AWR1642: RCS computation

I assume you have a stationary object at some distance (not necessarily at the max detectable range) and based on the demo data, you are trying to estimate the RCS of that object using the sensor. Typically the formula you quoted is used for maximum range calculation and/or (min) RCS at the maximum range for which the minimum detection SNR is assumed, I guess you are trying to measure the SNR of a detected object (probably kept much closer than max range, so SNR measured will be much higher than the minimum detection SNR) and then find RCS of the object. There is no calculation of SNR in dss_data_path.c but I assume you have changed it or offline using the range profile (doppler at zero velocity, assuming your object is stationary will show up in the range profile) and noise profile in the visualizer to find out the SNR at the detected object i.e you look at where you object is showing up in the range profile and see the value at that point and find the baseline value of noise profile (doppler at max velocity, assuming no real objects in scene are moving at max speeds, noise profile data in same format as range profile), and if they are in dB scale, then subtract to find the SNR in dB.

There are a few complexities to take into consideration beyond the simple formula you are using. For that I would like you to refer to  [related code is also in the visualizer]

If you run the estimator tool in your browser (link below the training video), you can see the source code to calculate the maximum range or the RCS corresponding to the max range by pressing F12, you can locate as below 

-----------

var min_rcs_detectable_at_max_range = function(maximum_detectable_range, cube_4pi, kB,
ambient_temperature, combined_factor_linear, lambda, num_virtual_rx, chirp_time, min_num_of_chirp_loops) {
// K9 =(0.9*E23*E23*E23*E23*F48*F47*F75*1000000*1000000)/(0.001*F83*K26*K26*F77*K14*K30)
return (0.9*Math.pow(maximum_detectable_range, 4)*
cube_4pi*kB*ambient_temperature*1e12)/(0.001*combined_factor_linear*lambda*lambda*num_virtual_rx*chirp_time*min_num_of_chirp_loops);
};

----------------

You can use the formula above for your calculations based on your chirping profile. The Tmeas in the formula you quoted corresponds to the total integration time = num_virtual_rx*chirp_time*min_num_of_chirp_loops in the above formula. This term assumes coherence integration but there is non-coherent combining across virtual antennas and so there is loss, this loss is taken into account in the one of the terms of "combined_factor_linear" (you can trace back the code to see), and this factor also has the gain terms etc incorporated and default values are what you see in the estimator tool GUI when you select xwr1642 in the device option, which you can verify against your assumed numbers [I think most of what you assumed is correct]. The SNR is embedded in the combined_factor_linear where you can plug your measured number and similarly the measured number for the range, this way you will incorporate other complexities (losses etc that you can additionally specify). See also 

The range index to meters conversion is described in the doxygen documentation of the out of box demo in section "Output infromation sent to host" (subsection is "List of detected objects"). Assumed temperature is also in the code and you can set to your environment where you are doing the measurement.

Note that this way of measuring will incorporate all energy in the signal corresponding to all objects present at (different angles at) the same radius as the object of interest (i.e no angle separation is there) so I guess assumption is your measurement environment is such that the clutter at this radius is low (ideally an anechoic chamber).