Femtocell Based Economic Health Monitoring Scheme Using Mobile Cloud Computing

Debashis De, Anwesha Mukherjee

Department of Computer Science and Engineering, West Bengal University of Technology,

BF-142, Sector -I, Salt Lake City, Kolkata -700064, West Bengal, India. dr.debashis.de@gmail.com

Abstract— This paper presents a mobile health monitoring scheme based on femtocell and mobile cloud computing. In this scheme, the health information of each user is captured by sensors and sent to the corresponding mobile device. From the mobile device the health data is transferred to the femtocell under which the mobile device is registered. In femtocell it is verified whether the user’s health is normal using a database stored inside the femtocell. If any abnormality is detected the data are sent to the cloud. The health data are securely stored on the cloud and accessed by the health centre. Based on health data, the corresponding health centre takes proper action to cure the patient. The monetary cost required to access the data on the cloud through the proposed scheme is calculated. Simulation results present that using femtocell results in achieving approximately 28-70% and 30-75% reduction in cost consumption for accessing medium and large amount of data on cloud respectively than using macrocell, microcell or picocell base station.

Keywords— femtocell; mobile cloud computing; health monitoring; cost consumption

I. INTRODUCTION

E-healthcare system or services are gaining popularity day by day. The healthcare systems or services that are supported by electronic processes and communication are known as e-health care systems [1-3]. Telemedicine and mobile health (m-health) are the essential form of e-health by which a person can aware of his or her physical and psychological fitness at a distance [4-7]. E-healthcare applications require a wireless body sensor network (BSN) to support multiple data rates with reliable and energy efficient data transmission. Wireless BSN provides a secure, efficient and reliable platform for e-health monitoring service over the traditional health monitoring services [1]. Mobile Cloud Computing (MCC) [8-10] takes an important role for mobile health (m-health) application in respect to the limitations like quality of service, physical storage, security, privacy, first response, medical error due to which the traditional healthcare system suffers [11-12]. A real time health monitoring system has been described in [13] where a sensor is attached with the existing medical equipments that are inter-connected to exchange service with the help of cloud environment. To support e-health monitoring, a pervasive environment is implemented in [14] using which data accessing, emergency management system, networking problems for heterogeneous network are solved. But still some major issues for quality of service like low bandwidth, latency, security, privacy, and context awareness are not resolved.

Femtocell which is also known as Home Node Base Station (HNB) [15-18] is a low power base station with the help of which the solutions to these problems can be determined. In this paper we have introduced a femtocell based secure mobile cloud computing service for m-health monitoring. This paper is organized as follows: section II describes the femtocell and mobile cloud computing based proposed m-heath monitoring scheme, section III and IV present the monetary cost consumption model and performance analysis of proposed approach respectively, and section V presents the conclusion.

II. FEMTOCELL AND MCC BASED M- HEALTH MONITORING SCHEME

A. Working Principle of Proposed Scheme

Femtocell and MCC based proposed m-health monitoring scheme requires the following components:

• Body sensor network • Mobile station (MS) • Femtocell i.e. HNB • Internet connectivity

The working model of proposed m-health scheme is pictorially depicted in Fig.1. Mobile user and HNB are connected via Uu interface [16]. HNB is connected to the internet and Home Node Base Station-Gateway (HNB-GW) via Iuh interface. By HNB-GW, the HNB is connected with the core network [16]. HNB-GW is connected with the core network via Iucs/Iups interface [16]. To provide proper security between HNB and HNB-GW over the internet a security gateway (SeGW) is maintained [16]. The working principle of the proposed scheme is divided into five phases and described in TABLE I.

TABLE I. WORKING PRINCIPLE OF THE PROPOSED SCHEME

Phase No

Function Description

1 Health data capturing by BSN and transmission

to MS

The BSN captures the physiological information (body temperature, blood sugar level, blood pressure, respiration rate, ECG etc) of the mobile user and

sends it to the corresponding MS 2 Health data

transmission from MS to HNB

When the MS registered under a HNB receives the physiological information

from the BSN, it sends the physiological information with the

location information captured using the GPS stored in the MS, phone number

385978-1-4799-2572-8/14/$31.00 c©2014 IEEE

and IMEI number (International Mobile Equipment Identity) to the HNB

3 Health data checking at HNB

HNB compares the received health data against the threshold values of the

respective health parameters stored in a database which is stored inside the

HNB, this phase is described in sub-section B

4 Health data transmission from

HNB to cloud

If the health data seems to be abnormal, then the HNB sends the health data to

the cloud 5 Data stored in

cloud for access by health centre

After the health data are received, stored in cloud with the location

information and IMEI number, so that health centre can access the data and

take necessary action like sending advice to the patient via a voice

call/SMS or sending ambulance to the patient location to take care of his or

her

Five phases described in TABLE1 are performed in the proposed health monitoring scheme.

B. Verification of User Health at HNB

In HNB a database is maintained which stores the range of values for the health parameters of a normal and healthy person. If the health data values obtained from the MS do not fall in the range stored in the database it indicates abnormal health condition. Then the health data are sent to the cloud.

C. Security in Proposed Scheme

The HNB is connected to the HNB-gateway through a security gateway. As the user health data are transmitted from the HNB to the cloud via the security gateway, secure data transmission is achieved. On the other hand to provide health data security in cloud, a user id and password are generated when for the first time the data are received from the user. The generated user id and password are sent to the user so that the user can access the data on cloud. To achieve high security a two-way verification is also introduced. When the user gives the corresponding user id and password to access the data, a verification code is sent to the mobile phone of the user. After giving the correct verification code, the user can access his or her data on cloud. The id of the health care centre which first accesses the data of the patient is attached to patient information stored in the cloud and phone number of the health care centre is sent to the mobile phone of the corresponding patient. For each health care centre a user id and password are maintained, so that no one except that particular health centre can access or see the data. As except the intended health care centre and the user no one can access the data, privacy, authentication and integrity are guaranteed from the view point of user and health centre both. If the data of a patient is not updated in the cloud for more than one year, the data values are erased from the health database maintained in cloud.

III. MONETARY COST CONSUMPTION IN PROPOSED SCHEME

In this section the cost consumed in the proposed m-health scheme in terms of money is determined and compared to that of the m-health schemes of macrocell, microcell or picocell

network. The parameters used in monetary cost consumption are presented in TABLE II.

TABLE II. PARAMETERS IN COST CALCULATION

Parameters Definition CC Monetary cost consumption per instruction execution

Cmtfi Monetary cost consumption for i th message transmission from MS to HNB

Cmrfi Monetary cost consumption for i th message reception by MS from HNB

Cmtmi Monetary cost consumption for i th message transmission from MS to macrocell base station (BS)

Cmrmi Monetary cost consumption for i th message reception by MS from macrocell BS

Cmtmii Monetary cost consumption for i th message transmission from MS to microcell BS

Cmrmii Monetary cost consumption for i th message reception by MS from microcell BS

Cmtpi Monetary cost consumption for i th message transmission from MS to picocell BS

Cmrpi Monetary cost consumption for i th message reception by MS from picocell BS

CCti Monetary cost consumption for i th message transmission from BS (macrocell BS/microcell BS/picocell BS/HNB)

to server CCri Monetary cost consumption for i th message reception by

BS (macrocell BS/microcell BS/picocell BS/HNB) from server

Mtf Number of message transferred from MS to HNB Mrf Number of message received by MS from HNB Mtm Number of message transferred from MS to

macrocell/microcell/picocell BS Mrm Number of message received by MS from

macrocell/microcell/picocell BS Mt Number of message transferred from

macrocell/microcell/picocell/HNB to server Mr Number of message received by

macrocell/microcell/picocell /HNB from server DaC Total health data in cloud Daf Total health data in HNB database

A. Cost consumption in Proposed Scheme

Monetary Cost required in message transmission from MS to HNB is given by,

1

tf

totmtf mtfii

M

C C=

= (1)

Required monetary cost in message reception by MS from HNB is given by,

1

rf

totmrf mrfii

M

C C=

= (2)

Total monetary cost involved in message transmission and reception between MS and HNB is given by,

totmf totmtf totmrfC CC += (3)

386 2014 IEEE International Advance Computing Conference (IACC)

Fig. 1. Working model of proposed m-health monitoring scheme.

For accessing the health data in cloud the monetary cost consumption is determined as,

insC C CC C DaC ×= × (4)

Total monetary cost consumption for accessing the health data in HNB database is determined as,

insf C fC C DaC ×= × (5)

Monetary cost needed for message transmission from macrocell/microcell/picocell BS/HNB to the server is given by,

1

tM

CtitotmtCi

C C=

= (6)

Monetary cost involved in message reception by macrocell/microcell/picocell BS/HNB from the server is given by,

1

rM

CritotmrCi

C C=

= (7)

Total monetary cost consumption in message transmission and reception between macrocell/microcell/picocell BS/HNB and the server is given by,

totmC totmtC totmrCC CC += (8)

In the proposed scheme if only the health data seems to be abnormal then the data are sent to the cloud through HNB. The health data are accessed on the cloud through the HNB in a network. If the health is normal, then no health data is sent to the cloud. Thus the monetary cost consumption in case of normal health condition is the sum of the cost consumed in message transmission and reception between MS and HNB, and the total cost for accessing the health data in HNB database, given as,

t o t h fn o r to t m f in s fC C C= + (9)

The monetary cost consumption in case of abnormal health condition is the sum of the cost required for message transmission and reception between MS and HNB, between HNB and the server, and the total cost for accessing the health data in HNB and in cloud, given as,

tothfabnor totmf totmC insf insCC C C C C= + ++ (10)

Thus considering both the normal and abnormal health condition, the total monetary cost consumed in the proposed scheme for accessing health data on cloud using femtocell BS i.e. HNB is given by,

(1 )

)( (1 )

( )

tothf tothfnor S tothfabnor S

totmf

totmf t

S

o

C f S

Ctm f CC C

P C C

C C P C

C Da P

C C C C Da C C Da

P

= × + × × + −

× + + × ×

×

+ × ×

= + × −

(11)

where the probability of patient condition is normal is PS and 1SP ≤ .

2014 IEEE International Advance Computing Conference (IACC) 387

B. Cost Consumption in Macrocell Network

Monetary cost consumption for message transmission from MS to macrocell BS is given by,

1

tm

totmtm mtmi

M

i

C C=

= (12)

Monetary cost involved for message reception by MS from macrocell BS is given by,

1

rm

totmrm mrmi

M

i

C C=

= (13)

Monetary cost consumption for message transmission and reception between MS and macrocell BS is given by,

totmm totmtm totmrmC CC += (14)

If femtocell is not used, the user can access the health data on the cloud using the macrocell/microcell/picocell base station in a macrocell/microcell/picocell network. Hence the data checking at the HNB will not occur whereas all checking will take place in the cloud. Thus the total monetary cost consumption for health monitoring using a macrocell BS is the sum of the cost consumption in message transmission and reception between MS and macrocell BS, between macrocell BS and the server and the total cost for accessing the health data on cloud, calculated as,

( )insCtothm totmm totmC

totmm totmC C CC DaC C C C

C C C= +

× ×+=+

+ (15)

Using equation (15) the total monetary cost for accessing the health data in cloud by a MS through a macrocell BS in a macrocell network is calculated and comparing with the proposed scheme Proposition 1 is given in sub-section C.

C. Proposition 1: Proposed m-health scheme is economic than the m-Health Scheme of Macrocell Network

Proof: The database of HNB stores small amount of data for checking the health condition where large amount of health data is stored in the cloud. Then f CDa Da<< which implies,

C f C CC C Da C C Da× × << × × => insf insCC C<<

Moreover 1SP ≤ , hence comparing equation (11) with equation (15) it is observed that,

tothf tothmC C< (16)

Hence it is established that using the proposed scheme cost consumption to access health data on cloud can be reduced than macrocell based network which in turn proves that the

proposed m-health scheme is economic than the m-Health scheme of macrocell network.

D. Cost Consumption in Microcell Network

Total monetary cost in message transmission from MS to microcell BS is given by,

1

tm

totmtmi mtmiii

M

C C=

= (17)

Monetary cost consumption in message reception by MS from microcell BS is given by,

1

rm

totmrmi mrmii

M

i

C C=

= (18)

Total monetary cost in message transmission and reception between MS and microcell BS is given by,

totmmi totmtmi totmrmiC CC += (19)

The total monetary cost consumption for health monitoring using a microcell BS is determined as the sum of the cost consumed in message transmission and reception between MS and microcell BS, between microcell BS and the server and the total cost for accessing the health data on cloud, given by,

( )insCtothmi totmmi totmC

totmmi totmC C C

C C C CC C C DaC

= + ++= + × ×

(20)

Using equation (20) the total monetary cost for accessing the health data on cloud by a MS through a microcell BS in microcell network is calculated and comparing with proposed scheme Proposition 2 is established in sub-section E.

E. Proposition 2: Proposed m-health scheme is economic than the m-Health Scheme of Microcell Network

Proof: Cloud stores large amount of health data. But the HNB contains small amount of data in its database to check the health condition which implies f CDa Da<< . Hence,

C f C CC C Da C C Da× × << × × => insf insCC C<<

As 1SP ≤ , comparing equation (11) with equation (20) it is observed that,

tothf tothmiC C< (21)

Hence it is proved that using the proposed scheme cost consumption to access the health data on cloud can be reduced than microcell based network which in turn proves that the proposed m-health scheme is economic than that of the microcell network.

388 2014 IEEE International Advance Computing Conference (IACC)

F. Cost Consumption in Picocell Network

Required monetary cost in message transmission from MS to picocell BS is given by,

1

tm

totmtp mtpii

M

C C=

= (22)

Total monetary cost in message reception by MS from picocell BS is given by,

1

rm

totmrp mrpi

M

i

C C=

= (23)

The total monetary cost consumption for health monitoring using a picocell BS is determined as the sum of the cost consumed in message transmission and reception between MS and picocell BS, between picocell BS and the server and the total cost for accessing the health data on cloud, given by,

( )in sCto thp to tm p to tm C

totm p to tm C C CC D a

C C C C

C C C

= +

= + × ×

+

+ (24)

Using equation (24), the total monetary cost for accessing the health data on cloud by a MS through a picocell BS in picocell network is determined and comparing with the proposed scheme Proposition 3 is given in sub-section G.

G. Proposed m-health scheme is economic than the m-Health Scheme of Picocell Network

Proof: Large amount of health data is stored in cloud where HNB contains small amount of data which is used for checking the health condition i.e. f CDa Da<< . Therefore,

C f C CC C Da C C Da× × << × ×

=> insf insCC C<<

As 1SP ≤ , comparing equation (11) with equation (24) we obtain,

tothf tothpC C< (25)

Hence it is proved that using proposed m-health scheme cost consumption to access health data on cloud can be reduced than using picocell BS in a picocell network which in turn proves that the proposed m-health scheme is economic than the m-Health scheme of picocell network.

IV. PERFORMANCE ANALYSIS OF PROPOSED M-HEALTH SCHEME

In this section the monetary cost consumed in proposed health monitoring scheme is calculated. Let approximately 106 numbers of instructions are executed in cloud to access per unit amount of health data. Let total amount of health data

stored in the cloud is 1KB-10000GB and total amount of health data stored in the femtocell BS is 10bytes-10KB. It is assumed that the monetary cost per instruction execution is 0.01 units. In case of the m-health monitoring scheme of a macrocell network, the total monetary cost for accessing the health data on the cloud by a MS is calculated using equation (15) based on different data sets and presented in Fig.2 and Fig.3. In case of m-health monitoring scheme of a microcell network and a picocell network, the total monetary cost consumption for accessing the health data on the cloud by a MS are calculated using equation (20) and (24) respectively based on different data sets and presented in Fig.2 and Fig.3. In case of the femtocell BS based proposed m-health scheme, the total monetary cost consumed for accessing the health data on the cloud and femtocell by a MS according to the need is calculated using equation (11) based on different data sets and presented in Fig.2 and Fig.3.

Fig. 2. Total monetary cost consumption for accessing the medium amount of health data (in KB) on cloud by the MS.

Fig. 3. Total monetary cost consumption for accessing the large amount of health data (in GB) on cloud by the MS.

The number of message transmission and reception in macrocell network, microcell network and picocell network for health data accessing on cloud is almost same and huge amount of health data is stored on the cloud. So the cost

2014 IEEE International Advance Computing Conference (IACC) 389

consumption for health data accessing on cloud in macrocell network, microcell network and picocell network becomes almost same as observed from Fig.2 and Fig.3. In proposition 1, 2 and 3 we have already proved that the proposed m-health scheme is more economic than the m-health scheme of macro, micro or picocell based network. Now, from the simulation results it is also demonstrated that using the femtocell and MCC based proposed m-health monitoring scheme, monetary cost consumption to access the health data on cloud by a MS can be reduced and hence an economic m-health care can be provided.

V. CONCLUSION

In this paper we have proposed femtocell and mobile cloud computing based a mobile health monitoring scheme. The physiological condition of the user is captured using body sensor network. The health data are sent to the femtocell by the mobile device. Femtocell checks the data and if any abnormality is detected, the data are sent to the cloud where user id and password are generated and sent to the respective user. The id of the health care centre that first tries to access the data is attached to the corresponding patient data in the cloud. For each health care centre a user id and password are maintained, so that no one except that particular health centre can access or see the data. The health care centre takes necessary action either by giving advice through a phone call or sending message or by sending ambulance to the victim based on the health condition. Thus health care service is provided to the patient as consumer where both the provider and corresponding consumer have access to the health data with the help of the user id and password and users can easily interact with their health care providers using the proposed scheme. Data transmission from HNB to the server occurs through a security gateway. Hence during transmission authentication, privacy and integrity of the patient data are achieved. User id and password are maintained so that only the respective user and the corresponding health care centre can access the data in the cloud and thus patient data security is achieved in the cloud. The monetary cost consumption involved in the proposed scheme is calculated and compared with the cost consumption in m-health monitoring schemes of macrocell network, microcell network and picocell network. Simulation and theoretical results present that using the proposed scheme monetary cost consumption can be reduced than that of m-health monitoring schemes of macrocell network, microcell network and picocell network. According to the simulation results for accessing medium and large amount of health data on cloud, approximately 28-70% and 30-75% reduction in cost consumption are achieved respectively than that of the macrocell, microcell and picocell network.

ACKNOWLEDGMENT

Authors are grateful to Department of Science and Technology (DST) for sanctioning a research Project entitled “Dynamic Optimization of Green Mobile Networks: Algorithm, Architecture and Applications” under Fast Track Young Scientist scheme reference no.: SERB/F/5044/2012-2013 under which this paper has been completed.

REFERENCES [1] H. Wang, D. Peng, W. Wang, H. Sharif, H. Chen, and A.

Khoynezhad, “Resource-Aware Secure ECG Healthcare Monitoring Through Body Sensor Networks,” IEEE Wireless Communications, vol. 17, no. 1, pp. 12-19, 2010.

[2] A. Ray and D. De, “Intelligent Body Sensor Network for Pervasive Health Monitoring: A Survey,” Journal of Computational Intelligence and Electronic Systems, American Scientific Publishers, vol. 1, no. 1, pp. 67-81, 2013.

[3] S. E. Kern and D. Jaron, “Healthcare Technology, Economics and Policy: An Evolving Balance,” IEEE Engineering in Medicine Biology Magazine, vol. 22, no. 1, pp. 16–19, 2003.

[4] J. C. Lin, “Applying Telecommunications Technology to Health-care Delivery,” IEEE Engineering in Medicine Biology Magazine, vol. 18, no. 4, pp. 28-31, 1999.

[5] J. Zhang, J. N. Stahl, H. K. Huang, X. Zhou, S. L. Lou, and K. S. Song, “Real-Time Tele consultation with High-Resolution and Large-Volume Medical Images for Collaborative Healthcare,” IEEE Transactions on Information Technology Biomedicine, vol. 4, no.2, pp. 178–185, 2000.

[6] R.G. Lee, H.S. Shen, C.C. Lin, K.C. Chang, and J.H. Chen, “Home Telecare System using Cable Television Plants—An Experimental Field Trial,” IEEE Transactions on Information Technology Biomedicine, vol. 4, no.1, pp. 37–44, 2000.

[7] R. Holle and G. Zahlmann, “Evaluation of Telemedical Services,” IEEE Transactions on Information Technology Biomedicine, vol. 3, no. 2, pp. 84–91, 1999.

[8] H. T. Dinh, C. Lee, D. Niyato, and P. Wang, “A Survey of Mobile Cloud Computing: Architecture, Applications, and Approaches,” Wireless Communications-Wiley Online Library, 2011, pp. 1-38.

[9] S. S. Qureshi, T. Ahmad, K. Rafique, and S. U. Islam, “Mobile Cloud Computing As Future For Mobile Applications – Implementation Methods And Challenging Issues,” IEEE CCIS, 2011, pp. 467-471.

[10] L. Guan, X. Ke, M. Song, and J. Song, “A Survey of Research on Mobile Cloud Computing,” IEEE/ACIS, 2011, pp. 387-392.

[11] D. Kopec, M.H. Kabir, D. Reinharth, O. Rothschild, and J.A. Castiglione, “Human Errors in Medical Practice: Systematic Classification and Reduction with Automated Information Systems,” Journal of Medical Systems, vol. 27, no. 4, pp. 297 – 313, 2003.

[12] L. T. Kohn, J. M. Corrigan, and S. Donaldson, “To Err Is Human: Building a Safer Health System,” National Academy Press, Washington, 1999.

[13] T.W. Tang, M.C. Hu, and Y.C. Hsu, “A Mobile Phone Based Homecare Management System on The Cloud,” International Conference on Biomedical and Informatics, 2010, pp. 24-42.

[14] U. Varshney, “Pervasive Healthcare and Wireless Health Monitoring,” Journal on Mobile Networks and Applications, vol. 12, no. 2, 2007, pp. 113 – 127

[15] D. De, B. Mahata, and S. Jana, “Effective load distribution algorithm considering power and interference mitigation in a dense femtocell based network,” Second International Conference on Computer, Communication, Control and Information Technology, 2012, pp. 692-697.

[16] J. Zhang, G. Roche, and L. De, “Femtocells technologies and deployment,” A John Wiley and Sons, Ltd., Wiley, 2010.

[17] A. Mukherjee, S. Bhattacherjee, S. Pal, and D. De, “Femtocell Based Green Power Consumption Methods for Mobile Network,” Computer Networks, Elsevier, vol. 57, no.1, pp. 162-178, 2012.

[18] A. Mukherjee and D. De, “Congestion Detection, Prevention and Avoidance Strategies for an Intelligent, Energy and Spectrum Efficient Green Mobile Network,” Journal of Computational Intelligence and Electronic Systems, American Scientific Publishers, vol. 2, no. 1, pp. 1-19, 2013.

390 2014 IEEE International Advance Computing Conference (IACC)