Chapter 3: Zone-Based Link Stability Model and Analysis
3.7 The Comparison of Theoretical Analysis and Simulation
In order to verify the analytically derived expressions for link lifetime in different zones, we compare each of them by using ns2 network simulator collected the statistic data from simulation. Figure9. (with strong zone range is 0.8R) and Figure10. (with strong zone range is 0.9R) compare the corresponding theoretical analysis with simulation results. As we expected, we can observe apparently the simulation results are in completely good agreement with the theoretical analysis.
From these figures, we can see the closest transmission range (r=0.9R), the longer link lifetime in strong zone as a shorter link lifetime in weak zone with respect to a particular relative velocity.
0 5 10 15 20 25 30 35 40
Expected Link Lifetime v.s. Relative Velocity in different zone
Node Velocity Vr (m/s)
Figure 13 Expected link lifetime in transmission, strong, and weak zone (r=0.8R)
0 5 10 15 20 25 30 35 40 0
5 10 15 20 25 30 35
Expected Link Lifetime v.s. Relative Velocity in different zone
Node Velocity Vr (m/s)
Expectied Link Lifetime (s)
Analysis Transmission Zone Analysis Strong Zone Analysis W eak Zone
Simulation Transmission Zone Simulation Strong Zone Simulation W eak Zone
Figure 14 Expected link lifetime in transmission, strong, and weak zone (r=0.9R)
Chapter 4: An Adaptive Routing Strategy
4.1 AODV Routing Protocol
AODV(Ad Hoc On Demand Distance Vector) [18] is a distance vector routing protocol. It does note require nodes to maintain routes to destinations that are not actively used. As long as the endpoints of a communication connection have valid routes to each other, AODV does not play a role. The protocol uses different messages to discover and maintain links: Route Requests (RREQs), Route Replies (RREPs), and Route Errors (RERRs). These message types are received via UDP and IP header processing applies.
AODV uses a destination sequence number for each route entry. The destination sequence number is created by the destination for any route information it sends to requesting nodes. Using destination sequence numbers ensures loop freedom and allows to know which of several routes is more "fresh". Given the choice between two routes to a destination, a requesting node always selects the one with the greatest sequence number.
When a node wants to send data packets to another one, it broadcasts a RREQ to all the network till either the destination is reached or another node is found with a
"fresh enough" route to the sequence (a "fresh enough" route is a valid route entry for the destination whose associated sequence number is as least as that contained in the RREQ). Then a RREP is sent back to the source and discovered route is made available.
Nodes that are part of an active route may offer connectivity information by broadcasting periodically local Hello messages (special RREP messages) to its immediate neighbors. If Hello messages stop arriving from a neighbor beyond some
given time threshold, the connection is assumed to disconnect.
When a node detects that a route to a neighbor node is not valid it removes the routing entry and sends a RERR message to neighbors that are active and use the route (this is possible by maintaining active neighbor lists). This procedure is repeated at nodes that receive RERR message. A source that receives an RERR can re-initiate RREQ message.
4.2 Problem Statement
Existing on-demand ad hoc routing protocols will re-initiate route discovery only after the links on the routing path are broken. Once this situation occurs, there will be a significant cost both in detecting the disconnection (for example: AODV will send Hello message to check the links on the routing path whether broken or not) and re-establishing a new routing path (after found that the link is broken). The overhead for on-demand routing protocol is less than table-driven routing protocols since table-driven protocols attempt to maintain consistent and up-to-date routing information among all mobile nodes in the network. But for on-demand routing protocols, the overhead will occur while finding one or more links broken on the routing path since it needs to be re-initiate a route request procedure. Thus, we focus on how to reduce the overhead by developing our efficient adaptive routing strategy based on our zone-based link stability model. Specifically, when a mobile node with respective to another node are moving far from the strong zone and entering the weak zone (an unstable weak), a reroute update message is needed to be considered. By this protection mechanism in advance, we can reduce the overhead as a fail path occurs.
With this early warning, the source can initiate route discovery early and switch to another more stable routing path.
4.3 An Adaptive Routing Strategy
In order to improve the drawback on reactive (on-demand) routing protocol, we develop Zone-Based Routing (ZBR) strategy based on our link stability model. Our basic idea is that we can obtain distinct knowledge that a mobile node is either in strong zone or in weak zone from our model to design a highly adaptive and efficient routing strategy. There are two stages in on-demand routing protocol: (1) Route Discovery; (2) Route Maintenance.
In route discovery stage, we adopt the routing mechanism in SSA [2]. A source node tries to search a shortest routing path to destination node with strong links (that is mobile node in strong zone) according our link stability model. If we could not find a routing path with all-strong links, we would find the shortest path with all available links with either strong or weak links. By establishing such a stable routing path, we will have a robust communication for delivering data packets.
In route maintenance stage, the link on the routing path may break since each mobile node is movable. In AODV routing protocol, it monitors this routing path whether valid or not at every time unit (the default value is one second). If it found one or more links on this routing path fail, it will re-initiate a route request message to re-establish a new routing path if the data packets desire this transmission. In our routing strategy: Zone-Based Routing (ZBR) will monitor the link status at every period: d second. Our scheme adopts “half link lifetime in strong zone” while mobile node is in strong zone and “half link lifetime in weak zone” while mobile node is in weak zone to be the monitoring period time. However, in order to avoid the mobile node existing out the transmission range with respect to another mobile node too fast with high velocity, once the mobile node enters the weak zone, we will add another constraint to avoid this situation occurring. While a mobile node is in weak zone, we will not only monitor the link status in every “half link lifetime in weak zone” but also
add a mechanism: comparing the residual link lifetime (RLL) if RLL is greater than
“half link lifetime in weak zone” at each monitoring. If this monitoring mechanism found the average monitoring period “half link lifetime in weak zone” is greater than residual link lifetime (RLL), then a route update message will be generated and send to source. Upon receiving this message, the source will find another stable routing path and delivery the later data packets on this routing path.
More specifically, the procedure of our ZBR routing strategy is as following flow chart:
Figure 15 Zone-Based Routing (ZBR) Procedure Flow Chart
4.4 Simulation Results and Analysis
4.4.1 Performance Metrics:
There are three performance metrics to compare in our simulation. We evaluate our ZBR (Zone-Based Routing) scheme and AODV routing protocol according the following three metrics:
(1)Packet delivery ratio: The ratio between the number of packets originated by CBR sources and the number of packets received by the CBR sink at the final destination.
(2) Average end-to-end delay: The average time delay between the time when a data packet delivered from the source node and the time when the packet arrives at the destination node.
(3)Routing overhead: Packet overhead is the number of routing packets
“transmitted” per data packet that is “delivered” at the destination node.
Packet delivery ratio is important as it describes the loss rate that will be seen by the transport protocols, which in turn affects the maximum throughput that the network can support. This metric characterizes both the completeness and correctness of the routing protocol.
Average end-to-end delay is a major parameter to measure latency time while delivering a packet whose time it takes. It dependents to overhead since a larger overhead may lead longer delay. However, a shorter delay may not necessarily imply a higher packet delivery ratio since delay is only measured on those successfully delivered packets.
Routing overhead is an important metric as it measures the scalability of a protocol, the degree to which it will function in congested or low-bandwidth environments. Protocols that send large numbers of routing packets can also increase
the probability of packet collisions and may delay data packets in transmission queues of network interface.
4.4.2 Simulation Results and Analysis:
For all simulations, the communication patterns were peer-to-peer, with each run having either 10, 20, or 30 traffic sources sending 4 packets per second. Traffic sources are CBR (constant bit rate) traffics and each data packet size is 512 bytes. The source-destination pairs are distributed randomly over the network. The other simulation parameter setting is mentioned previously in Chapter 3.6.
Each node starts its journey from a random source location to a random destination with a random speed that is uniformly between 0–20 meter/second. Once the destination is arrived, another random destination will be chosen after a pause time. The total simulation time is 900 seconds and each data point in flowing figures is average of five runs with the same scenario configuration but with different random seeds.
In Figure 16、17、18, we plot the results we have obtained with respect to the packet delivery ratio with 10, 20, and 30 traffic sources. In all the testing scenarios, ZBR demonstrate high quality in delivering packets – most of all are more than 93%.
Since our routing path setup on the strong link which is more stable than the ones on AODV (which only consider establishing a shortest path). Furthermore, AODV has difficulty when nodes are moving fast (corresponding to smaller pause time), with a packet delivery ratio less than 90%. However, our ZBR routing can keep a high packet delivery ratio above in all scenarios more than 90%.
In Figure 19、20、21, we portray the results we have obtained to capture the performance with respect to average delay metrics. The AODV have a shortest
end-to-end delay since it setup the shortest path corresponding to less links on this routing path than our ZBR routing. Our ZBR routing path is composed of the strong link that may experience more hops. However, the average delay for our ZBR routing is more than AODV slightly. The delay cure for our ZBR routing is smoother than AODV that means we have a smaller jitter. Therefore, we have a more stable routing path and can provide some Quality of Service.
In Figure 22、23、24, we show the results we have obtained to present the overhead metrics to both of AODV and ZBR in 3 scenarios. As we can see, without the frequent periodic hello message to monitor the link status, our ZBR routing outperforms AODV. Compared to AODV (the default checking period to check link is one second each time), our ZBR routing strategy takes much less times to check link status by half expected link lifetime in strong and weak zones depending on the mobile node located in strong zone or weak zone. Moreover, under the scenario of 10 traffic sources, our ZBR routing perform very well, it will have a performance that approximates to zero overhead.
0 100 200 300 400 500 600 700 800 900
50 nodes, 10 connections (delivery ratio)
Pause Time (s)
Delivery Ratio
ZBR AODV
Figure 16 Packet delivery ratio with 10 connection traffic sources
0 100 200 300 400 500 600 700 800 900
50 nodes, 20 connections (delivery ratio)
Pause Time (s)
Delivery Ratio
ZBR AODV
Figure 17 Packet delivery ratio with 20 connection traffic sources
0 100 200 300 400 500 600 700 800 900
50 nodes, 30 connections (delivery ratio)
Pause Time (s)
Delivery Ratio
ZBR AODV
Figure 18 Packet delivery ratio with 30 connection traffic sources
Figure 19 Average delay with 10 connection traffic sources
0 100 200 300 400 500 600 700 800 900
7 x 10-3 50 nodes, 10 connections (average delay)
Pause Time (s)
Average Delay
ZBR AODV
0 100 200 300 400 500 600 700 800 900
7.5x 10-3 50 nodes, 20 connections (average delay)
Pause Time (s)
Average Delay
ZBR AODV
Figure 20 Average delay with 20 connection traffic sources
0 100 200 300 400 500 600 700 800 900
7.5x 10-3 50 nodes, 30 connections (average delay)
Pause Time (s)
Average Delay
ZBR AODV
Figure 21 Average delay with 30 connection traffic sources
Figure 22 Packet overhead with 10 connection traffic sources
Figure 23 Packet overhead with 20 connection traffic sources
0 100 200 300 400 500 600 700 800 900
Figure 24 Packet overhead with 30 connection traffic sources
0 100 200 300 400 500 600 700 800 900
0 1 2 3 4 5 6 7
50 nodes, 30 connections (overhead)
Pause Time (s)
Overhead
Chapter 5: Conclusion and Future Work
In this thesis, we derive an analytical formulation model about link lifetime considering the link stability in different zone. In this model, we develop a series of analytical expressions including: (1) An expected link lifetime in strong and weak zone；(2) An PDF and CDF distribution in strong and weak zone, respectively；(3) The residual link lifetime time distribution；At the same time, we design a highly efficient and adaptive routing strategy based on our zone-based link stability model.
By establishing such a zone-based link lifetime model, we can have distinct knowledge to know the mobile node is located either in strong zone or in weak zone and according to such information to determine a stabile and robust routing path and provide a protection mechanism to avoid the overhead while link failed.
By such a scheme, we can provide a stale routing path on the routing discovery stage since we establish the routing path based on strong links. Furthermore, on the route maintenance stage we can decrease the network overhead and increase the packet delivery efficiency by providing a protection mechanism: prior route update message while a node is far from the weak zone and immediately inform next packets transfer through another routing path before node leaves the transmission range according to our efficient link stability efficient monitoring mechanism before the current link really breaks.
Our results indicate that our scheme: adaptive routing based on zone-based link stability model can reduce the network overhead over 50% compared to AODV. Our ZBR can also increase packet delivery benefit efficiently (our ZBR have more than 90
% packet delivery ratio under all scenarios) since we can efficiently detect the link situation early and provide another stable routing path before link fails since decreasing the probability of packet retransmission.
Since the desirable features of routing protocols for MANETs include ability to adapt to frequent changing network conditions due to mobility and provide quality control mechanisms during the life time of a route. Hence, how to provide a suitable Quality of service (QoS) in MANETs is a challenge. In our zone-based link stability model, we can gather information about signal strength to find a more stable route during route discovery and uses this information in route choosing to provide a QoS guarantee communication depending on the desired service. In the future, we will try to design a QoS aware routing algorithm to provide a better QoS choosing metric under various QoS requirements.
Reference:
[1] P. Samar, S.B. Wicker,”On the Behavior of Communication Links of a Node in a Multi-Hop Mobile Environment," Proceedings of The Fifth ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc) 2004, Tokyo, May 2004
[2] R. Dube, C. Rais, K.-Y. Wang and S. Tripathi, “Signal stability based adaptive routing (SSA) for ad hoc networks," IEEE Personal Communications, Feb. 1997, pp. 36-45.
[3] S. M. Jiang, D. J. He, and J. Q. Rao, “A prediction-based link availability estimation for mobile ad hoc networks,” in Proc. IEEE INFOCOM, vol. 3, Anchorage, AK, Apr. 2001, pp. 1745–1752.
[4] Shengming Jiang, “An enhanced prediction-based link availability estimation for MANETs” Communications, IEEE Transactions on Volume 52, Issue 2, Feb.
2004 Page(s):183 – 186
[5] A. B. McDonald and T. F. Znabi, “A path availability model for wireless ad hoc networks,” in Proc. IEEE WCNC, New Orleans, LA, Sept. 1999, pp. 35–40.
[6] S. M. Jiang, D. J. He, and J. Q. Rao, “A Prediction-Based Link Availability Estimation for Routing Metrics in MANETs” [Online]. Available:
http://www1.i2r.a-star.edu.sg/~jiangsm
[7] M. Gerharz, C. de Waal, M. Frank, P. Martini, “Link Stability in Mobile Wireless Ad Hoc Networks," IEEE LCN, Nov. 2002.
[8] E.M. Royer and C-K Toh, “A review of current routing protocols for ad-hoc mobile wireless networks” IEEE Personal Communications, April 1999.
[9] G. Lim, K. Shin, S. Lee, H. Yoon, J.S. Ma, “Link Stability and Route Lifetime in Ad hoc Wireless Networks," 2002 International Conference on Parallel Processing Workshops (ICPPW'02), Aug. 2002, Vancouver, Canada.
[10] A. B. McDonald and T. Znati, “A mobility based framework for adaptive clustering in wireless ad-hoc networks” IEEE Journal on Selected Areas in Communications, Special Issue on Ad-Hoc Networks, in press, Aug. 1999.
[11] A. B. McDonald and T. Znati, “Link availability models for mobile ad-Hoc networks” Technical Report TR99-07, University of Pittsburgh, Department of Computer Science, May 1999.
[12] McDonald, A.B., Znati, T., “Predicting node proximity in ad-hoc networks: a least overhead adaptive model for selecting stable routes” Mobile and Ad Hoc Networking and Computing, 2000. MobiHOC. 2000 First Annual Workshop on 11 Aug. 2000 Page(s):29 - 33
[13] Dan Yu, Hui Li, Gruber, I., “Path availability in ad hoc network,” ICT 2003.
10th International Conference Telecommunications on Volume 1, 23 Feb.-1 March 2003 Page(s):383 - 387 vol.1
[14] Gerharz, M., de Waal, C., Frank, M., Martini, P., “Link stability in mobile wireless ad hoc networks”, Local Computer Networks, 2002. Proceedings. LCN 2002. 27th Annual IEEE Conference on 6-8 Nov. 2002 Page(s):30 - 39
[15] Sridhar, K.N., Lillykutty, J., Rajeev, S., “Performance evaluation and enhancement of link stability based routing for MANETs” Parallel Processing Workshops, 2004. ICPP 2004 Workshops. Proceedings. 2004 International Conference on 2004 Page(s):133 - 140
[16] A. Boukerche and L. Zhang, “A performance evaluation of a preemptive on-demand distance vector routing protocol for mobile ad hoc networks”, Journal of Wireless Communications and Mobile Computing 4(1) (2004) 99–108.
[17] J. Broch, D.A.Maltz, D.B. Johnson, Y.-C. Hu and J. Jetcheva, “A performance comparison of multi-hop wireless ad hoc network routing protocols”, in: Proc. Of MOBICOM’98 (October 1998).
[18] C.E. Perkins, E.M. Royer and S.R. Das, “Ad hoc on-demand distance vector (AODV) routing”, IETF MANETWorking Group, Internet-Draft (March 2000).
[19] UCB/LBNL/VINT Network Simulator, http://www-mash.cs.berkeley.edu/ns/
[20] S. Agarwal, A. Ahuja, J.P. Singh and R. Shorey, “Route-lifetime Assessment Based Routing (RABR) Protocol for Mobile Ad-hoc Networks," IEEE ICC 2000, vol. 3, pp.1697 - 1701, New Orleans.
[21] M. Gerharz, C. de Waal, M. Frank, P. Martini, “Link Stability in Mobile Wireless Ad Hoc Networks," IEEE LCN, Nov. 2002.
[22] N. Sadagopan, F. Bai, B. Krishnamachari, A. Helmy,“PATHS: “Analysis of PATH Duration Statistics andtheir Impact on Reactive MANET Routing Protocols,” MobiHoc 2003, Annapolis, MD, June 2003.
[23] J. Broch, D.A.Maltz, D.B. Johnson, Y.-C. Hu and J. Jetcheva, “A performance comparison of multi-hop wireless ad hoc network routing protocols,” in: Proc. of MOBICOM’98 (October 1998).
[24] Azzedine Boukerche, “Performance evaluation of routing protocols for ad hoc wireless networks,” Mobile Networks and Applications 9, 333-342, 2004, Kluwer Academic Publishers, Manufactured in The Netherlands.