Cognitive Power Adjustment for OFDM-Based Wireless Communication System: An Empirical Study

In this paper, we describe an empirical study on the impact of throughput performance when transmission power is adjusted. This experiment was conducted in an office environment for an OFDM-based wireless communication system such as WiFi 802.11g protocol network. We present a power selection mechanism based on intuitively learning one of the attributes which then triggers the power adjustment. We show that client's similar throughput performance can be obtained at reduced transmission power thus prevent an access point from maximizing power unnecessarily. Our cognitive power adjustment method is potentially useful in extremely dense wireless networks deployment since it is able to prevent from any interference and also to reduce energy consumption of the devices. Our evaluation of this mechanism can reduce the power to 5dB while maintaining the same level of throughput.