Beamforming Software Defined Radios for Wireless Sensor Network Interference Evaluation

William Zhao

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2020-101

May 29, 2020

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2020/EECS-2020-101.pdf

Wireless communication on the 2.4GHz ISM band is becoming ever more prevalent in buildings. Additionally, various physical layer technologies share this RF spectrum (e.g IEEE 802.11 WiFi, Bluetooth, IEEE 802.15.4). As device density increases, devices sharing different physical layer technologies will be forced to transmit on the similar frequencies as each other, causing interference with each other's transmissions. Low power wireless sensor networks (WSN) utilizing IEEE 802.15.4 transmissions are particularly impacted by comparatively high powered WiFi emissions. As a result, previous research work has been done on network protocols to allow WSNs to mitigate the effect of cross-technology interference (CTI) produced from sources such as WiFi. However, the evaluation for these protocols differ protocol to protocol including the method of WiFi interference generation. Many factors that affect the CTI observed by the WSN such as testbed layout, testbed container shape, CTI location, and antenna parameters makes it difficult to precisely adjust the levels of CTI on a WSN.

In order to provide a dynamic CTI generation platform that can be tuned to impact different devices within a WSN with different levels of interference, we develop a software defined radio (SDR) solution. We propose Software Defined Interference Generation (SDIG) as a CTI generation method to induce WiFi CTI on motes instead of using real WiFi transceivers. Notably SDIG uses an SDR array which allows beamforming techniques to steer the interference signal to impact different devices with varying magnitudes of interference.

We evaluate SDIG in comparison to real and arbitrarily placed CTI transmitters. We demonstrate the system's ability to target individual nodes and deliver precise amounts of interference with minimal effects on other devices. However, we also reveal the scaling limitations of using SDIG to simultaneously impact many devices, each with varying amounts of interference. We recommend the use of beamforming in SDIG as it provides testbeds a way to carefully induce a level of CTI to a specific device.

Advisors: David E. Culler

BibTeX citation:

@mastersthesis{Zhao:EECS-2020-101,
    Author= {Zhao, William},
    Title= {Beamforming Software Defined Radios for Wireless Sensor Network Interference Evaluation},
    School= {EECS Department, University of California, Berkeley},
    Year= {2020},
    Month= {May},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2020/EECS-2020-101.html},
    Number= {UCB/EECS-2020-101},
    Abstract= {Wireless communication on the 2.4GHz ISM band is becoming ever more prevalent in buildings. Additionally, various physical layer technologies share this RF spectrum (e.g IEEE 802.11 WiFi, Bluetooth, IEEE 802.15.4). As device density increases, devices sharing different physical layer technologies will be forced to transmit on the similar frequencies as each other, causing interference with each other's transmissions. Low power wireless sensor networks (WSN) utilizing IEEE 802.15.4 transmissions are particularly impacted by comparatively high powered WiFi emissions. As a result, previous research work has been done on network protocols to allow WSNs to mitigate the effect of cross-technology interference (CTI) produced from sources such as WiFi. However, the evaluation for these protocols differ protocol to protocol including the method of WiFi interference generation. Many factors that affect the CTI observed by the WSN such as testbed layout, testbed container shape, CTI location, and antenna parameters makes it difficult to precisely adjust the levels of CTI on a WSN.

In order to provide a dynamic CTI generation platform that can be tuned to impact different devices within a WSN with different levels of interference, we develop a software defined radio (SDR) solution. We propose Software Defined Interference Generation (SDIG) as a CTI generation method to induce WiFi CTI on motes instead of using real WiFi transceivers. Notably SDIG uses an SDR array which allows beamforming techniques to steer the interference signal to impact different devices with varying magnitudes of interference.

We evaluate SDIG in comparison to real and arbitrarily placed CTI transmitters. We demonstrate the system's ability to target individual nodes and deliver precise amounts of interference with minimal effects on other devices. However, we also reveal the scaling limitations of using SDIG to simultaneously impact many devices, each with varying amounts of interference. We recommend the use of beamforming in SDIG as it provides testbeds a way to carefully induce a level of CTI to a specific device.},
}

EndNote citation:

%0 Thesis
%A Zhao, William 
%T Beamforming Software Defined Radios for Wireless Sensor Network Interference Evaluation
%I EECS Department, University of California, Berkeley
%D 2020
%8 May 29
%@ UCB/EECS-2020-101
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2020/EECS-2020-101.html
%F Zhao:EECS-2020-101