1
\$\begingroup\$

This will be my first try at USB intefacing !

May I know how to transfer the data from pins : 11-15 of KT3170 IC to USB ? Data sheet : http://www.datasheetcatalog.org/datasheet/SamsungElectronic/mXuusvq.pdf I guess I will need to configure a Microprocessor for making the OS uderstand what profile of device is this , In that Case what will be my device profile ? Mine is an RF remote control receiver using DTMF which I want to connect to the PC.

I also have a doubt on fast can The KT3170 can interpret the DTMF tone. That is if I press a button (say 1) and immediately within the same second if I press key 2 Will it be able to recieve and decode the DTMF correctly with proper time synchronization ? What will be the latency if I transmit this signal through Frequency Modulation (FM) ?

\$\endgroup\$
4
\$\begingroup\$

I'd suggest a microprocessor with built-in USB, such as the PIC 18F2455 (or one of its variants). The USB interface to your PC is very simple and ends up looking like an RS-232 serial port to the PC. The driver is probably already in your OS if you're using recent Windows.

The way I interpret the data sheet, I would connect Q1 - Q4 to digital inputs on the microcontroller. Then connect the DSO (pin 15) to a digital input with an interrupt-on-change and process when it goes high. When you get the interrupt, read the Q1 - Q4 data, do whatever pre-processing you may want and send it out the USB port.

I can't definitively tell by the data sheet if the KT3170 output would change if you pressed a second key while the first one is still pressed, but I'd guess that the fist output would stay latched unless you fully released the first key prior to the second press. You might have to just try and see.

If you press one key and then quickly press the next, you should be okay and get the second keypress as the maximum interdigit pause time is 40 ms. As long as you aren't faster than 40 ms between presses, you'll get the second press.

I don't know what your transmit latency will be. You'll have to look for that in the transmitter/receiver pair documentation.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ The USB Bit Whacker - 18F2553 Development Board might be a good starting point, if your goal is to learn how to communicate using USB and 8 bit Microchip PIC microcontrollers. sparkfun.com/products/762 \$\endgroup\$ – davidcary Nov 20 '10 at 4:31
1
\$\begingroup\$

I'd have a look at the AVR USB chips and the Teensy dev boards.

How to build a USB controller having knobs, sliders, and switches

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Thanks a lot for suggesting the Teensy board, Joby :) I would have missed it otherwise :) \$\endgroup\$ – NS Gopikrishnan Oct 6 '10 at 5:45
1
\$\begingroup\$

If reading the decoded data from that DTMF signal is all you want to do, programming a microcontroller seems unnecessarily complicated.

Instead, I would use one of the USB-to-parallel interface devices available from many companies, such as the Breakout Board for FT245RL USB to FIFO.

With the (relatively cheap) DTMF transmitters I have, if you hold down one button for 0.1 seconds, then continue holding it down while pushing a second button, it transmits an (impossible to decode) mishmash of more than 2 frequencies.

If I'm reading your data sheet right, it ignores such impossible-to-decode signals, and during such times acts the same as when it hears nothing but silence.

If I'm reading your data sheet right, the latency from pushing a button on a hard-wired DTMF encoder to the rising edge on DSO is less than 0.1 second. I don't know anything about the delay your FM modulator and demodulator will add to the system, but I suspect the total latency will still be well under 0.2 second.

| improve this answer | |
\$\endgroup\$
0
\$\begingroup\$

Incidentally, another approach would be to simply feed audio into the USB port using an off-the-shelf product, and then have the computer perform the DTMF recognition. Depending upon the exact application, a computer may be able to do a better job of DTMF recognition than a simple IC could manage. For example, a computer could take advantage of the fact that DTMF is generally gated on and off relatively cleanly. If the computer sees something that starts out looking like a DTMF tone but the frequencies change after 100ms, it might be able to (depending upon the application) retroactively determine that the tones weren't really DTMF after all. There's no way a DTMF chip could do that.

| improve this answer | |
\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.