HEIR: Heterogeneous interference recognition for wireless sensor networks

With the rapid development of wireless communication technology, a large number of wireless networks and devices that have different PHY and MAC layers coexist with each other. 2.4GHz Industrial, Scientific and Medical (ISM) band is becoming increasingly crowded. Wireless Sensor Networks (WSNs) which use low-power communication standard IEEE802.15.4 share the unlicensed spectrum with a plethora of other devices and technologies, such as WiFi systems underlying IEEE802.11, Bluetooth systems underlying IEEE802.15.1, and even non-communication appliance like microwave ovens. The ability to detect what radios are operating in the neighborhood is a fundamental need of WSNs, ranging from network management to network security. Since there are no explicit mechanisms to recognize such heterogeneous interference sources, WSNs often have no reasonable way to guard against them. In this paper, we describe the main working principle of heterogeneous interference sources, and present HEIR, a detector that is able to accurately identify heterogeneous interference sources. HEIR builds on the insight that there are hidden repeating patterns of signals which can be used to construct unique signatures and identify different types in most wireless protocols. The method can be implemented through signal sampling, interference estimation, feature extraction, and device classification. We show the experimental evaluation in an indoor testbed that HEIR is accurate in several different scenarios, and can live on sensor nodes´ hardware. Since no any channel changes, the network topology is not interrupted, and the stable communication in real-time is ensured then.