Exploring Bio-inspired Algorithm for Service Discovery in Pervasive Environments

In pervasive computing, one of the key open challenges is the efficient discovery of the device that provides target services. This paper proposes a novel distributed service discovery algorithm in pervasive computing environments. This method is based on the basic principle of social network and bio-inspired intelligence. It exploits a social behavior and ant inspired query routing mechanism, which views query messages as artificial ants, and utilizes pheromone and social utility as routing hints that direct query messages to nodes owning more services. This paper presents generation and update rule of pheromone, routing policy for artificial ants as well. In order to avoid getting into local optimization, the roulette wheel technique and pheromone limitation policy are used in our algorithm. Simulation results show that our method significantly improves the search quality as well as reduces traffic cost compared with classical approaches, which supports high mobility and is suitable for pervasive environments.