Decrease time in plotting the graph

4 vues (au cours des 30 derniers jours)
Nishant Gupta
Nishant Gupta le 8 Avr 2012
% This code is to make a sparse matrix of the roof, from where the light can come
% in the receiver (Rx). This code would work for receiver at any position, orientation, field of view, etc.
clear;clc;
t0 = tic;
% ----------------Dimensions of the room------------------------------------------
length = 10.0; % Length of Room [m]
breadth = 10.0; % Breadth of Room [m]
height = 5.0; % height of Room [m]
% ---------------------------------------------------------------------------------
% -----------------Characteristics of Transmitter/Source - LED--------------------
n = 4; % Number of LEDs is given by
LED_VA = 60.0; % <Find out what is it called> ????
Ntx = [0 0 -1]; % Normal to the Tx's plane (Orientation)
% Let the position of Tx be Pos_Tx
i=0; j=0; k=5;
Pos_Tx = [i j k];
% ---------------------------------------------------------------------------------
% ----------------Characteristics of Receiver - Photodiode-------------------------
Rtx = [0 0 1]; % Normal to the Rx's plane (Orientation)
Rx_FOV = 36.0; % FOV (field of view)
psi_c = (Rx_FOV*pi)/180; % FOV (field of view) in Radians
% Let the position of Rx be Pos_Rx
Pos_Rx = [5 5 0];
% ---------------------------------------------------------------------------------
space = 0.5;
LOS = zeros(max(size(0:space:length)), max(size(0:space:breadth)));
toc(t0)
hold on;
t1 = tic;
r=1;s=1;
for i=0:space:length
r=1;
for j=0:space:breadth
Pos_Tx = [i j k];
% ---------------------------------------------------------------------------------
DD_Tx_Rx = Pos_Rx - Pos_Tx; % Direction from Tx to Rx
DD_Rx_Tx = -1*DD_Tx_Rx; % Direction from Rx to Tx
% Distance between Tx and Rx
R = sqrt((Pos_Tx(1,1) - Pos_Rx(1,1)).^2 +(Pos_Tx(1,2) - Pos_Rx(1,2)).^2 +(Pos_Tx(1,3) - Pos_Rx(1,3)).^2);
% Angle between orientation of Tx and direction of Rx
Phi = acosd( dot(DD_Tx_Rx,Ntx)./(R.*(sqrt((Ntx(1,1)^2 + Ntx(1,2)^2 + Ntx(1,3)^2)))));
% Angle between orientation of Rx and direction of Tx
Psi = acosd( dot(DD_Rx_Tx,Rtx)./(R.*(sqrt((Rtx(1,1)^2 + Rtx(1,2)^2 + Rtx(1,3)^2)))));
phi = (Phi*pi)/180; % Angle of irradiance (phi) in Radians
psi = (Psi*pi)/180; % Angle of incidence (psi) in Radians
% ---------------------------------------------------------------------------------
% ----------------- For Direct Light ----------------------------------------------
% LOS (line of sight) propagation path channel transfer function - HLOS
if (psi>=0) && (psi<=psi_c) && (Phi<=LED_VA)
HLOS = 1.0;
elseif (psi>psi_c)
HLOS = 0.0;
elseif (Phi>LED_VA)
HLOS = 0.0;
elseif (Psi>Rx_FOV)||(Phi>Rx_FOV)
HLOS = 0.0;
end;
LOS(r,s) = HLOS;
% ---------------------------------------------------------------------------------
plot3(i,j,LOS(r,s),'.b');
r = r+1;
end;
s = s+1;
end;
grid on;
toc(t1)
Nothing much to understand in this code.
The main thing is that plot3 function is taking 75% (almost 2.5 sec.) of all time. Is there a better way to plot the function in much less time than that?
Any other idea for optimizing the code is welcome... Please help out. Thanks Nishant

Réponses (1)

Daniel Shub
Daniel Shub le 8 Avr 2012
You are calling plot3 in your loops. You would probably be better off saving all the data and calling plo3 once.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by