Title: QoS Routing for mobile ad hoc networks
1QoS Routing for mobile ad hoc networks
- Chenxi Zhu, M. Scott Corson
- INFOCOM02
- 2002. 09. 18
- Presented by Dong Wook Kim
2Contents
- Introduction
- The Network Model
- The Path Bandwidth Calculation Problem
- The QoS Routing Protocol
- Simulation Results
- Discussions of the QoS and Be Protocols
- Conclusion
- Discussions
3Introduction 1/2
- Quality of service is more difficult to guarantee
in ad hoc network - Low capacity
- Wireless bandwidth is shared among adjacent nodes
- Route change
- Network topology changes as the nodes move
- The ability to provide QoS is heavily dependent
on how well the resources are managed at the MAC
layer
4Introduction 2/2
- The object
- Develop a QoS routing protocol for ad hoc
networks using TDMA - Establish bandwidth guaranteed QoS routes in
small networks whose topologies change at low to
medium rate - The protocol is based on AODV, and builds QoS
routes on an On-demand basis - (QoS requirement bandwidth of time slots)
5The Network Model 1/3
6The Network Model 2/3
7The Network Model 3/3
8The Path Bandwidth Calculation Problem 1/2
9(No Transcript)
10(No Transcript)
11(No Transcript)
12The Path Bandwidth Calculation Problem 2/2
- For the link (nk1 ? nk), only three sets of
slots, PBk1k ? PBk1k-1 ? PBk1k-2 are
calculated - this is sufficient because transmissions of links
further downstream do not interfere with
transmissions of (nk1 ? nk)
13(No Transcript)
14(No Transcript)
15PB3 calculation
16The QoS Routing Protocol 1/2
- Routing protocol is AODV
- Bandwidth is calculated on its path as a RREQ
packet is forwarded hop by hop - This excludes any node other than destination to
generate a route reply (RREP) - Except local reply when the route is
re-established - If BW(FP) lt R (required bandwidth), RREQ is
dropped and no RREP is generated - If a RREQ reaches the destination via a path P, a
route satisfying the bandwidth requirement has
been found
17When a source node (S) wants to setup a QoS route
for a flow a destination (D),
18(No Transcript)
19The QoS Routing Protocol soft-state of a QoS
route 2/2
- The state of a QoS route at a node
- NONE
- REQ
- A RREQ to set up the QoS route has been
processed, but the QoS route is not established
yet - RESV
- The QoS route has been set up and is used to
forward data packets - BRK_U
- The QoS route is broken at upstream of this node
- BRK_D
- The QoS route is broken at downstream of this
node - Transitions among these states are triggered
- receiving or transmitting a packet
- expiration of the timer associated with the state
- Link layer feedback
201) NONE ? REQ i) When the source sends a RREQ
When a non-source node receives and forwards a
RREQ When the destination receives a RREQ and
verified ii) These nodes record reverse
pointer iii) The length of the timer is
Route_setup_time 2) REQ ? NONE i) When no route
is setup and the timer expires 3) REQ ? RESV i)
When the destination sends out a RREP ii) When a
node on the route receives a RREP iii) These
nodes record forward pointer iv) The length of
the timer is Route_setup_time 4) RESV ? RESV
(refreshed) i) When the route is used to transmit
a data packet belonging to this flow ii) The
timer is reset to Route_life_time 5) RESV ?
BRK_U i) When no data packet arrives for
Route_life_time ii) QoS route is broken at the
upstream iii) the time is set to Route_setup_time
21- 6) BRK_U ? RESV
- When the QoS route which was broken at upstream
is restored - 1gt a data packet belonging to this flow arrives
(link is recovered) - 2gt a node nk receives a RREQ packet from node
nk1 - ? after calculating the bandwidth of the
path FPk and verify - ? sends out a RREP back to nk1 local
reply - (even though it may not be destination)
- what makes the local reply feasible
is the part of the original QoS route from - this node to the destination still
exists - local reply reduces the delay to
restore a broken route - ? a node sending a local reply also sends a
route hold packet (RT_HLD) towards - the destination
22- 7) BRK_U ? NONE
- i) If it cannot be restored when the timer
expires - ii) The slots TSkP are released
- 8) RESV ? BRK_D
- i) When a node finds the link to its downstream
breaks - ii) It sends a RERR towards the source
- iii) So, When it receives a RERR packet from its
downstream neighbor - 9) BRK_D ? REQ
- i) The source sends out a new RREQ as soon as it
receives the RERR - ii) When the node on the route receives and
forwards a RREQ packet - iii) The timer is set to Route_setup_time
- 10) BRK_D ? NONE
- i) If no RREQ arrives before the timer expires
- ii) The slots TSkP are released
- 11) RESV ? NONE
23Simulation Results
24(No Transcript)
25Discussions of the QoS and BE Protocols
- BE protocol (original AODV protocol)
- Whether QoS can be achieved in a highly mobile
network is questionable - All the packets are sent through at most one
route - Cause congestion on this route under heavy
traffic - QoS routing protocol
- Packets transmitted on QoS routes are guaranteed
of bandwidth - Provide load balancing
- Disadvantages
- A QoS RREQ often travels further than a BE RREQ
- A long QoS route is more difficult to build and
to maintain than a short one, especially under
mobility - It is not considered multiple QoS routes can be
set simultaneously - They may compete for the same set of slots and
interfere with on another
26Conclusion
- An on-demand QoS routing protocol based on AODV
is developed for TDMA-based mobile ad hoc network - The QoS routing protocol is likely useful only
for short routes, and in networks of low mobility - QoS routes should be built and used as complement
to, not substitute for, best-effort routes
27Discussions
- Correctness and reliability of the FA
- Why is it sufficient if we consider adjacent
three sets of slots? - If we use the FA, can we calculate correctly a
bandwidth between adjacent nodes? - Is there a obvious basis we are able to believe?
- The FA is made by these authors and doesnt be
proved any theories - Until now, most research areas have been limited
only to use TDMA as the MAC layer protocol - Generally, a node uses IEEE 802.11
- This area is worth researching with hard
perspiration!!!