Self-Adaptation of Mobile Agent Population in Dynamic Networks: A Biologically Inspired Approach

In this paper, we consider the mobile agent population control problem in dynamic networks, and present biologically inspired solutions for the problem. The goal of this problem is to adjust the agent population a(t) at time i to a given constant fraction of the current network size n(t): a(t) = gamma middot n(t) for some constant gamma(0 < gamma les 1). To realize self-adaptation, we borrow an idea from the single species population model, which is a well-known population ecology model. This model shows that the population p(t) of a single species in an environment automatically converges to and stabilizes at f_a(t)/f, where f_a(t) is the amount of supplied food by the environment and f is the amount of food consumed by one individual to survive. In the proposed algorithms, agents are regarded as individuals of a single species, and nodes supply food for agents