Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016
The ACTHEON project is being financed by the Ministerio de Economía, Industria y Competitividad as part of the programa de estatal de investigación, desarrollo e innovación orientada a los retos de la sociedad.
The concern about energy consumption and energy efficiency as a paramount resource in the design and operation of telecommunication networks has emerged as a research challenge during recent years, as a result of the global coverage and massive penetration of information and communication technology (ICT). The challenge of reducing as much as possible the energy expenditure of ICT infrastructures has motivated the development of new, efficient operating procedures in many devices, networking equipment and network architectures, leading to a substantial reduction of the overall energy consumption without noticeable impact on their basic performance metrics. These mechanisms, however, have been usually added as an afterthought, a feature added to the normal operating state. In the next-generation networks envisioned today, the need for energy-efficient operating techniques must be an upfront design requirement, on the same level as other traditional design requirements.
In this research project we aim at innovating and devising the mechanisms, mathematical models and practical algorithms suitable for a substantial increase in the achievable energy savings in the ultra-dense data networks that will be installed during the next few years. Our approach combines three tools: optimization of the duty cycle, matched to the actual traffic load, in different types of devices; optimization of network-level algorithms, particularly routing and traffic aggregation schemes; and the proposal of anticipatory networking, i.e., the massive use of metering, sampling and data analysis techniques to predict the amount of resources needed in the network and to plan accordingly an optimal resource allocation. Specifically, the project will address these objectives:
- Design and assess energy-efficient algorithms based on the optimization of the duty cycle to the real traffic load, in several elements of ultra-dense networks (backhaul subnetwork, wireless access network, users’ terminals).
- Design and assess new energy-efficiency techniques by exploiting machine learning and pattern recognition of traffic and user mobility in the spatio-temporal domain.
- Design and evaluate global network algorithms for energy saving, particularly in the areas of efficient routing and optimal traffic aggregation using the capabilities provided by software defined networks.
The results of this project will bring a significant improvement over current energy consumptions standards in ultra dense networks, expected to reach up to 50%, and will be a step forward to make these networks able to operate with energy consumption genuinely proportional to the traffic load.