Spectral Leakage Pp289 jdsp. Freq of kth sample, No centering.

Slides:



Advertisements
Similar presentations
| Page Angelo Farina UNIPR | All Rights Reserved | Confidential Digital sound processing Convolution Digital Filters FFT.
Advertisements

DCSP-13 Jianfeng Feng Department of Computer Science Warwick Univ., UK
Islamic university of Gaza Faculty of engineering Electrical engineering dept. Submitted to: Dr.Hatem Alaidy Submitted by: Ola Hajjaj Tahleel.
Department of Kinesiology and Applied Physiology Spectrum Estimation W. Rose
Ch 3 Analysis and Transmission of Signals
Digital Kommunikationselektronik TNE027 Lecture 5 1 Fourier Transforms Discrete Fourier Transform (DFT) Algorithms Fast Fourier Transform (FFT) Algorithms.
Thursday, October 12, Fourier Transform (and Inverse Fourier Transform) Last Class How to do Fourier Analysis (IDL, MATLAB) What is FFT?? What about.
ACHIZITIA IN TIMP REAL A SEMNALELOR. Three frames of a sampled time domain signal. The Fast Fourier Transform (FFT) is the heart of the real-time spectrum.
1 Chapter 16 Fourier Analysis with MATLAB Fourier analysis is the process of representing a function in terms of sinusoidal components. It is widely employed.
Let’s go back to this problem: We take N samples of a sinusoid (or a complex exponential) and we want to estimate its amplitude and frequency by the FFT.
Techniques in Signal and Data Processing CSC 508 Fourier Analysis.
LINEAR-PHASE FIR FILTERS DESIGN
Discrete Fourier Transform(2) Prof. Siripong Potisuk.
Continuous Time Signals All signals in nature are in continuous time.
Spectral Analysis Spectral analysis is concerned with the determination of the energy or power spectrum of a continuous-time signal It is assumed that.
Introduction To Signal Processing & Data Analysis
EEE422 Signals and Systems Laboratory Filters (FIR) Kevin D. Donohue Electrical and Computer Engineering University of Kentucky.
Relationship between Magnitude and Phase (cf. Oppenheim, 1999)
Leakage & Hanning Windows
Goals For This Class Quickly review of the main results from last class Convolution and Cross-correlation Discrete Fourier Analysis: Important Considerations.
EE513 Audio Signals and Systems Digital Signal Processing (Systems) Kevin D. Donohue Electrical and Computer Engineering University of Kentucky.
Topic 7 - Fourier Transforms DIGITAL IMAGE PROCESSING Course 3624 Department of Physics and Astronomy Professor Bob Warwick.
Discrete-Time and System (A Review)
DTFT And Fourier Transform
Lecture 1 Signals in the Time and Frequency Domains
Motivation Music as a combination of sounds at different frequencies
Power Spectral Density Function
Chapter 4 The Frequency Domain of Signals and Systems.
TELECOMMUNICATIONS Dr. Hugh Blanton ENTC 4307/ENTC 5307.
Pg. 335 Homework Pg. 346#1 – 14 all, 21 – 26 all Study Trig Info!! #45#46 #47#48 #49Proof#50Proof #51+, +, + #52 +, –, – #53–, –, + #54 –, +, – #55 #56.
The Story of Wavelets.
Transforms. 5*sin (2  4t) Amplitude = 5 Frequency = 4 Hz seconds A sine wave.
Preprocessing Ch2, v.5a1 Chapter 2 : Preprocessing of audio signals in time and frequency domain  Time framing  Frequency model  Fourier transform 
Spectral Analysis AOE March 2011 Lowe 1. Announcements Lectures on both Monday, March 28 th, and Wednesday, March 30 th. – Fracture Testing –
1 BIEN425 – Lecture 10 By the end of the lecture, you should be able to: –Describe the reason and remedy of DFT leakage –Design and implement FIR filters.
ECE 4710: Lecture #6 1 Bandlimited Signals  Bandlimited waveforms have non-zero spectral components only within a finite frequency range  Waveform is.
EE104: Lecture 5 Outline Review of Last Lecture Introduction to Fourier Transforms Fourier Transform from Fourier Series Fourier Transform Pair and Signal.
Chapter 6 Spectrum Estimation § 6.1 Time and Frequency Domain Analysis § 6.2 Fourier Transform in Discrete Form § 6.3 Spectrum Estimator § 6.4 Practical.
Real time DSP Professors: Eng. Julian S. Bruno Eng. Jerónimo F. Atencio Sr. Lucio Martinez Garbino.
Noise and Sensitivity of RasClic 91
0 - 1 © 2007 Texas Instruments Inc, Content developed in partnership with Tel-Aviv University From MATLAB ® and Simulink ® to Real Time with TI DSPs Spectrum.
Lecture#10 Spectrum Estimation
GG313 Lecture 24 11/17/05 Power Spectrum, Phase Spectrum, and Aliasing.
STORM WP7 Software library for nonlinear analysis of fluctuations in space plasma time series P. Kovács, A. Koppán C. Munteanu, M. Echim, STORM Annual.
Professor A G Constantinides 1 Finite Wordlength Effects Finite register lengths and A/D converters cause errors in:- (i) Input quantisation. (ii)Coefficient.
Vibrationdata 1 Power Spectral Density Function PSD Unit 11.
Fourier Analyses Time series Sampling interval Total period Question: How perturbations with different frequencies contribute to the turbulent kinetic.
Learning from the Past, Looking to the Future James R. (Jim) Beaty, PhD - NASA Langley Research Center Vehicle Analysis Branch, Systems Analysis & Concepts.
Complex Form of Fourier Series For a real periodic function f(t) with period T, fundamental frequency where is the “complex amplitude spectrum”.
Chapter 4 Discrete-Time Signals and transform
Quadratic Functions.
CS 591 S1 – Computational Audio
EEE422 Signals and Systems Laboratory
Spectral Analysis Spectral analysis is concerned with the determination of the energy or power spectrum of a continuous-time signal It is assumed that.
MECH 373 Instrumentation and Measurements
FFT-based filtering and the
NONPARAMETRIC METHODs (NPM) OF POWER SPECTRAL DENSITY ESTIMATION P. by: Milkessa Negeri (…M.Tech) Jawaharlal Nehru Technological University,India December.
Fourier Analyses Time series Sampling interval Total period
OptiSystem-MATLAB data formats (Version 1.0)
Wavelet transform Wavelet transform is a relatively new concept (about 10 more years old) First of all, why do we need a transform, or what is a transform.
Fourier Analyses Time series Sampling interval Total period
Hanning and Rectangular Windows
Leakage Error in Fourier Transforms
Chapter 8 The Discrete Fourier Transform
Uses of filters To remove unwanted components in a signal
Lecture 5A: Operations on the Spectrum
Lec.6:Discrete Fourier Transform and Signal Spectrum
Electrical Communication Systems ECE Spring 2019
ENEE222 Elements of Discrete Signal Analysis Lab 9 1.
Electrical Communications Systems ECE
Presentation transcript:

Spectral Leakage Pp289 jdsp

Freq of kth sample, No centering

Centering and the PSD PSDs are centered if the DC component is at N/2, where N= number of samples. Before and after centering

How do I center the FFT? UlawCodec ulc = new UlawCodec(); final double[] doubleArray = ulc.getDoubleArray(); double d[] = Mat1.truncateToIntegralPowerOfTwo(doubleArray); FFT1dDouble f = new FFT1dDouble(); double in_im[] = new double[d.length]; Mat1.centering(d); f.computeForwardFFT(d, in_im); f.normalize(); final double[] psd = f.getPSDNotNormalized(); Graph.graph(psd, "", "", "psd"); f.computeBackwardFFT(d, in_im); Mat1.centering(d); UlawCodec ulc2 = new UlawCodec(f.getRealData()); ulc2.play();

How do I implement centering? public static void centering(double r[]) { int s = 1; for (int i = 0; i < r.length; i++) { s = -s; r[i] *= s; }

How do I compute freqs?

Computing Freqs The center freq should be zero. But what about 4001 and 3999?

Thus every bin gives one freq Bins give a unique freq. K=n/2=0 hz K=n = 4000 hz K=0= hz

Real signals Real signals have zero complex components Real signals have symmetrical PSDs Real signals are positive on the right Real signals are negative on the left One bin maps to one frequency

Spectral leakage frequency analysis effectfrequency analysis Applies to: – finite-length signals –finite-length segments energy has "leaked" out of the original signal spectrum into other buckets.

Windowing window function (also known as an apodization function or tapering function) a function that is zero-valued outside of some chosen intervalfunctioninterval

Windowing windowing is the root cause of spectral leakage

Does Windowing Really Help? Non-rectangular window functions actually increase the total leakagewindow functions they can also redistribute it to places where it does the least harm, depending on the application windowing predistorts the input samples so that the spectral leakage is evened out

For example

Frequency Bin Often the kth array element in the frequency domain is referred to as a frequency bin

The 400 Hz spectrum we expect the maximum amplitude to occur at. k=102, and 103<- energy spreads around two bins

Spectral Leakage When a single freq leaks into more than one bucket in the fft. This occurs because of rectangular windowing function use in a STFT It smearing the PSD across multiple buckets.

To limit spectral leakage Use a non-rectangular windowing function A function that is limited in time and zero everywhere else. Finite support. An a-periodic function

Select a windowing function Use either speed Or accuracy Or ease of implementation Do a straight multiply of the window and your input signal. You may need to do this several times, if you don’t want to loose data…

But I need speed! If you don’t overlap, you will loose data, but you will gain speed. You need to know the nature of the input signal!

There are many windows Hanning Blackman Bartlet Hamming Etc.

Windows

Code for Hanning public static double[] makeHanning(int n) { double window[] = new double[n]; double arg = 2.0 * Math.PI / (n - 1); for (int i = 0; i < n; i++) window[i] = * Math.cos(arg * i); return window; }

Lyon Window

Lyon Window Code public static double[] makeLyon(int n) { double window[] = new double[n]; double u = 0.0; double du = 2.0 / n; for (int i = 0; i < n / 2; i++) { window[i] = y5(0, 1.0, u); u += du; } u = 0; for (int i = n / 2; i < n; i++) { window[i] = y5(1.0, 0.0, u); u += du; } return window; }

y5 public static double y5(double y0, double y1, double u) { double t2 = u * u; double t3 = t2 * t2; return (6 * y1 - 6 * y0) * t3 * u + (-15 * y * y0) * t3 + (10 * y * y0) * t2 * u + y0; }

Lyon vs Hanning Window