Adaptive Parent Change and Cell Usage Aware Objective Function for RPL in 6TiSCH Network
Dzaky Zakiyal Fawwaz, Sang–Hwa Chung · 2024
The 6TiSCH wireless sensor network architecture combines the time-slotted channel hopping (TSCH) medium access layer (MAC) with the routing protocols for low-power and lossy networks (RPL). However, most RPL studies overlook the integration with TSCH MAC or focus solely on carrier-sense multiple access (CSMA) MAC. Existing RPL methods use an objective function (OF) based on an expected transmission count (ETX) link metric, yet this fails to fully capture network conditions and traffic load representation. Moreover, the RPL hysteresis method uses a static threshold without considering dynamic network conditions. Thus, it may not hold excessive parent changes and prevent better parent selection. To address these issues, we propose an adaptive parent change and cell usage aware objective function (AC) RPL method, which utilizes Q-learning to decide the optimal policy to change parent and an RPL OF based on traffic and cell usage. Experiments were conducted over simulation using the 6TiSCH simulator and real hardware tests using a testbed prepared by FIT IoT-Lab with OpenWSN firmware. The result demonstrates that AC-RPL outperforms the benchmarks. Compared to the standard RPL, AC-RPL can increase the packet delivery ratio and received packets by 9% and 13%. Also, it reduces energy consumption and latency by -21% and -8%.