Post on 17-Dec-2015
Lecture 2: Functions, Plots & Symbolic Math
Instructor: Hamze Msheik
LEBANESE AMERICAN UNIVERSITYSchool of Engineering
Department of Electrical and Computer Engineering
Fall 2014
Function Files
Creating MATLAB Functions
Simple programming in MATLAB
Introduction
3
Script files are used to write programs, to save and to run them using MATLAB commands.
Script files contain list of MATLAB commands. Script are saved in “filename.m” Select File menu in the MATLAB toolbar:
File>New>M-file The Editor window will open.
Creating Script Files
4
Commands in M-files are executed in the order they are listed.
Script files can be edited and executed many times. The program is automatically saved when Run button
is pressed.
Run command
Breakpoints
5
Breakpoints can be placed in script files for debugging.
A breakpoint can be set for every line. Place the cursor on the desired line. Press F12 or choose Debug>Set/Clear Breakpoint
from Editor toolbar.
Program stops and Workspace variables are updated once the program reaches a breakpoint press continue button to resume the program
Example of M-file
6
Function files
7
MATLAB has its own set of built-in and toolbox functions Example: sin, exp, rand, plot…
User can write his own function and executes it.
Function files are written using the Editor Window.
A function has a name, can have input arguments and output arguments.
The first line in Function files has the following format: function [out1,…,outm] =function-name(arg1,
…,argn) [out1,…,outm] is an array that returns m
variables (arg1,…,argn) is a list of n arguments that
the function takes.
Function Files
8
function[out1,…,outm]=function-name(arg1,…,argn)
Function-name It’s the function name. The M-file should have the same name as
function-name M-files that contain functions are saved as
Function-name.m In order to call a user defined function,
function-name.m should exist in the Current Directory.
Example of function files
9
Let’s define a function that transforms the polar coordinates into their corresponding Cartesian coordinates.
Let’s name the function pol_to_cart This function has 2 input arguments:
r and theta This function has 2 output arguments:
x and y
Example of function files
10
Calling User defined functions
11
>> [x,y]=pol_to_cart(1,pi/6)x = 0.86603y = 0.5User defined functions can run in Command
Window, or in Script Files or in other Function Files.
Function Overloading
12
The behavior of a function can be modified depending on the number of input and output arguments.
Commands: nargin: Number of Input arguments
nargout: Number of Output arguments Example:
function [P]=integral(t,x,flag)if nargin==2
dt=t(2)-t(1); P=sum(x)*dt;
elseif nargin==3 dt=t(2)-t(1); P=cumsum(x).*dt;
end
Saving Data
13
MATLAB uses its own platform independent file format for saving data files. Files have a “.mat” extension save is used to save variables from the workspace
to a named file (by default: matlab.mat if no filename is given) save filename – saves entire workspace to
filename.mat save var1 var2 … filename – saves named variables
to filename.mat By default save overwrites an existing file of the
same name, use –append to append data to an existing file Variables of the same name are always overwritten!!! save var1 var2 … filename -append
Data is recovered using load command load filename – loads entire .mat file load filename var1 var2 … - loads named vars
Prompting for User input
14
The input function can be used to prompt the user for numeric or string input.
x = input(‘Enter a value for x’);YourName = input(‘Enter your
name’, ‘s’);
MatLab Programming
15
Program Control Statements:
Conditional Control (if, switch) Loop Control (for, while, continue,
break) Error Control (try, catch) Program Termination (return)
Conditional Control (if statement)
16
clear;x=-2;y=10;if(x<0) angle=180+atand(y/x)else angle=atand(y/x)end
Loop Control (For loop)
17
Loop Expression Format:for index = start:increment:end statementsend
Example:for n = 2:size(x,2) x(n) = 2 * x(n - 1);end
The command break exits a loop
Comparison of break and return
18
break is used to escape the current loop. return is used to escape the current function.
Remark: an infinite loop can be broken by typing “Ctrl+C”
Plotting in MATLAB
Plot Function
Subplot function
Plotting Properties
Plotting
20
MATLAB is used extensively to plot graphs. Different parameters can be modified in
MATLAB figures: Number of subplots in a figure Scale (i.e. Linear or Logarithmic) Grid, colors, labels and legends.
Consider the plotting example: t=0:0.001:20; f=exp(-t).*sin(2*pi*10*t); plot(t,f), grid
Note: t and f have the same size
Plot command
21
Plotting a function
22
The Figure can be edited and labeled such that:
ylabel('F(t)')xlabel('t')title('F(t)=e^{-t}.sin(20\pit)')axis([0,2,-1,1]) % axis([xmin, xmax, ymin,
ymax])legend('F(t)')
Plotting a function
23
Plotting 2 curves on the same graph
24
Use the help command to get more information on the plot command.
t=0:0.0001:3;f=exp(-t).*sin(2*pi*10*t);g=cos(2*pi*t);plot(t,f,'r‚)holdplot(t,g,'b--')legend('f(t)','g(t)')
t=0:0.0001:3;f=exp(-t).*sin(2*pi*10*t);g=cos(2*pi*t);plot(t,f,'r',t,g,'b--')legend('f(t)','g(t)')
2 ways
Plotting 2 curves on the same graph
25
Plotting 2 curves in 1 graph window
26
t=0:1e-4:3;f=exp(-t).*sin(2*pi*10*t);g=cos(2*pi*t);subplot(2,1,1),plot(t,f),gridsubplot(2,1,2),plot(t,g),grid
subplot(M,N,n) creates an array of M-by-N graphs in a figure window, where n is the number of a selected graph in the array.
Plotting 2 curves in 1 graph window
27
Plotting 4 curves in 1 graph window
28
t=0:1e-4:3;f= exp(-t).*sin(2*pi*10*t);g= cos(2*pi*t);y= tan(2*pi*t);w= sinc(t); subplot(2,2,1),plot(t, f),gridsubplot(2,2,2),plot(t, g),gridsubplot(2,2,3),plot(t, w),gridsubplot(2,2,4),plot(t, y),grid
Plotting 4 curves in 1 graph
29
Plotting curves in different graph windows
30
To open a new window and to plot a curve, we use the command: figure
Example:
t=0:0.1:10;y1=sin(2*pi*t);y2=exp(-0.5*t)*sin(2*pi*t);plot(t,y1)figure,plot(t,y2)
In this way, the first graph stays in the first window, and a new window opens and displays the second graph.
Plotting curves in different windows
31
Semi-logarithmic plot
32
Semi-logarithmic plot is the same as plot, except a logarithmic (with base 10) scale is used for the Y-axis.
Semi-logarithmic plots are used extensively in communication systems to plot the error probability versus the Signal-to-Noise ratio (SNR) Referred to as Bit Error Rate (BER) graphs. Such plots are very useful in comparing
corresponding communication systems
See also: semilogx, loglog
Semi-logarithmic plot - Example
33
Consider the below example, where two types of signaling are to be compared:
SNR=[-200:10:-110];Pe1=[0.498 0.442 0.378 0.221 0.109 0.0254 0.0124 0.00213 0.00118
0.000175];Pe2=[0.748 0.582 0.378 0.121 0.059 0.0154 0.0064 0.00113
0.000568 0.0000175];
semilogy(SNR, Pe1,'ro-',SNR,Pe2,'b^-'),gridtitle('Symbol Error Probability')xlabel('SNR(dB)')ylabel('P_S')legend('Binary Signaling','4-ary Signaling')
Semi-logarithmic plot - Example
34
Multidimensional Functions
35
Consider the function f(x, y):
f(x,y) is defined over the range: -2<x<2 -3<y<3
To define the domain of definition of this multidimensional function, we use the command: ndgrid
22
),( yxxeyxf
Multidimensional Functions
36
Plotting f(x,y) is done by:
x=-2:0.2:2;y=-3:0.2:3;[X,Y]=ndgrid(x,y);f=X.*exp(-X.^2-Y.^2);mesh(X,Y,f)
Use the help command to further understand ndgrid and mesh commands.
It plots in a 3-Dimensional Space the points whose coordinates are in matrices X, Y and f.
Multidimensional Functions
37
-2-1.5
-1-0.5
00.5
11.5
2
-3
-2
-1
0
1
2
3
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
xy
f
Symbolic Math Toolbox
Symbolic Math
Mathematical Operations Using Symbolic Math
Symbolic Math
39
It defines variables that don’t have necessarily a defined scalar or numerical value.
Symbolic objects can be used as independent variables. For Example we can use the command syms such that:
syms x y z Therefore x, y and z are three Symbolic Objects.
A symbolic object can be: A variable with no pre-assigned numerical value.
Ex: syms x y t A number.
Ex: a=sym(2) An expression made of symbolic variable/numbers.
Ex: syms x y; z=sqrt(x+y)
A symbolic expression is a mathematical expression made of one or several symbolic objects.
Creating Symbolic Objects
40
We use the commands sym and syms to create symbolic objects.
To create one Symbolic Object: a=sym(2)
a has a symbolic-numerical value b=sym(‘gamma’)
b has a symbolic-string value.
To create multiple symbolic objects: syms x y z
Symbolic v.s Numerical
41
Symbolic objects: a and b a=sym(1); b=sym(2); f=a/b f = ½
Numerical objects: A and B:A=1;B=2;F=A/BF =0.500
Symbolic to Numeric
42
Some symbolic expressions can have numerical values resulting from numerical operations.
To convert from symbolic to numerical objects, we use the command double(S).
Example:
a=sym(3);b=1/ab = 1/3B=double(b)B = 0.33333
findsym command
43
Used to find and enumerate symbolic variables present in an expression.
Command Syntax: findsym(S)
Displays all symbolic variables found in S, in alphabetical order.
findsym(S,n) Displays the first n symbolic variables found in S, in
default order Default order Start from x (first symbol in the expression) and
list the others in the order of their closeness to x.
subs command
44
To substitute a variable in a symbolic expression we use the command subs. Consider the following example: syms x y; f=2*x+log(y); subs(f,x,2); ans = 4+log(y) subs(f,[x y],[2 1]) ans = 4
The Factor command: sym x
G=-1/2*exp(-x)*cos(x)-1/2*exp(-x)*sin(x)
factor(G)
ans = -1/2*exp(-x)*(cos(x)+sin(x)) See also: expand, simplify, pretty.
Solving equations (Symbolic)
45
Solving algebraic equations. Example (2 equations, 2 unknowns)
[x y]=solve('2*x+a*y-1=0','b*x+2*y=0') x =-2/(b*a-4) y =1/(b*a-4)*b
Solving Differential equations: ‘Dn’ represents the nth order derivative operator Consider the Example: x’’+a2x=0, IC: x(0)=1,
x’(0)=1 x = dsolve('D2x+a^2*x=0','x(0)=1, Dx(0)=1')
x = 1/a*sin(a*t)+cos(a*t)
Calculating Derivatives (Symbolic)
46
Partial Derivatives Consider the function given by:
f = sin(x^2+y)
Partial derivative w.r.t. x f_x=diff(f,x) f_x =2*cos(x^2+y)*x
2nd Partial derivative w.r.t. x f2_x=diff(f,x,2) f2_x = -4*sin(x^2+y)*x^2+2*cos(x^2+y)
Calculating Integrals (Symbolic)
47
Integration Consider the function given by:
f=sin(x)*exp(-x)
Definite integral (i.e. Integrate f(x) over the range [0,2]) F=int(f,x,0,2) F =-1/2*exp(-2)*cos(2)-1/2*sin(2)*exp(-2)+1/2
Indefinite Integral G=int(f,x) G =-1/2*exp(-x)*cos(x)-1/2*sin(x)*exp(-x)
Inline function
48
Used for multivariable expressions. Returns a function that takes several
arguments. As an example:
D=inline('2*x*y+sin(x)','x','y') D = Inline function: D(x,y) = 2*x*y+sin(x)
D(1,2) ans = 4.8415
Linear Math
Expressing Polynomials in MATLAB
Polynomials
50
Polynomials are represented in MATLAB using row vectors.
Polynomial coefficients are ordered by descending power.
For example: F(x)=x4+2x-1 F=[1 0 0 2 -1]
Polynomial roots
51
To get the roots of a polynomial expression, we use the command roots.
Example: Finding the roots of F(x)=x3-x+2 (i.e. Solving
P(x)=0) F=[1 0 -1 2] roots(F)
ans = -1.5214 0.76069 + 0.85787i 0.76069 - 0.85787i
F(x) has one real and two complex conjugate solutions
Polynomial coefficients
52
If we want to form a polynomial having specific roots, we use the command poly.
Example: Form a polynomial having the following roots: 2, 1-j, 1+jpoly([2;1-j;1+j])ans = 1 -4 6 -4
The corresponding polynomial is: F(x)=x3-4x2+6x-4
Note: that the roots are stored in a column vector, and the polynomial coefficients are stored in a row vector.
Evaluating Polynomials
53
To evaluate polynomials for a specific value of the variable we use the command polyval.
Example: Calculate F(2). F = 1 0 -1 2
polyval(F,2)ans = 8
Multiplication of Polynomials (Convolving)
54
We use the command conv to multiply two polynomials.
Example: Given P(x)=x3+2x-1 and Q(x)=x2-1, calculate
P(x)*Q(x).P=[1 0 2 -1];Q=[1 0 -1];conv(P,Q)ans = 1 0 1 -1 -2 1
Then: P(x)*Q(x)=x5+x3-x2-2x+1
Division of Polynomials (Deconvolving)
55
The output of the division of two polynomials is represented in the following expression:
A(x)/B(x) = q(x) + r(x)/B(x)
q(x) is the quotient of the polynomial division and r(x) is the remainder.
To divide two polynomials, we use the command deconv.
Division of Polynomials
56
Consider A(x)=x4+2x2-1 and B(x)=x3+x. Find A(x)/B(x):
A=[1 0 0 2 -1];B=[1 0 1 0];[q,r]=deconv(A,B)q = 1 0r = 0 0 -1 2 -1
Where q(x) = x and r(x)=-x2+2x-1, the final answer, will look like:
xx
xxx
xB
xA
3
2 12
Polynomials operations
57
Polynomial derivative: Use the command polyder
Partial Fraction Expansion: Use the command residue to get the partial
fraction expansion.
Elaborate on these commands on your own.