I've done a lot of packet stuff using RF over IR and also half duplex uart
over a single wire or rs485. The packets generally take this form

0xaa
0x55 - two byte start of packet identifier
0xnn - one byte packet type
0xnn - one data length
xxxx - variable length data
Checksum or CRC

Sometimes I put a source and destination address after the 0xaa 0x55.

On the IR system, I'm sending closed captions in movie theaters. On the
transmit and receive side, I use a union of structures for different
packet types. The received data is put in the buffer on a byte by byte
basis. If the CRC is good, the data packet type determines which structure
to use in sorting out the data. The reverse process is used in forming the
packets. This is a one way system.

On two way systems, I use a master polling slave units. The slave always
responds to the master and never transmits otherwise. So, there's no
contention.

In each case, the receiver uses a state machine to keep track of where it
is. It waits for the first 0xaa, then waits for the 0x55, then gets the
packet type, the length, then knows how many more bytes to get until the
checksum or CRC is coming in.

Good luck!

Harold




--=20
FCC Rules Updated Daily at http://www.hallikainen.com - Advertising
opportunities available!
Not sent from an iPhone.
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.