ECE 454 and ECE 554 Supplemental reading by Don Johnson, et al - HTML preview

PLEASE NOTE: This is an HTML preview only and some elements such as links or page numbers may be incorrect.
Download the book in PDF, ePub for a complete version.

cn to make Equation true. If f[ n] is finite energy ( f[ n]∈ L 2[0, N] ), then the equality in Equation

holds in the sense of energy convergence; with discrete-time signals, there are no concerns for

divergence as there are with continuous-time signals.

The cn - called the Fourier coefficients - tell us "how much" of the sinusoid ⅇjω 0 kn is in f[ n] . The formula shows f[ n] as a sum of complex exponentials, each of which is easily processed by an LTI

system (since it is an eigenfunction of every LTI system). Mathematically, it tells us that the set

of complex exponentials { ⅇjω 0 kn , k∈ℤ } form a basis for the space of N-periodic discrete time

functions.

index-67_1.jpg

index-67_2.jpg

index-67_3.jpg

index-67_4.jpg

index-67_5.jpg

index-67_6.jpg

index-67_7.jpg

index-67_8.jpg

Equations

Now, in order to take this useful tool and apply it to arbitrary non-periodic signals, we will have to

delve deeper into the use of the superposition principle. Let sT( t) be a periodic signal having

period T. We want to consider what happens to this signal's spectrum as the period goes to

infinity. We denote the spectrum for any assumed value of the period by cn( T). We calculate the

spectrum according to the Fourier formula for a periodic signal, known as the Fourier Series (for

more on this derivation, see the section on Fourier Series.)

()

where

and where we have used a symmetric placement of the integration interval about the

origin for subsequent derivational convenience. We vary the frequency index n proportionally as

we increase the period. Define

()

making the corresponding Fourier Series

()

As the period increases, the spectral lines become closer together, becoming a continuum.

Therefore,

()

with

()

(3.16)

Discrete-Time Fourier Transform

(3.17)

Inverse DTFT

Warning

It is not uncommon to see the above formula written slightly different. One of the most

common differences is the way that the exponential is written. The above equations use the

index-68_1.jpg

index-68_2.jpg

index-68_3.jpg

radial frequency variable ω in the exponential, where ω=2 πf , but it is also common to include

the more explicit expression, ⅈ2πft , in the exponential. Sometimes DTFT notation is

expressed as F( ⅈω) , to make it clear that it is not a CTFT (which is denoted as F( Ω) ). Click

here for an overview of the notation used in Connexion's DSP modules.

DTFT Definition demonstration

Figure 3.9.

Click on the above thumbnail image (when online) to download an interactive Mathematica Player demonstrating Discrete Time

Fourier Transform. To Download, right-click and save target as .cdf.

DTFT Summary

Because complex exponentials are eigenfunctions of LTI systems, it is often useful to represent

signals using a set of complex exponentials as a basis. The discrete time Fourier transform

synthesis formula expresses a discrete time, aperiodic function as the infinite sum of continuous

frequency complex exponentials.

()

The discrete time Fourier transform analysis formula takes the same discrete time domain signal

and represents the signal in the continuous frequency domain.

()

3.5. Continuous-Time Fourier Transform*

index-69_1.jpg

index-69_2.jpg

index-69_3.jpg

index-69_4.jpg

3.5. Continuous-Time Fourier Transform*

In this lab, we learn how to compute the continuous-time Fourier transform (CTFT), normally

referred to as Fourier transform, numerically and examine its properties. Also, we explore noise

cancellation and amplitude modulation as applications of Fourier transform.

Properties of CTFT

The continuous-time Fourier transform (CTFT) (commonly known as Fourier transform) of an

aperiodic signal x( t) is given by

(3.18)

The signal x( t) can be recovered from X( ω) via this inverse transform equation

(3.19)

Some of the properties of CTFT are listed in Table 3.1.

Table 3.1. Properties of CTFT

Properties

Time domain

Frequency domain

Time shift

x ( tt 0 )

X ( ω ) ejωt 0

Time scaling

Linearity

a 1 x 1 ( t ) + a 2 x 2 ( t ) a 1 X 1 ( ω ) + a 2 X 2 ( ω ) Time convolution

x ( t ) ∗ h ( t )

X ( ω ) H ( ω )

Frequency convolution x ( t ) h ( t )

X ( ω ) ∗ H ( ω )

Refer to signals and systems textbooks m31521 - m31521 for more theoretical details on this transform.

Numerical Approximations to CTFT

Assuming that the signal x( t) is zero for t<0 and tT, we can approximate the CTFT integration in Equation (1) as follows:

index-70_1.jpg

index-70_2.jpg

index-70_3.jpg

index-70_4.jpg

(3.20)

where T= and N is an integer. For sufficiently small τ, the above summation provides a close

approximation to the CTFT integral. The summation

is widely used in digital signal

processing (DSP), and both LabVIEW MathScript and LabVIEW have a built-in function for it

called fft. In a .m file, if N samples x( ) are stored in a vector x, then the function call

>>xw=tau*fft (x)

calculates

(3.21)

where

(3.22)

with N assumed to be even. The fft function returns the positive frequency samples before the

negative frequency samples. To place the frequency samples in the right order, use the function

fftshift as indicated below:

>>xw=fftshift(tau*fft (x ) )

Note that X( ω) is a vector (actually, a complex vector) of dimension N. X( ω) is complex in

general despite the fact that x( t) is real-valued. The magnitude of X( ω) can be computed using the

function abs and the phase of X( ω) using the function angle.

3.6. Introduction*

Contents of Sampling chapter

Introduction(Current module)

Proof

Illustrations

index-71_1.jpg

Matlab Example

Hold operation

System view

Aliasing applet

Exercises

Table of formulas

Why sample?

This section introduces sampling. Sampling is the necessary fundament for all digital signal

processing and communication. Sampling can be defined as the process of measuring an analog

signal at distinct points.

Digital representation of analog signals offers advantages in terms of

robustness towards noise, meaning we can send more bits/s

use of flexible processing equipment, in particular the computer

more reliable processing equipment

easier to adapt complex algorithms

Claude E. Shannon

Figure 3.10.

Claude Elwood Shannon (1916-2001)

Claude Shannon has been called the father of information theory, mainly due to his landmark papers on the "Mathematical theory of communication" . Harry Nyquist was the first to state

index-72_1.jpg

the sampling theorem in 1928, but it was not proven until Shannon proved it 21 years later in the

paper "Communications in the presence of noise" .

Notation

In this chapter we will be using the following notation

Original analog signal x( t)

Sampling frequency Fs

Sampling interval Ts (Note that:

)

Sampled signal xs( n) . (Note that xs( n)= x( nTs) )

Real angular frequency Ω

Digital angular frequency ω. (Note that: ω= ΩTs )

The Sampling Theorem

The Sampling theorem

When sampling an analog signal the sampling frequency must be greater than twice the

highest frequency component of the analog signal to be able to reconstruct the original signal

from the sampled version.

Finished? Have at look at: Proof; Illustrations; Matlab Example; Aliasing applet; Hold

operation; System view; Exercises

3.7. Proof*

Sampling theorem

In order to recover the signal x( t) from it's samples exactly, it is necessary to sample x( t) at a

index-73_1.jpg

index-73_2.jpg

index-73_3.jpg

rate greater than twice it's highest frequency component.

Introduction

As mentioned earlier, sampling is the necessary fundament when we want to apply digital signal

processing on analog signals.

Here we present the proof of the sampling theorem. The proof is divided in two. First we find an

expression for the spectrum of the signal resulting from sampling the original signal x( t). Next we

show that the signal x( t) can be recovered from the samples. Often it is easier using the frequency

domain when carrying out a proof, and this is also the case here.

Key points in the proof

We find an equation for the spectrum of the sampled signal

We find a simple method to reconstruct the original signal

The sampled signal has a periodic spectrum...

...and the period is 2π Fs

Proof part 1 - Spectral considerations

By sampling x( t) every Ts second we obtain xs( n). The inverse fourier transform of this time

discrete signal is

()

For convenience we express the equation in terms of the real angular frequency Ω using ω= ΩTs .

We then obtain

()

The inverse fourier transform of a continuous signal is

()

From this equation we find an expression for x ( nTs)

index-74_1.jpg

index-74_2.jpg

index-74_3.jpg

index-74_4.jpg

index-74_5.jpg

index-74_6.jpg

index-74_7.jpg

index-74_8.jpg

index-74_9.jpg

()

To account for the difference in region of integration we split the integration in Equation into

subintervals of length

and then take the sum over the resulting integrals to obtain the complete

area.

()

Then we change the integration variable, setting

()

We obtain the final form by observing that ⅇⅈkn= 1 , reinserting η= Ω and multiplying by

()

To make xs( n)= x( nTs) for all values of n, the integrands in Equation and Equation have to agreee, that is

()

This is a central result. We see that the digital spectrum consists of a sum of shifted versions of

the original, analog spectrum. Observe the periodicity!

We can also express this relation in terms of the digital angular frequency ω= ΩTs

()

This concludes the first part of the proof. Now we want to find a reconstruction formula, so that

we can recover x( t) from xs( n).

Proof part II - Signal reconstruction

For a bandlimited signal the inverse fourier transform is

index-75_1.jpg

index-75_2.jpg

index-75_3.jpg

index-75_4.jpg

index-75_5.jpg

index-75_6.jpg

index-75_7.jpg

index-75_8.jpg

()

In the interval we are integrating we have:

. Substituting this relation into Equation

we get

()

Using the DTFT relation for Xs( ⅇⅈΩTs) we have

()

Interchanging integration and summation (under the assumption of convergence) leads to

()

Finally we perform the integration and arrive at the important reconstruction formula

()

(Thanks to R.Loos for pointing out an error in the proof.)

Summary

Spectrum sampled signal

Reconstruction formula

Go to Introduction; Illustrations; Matlab Example; Hold operation; Aliasing applet; System

view; Exercises ?

3.8. Illustrations*

In this module we illustrate the processes involved in sampling and reconstruction. To see how all

these processes work together as a whole, take a look at the system view. In Sampling and

reconstruction with Matlab we provide a Matlab script for download. The matlab script shows

the process of sampling and reconstruction live.

Basic examples

Example 3.6.

To sample an analog signal with 3000 Hz as the highest frequency component requires

sampling at 6000 Hz or above.

Example 3.7.

The sampling theorem can also be applied in two dimensions, i.e. for image analysis. A 2D

sampling theorem has a simple physical interpretation in image analysis: Choose the sampling

interval such that it is less than or equal to half of the smallest interesting detail in the image.

The process of sampling

We start off with an analog signal. This can for example be the sound coming from your stereo at

home or your friend talking.

The signal is then sampled uniformly. Uniform sampling implies that we sample every Ts seconds.

In Figure 3.11 we see an analog signal. The analog signal has been sampled at times t= nTs .

index-77_1.jpg

index-77_2.jpg

index-77_3.jpg

index-77_4.jpg

Figure 3.11.

Analog signal, samples are marked with dots.

In signal processing it is often more convenient and easier to work in the frequency domain. So

let's look at at the signal in frequency domain, Figure 3.12. For illustration purposes we take the

frequency content of the signal as a triangle. (If you Fourier transform the signal in Figure 3.11

you will not get such a nice triangle.)

Figure 3.12.

The spectrum X( ⅈΩ) .

Notice that the signal in Figure 3.12 is bandlimited. We can see that the signal is bandlimited

because X( ⅈΩ) is zero outside the interval [– Ωg, Ωg] . Equivalentely we can state that the signal has no angular frequencies above Ωg, corresponding to no frequencies above

.

Now let's take a look at the sampled signal in the frequency domain. While proving the sampling

theorem we found the the spectrum of the sampled signal consists of a sum of shifted versions of

the analog spectrum. Mathematically this is described by the following equation:

()

Sampling fast enough

index-78_1.jpg

index-78_2.jpg

index-78_3.jpg

In Figure 3.13 we show the result of sampling x( t) according to the sampling theorem. This means that when sampling the signal in Figure 3.11/Figure 3.12 we use Fs≥2 Fg . Observe in

Figure 3.13 that we have the same spectrum as in Figure 3.12 for Ω∈[-Ω