Message Specification and Encoding – A Never Ending Story

Some 30 years ago I started studying digital communication like IEEE 802.4 (Token Passing) and IEC 802.3 (Ethernet) when I worked for Siemens. The “war” between the Token Passing supporters and the Ethernet friends caused a lot of struggle and generated many solutions to find a common way for information exchange in automation systems. One of the approaches was to make Ethernet deterministic. I invented a simple algorithm that used the three states of the transmission line: no-carrier, carrier and collision. The solution allowed a semi-deterministic behavior of Ethernet … the solution was patented in 1985.

Click HERE for the European Patent EP 0110015.

Unfortunately Siemens did not implement the patent – the direction was towards Token Passing

The international project MAP (Manufacturing Application Protocols) preferred Token Passing … then fieldbusses arrived … MAP opened the way for Ethernet – too late (or too early).

Ethernet “came” back some 10 years later, when it was obvious that traditional fieldbusses were quite limited in many aspects. Today Ethernet is the preferred solution in automation systems (industrial and power domain). Switched Ethernet has ended – to some extent – the wars on Data Link Layer solutions.

The communication wars for crucial automation domains are still ongoing! There are two issues discussed again and again: the approach of services and message encoding. What services are required (Get, Set, Event reporting, Event logging, Control, …) and what is the best message encoding method (abstract/concrete, XML schema, ASN.1, XML encoding or binary encoding like ASN.1 BER)?

The questions that are most crucial: What should be carried in a message? How many implicit or explicit content is needed to be carried in each message? How often has a message be sent/repeated? … and many other questions have a huge impact on the SYSTEM behavior. The focus should be on the SYSTEM and not on the encoding or message schema.

The most efficient encoding is to send an empty message! Really!

Click HERE for the story about a very efficient message.

It is quite interesting that old standards like Ethernet, TCP/IP and MMS have survived – even they are not the most efficient ways to communicate!

Why are they so successful – even in the electric power automation world, which is one of the most conservative markets? These standards are open and accepted globally!

I hope that the developer and users of automation systems will focus on the APPLICATION and SYSTEM ASPECTS – and not on communication layers 1 to 7!! The systems and applications based on open international standards will help to keep the power – more efficiently – flowing. IEC 61850 is one of the most crucial open standard.

The fastest way to get your information flowing with IEC 61850 information models, services and messages is: Develop your application using a simple IEC 61850 API. I have trained many experts in how to use IEC 61850 and a simple API to solve their APPLICATION needs!

Click HERE to evaluate such a simple API.

The power industry is short of experts – hope the remaining resources will focus on the applications and not on questions like: How can we optimize the message encoding, how can we save bits on the wire, … ?

Focus on the Applications!