Tell me more ×
Electrical Engineering Stack Exchange is a question and answer site for electronics and electrical engineering professionals, students, and enthusiasts. It's 100% free, no registration required.

Hypothetically, lets say you had a radio plugged into your computers microphone port. You record a certain specific signal in audio format to an .wav file. This signal is meant to be a small digital communication between small devices. They are using standard protocols, so no proprietary crap to worry about.

Does anything exist to be able to analyze this signal in this format?

I have been trying to manually analyze the captured signal just by looking at the waveform and comparing it to other waveforms I have seen on the internet, this is not working. I have looked everywhere for an automatic way of doing this, but all solutions seem to be hardware.

So I guess what I am looking for is a program that will take an audio file, analyze the waveform and spit out the original binary data.

share|improve this question
I have edited your question to what it seems you were asking, I was about to migrate it for looking for a program to analyze the audio and then realized I think your confusion comes from what a radio receives. Please let me know if this is what you were trying to ask.(If anyone wonders, before a migration you try to remove comments, i did so and then realized it may not be a good migration candidate as I saw a different meaning from the poster) – Kortuk Jul 23 '11 at 22:14
2  
@Kortuk: That totally changes the meaning of the question. I think he has some device that transmits data over RF, and wants to know whether he can intercept the transmission with a "radio" (whatever he thinks that means), record the result to a computer, then analyze the audio to try to interpret the data. At least that's what I tried to answer. Your interpretation is very different. – Olin Lathrop Jul 23 '11 at 22:45
1  
@Kortuk I have to agree with Olin, this is totally different from the original question. – Kellenjb Jul 24 '11 at 0:41
well, something ought to change regarding 'OCR' in the question title. either way, it certainly isn't about optical character recognition. – JustJeff Jul 24 '11 at 0:55
@JustJeff I wasn't sure of a great title for it, I just knew the question wasn't about being able to play music so I just rolled it back. I have taken my best guess for a title, feel free to edit if you have something better. – Kellenjb Jul 24 '11 at 1:37
show 5 more comments

2 Answers

You have to find the protocol being used somehow. Preferably you find a document that tells you outright. You might be able to reverse engineer it, but less likely for someone that has to ask about basic decoding here. You say the protocol is standard, so you obviously know something about it. If it's truly standard, you should be able to look up the standard.

Additionally for your scheme to work, the radio you use has to demodulate the RF signal according to this standard, and the resulting baseband signal has to be in the audio range, perhaps less depending on the radio. For example, if the radio is doing AM demodulation then this isn't going to work if the device is using frequency shift keying. Also, if the device is transmitting data at 30Kbits/s it also isn't going to work if it's a normal audio radio.

Added:

I just noticed you plan on plugging the radio output into the microphone input of the computer. That will most likely overdrive the microphone input amp. You should plug the radio into the "line in" jack instead of the microphone in jack. Most sound cards (or motherboards) these days have both choices.

share|improve this answer

Pretty much any modulation scheme can be decoded in software (though not necessarily in real time) if you can record the signal in samples capturing sufficient information.

Recording the audio output of a receiver's (AM or FM) demodulator is not necessarily going to be sufficient, unless it's a modulation mode such as AFSK designed to be used with voice radios.

Much more common is to have a receiver which outputs narrowband signals in IQ format, by mixing the final IF with two phases of a local oscillator to produce an audio-range intermediate frequency, often centered at 0 Hz. This is not a demodulated signal, but a form of digital IF which is suitable for digital demodulation. By using two phases, you can tell the positive and negative frequency components apart.

Stereo sound cards happen to provide you with two input channels, so people have tried to use them for IQ sampling of narrowband signals, however the input circuits of most sound cards aren't good all the way down to zero frequency, so it's sometimes preferable to use a non-zero center frequency and mix down to 0 in the digital domain. At higher bandwidths the same techniques apply, only higher performance A/Ds are used, typically feeding into ASICs, FPGAs, or DSPs.

There have been a lot of amateur radio, hobbyist, and professional receivers built using audio-frequency IQ sampling over the past decade - some web searching will turn up many projects including demodulation software, much of it open source. You will also find that some of the receiver front ends are pretty simple - often just an oscillator, a digital bus switch as a mixer (!) and some op amps.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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