Decentralized P2P Network Coordination with an Adaptive Transmission Cycle Decision mechanism and a simplified pulse-coupled oscillator

This paper describes a P2P wireless network framework with numerous nodes that can communicate with their neighbors in a stable condition with no collisions of transmissions. Under the network convergence state, every node transmits periodically in order with a cycle shared in the network. To realize that situation, we apply a pulse-coupled oscillator technique to shift the timing of transmission of each node to avoid overlapping with neighboring nodes´ transmissions. Additionally, we propose an algorithm to find the suboptimal transmission cycle, preferred small for faster communication, depending on the number of nodes and the network topology based on a class of chromatic number problem in graph theory. We developed a multi-thread simulation platform for multiple node communications to verify the algorithm. Results demonstrate its feasibility both with simulations and with real devices by implementing the method onto an omnidirectional infrared communication device and an RF long-distance communication device.