Thinking about ECAN (Extra Cheap And Nasty) MODEM ideas.

I've got lots of ideas myself but someone may have practical 
experience with such or some other OTW ideas.

I want to trial the use of cheap PRS (UHF CB) transceivers as data 
transceivers. This is legal in my administration but may not be in 
some others. I have a pair of transceivers which cost $NZ78 (about 
$US50) for the pair including 4 rechargeable batteries per unit, a 
dual bay mains powered charger and a "hands free" ear piece/mike combo 
per transceiver. Remarkable value even if the quality is not top 
shelf. The key enabling feature is that these units have VOX (voice 
operated transmit) and a 2.5mm microphone/earphone jack. This means 
that audio encoded data can be fed into the microphone circuit and 
recovered from the earphone circuit thereby allowing 3 wire plug in 
data connection with no alteration to the transceiver.

Part of the reason for doing it this way is to see if doing it this 
way proves as advantageous as it seems it might.
While the "sensible" / professional way to do short range data 
transceiving is to use any of a large number of available COTS 
modules, the PRS transceiver approach has some attractions. For $US25 
per end you get.

- 500 mW transmitter
- Aerial
- Complete transceiver with battery box
- Easily selectable over 40 channels (although data is usually used on 
a limited number of channels).
- Selective tone calling if desired.
- Easy access for the uninitiated.
- Licence free operation
- Ranges of 10's of metres almost anywhere and several km line of 
sight may be expected.

Envisaged uses are various ranging from car alarms (and can send data 
and audio if desired from car), low data rate telemetry, robot 
control, ...

To interface with typical microprocessor systems a "modem"is required 
to generate tones for the mic channel and decode received tones. 
Existing 1200 baud and possibly 2400 baud modem ICs would work fine. 
These add hardware cost and circuitry to what otherwise might be a 
trivially simple design. I'm aware of the advantages of "real" modem 
ICs. I'm interested in the possibility of cheaper alternatives. These 
may be false economy depending on application etc. A few ideas for 
ECAN modems are

1.    On/Off tone bursts. Processor generates tones or not. Could be 1 
pin square wave or PWM shaped wave or ... . Receiver uses diode into 
resistor load with a parallel cap. Simpler and cheaper you almost 
cannot get. Nastier would also be hard. System AGC rate must be 
substantially slower than bit rate. Channel noise will create mayhem.

2.    TX generates dual tone FM or mark space modulated frequency or 
some other tone generation system. receiver processor detects this 
directly. Hard on receiving processor. Noise immunity poor. Possibly 
cheaper than 1.

3.    Like 2. but receiver uses a tone detector IC or a phase lock 
loop or some other  tone detection system. superior to 2. More 
complex. Parts count for non-dedicated receiver IC may grow 
alarmingly. Various XR tone detection ICs and modem ICs of yore come 
to hand.

4.    TX/RX gets real in processor and does "proper" dual tone 
detection etc. AFAIR Scenix have / had a reference design (with NDA) 
that would give the general requirements.And this could easily enough 
be derived from scratch. More computationally intensive.

5.    ???

Method 1. or 2. may well work well enough to be used for non critical 
applications over moderate ranges. I can see this being attractive for 
hobby robot type use.

Thoughts?



        RM






-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist