fourier analysis part 1: fourier series
DESCRIPTION
FOURIER ANALYSIS PART 1: Fourier Series. Dr.M.a.Kashem Asst. Professor,CSE,DUET. TOPICS. 1. Frequency analysis: a powerful tool 2. A tour of Fourier Transforms 3. Continuous Fourier Series (FS) 4. Discrete Fourier Series (DFS) 5. Example : DFS by DDCs & DSP. analysis. - PowerPoint PPT PresentationTRANSCRIPT
FOURIER ANALYSISPART 1: Fourier Series
Dr.M.a.Kashem
Asst. Professor,CSE,DUET
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 2 / 24
TOPICSTOPICS
1. Frequency analysis: a powerful tool
2. A tour of Fourier Transforms
3. Continuous Fourier Series (FS)
4. Discrete Fourier Series (DFS)
5. Example: DFS by DDCs & DSP
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 3 / 24
Frequency analysis: why?Frequency analysis: why? Fast & efficient insight on signal’s building blocks.
Simplifies original problem - ex.: solving Part. Diff. Eqns. (PDE).
Powerful & complementary to time domain analysis techniques.
Several transforms in DSPing: Fourier, Laplace, z, etc.
time, t frequency, fF
s(t) S(f) = F[s(t)]
analysianalysiss
synthesissynthesis
s(t), S(f) : Transform Pair
General Transform General Transform as problem-solving as problem-solving
tooltool
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 4 / 24
Fourier analysis - applicationsFourier analysis - applicationsApplications wide ranging and ever present in modern Applications wide ranging and ever present in modern
lifelife
• TelecommsTelecomms - GSM/cellular phones,
• Electronics/ITElectronics/IT - most DSP-based applications,
• EntertainmentEntertainment - music, audio, multimedia,
• Accelerator controlAccelerator control (tune measurement for beam steering/control),
• Imaging, image processing,Imaging, image processing,
• Industry/researchIndustry/research - X-ray spectrometry, chemical analysis (FT spectrometry), PDE solution, radar design,
• MedicalMedical - (PET scanner, CAT scans & MRI interpretation for sleep disorder & heart malfunction diagnosis,
• Speech analysisSpeech analysis (voice activated “devices”, biometry, …).
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 5 / 24
Fourier analysis - tools Fourier analysis - tools Input Time Signal Frequency
spectrum
1N
0n
Nnkπ2
jes[n]
N
1kc~
Discrete
DiscreteDFSDFSPeriodic (period T)
ContinuousDTFTAperiodic
DiscreteDFTDFT
nfπ2jen
s[n]S(f)
0
0.5
1
1.5
2
2.5
0 2 4 6 8 10 12
time, tk
0
0.5
1
1.5
2
2.5
0 1 2 3 4 5 6 7 8
time, tk
1N
0n
Nnkπ2
jes[n]
N
1kc~
**
**
Calculated via FFT**
dttfπj2
es(t)S(f)
dtT
0
tωkjes(t)T
1kc Periodic
(period T)Discrete
ContinuousFTFTAperiodic
FSFSContinuous
0
0.5
1
1.5
2
2.5
0 1 2 3 4 5 6 7 8
time, t
0
0.5
1
1.5
2
2.5
0 2 4 6 8 10 12
time, t
Note: j =-1, = 2/T, s[n]=s(tn), N = No. of samples
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 6 / 24
A little historyA little history Astronomic predictions by Babylonians/Egyptians likely via trigonometric sums.
16691669: Newton stumbles upon light spectra (specter = ghost) but fails to recognise “frequency” concept (corpuscularcorpuscular theory of light, & no waves).
1818thth century century: two outstanding problemstwo outstanding problems
celestial bodies orbits: Lagrange, Euler & Clairaut approximate observation data with linear combination of periodic functions; Clairaut,1754(!) first DFT formula.
vibrating strings: Euler describes vibrating string motion by sinusoids (wave equation). BUTBUT peers’ consensus is that sum of sinusoids onlyonly represents smooth curves. Big blow to utility of such sums for all but Fourier ...
18071807: Fourier presents his work on heat conduction Fourier presents his work on heat conduction Fourier analysis born. Fourier analysis born. Diffusion equation series (infinite) of sines & cosines. Strong criticism by peers
blocks publication. Work published, 1822 (“Theorie Analytique de la chaleur”).
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 7 / 24
A little history -2A little history -2 1919thth / 20 / 20thth century century: two paths for Fourier analysis - Continuous & Discrete.two paths for Fourier analysis - Continuous & Discrete.
CONTINUOUSCONTINUOUS
Fourier extends the analysis to arbitrary function (Fourier Transform).
Dirichlet, Poisson, Riemann, Lebesgue address FS convergence.
Other FT variants born from varied needs (ex.: Short Time FT - speech analysis).
DISCRETE: Fast calculation methods (FFT)DISCRETE: Fast calculation methods (FFT)
18051805 - Gauss, first usage of FFT (manuscript in Latin went unnoticed!!! Published 1866).
19651965 - IBM’s Cooley & Tukey “rediscover” FFT algorithm (“An algorithm for the machine calculation of complex Fourier series”).
Other DFT variants for different applications (ex.: Warped DFT - filter design & signal compression).
FFT algorithm refined & modified for most computer platforms.
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 8 / 24
Fourier Series (FS)Fourier Series (FS)
** see next slide see next slide
A A periodicperiodic function s(t) satisfying Dirichlet’s conditions function s(t) satisfying Dirichlet’s conditions ** can be can be
expressed as a Fourier series, with harmonically related sine/cosine expressed as a Fourier series, with harmonically related sine/cosine
terms.terms.
1kt)ω(ksinkbt)ω(kcoska0as(t)
a0, ak, bk : Fourier coefficients.
k: harmonic number,
T: period, = 2/TFor all t but For all t but
discontinuitiesdiscontinuities
Note: {cos(kωt),
sin(kωt) }k form orthogonal base of function space.
T
0
s(t)dtT
10a
T
0
dtt)ωsin(ks(t)T
2kb-
T
0
dtt)ωcos(ks(t)T
2ka
(signal average over a period, i.e. DC term & zero-frequency component.)
analysis
analysis
synthesis
synthesis
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 9 / 24
FS convergence FS convergence
s(t) piecewise-continuous;
s(t) piecewise-monotonic;
s(t) absolutely integrable , T
0
dts(t)
(a)
(b)
(c)
Dirichlet conditions
In any period:
Example: square wave
T
(a)
(b)
T
s(t)
(c)
if s(t) discontinuous then |ak|<M/k for large k (M>0)
Rate of Rate of convergenceconvergence
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 10 / 24
FS analysis - 1FS analysis - 1
* Even & Odd functions
Odd :
s(-x) = -s(x)
x
s(x)
s(x)
x
Even :
s(-x) = s(x)
FS of odd* function: square wave.
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10 t
sq
ua
re s
ign
al,
sw
(t)
0π
0
2π
π
1)dt(dt2π
10a
0π
0
2π
π
dtktcosdtktcosπ
1ka
kπcos1πk
2...
π
0
2π
π
dtktsindtktsinπ
1kb-
evenk,0
oddk,πk
4
1ω2πT
(zero average)(zero average)
(odd function)(odd function)
...t5sinπ5
4t3sin
π3
4tsin
π
4sw(t)
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 11 / 24
FS analysis - 2FS analysis - 2
-π/ 2
f 1 3f 1 5f 1 f
f 1 3f 1 5f 1 f
rk
θk
4/π
4/ 3π
fk=k /2
rK = amplitude,
K = phase
rk
k
ak
bkk = arctan(bk/ ak)
rk = ak2 + bk
2
zk = (rk , k)
vk = rk cos (k t + k)
Polar
Fourier spectrum Fourier spectrum representationsrepresentations
0k(t)kvs(t)
Rectangular
vk = akcos(k t) - bksin(k t)
4/π
f 1 2f 1 3f 1 4f 1 5f 1 6f 1 f
f 1 2f 1 3f 1 4f 1 5f 1 6f 1 f
4/ 3π
ak
-bk
Fourier spectrum Fourier spectrum of square-wave.of square-wave.
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 12 / 24
7
1ksin(kt)kb-(t)7sw
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10t
sq
ua
re s
ign
al,
sw
(t)
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10t
sq
ua
re s
ign
al,
sw
(t)
5
1ksin(kt)kb-(t)5sw
3
1ksin(kt)kb-(t)3sw
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10t
sq
ua
re s
ign
al,
sw
(t)
1
1ksin(kt)kb-(t)1sw
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10t
sq
ua
re s
ign
al,
sw
(t)
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10t
sq
ua
re s
ign
al,
sw
(t)
9
1ksin(kt)kb-(t)9sw
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10t
sq
ua
re s
ign
al,
sw
(t)
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10t
sq
ua
re s
ign
al,
sw
(t)
11
1ksin(kt)kb-(t)11sw
FS synthesisFS synthesisSquare wave Square wave
reconstruction from reconstruction from spectral termsspectral terms
Convergence may be slow (~1/k) - ideally need infinite terms.Practically, series truncated when remainder below computer tolerance
( errorerror). BUTBUT … Gibbs’ Phenomenon.
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 13 / 24
Gibbs phenomenonGibbs phenomenon
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10t
sq
ua
re s
ign
al,
sw
(t)
79
1kk79 sin(kt)b-(t)sw
Overshoot exist @ Overshoot exist @ each discontinuityeach discontinuity
• Max overshoot pk-to-pk = 8.95% of discontinuity magnitude.Just a minor annoyance.
• FS converges to (-1+1)/2 = 0 @ discontinuities, in this casein this case.
• First observed by Michelson, 1898. Explained by Gibbs.
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 14 / 24
FS time shifting FS time shifting
-1.5
-1
-0.5
0
0.5
1
1.5
0 2 4 6 8 10 t
sq
ua
re s
ign
al,
sw
(t)
FS of even function:FS of even function: /2-advanced square-/2-advanced square-
wavewave
π
f 1 3f 1 5f 1 7f 1 f
f 1 3f 1 5f 1 7f 1 f
rk
θk
4/ π
4/ 3π
00a
0kb-
even.k,0
11...7,3,kodd,k,πk
4
9...5,1,kodd,k,πk
4
ka
(even function)
(zero average)
phas
phas
ee
ampl
itude
ampl
itude
Note: amplitudes unchanged BUTBUT phases advance by k/2.
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 15 / 24
Complex FSComplex FS
Complex form of FS (Laplace 1782). Harmonics ck separated by f = 1/T on frequency plot.
r
a
b = arctan(b/ a)
r = a2 + b2
z = r ej
2
eecos(t)
jtjt
j2
eesin(t)
jtjt
Euler’s notation:
e-jt = (ejt)* = cos(t) - j·sin(t) “phasor”
NoteNote: c-k = (ck)*
kbjka2
1kbjka
2
1kc
0a0c Link to FS real Link to FS real
coeffs.coeffs.
k
tωkjekcs(t)
T
0
dttωkj-es(t)T
1kc
analysis
analysis
synthesis
synthesis
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 16 / 24
FS propertiesFS properties Time FrequencyTime Frequency
** Explained in next week’s lectureExplained in next week’s lecture
Homogeneity a·s(t) a·S(k)
Additivity s(t) + u(t) S(k)+U(k)
Linearity a·s(t) + b·u(t) a·S(k)+b·U(k)
Time reversal s(-t) S(-k)
Multiplication * s(t)·u(t)
Convolution * S(k)·U(k)
Time shifting
Frequency shifting S(k - m)
m
m)U(m)S(k
td)tT
0
u()ts(tT
1
S(k)e Ttk2π
j
s(t)Ttm2π
je
)ts(t
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 17 / 24
FS - “oddities”FS - “oddities”
k = - , … -2,-1,0,1,2, …+ , ωk = kω, k = ωkt, phasor turns anti-clockwise.
Negative k phasor turns clockwise (negative phase k ), equivalent to negative time t,
time reversal.
Negative frequencies & time reversal
Fourier components {Fourier components {uukk} form orthonormal base of signal space:} form orthonormal base of signal space:
uk = (1/T) exp(jkωt) (|k| = 0,1 2, …+) Def.: Internal product :
uk um = δk,m (1 if k = m, 0 otherwise). (Remember (ejt)* = e-jt )
Then ck = (1/T) s(t) uk i.e. (1/T) times projectionprojection of signal s(t) on component uk
Orthonormal base
T
o
*mkmk dtuuuu
CarefulCareful: phases important when combining several signals!
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 18 / 24
0 50 100 150 200
k f
Wk/W0
10-3
10-2
10-1
12
Wk = 2 W0 sync2(k )
W0 = ( sMAX)2
1k 0WkW
10WW
FS - power FS - power
• FS convergence FS convergence ~1/k~1/k
lower frequency terms
Wk = |ck|2 carry most power.
• WWkk vs. ω vs. ωkk: Power density spectrum: Power density spectrum.
Example
sync(u) = sin( u)/( u)
Pulse train, duty cycle Pulse train, duty cycle = 2 = 2 / T/ T
T
2
t
s(t)
bk = 0 a0 = sMAX
ak = 2sMAX sync(k )
Average power WAverage power W : s(t)s(t)T
o
dt2s(t)T
1W
1k
2kb2
ka2
120a
k
2kcW
Parseval’s TheoremParseval’s Theorem
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 19 / 24
FS of main waveformsFS of main waveforms
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 20 / 24
Discrete Fourier Series (DFS)Discrete Fourier Series (DFS)
N consecutive samples of N consecutive samples of s[n] completely describe s s[n] completely describe s in time or frequency in time or frequency domains.domains.
DFS generate periodic ck with same signal period
1N
0k
Nnk2π
jekcs[n] ~
Synthesis: finite sum band-limited s[n]
Band-limited signal s[n], period = N.
mk,δ1N
0n
N-m)n(k2π
je
N
1
Kronecker’s delta
Orthogonality in DFS:
synthesis
synthesis
1N
0n
Nnk2π
jes[n]
N
1kc~
Note:Note: ck+N = ck same period N
i.e. time periodicity propagates to frequencies!i.e. time periodicity propagates to frequencies!
DFS defined as:DFS defined as:
~~~~
analysis
analysis
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 21 / 24
DFS analysis DFS analysis DFS of periodic discreteDFS of periodic discrete
1-Volt square-wave1-Volt square-wave
otherwise,
Nkπ
sin
NkLπ
sin
N
N1)(Lkπ
je
2N,...N,0,k,N
L
kc~
0.6
0 1 2 3 4 5 6 7 8 9 10 k
1
0 2 4 5 6 7 8 9 10 n
k
-0.4
0.2 0.24 0.24
0.6
0.6 0.24
1
0.24
-0.2
0.4
0.2
-0.4
-0.2
0.4
0.2
0.6 ck ~
ampl
itude
ampl
itude
phas
e
phas
e
Discrete signals Discrete signals periodic frequency spectra. periodic frequency spectra.Compare to continuous rectangular function (slide # 10, “FS analysis - 1”)
-5 0 1 2 3 4 5 6 7 8 9 10 n 0 L N
s[n] 1
s[n]: period NN, duty factor L/NL/N
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 22 / 24
DFS propertiesDFS propertiesTime FrequencyTime Frequency
Homogeneity a·s[n] a·S(k)
Additivity s[n] + u[n] S(k)+U(k)
Linearity a·s[n] + b·u[n] a·S(k)+b·U(k)
Multiplication * s[n] ·u[n]
Convolution * S(k)·U(k)
Time shifting s[n - m]
Frequency shifting S(k - h)
1N
0h
h)-S(h)U(kN
1
* * Explained in next week’s lectureExplained in next week’s lecture
1N
0m
m]u[ns[m]
S(k)e Tmk2π
j
s[n]Tth2π
je
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 23 / 24
QQ[[ttpp]]:: ““QQuuaaddrraattuurree””
s[tn] s(t)
cos[LO tn]
II [[ttnn]]
-sin[LO tn]
QQ[[ttnn]]
ADC (f S)
II [[ttpp]]:: ““II nn--pphhaassee””
LPF &
DECI MATI ON
N = NS/ NT
TO DSP (next slide)
DDII GGII TTAALL DDOOWWNN CCOONNVVEERRTTEERR
DFS analysis: DDC + ...DFS analysis: DDC + ...s(t) periodic with period Ts(t) periodic with period TREV REV (ex: particle bunch in “racetrack” (ex: particle bunch in “racetrack” accelerator)accelerator)
tn = n/fS , n = 1, 2 .. NS , NS = No. samples
(1)
(1)
(2)
(2) I[tn ]+j Q[tn ] = s[tn ] e -jLOtn
(3)
(3)I[tp ]+j Q[tp ] p = 1, 2 .. NT , Ns / NT = decimation. (Down-converted to baseband).
0 f
BBB
BBB
f LO f
M. E. Angoletta - DISP2003 - Fourier analysis - Part 1: Fourier Series 24 / 24
... + DSP... + DSP
DDCs with different fDDCs with different fLOLO yield more DFS yield more DFS componentscomponents
LO = REV
harmonic 1 harmonic 2 harmonic 3
LO = 2 REV LO = 3 REV
DSP
Fourier coefficients a Fourier coefficients a k*k*, b , b k*k*
harmonic k* = LO/REV
TN
1ppI
TN
1*ka
TN
1ppQ
TN
1*kb
Parallel digital input from ADC
TUNABLE LOCAL OSCI LLATOR
(DI RECT DI GI TAL SYNTHESI ZER)
Clock from ADC
sin cos
COMPLEX MI XER fS fS fS/N
Decimation factor N
I
Q
Central frequency
LOW PASS FI LTER
(DECI MATI ON)
Example: Real-life DDC