Wireless Channel Modeling for Drone to Ground 2.4 GHz link

Abstract

Use of Drones for communication is picking up the pace and expected to grow exponentially in the coming years. These drones will be used as access points for providing WiFi services in the near future. It is well known that WiFi was designed for low mobility scenario and the conventional transceivers do not work when drones move resulting in significantly higher Doppler frequency. This motivates us to design a modified physical (PHY) layer characteristics for a drone to ground communication in the WiFi band. Wireless Channel modeling of such drone-to-ground links is very important for this objective. The 2.4 GHz and 5 GHz bands (typically known as the WiFi bands) are unlicensed bands and hence most of the wireless devices use this band for communication. Developing a channel model for this band with a transmitter in motion is a challenging task. In this project, we have designed and developed a channel measurement system at 2.4 GHz link by conducting measurement campaigns to collect data and post-processing it. We model the channel as a finite impulse response (FIR) filter having tap coefficients that are stochastic in nature. The transmitter device HackRF/BladeRF carried by the drone, continuously transmits pilot data and the Universal Serial Radio Peripheral (USRP) receives the pilots. The drone is made to hover at a given distance and data is collected by USRP placed on the ground. The received data is post processed and the channel is estimated using known estimators like Maximum Likelihood (ML) and Minimum Mean Square Error (MMSE) estimators. The estimated channel is represented by 20 tap FIR filter. The histograms of the magnitude of estimated filter tap are fitted to known distribution with suitable parameters. The above procedure is repeated for different terrains and for varying distances between the drone and the ground at several different locations in IISc campus. A draft channel model is presented based on the measurement data