Raw data sampling rate information

I am collecting data from two sensors. Although the sensor data sheet provides sampling rate information, I want to confirm sampling rate independently from the raw data. Unfortunately, I do not have time stamp data with the raw data. Is there a better method you recommend to cross check the sampling rate of a time series data from the raw data itself?

6 commentaires

Walter Roberson
Walter Roberson le 23 Jan 2017
How are you collecting the data? The methods we suggest might be different for "reading from serial port" compared to "using Instrument Control Toolbox with a VISA driver" compared to "using Session-based Data Acquisition Toolbox with an NI-DAQmx device"
Madhu Kodappully
Madhu Kodappully le 23 Jan 2017
Modifié(e) : Madhu Kodappully le 23 Jan 2017
Dear Mr.Roberson,
Thank you for your response.Here is how the data is collected.
Sensor -1 : COM PORT (Windows Hyper terminal) Sensor -2 : Bluetooth terminal on smartphone
Guillaume
Guillaume le 23 Jan 2017
Unless the sensor does its own data acquisition, the sampling rate is independent of the sensor and only dependent on the device that does the acquisition (as long as you're within the frequency response of the sensor of course).
Madhu Kodappully
Madhu Kodappully le 23 Jan 2017
Dear Guillaume,
Thanks. My fundamental question was to extract sampling rate from raw data. I provided the acquisition details to aid Mr.Roberson. Any thoughts on how to extract sampling rate from time series?
Guillaume
Guillaume le 23 Jan 2017
Either I'm misunderstanding something or your question makes absolutely no sense. The sampling rate is an intrinsic property of your acquisition system. If you don't have the time at which the samples are acquired there's no way for you to know after the fact how frequently it was acquired.
e.g: I'm measuring the temperature of my office. I've measured [16 17 17 18 18 17 17]. Can you find the sampling rate from these values? No, you can't. Only I, the acquisition system, can tell you how frequently I measured it.
Madhu Kodappully
Madhu Kodappully le 24 Jan 2017
Dear Guillaume,
Thanks. This was helpful.

Connectez-vous pour commenter.

Réponses (1)

Jan
Jan le 23 Jan 2017

0 votes

You can check this with a defined signal. If you meaure e.g. a pressure, press the sensore for 1 second and count the length of the signal. Repeat this measurement to get a more accurate mean value.

4 commentaires

Madhu Kodappully
Madhu Kodappully le 23 Jan 2017
Dear Mr.Simon, The data is already collected. It is available as a time series and behavior does make sense. However sampling rate information is missing. Is there a data based approach to derive sampling rate?
Jan
Jan le 23 Jan 2017
No. If the rate has not been recorded, there is no way to guess it.
Madhu Kodappully
Madhu Kodappully le 24 Jan 2017
Dear Simon,
Thanks. This was helpful.
Walter Roberson
Walter Roberson le 24 Jan 2017
If your data has already been recorded and does not have time stamps then it is not possible to figure out the sampling rate.
If you were recording data "live" with MATLAB but wanted to check the rate at which data was actually being recorded, then there might be some possibilities. For example even though you might have instructed a device to send at 20 Hz, that does not mean that the device is really sending at 20 Hz and it does not mean that MATLAB is able to read at 20 Hz without losing data, so asking about the real rate that data is coming across is meaningful. However, that only works when the data is not pre-recorded.

Connectez-vous pour commenter.

Catégories

En savoir plus sur Data Acquisition Toolbox Supported Hardware dans Centre d'aide et File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by