Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos 1 “Continuous All k Nearest Neighbor Queries in Smartphone Networks” Tuesday, July 24, th IEEE Int. Conference on Mobile Data Management (MDM’12), Bangalore, India Georgios Chatzimilioudis Demetrios Zeinalipour-Yazti Marios D. Dikaiakos Wang-Chien Lee

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Smartphone: A powerful sensing device! –Processing: 1.4 GHz quad core (Samsung Exynos) –RAM & Flash Storage: 2GB & 48GB, respectively –Networking: WiFi, 3G (Mbps) / 4G (100Mbps–1Gbps) –Sensing: Proximity, Ambient Light, Accelerometer, Microphone, Geographic Coordinates based on AGPS (fine), WiFi or Cellular Towers (coarse). In-House Applications! Smartphones SmartTrace (ICDE’09,MDM’09,TKDE’12) Airplace (MobiSys’12, MDM’12) SmartP2P (MDM’11, MDM’12) 2

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Crowdsourcing with Smartphones –A smartphone crowd is constantly moving and sensing providing large amounts of opportunistic data that enables new services and applications Motivation "Crowdsourcing with Smartphones", Georgios Chatzimiloudis, Andreas Konstantinides, Christos Laoudias, Demetrios Zeinalipour-Yazti, IEEE Internet Computing, Special Issue: Crowdsourcing (Sep/Oct 2012), accepted May IEEE Press,

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Continuous k Nearest Neighbor Queries Find 2 Closest Neighbors for 1 User

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Motivation 5 “Create a Framework for Efficient Proximity Interactions” Screenshots from a prototype system we've implemented for Windows Phone

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Continuous All k Nearest Neighbor Queries Find 2 Closest Neighbors for ALL User

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Motivation System Model and Problem Formulation Proximity Algorithm Experimental Evaluation Future Work 7 Presentation Outline

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos A set of users moving in the plane of a region (u 1, u 2, … u n ) Area covered by a set of Network Connectivity Points (NCP) –Each NCP creates the notion of a cell –W.l.o.g., let the cell be represented by a circular area with an arbitrary radius A mobile user u is serviced at any given time point by one NCP. There is some centralized service, denoted as QP (Query Processor), which is aware of the coverage of each NCP. Each user u reports its positional information to QP regularly 8 System Model C.u7.u7.u4.u4 u5.u5. u2.u2. u6.u6. u0.u0. u3.u3..u1.u1 Query Processor QP

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Query Processor Definition of the CAkNN problem: Given a set U of n users and their location reports r i,t ∈ R at timestep t ∈ T, then the CAkNN problem is to find in each timestep t ∈ T and for each user u i ∈ U the k objects U sol ⊆ U − u i such that for all other objects u o ∈ U − U sol − u i, dist(u k, u i ) ≤ dist(u o, u i ) holds 9 Problem Definition C.u7.u7.u4.u4 u5.u5. u2.u2. u6.u6. u0.u0. u3.u3..u1.u1

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Query Processor 10 Naïve Solution / Related Work Find 2-NN for u 0 at timestep t. C.u7.u7.u4.u4 u5.u5. u2.u2. u6.u6. u0.u0. u3.u3..u1.u1 For u 5 ? Look inside your cell! Perform iterative deepening! TOO EXPENSIVE! WRONG! (u1 closer)

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos 11 Related Work Existing algorithms for CkNN (not CAkNN) Yu et al. (YPK) 11 and Mouratidis et al. (CPM) 12 Stateless Version: Iteratively expand the search space for each user into neighboring cells to find the kNN (like previous slide) Stateful Version: Improve the stateless version by utilizing previous state, under the following assumptions: –i) Static Querying User (i.e., designated for point queries) –ii) Target users move slowly (i.e., state does not decay) –iii) Few Target users 11. Yu, Pu, Koudas. “Monitoring k-nearest neighbor queries over moving objects,” ICDE ’ Mouratidis, Papadias, Hadjieleftheriou, “Conceptual partitioning: an efficient method for continuous nearest neighbor monitoring,” SIGMOD ’05. Not Appropriate for CAkNN

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Motivation System Model and Problem Formulation Proximity Algorithm Experimental Evaluation Future Work 12 Presentation Outline

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Proximity Overview The first specialized algorithm for Continuous All k-NN (CAkNN) queries Important characteristics: –Stateless (i.e., optimized for high mobility) –Batch processing (i.e., with search space searching) –Parameter-free (i.e., no tuning parameters) Generic operator for proximity-based queries –See Crowdcast application presented later. 13

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Query Processor C.u7.u7.u4.u4 u5.u5. u2.u2. u6.u6. u0.u0. u3.u3..u1.u1 Proximity Outline For every timestep: 1.Initialize a k + -heap for every cell 2.Insert every user’s location report to every k + -heap Notice that k + -heap is a heap-based structure and most location reports will be dropped as a result of an insert operation 3.For every user scan the k + -heap of his cell to find his k-NN 14

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Query Processor C.u7.u7.u4.u4 u5.u5. u2.u2. u6.u6. u0.u0. u3.u3..u1.u1 Intuition behind Proximity Users in a cell will share the same search space (search space sharing) Compute 1 search space per cell only! 15 Search Space (contains the right answers for all users in C) Cell

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos The k + -heap structure for a cell Proximity k + -heap 16 All users inside the cell K nearest users outside the cell Beyond k nearest outside users (correctness) k + -heap structure K top-k heap B ordered list U list

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos k + -heap structure K top-k heap B ordered list U list k + -heap Construction (for Cell C) 17 Arriving Reports Structure U Structure K Structure B u6u6 u6u6 u4u4 u6u6 u4u4 u7u7 u6u6 u 7, u 4 u2u2 u6u6 u 4, u 2 u7u7 u3u3 u6u6 u 3, u 2 u 4, u 7 u1u1 u6u6 u 2, u 1 u 3, u 4 u5u5 u6u6 u 2, u 1 u 3, u 4, u 5 u0u0 u 6, u 0 u 2, u 1 u 3, u 4, u 5 Assume k=2. C.u7.u7.u4.u4 u5.u5. u2.u2. u6.u6. u0.u0. u3.u3..u1.u1

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos k + -heap structure K top-k heap B ordered list U list k + -heap Construction (for Cell C) 18 Arriving Reports Structure U Structure K Structure B u6u6 u6u6 u4u4 u6u6 u4u4 u7u7 u6u6 u 7, u 4 u2u2 u6u6 u 4, u 2 u7u7 u3u3 u6u6 u 3, u 2 u 4, u 7 u1u1 u6u6 u 2, u 1 u 3, u 4 u5u5 u6u6 u 2, u 1 u 3, u 4, u 5 u0u0 u 6, u 0 u 2, u 1 u 3, u 4, u 5 Assume k=2. C.u7.u7.u4.u4 u5.u5. u2.u2. u6.u6. u0.u0. u3.u3..u1.u1 2 nearest users to cell x.x. Diameter + d k

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Motivation System Model and Problem Formulation Proximity Algorithm Experimental Evaluation Future Work 19 Presentation Outline

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Experimental Methodology Dataset: –Oldenburg Dataset: Spatiotemporal generator with roadmap of Oldenburg as input (25km x 25km) 5000 maximum users (Oldenburg population: ) –Manhattan Dataset: Vehicular mobility generator with roadmap of Manhattan as input (3km x 3km) 500 users Network connectivity points (NCPs) uniformly in space –Communication ranges 1km, 4km and 16km for the Oldenburg dataset and ranges 1km and 4km for the Manhattan dataset Query: CAkNN Comparison: Proximity vs YPK vs CPM (using the optimal parameter value for YPK and CPM) 20

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Proximity - Experiments Bottleneck for Proximity: build time Bottleneck for the adapted YPK and CPM: search time 21

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Proximity - Experiments Benefit of Proximity in the search time Proximity is scalable with respect to k –Search space for Proximity is not proportional to k like YPK, CPM 22

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Proximity - Experiments Proximity: build time drops as k increases! Machine’s memory scheduler makes more efficient use of buffers when the search spaces are larger 23

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Proximity - Experiments Proximity scales with the number of users Proximity outperforms YPK, CPM by an order of magnitude Proximity does not converge to YPK, CPM for higher values of Nmax. 24

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Crowdcast Suite The Crowdcast Application Suite PROXIMITY WIN7 API … Ranked 3 rd in Microsoft’s ImagineCup contest local competition (to appear in the next demo session!)

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Crowdcast: MsgCast "Get your location-based questions answered!" MsgCast - Location-based Chat Channel (Post / Follow) - Provide Guidelines

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Crowdcast: HelpCast HelpCast "The Ubiquitous Help Platform for Anyone in Need”

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Crowdcast: EyeCast "Extend your Vision beyond your 2 eyes!" EyeCast - Disaster Recovery Operations - Indoor Video Conferencing Network

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Motivation System Model and Problem Formulation Proximity Algorithm Experimental Evaluation Future Work 29 Presentation Outline

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Proximity – Future Work Privacy extensions (e.g., spatial cloaking, strong privacy) Parallelizing server computation Proximity for cloud server User-defined k values 30

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Future Work: SmartLab 31 SmartLab.cs.ucy.ac.cy Install APK, Upload File, Reboot, Screenshots, Monkey Runners, etc.… Programming cloud for the development of smartphone network applications & protocols as well as experimentation with real smartphone devices. "Demo: A Programming Cloud of Smartphones", A. Konstantinidis, C. Costa, G. Larkou and D. Zeinalipour-Yazti, "Demo at the 10th International Conference on Mobile Systems Applications and Services" (Mobisys '12), Low Wood Bay Lake District UK, 2012

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Future Work: SmartLab 32

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos 33 “Continuous All k Nearest Neighbor Queries in Smartphone Networks” Tuesday, July 24, th IEEE Int. Conference on Mobile Data Management (MDM’12), Bangalore, India Georgios Chatzimilioudis Demetrios Zeinalipour-Yazti Marios D. Dikaiakos Wang-Chien Lee Thanks! Questions?

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos Crowdcast: Website

Dagstuhl Seminar 10042, Demetris Zeinalipour, University of Cyprus, 26/1/2010 MDM 2012 © Chatzimilioudis, Zeinalipour-Yazti, Lee, Dikaiakos k + -heap Insertion Algorithm 35 Input: new user ux 1.IF ux is inside the cell THEN insert in list U 2.ELSE IF ux is closer to the border of the cell than the head of the heap THEN 3. insert in heap K 4. IF K has more than k elements THEN 5. pop the user uh from the head of the heap 6. insert uh into the ordered list B 7. update candidate boundary 8. prune list B according to the new head of the heap 9. END IF 10.ELSE IF ux is closer to the boundary of the cell than candidate boundary THEN insert ux to ordered list B 11.ELSE discard ux END IF