Bayesian Route Selection in Wireless Ad-Hoc Networks

The patent applies preliminary action by pre-establishing and maintaining routes to all potential destination nodes before actual communication occurs. Nodes proactively update their routing tables with route information, ensuring that when communication is needed, routes are already available without requiring reactive route discovery. This resolves the contradiction by preparing route information in advance, making route availability high while allowing selective updates only when topology changes occur, thus managing bandwidth consumption efficiently.

The patent combines reactive route establishment with preliminary action by maintaining observation tables that proactively collect and store route information from broadcast messages as nodes join or move in the network. When a route is needed, the system can quickly query these pre-collected observations rather than performing complete route discovery, thus reducing route establishment delay while avoiding continuous routing updates, thereby managing bandwidth consumption efficiently.

The patent extracts and separates route information from control messages by utilizing broadcast traffic (data or other control messages) to carry additional route observations. Instead of requiring dedicated routing control messages for every route update, the system extracts route information from existing broadcast messages and stores them in observation tables. This reduces the number of explicit routing control messages needed while maintaining route establishment reliability through accumulated observational data.

The patent applies local quality by maintaining observation tables with differentiated levels of detail for different destination nodes. Rather than uniformly tracking all possible route parameters for all nodes, the system maintains detailed observation data for frequently accessed or critical destinations while using simplified representations for less important routes. This selective approach improves connectivity probability accuracy for important routes while reducing the overall computational and storage complexity of observation table management.