{"id":18200,"date":"2025-08-29T15:31:18","date_gmt":"2025-08-29T19:31:18","guid":{"rendered":"https:\/\/blogs.mathworks.com\/simulink\/?p=18200"},"modified":"2025-09-02T14:23:04","modified_gmt":"2025-09-02T18:23:04","slug":"extracting-timetables-simulink-style","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/simulink\/2025\/08\/29\/extracting-timetables-simulink-style\/","title":{"rendered":"Extracting Timetables… Simulink style"},"content":{"rendered":"
\"\"<\/img><\/a><\/div>
Today I want to share a simple tip to interact with data logged from a Simulink model<\/span><\/div>

The Problem<\/span><\/h2>
Let's take this simple model that logs 3 signals with different sample times:<\/span><\/div>
\"\"<\/img><\/div>
After simulating, it is possible to use <\/span>extractTimetable<\/span><\/a> to create a single <\/span>timetable<\/span><\/a> with all the logged signals:<\/span><\/div>
out = sim(<\/span>'testLog'<\/span>);<\/span><\/span><\/div><\/div>
TT = out.logsout.extractTimetable<\/span><\/span><\/div>
<\/div>
<\/div>
<\/div>
<\/div>
TT = <\/span>11\u00d73 timetable <\/span><\/div>
<\/span><\/th>Time<\/span><\/th>sineTs1<\/span><\/th>sineTs2<\/span><\/th>sineTs3<\/span><\/th><\/tr><\/thead>
1<\/span><\/th>0 sec<\/span><\/td>0<\/span><\/td>0<\/span><\/td>0<\/span><\/td><\/tr>
2<\/span><\/th>1 sec<\/span><\/td>0.0628<\/span><\/td>NaN<\/span><\/td>NaN<\/span><\/td><\/tr>
3<\/span><\/th>2 sec<\/span><\/td>0.1253<\/span><\/td>0.1253<\/span><\/td>NaN<\/span><\/td><\/tr>
4<\/span><\/th>3 sec<\/span><\/td>0.1874<\/span><\/td>NaN<\/span><\/td>0.1874<\/span><\/td><\/tr>
5<\/span><\/th>4 sec<\/span><\/td>0.2487<\/span><\/td>0.2487<\/span><\/td>NaN<\/span><\/td><\/tr>
6<\/span><\/th>5 sec<\/span><\/td>0.3090<\/span><\/td>NaN<\/span><\/td>NaN<\/span><\/td><\/tr>
7<\/span><\/th>6 sec<\/span><\/td>0.3681<\/span><\/td>0.3681<\/span><\/td>0.3681<\/span><\/td><\/tr>
8<\/span><\/th>7 sec<\/span><\/td>0.4258<\/span><\/td>NaN<\/span><\/td>NaN<\/span><\/td><\/tr>
9<\/span><\/th>8 sec<\/span><\/td>0.4818<\/span><\/td>0.4818<\/span><\/td>NaN<\/span><\/td><\/tr>
10<\/span><\/th>9 sec<\/span><\/td>0.5358<\/span><\/td>NaN<\/span><\/td>0.5358<\/span><\/td><\/tr>
11<\/span><\/th>10 sec<\/span><\/td>0.5878<\/span><\/td>0.5878<\/span><\/td>NaN<\/span><\/td><\/tr><\/tbody><\/table><\/div><\/div><\/div><\/div><\/div><\/div>
What's the problem? In one word: <\/span>NaN<\/span>.<\/span><\/div>
By default, <\/span>extractTimetable<\/span> inserts <\/span>NaN<\/span><\/a> for steps where signals with a sample time slower than the fastest rate did not execute. While this is technically correct because there was no data point at this time, I find this not ideal for data logged from Simulink. <\/span><\/div>
In Simulink, if I was to connect a <\/span>Rate Transition<\/span><\/a> block and other blocks downstream from this logged signal, they would not receive NaNs. Instead, the previous value would be held.<\/span><\/div>
Those NaNs also make plots look strange. Here is the data as seen in the <\/span>Simulation Data Inspector<\/span><\/a>:<\/span><\/div>
\"\"<\/img><\/div>
When plotting the timetable, the NaNs cause gaps in the signals:<\/span><\/div>
stackedplot(TT);<\/span><\/span><\/div>
<\/div>
<\/div>
<\/div>
<\/div>
<\/div><\/div><\/div><\/div><\/div>

The Solution<\/span><\/h2>
To obtain a timetable that I find better aligned with Simulink's behavior, you can first extract the data as a cell array that contains one timetable per signal, and then combine them using the synchronize function:<\/span><\/div>
TT = out.logsout.extractTimetable(OutputFormat=<\/span>\"cell-by-signal\"<\/span>);<\/span><\/span><\/div><\/div>
TT = synchronize(TT{:},<\/span>\"union\"<\/span>,<\/span>\"previous\"<\/span>)<\/span><\/span><\/div>
<\/div>
<\/div>
<\/div>
<\/div>
TT = <\/span>11\u00d73 timetable <\/span><\/div>
<\/span><\/th>Time<\/span><\/th>sineTs1<\/span><\/th>sineTs2<\/span><\/th>sineTs3<\/span><\/th><\/tr><\/thead>
1<\/span><\/th>0 sec<\/span><\/td>0<\/span><\/td>0<\/span><\/td>0<\/span><\/td><\/tr>
2<\/span><\/th>1 sec<\/span><\/td>0.0628<\/span><\/td>0<\/span><\/td>0<\/span><\/td><\/tr>
3<\/span><\/th>2 sec<\/span><\/td>0.1253<\/span><\/td>0.1253<\/span><\/td>0<\/span><\/td><\/tr>
4<\/span><\/th>3 sec<\/span><\/td>0.1874<\/span><\/td>0.1253<\/span><\/td>0.1874<\/span><\/td><\/tr>
5<\/span><\/th>4 sec<\/span><\/td>0.2487<\/span><\/td>0.2487<\/span><\/td>0.1874<\/span><\/td><\/tr>
6<\/span><\/th>5 sec<\/span><\/td>0.3090<\/span><\/td>0.2487<\/span><\/td>0.1874<\/span><\/td><\/tr>
7<\/span><\/th>6 sec<\/span><\/td>0.3681<\/span><\/td>0.3681<\/span><\/td>0.3681<\/span><\/td><\/tr>
8<\/span><\/th>7 sec<\/span><\/td>0.4258<\/span><\/td>0.3681<\/span><\/td>0.3681<\/span><\/td><\/tr>
9<\/span><\/th>8 sec<\/span><\/td>0.4818<\/span><\/td>0.4818<\/span><\/td>0.3681<\/span><\/td><\/tr>
10<\/span><\/th>9 sec<\/span><\/td>0.5358<\/span><\/td>0.4818<\/span><\/td>0.5358<\/span><\/td><\/tr>
11<\/span><\/th>10 sec<\/span><\/td>0.5878<\/span><\/td>0.5878<\/span><\/td>0.5358<\/span><\/td><\/tr><\/tbody><\/table><\/div><\/div><\/div><\/div><\/div><\/div>
This looks better.<\/span><\/div>
stackedplot(TT);<\/span><\/span><\/div>
<\/div>
<\/div>
<\/div>
<\/div>
<\/div><\/div><\/div><\/div><\/div>

Now it's your turn<\/span><\/h2>
Do you prefer the default behavior of <\/span>extractTimetable<\/span>? Would you prefer if the \"zero-order hold\" behavior shown in this post was the default? Let us know in the comments below.<\/span><\/div>\r\n
\r\n<\/div>