How to hold different constellations using scatterhistogram?

Dear all,

I hope the title is explicit enough, does anyone have an idea?

Thanks in advance.

Best regards,

louis

Example of code:

% signal vector
X = rand(1,1e3)+1i*rand(1,1e3);

% power in signal
PX= mean(abs(X).^2);

% noise of variance power in signal/10
B = PX/10*(randn(1,1e3)+1i*randn(1,1e3));

% signal + noise
Y = X+B;

% plot
hold all

scatterhistogram(real(X),imag(X))

Error using scatterhistogram
Adding ScatterHistogramChart to axes is not supported. Turn hold off.

scatterhistogram(real(Y),imag(Y))

Answer 1

[Adam Danz]

scatterhistogram produces a ScatterHistogramChart object within a figure. When you call hold on, an axes is created if it doesn't already exist, and the axes is prepared to accept new objects. scatterhistogram tells you that you cannot assign the ScatterHistogramChart object to the axes.

Remove hold on so scatterhistogram can create the plot in a figure. Scatterhistogram() is not set up to combine two ScatterHistogramChart objects within the same set of axes.

[Louis Tomczyk]

Dear @Adam Danz,

Yes I misplaced the 'hold on' line.

That's really a pity that we cannot combine more scatterhistograms charts.

Best,

louis

[Adam Danz]

@Louis Tomczyk, you can, however, group your data together in a table and use a grouping variable to plot multiple sets of data into the same scatterhistogram chart. There are some examples in the documentation: scatterhistogram.

load patients
tbl = table(LastName,Diastolic,Systolic,Smoker)

tbl = 100×4 table
      LastName      Diastolic    Systolic    Smoker
    ____________    _________    ________    ______

    {'Smith'   }       93          124       true  
    {'Johnson' }       77          109       false 
    {'Williams'}       83          125       false 
    {'Jones'   }       75          117       false 
    {'Brown'   }       80          122       false 
    {'Davis'   }       70          121       false 
    {'Miller'  }       88          130       true  
    {'Wilson'  }       82          115       false 
    {'Moore'   }       78          115       false 
    {'Taylor'  }       86          118       false 
    {'Anderson'}       77          114       false 
    {'Thomas'  }       68          115       false 
    {'Jackson' }       74          127       false 
    {'White'   }       95          130       true  
    {'Harris'  }       79          114       false 
    {'Martin'  }       92          130       true  

s = scatterhistogram(tbl,'Diastolic','Systolic','GroupVariable','Smoker');

[Adam Danz]

Alternatively, you can append the existing data using set(__). Here's a demo.

x = randn(1,50)*2+9; 
y = randn(1,50)+5; 

s = scatterhistogram(x,y);

Append s with additional data and create a grouping variable g to identify which group each datapoint belongs to. I created a new figure here for demonstration purposes.

figure
s = scatterhistogram(x,y);
x2 = rand(1,40)+6;
y2 = rand(1,40)*2+4; 

g = [ones(numel(x),1); ones(numel(x2),1)+1];

set(s, 'XData', [x,x2], 'YData', [y,y2], 'GroupData', g);