Mesh Networks: Commodity Multihop Ad Hoc Nets

Although much research has been done in the field of ad hoc networks, they have not seen common usage in the general populace, the DoD and disaster recovery have used ad hoc networks but in general they are unused. The system described uses wireless routers or bridges to extend the range of the current network and act as a router without the need for a networking cable. The paper discusses existing networks that do this and say that the current systems if integrating MANET algorithms would be able to easily extend their network quickly. The system uses sets of multiphop information to layer the levels of sending data along to other nodes in the network. The network relies on a wireless backbone to provide service for the mobile nodes and maintaining a complete graph creating a wireless backhaul communication system. The network would be simple to deploy and self managing allowing users to join or drop off the network quickly and easily. The systems are not well used because the availability of such a network is not easily acquirable but could be given little work. There are several IEEE specs that define addressing for small quickly changing networks that allow for maximal flexibility on these networks. One of the solutions is the WiMAX towers that are nowadays being used for 4G access on cell phones as well as internet like Verizon fios's wireless service. The combination of WiMAX and wifi can enable longer range high speed internet to smaller local wireless networks and form a large mesh network. The paper also describes various methods for reducing interference when various nodes can collide waves in the same space. Finally for routing data the network would determine at the time of sending what would be the 'best' route for the data to follow. The system is again designed for fair sharing which most current protocols are not.
I find the paper to have been rather uninformative save for combining the findings of various existing plans and ideas. Many of the technologies described are being used now but do not support the ad-hoc capabilites as of yet. Some of the design constraints described in the paper are good ideas for our network like ensuring routes if there exist any in case of drop out nodes or unexpected interference. The idea of a framework of wireless nodes with various ad hoc users sounds a rather lot like tamulink-wpa but with potential to be more versatile.