Ultrasonic Communications

[sticks]

Hi all, I am just getting into this wonderful world, and I have kind of ambitious project. I'd like to set me up something that will do two-way communication via ultrasonic stuff. Not a lot of bandwidth is necessary, say two-digit numbers and whatever other overhead is necessary for checksums and the like. I wanna do this with ultrasonic in hopes that I can transfer this to an underwater application.<p>I'm having trouble getting started. I've seen all sorts of ultrasonic projects using PICs (my choice in MP) but they are pretty much all range finding type things. Can anyone tell me if it'll be possible to take this equipment/knowledge and turn it into what I want? Anyone know of any projects/hardware/software that might help out?<p>Thanks for any insight, help or pointers.

[chessman]

Well, this is by far a much more ambitious project than you might think. It can be done, but I think it's going to be hard.<p>First of all, it would be easy to create an ultra-sonic one-way communication system. You speak of 2-digit numbers, so we're talking about 7 bits of transmission, allowing you to transmit up to 128 as one byte. To make it more useful, you might as well plan on making it an 8-bit system, giving up to decimal 256 as one byte.<p>If it were to be a one-way system, you could implement a simple protocol of one start bit (low) followed by a delay, follow by the msb of the data, followed by an equal delay, etc. At the end of the transmission word could be a stop bit (low) to signify end-of-transmission.<p>Unfortunately, that only works for a one-way system. For a two-way, as you suggest, there are a few more problems.
The main problem with two-way is the basis of unltrasonic communication itself. This being the fact that a receiving ultra-sonic transducer responds to ANY ultra-sonic wavelength it encounters. That includes the serial data that EITHER side of the two-way system could be transmitting.<p>So basically, either PIC could not be transmitting and receiving at the same time, and there must be a short delay between the transmit and receive modes of operation. In essence, fairly precise delay routines need to be implemented, although that task shouldn't be too hard to accomplish.<p> Drive circuitry for the actual transducers wouldn't be that difficult, although Gordan McComb tends to say that for a short period, a transmitting transducer could draw up to 6A of current at 5V, so drivers would have to handle an 8A peak to be safe.<p> I do believe you could use one of the cool Morse Code Keyers and Morse Code Readers. These are found on varying places of interest on the internet, and I can dig for the links if you wish. The output device of the circuit could just be connected to a transducer drive circuit. Text is displayed via LCD screen, and text is entered via AT computer keyboard. The basic principle could be adapted to other data aquisition systems.<p>Feel free to email me if you wish,
~Kyle

[sticks]

Thanks for the replies.<p>What I am trying to do at first is "simply" to type in a two digit number on one end of the link and for that number to show up on the other end. Both ends can transmit and both can receive. <p>Ideally, there'd be the ability for there to be several of these links in the same area and not interfere with each other. <p>Since there'd be the chance of losss and confusion with tx/rx, I figure that in addition to whatever data I sent, I'd also have to do some sort of error checking/filtering. (Hence the checksums)<p>I chose ultrasonic, because from what I can tell, thats the best medium for wireless underwater communication and I want this to eventually work underwater. I figure (possibly incorrectly) that with all the ultrasonic range-finding projects out there that it'd be a lot easier to prototype the thing and get it working in air first.<p>As for communication methods, I've found a couple of references to FSK and even found some PIC code for using an RF tx/rx system, but I don't know enough to know whether or not this would be a good way to communicate or not. Maybe and RS-232 type thing? or FSK or BASK or....?<p>Anyway, I know that this is a big project, and I appreciate any and all comments.<p>-nathan

[Dimbulb]

you could start with a beacon such as:
http://www.ussc.com/~turner/am_tx1.gif<p>waterproof and testing is a challenge