IQ sample interpretation

[idel]

Hi!

We're working on an AoA application using a custom nRF5340 board and antenna array. Just trying to fully understand the interpretation of IQ data, there are a couple fine details that are not clearly to me from the product documentation.

On a 1M Phy, the CTE is offset 250KHz from the carrier frequency. So, for example, for the channel at 2.44GHz, the CTE will be at 2 440 250 000 Hz.

For IQ sampling, are the local in-phase and quadrature signals tuned to 2.44GHz, or 2.44GHz+250KHz?

If the local oscillator is set to the carrier frequency, and is not modulated by 250KHz, then we would expect phase samples at 1us apart to be offset by 90 degrees, is this correct?

Otherwise, if the local oscillator is tuned to match the CTE, then consecutive phase samples should be close to each other (offset due to subtle clock differences between Tx and Rx), correct?

The reason for asking this is that we're seeing larger-than-expected offset on consecutive IQ samples, and we can't yet determine if this is due to:

• A bigger-than-expected difference in Tx/Rx clocks
• Offset between CTE and Carrier frequency, if carrier is used as IQ reference
• A combination of both

Would appreciate anyone with knowledge of the internals of the IQ modulator to share any insight.

Thanks!

[Dmitry]

Hi,

idel said:

For IQ sampling, are the local in-phase and quadrature signals tuned to 2.44GHz, or 2.44GHz+250KHz?

As spec says, "The IQ samples are recorded with respect to the RX carrier frequency", that's 2.44GHz.

idel said:

If the local oscillator is set to the carrier frequency, and is not modulated by 250KHz, then we would expect phase samples at 1us apart to be offset by 90 degrees, is this correct?

In theory, yes. In real world, 32M oscillators of RX and TX side have some offset. For example, 40ppm will be multiplied by PLL and translated to ~100kHz offset for carrier frequency, so you can see 150kHz or 350kHz. That's what the reference period is for: you can measure a real frequency of CTE during 8 us then use this value in phase calculations.