Long range radio communications

G

gbbickerton

Member
Manchester encoding

Quite some time ago I posted a simple manchester encoding and decoding program and Hippy kindly showed how it could be done more easily. However I do not think that most people realise that the manchester encoding is needed for the receiver modules to work correctly they use what they call bit slicing, this uses the average value of the receiver output as the switching level for a comparator to determine what is a 1 and what is a 0 on one of my receivers this level has a such a short time period that it would not be able to receive 8 0 bits or 8 1 bits at 2400 baud without error. This means that you must send a peamble of equal 1's and 0's to get the comparator level in the receiver set up before any data and then use manchester encoding to keep the receiver operating at the correct level for the data I have found that I can double the reliable range of my system using this method. This also means that slow serial links would not work either.Has any one looked at AFSK as used by radio teletype the chips may not be available but the tones should be possible in software although decoding them will probably need hardware.
 
M

manuka

Senior Member
Basic 2 tone AFSK (Audio Freq. Shift Keying),with mark and space tones 2125 Hz & 2295 Hz, dates from the late 1950s & is small beer these days due to more effective "fuzzy" & digital modes. Radio hams have been at the forefront of this, dreaming up bullet proof creations such as such as Throb, Clover & MSK. Suggest you check => www.rac.ca/opsinfo/infodig.htm etc!

This is not to totally dismiss FSK/AFSK now of course. I'd quite a PICAXE slanted AFSK workout some years ago, using soundcard decoding, & concluded RTTY (Radio Teletype) better suited quick messaging (where errors may be tolerated) rather than PICAXE telemetry/telecontrol.
 
hippy

hippy

Technical Support
Staff member
There's no real way round sub bit-time problems at slow baud rates, but it does seem that straight ASCII after a suitable length of "U" preambles works quite well, but maybe not for long range usage. I suppose having a start bit and a stop bit helps as that means 20% of the data stream is already balanced and there's a good bet that most characters even out overall; numbers are of the form %0011xxxx so around 50% of them are balanced to a fair degree. I've never really thought about why it worked, but was surprised it works as well as it does.

Again, if bit-time isn't a problem, poor man's Manchester Encoding can be achieved over two bytes by sending the inverse of the first,a ndinserting the occassional "U". A big key to keeping the receiver tracking the data is not to have too many 1's or 0's and especially miimal inter-bytes gaps; build the data up in memor (SFR) then spit it out as fast as possible.
 
M

manuka

Senior Member
Hippy- your " Poor man's Manchester encoding" makes a very nice phrase! Most exotic wireless data techniques use PC sound card decoding,which naturally means more horsepower that straight PICAXE-PICAXE. Duty calls,so more later...
 
You must log in or register to reply here.
Top