English Deutsch Français Italiano Español Português 繁體中文 Bahasa Indonesia Tiếng Việt ภาษาไทย
All categories

I have many signals that I'm using to determine the temperature of a material. There are a few hundred signals. Each signal has a different amount of signal to noise. Knowing the signal to noise ratio how can I optimally average all the signals together to get the best overall measurement?

2006-10-29 12:45:58 · 1 answers · asked by professional student 4 in Science & Mathematics Engineering

1 answers

There's a couple of ways I can think of doing this.

1. Set a SNR threshold. Use only the samples above that threshold. Average the values with equal weight.

2. Use all the values with a weighted average. Have the weight for a particular value be 10^{-SNR} (assuming here that SNR is in dB)

Of course I don't know the complete set up, but I would use #1. Reason being I can't imagine a large variance in SNR if you are measuring the temperature of a single material. So basically you are just looking the bad samples and don't use them.

2006-10-29 13:18:00 · answer #1 · answered by cw 3 · 0 0

fedest.com, questions and answers