PROJECT TITLE :

DaRe: Data Recovery through Application Layer Coding for LoRaWAN

ABSTRACT:

Internet of Things (IoT) solutions are being implemented for smart applications in an increasing number of cases. In order to ensure efficient communication for the growing number of smart applications, there is a pressing need for Low Power Wide Area Network (LPWAN) technologies that are both cost-effective and have a broad range. LoRaWAN is a low-cost and energy-efficient low-power wide-area network (LPWAN) solution that is gaining popularity rapidly all over the world. However, in its default configuration, LoRaWAN does not guarantee a reliable communication link between devices. There is a risk of losing transmitted frames due to the effects of the channel and the mobility of the end devices. In this study, we perform extensive measurements on a new LoRaWAN network in order to characterize the spatial and temporal properties of the LoRaWAN channel. Specifically, we are interested in the effect that interference has on data transmission. In our deployment, the empirical outage probability for the farthest measured distance of 7.5 kilometers from the closest gateway is as low as 0.004, but the frame loss measured at this distance was as high as 70 percent. In addition, we demonstrate that burstiness in frame loss is something that should be anticipated in both mobile and stationary scenarios. Due to the fact that each frame in the basic configuration is only transmitted once, losing a frame will result in the loss of data. We design a novel coding scheme for data recovery in LoRaWAN called DaRe in order to cut down on the amount of data that is lost. This scheme extends frames with redundant information that is calculated based on the data from the frames that came before it. DaRe combines techniques from convolutional codes and fountain codes. We develop an implementation for DaRe and show that it is possible to recover 99% of the data with a code rate of 1/2 for up to 40% of the original frame rate being lost. In comparison to repetition coding, DaRe is capable of recovering 21% more data and can reduce the amount of energy required for transmission of 10 byte data units by up to 42%. In addition to this, DaRe offers improved resistance to sudden frame loss. Both developers of LoRaWAN applications and operators of LoRaWAN networks can benefit from the findings of this study. The results of the characterisation can be used by network operators to identify potential vulnerabilities in the network, and an instrument is provided to application developers to help them prevent the potential loss of data.