Meta-routing over heterogeneous networks in M2M systems

Abstract

Machine-to-machine (M2M) allows autonomous computation systems to communicate with each other and to collaborately accomplish the works without human intervention. Each node in M2M network may have different communication interface such as Zigbee, Z-wave and Wi-Fi. We design a routing framework that can enable M2M to transmit data across heterogeneous network interfaces via selecting a routing path under given transmission policy. The framework can work in the dynamic network environment, such as including a new node or losting an existin node, the framework will reconstruct the route automatically without developer/administrator instruction at run-time. A new node broadcasts a route exploration message to the network when it joins to the network. Every node updates the routing table to the new node according to the exploration message. Then the new node will send request message to the neighbors to receive the reply messages from them. After collecting the reply messages, the new node can then build the routing table itself. When a node detects a connection failure, it can recover the routing table from nearby neighbors. The experiment results show that we show that the meta-route framework determine throughput policy with high throughput and high energy consumption, and energy policy with low throughput and low energy consumption.