This paper presents the Expected Transmission Count (ETX) metric, discussing its advantages and describing its design and implementation for DSDV and DSR routing protocols. Unlike the minimum hop-count metric, the ETX metric accounts for (potentially asymmetric) link loss ratios and interference between successive links to find the optimal path - the one with the highest throughput. The EXT metric will make such a network as the multi-hop rooftop (discussed in the previous paper) practical to implement. Initially, the authors point out the various shortcomings of the minimum hop-count metric.
Upon describing the use of the ETX metric with DSDV and DSR routing protocols, the performance is evaluated with respect to link loss ratios and interference. One thing that confused me is why Figures 2 and 6 both display crossing lines - how can the DSDV hopcount route achieve higher throughput in that small period than the best static route? Overall, this paper was interesting but several points were rather obvious. The analysis of the effect of link loss ratios was especially thorough. This probably belongs in the syllabus.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment