Long-Term LPWAN Sub-GHz Deep Indoor-to-Outdoor Channel Model

In the context of the Internet-of-Things (IoT), wireless connectivity of wide spread sensor networks for infrastructure and environment monitoring is one of the essential factors for implementation. Low-power wide-area networks (LPWAN) are cost-effective solutions for such applications. Their high maximum coupling loss makes them also appealing for deep indoor-to-outdoor applications. However, no suitable channel models for these scenarios are available, especially taking the long-term effects of stationary transmitters and receivers into account. Such long-term effects are e.g. generated by the opening or closing of doors. The consideration of these long-term effects are important for LPWAN as many use-cases only require a single successful transmission per day. To develop a suitable channel model we performed extensive long-term measurements in a suburban, a commercial and an industrial area. For these we used up to 50 special transmitters that enabled us to measure the channel characteristics over a period of several weeks with a temporal resolution of five minutes. Using thousands of measurement data we developed a generic indoor-to-outdoor channel model for LPWANs operated in license-exempt frequency bands based on statistical analyses of the obtained data.