Buriedcode
Full Member level 6
Hi, I wasn't sure which forum to post this question in, I don't want to waste peoples time. But here it is.
I've got 2 AV 2.4Ghz modules (Tx/Rx). The system can send a video signal (in this case, its designed for PAL-I) and a stereo audio signal. I was thinking of using the system for sending data, namely digital audio. The two audio channels can be used for control signals, since they are low bandwidth of approx. 20Khz each. Really my question relates to the bandwidth of the video channel. I know that PAL needs about 5-6Mhz for a fairly decent signal, also, there are some components in the PAL signal which are above this, but these are filtered out inside a TV/Video player. So, if I was to send digital data, it needs to be bandwidth-limited to 5Mhz. For a digital signal (square wave) a low-pass-filter is needed to reduce the harmonics of the signal. I was initially going to use an analogue (2-4 pole) filter but it seems that it would not provide enough attenuation. So I was thinking of making a sort of 'cheap' FIR filter, using a shift register, some logic, and a resistor ladder. I've built a test circuit and I can get a pretty smooth sine wave (after filtering with a 2-pole analogue LPF) when I put 1.,0,1,0,1 etc.. into the system. I was hoping to see the spectrum of this output but I don't have access to a spectrum analyzer, and the prospect of doing all the fourier transform by hand doesn't fill me with confidence, although it can be done. So, if you want me to get to the point:
Do these wireless AV modules have a built-in low pass filter for the video input (Tx)? Otherwise one could put in a 1V p-p square wave at 5Mhz and the output spectrum would be very messy, or would it simply distort the signal?? ie, the fact that the bandwidth of the singal being sent is simply limited somehow to 5Mhz, in effect, acting as a lowpass filter.
Also, does anyone know of a simple way of creating a gaussian/raised root cosine function from a digital data stream? Without the need for high speed DACs and DSP.
Finally, what, in your opinion do you think the maximum datarate (symbol rate) for a 5Mhz bandwidth channel? Taking into account that I want to use simple filters, whether they're analogue, or digital.
Really long post, and probably a bit confusing, but any input on this subject is welcome. Again, sorry if this is in the wrong forum, since it involves RF, analogue and digital.
Buried(in)Code.
I've got 2 AV 2.4Ghz modules (Tx/Rx). The system can send a video signal (in this case, its designed for PAL-I) and a stereo audio signal. I was thinking of using the system for sending data, namely digital audio. The two audio channels can be used for control signals, since they are low bandwidth of approx. 20Khz each. Really my question relates to the bandwidth of the video channel. I know that PAL needs about 5-6Mhz for a fairly decent signal, also, there are some components in the PAL signal which are above this, but these are filtered out inside a TV/Video player. So, if I was to send digital data, it needs to be bandwidth-limited to 5Mhz. For a digital signal (square wave) a low-pass-filter is needed to reduce the harmonics of the signal. I was initially going to use an analogue (2-4 pole) filter but it seems that it would not provide enough attenuation. So I was thinking of making a sort of 'cheap' FIR filter, using a shift register, some logic, and a resistor ladder. I've built a test circuit and I can get a pretty smooth sine wave (after filtering with a 2-pole analogue LPF) when I put 1.,0,1,0,1 etc.. into the system. I was hoping to see the spectrum of this output but I don't have access to a spectrum analyzer, and the prospect of doing all the fourier transform by hand doesn't fill me with confidence, although it can be done. So, if you want me to get to the point:
Do these wireless AV modules have a built-in low pass filter for the video input (Tx)? Otherwise one could put in a 1V p-p square wave at 5Mhz and the output spectrum would be very messy, or would it simply distort the signal?? ie, the fact that the bandwidth of the singal being sent is simply limited somehow to 5Mhz, in effect, acting as a lowpass filter.
Also, does anyone know of a simple way of creating a gaussian/raised root cosine function from a digital data stream? Without the need for high speed DACs and DSP.
Finally, what, in your opinion do you think the maximum datarate (symbol rate) for a 5Mhz bandwidth channel? Taking into account that I want to use simple filters, whether they're analogue, or digital.
Really long post, and probably a bit confusing, but any input on this subject is welcome. Again, sorry if this is in the wrong forum, since it involves RF, analogue and digital.
Buried(in)Code.