SlideShare a Scribd company logo
1 of 29
Download to read offline
Institut Mines-Télécom
H. Xiong1, D. Zhang1, D. Zhang1,
V. Gauthier1, K. Yang2, M. Becker1
1.  Institut Mines-Telecom, TELECOM SudParis
2.  Network Convergence Laboratory, University of Essex
MPaaS: Mobility Prediction as
a Service of Telecom Cloud
Institut Mines-Télécom
Agenda
1.  Introduction
2.  MPaaS System Overview
3.  Problem Formulation; Empirical Observations on
Mobility Trajectories and MPaaS Prediction
Algorithms Design
4.  Evaluation Result
5.  Discussion
Institut Mines-Télécom
Objectives and Motivation
■ Overall Research Objectives:
●  In order to improve the performance of mobile services,
●  Predicting the future locations of each mobile user by
leveraging the telecommunication system and the
cloud computing facility.
■  Our research is motivated by three observations:
1.  Mobility prediction service
2.  Telecommunication System
3.  Telecommunication cloud
Institut Mines-Télécom
Observation 1
■ Mobility Prediction Service
● Predicting users’ future locations can help to
improve the performance of many mobile
services
− E.g.,: an user watching the online video in a fast-
moving train èpredicting the user’s next cell towers
èbuffering the video content on the next cell towers
in advance, in order to provide seamless handoff.
● Mobility prediction should be a fundamental
service for many mobile services.
Institut Mines-Télécom
Observation 2
■ Telecom System
● The telecommunication system is promising to
continuously track users’ location trajectories,
which could be used to predict users’ future
locations:
− When a user appears in a cell tower, the telecom
system would leave a signaling log (user id, time-
stamp) in the cell tower.
− Given the stream of signaling logs from each cell
tower, it is possible to track the continuous location
trajectory (e.g., cell tower id sequence) of each
mobile user
Institut Mines-Télécom
Observation 3
■ Telecom Cloud
●  Telcos in nowadays establish their own cloud computing
platform to facilitate the mobile service deployment.
●  The telecom cloud combining telecommunication/cloud
computing facilities:
−  Continuously tracks each mobile user’s location in real-time.
−  Provides the computational power to enable large-scale real-
time data mining/machine learning for mobility prediction.
Institut Mines-Télécom
Motivating Examples
•  Predictive Telecom Resource Management
•  Using users’ future locations to dynamically manage the resource
(e.g., power, bandwidth and storage) of communication systems
•  Mobility-based Service Personalization
•  Using users’ future locations to personalize the telecom services (e.g.,
location-aware recommendation, and location-based advertising )
Institut Mines-Télécom
Research Assumptions & Objectives
Given the stream of signaling logs real-timely-generated by
each cell tower; Given a set of mobile users;
For each mobile user, Predicting the cell towers that the user
would pass-by in next few hours of each mobile user.
Institut Mines-Télécom
Research Issues
■  Given the stream of signaling logs in each cell tower,
querying the cell traces of each individual mobile
user
■  Given the each user’s cell traces, extracting the
mobility trajectory (the sequence of locations)
■  Given each user’s historical trajectory, predicting the
user’s future locations
●  The low accuracy of predicting each user’s future locations
using the individual user’s mobility trajectory
●  Using the social interplay between mobile users to enhance
the mobility prediction.
Institut Mines-Télécom
Contributions
■  Investigating the research issues of mobility
prediction leveraging the telecom cloud; Proposing
MPaaS system.
■  Formulating the mobility prediction problem of the
MPaaS system; Proposing the mobility prediction
algorithm using the individual/collective mobility
patterns.
■  Evaluating the mobility prediction algorithms using
the real-world data traces.
Institut Mines-Télécom
Agenda
1.  Introduction
2.  MPaaS System Overview
3.  Problem Formulation; Empirical Observations on
Cellular Mobility Trace and MPaaS Prediction
Algorithms Design
4.  Evaluation Result
5.  Discussion
Institut Mines-Télécom
MPaaS System Overview
Signaling Logs of
each cell tower
Cellular traces
of each user
Mobility
Trajectory of each
user
Future Mobility Traje-
ctory of each user
Institut Mines-Télécom
Cellular Trace Query from Telecom Sys.
Querying each user’s cellular traces from each cell tower
using a Publisher/Subscriber system
Institut Mines-Télécom
Example of Cellular Traces
•  Ping-Pong Noise in
cellular traces (overlapped
area)
•  Example of cellular traces
Institut Mines-Télécom
Mapping Cellular Traces (with Ping-Pong
Noise) to Mobility Trajectory
■ Two Steps
●  Splitting the overlapped area among multiple cell
towers into the unoverlapped subregions
●  Mapping the cell tower id sequence (with ping-pong
noise) into the sequence of unoverlapped subregions
(mobility trajectories).
Institut Mines-Télécom
Mapping Cellular Traces to Mobility
Trajectory
■  Splitting the overlapped regions between cell towers
into subareas
E.g., three cell towers
A, B and C
1.  A,B,C
2.  AB, BC, AC
3.  ABC
Institut Mines-Télécom
Mapping Cellular Traces to Mobility
Trajectory
■  Given the cellular traces (with ping-pong noise) of
each individual user; e.g.,:
■  Finding the subregion where the user is most
likely to receive the cell tower id sequence
Institut Mines-Télécom
Agenda
1.  Introduction
2.  MPaaS System Overview
3.  Problem Formulation; Empirical Observations on
Cellular Mobility Trace; and MPaaS Prediction
Algorithms Design
4.  Evaluation Result
5.  Discussion
Institut Mines-Télécom
MPaaS Problem Formulation
Institut Mines-Télécom
Data Analysis
■  Before introducing our algorithms, we first pay a glance
on our mobility dataset
■  We select 13 users’ 3-month cellular traces from the
dataset, and extract the mobility trajectories (sequence
of regions) of these 13 users from the cellular traces
(using the aforementioned algorithm).
■  Putting 13 users’ trajectories together, we observe the
collective behavior patterns.
Institut Mines-Télécom
Data Analysis: Observing Collective
Behaviors in Cellular Mobility Traces
Users*regions
Time slots
e.g.,
Institut Mines-Télécom
MPaaS Predictor Design
•  Markov-based Predictor using each individual’s historical trajectories.
•  CBP-based Predictor using other users’ current locations by leveraging
collective behavior pattern mining.
Institut Mines-Télécom
MPaaS Prediction Algorithms—Markov-
based Predictor
■ Markov-based Predictor
●  Given the region sequence of each user’s historical
trajectory
●  Given the current region, and finding the list of neighboring
region (including current subregion)
●  For each neighboring region, Calculating the probability of
the user moving to the region using Markov-Chain and
historical region sequence
The highest probability region is the result
Institut Mines-Télécom
MPaaS Prediction Algorithms—CBP-
based Predictor
■ CBP-based Predictor
● Learning Phase
1.  Given the historical trajectories of all mobile users
2.  For each user and each time-slot (in history), learning
the association pattern from all rest users’ regions
in the time-slot to the user’s region in the next
time-slot.
● Predicting Phase
1.  Given a user, and the current regions of all users;
Finding the list of the user’s neighboring regions;
2.  For each neighboring region, Calculating the
probability of the user appearing in the region in
next time-slot.(using the association patterns)
3.  Finding the highest probability region as the
prediction result.
Institut Mines-Télécom
MPaaS Prediction Algorithms—Predictor
Fusion
■ Predictor Fusion
● Using DS-Evidence theory to fuse two
predictors;
− Combining the probability distributions
(probability on each region) from two
predictors è the joint probability
distribution
− Finding the highest joint probability region
as the prediction result
Institut Mines-Télécom
Agenda
1.  Introduction
2.  MPaaS System Overview
3.  Problem Formulation; Empirical Observations on
Cellular Mobility Trace; and MPaaS Prediction
Algorithms Design
4.  Evaluation Result
5.  Discussion
Institut Mines-Télécom
Evaluation
Institut Mines-Télécom
Discussion
■  The realistic deployment on the telecom cloud is
required.
●  Will deploy a prototype system in SFR network…
■  Need large dataset to evaluate the mobility prediction
algorithms
●  The current evaluation dataset is acquired from a small
group of mobile users, where collective behavioral
patterns could be easily observed/extracted.
Institut Mines-Télécom
Many Thanks!

More Related Content

What's hot

CCS SoftBank Tokyo Case Study
CCS SoftBank Tokyo Case StudyCCS SoftBank Tokyo Case Study
CCS SoftBank Tokyo Case StudyCCS
 
International Journal of Computational Engineering Research (IJCER)
International Journal of Computational Engineering Research (IJCER) International Journal of Computational Engineering Research (IJCER)
International Journal of Computational Engineering Research (IJCER) ijceronline
 
Mobile positioning for location dependent services in GSM networks
Mobile positioning for location dependent services in GSM networks Mobile positioning for location dependent services in GSM networks
Mobile positioning for location dependent services in GSM networks marwaeng
 
2 g&3g planning & optimization
2 g&3g planning & optimization 2 g&3g planning & optimization
2 g&3g planning & optimization Abdou Obado
 
Modulation classification of single input multiple-output signals using async...
Modulation classification of single input multiple-output signals using async...Modulation classification of single input multiple-output signals using async...
Modulation classification of single input multiple-output signals using async...jpstudcorner
 

What's hot (6)

Ab03301680176
Ab03301680176Ab03301680176
Ab03301680176
 
CCS SoftBank Tokyo Case Study
CCS SoftBank Tokyo Case StudyCCS SoftBank Tokyo Case Study
CCS SoftBank Tokyo Case Study
 
International Journal of Computational Engineering Research (IJCER)
International Journal of Computational Engineering Research (IJCER) International Journal of Computational Engineering Research (IJCER)
International Journal of Computational Engineering Research (IJCER)
 
Mobile positioning for location dependent services in GSM networks
Mobile positioning for location dependent services in GSM networks Mobile positioning for location dependent services in GSM networks
Mobile positioning for location dependent services in GSM networks
 
2 g&3g planning & optimization
2 g&3g planning & optimization 2 g&3g planning & optimization
2 g&3g planning & optimization
 
Modulation classification of single input multiple-output signals using async...
Modulation classification of single input multiple-output signals using async...Modulation classification of single input multiple-output signals using async...
Modulation classification of single input multiple-output signals using async...
 

Similar to Technical Note of EU FP7 MONICA Project: Mobility Prediction as a Service of Telecom Cloud

Mobility prediction in telecom cloud using telecom calls.
Mobility prediction in telecom cloud using telecom calls.Mobility prediction in telecom cloud using telecom calls.
Mobility prediction in telecom cloud using telecom calls.Afiya Rajee
 
IEEE 2014 DOTNET PARALLEL DISTRIBUTED PROJECTS Multicast capacity-in-manet-wi...
IEEE 2014 DOTNET PARALLEL DISTRIBUTED PROJECTS Multicast capacity-in-manet-wi...IEEE 2014 DOTNET PARALLEL DISTRIBUTED PROJECTS Multicast capacity-in-manet-wi...
IEEE 2014 DOTNET PARALLEL DISTRIBUTED PROJECTS Multicast capacity-in-manet-wi...IEEEMEMTECHSTUDENTPROJECTS
 
2014 IEEE DOTNET PARALLEL DISTRIBUTED PROJECT Multicast capacity-in-manet-wit...
2014 IEEE DOTNET PARALLEL DISTRIBUTED PROJECT Multicast capacity-in-manet-wit...2014 IEEE DOTNET PARALLEL DISTRIBUTED PROJECT Multicast capacity-in-manet-wit...
2014 IEEE DOTNET PARALLEL DISTRIBUTED PROJECT Multicast capacity-in-manet-wit...IEEEGLOBALSOFTSTUDENTSPROJECTS
 
Accurate and Energy-Efficient Range-Free Localization for Mobile Sensor Networks
Accurate and Energy-Efficient Range-Free Localization for Mobile Sensor NetworksAccurate and Energy-Efficient Range-Free Localization for Mobile Sensor Networks
Accurate and Energy-Efficient Range-Free Localization for Mobile Sensor Networksambitlick
 
multicast capacity in manet with infrastructure support
multicast capacity in manet with infrastructure supportmulticast capacity in manet with infrastructure support
multicast capacity in manet with infrastructure supportswathi78
 
enhanching rail worker safety: nrf based wireless sensing approach
enhanching rail worker safety: nrf based wireless sensing approachenhanching rail worker safety: nrf based wireless sensing approach
enhanching rail worker safety: nrf based wireless sensing approachMaSameer4
 
Claude perruche rev5 english
Claude perruche rev5 englishClaude perruche rev5 english
Claude perruche rev5 englishClaude Perruche
 
Claude perruche rev5 english
Claude perruche rev5 englishClaude perruche rev5 english
Claude perruche rev5 englishClaude Perruche
 
Training presentation
Training presentationTraining presentation
Training presentationSagnik Saha
 
Teletraffic Analysis of Overflowed Traffic with Voice only in Multilayer 3G W...
Teletraffic Analysis of Overflowed Traffic with Voice only in Multilayer 3G W...Teletraffic Analysis of Overflowed Traffic with Voice only in Multilayer 3G W...
Teletraffic Analysis of Overflowed Traffic with Voice only in Multilayer 3G W...IRJET Journal
 
a data mining approach for location production in mobile environments
a data mining approach for location production in mobile environments a data mining approach for location production in mobile environments
a data mining approach for location production in mobile environments marwaeng
 
JPD1428 Multicast Capacity in MANET with Infrastructure Support
JPD1428  Multicast Capacity in MANET with Infrastructure SupportJPD1428  Multicast Capacity in MANET with Infrastructure Support
JPD1428 Multicast Capacity in MANET with Infrastructure Supportchennaijp
 
Wap based seamless roaming in urban environment with wise handoff technique
Wap based seamless roaming in urban environment with wise handoff techniqueWap based seamless roaming in urban environment with wise handoff technique
Wap based seamless roaming in urban environment with wise handoff techniqueijujournal
 
On quality of monitoring for multi channel wireless
On quality of monitoring for multi channel wirelessOn quality of monitoring for multi channel wireless
On quality of monitoring for multi channel wirelessShakas Technologies
 
Mimo noma design for small packet transmission in the internet of things
Mimo noma design for small packet transmission in the internet of thingsMimo noma design for small packet transmission in the internet of things
Mimo noma design for small packet transmission in the internet of thingsredpel dot com
 
Urban Traffic Estimation & Optimization: An Overview
Urban Traffic Estimation & Optimization: An OverviewUrban Traffic Estimation & Optimization: An Overview
Urban Traffic Estimation & Optimization: An OverviewRakedet
 
Traffic models and estimation
Traffic models and estimation Traffic models and estimation
Traffic models and estimation Mina Yonan
 

Similar to Technical Note of EU FP7 MONICA Project: Mobility Prediction as a Service of Telecom Cloud (20)

Mobility prediction in telecom cloud using telecom calls.
Mobility prediction in telecom cloud using telecom calls.Mobility prediction in telecom cloud using telecom calls.
Mobility prediction in telecom cloud using telecom calls.
 
IEEE 2014 DOTNET PARALLEL DISTRIBUTED PROJECTS Multicast capacity-in-manet-wi...
IEEE 2014 DOTNET PARALLEL DISTRIBUTED PROJECTS Multicast capacity-in-manet-wi...IEEE 2014 DOTNET PARALLEL DISTRIBUTED PROJECTS Multicast capacity-in-manet-wi...
IEEE 2014 DOTNET PARALLEL DISTRIBUTED PROJECTS Multicast capacity-in-manet-wi...
 
2014 IEEE DOTNET PARALLEL DISTRIBUTED PROJECT Multicast capacity-in-manet-wit...
2014 IEEE DOTNET PARALLEL DISTRIBUTED PROJECT Multicast capacity-in-manet-wit...2014 IEEE DOTNET PARALLEL DISTRIBUTED PROJECT Multicast capacity-in-manet-wit...
2014 IEEE DOTNET PARALLEL DISTRIBUTED PROJECT Multicast capacity-in-manet-wit...
 
Accurate and Energy-Efficient Range-Free Localization for Mobile Sensor Networks
Accurate and Energy-Efficient Range-Free Localization for Mobile Sensor NetworksAccurate and Energy-Efficient Range-Free Localization for Mobile Sensor Networks
Accurate and Energy-Efficient Range-Free Localization for Mobile Sensor Networks
 
multicast capacity in manet with infrastructure support
multicast capacity in manet with infrastructure supportmulticast capacity in manet with infrastructure support
multicast capacity in manet with infrastructure support
 
enhanching rail worker safety: nrf based wireless sensing approach
enhanching rail worker safety: nrf based wireless sensing approachenhanching rail worker safety: nrf based wireless sensing approach
enhanching rail worker safety: nrf based wireless sensing approach
 
Claude perruche rev5 english
Claude perruche rev5 englishClaude perruche rev5 english
Claude perruche rev5 english
 
Claude perruche rev5 english
Claude perruche rev5 englishClaude perruche rev5 english
Claude perruche rev5 english
 
Training presentation
Training presentationTraining presentation
Training presentation
 
Teletraffic Analysis of Overflowed Traffic with Voice only in Multilayer 3G W...
Teletraffic Analysis of Overflowed Traffic with Voice only in Multilayer 3G W...Teletraffic Analysis of Overflowed Traffic with Voice only in Multilayer 3G W...
Teletraffic Analysis of Overflowed Traffic with Voice only in Multilayer 3G W...
 
a data mining approach for location production in mobile environments
a data mining approach for location production in mobile environments a data mining approach for location production in mobile environments
a data mining approach for location production in mobile environments
 
MU- mimo [autosaved]
MU- mimo [autosaved]MU- mimo [autosaved]
MU- mimo [autosaved]
 
JPD1428 Multicast Capacity in MANET with Infrastructure Support
JPD1428  Multicast Capacity in MANET with Infrastructure SupportJPD1428  Multicast Capacity in MANET with Infrastructure Support
JPD1428 Multicast Capacity in MANET with Infrastructure Support
 
Wap based seamless roaming in urban environment with wise handoff technique
Wap based seamless roaming in urban environment with wise handoff techniqueWap based seamless roaming in urban environment with wise handoff technique
Wap based seamless roaming in urban environment with wise handoff technique
 
Next-Generation Optical Access Architecture
Next-Generation Optical Access ArchitectureNext-Generation Optical Access Architecture
Next-Generation Optical Access Architecture
 
On quality of monitoring for multi channel wireless
On quality of monitoring for multi channel wirelessOn quality of monitoring for multi channel wireless
On quality of monitoring for multi channel wireless
 
Mimo noma design for small packet transmission in the internet of things
Mimo noma design for small packet transmission in the internet of thingsMimo noma design for small packet transmission in the internet of things
Mimo noma design for small packet transmission in the internet of things
 
B04920920
B04920920B04920920
B04920920
 
Urban Traffic Estimation & Optimization: An Overview
Urban Traffic Estimation & Optimization: An OverviewUrban Traffic Estimation & Optimization: An Overview
Urban Traffic Estimation & Optimization: An Overview
 
Traffic models and estimation
Traffic models and estimation Traffic models and estimation
Traffic models and estimation
 

Recently uploaded

Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfOrbitshub
 
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot ModelMcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot ModelDeepika Singh
 
Vector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptxVector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptxRemote DBA Services
 
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Bhuvaneswari Subramani
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Victor Rentea
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024The Digital Insurer
 
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...apidays
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Jeffrey Haguewood
 
Six Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal OntologySix Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal Ontologyjohnbeverley2021
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDropbox
 
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Victor Rentea
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...Zilliz
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfsudhanshuwaghmare1
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoffsammart93
 
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamUiPathCommunity
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodJuan lago vázquez
 
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusExploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusZilliz
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...DianaGray10
 

Recently uploaded (20)

Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
 
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot ModelMcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
Mcleodganj Call Girls 🥰 8617370543 Service Offer VIP Hot Model
 
Vector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptxVector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptx
 
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
 
Six Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal OntologySix Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal Ontology
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
 
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusExploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with Milvus
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 

Technical Note of EU FP7 MONICA Project: Mobility Prediction as a Service of Telecom Cloud

  • 1. Institut Mines-Télécom H. Xiong1, D. Zhang1, D. Zhang1, V. Gauthier1, K. Yang2, M. Becker1 1.  Institut Mines-Telecom, TELECOM SudParis 2.  Network Convergence Laboratory, University of Essex MPaaS: Mobility Prediction as a Service of Telecom Cloud
  • 2. Institut Mines-Télécom Agenda 1.  Introduction 2.  MPaaS System Overview 3.  Problem Formulation; Empirical Observations on Mobility Trajectories and MPaaS Prediction Algorithms Design 4.  Evaluation Result 5.  Discussion
  • 3. Institut Mines-Télécom Objectives and Motivation ■ Overall Research Objectives: ●  In order to improve the performance of mobile services, ●  Predicting the future locations of each mobile user by leveraging the telecommunication system and the cloud computing facility. ■  Our research is motivated by three observations: 1.  Mobility prediction service 2.  Telecommunication System 3.  Telecommunication cloud
  • 4. Institut Mines-Télécom Observation 1 ■ Mobility Prediction Service ● Predicting users’ future locations can help to improve the performance of many mobile services − E.g.,: an user watching the online video in a fast- moving train èpredicting the user’s next cell towers èbuffering the video content on the next cell towers in advance, in order to provide seamless handoff. ● Mobility prediction should be a fundamental service for many mobile services.
  • 5. Institut Mines-Télécom Observation 2 ■ Telecom System ● The telecommunication system is promising to continuously track users’ location trajectories, which could be used to predict users’ future locations: − When a user appears in a cell tower, the telecom system would leave a signaling log (user id, time- stamp) in the cell tower. − Given the stream of signaling logs from each cell tower, it is possible to track the continuous location trajectory (e.g., cell tower id sequence) of each mobile user
  • 6. Institut Mines-Télécom Observation 3 ■ Telecom Cloud ●  Telcos in nowadays establish their own cloud computing platform to facilitate the mobile service deployment. ●  The telecom cloud combining telecommunication/cloud computing facilities: −  Continuously tracks each mobile user’s location in real-time. −  Provides the computational power to enable large-scale real- time data mining/machine learning for mobility prediction.
  • 7. Institut Mines-Télécom Motivating Examples •  Predictive Telecom Resource Management •  Using users’ future locations to dynamically manage the resource (e.g., power, bandwidth and storage) of communication systems •  Mobility-based Service Personalization •  Using users’ future locations to personalize the telecom services (e.g., location-aware recommendation, and location-based advertising )
  • 8. Institut Mines-Télécom Research Assumptions & Objectives Given the stream of signaling logs real-timely-generated by each cell tower; Given a set of mobile users; For each mobile user, Predicting the cell towers that the user would pass-by in next few hours of each mobile user.
  • 9. Institut Mines-Télécom Research Issues ■  Given the stream of signaling logs in each cell tower, querying the cell traces of each individual mobile user ■  Given the each user’s cell traces, extracting the mobility trajectory (the sequence of locations) ■  Given each user’s historical trajectory, predicting the user’s future locations ●  The low accuracy of predicting each user’s future locations using the individual user’s mobility trajectory ●  Using the social interplay between mobile users to enhance the mobility prediction.
  • 10. Institut Mines-Télécom Contributions ■  Investigating the research issues of mobility prediction leveraging the telecom cloud; Proposing MPaaS system. ■  Formulating the mobility prediction problem of the MPaaS system; Proposing the mobility prediction algorithm using the individual/collective mobility patterns. ■  Evaluating the mobility prediction algorithms using the real-world data traces.
  • 11. Institut Mines-Télécom Agenda 1.  Introduction 2.  MPaaS System Overview 3.  Problem Formulation; Empirical Observations on Cellular Mobility Trace and MPaaS Prediction Algorithms Design 4.  Evaluation Result 5.  Discussion
  • 12. Institut Mines-Télécom MPaaS System Overview Signaling Logs of each cell tower Cellular traces of each user Mobility Trajectory of each user Future Mobility Traje- ctory of each user
  • 13. Institut Mines-Télécom Cellular Trace Query from Telecom Sys. Querying each user’s cellular traces from each cell tower using a Publisher/Subscriber system
  • 14. Institut Mines-Télécom Example of Cellular Traces •  Ping-Pong Noise in cellular traces (overlapped area) •  Example of cellular traces
  • 15. Institut Mines-Télécom Mapping Cellular Traces (with Ping-Pong Noise) to Mobility Trajectory ■ Two Steps ●  Splitting the overlapped area among multiple cell towers into the unoverlapped subregions ●  Mapping the cell tower id sequence (with ping-pong noise) into the sequence of unoverlapped subregions (mobility trajectories).
  • 16. Institut Mines-Télécom Mapping Cellular Traces to Mobility Trajectory ■  Splitting the overlapped regions between cell towers into subareas E.g., three cell towers A, B and C 1.  A,B,C 2.  AB, BC, AC 3.  ABC
  • 17. Institut Mines-Télécom Mapping Cellular Traces to Mobility Trajectory ■  Given the cellular traces (with ping-pong noise) of each individual user; e.g.,: ■  Finding the subregion where the user is most likely to receive the cell tower id sequence
  • 18. Institut Mines-Télécom Agenda 1.  Introduction 2.  MPaaS System Overview 3.  Problem Formulation; Empirical Observations on Cellular Mobility Trace; and MPaaS Prediction Algorithms Design 4.  Evaluation Result 5.  Discussion
  • 20. Institut Mines-Télécom Data Analysis ■  Before introducing our algorithms, we first pay a glance on our mobility dataset ■  We select 13 users’ 3-month cellular traces from the dataset, and extract the mobility trajectories (sequence of regions) of these 13 users from the cellular traces (using the aforementioned algorithm). ■  Putting 13 users’ trajectories together, we observe the collective behavior patterns.
  • 21. Institut Mines-Télécom Data Analysis: Observing Collective Behaviors in Cellular Mobility Traces Users*regions Time slots e.g.,
  • 22. Institut Mines-Télécom MPaaS Predictor Design •  Markov-based Predictor using each individual’s historical trajectories. •  CBP-based Predictor using other users’ current locations by leveraging collective behavior pattern mining.
  • 23. Institut Mines-Télécom MPaaS Prediction Algorithms—Markov- based Predictor ■ Markov-based Predictor ●  Given the region sequence of each user’s historical trajectory ●  Given the current region, and finding the list of neighboring region (including current subregion) ●  For each neighboring region, Calculating the probability of the user moving to the region using Markov-Chain and historical region sequence The highest probability region is the result
  • 24. Institut Mines-Télécom MPaaS Prediction Algorithms—CBP- based Predictor ■ CBP-based Predictor ● Learning Phase 1.  Given the historical trajectories of all mobile users 2.  For each user and each time-slot (in history), learning the association pattern from all rest users’ regions in the time-slot to the user’s region in the next time-slot. ● Predicting Phase 1.  Given a user, and the current regions of all users; Finding the list of the user’s neighboring regions; 2.  For each neighboring region, Calculating the probability of the user appearing in the region in next time-slot.(using the association patterns) 3.  Finding the highest probability region as the prediction result.
  • 25. Institut Mines-Télécom MPaaS Prediction Algorithms—Predictor Fusion ■ Predictor Fusion ● Using DS-Evidence theory to fuse two predictors; − Combining the probability distributions (probability on each region) from two predictors è the joint probability distribution − Finding the highest joint probability region as the prediction result
  • 26. Institut Mines-Télécom Agenda 1.  Introduction 2.  MPaaS System Overview 3.  Problem Formulation; Empirical Observations on Cellular Mobility Trace; and MPaaS Prediction Algorithms Design 4.  Evaluation Result 5.  Discussion
  • 28. Institut Mines-Télécom Discussion ■  The realistic deployment on the telecom cloud is required. ●  Will deploy a prototype system in SFR network… ■  Need large dataset to evaluate the mobility prediction algorithms ●  The current evaluation dataset is acquired from a small group of mobile users, where collective behavioral patterns could be easily observed/extracted.