Simultaneous coding: Are there any options between simple bundle control and hemming code?
I need to encrypt the package (8 bat) in such a way as to detect, if any, errors of transmission at the receiving end.
The task of correcting errors is not worthwhile, but it would be great, of course.
I see two options now:
Simple and least sustainable (single error)♪ It's honor control.
It's rather complicated, but very steady.♪ The Hemming Code 8 is coded in a 12-bit package (4 excess bat). It's actually a classic, like an 11-bit delivery that makes seven data bats, but it's kind of like a coded 8th battle.
I've overwhelmed a lot of Hemming coding decisions, but I don't understand the very principle, whether it's a really complicated task. I don't have much time to clean up, the resources of the processor (ATMega48PA @ 1MHz) don't matter, even though in my case there's no clear limit.
It's purely intuitive to me that the Hemming code for me is out of proportion (and the 33% reduction in the speed of transmission is also tightening) and the integrity control is inadequate.
Are there any decisions in the middle?
morde last edited by
First, you need to figure out what real mistakes you're making and how often. After that, how difficult it is to re-examine and pass the damaged data.
For example, if you have one mistake for a million bat, a mistake is one damaged bat, and the data can be requisitioned, the honor control is over your eyes. If the requisition is very expensive and the processing is as fast as possible, and it does not tolerate delays, then we should look towards the Hemming code.
Record control can be extended by any number of bats, if you have a real need, for example, you can take the balance of separation by 2 and the balance of separation by 4. Then the algorithm will find two rather than one battle on the White. It is also possible to act in the opposite direction, for example, to consider the nuance of not one bayth, but of blocks greater size: if errors occur very rarely and speed is very critical, it may be beneficial.