In this paper we describe an approach for optimizing multi-hop broadcast protocols in ad-hoc mobile networks with an online, distributed machine intelligence solution. In our proposed framework not only runtime parameters of a predefined protocol are optimized, but the protocol logic itself also emerges dynamically. The model is based on genetic programming and natural selection: protocol candidates compete for being picked (natural selection), then survivors get combined with each other and/or mutated (genetic operators), forming the next generation of protocol instances. To achieve this we created (i) a genetic programming language to describe protocols, and (ii) defined a distributed, communication-wise non-intensive, stigmergic feed-forward evaluation and selection mechanism over protocol instances, and (iii) a budget based fair execution model for competing protocols. We show that the result of the online, autonomous protocol evolution outperforms traditional approaches, by adapting to the local situation, when used for multi-hop broadcast problem in ad-hoc mobile networks. Experiments confirmed 50% improvement with a random movement mobility pattern, and 66% improvement with a group based mobility pattern. The evolution also protected the system from the negative effects of initially present harmful protocols.
|Number of pages||15|
|Publication status||Published - Apr 1 2012|
ASJC Scopus subject areas
- Computer Science(all)
- Electrical and Electronic Engineering