SlideShare a Scribd company logo
1 of 26
ENDA: Embracing Network Inconsistency for
Dynamic Application Offloading
in Mobile Cloud Computing
Jiwei Li Kai Bu Xuan Liu Bin Xiao
The Hong Kong Polytechnic University
Presenter: Jiwei Li
Outline
• Background & Problem
• Proposed Solution
• Preliminary Results
• Conclusion
Outline
• Background & Problem
• Proposed Solution
• Preliminary Results
• Conclusion
Mobile Cloud Computing
• Applications
– Apple’s iCloud, Dropbox
• Technical problems
– MCC architecture & infrastructure
– Network connectivity
– Energy efficiency
• One important research topic - offloading
Offloading Strategy
• Previous work
– MAUI, CloneCloud, Odessa, COMET
Offload to Cloud
Compute-intensive
applications
Computed results
High latency (100-300ms)
Limited bandwidth (386 Kbs to 3.6 Mbs)
High energy consumption
Nearly unlimited resources
Offload to Cloudlet
Compute-intensive
applications
Computed results
Low latency (23-50ms)
High bandwidth (54 Mbs)
Limited coverage of Wi-Fi (20-100m)
Resource constraint
Uninvestigated Issues in Offloading
• Offloading at mobile environments
• Balancing workloads among multiple cloudlets
Our research is focused on
offloading to cloudlets through Wi-Fi
at mobile environments.
A B
C
D
A Motivating Example
Re-connection Matters
• Re-connection includes
– Scanning
– Connecting
– Assigning IP and network ID
• Takes long time (1-12s)
• Consumes additional power
Reducing re-connection times means increasing energy efficiency.
Our Studied Problems
• How to predict user’s trajectory?
• How to select Wi-Fi access points (AP)?
• How to balance workload among cloudlets?
Problem Formulation
• Minimize:
– Communication overheads during offloading at
mobile environments
• Must satisfy requirements:
– App-specific network latency
– App-specific response time
To put it simply, we aim to
select the most energy-efficient Wi-Fi access point,
taking user mobility and server load into account.
Outline
• Background & Problem
• Proposed Solution
• Preliminary Results
• Conclusion
Answering a few questions …
• Is it feasible to deploy cloudlets at large scale?
• Bind current public Wi-Fi hotspots with cloudlets.
• How do we overcome resource constraints on
cloudlets?
• Adopt workload balance management mechanism among
participating cloudlets.
• How do we conquer Wi-Fi’s limited coverage range
issue?
• Propose mobility-aware Wi-Fi AP selection scheme.
A Real Scenario
ENDA
• Three-tier architecture Design
– Cloud
– Cloudlet
– Smartphone
• Objective:
– Make the most energy efficient offloading
decision
Clouds
CloudletsSmartphones
VM on cloudlets
Profilers
Wi-Fi
2G/3G
WAN
User Track
Prediction
Wi-Fi AP
Distribution
and Status
Wi-Fi AP
Selector
GPS
Runtime System
Wi-Fi
Adapter
FINAL DECISION
OFFLOADING
INPUT INPUTREPORT
REPORT
APPINFO
Our work will be focused on
Advantages
• Minimize end-to-end communication
overheads
• Exempt smartphones from complex
computation of making decisions
• Improve energy efficiency for offloading
Demo Scenario
Predicted user track
(will be pruned based on
app info & network conditions)
Effective routes:
N1 -> (S, A)
N2 -> (S, B)
N3 -> (S, D)
N4 -> (C, D)
ENDA chooses the most energy-efficient Wi-Fi AP
according to the specific predicted track
Start offloading at location S
Outline
• Background & Problem
• Proposed Solution
• Preliminary Results
• Conclusion
GUI-based Simulation
Add routers
Add walking path
Calculate effective path
Simulation Results
Wi-Fi B Wi-Fi A Wi-Fi C
Outline
• Background & Problem
• Proposed Solution
• Preliminary Results
• Conclusion
Conclusion
• ENDA
– Difference from previous work
– Minimize communication overheads
– Potential to apply to real offloading systems
• Future work
– Thorough mathematical analysis
– Implementation
– More complex scenarios
Thank you!
Q&A

More Related Content

What's hot

Optimising Service Deployment and Infrastructure Resource Configuration
Optimising Service Deployment and Infrastructure Resource ConfigurationOptimising Service Deployment and Infrastructure Resource Configuration
Optimising Service Deployment and Infrastructure Resource ConfigurationRECAP Project
 
A view of Cloud Computing
A view of Cloud ComputingA view of Cloud Computing
A view of Cloud ComputingAsli Yazagan
 
RECAP at ETSI Experiential Network Intelligence (ENI) Meeting
RECAP at ETSI Experiential Network Intelligence (ENI) MeetingRECAP at ETSI Experiential Network Intelligence (ENI) Meeting
RECAP at ETSI Experiential Network Intelligence (ENI) MeetingRECAP Project
 
Extending Network Virtualization into the Optical Domain
Extending Network Virtualization into the Optical DomainExtending Network Virtualization into the Optical Domain
Extending Network Virtualization into the Optical DomainADVA
 
Getting the hybrid cloud right the first time
Getting the hybrid cloud right the first timeGetting the hybrid cloud right the first time
Getting the hybrid cloud right the first timeDavid Linthicum
 
RECAP: The Simulation Approach
RECAP: The Simulation ApproachRECAP: The Simulation Approach
RECAP: The Simulation ApproachRECAP Project
 
The Science DMZ
The Science DMZThe Science DMZ
The Science DMZJisc
 
Unlock your power with 4 dimensional actionable intelligence
Unlock your power with 4 dimensional actionable intelligence Unlock your power with 4 dimensional actionable intelligence
Unlock your power with 4 dimensional actionable intelligence BSP Media Group
 
Green Cloud Computing
Green Cloud ComputingGreen Cloud Computing
Green Cloud ComputingSabiha M
 
To minimize energy consumption in virtualization based on a computing cloud
To minimize energy consumption in virtualization based on a computing cloudTo minimize energy consumption in virtualization based on a computing cloud
To minimize energy consumption in virtualization based on a computing cloudArumugam Reddy
 
A Modelling Language for Defining Cloud Simulation Scenarios in RECAP Project...
A Modelling Language for Defining Cloud Simulation Scenarios in RECAP Project...A Modelling Language for Defining Cloud Simulation Scenarios in RECAP Project...
A Modelling Language for Defining Cloud Simulation Scenarios in RECAP Project...RECAP Project
 
“A Practical Guide to Implementing ML on Embedded Devices,” a Presentation fr...
“A Practical Guide to Implementing ML on Embedded Devices,” a Presentation fr...“A Practical Guide to Implementing ML on Embedded Devices,” a Presentation fr...
“A Practical Guide to Implementing ML on Embedded Devices,” a Presentation fr...Edge AI and Vision Alliance
 
Cloud Computing in the Enterprise
Cloud Computing in the EnterpriseCloud Computing in the Enterprise
Cloud Computing in the EnterpriseDavid Ramirez
 
COMIT Community Day Autumnn 2018 - Trellisworks ipcctv
COMIT Community Day Autumnn 2018 - Trellisworks ipcctvCOMIT Community Day Autumnn 2018 - Trellisworks ipcctv
COMIT Community Day Autumnn 2018 - Trellisworks ipcctvComit Projects Ltd
 
Energy efficient computing & computational services
Energy efficient computing & computational services Energy efficient computing & computational services
Energy efficient computing & computational services David Wallom
 
Cloud Modeling vs Internal vs Global Market using Burstorm Platform
Cloud Modeling vs Internal vs Global Market using Burstorm PlatformCloud Modeling vs Internal vs Global Market using Burstorm Platform
Cloud Modeling vs Internal vs Global Market using Burstorm PlatformScott Goessling
 
Cloud computing & energy efficiency using cloud to decrease the energy use in...
Cloud computing & energy efficiency using cloud to decrease the energy use in...Cloud computing & energy efficiency using cloud to decrease the energy use in...
Cloud computing & energy efficiency using cloud to decrease the energy use in...Puru Agrawal
 
Resource Management in Cloud Computing
Resource Management in Cloud ComputingResource Management in Cloud Computing
Resource Management in Cloud ComputingCristian Klein
 

What's hot (20)

Optimising Service Deployment and Infrastructure Resource Configuration
Optimising Service Deployment and Infrastructure Resource ConfigurationOptimising Service Deployment and Infrastructure Resource Configuration
Optimising Service Deployment and Infrastructure Resource Configuration
 
A view of Cloud Computing
A view of Cloud ComputingA view of Cloud Computing
A view of Cloud Computing
 
RECAP at ETSI Experiential Network Intelligence (ENI) Meeting
RECAP at ETSI Experiential Network Intelligence (ENI) MeetingRECAP at ETSI Experiential Network Intelligence (ENI) Meeting
RECAP at ETSI Experiential Network Intelligence (ENI) Meeting
 
Extending Network Virtualization into the Optical Domain
Extending Network Virtualization into the Optical DomainExtending Network Virtualization into the Optical Domain
Extending Network Virtualization into the Optical Domain
 
Getting the hybrid cloud right the first time
Getting the hybrid cloud right the first timeGetting the hybrid cloud right the first time
Getting the hybrid cloud right the first time
 
RECAP: The Simulation Approach
RECAP: The Simulation ApproachRECAP: The Simulation Approach
RECAP: The Simulation Approach
 
The Science DMZ
The Science DMZThe Science DMZ
The Science DMZ
 
Unlock your power with 4 dimensional actionable intelligence
Unlock your power with 4 dimensional actionable intelligence Unlock your power with 4 dimensional actionable intelligence
Unlock your power with 4 dimensional actionable intelligence
 
Green Cloud Computing
Green Cloud ComputingGreen Cloud Computing
Green Cloud Computing
 
To minimize energy consumption in virtualization based on a computing cloud
To minimize energy consumption in virtualization based on a computing cloudTo minimize energy consumption in virtualization based on a computing cloud
To minimize energy consumption in virtualization based on a computing cloud
 
A Modelling Language for Defining Cloud Simulation Scenarios in RECAP Project...
A Modelling Language for Defining Cloud Simulation Scenarios in RECAP Project...A Modelling Language for Defining Cloud Simulation Scenarios in RECAP Project...
A Modelling Language for Defining Cloud Simulation Scenarios in RECAP Project...
 
“A Practical Guide to Implementing ML on Embedded Devices,” a Presentation fr...
“A Practical Guide to Implementing ML on Embedded Devices,” a Presentation fr...“A Practical Guide to Implementing ML on Embedded Devices,” a Presentation fr...
“A Practical Guide to Implementing ML on Embedded Devices,” a Presentation fr...
 
Cloud Computing in the Enterprise
Cloud Computing in the EnterpriseCloud Computing in the Enterprise
Cloud Computing in the Enterprise
 
COMIT Community Day Autumnn 2018 - Trellisworks ipcctv
COMIT Community Day Autumnn 2018 - Trellisworks ipcctvCOMIT Community Day Autumnn 2018 - Trellisworks ipcctv
COMIT Community Day Autumnn 2018 - Trellisworks ipcctv
 
Energy efficient computing & computational services
Energy efficient computing & computational services Energy efficient computing & computational services
Energy efficient computing & computational services
 
Cloud Modeling vs Internal vs Global Market using Burstorm Platform
Cloud Modeling vs Internal vs Global Market using Burstorm PlatformCloud Modeling vs Internal vs Global Market using Burstorm Platform
Cloud Modeling vs Internal vs Global Market using Burstorm Platform
 
High–Performance Computing
High–Performance ComputingHigh–Performance Computing
High–Performance Computing
 
Cloud computing & energy efficiency using cloud to decrease the energy use in...
Cloud computing & energy efficiency using cloud to decrease the energy use in...Cloud computing & energy efficiency using cloud to decrease the energy use in...
Cloud computing & energy efficiency using cloud to decrease the energy use in...
 
Chap 1(one) general introduction
Chap 1(one)  general introductionChap 1(one)  general introduction
Chap 1(one) general introduction
 
Resource Management in Cloud Computing
Resource Management in Cloud ComputingResource Management in Cloud Computing
Resource Management in Cloud Computing
 

Viewers also liked

MY CONTINUATION RESEARCH 1
MY CONTINUATION RESEARCH 1MY CONTINUATION RESEARCH 1
MY CONTINUATION RESEARCH 1Binod Panthee
 
Agenda Cultural de Valledupar : Abril 2014
Agenda Cultural de Valledupar : Abril 2014Agenda Cultural de Valledupar : Abril 2014
Agenda Cultural de Valledupar : Abril 2014aracataka
 
組織を変革する(2016年度SPH授業)
組織を変革する(2016年度SPH授業)組織を変革する(2016年度SPH授業)
組織を変革する(2016年度SPH授業)Daisuke Son
 
Recomendaciones sobre evaluación de desempeño
Recomendaciones sobre evaluación de desempeñoRecomendaciones sobre evaluación de desempeño
Recomendaciones sobre evaluación de desempeñoJesus Villa
 
The role of the new ISO 9001:2015 leadership requirements in companies
The role of the new ISO 9001:2015 leadership requirements in companiesThe role of the new ISO 9001:2015 leadership requirements in companies
The role of the new ISO 9001:2015 leadership requirements in companiesPECB
 
Quick guide to ISO9001:2015
Quick guide to ISO9001:2015Quick guide to ISO9001:2015
Quick guide to ISO9001:2015Ellen Willoughby
 
Chris Argyris
Chris ArgyrisChris Argyris
Chris Argyrismrnelson
 

Viewers also liked (12)

MY CONTINUATION RESEARCH 1
MY CONTINUATION RESEARCH 1MY CONTINUATION RESEARCH 1
MY CONTINUATION RESEARCH 1
 
2016ls1 day9&10
2016ls1 day9&102016ls1 day9&10
2016ls1 day9&10
 
TRI and OPRA Overview
TRI and OPRA OverviewTRI and OPRA Overview
TRI and OPRA Overview
 
2016ls1 day5&6
2016ls1 day5&62016ls1 day5&6
2016ls1 day5&6
 
Admin interface login
Admin interface loginAdmin interface login
Admin interface login
 
Agenda Cultural de Valledupar : Abril 2014
Agenda Cultural de Valledupar : Abril 2014Agenda Cultural de Valledupar : Abril 2014
Agenda Cultural de Valledupar : Abril 2014
 
Sectional delegate
Sectional delegateSectional delegate
Sectional delegate
 
組織を変革する(2016年度SPH授業)
組織を変革する(2016年度SPH授業)組織を変革する(2016年度SPH授業)
組織を変革する(2016年度SPH授業)
 
Recomendaciones sobre evaluación de desempeño
Recomendaciones sobre evaluación de desempeñoRecomendaciones sobre evaluación de desempeño
Recomendaciones sobre evaluación de desempeño
 
The role of the new ISO 9001:2015 leadership requirements in companies
The role of the new ISO 9001:2015 leadership requirements in companiesThe role of the new ISO 9001:2015 leadership requirements in companies
The role of the new ISO 9001:2015 leadership requirements in companies
 
Quick guide to ISO9001:2015
Quick guide to ISO9001:2015Quick guide to ISO9001:2015
Quick guide to ISO9001:2015
 
Chris Argyris
Chris ArgyrisChris Argyris
Chris Argyris
 

Similar to Embracing Network Inconsistency for Dynamic Application Offloading in Mobile Cloud Computing

Towards the extinction of mega data centres? To which extent should the Clou...
 Towards the extinction of mega data centres? To which extent should the Clou... Towards the extinction of mega data centres? To which extent should the Clou...
Towards the extinction of mega data centres? To which extent should the Clou...Thierry Coupaye
 
Networking Challenges for the Next Decade
Networking Challenges for the Next DecadeNetworking Challenges for the Next Decade
Networking Challenges for the Next DecadeOpen Networking Summit
 
Research portfolio
Research portfolio Research portfolio
Research portfolio Mehdi Bennis
 
(R)evolution of the computing continuum - A few challenges
(R)evolution of the computing continuum  - A few challenges(R)evolution of the computing continuum  - A few challenges
(R)evolution of the computing continuum - A few challengesFrederic Desprez
 
Parallel Wireless - Making carrier-grade cellular network deployments as easy...
Parallel Wireless - Making carrier-grade cellular network deployments as easy...Parallel Wireless - Making carrier-grade cellular network deployments as easy...
Parallel Wireless - Making carrier-grade cellular network deployments as easy...Small Cell Forum
 
Gearing up of resource poor mobile devices using cloud
Gearing up of resource poor mobile devices using cloudGearing up of resource poor mobile devices using cloud
Gearing up of resource poor mobile devices using cloudamelpakkath
 
Assessment to Delegate the Task to Cloud for Increasing Energy Efficiency of ...
Assessment to Delegate the Task to Cloud for Increasing Energy Efficiency of ...Assessment to Delegate the Task to Cloud for Increasing Energy Efficiency of ...
Assessment to Delegate the Task to Cloud for Increasing Energy Efficiency of ...IRJET Journal
 
Cloud_Computing.pptx
Cloud_Computing.pptxCloud_Computing.pptx
Cloud_Computing.pptxYash771676
 
Optimization of Fog computing for Industrial IoT applications
Optimization of Fog computing for Industrial IoT applicationsOptimization of Fog computing for Industrial IoT applications
Optimization of Fog computing for Industrial IoT applicationsSabelo Dlamini
 
Mobile cloud computing
Mobile cloud computingMobile cloud computing
Mobile cloud computing402chandan
 
Cumulus Networks Overview
Cumulus Networks OverviewCumulus Networks Overview
Cumulus Networks OverviewAdam Lorts
 
Energy Cost Models of Smartphones for Task Offloading to the Cloud
Energy Cost Models of Smartphones for Task Offloading to the CloudEnergy Cost Models of Smartphones for Task Offloading to the Cloud
Energy Cost Models of Smartphones for Task Offloading to the CloudAswin Vp
 
Cloudlet-Based Cyber-Foraging in Resource-Constrained Environments
Cloudlet-Based Cyber-Foraging in Resource-Constrained EnvironmentsCloudlet-Based Cyber-Foraging in Resource-Constrained Environments
Cloudlet-Based Cyber-Foraging in Resource-Constrained EnvironmentsPatricia Lago
 
presentationGAATT
presentationGAATTpresentationGAATT
presentationGAATTKellyCheah
 
云计算及其应用
云计算及其应用云计算及其应用
云计算及其应用lantianlcdx
 

Similar to Embracing Network Inconsistency for Dynamic Application Offloading in Mobile Cloud Computing (20)

Towards the extinction of mega data centres? To which extent should the Clou...
 Towards the extinction of mega data centres? To which extent should the Clou... Towards the extinction of mega data centres? To which extent should the Clou...
Towards the extinction of mega data centres? To which extent should the Clou...
 
Cloud computing
Cloud computingCloud computing
Cloud computing
 
Cloud computing managing
Cloud computing managingCloud computing managing
Cloud computing managing
 
Networking Challenges for the Next Decade
Networking Challenges for the Next DecadeNetworking Challenges for the Next Decade
Networking Challenges for the Next Decade
 
Research portfolio
Research portfolio Research portfolio
Research portfolio
 
(R)evolution of the computing continuum - A few challenges
(R)evolution of the computing continuum  - A few challenges(R)evolution of the computing continuum  - A few challenges
(R)evolution of the computing continuum - A few challenges
 
Parallel Wireless - Making carrier-grade cellular network deployments as easy...
Parallel Wireless - Making carrier-grade cellular network deployments as easy...Parallel Wireless - Making carrier-grade cellular network deployments as easy...
Parallel Wireless - Making carrier-grade cellular network deployments as easy...
 
Gearing up of resource poor mobile devices using cloud
Gearing up of resource poor mobile devices using cloudGearing up of resource poor mobile devices using cloud
Gearing up of resource poor mobile devices using cloud
 
Assessment to Delegate the Task to Cloud for Increasing Energy Efficiency of ...
Assessment to Delegate the Task to Cloud for Increasing Energy Efficiency of ...Assessment to Delegate the Task to Cloud for Increasing Energy Efficiency of ...
Assessment to Delegate the Task to Cloud for Increasing Energy Efficiency of ...
 
Cloud_Computing.pptx
Cloud_Computing.pptxCloud_Computing.pptx
Cloud_Computing.pptx
 
Optimization of Fog computing for Industrial IoT applications
Optimization of Fog computing for Industrial IoT applicationsOptimization of Fog computing for Industrial IoT applications
Optimization of Fog computing for Industrial IoT applications
 
Mobile cloud computing
Mobile cloud computingMobile cloud computing
Mobile cloud computing
 
Cumulus Networks Overview
Cumulus Networks OverviewCumulus Networks Overview
Cumulus Networks Overview
 
Energy Cost Models of Smartphones for Task Offloading to the Cloud
Energy Cost Models of Smartphones for Task Offloading to the CloudEnergy Cost Models of Smartphones for Task Offloading to the Cloud
Energy Cost Models of Smartphones for Task Offloading to the Cloud
 
Evolution of internet by Ali Kashif
Evolution of internet  by Ali KashifEvolution of internet  by Ali Kashif
Evolution of internet by Ali Kashif
 
Cloud ppt
Cloud pptCloud ppt
Cloud ppt
 
Cloudlet-Based Cyber-Foraging in Resource-Constrained Environments
Cloudlet-Based Cyber-Foraging in Resource-Constrained EnvironmentsCloudlet-Based Cyber-Foraging in Resource-Constrained Environments
Cloudlet-Based Cyber-Foraging in Resource-Constrained Environments
 
presentationGAATT
presentationGAATTpresentationGAATT
presentationGAATT
 
云计算及其应用
云计算及其应用云计算及其应用
云计算及其应用
 
Introduction to Cloud computing
Introduction to Cloud computingIntroduction to Cloud computing
Introduction to Cloud computing
 

Embracing Network Inconsistency for Dynamic Application Offloading in Mobile Cloud Computing

  • 1. ENDA: Embracing Network Inconsistency for Dynamic Application Offloading in Mobile Cloud Computing Jiwei Li Kai Bu Xuan Liu Bin Xiao The Hong Kong Polytechnic University Presenter: Jiwei Li
  • 2. Outline • Background & Problem • Proposed Solution • Preliminary Results • Conclusion
  • 3. Outline • Background & Problem • Proposed Solution • Preliminary Results • Conclusion
  • 4. Mobile Cloud Computing • Applications – Apple’s iCloud, Dropbox • Technical problems – MCC architecture & infrastructure – Network connectivity – Energy efficiency • One important research topic - offloading
  • 5. Offloading Strategy • Previous work – MAUI, CloneCloud, Odessa, COMET
  • 6. Offload to Cloud Compute-intensive applications Computed results High latency (100-300ms) Limited bandwidth (386 Kbs to 3.6 Mbs) High energy consumption Nearly unlimited resources
  • 7. Offload to Cloudlet Compute-intensive applications Computed results Low latency (23-50ms) High bandwidth (54 Mbs) Limited coverage of Wi-Fi (20-100m) Resource constraint
  • 8. Uninvestigated Issues in Offloading • Offloading at mobile environments • Balancing workloads among multiple cloudlets Our research is focused on offloading to cloudlets through Wi-Fi at mobile environments.
  • 10. Re-connection Matters • Re-connection includes – Scanning – Connecting – Assigning IP and network ID • Takes long time (1-12s) • Consumes additional power Reducing re-connection times means increasing energy efficiency.
  • 11. Our Studied Problems • How to predict user’s trajectory? • How to select Wi-Fi access points (AP)? • How to balance workload among cloudlets?
  • 12. Problem Formulation • Minimize: – Communication overheads during offloading at mobile environments • Must satisfy requirements: – App-specific network latency – App-specific response time To put it simply, we aim to select the most energy-efficient Wi-Fi access point, taking user mobility and server load into account.
  • 13. Outline • Background & Problem • Proposed Solution • Preliminary Results • Conclusion
  • 14. Answering a few questions … • Is it feasible to deploy cloudlets at large scale? • Bind current public Wi-Fi hotspots with cloudlets. • How do we overcome resource constraints on cloudlets? • Adopt workload balance management mechanism among participating cloudlets. • How do we conquer Wi-Fi’s limited coverage range issue? • Propose mobility-aware Wi-Fi AP selection scheme.
  • 16. ENDA • Three-tier architecture Design – Cloud – Cloudlet – Smartphone • Objective: – Make the most energy efficient offloading decision
  • 17. Clouds CloudletsSmartphones VM on cloudlets Profilers Wi-Fi 2G/3G WAN User Track Prediction Wi-Fi AP Distribution and Status Wi-Fi AP Selector GPS Runtime System Wi-Fi Adapter FINAL DECISION OFFLOADING INPUT INPUTREPORT REPORT APPINFO Our work will be focused on
  • 18. Advantages • Minimize end-to-end communication overheads • Exempt smartphones from complex computation of making decisions • Improve energy efficiency for offloading
  • 19. Demo Scenario Predicted user track (will be pruned based on app info & network conditions) Effective routes: N1 -> (S, A) N2 -> (S, B) N3 -> (S, D) N4 -> (C, D) ENDA chooses the most energy-efficient Wi-Fi AP according to the specific predicted track Start offloading at location S
  • 20. Outline • Background & Problem • Proposed Solution • Preliminary Results • Conclusion
  • 21. GUI-based Simulation Add routers Add walking path Calculate effective path
  • 22. Simulation Results Wi-Fi B Wi-Fi A Wi-Fi C
  • 23. Outline • Background & Problem • Proposed Solution • Preliminary Results • Conclusion
  • 24. Conclusion • ENDA – Difference from previous work – Minimize communication overheads – Potential to apply to real offloading systems • Future work – Thorough mathematical analysis – Implementation – More complex scenarios
  • 26. Q&A