Enhancement of Emergency Telemedicine Diagnosis Using 3G+

Enhancement of Emergency Telemedicine Diagnosis Using 3G+ Mobile Systems Venkatapathy Prithiviraj, Bharani Kumar Gnanasekaran, Mohan Kumar Murthy and Mohan Devanathan Department of Electronics and Communication Engineering, Pondicherry Engineering College, Puducherry 605014, India; e-mail: {principal, bharaniece07, mohankumarece07, mohanece07}@pec.edu Received 3 October 2011; Accepted: 30 November 2011

Abstract Mobile telemedicine is used in cases of emergency when an ambulance needs to communicate with an expert doctor present at a distant location. With the present day telemedicine systems lacking a framework that can handle all the types of data like real time video, audio and heavy data like Magnetic Resonance Image (MRI), X-ray scans, live cardiograms, and other types of medical data, it becomes necessary to design a telemedicine system that can tackle all the problems and provide a robust telemedicine system. This paper focuses on providing an efficient mobile wireless telemedicine system using 3G+ mobile systems based on High Speed Downlink Packet Access (HSDPA) technology. The telemedicine system can handle real time audio, video streaming, interactive data and background traffic which are essential data traffic involved in a telemedicine system. The latency, jitter associated with the network is reduced without affecting the throughput by analyzing the packet scheduler available at the network layer of the UMTS-HSDPA network and various packet scheduling algorithms are tested. The mobile telemedicine system along with real time channel conditions and background traffic has been simulated in QualNet Developer Network Simulator. Keywords: HSDPA, jitter, latency, packet scheduler, wireless.

Journal of Green Engineering, Vol. 2, 139–154. c 2012 River Publishers. All rights reserved.

140

V. Prithiviraj et al.

1 Introduction Telemedicine is a rapidly growing application of clinical medicine that enables the transfer of medical information through an interactive audio-visual media for the purpose of consulting and for remote examinational procedures. Existing telemedicine systems which use fiber optic broadband systems lack mobility to the patient or the user and satellite communication based systems exhibit large latency. One major constraint for telemedicine systems is the latency when a real time video or real time audio is transmitted, which needs to be maintained well below the threshold levels to provide satisfactory communication link between the patient and the doctor. The wireless communication system can solve the mobility issues of the telemedicine system while bandwidth offered by wireless systems is restricted and also has drawbacks like packet loss, high latency, and jitter. To solve such issues an efficient telemedicine system using the High Speed Downlink Packet Access (HSDPA) system based on the 3rd Generation Partnership Project (3GPP) has been proposed which provides high data rates of the order of up to 14.4 Mbps on the downlink. The cause for the latency is the management of queue present between the MAC layer and the network layer in the core UMTS network. The goal of the packet scheduler is to maximize the network throughput while satisfying the Quality of Service (QoS) of the users. The latency and jitter of several of types of data over the telemedicine link has been reduced using proper IP Packet scheduling algorithms in the UMTS core network. In the proposed system in this paper, the QoS classes are mentioned for different types of flows of data in the telemedicine link and priority bits are attached to their IP headers and weights are assigned to each type of flow. The scenario of a moving ambulance with Rayleigh physical channel conditions communicating with a doctor available on the IP network is considered in the proposed system. The proposed telemedicine system is shown in Figure 1.

2 Packet Scheduling Algorithms The HSDPA scheduler, which resides in Node B, is the key to resource management in the UTRAN downlink because it decides which user is to be scheduled in each transmission time interval (TTI). The scheduling decision does not take into account only the channel conditions, but additional factors such as fairness between users, cell throughput, and QoS parameters are typically considered in the scheduling mechanism. The goal of the HSDPA

Emergency Telemedicine Diagnosis Using 3G+ Mobile Systems

141

Figure 1 Proposed telemedicine system.

scheduler is to maximize the spectrum efficiency of the cell while maintaining the QoS requirements for different data services. It is a fast scheduler which can make decisions on per TTI (2 ms) basis. 2.1 Slow Scheduling Methods In Slow Scheduling Methods, the scheduling decisions rely on the average user’s signal quality (or they do not use any user’s performance metric at all). 2.1.1 Round Robin Scheduling The Round Robin (RR) packet scheduler algorithm is the simplest one which distributes the scheduling turns equally among all active HSDPA users, regardless of the radio channel condition and the QoS requirements of the application running on the mobile devices. The fairness in time sharing of the system resource creates unfairness to those UEs which are under good radio conditions and starving for throughput. 2.1.2 Strict Priority Queuing Strict priority queuing assumes that types of traffic can be differentiated and treated preferentially. Separate FIFO queues are created for each defined priority level and the arriving traffic is sorted into its proper queue as it arrives. Thus the first task of configuring strict priority queuing is to determine the traffic classifications. More queues mean more complexity in running the algorithm. At the service side of the queue, the processing rule is simple: higher priority FIFO queues are always processed to completion before lower priority queues are processed. For example, in a three-queue system, if the two highest priority queues had no packets buffered, then the lowest priority queue would be serviced. The moment a higher priority packet arrived in its FIFO queue,

142


however, servicing of the lower priority packets would be interrupted in favor of the higher priority queue. Strict priority queuing is the gold standard for high priority traffic. The greatest disadvantage relates to the way strict priority queuing treats queues. High-priority packets are always processed before those of less priority. If the amount of high-priority traffic is great, other queues might never empty, leading to worse performance for the low-priority and medium-priority traffic than if a single FIFO queue were used. 2.2 Fast Scheduling Methods In Fast Scheduling Methods, the scheduling decisions rely on UE channel quality measurements (i.e. executed on a TTI basis) that allow tracking the instantaneous variations of the user’s supportable data rate. These algorithms have to be executed in Node B in order to acquire the recent channel quality information. These methods can exploit the multiuser selection diversity, which can provide a significant capacity gain when the number of time multiplexed users is sufficient. 2.2.1 Weighted Fair Queuing Weighted Fair Queuing (WFQ) is a packet scheduling technique allowing guaranteed bandwidth services. The purpose of WFQ is to let several sessions share the same link. WFQ is an approximation of Generalized Processor Sharing (GPS). It is based on a fluid model, so it assumes that the input traffic is infinitely divisible and that all sessions can be served at the same time. Since each session has its own queue, an ill-behaved session (which is sending a lot of data) will only ‘punish’ itself and not other sessions. This is a work-conserving server and it guarantees that each session receives a service rate gi of at least: θi r gi = N j =1 θj where r is the server rate and θ i is the weight for the ith session. The scheduler picks a small piece of data from each session and transmits it to the output link. From an implementation point of view, the slow scheduling methods have a lower degree of complexity than fast scheduling ones, because the latter require the information of the supportable data rate from the UE channel quality measurements for all the users in the cell, and later compute their


143

Figure 2 Comparison of Round Robin, strict priority and weighted fair scheduling algorithms.

priority on a 2 ms basis. In order to have access to such frequent and recent channel quality measurements, the Packet Scheduler must be located in Node B. The graphs in Figure 2 show the theoretical value of end-to-end delay and throughput of different scheduling algorithms. • For weighted fair queuing, the end-to-end delay is less among the three algorithms and the throughput is comparatively high. • For strict priority queuing, the throughput is high than WFQ but the end-to-end delay is high than WFQ. • For Round Robin, the end-to-end delay and the throughput is very high among the three algorithms.

3 HSDPA Based Mobile Telemedicine System The High Speed Downlink Packet Access (HSDPA) system based on the packet data network has the ability to connect with the IP based Local Area Network (LAN) which can be exploited to provide an efficient data connectivity between a mobile user and immobile user. The system is designed to handle all types of data real time audio, video, heavy data and background data for applications like email and the queue management is done by using the priority based scheduling algorithm. The telemedicine system should be fault tolerant and a doctor should always be connected to the patient. In case any failure occurs in the network,

144


Figure 3 Routing between UMTS and IP network.

then an alternate connection should be provided. Such a system is proposed in this paper. 3.1 Routing between UMTS and IP Networks Usually, a routing protocol is not needed within the access network. The nodes in the UTRAN, including UEs, Node B’s and RNCs learn their connectivity through UMTS signaling. However, the scenario has to be configured to enable routing of packets beyond the UTRAN. There are two aspects in routing: • A routing protocol is needed in the backbone core network of UMTS to enable routing between SGSNs, GGSNs, and HLRs, which may be several hops away from each other. • IP Packet Data Networks (PDNs) have their own IP routing architecture. In order to route packets between the UMTS PLMN and IP networks, necessary routing information is needed, especially at the nodes on the boundary between the PLMN and PDN. These requirements for boundary nodes are described below. 3.1.1 Boundary Node Requirements Routing from UMTS PLMN to IP networks: The GGSN node needs to route packets to the proper IP gateway node. Routing from IP networks to UMTS PLMN: The following considerations apply for routing packets from IP networks to the UMTS PLMN: • The IP routers usually do not know the individual IP addresses assigned to UMTS UEs. However, they need to know how to reach the UMTS PLMN network as a whole.


145

• The IP address segments allocated to a UMTS PLMN are usually statically configured. • The aggregate routing information for IP address segments of each UMTS PLMN needs to be advertised to IP networks. 3.2 Inter Domain Routing Routing by Making GGSN a Part of IP Network: In this configuration, the GGSN node of the UMTS PLMN is also included in the routing domain of the IP network. The GGSN node is also configured to be a gateway for the IP network. In order to configure the GGSN node as an IP gateway, it should have at least one interface in each of the routing domains (UMTS and IP network). The GGSN learns the routing information to IP nodes via the dynamic routing protocol configured for the IP network. This method works for routing packets from the UMTS PLMN to the IP network. However, packets cannot be routed from the IP network to the UMTS PLMN because the IP nodes do not know that the packets have to be routed to the GGSN node. In this case, static or default routes can be used to route packets from IP nodes to the UMTS PLMN. 3.3 Mobile Telemedicine Scenario in Qualnet A mobile telemedicine system is the system in which the patient is mobile and is connected to the expert doctor at the other and all types of data needs to be transferred to the expert doctor to know the condition of the patient. The scenario considered here is that a patient is moving in an ambulance which is present on a HSDPA network and a doctor is available on the LAN network. The LAN network is introduced because the usage of the wireless HSDPA network for the doctor who is immobile is inefficient as there is an increased packet loss and latency due to the wireless channel conditions. The IP based LAN network is connected to the UTRAN network using the method described in Section 5.3.2. In Figure 4, Node 20 represents the HSDPA UE in the moving ambulance and Node 26 represents terminal present in LAN network of the hospital premises. 3.4 Priority Assignment and Scheduling The telemedicine system proposed can handle all the types of data that are needed in a telemedicine system i.e. real time audio, video, heavy data and

146


Figure 4 Mobile telemedicine scenario using HSDPA and LAN in QualNet simulator.

background data. The problem that is encountered while handling all such data is the queue management at the network layer. An improperly managed queue leads to extremely large latency and low throughput which will make the system not suitable for operation. The IP headers of the data packets of the different types of flows are assigned priority as shown in the following table. The strict priority and weighted fair queue algorithms are the algorithms that are based on the priority which are used in the network for the purpose of packet scheduling. 3.5 Fault Tolerant System The telemedicine system is a very crucial system and should be kept away from failures. But there are chances that the data network may fail in the UTRAN or in LAN system. In such cases the doctor should not lose connectivity with the patient. For this purpose, if the telemedicine data system fails at any time the system should at least have a circuit switched voice call between the


147

Figure 5 Flow diagram of the proposed telemedicine system.

doctor and user for immediate necessary communication. The flow diagram below summarizes the proposed telemedicine system.

4 Results and Simulations The simulation commences with the setting up of the backbone UMTS network enabled with HSDPA capability. It is followed by the configuring of the Node B’s attached to the Radio Network Controller (RNC) for the handling

148

V. Prithiviraj et al. Table 1 Handover parameters. Parameter Description Minimum cell selection RX level –95 dBm Cell search threshold –80 dBm Cell reselection hysteretic (dB) 3.0

Figure 6 Mobile telemedicine scenario.

of the soft handover. The HSDPA UE is attached to the respective Node B’s. The wireless channel is configured to be a Raleigh’s fading channel. The Local Area Network (LAN) is connected to the HSDPA backbone network by configuring the GGSN as the default gateway and by making it as a part of the IP network. The simulation parameters are listed in Table 1. Figure 6 shows a snapshot of the QualNet scenario of the mobile telemedicine system simulated for a dimension of 1500 m × 1500 m. To introduce congestion two UMTS phone calls and one streaming CBR application were configured in the network between the rest of the users in the network. The UE (mobile ambulance) is allowed to move with varying velocity in the path shown in figure below with an average velocity of

149


S. No 1 2 3 4

Table 2 Telemedicine data traffic. Type of data Data rate (Kbps) TTI (ms) Conversational audio 14 10 Streaming video 128 10 Heavy data 47 10 Background data 47 10

Priority 6–7 4–5 1–3 0

Figure 7 Average end-to-end delay in the network.

60 km/h. The weighted fair queue algorithm is implemented in the different stages of the backbone network. The fault tolerant system is simulated with an assumption that the data network fails after 3 seconds of the simulation time and a voice call is established after the data flow stops. In Figure 6 Node 20 represents the mobile ambulance and Node 26 represent the terminal at the doctor’s end. Table 2 shows the types of data that flows thorough the network and their data rates and their transmit time intervals (TTI). 4.1 Simulation Results of Mobile Telemedicine System The simulation results of the mobile telemedicine system based on the HSDPA and IP networks using the weighted fair queue algorithm for the packet scheduling have been shown in following figures. Table 3 shows the notation used in Figures 7–9. The average end-to-end delay, jitter and the throughput of the network has been observed for the proposed telemedicine system.

150

V. Prithiviraj et al. Table 3 Representation of application. Type of data Description Conversational audio 1027 Streaming video 1026 Heavy data 1025 Background data 1024

Figure 8 Average jitter in the network.

Figure 9 Throughput of the network.

4.2 Simulation Results of Fault Tolerant System Figure 10 shows the snapshot of the fault tolerant system with Node 8 representing the mobile ambulance and Nodes 10 and 14 represents the terminals present at the doctor’s end. Figures 11 and 12 show the results of the fault tolerant system that has been simulated on QualNet simulator.

5 Conclusion In this paper we have proposed a HSDPA based emergency telemedicine system has been proposed with reduced latency and jitter without affecting the throughput of the network. Simulation Results using Qualnet Simulator show that the system fault tolerant, robust and efficient. We plan to conduct an in-depth analysis by considering two types of handoffs: soft (already supported in HSDPA network) and hard handoff (by


Figure 10 Fault tolerant telemedicine system.

Figure 11 Total packets received at the doctor’s end.

Figure 12 Establishment of a UMTS phone call.

151

152


forcing the radio link cut). With such tests we expect to refine the conclusions made in the present paper.

References [1] Sung Hye Hong, Sang Yong Kim, Jungchae Kim, Dong Kyu Lim, Seok Myung Jung, Dong Keun Kim, et al. Portable emergency telemedicine system over wireless broadband and 3G networks. In Proceedings of 31st Annual International Conference of the IEEE EMBS, Minneapolis, Minnesota, USA, September 2–6, pp. 1250–1253, 2009. [2] K. Perakis, K. Banitsas, G. Konnis, and D. Koutsouris. 3G networks in emergency telemedicine – An in-depth evaluation & analysis. In Proceedings of 27th Annual Conference on Engineering in Medicine and Biology, Shanghai, China, September 1-4, pp. 2163–2166, 2005. [3] Pu Zhang, Yuichi Kogure, Hiroki Matsuoka, Masatake Akutagawa, Yohsuke Kinouchi, and Qinyu Zhang. Video a remote patient monitoring system using a Java-enabled 3G mobile phone. In Proceedings of 29th Annual International Conference of the IEEE EMBS Cit´eInternationale, Lyon, France, August 23–26, pp. 2157–2160, 2007. [4] P.O. Amimo-Rayolla, M.O. Odhiambo, and A.M. Kurien. Video streaming over HSDPA for delivery of essential services in under serviced areas in Africa. In Proceedings of 2008 Third International Conference of Broadband Communications, Information Technology & Biomedical Applications, Gauteng, November 23–26, p. 262, 2008. [5] Young Ik, Seo Dan, and Keun Sung. Analytical comparison of three packet scheduling schemes under a per-user minimum throughput assurance requirement in HSDPA. In Proceedings of IEEE 16th International Symposium, Berlin, July 24, p. 1708, 2006. [6] A. Pahalawatta, P. Berry, R. Pappas, and T. Katsaggelos, Content-aware resource allocation and packet scheduling for video transmission over wireless networks. IEEE Journal on Selected Areas in Communications, 25(4):749–759, 2007. [7] M. Sarvagya, R.V.R. Kumar, and B.N. Bhandari. A new packet scheduling scheme and performance analysis of high speed packet data access. In Proceedings of 2008 IEEE Region 10 Conference (TENCON 2008), Hyderabad, India, November 19–21, p. 1, 2008. [8] C. Kugean, S.M. Krishnan, O. Chutatape, S. Swaminathan, N. Srinivasan, and P. Wang. Design of a mobile telemedicine system with wireless LAN. In Proceedings of Circuits and Systems, Asia-Pacific Conference, Singapore, 6 January, p. 313, 2003. [9] L. Bianchi, M. Luglio, M.G. Marciani, and F. Zampognaro. Multiservice temedicine application over an emulated DVB-RCS satellite link. In Proceedings of 3rd International Symposium on Applied Sciences in Biomedical and Communication Technologies (ISABEL), Rome, Italy, 7–10 November, p. 1, 2010. [10] M. Elena, J.M. Quero, C.L. Tarrida, and L.G. Franquelo. Design of a mobile telecardiology system using GPRS/GSM technology. In Proceedings of 24th Annual Conference and the Annual Fall Meeting of the Biomedical Engineering Society EMBS/BMES Conference, Sevilla, Spain, 23–26 October, p. 1859, 2002. [11] I. Sachpazidis, D. Rizou, W. Menary, and I.G.D. Fraunhofer. Satellite based health network in Peru and Brazil. In Proceedings of Information Technology and Applications in Biomedicine, Darmstadt, Germany, 30–31 May, p. 309, 2008.


153

Biographies Venkatapathy Prithiviraj Dr. Prithviraj is the Principal of Pondicherry Engineering College. He completed his Bachelor of Engineering (Electronics and Communication Engineering) in 1972 from the College of Engineering, Guindy Madras University, his M.S. by research in 1982 from IIT, Madras, and his Ph.D in 1999 from IIT, Kharagpur (research area – Signal Processing Techniques in Array Antennas Systems). He is one of the founding members of Pondicherry Engineering College. At present he is holding the position of Dean-in-charge, School of Engineering, Pondicherry University. He has been teaching for over 28 years and held the position of Head of the Electronics and Communications Engineering Department. He has published over 70 technical research papers. He has also held the position of Director, IT for Government of Pondicherry (2002–2005). Currently, he is a Member of Expert Committee for monitoring International Indo-French projects in the field of Information Technology as well as Regional Committee of AICTE for Tamil Nadu and Pondicherry. He has a keen interest in research and development projects and provided leadership in many successful projects sponsored by various organizations such as DRDO, ISRO, Department of Electronics and Department of Information Technology at IITs and PEC. He is the recipient of the IEEE International Student Branch Award in 1984 and the EDI Award for best technical paper, entitled “COFDM for Telemedicine Applications”, in 2007. He is a Life Member of ISTE and Member of EMC Engineers. His areas of interest include Broad band and Wireless Communication, Mobile Computing, VLSI for Wireless Applications, Tele Medicine and e-Governance Applications. Bharani Kumar Gnanasekaran received his B.Tech degree in Electronics and Communication Engineering from Pondicherry Engineering College Affiliated to Pondicherry University, Puducherry, India in 2011. He is currently working on Session Border Controllers in the Research and Development wing of Sonus Networks. His primary research interests are in the areas of VOIP, Wireless Communications and Networking.

154


Mohan Kumar Murthy received his B.Tech degree in Electronics and Communication Engineering from Pondicherry University, Puducherry, India in 2011. He is currently working for Tata Consultancy Services as Assistant Systems Engineer. His research interests include areas of communication systems and Networking. Mohan Devanathan has completed his B.Tech in Electronics and Communication at Pondicherry University, India in 2011. His areas of interests include cellular mobile communication, embedded systems and computer networks.