Reference Paper:  A reinforcement learning-based link quality estimation strategy for RPL and its impact on topology management-[2017]

Research Area:

IOT

Research Topics:

Contiki Cooja Simulator Projects in RPL Routing Protocol

Tools Languages:  Contiki-Cooja / Contiki NG simulator, Front End: Java, Back End: C

Software Requirement:  Vmware workstation player, Instant Contiki-3.0

Aim and Objectives:  
This project aims to develop a reinforcement learning-based link quality estimation strategy for RPL and its impact on topology management using the contiki-cooja simulator.

Contribution:  
1. This project presents Reinforcement Learning - Probe(RL - Probe), a novel strategy for link quality monitoring in RPL.
2. To achieve this project, RL-Probe leverages synchronous and asynchronous monitoring schemes to maintain up-to-date information on link quality and promptly react to sudden topology changes.

Performance Evalution:  
A simulation is needed to investigate the performance in controllable and easily reproducible network conditions. It uses Cooja to simulate Tmote Sky nodes with different network sizes (30, 60) nodes.
Performance Metrics:
 •  PDR
 •  Energy Consumption
 •  Throughput
 •  Delay
 •  Control Packet Overhead
 •  Execution Time
 •  CPU Energy Consumption