2.4 GHz IEEE 802.15.4 Channel Interference Classification Algorithm Running Live on a Sensor Node Sven Zacharias, Thomas Newe, Sinead O’Keeffe, Elfed Lewis Optical Fibre Sensors Research Centre, Department of Electronic and Computer Engineering University of Limerick Limerick, Ireland {Sven.Zacharias, Thomas.Newe, Sinead.OKeeffe, Elfed.Lewis}@ul.ie Abstract—IEEE 802.15.4, the basic standard behind ZigBee, is a low-power communication standard mostly operating in the free 2.4 GHz Industrial, Scientific and Medical (ISM) frequency band. This band is heavily used by many technologies sending with more power, therefore avoiding interference is an important task and knowledge of the spectrum is essential. This paper presents an algorithm and its actual implementation that classifies sources of interference from a Received Signal Strength Indication (RSSI) noise floor reading of a single IEEE 802.15.4 channel. This algorithm can identify WLAN and Bluetooth traffic as well as the interference generated by microwave ovens with a second of RSSI measurements. It can run on off-the-shelf sensor node hardware. Since there is no need to change the channel, the node is continuously connected to its network with full message receiving capability. I. INTRODUCTION The IEEE 802.15.4 standard [1] provides low-power wireless connectivity among inexpensive devices for different topologies and applications. In the Physical Layer the worldwide available 2.4 GHz Industrial, Scientific and Medical (ISM) frequency band is mostly used, due to its availability and data rate (250 kbps). However, this frequency band is shared with many other technologies, namely IEEE 802.11 based Wireless Local Area Networks (WLANs) [2], Bluetooth (BT) [3] and microwave ovens (MWOs), as shown in Fig. 1. Due to this coexistence, a blocked medium or occurring collisions decrease the chance of IEEE 802.15.4 devices to communicate undisturbed. This leads to retransmissions increasing the power consumption, latency, and the probability of lost packets or violated application timing constraints. The ZigBee standard [4] recommends an energy scan on all channels if more than 25% of the transmission attempts fail. This takes the node off the network. Instead of measuring the noise floor on all channels, this work suggests an algorithm to scan a single channel for one second and to classify the sources of interference into one of the following groups: WLAN, BT or MWO. This has the following advantages:  Only sampling RSSI values of a single channel without changing the frequency allows to stay connected to the network all the time.  Sampling the noise floor (as for example suggested by ZigBee) is insufficient for long term planning. While MWOs use multiple channels heavily, they are seldom used longer than half an hour. On the other hand WLANs without traffic generate little interference, but they will stay on their channel and data traffic can increase depending on the network usage. Thus, it might be better in the long run to stay on a temporarily fully blocked channel interfered by a MWO and to change from a weakly interfered channel used by a co-existing WLAN.  After some classification results, further sensing can be stopped. If BT is detected as interference source, it can be assumed that no considerably less interfered channel will be found due to the frequency hopping of BT.  The result of the sampling is compressed to a class, allowing an easy exchange of results in a network, e.g. for central decision making and channel management. Spectrum sensing or channel classification (as done by the present algorithm) is a very important task for reliable, energy-conserving Wireless Sensor Networks (WSNs). The authors wish to thank the following for their financial support: The Embark Initiative (IRCSET) and Intel, who fund this research and Cost Action TD1001 for additional support. 11 13 12 14 15 16 17 18 19 20 21 22 23 25 24 26 2405 2410 2415 2420 2425 2430 2435 2440 2445 2450 2455 2460 2465 2470 2475 2480 (a) depending on model (c) (d) channel MHz 6 2437 11 2462 1 2412 channel MHz adaptive frequency hopping between 79 channels, each 1 MHz wide (b) ... ... Figure 1. Overview of the usage of the 2.4 GHz ISM frequency band: (a) microwave oven (b) Bluetooth (c) IEEE 802.11 (d) IEEE 802.15.4 978-1-4577-1767-3/12/$26.00 ©2012 IEEE Authorized licensed use limited to: University of Limerick. Downloaded on April 09,2021 at 08:26:34 UTC from IEEE Xplore. Restrictions apply.