MONITORING QoS OVER WIRELESS SENSOR NETWORKS

For Medical Applications

Carlos Abreu

Polytechnic Institute of Viana do Castelo, Av. Atlântico, Viana do Castelo, Portugal

Manuel Ricardo

, Paulo Mendes

INESC Porto, Faculty of Engineering, University of Porto, Porto, Portugal

Industrial Electronics Dept., University of Minho, Guimarães, Portugal

Keywords: Quality of Service, Medical Applications, Wireless Sensor Networks.

Abstract: Wireless Sensor Networks (WSNs) for medical purposes enables real-time (RT) acquisition of vital signals.

In this context the network reliability, data integrity and RT delivery have extreme importance. The network

should guarantee an appropriate level of Quality of Service (QoS). This paper presents the QoS parameters

and metrics of standard telemedicine and WSNs based Medical Applications (MA), and a strategy, based on

QoS monitoring, to prevent QoS degradation in WSNs for MA.

1 INTRODUCTION

A Biomedical WSN (BWSN) is a small-size WSN

for MA. BWSNs have the potential to promote new

applications and services in different healthcare

scenarios, e.g., mobile telemedicine, emergence

response and management, location services and

patient monitoring (Varshney, 2009). In the case of

patient monitoring scenarios, a BWSN can be used

to monitor a wide range of biological and vital

signals.

This monitorisation can be done through various

situations, e.g., in emergency response providing

information about the health condition of patients,

monitoring physiological signals in people with

chronic illness or measuring the activity level of

elderly or disabled people (Gao et al., 2008). In any

of these situations, BWSN presents itself as a key

technology to the success of many ongoing

investigations. They are responsible to transport

information, which in MA have to obey to very strict

quality criteria (Liang, 2009). As noted in (Liang &

Balasingham, 2007), the main function of a BWSN

is to ensure that sensed signals are delivered reliably

and efficiently. There is a set of QoS requirements

that must be fulfilled.

In the rest of this paper we discuss the QoS needs

of MA and briefly present the main QoS parameters

and metrics in traditional networks and propose a

strategy to QoS monitoring in order to improve the

performance (prevent the degradation) of BWSNs.

2 QoS REQUIREMENTS OF

MEDICAL APPLICATIONS

Currently, most medical procedures are performed

based on information obtained from electronic

or/and informatics systems. This information, about

the health condition of an individual, must have

medical quality. According to the (American

College of Medical Quality, 2010), medical quality

can be defined as “the degree to which health care

systems, services and supplies for individuals and

populations increase the likelihood for positive

health outcomes and are consistent with current

professional knowledge”. This definition makes

clear that, healthcare systems and its communication

networks have a key role in the quality of healthcare

services provided to citizens. So, it is extremely

important that the communication networks used to

transport medical information ensure a service with

382

Abreu C., Ricardo M. and Mendes P..

MONITORING QoS OVER WIRELESS SENSOR NETWORKS - For Medical Applications.

DOI: 10.5220/0003163603820385

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2011), pages 382-385

ISBN: 978-989-8425-37-9

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

quality. In the context of communication networks,

this characteristic is expressed in terms of QoS.

Recommendation E.800 of the International

Telecommunication Union (ITU) introduced the

following QoS definition: "totality of characteristics

of a telecommunication service that bear on its

ability to satisfy stated am implied needs of the user

of the service". According to the definitions

presented above, communication networks used to

transport information used by health providers are a

keystone to the quality of patient health and care

according to its medical condition.

The term QoS is understood and interpreted in

various ways by the scientific community; most

often it refers to the ability of a network to deliver

data efficiently and reliably (Liang, 2009). RFC2386

defines QoS as: "A set of parameters that the

network should ensure during the transport of a data

stream". Given this definition, it is necessary to

specify what parameters should the network ensure,

who imposes them and in what situations.

2.1 QoS Support in BWSN

BWSNs can be considered small size WSNs, and

therefore they have the same constrains, e.g., limited

computational power, limited memory and severe

power supply limitations (in implantable nodes this

problem worsens) (Khan et al., 2009). However,

from the network viewpoint there are some

important differences, such as, the number of sensor

nodes on the network, its localization and the

number of hops between nodes, e.g., compared with

normal WSNs, BWSNs contain a reduced number of

nodes; nodes localization in BWSNs is well known

or restricted to a limited area, on the other hand, in a

normal WSN the sensor nodes are randomly

deployed in large areas. These differences, although

significant, do not invalidate the utilization of the

same QoS parameters and metrics. However, if

taken into account, they may facilitate the

development of more efficient QoS mechanisms.

Most of the ongoing investigation in BWSNs takes

advantage on these characteristics to implement QoS

mechanisms to specific monitoring applications.

There are several published works about BWSNs

based monitoring systems that, in most cases, are

individual or for a small number of patients (Khan et

al., 2009) (Alemdar & Ersoy, 2010). The authors of

(Liang & Balasingham, 2007), present a priorities

based routing protocol with QoS guaranties on the

following parameters, end-to-end delay, delivery

ratio and power consumption. This protocol was

validated by simulation with 20 sensor nodes.

According to scholarly information, this was the first

QoS enabled routing protocol to healthcare

monitoring applications. In (Ko et al., 2010), the

authors present a monitoring system to hospital use

where the nodes localization is precise and well

known. In this case, the authors conclude that the

existence of a backbone of static network nodes

improves the network performance.

2.2 QoS Support in WSN

WSNs differ from “traditional” networks in several

aspects, e.g., WSN nodes have strong constrains in

power consumption and consequently in the power

of RF signal emitted which limits its communication

range and transmission rate (Zhang et al., 2010);

they have strong computational and memory

limitations, so that the algorithms used must be

lightweight and efficient; the dynamic nature of

WSNs and the lack of centralized control present

additional challenges in the development of efficient

QoS solutions (Ben-Othman & Yahya, 2010).

Due to its specificities, in WSNs, the QoS must

be ensured in all layers of the protocol stack in use,

starting at physical and ending in application layer.

The authors of (Wang et al., 2006), present an

analysis of the QoS requirements at each layer based

on the Open System Interconnection (OSI) model.

Here, our concern is to discover how a WSN can

guarantee QoS to applications that depend on it.

Thus, in order to obtain some independence from the

network technology used we will focus on the

mechanisms to guarantee QoS at the network layer.

In (Asokan, 2010) the author analyses the most

relevant routing protocols in mobile Ad Hoc

networks (WSN can be regarded as an Ad Hoc

network). In his analysis he highlights that the

studied protocols take into account only a small set

of the total QoS parameters. Finally, he concludes

by saying that routing protocols are needed, namely,

those that take into account more ample QoS

parameters. Also among the authors of (Asokan,

2010) it’s consensual that there are still many

unsolved issues regarding routing protocols with

QoS guarantees.

3 QoS PARAMETERS

AND METRICS

QoS requirements are imposed to network by

applications that use it. They don't depend only on

the intrinsic characteristics of data to be transmitted

but also on its use, in other words, on its application.

MONITORING QoS OVER WIRELESS SENSOR NETWORKS - For Medical Applications

383

3.1 QoS Parameters Concerning the

Application Viewpoint

In (Ruiz, 2006) the author identifies the following

parameters, as the most important at this level: Peak

Data Rate (PDR), Sustainable Data Rate (SDR) and

Maximum Burst Size (MBS). Drawing upon these

parameters one can define traffic classes which can

be applied to different data flows and applications.

Different network technologies have adopted

different QoS classes. However, there is a common

feature among them, the distinction between classes

for applications with and without RT requirements.

Being the most widely used network technology, in

Table 1 we present the traffic classes defined for IP

networks as defined in (Marchese, 2007).

Table 1: IP QoS Classes.

QoS class Characteristics

0 RT, jitter sensitive, highly interactive.

1 RT, jitter sensitive, interactive.

2 Transaction data, highly interactive.

3 Transaction data, interactive.

Low loss only (short transactions, bulk

data, video streaming).

5 Traditional applications of IP networks.

Other technologies have defined different

classes. However, in essence, they have the same

division taking into account the application

characteristics.

3.2 QoS Parameters Concerning the

Network Viewpoint

According to (Marchese, 2007) and (Ruiz, 2006),

from the communication network viewpoint, the

most important metrics used to measure and ensure

that the QoS required for a given application is

guaranteed, are: Packet Transfer Delay (PTD),

Packet Delay Variation (PDV), Packet Loss Ratio

(PLR), Packet Error Ratio (PER) and Bandwidth

(BW). There are no unique values to these metrics.

Rather, they depend on the network characteristics

and particular requirements of each application.

3.3 QoS Parameters in WSN

In the context of WSNs, due to its specificities, the

QoS metrics presented earlier are insufficient. In

(Chen & Varshney, 2004) the authors identify new

QoS metrics that reflect the collective effort of all

network nodes to perform a given task. They are:

collective delay or latency, collective packet loss

rate, collective bandwidth, collective data rate and

information throughput. WSNs routing protocols

with QoS requirements have to take into account

these metrics in order to provide more efficient QoS

mechanisms.

3.4 Telemedicine QoS Metrics

QoS requirements depend on the application

characteristics and network technology in use. Since

we are interested in the development of mechanisms

for QoS support in BWSNs, a good starting point is

to study how 'traditional' networks are used in

medicine. There are countless applications and

services using information systems and

communications networks in medicine. However, in

this work, the focus is on applications and services

that allow remote monitoring of biological and vital

signals, such as, ECG, EEG, BP and T. According

(Ruiz, 2006) and (Varshney, 2009), taking into

account the characteristics of each one of these

signals and the requirements of the application at

stake, the following metrics,

Table 2, are sufficient to

telemedicine and telemonitoring applications and

services in “traditional” IP networks.

Table 2: QoS Metrics to Telemedicine.

PTD Application or service characteristics

50ms Audio transmission with interaction.

100ms Interactive application.

150ms Video transmission with interaction.

400ms Default value to real-time applications.

PLR

3% Tx of medical images and signals.

10%

Tx of radiology images with compression

ratios of 10:1 and 20:1.

15% Interactive video and audio transmission.

20% Default to real-time applications.

15Kb/s Tx of medical images and signals.

60Kb/s

Tx of radiology images with compression

ratios of 10:1 and 20:1.

100Kb/s Interactive video and audio transmission.

200Kb/s Default value to real-time applications.

4 QoS MONITORING

As mentioned in (Asokan, 2010), QoS provision

over low-rate Wireless Personal Area Networks

(LoWPANs) can be done over different layers on the

communication protocol stack. However, by

themselves, these mechanisms do not provide QoS

in all situations, e.g., in individual monitoring

applications it is necessary to know how many

signals we can transmit without degrading the

BIODEVICES 2011 - International Conference on Biomedical Electronics and Devices

384

network QoS, if we need to monitor multiple

patients then we also need to know how many

patients may be on the network. In a more complex

scenario, the patients are distributed in a given area:

in this case, it is necessary to determine if and where

the QoS can be guaranteed.

It is necessary to develop management and

supervision tools to manage and control the network.

These management and supervision tools must

provide sufficient information to add new

monitoring services or applications to network

without degrading the existing QoS.

Figure 1: QoS Monitoring System.

Figure 1 presents, in terms of high level blocks,

the QoS monitoring system architecture and

principal components. Each sensor node has, stored

in a QoS Database (QoS DB), information about its

QoS requirements. This information can be accessed

by the QoS Manager (QoSM) and transmitted

through the LoWPAN to the QoS Monitoring

Software (QoSMoS). To add a new patient to the

network, the healthcare provider must insert the

patient QoS needs on the QoSMoS. Then, taking

into account the network QoS status (based on

information collected from the sensor nodes in the

network) and new patient QoS needs the Patient

Admission and Control (PAC), decides if and where

the patient can be inserted in the network.

5 CONCLUSIONS

This paper provides a review of the state-of-the-art

in QoS for telemedicine and BWSNs, presenting the

principal QoS parameters and metrics. Finally, we

have argued that existing QoS mechanisms do not

guarantee QoS in all utilization scenarios and a

different approach to increase QoS or prevent its

degradation has been proposed. Due limited capacity

of WSNs it’s necessary to develop monitoring tool

to allow QoS management and supervision.

ACKNOWLEDGEMENTS

Portuguese Foundation for Science and Technology,

FCT, PhD Grant SFRH/BD/61278/2009.

REFERENCES

Alemdar, H. & Ersoy, C., 2010. Wireless sensor networks

for healthcare: A survey. Computer Networks.

American College of Medical Quality, 2010. Index of

Policies. [Online] Available at: http://www.acmq.org/

policies/policyindex.cfm [Accessed 17 July 2010].

Asokan, R., 2010. A review of Quality of Service (QoS)

routing protocols for mobile Ad hoc networks., 2010.

Ben-Othman, J. & Yahya, B., 2010. Energy efficient and

QoS based routing protocol for wireless sensor

networks. Journal of Parallel and Distributed

Computing, 70(8), pp.849-57.

Chen, D. & Varshney, P. K., 2004. QoS Support in

Wireless Sensor Networks A Survey., 2004.

Gao, T. et al., 2008. Wireless Medical Sensor Networks in

Emergency Response Implementation and Pilot

Results., 2008.

Khan, P., Hussain, M. A. & Kwak, K. S., 2009. Medical

Applications of Wireless Body Area Networks.

JDCTA: International Journal of Digital Content

Technology and its Applications, 3(3), p.185 ~ 193.

Ko, J., Gao, T., Rothman, R. & Terzis, A., 2010. Wireless

Sensing Systems in Clinical Environments: Improving

the Efficiency of the Patient Monitoring Process.

Engineering in Medicine and Biology Magazine,

IEEE, 29(2), pp.103-09.

Liang, X., 2009. QoS Provisioning for Wireless Sensor

Networks: Algorithms, Protocols and Modeling.

Liang, X. & Balasingham, I., 2007. A QoS-aware Routing

Service Framework for Biomedical Sensor Networks.,

2007.

Marchese, M., 2007. QoS Over Heterogeneous Networks.

John Wiley \& Sons Ltd.

Ruiz, I.M., 2006. Contribuciones a Modelos de Tráfico y

Control de QoS en los Nuevos Servicios Sanitarios

Basados en Telemedicina.

Varshney, U., 2009. Pervasive Healthcare Computing:

EMR/EHR, Wireless and Health Monitoring. 1st ed.

Springer.

Wang, Y., Liu, X. & Yin, J., 2006. Requirements of

Quality of Service in Wireless Sensor Network., 2006.

Zhang, X. et al., 2010. An Energy-Efficient ASIC for

Wireless Body Sensor Networks in Medical

Applications. Biomedical Circuits and Systems, IEEE

Transactions on, 4(1), pp.11-18.

MONITORING QoS OVER WIRELESS SENSOR NETWORKS - For Medical Applications

385