Real-Time Investigation of Cross-Technology Interference in Heterogeneous IoT Networks

Abstract

The use of unlicensed frequency bands has gathered much interest recently to meet the bandwidth requirements of beyond 5G networks. However, the frequency overlap between different radio technologies such as LTE-U, Wi-Fi, and ZigBee can cause severe interference and hamper communication reliability. This inter-technology interference is particularly unfavorable for IoT devices operating on the ZigBee network that shares the same unlicensed band with Wi-Fi which is capable of transmitting at 10-100 times higher power. Our real-time experiments study this transmission power asymmetry and examine the interference patterns between IEEE 802.11 Wi-Fi and IEEE 802.15.4 ZigBee networks at bitlevel granularity, the simulations were performed on the Zolertia Z1 platform. In this paper, we evaluate the system performance under different settings and the scalability of a real-world ZigBee network. We observed that under certain conditions, Wi-Fi activity can trigger a nearby ZigBee device to back off for a certain time and starve. We also demonstrate that most of the link disconnectivity and packet loss are mainly due to channel access failure rather than errors in data transmission. The observations illustrated in this work would help researchers to propose efficient solutions for heterogeneous IoT networks and design modern data communication protocols for ZigBee devices that are likely to operate in an environment that coexists with dense Wi-Fi deployment