Mobile Ad Hoc Network

Friday, October 13, 2006

Proposal to New Route Caching Strategies in T-DSR

In the original implemention of DSR as outline in manet-dsr-ietf version 10, mobicache and linkcache are included into the ns-2 DSR implementation. However, as stated in [1], three main drawbacks have been identified:
  1. Incomplete error notification
  2. No expiry
  3. Quick pollution

Although [1] proposed three new mechanisms for the on-demand routing protocol, their results seem inadequate to further address the shortcomings of the routing overheads. The problems of keeping the caches up-to-date are still being studied. Constructive suggestion on improving the caching strategies in DSR are welcomed.

[1] Mahesh K. Marina, Samir R. Das, "Performance of Route Caching Strategies in Dynamic Source Routing", NSF Networking 1999


Thursday, June 08, 2006

Behavioural Ad Hoc Routing

Firstly, I would like to briefly walk you through the nature of the original DSR protocol:
This protocol is truly based on source routing whereby all the routing information is maintained (continually updated) at mobile nodes.
It has only 2 major phases which are Route Discovery and Route Maintenance.
Route Reply would only be generated if the message has reached the intended destination node (route record which is initally contained in Route Request would be inserted into the Route Reply).
To return the Route Reply, the destination node must have a route to the source node. If the route is in the Destination Node's route cache, the route would be used. Otherwise, the node will reverse the route based on the route record in the Route Reply message header (symmetric links).
In the event of fatal transmission, the Route Maintenance Phase is initiated whereby the Route Error packets are generated at a node. The erroneous hop will be removed from the node's route cache, all routes containing the hop are truncated at that point. Again, the Route Discovery Phase is initiated to determine the most viable route.
Proposal of my T-DSR:
At the moment of research, DSR is not yet a power-aware routing protocol. This feature could be implemented in ns-2's DSR whereby the routing protocol will only pick the mobile nodes with sufficient battery power for during route construction.
Improvement to the Route Reply process: I notice that the main drawback of this protocol the lack of support for alternative route in transporting the Route Reply message. This can be done by implementing the inter-Route Discovery feature during the Route Reply process. This is considered as an innovation to improve the performance of DSR. Somehow, I am still working around of realising the Theorectical Model into the C++ routing protocol. Another key issue that might be arising would be the enormous flood of the packets during route discovery. This is because at the same time there might be other mobile nodes in the same network initiating Route Discovery as well. Consequently, there must be a solution to reduce packet flooding and increase higher hit rate.
Restructural of Route Maintenance, instead of sending a route error message back to the source node, the intermediate node should at least probe for available intermediate nodes which is able of joining the broken route temporarily to transport the data packet to the destination node. Should the temporary route does not exist, generate route error back to the source node or the data packet has been sent to the destination node through the temporary route, generate route error back to the source node and initiate route discovery again to look for viable route. (Note: The temporary route will always be the critical route which might have longer hops to reach the destination).
Causes of Broken Link/Destination Unreachable
intermediate nodes moved away from the line of transmission of other mobile nodes which is involved in routing information,
intermediate nodes is busy transmitting other data packets for other source nodes,
route looping whereby the data packet looping around and yet has not reached the destination node.

Radio Propagation Model

Before redesigning the PacketHandler() in ns-2, there is one issue that come across my mind which is the radio propagation model. They are free space model, two ground reflection model and the shadowing model. Selecting the right propagation model is quite problematic due to unique properties of each propagation model.
Free Space Model assumes both linked nodes are in the line of sight of each other.
Two Ground Reflection Model considers both direct path and ground reflection path, thus more accurate prediction at long distance.
Shadowing Model takes into account of multi-path fading (with obstacles) whereby nodes are not transmitting in an ideal circle.
Again, there is somehow related to the equating of deriving number of hops, 2-D or 3-D.

Any suggestion on the propagation model?

Optimum Number of Hops

In general, a multi-hop route could have any number of hops, n. In order to assess the average network performance, I constructed a mathematical model of two planary equation whereby the optimum number of hop, n = diameter of the corresponding circular area of the network / distance between the 2 linked nodes. This equation does apply on 2-dimensional area. Does it apply to 3-dimensional area as well? Assume that each node is beaconing on unidirectional antenna. Your view on this problem is highly appreciated. Thank you.