Skip to content

Files

Latest commit

 Cannot retrieve latest commit at this time.

History

History
25 lines (19 loc) · 2.37 KB

temporalwindow.md

File metadata and controls

25 lines (19 loc) · 2.37 KB

Temporal Window

Temporal Window

Using the Temporal Window

This window gives a dual graphical representation of CAN IDs over time. For each frame in the capture a circle is placed on the graph at a point given by the time when the frame came in on the X axis and the frame ID on the Y axis. That is, a frame with ID 0x300 that comes in at 23.543 seconds would be plotted at the point 23.543, 0x300. The graph background is color coded for "density." That brings up the question "What do you mean by density?" The answer is somewhat nebulous but essentially the color coded density maps where there is a lot of traffic within a small space on the graph. This could be because many frames with the same ID came in rapid fire or it could be because many frames with similar IDs came in very close to each other - or both.

The general point of this window is to show how active the bus is at any given point in time and which IDs are most active (they'll create bright streaks in the background color). There isn't a lot that can be done to modify the functionality of this window. All that can be done is zooming and panning the view. If you get the view too messed up the R key will reset the view back to standard. As with the Graphing Window, it is possible to select just the X or Y axis and zoom/pan that axis without affecting the other one. To do so, click on the axis marks of the axis you'd like to independently control. To control both again click in the graph itself.

So, what is it good for?

The idea here is to find areas of high traffic so that you can investigate those areas. Chances are, things that are sent rapidly are important. Steering data is usually sent rapidly, low level engine performance data is usually sent rapidly. These things can change rapidly and are safety critical so they tend to have a high rate of transmission. This window can also show where "clusters" of IDs are. Sometimes a given device will send long form data over a range of ideas, say, 0x102, 0x103, 0x104. These clusters could be visible on the temporal graph. A reasonable rule of thumb is that the "busy" areas of the temporal graph are probably important. The question is, for what? This window thus takes a supporting role and other windows would help answer the question of "why is there so much traffic there?"