Introduction to Digital Signal Processing (Discrete-time Signal Processing)

Transcription

1 Introduction to Digital Signal Processing (Discrete-time Signal Processing) Prof. Chu-Song Chen Research Center for Info. Tech. Innovation, Academia Sinica, Taiwan Dept. CSIE & GINM National Taiwan University Fall 2013

2 In our technical society we often measure a continuously varying (analog) quantity. eg. Blood pressure, earthquake displacement, population of a city, waves falling on a beach, and the probability of death. All these measurement varying with time; we regard them as functions of time: x(t) in mathematical notation.

3 Signals flow of information measured quantity that varies with time (or position) electrical signal received from a transducer (microphone, thermometer, accelerometer, antenna, etc.) electrical signal that controls a process

4 For technical reasons, instead of the signal x(t), we usually record equally spaced samples x n of the function x(t). (discrete-time) The sampling theorem gives the conditions on the signal that justify this sampling process. i.e., discrete-time signal is a sequence of numbers Moreover, when the samples are taken they are not recorded with infinite precision but are rounded off (sometimes chopped off) to comparatively few digits. This procedure is often called quantizing the samples. (digital)

5 Discrete-time signal sequences can often arise from periodic sampling of an analog signal x a (t) x xa ( nt ), - n

6

7 Signal Source where it comes Continuous-time signals: voltage, current, temperature, speed,... Discrete-time signals: daily minimum/maximum temperature, daily stock price, sampled continuous signals,... Electronics can only deal easily with timedependent signals; therefore spatial signals, such as images, are typically first converted into a time signal with a scanning process (TV, fax, etc.).

8 Type of signals

9 Example: Acoustic Signal (analog)

10 Example: Number of Sunspots (discrete-time)

11 Example: Image Signal (discrete-time)

12 The concept of System The set consisting of all signals (of a type) forms a functional space. Signal Processing System: map an input signal to an output signal Continuous-time systems Systems for which both input and output are continuous-time signals Digital system Both input and output are digital signals x[n] T{} y[n] System: a function (or mapping) whose input and output are both functions.

13

14

15

16 Example: Microphone and Speaker Analog system Discrete-time system

17

18

19

20

21 Course Outline Basic topics Z-transform Discrete-time Fourier transform (DTFT) Sample of continuous-time signals Discrete-time linear systems & its transform domain analysis Structure for discrete-time systems Digital filter Discrete Fourier transform (DFT) Fast computation of discrete Fourier transform Fourier analysis of signals using DFT Random signals and systems Wavelets

22 Reference Textbooks James H. McClellan, Ronald W. Schafer, and Mark A. Yoader, Signal Processing First Alan V. Oppenheim and Ronald W. Schafer, Discrete- Time Signal Processing, Prentice-Hall. Boaz Porat, A Course in Digital Signal Processing Dimitris G. Manolakis and Vinay K. Ingle, Applied Digital Signal Processing Main Journals IEEE Trans. Signal Processing IEEE Signal Processing Magazine Main Conferences IEEE International Conference on ASSP (ICASSP)

23 Course Information Teaching assistant: Yin-Tzu Lin 林映孜 Course webpage: Grades Homework x several (30%) Test x 2~3 (40~45%) Term project (25~30%)

24 Frequency and Sinusoids Signal processing is originated form the processing of frequency. Understanding the frequency: better from sinusoidal functions. Important to the field of broadcasting, wireless communication, music analysis, etc.

25 Sinusoids Cosine functions x(t)=a cos(2πf 0 t+φ)

26 Sinusoids Cosine functions x(t)=a cos(2πf 0 t+φ) Where x(t) is a function of the time variable t. The amplitude of the cosine is given by the real number A. The frequency of the of the cosine wave is f 0, The phase of the sinusoid is given by the parameter φ.

27 Sound of a clay whistle Example: CLAY whistle

28 CLAY whistle BLUE-BIRD: 1000 to 1009 milliseconds In this case, the audio file is the analog-to-digitalconverted samples of the microphone output.

29 Frequency In the audio experiments that follow, it is the frequency that determines what we hear. Eg., the waveform of a tuning fork Amplitude determines the loudness; Frequency determines the sound we heard

30 Tuning Fork Signal: first 14 milliseconds

31 Frequency and period of sinusoids Cosine functions x(t)=a cos(2πf 0 t+φ) It can also be written as x(t)=a cos(w 0 t+φ) where w 0 = 2πf 0 f 0 : Hertz (cycles/sec); w 0 :(Radians/sec) T 0 = 1/f 0 is the period. It can be verified that x(t+nt 0 ) = x(t) for all integers n. So, T 0 = 2π/w 0

32 Example

33 Example What is phase? time shift vs. phase

34 Time Shift & Phase

35 Range of phase So, we usually consider the phase only in a range [-, ] or [0,2]

36 Other sinusoids Sine function can be represented as a phase shift of -/2 of the cosine function: sin() = cos( - /2) Cosine function can also be represented as phase shift of -/2 of the sine function cos() = sin( + /2)

37 Basic Properties of Sine and Cosine

38 Some basic trigonometric identities Sine and cosine are derivatives to each other

39 Addition of Sinusoids of The Same Frequency

40 Here, we can see that sine waves form a set of orthogonal bases that could be used for signal representation.

41 Basis Functions for Decompositions As mentioned, signal processing is originated form the processing of frequency. We hope to decompose the signals by extracting its components with respect to different frequencies. In other words, we hope to represent a signal as a linear combination of bases functions. What are the proper basis functions for frequency decomposition? Sine or Cosine?

42 Bases functions in signal processing In signal processing, instead of using either sine or cosine, we use the complex exponential for bases representation. Compared to sine and cosine, complex exponential has several advantages for signal representations. Reasons will be made clearer later. complex exponentials: where e j w j w e 0 t 0t cos( w 0t ) jsin( w 0t ) W 0 : frequency of the complex exponential; : phase.

43 Review of complex numbers

44

45

46

47

48

49 Explanation of the complex exponential e jwt Imagery (Vertical) axis Time axis Real (Horizontal) axis

50 Projection of e jwt Real and Imagery axes Imagery (Vertical) axis Z Real (Horizontal) axis, X

51 Analogous to electromagnetic wave

52

53 Unifying operation by multiplication By using complex exponential, amplitude and phase changes can both be represented by multiplications: j w t jw t j Ae 0 ( Ae ) e 0 Amplitude and Phase: by multiplication complex number multiplication can represent both scaling (amplitude variation) and rotation (phase shift)

54 Further advantage of complex exponential Geometric Sequence Evaluation Let x be a complex number. Geometric series is used quite often to simplify expressions in DSP. if the magnitude of x is less than one, then x x x x x x N N n N n , x x x n n

55 Note that trigonometric functions, especially sine and cosine functions, appear in different combinations in all kinds of harmonic analysis: Fourier series, Fourier transforms, etc. Advantages of complex exponential The identities that give sine and cosine functions in terms of exponentials are important because they allow us to find sums of sines and cosines using the geometric series. Eg. we know N 1 n0 sin 2n N 0 N 1 n0 cos 2n N 0 ie. a sum of equally spaced samples of any sine or cosine function within 2 is zero, provided the sum is over a cycle (or a number of cycles), of the function.

56 They can be easily verified by using the geometric sequence of complex exponentials: N n 1 0 e 2n j N 1 e 1 e j2 2n j N 0