Intensity Graph - X axis (time) Y axis (Frequency) Z value (intensity)

[26314362]

Dear All,

            I have been trying since last week to find solution to use  intensity graph in order to plot color intensity (z value) for X Y coordinates. X axis is time and Y axis is frequency which has range of (100 to 20,000). Now, I am able to interpolate index value of X array index (0 to 50k) in time after adding offset in the axis properties, but for Y axis there is no such relation between Y index and frequency value. I have been searching for solutions, but not able to figure it out. Finally I want to ask if what I want to do is it really possible with intensity graph or I should look for other solution?

Example : Array X (time) [10] = {1,2,3,4,5,6,7,8,9,10} ----> scaled to time{13:00:01 to 13:00:10} adding offset and changing display to time

                Array Y (freq.) [10] = {226, 248,300,375,430,575,690,825,925,1200} --> 10 data points of freq. to be mapped in Y axis

               Array Z (Intensity) [10]= {20, 35, 50,10,36,60,5,28,80,65} --> Z values of intensity to be mapped at (X,Y) as color intensity based on its value.

I hope it is clear what I need to do.

Thanks

[altenbach]

Intensity graphs take 2D arrays and since your data is not spaced equally in the Y axis you could initialize a 2D array of sufficient size and map your data into it by changing certain elements. If Z is a known function of x and y, you can do a nonlinear fit to the model and generate data for all array values.

[altenbach]

Another possibility would be  e.g. one of the 3D graphs (e.g. scatter)

[26314362]

Thank you for reply.

Z is not function of x and y. In reality, I need to plot 50k data points in single graph.  

How I can map Y index to its values since I only know the range in which the value would be. I mean I can't make any relation between Y index and value... Since intensity graph uses array index in Y axis, I don't understand how to scale to values...

Regards

[altenbach]

---

@26314362 wrote:

Since intensity graph uses array index in Y axis, I don't understand how to scale to values...

---

For each axis you can define offset and increment. For example if you set x0=100 and dx=2, the elements will correspond to 100, 102, 104, etc.

If you would attach some typical data, we can get a better idea what you are trying to do. If you have 50k data, how many in x and y? How irregular is the spacing? How smooth is the Z surface? Can you guarantee that there are no duplicate xy points? What does the data represent?

[26314362]

Thank you again for reply and apology for delayed response.

1. X axis is defined and its absolute time with delta of 0.1 sec.

2. Y axis values ranging from 100 to 20k, and it has repeat values every 50 iteration but there is no specific relation that can be defined in context to index values. Y values will be duplicated but not xy together.

3. Z is intensity values likely to have large contrast in terms of xy indexing. So its not smooth. 

I have attached sample data of 100 points. Column A,B,C are X,Y,Z.

[altenbach]

• So do you want to reset the time for each stretch of Y repeat?
• Your repeat seems to be 100 points and a triangle wave on the log scale.

So your x-axis should just have an increment of 0.1 and repeat for each Y repeat.

Your Y scale mapping could be set to logarithmic to have approximately linear mapping into points.

I assume that you probably want to start with the graph at the start of a new y ramp.

[26314362]

Let me put this way, 

I get about 50k or more data points per plot. 

• No reset of time or x axis on repeat , since xy combination would be unique
• Repeat of Y is not always equidistance as you can see from data... its 50 values in stair case pattern... like 1 to 50 and 50 to 1
• Can I still look for intensity graph as solution or figure out something else? 

[altenbach]

As I said, the intensity graph is always a rectangular grid of values and thus require equal spacing. 

If there is only one linear x-ramp (time), you don't have any second dimension. You can have unique xy pairs if the time would reset for each new Y repeat.

Do the Y repeats mean anything or could you remap the Y data into an equally spaced (in logarithmic units) linear Y ramp? (fitting, interpolation, etc.)