Enhancing security of data transmission in LoRaWAN

Abstract

Long-Range (LoRa) is a commonly used physical layer in the Internet of Things (IoT), to transmit data with low power and wide area coverage. LoRa wide area network (LoRaWAN), is a commonly applied network protocol based on LoRa modulation, consists of end-devices, gateway, network server, and application server. However, some security flaws are possible to occur in LoRaWAN during the data transmission. For instance, the packets that transmit between gateway and network server, are possible to be tampered and sniffed. Moreover, network servers are vulnerable to malicious gateway attacks. Recently, existing literature is using blockchain technology to address those security issues, but it may not be suitable for LoRaWAN with a critical latency. In the project, directed acyclic graph (DAG) peer network is proposed to apply in LoRaWAN which connects all gateways and the network server. It can ensure both uplink and downlink packets are transmitted without tampering. In addition, the project implements cryptography and digital signature to prevent packet sniffing and malicious gateway attacks. To evaluate the performance of LoRaWAN with DAG network, a customized testbench is established to perform analysis. The result showed that packet tampering, sniffing and malicious gateway attacks can be prevented with lower latency.