AWR2243: Amplitude differences between channels

Hello Miles,

Apologies, If there is a single object, and the radar is perfectly calibrated, all channels would have the same amplitude (but different phase), The fourier transform of the antenna samples would yield a spectrum whose peak is a function of the angle of arrival. 

However, I think you are getting these values from the post-processing window?  If these are linear values, then the difference is not very large.  

Also, this is a good general reference on MIMO radars: https://www.ti.com/lit/an/swra554a/swra554a.pdf?ts=1728671312282&ref_url=https%253A%252F%252Fwww.google.com%252F 

Regards,

James

yes, this is the data before calibration and compensation.

Hi Miles,

These values are actually then not very different, and this shouldn't be a cause for concern.

Regards,

James

Hi James,

Yes,

indeed, after calibration, the amplitude differences between the various channels became relatively low.

Regards,

Miles

Hi James,

Additionally, I would like to ask if the actual RX GAIN on this link will be affected by temperature. There is a

temperature matching parameter in the runtime calibration, so will the RX GAIN be affected as the ambient

temperature changes?

Regards,

Miles

CalTempIdxTx / CalTempIdxRx /CalTempIdxLodist

/**
    * @brief  Override temperature index is used to calibrate Tx front end \n
    *      Idx val             Definition
    *         0:  Index for temperature < -30 deg C \n
    *         1:  Index for to temperature [-30, -20) deg C \n
    *         2:  Index for to temperature [-20, -10) deg C \n
    *         3:  Index for to temperature [-10, 0) deg C \n
    *         4:  Index for to temperature [0, 10) deg C \n
    *         5:  Index for to temperature [10, 20) deg C \n
    *         6:  Index for to temperature [20, 30) deg C \n
    *         7:  Index for to temperature [30, 40) deg C \n
    *         8:  Index for to temperature [40, 50) deg C \n
    *         9:  Index for to temperature [50, 60) deg C \n
    *         10: Index for to temperature [60, 70) deg C \n
    *         11: Index for to temperature [70, 80) deg C \n
    *         12: Index for to temperature [80, 90) deg C \n
    *         13: Index for to temperature [90, 100) deg C \n
    *         14: Index for to temperature [100, 110) deg C \n
    *         15: Index for to temperature [110, 120) deg C \n
    *         16: Index for to temperature [120, 130) deg C \n
    *         17: Index for to temperature [130, 140) deg C \n
    *         18: Index for to temperature >= 140 deg C \n
    */

Hi Miles,

The gain will vary a little with temperature, the exact trend may depend on how you have configured the device.  However, generally it should only vary a few dB.  

Regards,

James

Hi James,

Thank you very much for your reply! Yes, I also agree that there exists a certain compensatory relationship between temperature and gain. Could you elaborate on how

temperature specifically influences gain?

Regards,

Miles

Hi Miles,

usually, the RX gain decreases slightly with higher temperatures.  

Regards,

James

Hi James,

I apologize for not expressing myself clearly earlier. What I meant to ask is about the relationship between the temperature index values [CalTempIdxTx /

CalTempIdxRx / CalTempIdxLodist] used in runtime calibration and the gain, or rather, how should I set these temperature index values?

For instance, if the temperature of the Tx, obtained through the radar's built-in temperature acquisition interface, is 60 degrees Celsius, should CalTempIdxTx

be set to 10 (corresponding to the range of 60 to 70 degrees Celsius)? I have conducted tests with this setting, however, the read RX Gain was smaller than

the set RX Gain. On the other hand, when I set CalTempIdxTx to 18 (above 140 degrees Celsius), the read RX Gain was closer to the set RX Gain.

Regards,

Miles

Hi Miles,

Additionally, please refer to the following app note, section 4 for variation across temp.

Self-Calibration of mmWave Radar Devices (Rev. C)

Regards,

Aydin 

Hi Miles,

You can set this however you want given your requirements, but please note when you maximize the gain this way at 60C, you may have a problem when you go to a lower temperature and the gain becomes too high causing linearity issues.  As it is, the chip tries to pick the best code to give good performance over the entire temperature range. 

Regards,

Jamse

Hi James,

Yes, I'm just not clear about how to configure the temperature parameter. I couldn't find the specific section on how to configure this temperature parameter in

the document titled "Self-Calibration of mmWave Radar Devices (Rev. C)".

Could you please give me some detailed guidance?

Regards,

Miles

Hi James,

Thank you very much for your patient help and explanation!

Regards,

Miles

Our pleasure Miles, good luck in your development!

Regards,

James