Metrics on Energy Efficiency for Cognitive Green Equipment Based on FPGA Platform

Abstract

In order to efficiently manage the power consumption of communication devices, a management architecture is necessary to sense the working state of the equipment and the ambient environment. Therefore, it is important to properly select energy-efficient metrics that reflect the working environment. Then, on the basis of the metrics obtained, an optimal decision can be made and is followed by a reconfiguration action, the aim of which is to minimize the power dissipation while not compromising the performance. In this paper, we would like to introduce some useful metrics for a cognitive management architecture when it is implemented on a field-programmable gate array platform and the methods used to get them. We also study the use case of some of the metrics, and the results are useful in providing the necessary information for decision making. There is a tradeoff between the power consumption, the performance, and the resources. A decision maker has to choose an optimal solution according to the working environment.