• Aucun résultat trouvé

Power-and Delay-Aware Mobile Application-Data Flow Adaptation: the MobiHealth system case study

N/A
N/A
Info
Télécharger
Protected

Academic year: 2022

Partager "Power-and Delay-Aware Mobile Application-Data Flow Adaptation: the MobiHealth system case study"

Copied!
21
0
0

Chargement.... (Voir le texte intégral maintenant)

Texte intégral

(1)

Conference Presentation

Reference

Power-and Delay-Aware Mobile Application-Data Flow Adaptation:

the MobiHealth system case study

WAC, Katarzyna, et al.

Abstract

Emerging healthcare applications rely on personal mobile devices to monitor patient vital signs and to send it to the hospitals-backend servers for further analysis. However, these devices have limited resources that must be used optimally in order to meet the requirements of healthcare applications end-users: healthcare professionals and their patients. This paper reports on a case study of a cardiac telemonitoring application delivered by the so-called MobiHealth system. This system relies on a commercial device with multiple (wireless) network interfaces (NI). Our study focuses on how the choice of a NI affects the end-to-end applicationpsilas data delay (extremely important in case of patientpsilas emergency) and the energy consumption of the device (relating to the service sustainability while a patient is mobile). Our results show the trade-off between the delay and battery savings achieved by various NI activation strategies in combination with application-data flow adaptation. For a given mobile device, our study shows a gain of 40-90% in battery savings, traded against the higher delays (therefore applicable [...]

WAC, Katarzyna, et al. Power-and Delay-Aware Mobile Application-Data Flow Adaptation: the MobiHealth system case study. In: Proceedings of the 10th International Conference on e-health Networking, Applications and Services - HealthCom 2008, Singapore

(Singapore), 7-9 July 2008, 2008, p. 20

Available at:

http://archive-ouverte.unige.ch/unige:72670

Disclaimer: layout of this document may differ from the published version.

1 / 1

(2)

Kate WAC, HealthCom’08

Power- and Delay-Aware Mobile Application-Data Flow Adaptation

the MobiHealth system case study

Katarzyna Wac (Kate), PhD student

Pravin Pawar, Bert-Jan van Beijnum, Richard Bults Mortaza Bargh, Arjan Peddemors

(3)

Kate WAC, HealthCom’08

Outline

• Introduction

m-health services: from MobiHealth project to MobiHealthTM system

• Problem Description

telemonitoring service: battery consumption, delays vs. NI choice

• Approach

measurements-based performance evaluation of service for different NIs

• Conclusions & Recommendations

which NI choice is best for which application flow?

(4)

Kate WAC, HealthCom’08

Introduction

m-health services

from MobiHealth project to MobiHealth

TM

system

(5)

health tele-monitoring services

general practitioner Mobile Network

Operator

medical specialist

Internet

mobile patient (BAN-MBU)

healthcare centre

e.g. 2.5/3G, WLAN

MobiHealth project

MobiHealth™ System

(6)

Kate WAC, HealthCom’08

MobiHealth System History

2002-2004: MobiHealth – EU IST-2001-36006 (5 countries) m-health services: technically feasible? (emerging 2.5G/3G) 2005-2006: HealthService24 – EU eTEN-517352 (4 countries) m-health services: clinically/commercially feasible?

2004-2008: Freeband-Awareness – Dutch BSIK-5902390

m-health services: proactively context-aware? (security/privacy?) from 2007: MobiHealth BV – University of Twente (NL) spin-off commercial m-health services: platform for any sensor system?

2007-2009: Myotel – EU eTen-C046230 (4 countries)

telemonitoring/teletreatment services: chronic neck-shoulder pain?

(7)

Kate WAC, HealthCom’08

Problem Description

telemonitoring service:

battery consumption, delays vs. NIs status

(8)

Kate WAC, HealthCom’08

Problem description

Focus: explorative study

• mobile: limited processing, communication, storage, battery capacity

• mobile health services need to support emergency & non-emergency cases

• health telemonitoring service performance:

• data delay

• battery consumption

How to choose NI and parameterize application flow to

• match delay requirement to emergency/non-emergency case and

• minimize battery consumption

=f (NIs status)

(9)

Kate WAC, HealthCom’08

Approach

measurements-based performance evaluation

of telemonitoring service for different NIs

(10)

Kate WAC, HealthCom’08

Measurements Setup

MobiHealth™ system used

• cardiac patient case: 3 leads ECG, heart rate*, SpO2, pleth, alarm (128 Hz)

• MBU: Qtek 9090, Windows Mobile® 2003 (!battery drain!)

• main battery: Li-ion polymer 1490 mAh

• NI: Bluetooth (always ON gathering data from MOBI™)

• NI: WLAN (802.11b, OS ‘best-battery’ setting)

• NI: WWAN-GPRS (class 10: 4+1/3+2 slots)

• Application flow: 5-14 Bytes, 128Hz

• aggregation: 1 second of data

• compression (ZIP): 38-85 %

• TCP-IP end-to-end path

• continuous: ~1.2-1.5, 5.5 or 7.7 kbps

• bursts: 5.5 or 7.7 kbps, ~ Mbps

*heart rate is derived from 3 leads ECG

NI status: ON-IDLE-OFF

MOBI

MBU

(11)

Kate WAC, HealthCom’08

Approach: Measurements

Application-delay: App-RTT

• system response time for: telemonitoring/teletreatment

• does not require MBU & BEsys clocks synchronization

• MBU: measures it every 10 seconds

(12)

Kate WAC, HealthCom’08

Approach: Measurements

Remaining battery level (Windows Mobile

®

)

• MBU: measures every 5 seconds

(13)

Measurements setup

focus

BAN BEsys

(14)

Kate WAC, HealthCom’08

Selected Findings

(15)

NI choice: consumed battery capacity

Min: GPRS ON-ACTIVE, WLAN OFF

Max: WLAN ON-ACTIVE, GPRS ON-IDLE

(16)

NI choice: App-RTT delay

Min: WLAN ON-ACTIVE, GPRS ON-IDLE Max: GPRS ON-ACTIVE, WLAN OFF

(inverted to the battery profile!)

emergency

emergency (if no WLAN)

?

(17)

NI activation strategies: power efficiency

S-EM emergency

continuous flow: WLAN ON-ACTIVE, GPRS ON-IDLE (S-EM) bursty flow: WLAN OFF/ON-ACTIVE, GPRS OFF

patient reachable?

(18)

Kate WAC, HealthCom’08

Conclusions & Recommendations

telemonitoring service:

which NI choice is best?

(19)

Kate WAC, HealthCom’08

Conclusions & Recommendations

• GPRS vs WLAN have complementary profiles

• GPRS: power consumption lower, App-RTT higher than WLAN

• App-RTT vs power consumption

• minimal App-RTT if continuous application flow

• minimal power consumption if application flow in bursts

• ? WLAN App-RTT lower when GPRS ON-IDLE than when GPRS-OFF

• Optimal choices:

• emergency: continuous flow (App-RTT efficient)

• WLAN ON-ACTIVE (GPRS ON-IDLE)

• GPRS ON-ACTIVE (WLAN-OFF)

• non-emergency: bursty flow (power efficient)

• WLAN ON-ACTIVE/-IDLE (GPRS ON-IDLE) Æ n=4 seconds of data

• GPRS ON-ACTIVE/-IDLE (WLAN-OFF) Æ n=6 seconds of data

• larger n are not power-efficient enough to be considered (+ patient unreachable)

(20)

Kate WAC, HealthCom’08

Future work

• More measurements

• NI activation-deactivation (ON-OFF) and NI-NI WLAN-GPRS handovers

• multiple MBU-devices, NIs, different locations (mobile!) and times

• detailed study on delay variation as f(NI)

• multiple application data flows with different App-RTT requirements

• NI activation strategy vs.

• monetary cost of networks usage

• security considerations

• Further QoS/QoE considerations for the Mobihealth system

• requirements & provisions

• towards dependable system Æ dynamic system adaptation e.g. self-healing

(21)

Kate WAC, HealthCom’08

www.healthservice24.com www.mobihealth.org

www.awareness.freeband.nl www.myotel.eu

www.mobihealth.com

Thank You!

Références

Télécharger maintenant ( PDF - 21 Page - 2.26 MB )

Documents relatifs

How to organize the data flow of the City: a case study with the Spatial Data Infrastructure CartoPOLIS

How to organize the data flow of the City: a case study with the Spatial Data Infrastructure CartoPOLIS?.

Linking Data in the Insurance Sector: A Case Study

Figure 1 illustrates how Insurable Object, Vehicle, Claim and Claim Folder entities can be defined in terms of attribute and can be linked to each other via relationships in

On the effective penetration depth of l-band radiometry and the validation of SMOS data: a case study

Maria-José Escorihuela, Yann H. Kerr, Andre Chanzy, Patricia de Rosnay, Jean-Pierre Wigneron. On the effective penetration depth of l-band radiometry and the validation of SMOS data:

Evaluation of an End-to-End Delay Estimation in the Case of Multiple Flows in SDN Networks

3 depicts the cumulative distribution for the relative error committed by our solution (σ Dec ), as well as by the trivial solution (σ T rivial ) and by the Conditional method (σ

Towards a Passive Measurement-based Estimator for the Standard Deviation of the End-to-End Delay

We propose a solution to estimate the first two moments of the end-to-end delay experienced by the packets of a flow in a wired network based only on delay measurements

Approximating the end-to-end delay using local measurements: a preliminary study based on conditional expectation

The obtained results show that the method is generally good, with a relative error on the estimated standard deviation of the end-to-end delay usually close to 5%, though it may

The Energy and Carbon Footprint of Training End-to-End Speech Recognizers

Table 1: Estimated CO 2 from training a single French (Com- monVoice) or English (LibriSpeech) state-of-the-art end-to-end speech recognizer on Nvidia Tesla V100 GPUs2. Models CO

Safety at the Sharp-End: A Case Study in the Gas Sector

This observation of a discrepancy between what seems to be necessary to effectively manage risk (i.e. the managerialisation of field workers achieved through the