A Traffic Injection Framework to Support the Evaluation of Effects of Reconfigurations on Energy Consumption in Multi-RAT Networks

With the ongoing growth of mobile Internet usage numbers and the continuous deployment of additional mobile network nodes, the power consumption in mobile networks is increasing at dramatic rates leading to a huge amount of carbon emissions. Therefore, improving energy efficiency in mobile radio networks is of increasing relevance. The reduction of energy consumption in mobile radio networks can be achieved by adaptively reconfiguring mobile network components (e.g., switching on/off base stations) based on user behavior and predicted capacity demands for a certain cell or area. In the project Communicate Green, we evaluate and develop concepts and algorithms that can be utilized to apply a context-aware power management in heterogeneous Multi-RAT networks. One of the biggest problems we encountered during the evaluation of the proposed mechanisms in our test bed is the fact that test beds do not resemble realistic conditions due to a lack of actual users and therefore traffic. In this paper, we propose a Traffic Injection Framework to simulate various load scenarios in test beds. The architecture is implemented in the Multi-RAT Test bed in Berlin, Germany, which consists of multiple distinct radio access technologies. The Traffic Injection Framework features a distributed approach, where client applications are installed on the nodes of the test bed and remotely managed by a control server.