Need advice on forward error correction..

Status
Not open for further replies.

Buriedcode

Full Member level 6
Joined
May 6, 2004
Messages
357
Helped
43
Reputation
86
Reaction score
7
Trophy points
1,298
Location
London
Activity points
8,887
Hi,
I've been reading up on error correction for a wireless design of mine and I'm a bit stuck at the moment. The design already uses 'hamming codes' implemented in hardware (schematic entry) using the 12,8 code, modified to a 13,8 code for double error detection. Seeing as I'm sending a 16-bit word, I simply split it up into two bytes, so each byte has its own error correcting capability (total of 2 errors, 1 in each byte corrected, 2 in each detected). So I'm sending 26 bits in total, but I have room for 32.

Now, this is quite simple to do, and works beautifully with the MACH4's I've got. I was originally going to use convolutional coding, but decoding requires extensive hardware/software, even when using the simplest algorithms. Also latency is a problem because I'm after <2ms from transmit to recieve.

I'm really just asking if anyone has heard of any less-known codes, block, systematic, cyclic, that can correct multiple errors in a 16-bit word with a maximum overhead of 16 bits. The encoding/decoding has to be simple (using 64 macrocell CPLD's), preferably 'on the fly'. A special 'reed-solomon' code looks OK (32,16,5) but the decoding is a bit of a nightmare.

Any links, advice, code, schematics or ideas are welcome. Its just a hobby of mine that I've become a bit obsessed with :?

Ps. I have also implented a (8,4) hamming code 4 times for the 16 bits, giving a pseudo (32,16) hamming code, but its messy.

Thankyou.

BuriedCode.
 

Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…