IWR6843ISK: Long range demo: Error convariance, g and confidence level meaning

Divya Teja Settimali

Part Number: IWR6843ISK

Hi,

while going through the source code of the people counting visualiser in python, I found out that the target structure contains an error convariance, confidence level and a value named g.

However, I couldn't find any documentation regarding these variables. Could you please explain what they are and how they could be useful in an application?

Kind regards,

Divya

over 4 years ago

+1 Jackson Thomas over 4 years ago

TI__Genius 14415 points

Hello Divya,

Those are the error covariance and gating gain of the tracking algorithm. They are not needed for normal tracking but can be used for info about the tracker decisions. Please see the following documentation for more information

3D_people_counting_demo_implementation_guide.pdf

Tracking_radar_targets_with_multiple_reflection_points.pdf

Regards,

Jackson

0 Divya Teja Settimali over 4 years ago

Intellectual 360 points

Hi Jackson,

thank you. Could you also tell me how to use the checksum from the header to check wether a frame is valid or not? I couldn't find any documentation.

Regards,

Divya

0 Jackson Thomas over 4 years ago in reply to Divya Teja Settimali

TI__Genius 14415 points

Hello Divya,

The logic can be seen in the uartTxTask in mss.main. It is copied below. Essentially it is the sum of all values in the header with some additional shifts.

headerPtr = (uint16_t *)&header;
for(n=0, sum = 0; n < sizeof(Pcount3DDemo_output_message_header)/sizeof(uint16_t); n++)
sum += *headerPtr++;
header.checkSum = ~((sum >> 16) + (sum & 0xFFFF));

Regards,

Jackson

Sensors

Sensors forum

IWR6843ISK: Long range demo: Error convariance, g and confidence level meaning