Routing Protocol Comparison Josh Broch David A Maltz David B Johnson Yih Chun Hu and Jorjeta Jetcheva A Performance Comparison of Multi Hop Wireless Ad Hoc Network Routing Protocols In Proceedings of the Fourth Annual International Conference on Mobile Computing and Networking MobiCom 98 ACM Dallas TX October 1998 Shaan Mahbubani Overview Focus Protocols Simulation Paper Evaluation Discussion Focus infrastructureless networking Ad hoc networks Multiple hops Routing protocols Performance comparison Protocols Destination Sequence Distance Vector Temporal Ordered Routing Algorithm Dynamic Source Routing Ad Hoc On Demand Distance Vector Destination Sequence Distance Vector Hop by hop Distance Vector Bellman Ford Loop freedom Per node routing table Destination Next hop Number of hops Sequence number Temporally Ordered Routing Algorithm Link reversal Minimize communication Avoid overhead rather than shortest path Query Update Internet MANET Encapsulation IMEP Reliable in order delivery Link status BEACON HELLO Dynamic Source Routing Source routing Each packet complete route Route Discovery Route Maintenance Caching optimizations Ad Hoc On Demand Distance Vector Like DSR Like DSDV hop by hop routing route sequence numbers Route Request Route Reply on demand route discovery maintenance Number of hops to destination Maintenance Periodic HELLO Simulation Methodology Results Other observations Related work Methodology Goal Measure the ability of the routing protocols to react to network topology changes while continuing to successfully deliver data packets to their destinations Scenarios Movement Random Waypoint Model Communication constant bit rate no TCP Metrics Packet Delivery Ratio Routing Overhead Path Optimality Results Packet Delivery Best to worst DSR AODV TORA DSDV DSR AODV load independent DSDV stale routing TORA routing loops from link reversal Packet Delivery Details Results Routing Overhead Least to most overhead DSR DSDV AODV TORA DSR decreasing incremental overhead AODV less caching DSDV constant overhead TORA positive feedback loop congestive collapse Routing Overhead details Other Observations Overhead DSR worse than AODV when measured in bytes DSDV updates upon new sequence or metric number Sequence more overhead better packet delivery Broadcast reliability ARP on demand protocol interaction Does not cache enough waiting packets Related Work Park Corson TORA Over simplified simulation no node mobility Johnson Maltz DSR Lacked radio propagation MAC Freisleben Jenson DSDV DSR Simulation deficiencies synchronization Paper Evaluation Strengths Weaknesses Further discussion points Evaluation Strengths Realistic Model 802 11 MAC Physical layer Detailed Comparison 210 scenarios Varied pause time of data sources Effective metrics Packet delivery ratio Overhead Detailed considerations Considered implementation specific optimizations Proposed protocol specific reasons for performance differences Evaluation Weaknesses Comparing very specific aspect of protocol Only changes to network topology No load stress normal case overhead Not real enough Flat space no obstructions Too much movement No device locality Constant bit rate Scalability Effect of TCP Evaluation Overall Achieved what they said they would Other information needed for a more complete comparison Discussion Implementation Which protocol is easier better Node behavior Malicious lazy nodes Power consumption Scalability Security
View Full Document
Unlocking...