Spectrum-sensing-based WiFi performance emulator for experimental evaluation of cognitive solutions

As the wireless medium is becoming more and more congested, a better use of the scarce frequency spectrum is needed to deliver the required Quality of Service (QoS) to clients. Especially in unlicensed bands, wireless systems suffer from intra-network interference [1], e.g., in professional wireless conferencing environments [2]. Therefore, cognitive solutions are being developed, both on radio and networking level [3, 4]. A way of approaching cognitive radio is by relying on spectrum sensing within the considered environment and building a real-time Radio Environment Map (REM), which visualizes Received Signal Strength Indicator (RSSI) values, power values, or even Mean Opinion Scores (MOS) scores. If nodes in the environment could be configured themselves, the REM and its corresponding output data can be used to test cognitive decision engine (CDE) interference scenarios. This paper will describe a system that allows experimentally emulating WiFi traffic in an actual (pseudo-shielded) environment and visualizing the related QoS and QoE (Quality of Experience) parameters. The emulator allows testing cognitive corrective actions of a CDE. The outline of the paper is as follows. In Section II, the architecture of the system and its different components are discussed. Section III describes the three emulation scenarios that are executed. Section IV discusses the emulation results and finally, Section V summarizes the conclusions of this paper.