I need to measure the receiver sensitivity of a COTS Transceiver ADF7021, but I have only a Synthesized Signal generator (Aplab 2131) with the following specifications
- 9Khz - 1GHz, 1Hz resolution
- Output power range of -127 dBm to 13dBm with 0.1dB resolution
- Modulation type: FM, PM, AM
I'm using FM modulation by feeding the signal generator with a square wave of appropriate frequency and amplitude so that I can get the desired data rate(1200 bps), frequency deviation (3KHz) as I needed (I verified these by measuring the modulated signal characteristics using a spectrum analyzer (RIGOL DSA815-TG)).
Then I connected the modulated wave to the receiver IC and I got stream of 1010.. or 0101.. as I expected. I measured the sensitivity to be around -118 dBm (below this level, the received data had significant number of errors (2 Bytes out of 50 Bytes received are erroneous)
I want to find out if these tests are reliable or is there any better way to test the receiver sensitivity?
Since this is a COTS item, you should be able to find its specification which would certainly include sensitivity. If you are trying to confirm that your unit meets the spec, you really need to use the same test setup and waveforms that the manufacturers used.
If you are trying find out the sensitivity with your own waveforms and are only interested in digital signals, a bit error rate tester (BER) would be more accurate but you should use your intended modem to get the right modulation. Applying a square wave just gives you FSK.
For an analogue measurement, you could also apply sine wave modulation (typically at 1kHz or over the modulation bandwidth) and measure the SNR of the receiver output versus RF signal level.
Hello,
this looks rather good. Though there is some space for improvement:
Currently you are testing with a square wave. This applies some limitations regarding the "test depth".
If you have access to a BERT (Bit Error Rate Tester) you should apply this one to get a better assessment of the receiver sensitivity. If not, you could build your own pseudo-random pattern generator (basically a shift register with some feedback) to apply a 'random' pattern to your transmission line. Without the BERT it will be a bit more complicated to check for errors, but is is doable with an oscilloscope and eventually some delay line.
Regards
- Badges
- Science topic