Synchronous and Asynchronous Data Sources in Data Acquisition Systems

Synchronous and Asynchronous Data Sources in Data Acquisition Systems video

Data acquisition systems typically gather data from two kinds of sources – synchronous and asynchronous sources. In this knowledge packets video, David Buckley talks about what this means and how to best capture this data to ensure data coherency.

Video Transcript

Hi, my name is Dave Buckley and today we are going to talk about data sources that you find on data acquisition units (DAU). In general, there are two types – synchronous and asynchronous. Synchronous data sources are those that are under the control of the DAU itself. They are typically samples of data that repeat periodically. This could be an analog channel such as an accelerometer, a strain gauge, or even a discrete channel that captures some kind of bi-level input. The DAU will control the sampling of these channels and it will gather the data and transmit it out so it can be analyzed.

The second type of input is asynchronous – these channels are not typically under the control of the DAU so the data will arrive in an asynchronous manner to the DAU itself. So, where you will have even spaced samples in the synchronous data, you will very often have uneven samples from the asynchronous data source. These are the two types of data sources that you will find in a DAU.

It’s very important for the sake of post-flight analysis and correlation of parameters that every single data point captured can be correlated so that you know when every sample was taken. If we take the synchronous example, you want to make sure that the sampling for all channels occurs at the same time. However, you may have a situation where you are sampling different channels at different rates.

If we look at an example of this and take a 1 Hz signal, a 2 Hz signal, and a 4 Hz signal, you will see that each channel is sampled at a different rate. If you take the 1st sample from each channel, they are all sampled at exactly the same time so the pattern of samples repeats over and over again. If you take the superset of all the samples – what we describe as an acquisition cycle – and you follow this sampling schema, you will be able to very easily correlate the samples from each channel together and know when they were sampled.

If you then took an asynchronous source and compared it to what was going on with the synchronous data, you will find, as mentioned before, that you’re not under control of when the messages arrive – they may arrive at unexpected intervals. All you can do is monitor the bus and wait until the samples arrive. If you want to correlate the asynchronous samples with the synchronous samples, you need to timestamp exactly when the asynchronous samples arrive.

The synchronous sampling data will be timed using an absolute time known by the DAU and you will use that time to put timestamps on the asynchronous samples. Then, by comparing the timestamps of the asynchronous data with the known time of the synchronous traffic, you will be able to correlate everything in your system together.