IWR6843: Range–Doppler heatmap values and dB conversion (IWR6843, mmWave SDK 3.6.2.0 LTS)

Chung-Hsun Lo

Part Number: IWR6843

Hi,

I’m currently using the IWR6843 (mmWave SDK 3.6.2.0 LTS) and would like to confirm how the Range–Doppler heatmap values are represented. Do these values correspond to power, or are they logarithmic? If they are not power, is there a complete conversion formula to express them as power in dB?

Best regards,

CH.

9 days ago

0 Josh Dye 8 days ago

TI__Mastermind 21511 points

Hi CH,

The information you are looking for is included in the SDK documentation. Please refer to the dopplerproc doxygen here:
C:/ti/mmwave_sdk_03_06_02_00-LTS/packages/ti/datapath/dpc/dpu/dopplerproc/docs/doxygen/html/index.html.m The section called Summation includes the dB conversion formula you are looking for.

Best regards,

Josh

0 Chung-Hsun Lo 2 days ago in reply to Josh Dye

Prodigy 30 points

Hi Josh,

Thank you for your reply. I’m using DopplerProcDSP and would like your help confirming whether my interpretation for converting detection-matrix values to dB is correct, because the converted numbers look high.

According to the Doppler DPU documentation, each cell in the detection matrix stores a Q8 fixed-point statistic of log2(|·|) accumulated along the virtual-antenna dimension.

My conversion to dB is as follows:

Recover the real value: S = value / 2^8
Convert to amplitude dB: dB = (20 / log2(10)) × S

Example (value = 4300):

dB = 6.0206 × 4300 / 256 = 101.127266 dB

Questions:

Are the above steps—and the result (~101.1 dB)—reasonable for a single cell?
In practice, for common targets or corner-reflector tests, what is the typical dB range after converting a single detection-matrix cell? Is a value around ~100 dB commonplace?

Best regards,
CH