Injecting noise to see Bit error rate response of coaxial cable
时间：04-04
整理：3721RD
点击：
Hi
I'm experimenting with power line communication by connecting 2 modems (Texas Instuments PLC developer's kits) through 100m of RG58 coaxial cable as a model power line and injecting white gaussian noise with a signal generator. The modems exchange modulated packets of data shown in fig 1.
I was hoping to see the response of the Bit error rate at different modulation schemes to increasing noise levels but the results from the GUI are never more than 0.00 Packet Error Rate (PER) when connected or 100% when disconnected (which is the spike at 200 in fig 3).
I've tried isolating the data packet on an oscilloscope (fig 1) and flooding the cable with noise (fig 2) which I'm sure should be causing some errors but is not shown in the GUI. In fact injecting large amounts of noise seems to have very little effect on Received signal (RSSI) and SNR which should definitely have noticeable variances.
Could anyone tell me why this might be the case? Or under what conditions would the zero configuration GUI read an actual PER and BER other than 0 or 100%?
Many Thanks!
I'm experimenting with power line communication by connecting 2 modems (Texas Instuments PLC developer's kits) through 100m of RG58 coaxial cable as a model power line and injecting white gaussian noise with a signal generator. The modems exchange modulated packets of data shown in fig 1.
I was hoping to see the response of the Bit error rate at different modulation schemes to increasing noise levels but the results from the GUI are never more than 0.00 Packet Error Rate (PER) when connected or 100% when disconnected (which is the spike at 200 in fig 3).
I've tried isolating the data packet on an oscilloscope (fig 1) and flooding the cable with noise (fig 2) which I'm sure should be causing some errors but is not shown in the GUI. In fact injecting large amounts of noise seems to have very little effect on Received signal (RSSI) and SNR which should definitely have noticeable variances.
Could anyone tell me why this might be the case? Or under what conditions would the zero configuration GUI read an actual PER and BER other than 0 or 100%?
Many Thanks!
