Mobility Prediction in Cellular Network Using Hidden Markov Model

Size: px
Start display at page:

Download "Mobility Prediction in Cellular Network Using Hidden Markov Model"

Transcription

1 Mobility Prediction in Cellular Network Using Hidden Markov Model Hongbo SI, Yue WANG, Jian YUAN and Xiuming SHAN Department of Electronic Engineering Tsinghua University, Beijing 184, P. R. China. Abstract In next generation networks, mobile communication calls for service with higher quality, which brings new challenge for mobility management. Thereinto, utilization and improvement of mobility prediction helps for preserving resource and providing better performance. So this paper aims to propose a theoretical and factual method to perform mobility prediction in cellular network. By analyzing the demand and character of this kind of personal mobility prediction in large spacial and temporal scale, it is concluded that Hidden Markov Model fits for system modeling. However, classical HMM algorithm will meet with numerical calculation problem when adopted to practical communication system. An improved algorithm is put forward to overcome possible calculating defects. Three different scenarios are set to testify HMM s efficiency and accuracy, using factual measurement data in cellular network. I. INTRODUCTION Mobile communication technology has felt the progressively increasing demand of users, since they do not entirely satisfy with the quality of current service. In order to improve communication condition and provide service of better performance, researchers and technologists have paid more attention to users movement patterns. Mobile terminal movements are never totally random since they are constrained by local terrain and traffic condition or since they usually have a precise purpose or habitual route. This ascertainment implicated behind uncertainty is just the foundation of mobility prediction. Thus a mathematical problem could be abstracted from application scenarios as a definition of mobility prediction: learning and inference from prior knowledge. This knowledge could be road information, movement history or user preference. In consideration of different research objects and spacialtemporal scales, mobility prediction can be divided into personal prediction [1] [2]. and community prediction [3] [4], macro-prediction [5] [6] and micro-prediction [7]. This paper is proposed to provide an effective method for solving personal prediction problem from a large point of view. Current researches have given some productions on this subject in theory [6], but its application in practical scenario is still lack of study. In this paper, Hidden Markov Model is adopted for mobile prediction and its factual effect is tested in cellular network system. The rest of paper is organized as follows. Section II gives a brief introduction on basic definition and calculation method of HMM. The next section describes HMM s application on mobility prediction in detail. Section IV exhibits the results of simulation using different forecasting methods, where prediction accuracy and consumed time are two major factors measuring the performance of algorithm.the last section gives a conclusion of the whole paper. II. HIDDEN MARKOV MODEL Hidden Markov Model (HMM) [9], as a classic part of the theory of Bayesian network [8], sets two kinds of stochastic variables, state variable (hidden variable) and output variable (observable variable). State variable (q 1:T in Fig.1) describes the factual situation of observed object, and the underling state-sequence forms a Markov chain. However, due to impossibility, difficulty or imprecision of observation, measured value may not reflect practical situation sometimes. So output variables (y 1:T in Fig.1) are distinguished from state variables, and the one-to-one mapping between them describes their corresponding relationship. Fig Structure of Hidden Markov Model Based on the structure of HMM (shown in Fig.1), there implicates two kinds of conditional independence statements. {q 1:t 2,y 1:t 1 } 1 y t {q 1:T \t,y 1:T \t } Using these assumptions, the joint probability distribution of all variables can be simplified as follow. T p(q 1:T,y 1:T ) p( 1 )p(y t ) (1) Since all other forms of probability distribution can be gained by marginalizing joint probability p(q 1:T,y 1:T ),itis believed that the whole model can be represented by p(q 1 ), p( 1 ) and p(y t ).Markλ (π, A, B) is introduced, where π is a N 1 vector and π q1 p(q 1 ); A is a N N matrix and A qt, p( ); B is a N N matrix and B qt,y t p(y t ) (N is the number of states). For HMM, there are three kinds of typical problems [9], which are also basically concerned in mobility prediction, so their solution are briefly discussed in next subsections /1/$ IEEE

2 A. Likelihood calculation Given a model λ (π, A, B), how to efficiently compute the likelihood of an output-sequence p(y 1:T λ)? Traditionally, the method of brute-force is often adopted. It can be estimated that the overall complexity of calculating p(y 1:T ) is O(N T ), thus this algorithm is simple and effective for small T, yet becoming invalid with the increase of T.For this, two denotations are introduced to reduce complexity by utilizing the redundancy of calculation. α( ) p(y 1:t, ) β( ) p(y t+1:t ) Then p(y 1:T, )α( )β( ). The problem of likelihood calculation is changed into computing α and β. Here recursive algorithm is adopted to reduce complexity. { Bq1,y 1 π q1 t α( ) A qt, α( ) t> (2) { A qt, β( ) t<t β( ) (3) 1 t T It can be summarized that complexity of this recursive method is O(TN 2 ), which is remarkably reduced comparing with O(N T ) of brute-force method. B. Decoding (recognition) Given an output-sequence y 1:T and a model λ, how to find the optimal state-sequence? max q 1:T p(q 1:T λ, y 1:T ) Getting q 1:T from y 1:T is necessary and important in inference theory of probabilistic graphical model [8]. In deed, only the distribution of is interested in mobility prediction. Thus, the calculation method is shown as follow and another two denotations are introduced. system structure. For HMM, using Baum-Welch algorithm [1], the answer to learning problem is shown as follow: πq 1 γ(q 1 ) (6) T 1 A, ξ(, ) T 1 γ(q (7) t) B,y t T γ()δ(y t,y t ) T γ() Where δ() is Kronecker delta function. Baum-Welch algorithm is an iterative algorithm and every time a group of parameters is calculated, the mount of p(y 1:T λ) and p(y 1:T λ ) should be compared. Only if p(y 1:T λ) p(y 1:T λ ), the algorithm is completed and the optimal parameters are acquired. III. MOBILITY PREDICTION USING HMM Hidden Markov Model is widely utilized in lots of domains [9], including automatic control, artificial intelligence, finance and biology. Here we attempt to adopt it in cellular communication system. A. Overview For cellular network, its hierarchical character lends itself to adopting HMM. In most cellular network systems, mobility management is performed in BSC. So what mobility prediction concerns is just user s path from entering the covering area of BSC to leaving. Thus the transfer matrix, distributed and managed by BSC for every user to record his movement habit, only thinks of the cells in this area, which brings about two advantages: 1) The number of states will not be too large. 2) The length of movement history will also be acceptable. B. System Modeling (8) ξ(, ) p(, y 1:T ) α( )β( )A qt,q t+1 B qt+1,y t+1 (4) α( )β( )A qt, qt γ( ) p( y 1:T ) ξ(, ) (5) Fig. 2. System modeling of cellular network (regular arrangement) Thus p( y 1:T )γ( ). C. Learning (parameter estimation) Given an output-sequence y 1:T, how to estimate the model parameters λ so as to best describe these data? max p(y 1:T λ) λ Target of learning is utilizing existent data to adjust model parameters, so as to make new parameters better describing Fig. 3. System modeling of cellular network (irregular arrangement) Structure of cellular network can be modeled as a graph. Nodes represent cells and edges represent the neighbouring relationship of cells. So the topology of cellular system, no matter regular (like Fig.2) or irregular arrangement (like Fig.3), can be abstracted into a undirected graph. From another

3 point of view, this graph is exactly the state-transition graph of every stochastic variable, and every one-step transition must be along the edge from one node to another. Based on this model, the path followed by a mobile is thus modeled as a string of symbols, called movement history. And the problem of mobility prediction in cellular network is converted into a problem of stochastic process. Usually, an assumption is introduced that the state sequence is stationary, based on general knowledge. C. Mobility Prediction Mobility prediction usually contains two major steps: 1) Parameter learning. The object of this process is to determine optimal parameters fitting history data. Using denotations of HMM, it means: max p(y 1:T λ) λ 2) Prediction. Actually, prediction process is inference in HMM. More concretely, in a mobility prediction process, three steps of inference are performed, including: max p( y 1:T ) max γ( ) (9) max p( +1 ) max A qt, +1 (1) max p(y T ) max B qt y T +1 y +1,y T +1 (11) T +1 In cellular network, not every changing of user s locating cell number will cost system resource. In fact, only handoff in communication process will lead system to distribute channel, so prediction is performed only for handoff. In another word, parameter learning happens as long as user steps into new cell, while prediction is only performed in communication. D. Algorithm Improvement According to the method above, using classic HMM will meet with problem when adopted in mobility prediction. Normally, location update performs frequently and it leads to long movement history. It is noticed that with the growing of T, the probability p(y 1:T ) is declining and finally less than the calculation precision of computer. Facing this problem, two classic numerical calculation solutions are worthy for reference. Normalized probability distribution. Using normalized distribution for α and β will keep their calculation precision in recursive processes. ᾱ( ) α() α( ) β( ) β() β( ) (12) Logarithmic summation. Replacing product by logarithmic summation also preserves calculation precision. This action ensures data apart from calculation limit of computers. These treatments must cause changes to classic algorithm. It can be imagined that the following three aspects should be improved in new algorithm. 1) The recursion of ᾱ and β. Using (2)(12), and noticing that α( )/ᾱ( ) α( ) p(y 1:t ) is independent of,wehave: ᾱ( ) α() α( ) A qt, ᾱ( ) A qt, ᾱ( ) And especially for q 1, ᾱ(q 1 ) (13) B q 1,y 1 π q1 q 1 B q1,y 1 π q1 (14) Similarly the recursion of β is as follow: A qt, β(qt+1 ) β( ) A qt, β(qt+1 ) And especially for, (15) β( ) 1 (16) T 2) The expression of likelihood p(y 1:T λ). Reconsider the denominator of calculating ᾱ( ) and give it a denotation μ t+1,wehave: μ t+1 A qt, ᾱ( ) q t α( ) p(y 1:t+1) p(y 1:t ) p(y 1:t ) Thus p(y 1:T )μ T p(y 1:T 1 ) T μ t (17) When using the logarithmic form, equation (17) can be expressed as: T 1 log p(y 1:T ) log T μ t log μ t (18) 3) The method of calculating π, A and B. Based on the definition of ᾱ and β, the form of calculating ξ is not changed. So the calculating method of γ and π, A, B is not radically changed accordingly. ξ(, ) ᾱ( ) β( )A qt, ᾱ( ) β( )A qt, (19) E. Algorithm Pseudocode The pseudocode of improved HMM algorithm can be summarized as follow. Learning and Prediction are two kernel functions for main program.

4 Algorithm 1 Main Program Require: observation information y 1:T Ensure: prediction results r 2:T +1 1: Algorithm starts up when user enters appointed area. 2: if user is the first time accessing to this area then 3: Initialize λ with random parameters 4: else 5: Initialize λ with history parameters 6: end if 7: T 1 8: Calculate α and β using (2) and (3) 9: while user is still in this area do 1: Add new cell information to y 1:T 11: if user is in communication then 12: λ Learning(λ, y 1:T ) 13: r T +1 P rediciton(λ,y 1:T ) 14: return r 2:T +1 15: else 16: λ Learning(λ, y 1:T ) 17: end if 18: T T +1 19: end while Algorithm 2 Function Learning Require: λ, y 1:T Ensure: λ 1: Calculate ᾱ, β, ξ and γ using (13) (15) (19) (5) 2: Calculate log p(y 1:T λ) using (18) 3: for r 1to c do 4: Parameter adjust to get π, A and B using (6) (7) (8) 5: Recalculate ᾱ, β, ξ and γ using (13) (15) (19) (5) 6: Recalculate log p(y 1:T λ ) using (18) 7: if log p(y 1:T λ ) > log p(y 1:T λ) then 8: Replace π, A and B with π, A and B 9: else 1: End circulation 11: Replace π, A and B with π, A and B 12: end if 13: end for 14: return λ (π,a,b ) Algorithm 3 Function Prediction Require: λ, y 1:T Ensure: r T +1 1: Get from max p( y 1:T ) using (9) 2: Get +1 from max p( +1 ) using (1) +1 3: Get r T +1 from max p(r T ) using (11) r T +1 4: return r T +1 A. Data Source IV. SIMULATION A database, called Reality Mining Project [11] from MIT media laboratory, is adopted in the simulation. This project records movement history and communication situation of 1 persons over one year. For personal prediction in large scale, what we interest in this database is locating cell tower information and corresponding staying time of a particular user. Using this database for mobility prediction, a particular location area with group of cells is chosen as the background of research. The system model of neighbouring relationship in this area is shown in Fig.4. Moreover, a person in this project is selected as our research object. Fig. 4. System model of location area in simulation B. Simulation Result In order to exhibit the performance of HMM in different application scenarios, two different actual environments are set in our simulation, where prediction accuracy and time consumed are two major elements of algorithm concerned. It is noted that in order to realize the performance of HMM, prediction is executed as long as user s locating cell number changes, yet in actual cellular system, only changes in communication cause mobility prediction. For comparison, Markov chain and order-2 Markov method are also contained in the simulation. Scenario One. User s first access to this location area. Since no history data for this user can be utilized, the initial parameters of HMM are randomly generated, using a distribution close to uniform. Scenario Two. The user has accessed this area frequently and left mature data for mobility modeling, yet he suddenly changes his moving habit. Here old parameters are still adopted when initializing HMM since the changing is unpredictable, and HMM s fitting ability is mainly concerned in this scenario. Accuracy of prediction HMM Markov order 2 Markov Fig. 5. Prediction accuracy in scenario one

5 Time consumed in prediction Accuracy of prediction Time consumed in prediction Fig. 6. Time consumed in scenario one HMM Markov order 2 Markov Fig. 7. Prediction accuracy in scenario two Fig. 8. C. Result Analysis Time consumed in scenario two From figures above, it is concluded that HMM surely has its advantage in prediction accuracy, comparing with Markov and order-2 Markov. A precise prediction algorithm is usually measured from two aspects: insensitive to outliers and adaptive to new changing. In fact, these two characters contradict with each other, so making moderate compromise is crucial. HMM and Markov method are both based on large data of history, which makes them insensitive to outliers. On the other hand, comparing with Markov method, HMM has more adaptive ability to user s change. In another word, HMM learns faster than Markov, which is determined by algorithm itself. Considering the calculation method of these algorithms, it can be analyzed that multiplication complexity of HMM is O(TN 2 ) while the one of Markov and order-k Markov is only O(1). So time consumed in prediction using Markov and order-2 Markov is far less than using HMM. This theoretical conclusion can be proved from figures, which only show the time consumed of HMM (Time curves of Markov and order-2 Markov are so close to that they cannot be seen in figures). It is seen that the trend of curve is basically linear with the increase of sequence. And the fluctuation is mainly due to different attempt times in parameter learning. V. CONCLUSION In order to improve communication performance in cellular, the technology of mobility prediction is introduced. Through analysis and simulation, it is concluded that Hidden Markov Model uses more cost in calculation time to exchange for prediction accuracy. However, in actual communication system, the movement path may not be so long as our simulation scenarios, so temporal cost in prediction is not comparable to staying time in corresponding cell. Thus, Hidden Markov Model is proved to be efficient and accurate as a solution to mobility prediction, and will exhibit more merit if adopted in factual communication system. ACKNOWLEDGMENT This work is supported in part by the National Basic Research Program of China (973 Program) under grants 27CB371 and 27CB3715. REFERENCES [1] Amiya Bhattacharya and Sajal K. Das, LeZi-update: an informationtheoretic approach to track mobile users in PCS networks, International Conference on Mobile Computing and Networking, pp.1-12, [2] Lucian Vintan, Arpad Gellert, Jan Petzold, and Theo Ungerer, Person Movement Prediction Using Neural Networks, First Workshop on Modeling and Retrieval of Context, 24. [3] Wee-Seng Soh and Hyong S. Kim, Dynamic Bandwidth Reservation in Cellular Networks Using Road Topology Based Mobility Predictions, INFOCOM 24, pp , 24. [4] Wee-Seng Soh and Hyong S. Kim, Dynamic guard bandwidth scheme for wireless broadband networks, INFOCOM 21, pp , 21. [5] Fei Yu and Victor Leung, Mobility-based predictive call admission control and bandwidth reservation in wireless cellular networks, Computer Networks, vol.38, pp , 22. [6] Arpad Gellert and Lucian Vintan, Person Movement Prediction Using Hidden Markov Models, Studies in Informatics and Control, vol.15, pp.17-3, 26. [7] Wei Cui and Xuemin Shen, User Movement Tendency Prediction and Call Admission Control for Cellular Networks, IEEE International Conference on Communications, pp , 2. [8] Micheal I. Jordan, Graphical Models, Statistical Science, vol.19, pp , 24. [9] Sherif Akoush and Ahmed Sameh, Mobile User Movement Prediction Using Bayesian Learning for Neural Networks, Proceeding of the 27 International Conference on Wireless Communications and Mobile Computing, pp , 27. [1] LE Baum and T Petrie, Statistical Inference for Probabilistic Functions of Finite State Markov Chains, Annals of Mathematical Statistics, vol.37, , [11] N Eagle and A Pentland, Reality Mining: Sensing Complex Social Systems, Personal and Ubiquitous Computing, Vol 1, pp , 26.

ECE521: Week 11, Lecture March 2017: HMM learning/inference. With thanks to Russ Salakhutdinov

ECE521: Week 11, Lecture March 2017: HMM learning/inference. With thanks to Russ Salakhutdinov ECE521: Week 11, Lecture 20 27 March 2017: HMM learning/inference With thanks to Russ Salakhutdinov Examples of other perspectives Murphy 17.4 End of Russell & Norvig 15.2 (Artificial Intelligence: A Modern

More information

Using HMM in Strategic Games

Using HMM in Strategic Games Using HMM in Strategic Games Mario Benevides Isaque Lima Rafael Nader Pedro Rougemont Systems and Computer Engineering Program and Computer Science Department Federal University of Rio de Janeiro, Brazil

More information

A Two-phase Distributed Training Algorithm for Linear SVM in WSN

A Two-phase Distributed Training Algorithm for Linear SVM in WSN Proceedings of the World Congress on Electrical Engineering and Computer Systems and Science (EECSS 015) Barcelona, Spain July 13-14, 015 Paper o. 30 A wo-phase Distributed raining Algorithm for Linear

More information

ε-machine Estimation and Forecasting

ε-machine Estimation and Forecasting ε-machine Estimation and Forecasting Comparative Study of Inference Methods D. Shemetov 1 1 Department of Mathematics University of California, Davis Natural Computation, 2014 Outline 1 Motivation ε-machines

More information

An embedded system of Face Recognition based on ARM and HMM

An embedded system of Face Recognition based on ARM and HMM An embedded system of Face Recognition based on ARM and HMM Yanbin Sun 1,2, Lun Xie 1, Zhiliang Wang 1,Yi An 2 1 Department of Electronic Information Engineering, School of Information Engineering, University

More information

INDIAN INSTITUTE OF MANAGEMENT CALCUTTA WORKING PAPER SERIES. WPS No. 644/ August A Markov-based Diurnal Mobility Model for 3G Cellular Networks

INDIAN INSTITUTE OF MANAGEMENT CALCUTTA WORKING PAPER SERIES. WPS No. 644/ August A Markov-based Diurnal Mobility Model for 3G Cellular Networks INDIAN INSTITUTE OF MANAGEMENT CALCUTTA WORKING PAPER SERIES WPS No. 644/ August 2009 A Markov-based Diurnal Mobility Model for 3G Cellular Networks by Samir K Sadhukhan SSA, IIM Calcutta, Diamond Harbour

More information

Clustering-Based Distributed Precomputation for Quality-of-Service Routing*

Clustering-Based Distributed Precomputation for Quality-of-Service Routing* Clustering-Based Distributed Precomputation for Quality-of-Service Routing* Yong Cui and Jianping Wu Department of Computer Science, Tsinghua University, Beijing, P.R.China, 100084 cy@csnet1.cs.tsinghua.edu.cn,

More information

Recurrent Neural Network (RNN) Industrial AI Lab.

Recurrent Neural Network (RNN) Industrial AI Lab. Recurrent Neural Network (RNN) Industrial AI Lab. For example (Deterministic) Time Series Data Closed- form Linear difference equation (LDE) and initial condition High order LDEs 2 (Stochastic) Time Series

More information

arxiv: v1 [cond-mat.dis-nn] 30 Dec 2018

arxiv: v1 [cond-mat.dis-nn] 30 Dec 2018 A General Deep Learning Framework for Structure and Dynamics Reconstruction from Time Series Data arxiv:1812.11482v1 [cond-mat.dis-nn] 30 Dec 2018 Zhang Zhang, Jing Liu, Shuo Wang, Ruyue Xin, Jiang Zhang

More information

Hidden Markov Models in the context of genetic analysis

Hidden Markov Models in the context of genetic analysis Hidden Markov Models in the context of genetic analysis Vincent Plagnol UCL Genetics Institute November 22, 2012 Outline 1 Introduction 2 Two basic problems Forward/backward Baum-Welch algorithm Viterbi

More information

A noninformative Bayesian approach to small area estimation

A noninformative Bayesian approach to small area estimation A noninformative Bayesian approach to small area estimation Glen Meeden School of Statistics University of Minnesota Minneapolis, MN 55455 glen@stat.umn.edu September 2001 Revised May 2002 Research supported

More information

Hidden Markov Models. Slides adapted from Joyce Ho, David Sontag, Geoffrey Hinton, Eric Xing, and Nicholas Ruozzi

Hidden Markov Models. Slides adapted from Joyce Ho, David Sontag, Geoffrey Hinton, Eric Xing, and Nicholas Ruozzi Hidden Markov Models Slides adapted from Joyce Ho, David Sontag, Geoffrey Hinton, Eric Xing, and Nicholas Ruozzi Sequential Data Time-series: Stock market, weather, speech, video Ordered: Text, genes Sequential

More information

Simultaneous Perturbation Stochastic Approximation Algorithm Combined with Neural Network and Fuzzy Simulation

Simultaneous Perturbation Stochastic Approximation Algorithm Combined with Neural Network and Fuzzy Simulation .--- Simultaneous Perturbation Stochastic Approximation Algorithm Combined with Neural Networ and Fuzzy Simulation Abstract - - - - Keywords: Many optimization problems contain fuzzy information. Possibility

More information

Shallow Parsing Swapnil Chaudhari 11305R011 Ankur Aher Raj Dabre 11305R001

Shallow Parsing Swapnil Chaudhari 11305R011 Ankur Aher Raj Dabre 11305R001 Shallow Parsing Swapnil Chaudhari 11305R011 Ankur Aher - 113059006 Raj Dabre 11305R001 Purpose of the Seminar To emphasize on the need for Shallow Parsing. To impart basic information about techniques

More information

Heuristic Algorithms for Multiconstrained Quality-of-Service Routing

Heuristic Algorithms for Multiconstrained Quality-of-Service Routing 244 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL 10, NO 2, APRIL 2002 Heuristic Algorithms for Multiconstrained Quality-of-Service Routing Xin Yuan, Member, IEEE Abstract Multiconstrained quality-of-service

More information

Volume 2, Issue 11, November 2014 International Journal of Advance Research in Computer Science and Management Studies

Volume 2, Issue 11, November 2014 International Journal of Advance Research in Computer Science and Management Studies Volume 2, Issue 11, November 2014 International Journal of Advance Research in Computer Science and Management Studies Research Article / Survey Paper / Case Study Available online at: www.ijarcsms.com

More information

A Boosting-Based Framework for Self-Similar and Non-linear Internet Traffic Prediction

A Boosting-Based Framework for Self-Similar and Non-linear Internet Traffic Prediction A Boosting-Based Framework for Self-Similar and Non-linear Internet Traffic Prediction Hanghang Tong 1, Chongrong Li 2, and Jingrui He 1 1 Department of Automation, Tsinghua University, Beijing 100084,

More information

Conditional Random Fields - A probabilistic graphical model. Yen-Chin Lee 指導老師 : 鮑興國

Conditional Random Fields - A probabilistic graphical model. Yen-Chin Lee 指導老師 : 鮑興國 Conditional Random Fields - A probabilistic graphical model Yen-Chin Lee 指導老師 : 鮑興國 Outline Labeling sequence data problem Introduction conditional random field (CRF) Different views on building a conditional

More information

Assignment 2. Unsupervised & Probabilistic Learning. Maneesh Sahani Due: Monday Nov 5, 2018

Assignment 2. Unsupervised & Probabilistic Learning. Maneesh Sahani Due: Monday Nov 5, 2018 Assignment 2 Unsupervised & Probabilistic Learning Maneesh Sahani Due: Monday Nov 5, 2018 Note: Assignments are due at 11:00 AM (the start of lecture) on the date above. he usual College late assignments

More information

The Basics of Graphical Models

The Basics of Graphical Models The Basics of Graphical Models David M. Blei Columbia University September 30, 2016 1 Introduction (These notes follow Chapter 2 of An Introduction to Probabilistic Graphical Models by Michael Jordan.

More information

Dynamic Programming. Ellen Feldman and Avishek Dutta. February 27, CS155 Machine Learning and Data Mining

Dynamic Programming. Ellen Feldman and Avishek Dutta. February 27, CS155 Machine Learning and Data Mining CS155 Machine Learning and Data Mining February 27, 2018 Motivation Much of machine learning is heavily dependent on computational power Many libraries exist that aim to reduce computational time TensorFlow

More information

An Adaptive Bandwidth Reservation Scheme for Multimedia Mobile Cellular Networks

An Adaptive Bandwidth Reservation Scheme for Multimedia Mobile Cellular Networks An Adaptive Bandwidth Reservation Scheme for Multimedia Mobile Cellular Networks Hong Bong Kim Telecommunication Networks Group, Technical University of Berlin Sekr FT5 Einsteinufer 25 1587 Berlin Germany

More information

Directed Graphical Models (Bayes Nets) (9/4/13)

Directed Graphical Models (Bayes Nets) (9/4/13) STA561: Probabilistic machine learning Directed Graphical Models (Bayes Nets) (9/4/13) Lecturer: Barbara Engelhardt Scribes: Richard (Fangjian) Guo, Yan Chen, Siyang Wang, Huayang Cui 1 Introduction For

More information

Intelligent Hands Free Speech based SMS System on Android

Intelligent Hands Free Speech based SMS System on Android Intelligent Hands Free Speech based SMS System on Android Gulbakshee Dharmale 1, Dr. Vilas Thakare 3, Dr. Dipti D. Patil 2 1,3 Computer Science Dept., SGB Amravati University, Amravati, INDIA. 2 Computer

More information

Graphical Models & HMMs

Graphical Models & HMMs Graphical Models & HMMs Henrik I. Christensen Robotics & Intelligent Machines @ GT Georgia Institute of Technology, Atlanta, GA 30332-0280 hic@cc.gatech.edu Henrik I. Christensen (RIM@GT) Graphical Models

More information

Prediction of traffic flow based on the EMD and wavelet neural network Teng Feng 1,a,Xiaohong Wang 1,b,Yunlai He 1,c

Prediction of traffic flow based on the EMD and wavelet neural network Teng Feng 1,a,Xiaohong Wang 1,b,Yunlai He 1,c 2nd International Conference on Electrical, Computer Engineering and Electronics (ICECEE 215) Prediction of traffic flow based on the EMD and wavelet neural network Teng Feng 1,a,Xiaohong Wang 1,b,Yunlai

More information

Constraints in Particle Swarm Optimization of Hidden Markov Models

Constraints in Particle Swarm Optimization of Hidden Markov Models Constraints in Particle Swarm Optimization of Hidden Markov Models Martin Macaš, Daniel Novák, and Lenka Lhotská Czech Technical University, Faculty of Electrical Engineering, Dep. of Cybernetics, Prague,

More information

MEMMs (Log-Linear Tagging Models)

MEMMs (Log-Linear Tagging Models) Chapter 8 MEMMs (Log-Linear Tagging Models) 8.1 Introduction In this chapter we return to the problem of tagging. We previously described hidden Markov models (HMMs) for tagging problems. This chapter

More information

Title Grid for Multimedia Communication Ne. The original publication is availabl. Press

Title Grid for Multimedia Communication Ne. The original publication is availabl. Press JAIST Reposi https://dspace.j Title Grid for Multimedia Communication Ne A Double Helix Architecture of Knowl Discovery System Based Data Grid and Author(s)Jing, He; Wuyi, Yue; Yong, Shi Citation Issue

More information

Modeling time series with hidden Markov models

Modeling time series with hidden Markov models Modeling time series with hidden Markov models Advanced Machine learning 2017 Nadia Figueroa, Jose Medina and Aude Billard Time series data Barometric pressure Temperature Data Humidity Time What s going

More information

Markov Decision Processes and Reinforcement Learning

Markov Decision Processes and Reinforcement Learning Lecture 14 and Marco Chiarandini Department of Mathematics & Computer Science University of Southern Denmark Slides by Stuart Russell and Peter Norvig Course Overview Introduction Artificial Intelligence

More information

Where Next? Data Mining Techniques and Challenges for Trajectory Prediction. Slides credit: Layla Pournajaf

Where Next? Data Mining Techniques and Challenges for Trajectory Prediction. Slides credit: Layla Pournajaf Where Next? Data Mining Techniques and Challenges for Trajectory Prediction Slides credit: Layla Pournajaf o Navigational services. o Traffic management. o Location-based advertising. Source: A. Monreale,

More information

A Verification Method for Software Safety Requirement by Combining Model Checking and FTA Congcong Chen1,a, Fuping Zeng1,b, Minyan Lu1,c

A Verification Method for Software Safety Requirement by Combining Model Checking and FTA Congcong Chen1,a, Fuping Zeng1,b, Minyan Lu1,c International Industrial Informatics and Computer Engineering Conference (IIICEC 2015) A Verification Method for Software Safety Requirement by Combining Model Checking and FTA Congcong Chen1,a, Fuping

More information

Supplementary file for SybilDefender: A Defense Mechanism for Sybil Attacks in Large Social Networks

Supplementary file for SybilDefender: A Defense Mechanism for Sybil Attacks in Large Social Networks 1 Supplementary file for SybilDefender: A Defense Mechanism for Sybil Attacks in Large Social Networks Wei Wei, Fengyuan Xu, Chiu C. Tan, Qun Li The College of William and Mary, Temple University {wwei,

More information

STA 4273H: Statistical Machine Learning

STA 4273H: Statistical Machine Learning STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! http://www.utstat.utoronto.ca/~rsalakhu/ Sidney Smith Hall, Room 6002 Lecture 5 Inference

More information

A Decision-Theoretic Rough Set Model

A Decision-Theoretic Rough Set Model A Decision-Theoretic Rough Set Model Yiyu Yao and Jingtao Yao Department of Computer Science University of Regina Regina, Saskatchewan, Canada S4S 0A2 {yyao,jtyao}@cs.uregina.ca Special Thanks to Professor

More information

A Study on Network Flow Security

A Study on Network Flow Security BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 8, No 3 Sofia 28 A Study on Network Flow Security Tsvetomir Tsvetanov, Stanislav Simeonov 2 Sofia University, Faculty of Mathematics

More information

Some Problems of Fuzzy Modeling of Telecommunications Networks

Some Problems of Fuzzy Modeling of Telecommunications Networks Some Problems of Fuzzy Modeling of Telecommunications Networks Kirill Garbuzov Novosibirsk State University Department of Mechanics and Mathematics Novosibirsk, Russia, 630090, Email: gartesk@rambler.ru

More information

Graphical Models. David M. Blei Columbia University. September 17, 2014

Graphical Models. David M. Blei Columbia University. September 17, 2014 Graphical Models David M. Blei Columbia University September 17, 2014 These lecture notes follow the ideas in Chapter 2 of An Introduction to Probabilistic Graphical Models by Michael Jordan. In addition,

More information

On Reduct Construction Algorithms

On Reduct Construction Algorithms 1 On Reduct Construction Algorithms Yiyu Yao 1, Yan Zhao 1 and Jue Wang 2 1 Department of Computer Science, University of Regina Regina, Saskatchewan, Canada S4S 0A2 {yyao, yanzhao}@cs.uregina.ca 2 Laboratory

More information

CS224W Final Report Emergence of Global Status Hierarchy in Social Networks

CS224W Final Report Emergence of Global Status Hierarchy in Social Networks CS224W Final Report Emergence of Global Status Hierarchy in Social Networks Group 0: Yue Chen, Jia Ji, Yizheng Liao December 0, 202 Introduction Social network analysis provides insights into a wide range

More information

Domain Based Approach for QoS Provisioning in Mobile IP

Domain Based Approach for QoS Provisioning in Mobile IP Domain Based Approach for QoS Provisioning in Mobile IP Ki-Il Kim and Sang-Ha Kim Department of Computer Science 220 Gung-dong,Yuseong-gu, Chungnam National University, Deajeon 305-764, Korea {kikim, shkim}@cclab.cnu.ac.kr

More information

Computer Based Image Algorithm For Wireless Sensor Networks To Prevent Hotspot Locating Attack

Computer Based Image Algorithm For Wireless Sensor Networks To Prevent Hotspot Locating Attack Computer Based Image Algorithm For Wireless Sensor Networks To Prevent Hotspot Locating Attack J.Anbu selvan 1, P.Bharat 2, S.Mathiyalagan 3 J.Anand 4 1, 2, 3, 4 PG Scholar, BIT, Sathyamangalam ABSTRACT:

More information

The Method of User s Identification Using the Fusion of Wavelet Transform and Hidden Markov Models

The Method of User s Identification Using the Fusion of Wavelet Transform and Hidden Markov Models The Method of User s Identification Using the Fusion of Wavelet Transform and Hidden Markov Models Janusz Bobulski Czȩstochowa University of Technology, Institute of Computer and Information Sciences,

More information

SIMULATION OF ARTIFICIAL SYSTEMS BEHAVIOR IN PARAMETRIC EIGHT-DIMENSIONAL SPACE

SIMULATION OF ARTIFICIAL SYSTEMS BEHAVIOR IN PARAMETRIC EIGHT-DIMENSIONAL SPACE 78 Proceedings of the 4 th International Conference on Informatics and Information Technology SIMULATION OF ARTIFICIAL SYSTEMS BEHAVIOR IN PARAMETRIC EIGHT-DIMENSIONAL SPACE D. Ulbikiene, J. Ulbikas, K.

More information

Research Article Optimization of Access Threshold for Cognitive Radio Networks with Prioritized Secondary Users

Research Article Optimization of Access Threshold for Cognitive Radio Networks with Prioritized Secondary Users Mobile Information Systems Volume 2016, Article ID 3297938, 8 pages http://dx.doi.org/10.1155/2016/3297938 Research Article Optimization of Access Threshold for Cognitive Radio Networks with Prioritized

More information

Representability of Human Motions by Factorial Hidden Markov Models

Representability of Human Motions by Factorial Hidden Markov Models Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems San Diego, CA, USA, Oct 29 - Nov 2, 2007 WeC10.1 Representability of Human Motions by Factorial Hidden Markov

More information

Link Lifetime Prediction in Mobile Ad-Hoc Network Using Curve Fitting Method

Link Lifetime Prediction in Mobile Ad-Hoc Network Using Curve Fitting Method IJCSNS International Journal of Computer Science and Network Security, VOL.17 No.5, May 2017 265 Link Lifetime Prediction in Mobile Ad-Hoc Network Using Curve Fitting Method Mohammad Pashaei, Hossein Ghiasy

More information

10-701/15-781, Fall 2006, Final

10-701/15-781, Fall 2006, Final -7/-78, Fall 6, Final Dec, :pm-8:pm There are 9 questions in this exam ( pages including this cover sheet). If you need more room to work out your answer to a question, use the back of the page and clearly

More information

Challenges in Ubiquitous Data Mining

Challenges in Ubiquitous Data Mining LIAAD-INESC Porto, University of Porto, Portugal jgama@fep.up.pt 1 2 Very-short-term Forecasting in Photovoltaic Systems 3 4 Problem Formulation: Network Data Model Querying Model Query = Q( n i=0 S i)

More information

Nonparametric Bayesian Texture Learning and Synthesis

Nonparametric Bayesian Texture Learning and Synthesis Appears in Advances in Neural Information Processing Systems (NIPS) 2009. Nonparametric Bayesian Texture Learning and Synthesis Long (Leo) Zhu 1 Yuanhao Chen 2 William Freeman 1 Antonio Torralba 1 1 CSAIL,

More information

A New Call Admission Control scheme for Real-time traffic in Wireless Networks

A New Call Admission Control scheme for Real-time traffic in Wireless Networks A New Call Admission Control scheme for Real-time traffic in Wireless Networks Maneesh Tewari and H.S. Jamadagni Center for Electronics Design and Technology, Indian Institute of Science, Bangalore, 5612

More information

Recognition of Human Body Movements Trajectory Based on the Three-dimensional Depth Data

Recognition of Human Body Movements Trajectory Based on the Three-dimensional Depth Data Preprints of the 19th World Congress The International Federation of Automatic Control Recognition of Human Body s Trajectory Based on the Three-dimensional Depth Data Zheng Chang Qing Shen Xiaojuan Ban

More information

Support Vector Regression for Software Reliability Growth Modeling and Prediction

Support Vector Regression for Software Reliability Growth Modeling and Prediction Support Vector Regression for Software Reliability Growth Modeling and Prediction 925 Fei Xing 1 and Ping Guo 2 1 Department of Computer Science Beijing Normal University, Beijing 100875, China xsoar@163.com

More information

Improving Time Series Classification Using Hidden Markov Models

Improving Time Series Classification Using Hidden Markov Models Improving Time Series Classification Using Hidden Markov Models Bilal Esmael Arghad Arnaout Rudolf K. Fruhwirth Gerhard Thonhauser University of Leoben TDE GmbH TDE GmbH University of Leoben Leoben, Austria

More information

A NOVEL APPROACH FOR PREDICTING MOVEMENT OF MOBILE USERS BASED ON DATA MINING TECHNIQUES

A NOVEL APPROACH FOR PREDICTING MOVEMENT OF MOBILE USERS BASED ON DATA MINING TECHNIQUES A NOVEL APPROACH FOR PREDICTING MOVEMENT OF MOBILE USERS BASED ON DATA MINING TECHNIQUES V.Nivedha 1, E. Karunakaran 2, J.Kumaran@Kumar 3 1 Student, Dept. of CSE, Pondicherry Engineering College, Puducherry,

More information

A Timer-based Session Setup Procedure in Cellular-WLAN Integrated Systems

A Timer-based Session Setup Procedure in Cellular-WLAN Integrated Systems his paper was presented as part of the Mobility Management in the Networks of the Future World (MobiWorld) Workshop at A -based Session Setup Procedure in Cellular-WLAN Integrated Systems Gwangwoo Park,

More information

The Use of Fuzzy Logic at Support of Manager Decision Making

The Use of Fuzzy Logic at Support of Manager Decision Making The Use of Fuzzy Logic at Support of Manager Decision Making The use of fuzzy logic is the advantage especially at decision making processes where the description by algorithms is very difficult and criteria

More information

Computer Vision Group Prof. Daniel Cremers. 4. Probabilistic Graphical Models Directed Models

Computer Vision Group Prof. Daniel Cremers. 4. Probabilistic Graphical Models Directed Models Prof. Daniel Cremers 4. Probabilistic Graphical Models Directed Models The Bayes Filter (Rep.) (Bayes) (Markov) (Tot. prob.) (Markov) (Markov) 2 Graphical Representation (Rep.) We can describe the overall

More information

Some questions of consensus building using co-association

Some questions of consensus building using co-association Some questions of consensus building using co-association VITALIY TAYANOV Polish-Japanese High School of Computer Technics Aleja Legionow, 4190, Bytom POLAND vtayanov@yahoo.com Abstract: In this paper

More information

A Modified Algorithm to Handle Dangling Pages using Hypothetical Node

A Modified Algorithm to Handle Dangling Pages using Hypothetical Node A Modified Algorithm to Handle Dangling Pages using Hypothetical Node Shipra Srivastava Student Department of Computer Science & Engineering Thapar University, Patiala, 147001 (India) Rinkle Rani Aggrawal

More information

Terminating Decision Algorithms Optimally

Terminating Decision Algorithms Optimally Terminating Decision Algorithms Optimally Tuomas Sandholm Computer Science Department Carnegie Mellon University Pittsburgh PA 15213 sandholm@cs.cmu.edu Abstract. Incomplete decision algorithms can often

More information

Structured Learning. Jun Zhu

Structured Learning. Jun Zhu Structured Learning Jun Zhu Supervised learning Given a set of I.I.D. training samples Learn a prediction function b r a c e Supervised learning (cont d) Many different choices Logistic Regression Maximum

More information

New Results on Simple Stochastic Games

New Results on Simple Stochastic Games New Results on Simple Stochastic Games Decheng Dai 1 and Rong Ge 2 1 Tsinghua University, ddc02@mails.tsinghua.edu.cn 2 Princeton University, rongge@cs.princeton.edu Abstract. We study the problem of solving

More information

A Data Classification Algorithm of Internet of Things Based on Neural Network

A Data Classification Algorithm of Internet of Things Based on Neural Network A Data Classification Algorithm of Internet of Things Based on Neural Network https://doi.org/10.3991/ijoe.v13i09.7587 Zhenjun Li Hunan Radio and TV University, Hunan, China 278060389@qq.com Abstract To

More information

Image Segmentation for Image Object Extraction

Image Segmentation for Image Object Extraction Image Segmentation for Image Object Extraction Rohit Kamble, Keshav Kaul # Computer Department, Vishwakarma Institute of Information Technology, Pune kamble.rohit@hotmail.com, kaul.keshav@gmail.com ABSTRACT

More information

Classification with Diffuse or Incomplete Information

Classification with Diffuse or Incomplete Information Classification with Diffuse or Incomplete Information AMAURY CABALLERO, KANG YEN Florida International University Abstract. In many different fields like finance, business, pattern recognition, communication

More information

Modelling and quantitative analysis of LTRACK A novel mobility management algorithm

Modelling and quantitative analysis of LTRACK A novel mobility management algorithm Mobile Information Systems (006) 1 50 1 IOS Press Modelling and quantitative analysis of LTRACK A novel mobility management algorithm Benedek Kovács, M. Szalay and S. Imre Department of Telecommunications,

More information

Markov Model Based Congestion Control for TCP

Markov Model Based Congestion Control for TCP Markov Model Based Congestion Control for TCP Shan Suthaharan University of North Carolina at Greensboro, Greensboro, NC 27402, USA ssuthaharan@uncg.edu Abstract The Random Early Detection (RED) scheme

More information

Texture Modeling using MRF and Parameters Estimation

Texture Modeling using MRF and Parameters Estimation Texture Modeling using MRF and Parameters Estimation Ms. H. P. Lone 1, Prof. G. R. Gidveer 2 1 Postgraduate Student E & TC Department MGM J.N.E.C,Aurangabad 2 Professor E & TC Department MGM J.N.E.C,Aurangabad

More information

NOVEL HYBRID GENETIC ALGORITHM WITH HMM BASED IRIS RECOGNITION

NOVEL HYBRID GENETIC ALGORITHM WITH HMM BASED IRIS RECOGNITION NOVEL HYBRID GENETIC ALGORITHM WITH HMM BASED IRIS RECOGNITION * Prof. Dr. Ban Ahmed Mitras ** Ammar Saad Abdul-Jabbar * Dept. of Operation Research & Intelligent Techniques ** Dept. of Mathematics. College

More information

Lecture 21 : A Hybrid: Deep Learning and Graphical Models

Lecture 21 : A Hybrid: Deep Learning and Graphical Models 10-708: Probabilistic Graphical Models, Spring 2018 Lecture 21 : A Hybrid: Deep Learning and Graphical Models Lecturer: Kayhan Batmanghelich Scribes: Paul Liang, Anirudha Rayasam 1 Introduction and Motivation

More information

An Introduction to Pattern Recognition

An Introduction to Pattern Recognition An Introduction to Pattern Recognition Speaker : Wei lun Chao Advisor : Prof. Jian-jiun Ding DISP Lab Graduate Institute of Communication Engineering 1 Abstract Not a new research field Wide range included

More information

An ELM-based traffic flow prediction method adapted to different data types Wang Xingchao1, a, Hu Jianming2, b, Zhang Yi3 and Wang Zhenyu4

An ELM-based traffic flow prediction method adapted to different data types Wang Xingchao1, a, Hu Jianming2, b, Zhang Yi3 and Wang Zhenyu4 6th International Conference on Information Engineering for Mechanics and Materials (ICIMM 206) An ELM-based traffic flow prediction method adapted to different data types Wang Xingchao, a, Hu Jianming2,

More information

Clustering: Classic Methods and Modern Views

Clustering: Classic Methods and Modern Views Clustering: Classic Methods and Modern Views Marina Meilă University of Washington mmp@stat.washington.edu June 22, 2015 Lorentz Center Workshop on Clusters, Games and Axioms Outline Paradigms for clustering

More information

Construction C : an inter-level coded version of Construction C

Construction C : an inter-level coded version of Construction C Construction C : an inter-level coded version of Construction C arxiv:1709.06640v2 [cs.it] 27 Dec 2017 Abstract Besides all the attention given to lattice constructions, it is common to find some very

More information

A Game Map Complexity Measure Based on Hamming Distance Yan Li, Pan Su, and Wenliang Li

A Game Map Complexity Measure Based on Hamming Distance Yan Li, Pan Su, and Wenliang Li Physics Procedia 22 (2011) 634 640 2011 International Conference on Physics Science and Technology (ICPST 2011) A Game Map Complexity Measure Based on Hamming Distance Yan Li, Pan Su, and Wenliang Li Collage

More information

Loopy Belief Propagation

Loopy Belief Propagation Loopy Belief Propagation Research Exam Kristin Branson September 29, 2003 Loopy Belief Propagation p.1/73 Problem Formalization Reasoning about any real-world problem requires assumptions about the structure

More information

EFFICIENT ATTRIBUTE REDUCTION ALGORITHM

EFFICIENT ATTRIBUTE REDUCTION ALGORITHM EFFICIENT ATTRIBUTE REDUCTION ALGORITHM Zhongzhi Shi, Shaohui Liu, Zheng Zheng Institute Of Computing Technology,Chinese Academy of Sciences, Beijing, China Abstract: Key words: Efficiency of algorithms

More information

Machine Learning A W 1sst KU. b) [1 P] Give an example for a probability distributions P (A, B, C) that disproves

Machine Learning A W 1sst KU. b) [1 P] Give an example for a probability distributions P (A, B, C) that disproves Machine Learning A 708.064 11W 1sst KU Exercises Problems marked with * are optional. 1 Conditional Independence I [2 P] a) [1 P] Give an example for a probability distribution P (A, B, C) that disproves

More information

Performance Study of Interweave Spectrum Sharing Method in Cognitive Radio

Performance Study of Interweave Spectrum Sharing Method in Cognitive Radio Performance Study of Interweave Spectrum Sharing Method in Cognitive Radio Abstract Spectrum scarcity is one of the most critical recent problems in the field of wireless communication One of the promising

More information

Hidden Markov Model for Sequential Data

Hidden Markov Model for Sequential Data Hidden Markov Model for Sequential Data Dr.-Ing. Michelle Karg mekarg@uwaterloo.ca Electrical and Computer Engineering Cheriton School of Computer Science Sequential Data Measurement of time series: Example:

More information

A MULTI-ROBOT SYSTEM FOR ASSEMBLY TASKS IN AUTOMOTIVE INDUSTRY

A MULTI-ROBOT SYSTEM FOR ASSEMBLY TASKS IN AUTOMOTIVE INDUSTRY The 4th International Conference Computational Mechanics and Virtual Engineering COMEC 2011 20-22 OCTOBER 2011, Brasov, Romania A MULTI-ROBOT SYSTEM FOR ASSEMBLY TASKS IN AUTOMOTIVE INDUSTRY A. Fratu 1

More information

SA-IFIM: Incrementally Mining Frequent Itemsets in Update Distorted Databases

SA-IFIM: Incrementally Mining Frequent Itemsets in Update Distorted Databases SA-IFIM: Incrementally Mining Frequent Itemsets in Update Distorted Databases Jinlong Wang, Congfu Xu, Hongwei Dan, and Yunhe Pan Institute of Artificial Intelligence, Zhejiang University Hangzhou, 310027,

More information

Markov Random Fields and Gibbs Sampling for Image Denoising

Markov Random Fields and Gibbs Sampling for Image Denoising Markov Random Fields and Gibbs Sampling for Image Denoising Chang Yue Electrical Engineering Stanford University changyue@stanfoed.edu Abstract This project applies Gibbs Sampling based on different Markov

More information

Analysis of Space-Ground Integrated Information Network Architecture and Protocol

Analysis of Space-Ground Integrated Information Network Architecture and Protocol 3rd International Conference on Mechatronics and Industrial Informatics (ICMII 2015) Analysis of Space-Ground Integrated Information Network Architecture and Protocol Yong Zhou 1,a*,Chundong She 2, b,ligang

More information

FEC Performance in Large File Transfer over Bursty Channels

FEC Performance in Large File Transfer over Bursty Channels FEC Performance in Large File Transfer over Bursty Channels Shuichiro Senda, Hiroyuki Masuyama, Shoji Kasahara and Yutaka Takahashi Graduate School of Informatics, Kyoto University, Kyoto 66-85, Japan

More information

A hierarchical network model for network topology design using genetic algorithm

A hierarchical network model for network topology design using genetic algorithm A hierarchical network model for network topology design using genetic algorithm Chunlin Wang 1, Ning Huang 1,a, Shuo Zhang 2, Yue Zhang 1 and Weiqiang Wu 1 1 School of Reliability and Systems Engineering,

More information

An Abnormal Data Detection Method Based on the Temporal-spatial Correlation in Wireless Sensor Networks

An Abnormal Data Detection Method Based on the Temporal-spatial Correlation in Wireless Sensor Networks An Based on the Temporal-spatial Correlation in Wireless Sensor Networks 1 Department of Computer Science & Technology, Harbin Institute of Technology at Weihai,Weihai, 264209, China E-mail: Liuyang322@hit.edu.cn

More information

Lecture 5: Markov models

Lecture 5: Markov models Master s course Bioinformatics Data Analysis and Tools Lecture 5: Markov models Centre for Integrative Bioinformatics Problem in biology Data and patterns are often not clear cut When we want to make a

More information

Adaptive Doppler centroid estimation algorithm of airborne SAR

Adaptive Doppler centroid estimation algorithm of airborne SAR Adaptive Doppler centroid estimation algorithm of airborne SAR Jian Yang 1,2a), Chang Liu 1, and Yanfei Wang 1 1 Institute of Electronics, Chinese Academy of Sciences 19 North Sihuan Road, Haidian, Beijing

More information

COS 513: Foundations of Probabilistic Modeling. Lecture 5

COS 513: Foundations of Probabilistic Modeling. Lecture 5 COS 513: Foundations of Probabilistic Modeling Young-suk Lee 1 Administrative Midterm report is due Oct. 29 th. Recitation is at 4:26pm in Friend 108. Lecture 5 R is a computer language for statistical

More information

Modified Metropolis-Hastings algorithm with delayed rejection

Modified Metropolis-Hastings algorithm with delayed rejection Modified Metropolis-Hastings algorithm with delayed reection K.M. Zuev & L.S. Katafygiotis Department of Civil Engineering, Hong Kong University of Science and Technology, Hong Kong, China ABSTRACT: The

More information

Learning the Three Factors of a Non-overlapping Multi-camera Network Topology

Learning the Three Factors of a Non-overlapping Multi-camera Network Topology Learning the Three Factors of a Non-overlapping Multi-camera Network Topology Xiaotang Chen, Kaiqi Huang, and Tieniu Tan National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy

More information

Sequential Dependency and Reliability Analysis of Embedded Systems. Yu Jiang Tsinghua university, Beijing, China

Sequential Dependency and Reliability Analysis of Embedded Systems. Yu Jiang Tsinghua university, Beijing, China Sequential Dependency and Reliability Analysis of Embedded Systems Yu Jiang Tsinghua university, Beijing, China outline Motivation Background Reliability Block Diagram, Fault Tree Bayesian Network, Dynamic

More information

Package HMMCont. February 19, 2015

Package HMMCont. February 19, 2015 Type Package Package HMMCont February 19, 2015 Title Hidden Markov Model for Continuous Observations Processes Version 1.0 Date 2014-02-11 Author Maintainer The package includes

More information

XI International PhD Workshop OWD 2009, October Fuzzy Sets as Metasets

XI International PhD Workshop OWD 2009, October Fuzzy Sets as Metasets XI International PhD Workshop OWD 2009, 17 20 October 2009 Fuzzy Sets as Metasets Bartłomiej Starosta, Polsko-Japońska WyŜsza Szkoła Technik Komputerowych (24.01.2008, prof. Witold Kosiński, Polsko-Japońska

More information

Skill. Robot/ Controller

Skill. Robot/ Controller Skill Acquisition from Human Demonstration Using a Hidden Markov Model G. E. Hovland, P. Sikka and B. J. McCarragher Department of Engineering Faculty of Engineering and Information Technology The Australian

More information

Stochastic Control of Path Optimization for Inter-Switch Handoffs in Wireless ATM Networks

Stochastic Control of Path Optimization for Inter-Switch Handoffs in Wireless ATM Networks 336 IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 9, NO. 3, JUNE 2001 Stochastic Control of Path Optimization for Inter-Switch Handoffs in Wireless ATM Networks Vincent W. S. Wong, Member, IEEE, Mark E. Lewis,

More information

CS Introduction to Data Mining Instructor: Abdullah Mueen

CS Introduction to Data Mining Instructor: Abdullah Mueen CS 591.03 Introduction to Data Mining Instructor: Abdullah Mueen LECTURE 8: ADVANCED CLUSTERING (FUZZY AND CO -CLUSTERING) Review: Basic Cluster Analysis Methods (Chap. 10) Cluster Analysis: Basic Concepts

More information