Desiging a Virtual Information Telescope using Mobile Phones and Social Participation Romit Roy Choudhury Asst.
Download ReportTranscript Desiging a Virtual Information Telescope using Mobile Phones and Social Participation Romit Roy Choudhury Asst.
Desiging a Virtual Information Telescope using Mobile Phones and Social Participation Romit Roy Choudhury Asst. Prof. (Duke University) 1 Virtual Information Telescope 2 Context Next generation mobile phones will have large number of sensors Cameras, microphones, accelerometers, GPS, compasses, health monitors, … 3 Context Each phone may be viewed as a micro lens Exposing a micro view of the physical world to the Internet 4 Context With 3 billion active phones in the world today (the fastest growing comuting platform …) Our Vision is … 5 A Virtual Information Telescope Internet 6 One instantiation of this vision through a system called Micro-Blog - Content sharing - Content querying - Content floating 7 Content Sharing Web Service Virtual Telescope Cellular, WiFi Visualization Service People Phones Physical Space 8 Content Querying Web Service Virtual Telescope Cellular, WiFi Visualization Service People Phones Physical Space Some queries participatory Is beach parking available? Others are not Is there WiFi at the beach café? 9 Content Floating [on physical space] superb sushi Safe@ Nite? 10 If designed carefully, a variety of applications may emerge on Micro-Blog 11 Applications Tourism View multimedia blogs … query for specifics Micro Reporters News service with feeds from individuals On-the-fly Ride Sharing Ride givers advertize intension w/ space-time sticky notes Respond to sticky notes once you arrive there Virtual order on physical disorder Land in a new place, and get step by step information RSS Feeds on Location Inform me when a live band is playing at the mall 12 MiroBlog Prototype Nokia N95 phones Symbian platform Carbide C++ code 13 Micro-Blog Beta live at http://synrg.ee.duke.edu/microblog.html 14 Prototype 15 Case Studies Micro-Blog phones distributed to volunteers 12 volunteers • 4 phones in 3 rounds • 3 weeks Not great UI • Basic training for users Exit interview revealed useful observations 16 From Exit Interview 1. “Fun activity” for free time Needs much “cooler GUI” 2. Privacy control vital, don’t care about incentives “more interesting to reply to questions … interested in knowing who is asking …” 3. Voice is personal, text is impersonal “Easier to correct text … audio blogs easier but …” 4. Logs show most blogs between 5:00 to 9:00pm Probably better for battery usage as well 17 Thoughts Micro-Blog: Rich space for applications and services But where exactly is the research here ???!!** 18 Problem I Energy Efficient Localization (EnLoc) 19 To GPS or not to GPS GPS is popular localization scheme Good error characteristics ~ 10m Apps naturally assume GPS Shockingly, first Micro-Blog demo lasted < 10 hours 20 Cost of Localization Performed extensive measurements GPS consumes 400 mW, AGPS marginally better Idle power consumption 55 mW 21 Alternate Localization WiFi fingerprinting, GSM triangulation Place Lab, SkyHook … Improved energy savings WiFi 20 hours GSM 40 hours At the cost of accuracy 40m + 200m + 22 Tradeoff Summary: 20 40 200 Research Question: Can we achieve the best of both worlds 23 Formulation L(t0) L(t2) L(t3) L(t4) L(t1) L(t5) L(t6) L(t7) Accuracy gain from GPS Error Accuracy gain from WiFi t0 t1 t2 t3 t4 GPS t5 t6 WiFi t7 Given energy budget, E, Trace T, and location reading costs, egps , ewifi , egsm : Schedule location readings to minimize avg. error 24 Dynamic Program Minimize the area under the curve By cutting the curve at appropriate points Number of (GPS + WiFi + GSM) cuts must cost < budget 25 Offline optimal offers lower bound on error Online algorithm necessary Online optimal difficult Need to design heuristics 26 Our Approach Do not invest energy if you can predict (even partially) 27 Predictive Heuristics Prediction opportunities exist Human users are not in brownian motion (exploit inertia) Exploit habitual mobility patterns Population distribution can be leveraged Prediction also incorporated into Dynamic Program Optimal computed on a given predictor Error Prediction generates different error curve t0 t1 t2 t3 t4 t5 t6 t6 28 Mobility Profiling Build logical mobility tree per-user Each link an uncertainty point (UP) Sample location only when uncertain Location predictable between UPs Home 8:00 8:15 8:30 12:00 Road crossing 8:05 12:05 Gym Exploit acclerometers Predict traffic turns Periodically localize to reset errors Office 3:30 5:30 6:00 Library 6:00 Grocery 29 Population Statistics Humans may deviate from mobility profile Predict based on population statistics Goodwin & Green U-Turn Straight Right Left E on Green 0 0.881 0.039 0.078 W on Green 0 0 0.596 0.403 N on Goodwin 0 0.640 0.359 0 S on Goodwin 0 0.513 0 0.486 30 Buy Accuracy with Energy Comparison of optimal with simple interpolation GPS clearly not the right choice 31 Thoughts Localization cannot be taken for granted Critical tradeoff between energy and accuracy Substantial room for saving energy While sustaining reasonably good accuracy However, physical localization May not be the way to go … Several motivations to pursue symbolic localization 32 Questions? 33 Problem 2 Symbolic localization (SurroundSense) 34 Symbolic Localization Services may not care about physical location Symbolic location often sufficient E.g., coffee shop, movie, park, in-car … Physical to Symbolic conversion Lookup location name based on GPS coordinate However, risky Starbucks RadioShack GPS Error range 35 Hypothesis Its possible to localize phones by sensing the ambience such as sound, light, color, movement, orientation… 36 SurroundSense Develop multi-modal fingerprint Using ambient sound/light/color/movement etc. Starbucks QuickTime™ and a TIFF (Uncompressed) decompressor are needed to see this picture. RadioShack Wall SurroundSense Server 37 SurroundSense Each individual sensor not discriminating enough Together, they are quite unique Use Support Vector Machines to identify uniqueness Location Classification Algorithm (SVM) Fingerprint Database 38 Should Ambiences be Unique Worldwide? GSM provides macro location (mall) SurroundSense refines to Starbucks B A C E D 39 Why will it work? The Intuition: Economics forces nearby businesses to be different Not profitable to have 5 chinese restaurants with same lighting, music, color, layout, etc. SurroundSense exploits this ambience diversity 40 Fingerprints Sound: Color: 41 Fingerprints Light: Movement: 42 Ambience Fingerprinting QuickTim e™ and a TIFF(U ncompressed) decom press or are needed to see this pi cture. Sound Color/Light Quic kTime™ a nd a TIFF (Un co mp res sed ) d ec omp re sso r ar e n eed ed to see thi s p ictu re. Fingerprint Filtering & Matching Test Fingerprint + Compass = RF/Acc. QuickTime™ and a TIFF (Uncompressed) decompressor are needed to see this picture. Macro Location Logical Location Fingerprint Database Candidate Fingerprints 43 Full System on Nokia N95 Experimented on 58 stores 10 different clusters Different parts of Duke campus and in Durham city 44 Full System on Nokia N95 Some classifications were incorrect But we wanted to know how much incorrect? We plotted Top-K accuracy Top-3 accuracy proved to be 100% for all stores 45 Issues and Opportunity Cameras may be inside pockets Now, we detect when its taken out Activate cameras, and take pictures Future phones will be flexible (wrist watch) - see Nokia Morph Electroic compasses can fingerprint layout Tables and shelves laid out in different orientations Users forced to orient in those ways Quick Time™ and a TIFF ( Uncompr ess ed) decompr ess or ar e needed to s ee this pic ture. QuickTime™ and a TIFF (Uncompressed) decompressor are needed to see this picture. a d na ™emi Tkci uQ ro sser pmo ced ) des serp mocn U( F FIT .erut cip s iht e es ot ded een era QuickTime™ and a TIFF (Uncompressed) decompressor are needed to see this picture. 46 Summary Ambience can be a great clue about location Ambient Sound, light, color, movement … None of the individual sensors good enough Combined they may be unique Uniqueness facilitated by economic incentive Businesses benefit if they are mutually diverse in ambience Ambience diversity helps SurroundSense 47 Conclusion The Virtual Information Telescope A generalization of mobile, location based, social computing Just developing apps Not enough Internet Many challenges Energy Localization Privacy Incentives, data distillation … 48 Conclusion Project Micro-Blog Addressing the challenges systematically Building a fully functional system with applications The project snapshot as of today, includes: Micro-Blog: Overall system and application EnLoc: Energy Efficient Localization SurroundSense: Context aware localization CacheCloak: Location privacy via mobility prediction 49 PhonePoint Pens Using phone accelerometers To write short messages in the air 50 Please stay tuned for more at http://synrg.ee.duke.edu Thank You 51 Several research challenges and opportunities 1. Energy-efficient localization 2. Symbolic localization through ambience sensing 3. Location privacy 4. 5. 6. 7. Our Research Incentives Spam Information distillation User Inerfacing … 52 Disclaimer All of our projects are ongoing, hence not fully mature Today’s talk more about the problems than about solutions 53 Today’s Talk Information Telescope Vision System and Challenges/Opporunities Applications Ongoing, Future Work 1. EnLoc 2. SurroundSense 3. CacheCloak 54