Dynamic power management in a wireless sensor network using predictive control

Abstract

Technological advances have made wireless sensor nodes cheap and reliable enough to be brought into various application domains. These nodes are powered by battery, thus they have a limited lifespan which is a major drawback for their acceptance. This paper addresses a power consumption control problem of wireless nodes equipped with batteries. Dynamic power management is used to dynamically re-configure the set of sensor nodes in order to provide given service and performance levels with a minimum number of active nodes and/or a minimum load on such components. The power control formulation is based on model predictive control with constraints and binary optimization variables, leading to a mixed integer quadratic programming problem. Simulations are performed to demonstrate the efficiency of the proposed control method.

Feedback scheduling of sensor network activity using a Hybrid Dynamical Systems approach

“…A comparison between the hybrid control scheme and the MPC strategy proposed in [6] is performed here. The same WSN benchmark is considered, with the same profile of the perturbation as in scenario 2.…”

“…Here, we focus on solutions to (2), (6), insisting that they evolve in a specific set O where the remaining battery energy is non-negative for all nodes, and the input vector u has components equal to 0 or 1. We will also insist that, within this set, the flow set C is the closed complement of the flow set D relative to O.…”

“…With the pairwise rules in (5), we may then define an overall jump rule gathering together all of the relations in (5), and associated to a jump set D corresponding to the set where at least one pair of nodes is ready for a swap operation (namely x ∈ D if x ∈ D ih for at least one pair (i, h)): (6) where by construction, G is a set-valued mapping (multiple pairs may be ready to swap at the same time) that possesses the useful property of having a closed graph because its graph is the union of the (closed) graphs of g ih . Discretetime dynamics (6) together with (2) correspond to a hybrid description of the WSN system under the specific control action, using the notation in [7].…”

“…Section IV evaluates in simulation the HDS control strategy on a realistic benchmark. Comparisons with an MPC approach [6] are also provided. Section V summarizes the main results and draws future work directions.…”

“…[5] provides an overview of these techniques. Model Predictive Control (MPC) [6] has been used in order to select the devices to be in Active mode both to limit the WSN overall energy consumption and to extend its lifespan while the WSN fulfills a given "mission". Even if the results are appealing, the control law requires a O.…”