Pulse generation

Hi JS1998,

I'm not sure I understand what you're trying to do.

If you want your baseband data to be a 500 ns pulse, you can do that by making a 10 sample long pulse at 20 MHz.

Kind Regards,
Michał

Hi Michal, I want to use a frequency hopping application, and the pulses at each frequency are 500ns, my current understanding is that if I use DMA to transmit a pulse on I and zero on Q, I can get the pulse width of the DMA transmission length, but my test found that I did not see the high frequency pulse on the oscilloscope. I want to know if I'm right. Looking forward to your reply.

Hi JS1998,

I'm afraid I still don't follow. When you say "the pulses at each frequency are 500ns" - are you talking about the baseband data containing pulses of 500 ns, or are you trying to enable the channel for 500 ns at a time?

Kind Regards,
Michał

I know these are two concepts, but what I think is that after I set the baseband signal to 500ns, the pulse will be multiplied by the local oscillator signal. This way I don't have to worry about the time it takes to enable the channel (500ns pulses can be produced even using 38.4MHz, no external clock source is required), so the final pulse emitted is still 500ns. The reason I'm thinking about doing this is because I use PS to control HOP signals and find that private timers using PS have a hard time determining the length of my hop cycle, so I can't control the Tx setup signal and HOP signal very smoothly.

Hi JS1998,

I am struggling to understand what you are after. Can you perhaps provide diagrams (including timing diagrams) of what you are trying to do?

Kind Regards,
Michał

This is just an idea of mine, I don't know if it will work. I don't use orthogonal up-conversion, I just use I. The I subband signal is a narrow pulse, while the timing of the local oscillator part provides a signal with a longer time width. After mixing the two, the result will be a narrow pulse of high frequency signal.Because 500ns is very narrow, it seems difficult to control the timing from the user document. So I think in terms of signal modulation. This is my idea map.

Hi,

We are currently auditing our engineerzone forums for questions that have not been fully answered. If any more information is needed, please reply to this post or create a new post.

For frequency hopping minimum timings, review the 'MODES OF OPERATION' section in the ADRV9002 user guide, frequency hopping chapter. The minimum timing will be significantly higher than 500ns in all cases.

It is of course possible to send/receive tones at different baseband frequencies. However, generating IQ data isn't something we can help with on this forum. Note that the maximum channel bandwidth of the ADRV9002 is 40MHz, so any generated tones must be within this range.

Regards,

Conrad