8/28/2012 1 Lecture 01: Introduction Instructor: Dr. Gleb V. Tcheslavski Contact: gleb@lamar.edu Office Hours: Room 2030 Class web site: http://www.ee.lamar.edu/ gleb/stat/Index.htm ELEN 5301 Stochastic Signals and Systems Fall 2008 2 Syllabus overview • Prerequisites: Signals and Systems, algebra, calculus • Meetings: MW 1:25-2:40; Room 2631C • Required book: A. Papoulis and S. Unnikrishna Pillai. Probability, random variables, and stochastic processes. McGraw-Hill, New York, 2002, 4th Edition; ISBN-13: 9780071199810. • Tests: The midterm and the final exams will be closed book/notes; frequent pop-up quizzes. • Homework/Projects: up to ten homework may be assigned. Solutions may be posted one week after the assignment. • Attendance policy: Class attendance is mandatory with exceptions for medical and family emergencies. • Honor System: Discussions on lecture subject material, to clarify your understanding, are highly encouraged. However, it is your personal/own understanding only that should be reflected in all work that you turn in. You may thus claim credit only for your own work. All graded work is covered by the Academic Honor Code; violations will be prosecuted. ELEN 5301 Stochastic Signals and Systems Fall 2008 1 8/28/2012 3 Styles, notations, legends… 1. Colors: Normal text and formulas Something more important (imho) Important formulas and results Very Important Formulas Miscellaneous Miscellaneous 2. Equations notations: (2.17.3) Lecture # Slide # Formula # Slides: all class material (handouts, assignments, possibly solutions) will be available, in form of pdf slides, through the course web site: http://www.ee.lamar.edu/gleb/stat/Index.htm ELEN 5301 Stochastic Signals and Systems Fall 2008 4 Introduction Processes (events) 1. Deterministic: One possible value for any time instance. Therefore, we can predict the exact value of the signal for a desired time instance. Barely exist in real world… ELEN 5301 Stochastic Signals and Systems 2. Random (stochastic): Many (infinitely) possible values for any time instance. Therefore, we can predict only the expected value of the signal for a desired time instance. Stock market, weather, gambling outcomes, most of real life signals such as speech, medical data, digital images, communication signals,… Fall 2008 2 8/28/2012 5 Definitions and preliminary considerations The probability theory calls every action or occurrence an event (me standing in front of you, raining outside, student “John Smyth” getting an “F” in this class). Every event either may or may not occur. The purpose of the theory is to predict events in terms of their probability. The probability of an event A is a number P{A} assigned to this event. We can interpret this number as follows: If the experiment is performed n times and the event A occurs nA times, then, with a high degree of certainty, the relative frequency nA/n of the occurrence of A is close to P{A}: (1.5.1) provided that n is sufficiently large. Different events have different probabilities: the probability of rain in June in Beaumont is much higher that the probability of hail, which, in turn, is much higher than the probability of snow. Probability indicates how frequently the event may occur. ELEN 5301 Stochastic Signals and Systems Fall 2008 6 Introduction This cone shows the most likely (at the time of its generation) path of the tropical storm Isaac… From http://www.nhc .noaa.gov/ Assessed on 8/27/2012 at noon ELEN 5301 Stochastic Signals and Systems Fall 2008 3 8/28/2012 7 Introduction In the applications of probability to real problems, three steps must be distinguished: 1. Observation (physical): we assign the probabilities P{Ai} of specific events Ai by an exact process (for instance, observation) according to (1.5.1). If a fair die is rolled 600 times and “5” is observed 100 times, then we assign P{“5”} = 100/600 = 1/6. 2. Deduction (conceptual): we assume that probability satisfies certain axioms, and by deductive reasoning we determine from the probabilities P{Ai} of specific events Ai the probabilities P{Bi} of other events Bi. In the game with a fair die, we deduce that the probability of “even” is 3/6 with the following reasoning: If P{“2”} = P{“4”} = P{“6”} = 1/6, then P{“even”} = 1/6 + 1/6 + 1/6 = 3/6. 3. Prediction (physical): we make a physical prediction based on the numbers P{Bi} we obtained according to (1.5.1). If we roll a fair die 500 times, our prediction is that “even” will show about 250 times. We must make a clear distinction between the data we have determined empirically and the results we have deduced logically. ELEN 5301 Stochastic Signals and Systems Fall 2008 8 Introduction Steps 1 and 3 are based on inductive reasoning. Say, for instance, we wish to determine the probability of “heads” of a given coin. Should we toss it 100 times or 1000 times? If we tossed it 1000 times and the number of “heads” is 493, what can we predict based on this observation? Can we deduce that at next 1000 tosses the number of “heads” will be 493? We can only answer such questions inductively. We will consider mainly step 2: from certain probabilities we derive deductively other probabilities. Moreover, we will interpret the probability as a characteristic of an event (similarly to the mass or resistance) and will not be concerned about the “physical meaning” of the probability in the theory development. For example, a resistor is commonly viewed as a 2-terminal device satisfying: (1.7.1) However, this is only a convenient abstraction: a real resistor is a complex device with distributed inductance and capacitance… Therefore, (1.7.1) can be claimed only with certain errors, in certain frequency ranges and with other assumptions. Nevertheless, these uncertainties are usually ignored and we are not concerned with the true meaning of R. ELEN 5301 Stochastic Signals and Systems Fall 2008 4