EADIN_Lite Network Protocol
Authored by Eliot Aretskin-Hariton: eliot.d.aretskin-hariton@nasa.gov, Intelligent Control and Autonomy Branch, NASA Glenn Research Center,
Network Protocol Summary Stats:
- Half-Duplex & Hard Real-Time, works with Wired or Wireless networks
- Command / Response protocol using 1 Master / Multiple Slave Architecture
- 8 Byte payload
- RTT Performance (See Table and Note below)
Speed | TYPICAL | WORSE CASE |
---|---|---|
4000000 baud | 943 +/- 13 | 981 +/- 13 |
921600 baud | 1,197 +/- 15 | 1,280 +/- 7 |
115200 baud | 4,467 +/- 12 | 4,907 +/- 7 |
9600 baud | 45,798 +/- 12 | 50,750 +/- 20 |
units | (micros 1-sigma) | (micros 1-sigma) |
Note: Performance based on message Round Trip Time (RTT), which includes formulation of the message by the master, receipt of message by slave and recept of respons from slave by master. master -> slave -> master. Time is expressed in microseconds.
- Memory Requirements
Memeory Req. | Program Storage (Bytes) | Dynamic Memory (Bytes) |
---|---|---|
Default | 9,496 | 900 |
Minimum | 8,604 | 388 |
- Successfull Deployment
- Arduino Yun
- Arduino Mega 2560 (the $10 version)
- Expected Compatibility
- Arduino Pro Micro
- Any Arduino with Hardware Serial ports
Overview: This code was created to support the Distributed Engine Control task as part of the Fixed Wing Aeronautics project. The purpose of this research was to enable multiple microcontrollers to speak with eacho ther per the protocol specified in the preliminary release of EADIN BUS. EADIN BUS is a candidate for distributed control systems on aircraft engines and is being worked on by the Distributed Engine Control Working Group (DECWG) http://www.decwg.org/. The primary use of this code was to assist in the modelling of small local networks which contain 16 or fewer nodes. Ultimately, we expect this network to be implemented in an FPGA or ASIC as opposed to it's current implementation on a microcontroller.
This communication protocol uses a master node which distributes information between nodes through a call and response system. The RS-485 network is simplex and thus does not allow multiple nodes to talk at the same time. No time synchronization between nodes is required for this network. These factors enable the master to request information from sensors and command actuators, one at a time. In the current implementation, no information is passed from individual nodes without first going through the master node.
While other communication protocols do exist like ModbusMaster and simple-modbus, the speed of these communication protocols on the RS-485 network was not sufficient for our needs which required message send to reply receipt times of 1 millisecond. Additionally, the other protocols did not implement the same message system as specified by the preliminary documents regarding the EADIN protocol.
Details: The EADIN protocal as implemented by this code has the following structure: Total Size: 18 bytes
- Preamble: 3 bytes
- 2 bytes start of message (0x00, 0x01)
- 1 byte a sync byte (0x55)
- Headder: 5 bytes
- 1 byte request type (ENQ/ACK/NAK = 0x05/0x06/0x15)
- 1 byte node_ID of desination
- 1 byte node_ID of sender
- 1 byte unique message number
- 1 byte extra space to be used in future development
- Data: X bytes (8 bytes Default)
- 8 bytes DATA_L (can be modified)
- Footer: 2 bytes
- 2 bytes CRCFast (a 16 bit CRC, Default)
Updates: Version 2 is incompatible with previous versions as it is constructed with different function calls using an object oriented programming approach for easier use. The code now contains built in timing functions which should enable the user to simply call OBJ.read() OBJ.write() functions without worrying about inserting delays between the write and read operations. These delays should scale with network speed selected from 9600 - 4000000 baud.
References: EADIN Lite Communication Network
Eliot Aretskin-Hariton, Benchmarking Variants of a Hardware-in-the-Loop Simulation System
Eliot Aretskin-Hariton, A Modular Framework for Modeling Hardware Elements in Distributed Engine Control Systems
Dennis Culley, Developing an Integration Infrastructure for Distributed Engine Control Technologies
Ross N. Williams, A painless guide to CRC Error Detection Algorithms