37 Pages

# HKUST ELEC210 Probability and Random Process

Course Number: ELEC ELEC210, Spring 2012

College/University: HKUST

Word Count: 2077

Rating:

###### Document Preview

ELEC 2600: Probability and Random Processes in Engineering Spring 2012 Instructor: Prof. Jun ZHANG Office: Rm 2448 (Lift 25-26) Tel: 2358 7050 Email: eejzhang@ust.hk Webpage: http://www.ece.ust.hk/~eejzhang/ 1 Elec 2600: Lecture 1 • Course Details • Models in Engineering  Their role  Types of models • • • • Relative Frequency Packet Voice Transmission Example Axiomatic Approach to Probability An Interesting...

##### Unformatted Document Excerpt
Coursehero >> China >> HKUST >> ELEC ELEC210

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

2600: Probability ELEC and Random Processes in Engineering Spring 2012 Instructor: Prof. Jun ZHANG Office: Rm 2448 (Lift 25-26) Tel: 2358 7050 Email: eejzhang@ust.hk Webpage: http://www.ece.ust.hk/~eejzhang/ 1 Elec 2600: Lecture 1 • Course Details • Models in Engineering  Their role  Types of models • • • • Relative Frequency Packet Voice Transmission Example Axiomatic Approach to Probability An Interesting Problem – Winning an iPad Elec2600 Lecture 1 2 Course Details • Weekly schedule • 2 lectures per week (Wed and Fri 4:30pm – 5:50pm, Rm 3008, Lift 3-4) • 2 tutorials per week (Each student only needs to attend one) • Wed 12:00pm – 12:50pm, Rm 4504, Lift 25-26 • Tue 6:00pm – 6:50pm, Rm 4504, Lift 25-26 • Grading: • Homework (20%) • Mid-Term (30%) + Final (50%) • There will be 4 homework assignments • Each worth 5% • Details of homework submission procedures and late penalties on website (see next slide) Elec2600 Lecture 1 3 Course Details • Website • http://course.ee.ust.hk/elec2600/ • Textbook: • Probability, Statistics and Random Processes for Electrical Engineering, Alberto Leon-Garcia, Addison Wesley, 3rd ed., 2009. Elec2600 Lecture 1 4 Course Details • Course Structure • Part I: Basic probability theory. (~2-3 weeks) • Part II: Single random variables. (~3-4 weeks) • Part III: Multiple random variables. (~5 weeks) • Part IV: Stochastic processes. (~3 weeks) • Objective of the course • Basic concepts of probability theory required to understand probability models used in engineering. • Basic techniques required to develop probability models. Elec2600 Lecture 1 Course Details • Instructional Assistant • ZHANG Xuning, eexuning@ust.hk • Teaching Assistants:  CHEN Shibo, schenab@ust.hk  SHI Yuanming, yshiac@ust.hk  LI Tian, tli@ust.hk  PANG Jiahao, jpang@ust.hk • Office location and consultation hours will be posted on website • TAs are there to help you – PLEASE USE THEM! Elec2600 Lecture 1 6 Elec 2600: Lecture 1 • Course details • Models in Engineering  Their role  Types of models • • • • Relative Frequency Packet Voice Transmission Example Axiomatic Approach to Probability An Interesting Problem – Winning an iPad Elec2600 Lecture 1 7 Modelling • • • • In engineering as well as in daily life, we are faced with choices Our decisions are based on our belief – our model – of how things will behave A model is an approximate representation of a physical situation which attempts to explain observed behavior Good models  are simple and understandable  predict all relevant aspects of the situation • Models can help us avoid costly (in time and money) experimentation Elec2600 Lecture 1 8 Mathematical and Computer Simulation Models • Mathematical model • A set of assumptions about how a system or physical process works. • Assumptions are stated in the form of mathematical relations between the important parameters and variables. • By solving these relations, we can predict the outcome of an experiment, e.g. V = I*R • The simpler the model, the easier it is to solve, understand and analyze. • However, simple models may not accurately describe the actual system of interest. • Computer simulation model • Assumptions are stated in the form of a computer program. • By running the program, we can predict the outcome of an experiment, e.g. Spice. • Computer models can represent systems in greater detail, and more accurately predict performance, but are generally more difficult to analyze precisely. Elec2600 Lecture 1 9 The Modelling Process • Models for components of a complex system • • Validation through observation of the physical system • • Hypothesis formulation Modification if necessary E.g., digital camera: • Acquisition (sensors) • Digitization (quantization noise models) Elec2600 Lecture 1 Curve fitting, statistical modelling, etc 10 Deterministic versus Probabilistic Models • Deterministic Model • • • • The conditions of the experiment determine the exact outcome If I do A, then B will happen. Example: Ohm’s Law (V = I*R), Kirchoff’s current and voltage laws. In practice, there will always be random deviations if you repeat the same experiment twice, but a deterministic model is OK if these are small. • Probabilistic Model • The conditions of the experiment determine the probability of different outcomes. • If I do A, then B will probably happen, but it is possible that C might happen instead. • Example: tossing of a coin, rolling a die • In many cases, probabilistic models are used because we cannot model all the relevant variables required by a deterministic model. Elec2600 Lecture 1 11 More on Probabilistic Models • • • • Allow designers to model very complex systems Nearly all practical systems cannot be described exactly, and have some sort of randomness Systems can be designed and optimized “off-line”, without physical implementation – save time and \$\$\$ Each complex system may involve MANY models! Elec2600 Lecture 1 12 Other Examples Elec2600 Lecture 1 13 Elec 2600: Lecture 1 • Course details • Models in Engineering • Their role • Types of models • • • • Relative Frequency Packet Voice Transmission Example Axiomatic Approach to Probability An Interesting Problem – Winning an iPad Elec2600 Lecture 1 14 Example: Ball Picking from an Urn • • • An urn contains 3 balls, labeled 1, 2 or 3. A ball is selected, its number recorded (outcome) and the ball returned. This process is repeated for n (e.g. 100) times (trials). • RANDOM Experiment: Outcome cannot be exactly determined! 5 4 3 2 1 0 -1 Elec2600 Lecture 1 15 Relative Frequency • Let N1(n), N2(n) and N3(n) be the number of times that we pick balls 1, 2, and 3 in n trials (events). N n  • Define the relative frequency of the outcome k as f k n   k n • This experiment exhibits statistical regularity: as n increases, the relative frequency approaches a constant value lim f k n   p k n  where pk indicates the probability of outcome k. Elec2600 Lecture 1 Provides a key connection between measurement of physical quantities and probability models! 16 Properties of the relative frequency (1) • Since 0 ≤ Nk(n) ≤ n, N k n  0  f k n   1 n The relative frequencies are between zero and one. K • Let K = the number of possible outcomes. Since K  k 1 f k n   K  k 1 N k n  1 n  k 1 N k n   n The relative frequencies sum to one. Elec2600 Lecture 1 17 Properties of the relative frequency (2) • Events are groups of the outcomes of an experiment (sets): • ONE = “the ball picked is labeled 1” = {1} • NOT_THREE = “the ball picked is not labeled 3” = {1 or 2} • ODD = “the ball picked is labelled with an odd number” = {1 or 3} • We can often derive the relative frequency of one event from the relative frequency of other events. Example: Since NODD(n) = N1(n) + N3(n), f ODD n   N ODD n  N 1 n   N 3 n    f 1 n   f 3 n  n n • More generally, if C = {A or B} and A and B cannot occur simultaneously exclusive), f (mutually C n   f A n   f B n  Elec2600 Lecture 1 Above 3 properties coincide with AXIOMS of probability (discussed next lecture) 18 Elec 2600: Lecture 1 • Course Details • Models in Engineering  Their role  Types of models • • • • Relative Frequency Packet Voice Transmission Example Axiomatic Approach to Probability An Interesting Problem – Winning an iPad Elec2600 Lecture 1 19 Packet Voice Transmission Example • • • • Suppose a communication system needs to transmit N = 48 simultaneous conversations using “packets” corresponding to 10ms of speech. Suppose each person speaks only 1/3 of the time. If we wish to transmit at most M packets every 10 ms, how should we choose M? Obviously, M = 48 guarantees that no packets are ever lost. However, this is expensive in terms of equipment/bandwidth. In addition, most (~2/3) of the packets will be empty. Elec2600 Lecture 1  On the other hand, if M < 48, some packets may get lost.  How should we choose M so that on average only a small fraction (1%) of the packets are lost? 20 Example: Performance Metric • We obtain good performance when the ratio below is small. discard fraction = average # of active packets lost average # of active packets where the average is over a large number (n) of intervals. • Let A(j) be the number of active packets (i.e. speakers that speak) during the jth 10ms time interval. 48 A( j ) 32 16 0 0 Elec2600 Lecture 1 20 40 j 60 80 100 (n  100) 21 Example: Average number of active packets • Define Nk(n) to be the number of intervals with k active packets. 48 15 N k (100) A( j ) 32 10 k  14 16 0 0 5 50 j 100 (n  100) 0 0 16 32 k 48 • The average number of active packets can be computed in two different ways: 1n 1 48 average # of active packets =  A( j )   k  N k (n) n j 1 n k 0 Elec2600 Lecture 1 total number of active packets 22 Example: Lost packets • Define B(j) to be the number of packets lost in interval j. • The number of lost packets is a function of the number of active packets A(j). • If A(j) ≤ M, then all packets can be transmitted and B(j) = 0. • Otherwise, B(j) = A(j) – M. • The maximum number of lost packets is 48 – M. 30 20 A(j) M = 18 10 B(j) 0 0 Elec2600 Lecture 1 20 40 60 80 j 100 23 Example: Average number of lost packets • Define Li(n) to be the number of times i packets are lost. • As before, there are two ways to compute the average number of lost packets 1n 1 48 M average # of active packets lost =  B( j )   i  Li (n) n j 1 n i 1 • We can also compute the average number of lost packets based on the number of active packets. • If the number of active packets is k and k > M, then the number of lost packets is i = k - M. Thus, 1 48 M # packets lost average # of active packets lost   i  Li (n) n i 1 # times k packets appear 48 1   ( k  M )  N k ( n) n k  M 1 Elec2600 Lecture 1 24 Example: Express as relative frequency • Summarizing: 1 48 1 (k  M ) N k (n) average # of active packets lost n k  M discard fraction   1 48 average # of active packets  k  N k ( n) n k 0 • Letting n → ∞, 48 48 N k ( n) 1 (k  M )  n 1 (k  M )  pk  k  M 48  discard fraction  k  M 48 n  N ( n) k k n  k  pk k 0 k 0 0.2 • Later in the course, we will learn how to calculate the pk. 0.15 0.1 0.05 0 0 Elec2600 Lecture 1 Nk(100)/100 pk 16 32 k 48 25 Example (4) • Given the pk, we can calculate the discard fraction • Even if we transmit at most M=24 packets, on average less than 1% of the active packets will be lost. Elec2600 Lecture 1 26 More Examples • Casino strategy: bet \$1, \$2 if losing the 1st one, \$4 if losing the 2nd one, and so on. • What is the winning/losing probability? • Communications over unreliable channels: bit “0” and bit “1” may be interpreted incorrectly with an error probability • Prediction of signals: use available signal portion to predict for the future, or to analyze and synthesize signals (e.g., LPC). Elec2600 Lecture 1 27 More Examples • System reliability: failure probability, average time to fail • Resource-sharing systems: banks, telephone, etc. • Internet systems: simple client-server, peer-to-peer • Many, many more... Elec2600 Lecture 1 28 Elec 2600: Lecture 1 • Course Details • Models in Engineering • Their role • Types of models • • • • Relative Frequency Packet Voice Transmission Example Axiomatic Approach to Probability An Interesting Problem – Winning an iPad Elec2600 Lecture 1 29 Axiomatic Approach to Probability • Efforts to build a mathematical theory based on the idea of the limit of the relative frequency run into difficulty because • • It is not clear how to define the limit mathematically. The limit cannot be found in practice. • The axiomatic approach does not define probability as a limit, but rather as any way of assigning numbers to events such that they behave like the relative frequency. • Note that the theory does not assign any meaning to the probability, or specify how they are obtained. It only specifies the properties it must satisfy in order for the theory to hold. • It is up to the designer (model builder) to determine how to assign the probability in a way that is useful. Elec2600 Lecture 1 30 Assumptions of the Axiomatic Approach • The set S of all outcomes has been identified. • e.g. the number of active packets is in the set S = {0,1,2,...,48} • A class of subsets of S called events has been specified. • e.g. the event that number of active packets is 15 • Each event A has been assigned a number pA that satisfies • 0 ≤ pA ≤ 1 • pS = 1 • If A and B are events that cannot occur simultaneously, then pA or B = pA + pB • Note the similarity with the properties of the relative frequency covered earlier. Elec2600 Lecture 1 31 Elec 2600: Lecture 1 • Course Details • Models in Engineering  Their role  Types of models • • • • Relative Frequency Packet Voice Transmission Example Axiomatic Approach to Probability An Interesting Problem – Winning an iPad Elec2600 Lecture 1 32 Something to think about (i): • Suppose exactly 1 out of 3 boxes contains an iPad Elec2600 Lecture 1 33 Something to think about (ii): • You pick one randomly Elec2600 Lecture 1 34 Something to think about (iii): • I take away one of the boxes which you did NOT select, and I tell you that one of the remaining boxes contains the iPad Elec2600 Lecture 1 35 Something to think about (iv): • I take away one of the boxes which you did NOT select, and I tell you that one of the remaining boxes contains the iPad Will you change your choice? Why / Why not? Elec2600 Lecture 1 36 Summary • This lecture we have discussed:   The concept of relative frequency  Practical example: Packet voice transmission  Introduction to probability axioms  37 The role and types of probability models in engineering  • Overall course details Example: Winning an iPad Next Lecture: Random Experiments and Probability Axioms (more)!

Find millions of documents on Course Hero - Study Guides, Lecture Notes, Reference Materials, Practice Exams and more. Course Hero has millions of course specific materials providing students with the best way to expand their education.

Below is a small sample set of documents:

University of Alabama - Huntsville - ME - 202
Wall Temperature DynamicsE X A M P L E 7.8.6■ ProblemConsider the wall shown in cross section in Figure 7.7.5 and treated in Example 7.7.1. In thatexample the thermal capacitances of the layers were neglected. We now want to develop a model
Oregon State - ECE - 576
er.Next header specifies the type of extension header used. The functionalityof the option field in IPv4 is specified in the extension header. In addition,the extension header is more flexible than the options field.Hop limit is the same as the t
Oklahoma State - HIST - 1103
caWhen—1450-1850Significance—allowed for repopulation of the Americas after decline of Native Americanpopulation, integrated African, European &amp; American societies5) Bacon’s Rebellion—Who—Nathanael Bacon, Governor Berkley, Indians, for
Maryland - BMGT - 370
d consolidation opportunitiesParticipation in the EPA’s “Smart Way TransportPartnership”Many opportunities for greener supply chainexist–Subway restaurants provide one example12Collaboration and Visibility:Art and Science•V
HKUST - ELEC - ELEC210
ELEC 2600 Lecture 2: Build a Probability Model Specifying Random Experiments Sample spaces and events Set Operations The Three Axioms of Probability Corollaries Probability Laws for Assigning ProbabilitiesElec2600 Lecture 21Specifying Random Expe
HKUST - ELEC - ELEC210
Elec 210: Lecture 3 Computing Probabilities using Counting Methods Sample Size Computation and Examples Probabilities and Poker!Elec2600 Lecture 31Computing Probabilities Using Counting Methods In experiments where the outcomes are equiprobable, we
HKUST - ELEC - ELEC210
Elec 2600: Lecture 4 Conditional Probability PropertiesThus far, we have looked at the probabilityof events occurring individually, withoutregards to any other event.However, if we KNOW that a particularevent occurs, then how does this change thep
HKUST - ELEC - ELEC210
Elec 2600: Lecture 5 Sequential Experiments Bernoulli trials Binomial probability law *Multinomial probability law Geometric probability law Sequences of dependent experiments Example: Bean Machine Game!Elec2600 Lecture 51Sequential Experiments
HKUST - ELEC - ELEC210
Elec 2600: Lecture 6 Random Variables Equivalent events Discrete Random Variables Probability mass functionElec2600 Lecture 61Random Variables A random variableX is a function that assigns a number to everyoutcome of an experiment. The function
HKUST - ELEC - ELEC210
Elec 2600: Lecture 7 Expectation of a random variable Expected value of a function of a randomvariable Variance of a random variable Moments of a random variableThe human brain weighs1500 grams, on average.Elec2600 Lecture 71Interesting Fact Ein
HKUST - ELEC - ELEC210
Elec 210: Lecture 8 Conditional Probability Mass Function Conditional Expected Value Conditional Poker HandElec210 Lecture 81Conditional Probability Mass Function The effect of partial information about the outcome of a randomexperiment on the pro
HKUST - ELEC - ELEC210
Elec 2600: Lecture 9 Important discrete random variables Summary of variables you know:BernoulliBinomialGeometricDiscrete Uniform New random variable: Poisson MATLAB commands for plotting probability mass functionsand generating discrete random v
HKUST - ELEC - ELEC210
Elec 2600: Lecture 10 Single random variables: discrete, continuous and mixedContinuous R.V. and Cumulative Distribution Function (CDF)Probability Density Function (PDF)Conditional CDFs and PDFsElec2600 Lecture 101Random Variables: ReviewA random
HKUST - ELEC - ELEC210
Elec2600 Lecture 11 Expectation of Continuous Random Variables Variance of Continuous Random Variables Important Continuous Random VariablesElec2600 Lecture 111Review: Expectation Interpretation The average value of a random variable if we repeat
HKUST - ELEC - ELEC210
Elec2600 Lecture 12 MATLAB functions for continuous random variables Functions (Transformations) of a Random VariableElec2600 Lecture 121MATLAB Functions for Continuous RVs Generating m by n arrays of random samplesunifrnd(a,b,m,n)exprnd(1/lambda,
HKUST - ELEC - ELEC210
Elec2600: Lecture 13 Multiple random variables (RVs) Joint probability mass function of two discrete RVs Marginal probability mass functionElec2600 Lecture 131Vector Random VariablesA vector random variable X is a function that assigns a vectorof
HKUST - ELEC - ELEC210
Elec2600: Lecture 14 Pairs of continuous random variables Review of 2D functions, differentiation and integrationJoint cumulative distribution functionJoint density functionElec2600 Lecture 141Two Random Variables One random variable can be consid
HKUST - ELEC - ELEC210
Elec2600: Lecture 15 Independence Expectation of Function of 2 Variables Joint MomentsElec2600 Lecture 151Independence Definition: Two random variables X and Y are said to beindependent or statistically independent if for any events, AXand AY, de
HKUST - ELEC - ELEC210
Elec2600: Lecture 16 Conditional Probability Conditional ExpectationElec2600 Lecture 161Conditional Probability Mass Functions Suppose that X and Y are discrete RVs assuming integer values. The conditional pmf of Y given X ispY | X (k | j ) P Y k
HKUST - ELEC - ELEC210
Elec2600: Lecture 17One function of two random variablesDiscrete random variablesContinuous random variablesUsing conditioningThus far, for Z = g(X,Y), with X and Yrandom variables, we know how tocompute the moments of Z.But how do we compute the
HKUST - ELEC - ELEC210
Elec2600: Lecture 18 Random Vectors Joint distribution/density/mass functions Marginal statistics Conditional densities Independence and Expectationhttp:/www.cs.princeton.edu/~cdecoro/eigenfaces/Elec2600 Lecture 181N Random Variables An N dimens
HKUST - ELEC - ELEC210
Elec2600: Lecture 19 Single Gaussian Random Variable Gaussian Random VectorsElec2600 Lecture 191Gaussian Random Variable The Gaussian random variable isused to model variables that tendto occur around a certain value, m,called the mean. This ran
HKUST - ELEC - ELEC210
Elec2600: Lecture 20 Sums of Random Variables Mean and Variance of Sample Means Useful Inequalities Laws of Large NumbersElec2600 Lecture 201Sums of Random Variables For any set of random variables, X 1 , X 2 ,., X nEVAR jX j Xjj 1n E[ Xj
HKUST - ELEC - ELEC210
Elec210: Lecture 21 Central Limit Theorem The PDF of sums of Random Variables The characteristic function Proof of the Central Limit TheoremElec210 Lecture 21http:/www.mathsisfun.com/data/quincunx.html1Central Limit TheoremAny distributionSuppos
HKUST - ELEC - ELEC210
Elec2600: Lecture 22 Definition of a Random Process Specification of a Random ProcessElec2600 Lecture 221Definition of a Random ProcessDefinition: A random process orstochastic process maps aprobability space S to a set offunctions, X(t,)It assi
HKUST - ELEC - ELEC210
Elec2600: Lecture 23 Mean and Variance Correlation and Covariance Functions Multiple Random ProcessesElec2600 Lecture 231Mean and Variance FunctionsMeanmX (t ) E[ X (t )] xf X ( t ) ( x) dx Variance X (t ) mX (t ) 2 ( x mX (t ) 2 f X (t ) ( x)dx
HKUST - ELEC - ELEC210
Elec2600: Lecture 24 Discrete Time Random Processes Sum Processes ISI ProcessesElec2600 Lecture 241Sum Random Processes Definition: A sum process Sn is obtained by taking the sum ofall past values of an i.i.d. random process Xn, i.e.,nS n X i X
HKUST - ELEC - ELEC210
Elec210: Lecture 25Continuous Time I.S.I. Random ProcessesPoisson Random ProcessAdditional Random Processes (FYI) Random Telegraph Process Shot Noise Process Weiner ProcessElec210 Lecture 251The Poisson ProcessConsider the following sequence of
HKUST - ELEC - ELEC210
Elec2600: Lecture 26 Stationary Random Processes Wide Sense Stationary (WSS) Random ProcessesElec2600 Lecture 261Stationary Random ProcessesDefinition: A process is stationary if the joint distribution of any set of samplesdoes not depend on the pl
HKUST - ELEC - 546
Diversity and Convolutional Code onWireless Communication Channels(Ref: Proakis, p. 470-492)1/2/20091BPSK:-AA1/2/20092PeConcavefunction(linearscale)Avg Peg1PePe(Avg g) ismuchsmaller thanAvg PePe(Avg g)g2bit SNR (linear scale)g1/2
HKUST - ELEC - 546
Vincent LauSimplified Version 1.01/1/20091Transmitter with t antennas &amp; receiver with r antennas.Transmitted Symbol, received symbolChannel Matrix (Flat Fading)1/1/20092Similar to the scalar case, we have Capacity Capacity Capacity Capacity
HKUST - ELEC - 546
Vincent LauSimplified version1/1/20091From information theory, the MIMO capacityincreases linearly with min(r,t) for the sameP,W. How to achieve the theoretical capacity?1/1/20092There are two favors of Space Time Coding Spatial diversity (ST C
HKUST - ELEC - 546
Vincent Lau1/25/101Space Time codes (STBC &amp; STTC)Layered Space Time Coding requires ML decoding at the receiver complexity growsexponentially with number of bits per symbol. In general, full rank ST codes are used to achieve spatial diversity. A s
HKUST - ELEC - 546
MIMO Channel Model and CapacityELEC546-Ch.9, Spring 2009, HKUST1Flat Fading MIMO Channel I/O relationshipELEC546-Ch.9, Spring 2009, HKUST2Singular Value Decomposition (SVD)Applying SVD, we haveELEC546-Ch.9, Spring 2009, HKUST3ELEC546-Ch.9, Sprin
HKUST - ELEC - 546
1ELEC546-Ch.1, Spring 2010, HKUSTPre-requisite: MUST HAVE DIGITAL COMMUNICATIONS BACKGROUND (ELEC 314, ELEC343) What is M-ary modulation and the demodulation methods? What is bit rate, BER, BW, Power Cellular Systems Basic mathematics E.g. Convex
HKUST - ELEC - 546
ELEC546-Ch.1, Spring 2009, HKUST1General 3-level model Path Loss Model Shadowing Multipath Fading (Rapid Small-scale Variation)Path Loss Model Free-Space Propagation 3 Basic Propagation Mechanisms Practical Path Loss ModelsShadowingELEC546-Ch.1
HKUST - ELEC - 546
ELEC546-Ch.3, Spring 2009, HKUST1Physical cause: constructive and destructive interference betweenmultiple TX-RX paths with different lengths(diffphases)RecSignalPowerLocation (l)ELEC546-Ch.3, Spring 2009, HKUST2Effects: Small scale Fading At
HKUST - ELEC - 546
1/1/200911/1/20092Properties of Entropy: Lower Bound: Upper Bound: Discrete X: Continuous X: Chain Rule: Conditioning reduces Entropy: Concavity of entropy: If X &amp; Y are independent, then Fanos Inequality: Given two random variables X &amp; Y, l
HKUST - ELEC - 546
Orthogonal Frequency Division Multiplexing(OFDM): Basic OFDM System1Motivations Diagonalization of Vector Channels Transmission of one OFDM Symbol Transmission of sequence of OFDMSymbols Sample and Symbol Time Synchronization Carrier Frequency Sy
HKUST - ELEC - 546
Multi-user Capacity andOpportunistic CommunicationsELEC546-Ch.9, Spring 2009, HKUST1Uplink AWGN ChannelDownlink AWGN Channel Successive Interference Cancellation for UplinkCommunication Superposition Coding for Downlink Communication Dirty Paper
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [01]Introduction: CourseInformation and Policiescourse notes byDr. Enamul HossainFall 2012 20131General Information Instructor :NameOfficeE-mailTel.
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMEdited by Foxit ReaderCopyright(C) by Foxit Corporation,2005-2009For Evaluation Only.ENGINEERINGDepartment of Petroleum EngineeringPETE-401: RESERVOIR DESCRIPTIONLecture [03]course notes byDr. Enamul HossainFall 20121Review of last lectur
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMEdited by Foxit ReaderCopyright(C) by Foxit Corporation,2005-2009For Evaluation Only.ENGINEERINGDepartment of Petroleum EngineeringPETE305: RESERVOIR DESCRIPTIONLecture [04]course notes byDr. Enamul HossainFall 20121Example 2.1.2 KFUPM |
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [05]course notes byDr. Enamul HossainFall 20121Review of last lecture:Statistics of Univariate DistributionSpatial Data Sets Data sets do not represent r
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMSep.25,2012ENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [06]course notes byDr. Enamul HossainFall 20121Cumulative Distribution Function (CDF)F( x) Pr obX xRandom Variable (RV)F(x)Limit, x-axis1A
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGSep.25,2012Department of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [07]course notes byDr. Enamul HossainFall 20121AnnouncementMajor Exam IonSeptember 29, 2012 @ 12:00 Noon (Saturday)Venue: Building 3; Room 1
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONSep.28,2012Lecture [08]course notes byDr. Enamul HossainFall 20121Review of last lecture: KFUPM |Fall 2012| PETE-305: Reservoir Description | Dr. Enamul Hossain2
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMEdited by Foxit ReaderCopyright(C) by Foxit Corporation,2005-2009For Evaluation Only.ENGINEERINGDepartment of Petroleum EngineeringSep.28,2012PETE-305: RESERVOIR DESCRIPTIONLecture [09 and 10]course notes byDr. Enamul HossainFall 20121Re
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMEdited by Foxit ReaderCopyright(C) by Foxit Corporation,2005-2009For Evaluation Only.ENGINEERINGSep.28,2012Department of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [11]Spatial Relationship: Estimation &amp;Modeling (Chapter 3)
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMEdited by Foxit ReaderCopyright(C) by Foxit Corporation,2005-2009For Evaluation Only.Nov.11,2012ENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [12]Spatial Relationship: Estimation &amp;Modelingcourse notes
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMEdited by Foxit ReaderCopyright(C) by Foxit Corporation,2005-2009For Evaluation Only.Oct.11,2012ENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [13]Spatial Relationship: Estimation &amp;Modelingcourse notes
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGNov.3,2012Department of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [14]Modeling Variogram 1course notes byDr. Enamul HossainFall 211213.5 Modeling Variogram We have learned in Lecture 6 various methods toestim
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMEdited by Foxit ReaderCopyright(C) by Foxit Corporation,2005-2009For Evaluation Only.ENGINEERINGDepartment of Petroleum EngineeringNov.3,2012PETE-305: RESERVOIR DESCRIPTIONLecture [15]Modeling Variogram 2course notes byDr. Enamul HossainF
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGNov.4,2012Department of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [16]Conventional Estimation Technique 1Chapter 4course notes byDr. Enamul HossainFall 201214.2 Linear Kriging ProceduresSimple krigingOrdinar
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMNov.4,2012ENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [17]Conventional Estimation Technique 2course notes byDr. Enamul HossainFall 20121Simple Kriging: Example 2P109 KFUPM| PETE-305: Reservoir Des
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGDepartment of Petroleum EngineeringNov.4,2012PETE-305: RESERVOIR DESCRIPTIONLecture [18]Conventional Estimation Technique 3course notes byDr. Enamul HossainFall 201214.2.2 Ordinary Kriging KFUPM| PETE-305: Reservoir Description
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [19]Conventional Estimation Technique 4course notes byDr. Enamul HossainFall 20121Review: Ordinary Kriging KFUPM| PETE-305: Reservoir Description | Dr Ena
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [20]Conventional Estimation Technique 5course notes byDr. Enamul HossainFall 20121Review: Cokriging KFUPM| PETE-305: Reservoir Description | Dr Enamul Hos
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [21]Conventional Estimation Technique 6course notes byDr. Enamul HossainFall 20121AnnouncementMajor Exam IIonNov 12, 2012 @ 12:00 Noon (Monday)Venue: B
King Fahd University of Petroleum & Minerals - PETE - 305
KFUPMENGINEERINGDepartment of Petroleum EngineeringPETE-305: RESERVOIR DESCRIPTIONLecture [22]Conventional Estimation Technique 7course notes byDr. Enamul HossainFall 20121Review: Nonlinear Kriging1.2.3.4.P128Log-Normal KrigingMulti-Gauss
King Fahd University of Petroleum & Minerals - PETE - 424
InstructorRock Mechanicsfor Petroleum EngineersDr. Abdulazeez AbdulraheemManaged research projects worth SR 10million in the past 10 years1Students2Students1 AL-HUSAIN, HASAN MUHAMMAD11 AL-ABDUL-MOHSEN, YOUNES A2 AL-RADHI, SALEH KHALIFAH12 AL