EKF-RPL: Enhanced Mobility Support for IoT

This project implements an Extended Kalman Filter (EKF) extension for the RPL (IPv6 Routing Protocol for Low-Power and Lossy Networks) protocol within the OMNeT++ simulation environment.

The goal is to improve packet delivery reliability in high-mobility Wireless Sensor Networks (WSN), such as drone swarms or autonomous robotic shuttles, by proactively predicting node movement and pruning unstable links before they break.

📂 Repository Structure

This workspace contains three main modules:

• inet/: The standard INET Framework (v4.2.8).
• omnetpp-rpl/: The core implementation module.
  • Modified: Contains the custom EkfRpl module, EKF.cc (Kalman logic), and the modified EkfRplRouter.ned.
• wsn_baseline/: The simulation project.
  • Contains omnetpp.ini configurations, WsnNetwork.ned, and analysis files (.anf) for comparing Baseline RPL vs. EKF-RPL.

🚀 Key Features

1. Decoupled 1D Kalman Filter

Unlike traditional implementations that use complex 2D matrix libraries, this project uses two lightweight, independent 1D Linear Kalman Filters (one for X-axis, one for Y-axis) to track node trajectory. This reduces computational overhead suitable for constrained IoT devices.

2. Proactive Drop Mechanism

The standard RPL protocol is reactive (waiting for a link to fail). This project makes it proactive:

• Intercepts DIO control packets.
• Predicts the sender's position 3 seconds into the future.
• Drops the packet if the node is predicted to move out of the 200m radio range (Safety threshold: 150m).
• Prevents the routing table from selecting a parent that is about to disconnect.

3. High-Mobility Tuning

Includes tuned Trickle Timer parameters in omnetpp.ini to handle speeds of 5–15 m/s:

• dioIntervalMin reduced to 0.1.
• dioRedundancyConstant increased to 10.

🛠️ Installation & Build

1. Prerequisites:

   • OMNeT++ 5.7
   • INET Framework 4.2.8

2. Clone the Repo:

   git clone [https://github.com/Alpsource/ekf-rpl-omnetpp.git](https://github.com/Alpsource/ekf-rpl-omnetpp.git)
   cd ekf-rpl-project

3. Install INET Framework:

   • Download INET 4.2.8 (ensure it is compatible with your OMNeT++ version).
   • Extract or clone it into the root of this workspace so the folder is named inet.
   • Your structure should look like this:

     ekf-rpl-project/
     ├── inet/           <-- You place this here
     ├── omnetpp-rpl/    <-- Included in repo
     └── wsn_baseline/   <-- Included in repo
     

   • Build INET:

     cd inet
     make makefiles
     make
     cd ..

4. Build RPL Module: Navigate to omnetpp-rpl/ and run:

   make makefiles
   make

5. Run Simulation: Navigate to wsn_baseline/ and execute the simulation:

   ./wsn_baseline -m -u Cmdenv -c EKF_RPL

📊 Performance

Simulation results (50 nodes, 500x500m area, Gauss-Markov mobility 5-15 m/s):

Metric	Baseline RPL	EKF-RPL (Proposed)
PDR	33.96%	36.32%
Packets Received	849	908
Avg Delay	554 µs	576 µs

While the PDR increase is modest (~2.4%), it proves that predictive logic effectively filters unstable parents in highly dynamic environments.

📜 Credits

• Original RPL Implementation: ComNetsHH/omnetpp-rpl
• INET Framework: inet-framework/inet

📄 License

This project follows the licensing of the underlying INET and OMNeT++ frameworks (LGPL/GPL).