I'm receiving a signal at a USRP (B200) and want to calculate the RX power, given that the I know the TX packet has 500 samples. In MATLAB, I calculated the RX power this way: mean(abs(TXsignal).^2). But, I got this number 2.4*10^5, which is clearly not a reasonable RX power. How to map this number into a usable value that reflects the true RX power.
Dear Mohanad,
Yes your value of the power is high. The calculated power is that of the transmitter as i understood from the question. By definition the power Pt= sum X(n)^2/ N where N is the number of samples assuming the signal in volt in a reference resistance of one Ohm. If you have 50 ohm resistance you must divide the sum on it. The sum must be over all samples N.
In addition you have to take into consideration the communication channel attenuation alpha which is normally much smaller than one. Assuming it is a scaler quantity, then the received signal will be alpha* X(n) and the power at the receiver Pr,will be alpha^2 Pt, Alpha is estimated from the propagation model across a predetermined distance in a given environment.
wish you success
Most likely your calculated power is off by some factor. You could do a calibration by injecting a known signal power and calculating the power like you did before, to determine this factor.
For example you could inject a signal of e.g. -30 dBm (one microwatt), depending on the power handling of your receiver. And you may calculate your power as e.g. 1.8*10^5. If you convert this number to dBs, you get 10*log10(1.8*10^5)=52.6 dB, so the factor would be -30-52.6=82.6. Your calculation is offset by 82.6 dB.
To check this approach, you could then inject another power, e.g. -20 dBm, calculate the power like you did, apply the factor (i.e. subtract 82.6 dB) and see if you get the correct value (-20 dBm).
Obviously there are limits to this power calculation: at low signal levels you will reacht the noise floor of your receiver and at high powers the front-end will go into saturation. It may be useful to do some measurements to determine these edges of the useful range.
Good luck!
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators