Utility max-min fair flow control for heterogeneous sensor networks

This paper considers flow control and resource allocation problem as applied to heterogeneous sensor networks with multipath routing. As heterogeneous sensor networks consist of diverse sensor types and applications, either elastic data collecting or real-time monitoring, the utility functions (a model of performance) of these applications do not consistently stay concave. By virtue of this, the traditional optimal flow control approach, which requires the utility function to be strictly concave, is no longer applicable. Therefore, we propose a novel utility-based flow control algorithm that is able to allocate the resources (both wireless channel usage and sensor node energy) efficiently and guarantee the application performance in a max-min fair manner. Furthermore, by combining a first order Lagrangian method and filtering mechanism, the typical oscillation behavior for the multipath network is alleviated and a rapid convergence property is possessed.