The goal of this paper is to provide a fair performance comparison between 4 major ad-hoc routing protocols (by 1998) under different mobility conditions. The routing protocols under study are: DSDV, TORA, DSR and AODV.
The comparisons are done through simulations and one of the main contributions of this paper is the extension of ns simulator to better model node mobility, physical layer interactions, radio interfaces and 802.11 MAC protocol. The 4 different protocols are tested over 210 different scenario files (10,20 or 30 source nodes over 70 different movement patterns) for 1m/s and 20m/s maximum speed which are fed to the simulator. The pause time is the constant time between two consecutive moves of a node and its variation partially characterizes the mobility of the environment.
The metrics used for comparing the protocols are packet delivery ratio, routing overhead and path optimality. Besides DSDV which is a loop free distance-vector routing protocol, the other 3 are on-demand protocols and have varying overhead as pause time is changed from min=0sec to max=900sec. In general the Dynamic Source Routing protocol shows to perform superior to others in most of the mobility conditions even though the gap is narrowed at lower mobile environments. AODV which combines DSDV and DSR in order to remove the need for encapsulating the route information in the packet performs pretty close to DSR in general but in the most mobile scenarios in which it performs worse than DSR.
This paper gives a good overview of the main ad-hoc routing protocols and gives a fair comparison among them and therefore I recommend it for the syllabus. It would have been interesting to discuss their propagation model and weather testing for an indoor or outdoor environment could have a slightly different result.
The comparisons are done through simulations and one of the main contributions of this paper is the extension of ns simulator to better model node mobility, physical layer interactions, radio interfaces and 802.11 MAC protocol. The 4 different protocols are tested over 210 different scenario files (10,20 or 30 source nodes over 70 different movement patterns) for 1m/s and 20m/s maximum speed which are fed to the simulator. The pause time is the constant time between two consecutive moves of a node and its variation partially characterizes the mobility of the environment.
The metrics used for comparing the protocols are packet delivery ratio, routing overhead and path optimality. Besides DSDV which is a loop free distance-vector routing protocol, the other 3 are on-demand protocols and have varying overhead as pause time is changed from min=0sec to max=900sec. In general the Dynamic Source Routing protocol shows to perform superior to others in most of the mobility conditions even though the gap is narrowed at lower mobile environments. AODV which combines DSDV and DSR in order to remove the need for encapsulating the route information in the packet performs pretty close to DSR in general but in the most mobile scenarios in which it performs worse than DSR.
This paper gives a good overview of the main ad-hoc routing protocols and gives a fair comparison among them and therefore I recommend it for the syllabus. It would have been interesting to discuss their propagation model and weather testing for an indoor or outdoor environment could have a slightly different result.
Nice summaries! I know it is hard, but please try to keep them up to date ;-)
ReplyDelete