Fully distributed algorithms for blind rendezvous in cognitive radio networks

Abstract

Rendezvous process is the cornerstone to construct Cognitive Radio Networks (CRNs), through which a secondary user can establish a link for communication with its neighbor on a common channel. Although many blind rendezvous algorithms have been proposed which do not rely on a central controller or a common control channel, all of these works still rely on the global parameters such as the number of licensed channels N and the number of users. This paper aims to design fully distributed blind rendezvous algorithms only based on each user's local information. We first give the Synchronous Check & Hop (SCH) algorithm for two synchronous users where they start the rendezvous process at the same time. The SCH algorithm guarantees rendezvous in O(min{ka,kb} N) time slots where ka,kb are the corresponding number of sensed channels of these two users. Our main contribution is a fully distributed algorithm called Conversion Based Hopping (CBH), where each user only uses its identifier (ID) and its number of sensed channels. CBH guarantees rendezvous between two asynchronous users in O((max{ka,kb})2) time slots. To our knowledge, this is the first result with rendezvous time independent of the global parameter N. We also derive a lower bound of rendezvous time between two users as Ω((ka-kg)(kb-kg)) where k_g is the number of their common channels. All of our results also apply to a more general blind rendezvous problem which we call Oblivious Blind Rendezvous where each user is free to assign their local labels to the sensed channels. Extensive simulation results compared with the state-of-the-art rendezvous algorithms corroborate our theoretical analyses.