximum number of paths that can be probed forsuitabilityIntermediate nodes allocate tickets among neighbors
Choose neighbors most likely to satisfy QoS constraintRequires state information
ExampleSource A transmits PKT with 3 tickets to neighbor BB transmits PKT with 2 tickets to C, PKT with 1 ticket to DC transmits PKT with 1 ticket to E, PKT with 1 ticket to FD transmits PKT with 1 ticket to GEtc.
Multiple suitable paths found, select “best” one
25
Ticket-based QoS Routing (cont.)Reliability
Multi-path routingData sent independently on all pathsDestination keeps first copy of data to arrive, discards rest
“Best” path selected as primary path, others kept as back-up paths With resources reservedWithout resources reserved best effort
Provides nice trade-off between control overhead (based on number of tickets) and finding good feasible pathIssues
Performance depends on ticket-issuing and ticket-splitting proceduresGlobal state required at each node not scalable
26
QoS MAC ProtocolsTwo approaches:
Guaranteed resource reservationProvide service differentiation
Allow real-time/high priority packets to access channel before non-real-time/lower priority packets
Still meet the goals of MAC protocols
27
Cluster TDMANetwork organized into clusters
Select cluster head via Lowest-IDHighest degreeLeast cluster change
Cluster head assigns TDMA slots to nodesInter-cluster interference avoided via TDMA or CDMAFrame times synchronized throughout networkCreate virtual connections via assigning slotsFree slots used for best-effort traffic via slotted-ALOHA
28
MACA/PRMACA with Piggyback ReservationData and ACK packets include reservation information in headerReceivers keep reservation tablesBandwidth reservation
Reservation information in Data packets inform neighbors of transmitter of next transmissionReservation information in ACK packets inform neighbors of receiver of next transmissionReservation tables also shared among nodes
29
Providing QoS in IEEE 802.11Provide “better than best effort” serviceService differentiation via
Prioritization of different packetsFair scheduling
Tunable parametersContention windowBackoff algorithmInterframe spacing (IFSs)
30
IEEE 802.11eEnhanced DCF (EDCF)All nodes use DCF MAC protocolParameters set according to traffic requirements
Prioritizes traffic according to access category (AC)Adjust
IFSsMinimum and maximum backoff window sizesMultiplication factor for adjusting backoff window
Probability of accessing channel affected by these parameters
Set parameters such that high priority data has higher probability of accessing channel earlier
31
QoS in IEEE 802.11Other techniques for prioritization
Persistent Factor DCF: backoff geometrically distributed with parameter P based on packet priority
Fair-scheduling techniquesProvide fairness in allocation of bandwidth to different traffic classesOften cannot be implemented in existing standardDistributed weighted fair queueDistributed fair schedulingDistributed deficit round robin
32
Discussion