Our objective was to research and design a long-range mesh network that would allow for broad coverage in large areas where other networks fail due to environment.
![](https://sites.psu.edu/lfshowcasesp20/files/2020/11/JD_gy_2c_RGB_h-002-1.png)
Sponsored By: John Deere
Team Members
Sean Sacchetti | Heather Moss | Kenneth Pan | Songmeng Wang | Noah Krieger | | | | | | |
Project Poster
Click on any image to enlarge.
Project Summary
Overview
John Deere offers smart equipment with data being sent to and received from vehicles. In some situations, there are environmental challenges that make traditional networks unreliable. Deere needs a network solution that can connect vehicles and supervisors, regardless of the environment. This solution would have to work in a variety of environments, have low latency, and have a high degree of reliability through redundancy built into the system.
Objectives
Our objective was to investigate if a mesh network is a more reliable solution by implementing one to transmit a message between equipment and supervisors where no other network access is available. The software goal of this project was to implement a “heartbeat” function into the network to allow the nodes to communicate their status to each other to assure that the network elements remain online. If one of the nodes loses power or connection to the network, we ensured that the mesh is self-healing, meaning that other nodes will automatically route the correct information through other nodes.
Approach
– Purchased BLE and Zigbee boards in order to analyse and compare each to find the best boards.
– Chose Zigbee because it was the most easily compatible with a mesh network due to a simplified software implementation, was a low-energy hardware, and it offered an increased range.
– These boards will act as nodes that use antennas that act as our core mesh network, these antennas allow the boards to communicate with each other wirelessly.
– Configured each board with XCTU software to run the mesh network with the same settings.
– Implemented a MicroPython program that will allow the boards to transmit data and confirm that all network elements remain online.
– Using the datasheets we compared board and antenna ranges while still having low power, we wanted low power due to keeping these boards powered all the time.
– Verified the operation of the mesh network with three nodes.
– Tested our code on the three boards in an indoor setting. We developed a test plan with intentions to test our mesh network in a variety of environments including open areas with trees, hills, and etc.
Outcomes
– Developed a test plan in order to verify that our prototype was working correctly. The sponsor will execute the test plan we developed to see if a mesh network is truly the more reliable way to go.
– Completed research on tradeoffs when choosing which boards to use.
– Created application code to run on the mesh network.
-Completed a configuration guide for our sponsors to use when recreating this project.