Evolution towards Fourth Generation Mobile Multimedia Communication

Carlos Rodrguez Casal (1,2), Frits Schoute (1,3), Ramjee Prasad (1) (1) Delft University of Technology P.O.Box 5031 Delft, The Netherlands Tel: +34.15.278.35.44 Fax: +34.15.278.18.43 carlos@ frits@ ramjee@mmc.et.tudelft.nl (2) Universidad Pblica de Navarra, Campus Arrosada s/n Pamplona, Spain (3) Philips Business Communications P.O.Box 32 Hilversum, The Netherlands

Abstract.
In this paper a view of the European developments in the field of wireless broadband multimedia communications is given following two lines: on one side evolution in Second and Third Generation of mobile communications; on the other side evolution of multimedia data communications. This two evolutions will merge into a new concept of communications: the Mobile Broadband Systems, or fourth generation of Mobile Multimedia Communications. Within this context, a multi-disciplinary project is being developed at Delft University of Technology. The MMC (Mobile Multimedia Communications) project. MMC holds a combination of: promising applications, novel user-interfaces, compression, protocols geared to the mobile user and transmission techniques that give abundant bandwidth to the user.

The unprecedented growth of world-wide mobile and wireless markets, coupled with advances in communications technology and the accelerated development of services taking place in fixed networks, point now to two possibilities: GSM evolution and the introduction of a Third Generation Mobile Communication System.

1.1. GSM Evolution.

The initial transmission rate of GSM systems is 9.6 kilobits per second (kbps) with a possible increase to 14.4 kbps by changing the error protection coding. High Speed Circuit Switched Data (HSCSD) allows the combination of multiple time slots and offers a raw data rate of up to 64kbps (38.4 kbps user data rate). This high-speed data functionality is available in GSM networks without any base station hardware modifications, only the software upgrade is needed [1]. Next step will go up to 164 kbps by using packet switched resource allocation; this means that resources are allocated only when data are to be sent or received. This technique is known as General Packet Radio Service (GPRS), also known as GSM2+. Although this may appear to be a perfect evolution from GSM, it is not as it is not possible to "commute" between high speed circuit switched data and general packet radio service, because each of them requiring a priori a dedicated specialised radio [2]. An improvement upon the former system is Enhanced Data rate for GSM Evolution, (EDGE), also known as Evolved GSM. EDGE technology is designed to be introduced in existing digital networks such as GSM and the North American Digital Advanced Mobile Phone Service (D-AMPS). It is based on a higher level modulation and allows transmission at data rates of 384 kbps [3]. This evolution is shown in figure 1.

1. Cellular Evolution.
The First Generation Cellular Systems were implemented in early eighties. They used direct analog voice modulation and were based on a cellular architecture. Transmission rates where around 2.4 kbps and there were different systems working at different countries. The Second Generation Cellular Systems were designed as the technical evolution of microelectronics enabled use of full digital communication over the radio channel with portable devices. From Europe the standard GSM (Global System for Mobile communications) has emerged as an European standard and it has became the main mobile system all around the world providing services to 70 million users in Europe, and over 100 million users word wide not only in Europe but in 110 countries with over 200 networks all over the word.

Figure 1: Cellular Systems Evolution

1.2. Mobile Systems Convergence: Third Generation.

Nowadays mobile communication systems can be defined in 5 groups: cordless, cellular, satellite, paging and private mobile radio systems. These mobile communication systems will all be included at a common system: the Universal Mobile Telecommunications System, (UMTS). While the 1st and 2nd generation cellular systems were all deployed at 1 GHz spectrum domain with some extensions to higher frequencies as the earlier allocations become congested. Radio regulation authorities did allocate new spectrum for mobile use in the 2 GHz area. Initial transmission will be at 2GHz with a rate of 2Mbps in low mobility conditions, independently of the environment. This rate will be only between 144 and 384 kbps for wide area services. 144 kbps Is the minimum achievable user bit rate in any mobility condition and the UMTS system will have to support user speed up to 500 km/h [4]. There will be available up to 10 Mbps for local area mobility services by 2.005, but only in a provisional way. The final extension, everywhere, will be available by 2.010.

Concurrently to the mobile telephony change, local area networks are also changing to make it possible to distribute all wireless services to the users. This will evolve the local area networks (LAN's) towards wireless local area networks (WLANs) that will provide multimedia services. The European Telecommunication Standards Institute (ETSI) supports projects, as the Broadband Radio Access Networks (BRAN) project, in order to develop HIPERLAN2, (HIgh PErformance Radio Local Area Network.) to be used as next generation of WLAN. Some other dimensions as HomeRF or Bluetooth are being developed in wireless communications. HomeRF: the Home Radio Frequency Working Group (HRFWG) [5] was formed in March 1998 to look for a specification for wireless digital communication between PCs and consumer electronic devices anywhere in and around the home. This new communication system will provide flexibility and mobility and will enable interoperability between many different consumer electronic devices from different manufacturers. A new protocol is being developed, it is called the Shared Wireless Access Protocol (SWAP). The system is designed to carry both voice and data traffic and to interoperate with the Public Switched Telephone Network and the Internet. It operates in the 2.4 GHz band and uses a digital frequency hopping (FH) spread spectrum (SS) radio. It supports both a Time Division Multiple Access (TDMA) service to provide delivery of interactive voice and other time-critical services, and a Carrier Sense Multiple Access / Collision Avoidance (CSMA/CA) service for delivery of high-speed packet data. The system data rate will be 1 or 2 Mbps and use 2 or 4 FSK modulation (Frequency Shift Keying) and it will support up to 127 devices per network.

2. Multimedia Data Communication.

Multimedia communication points out a communication with multiple ways of presenting the information, as a combination of text, data, graphics, animation, images, sound, speech and still or moving video. Interesting characteristics added to multimedia communication definition, are the media flexibility for different applications and the possibility of interaction between the communication subjects.

Bluetooth technology will allow the replacement of the many proprietary cables that connect one device to another with one universal short-range radio link. Bluetooth, "A Global Specification for Wireless Connectivity" [6], will replace the cable used today to connect printers, desktops, fax machines, cellular phones, laptops, keyboards, joysticks and virtually any other digital device. Bluetooth radios will operate in the unlicensed Industrial Scientific Medical (ISM) band at 2.45 GHz. (this frequency is available all over the world) with a gross data rate of 1 Megabit per second (Mbps). Spectrum spreading is accomplished by frequency hopping in 79 hops displaced by 1MHz, starting at 2.402 GHz and stopping at 2.480 GHz. Multimedia data communication and mobile telephony are evolving simultaneously. Future systems will make use of both technologies and a new concept for communications will become a reality. This future communication system is the so-called "Mobile Multimedia Communication". Systems convergence towards Mobile Multimedia Communication is shown in figure 2.

4. MMC Project:
For fourth generation mobile telecommunication systems, innovative concepts are needed that may not be found by evolutionary enhancements to existing systems, the approach taken at the Mobile Multimedia Communication project of the Delft University of Technology is to form a multi-disciplinary team in which user aspects get as much attention as the technological challenges. This MMC project has five research lines, some of their achieved results are here presented.

4.1. Applications: Work Coordination.

Multimedia was seen as most suitable for nonstandard work situations, i.e. those situations where the professional runs into a complex problem that cannot be solved by known procedures and for which the professional lacks some expertise. These situations are typically found in emergency services, medical assistance and disaster management. Two field studies have been performed so far: ambulance services and the repair and maintenance of photo-copying machines. During field observations the mobile workers and related actors were followed in their daily work. All actions and conversations were written down. Based on the field studies, scenarios are build to envision the role of different kinds of support for the daily work practice of ambulance and copier services. These especially concern the remote support through mobile communication via different modalities (speech/ audio/video) and ways of interaction (synchronous/ asynchronous) [9].

3. Towards a Fourth Generation ?

Only if the third generation becomes a success there will be other future generations. The successful introduction of third generation mobile systems is critically dependent on the matching of the offered services to the user needs (applications), and on cost-effective provision of those services. Between 1992 and 1995 there was a project on the European Community that was called Mobile Broadband System, MBS [7]. It was part of the second part of the RACE program (Research and development in Advanced Communications in Europe). It was a project, but also the Mobile Broadband Systems concept was born there, targeting future outdoor, cellular scenarios with high mobility and high data rates, as to carry Mobile Multimedia Communications. These systems will be the fourth mobile generation. No standardisation activity has so far been foreseen, although the European Radiocommunications Office, ERO, has already considered the spectrum needs of this new service [8].

4.2. User Interface and Transparency.

Communication from one office to another is relatively straightforward with today's technology. However when one of the parties wishing to communicate is based outside, problems of remoteness and insufficient infrastructure can limit the effectiveness of their communications.

MOBILE TELEPHONY Paging PMR Systems Cordless Satellite Cellular

Figure 2:

MULTIMEDIA DATA COMMUNICATION

MBS
Mobile Broadband Systems

HIPERLAN HomeRF Bluetooth

UMTS
Universal Mobile Telecommunication System

Mobile MultimediaCommunication

Mobile and Multimedia convergence.

The emergence of wearable computers offers a potential method of improving this situation particularly when a visual link is provided by a mobile video transmission system. The wearable equipment that is proposed at the MMC project is a head mounted display. Head mounted displays can be used to present visual information transparent as to the environment, (presenting transparent information means taking characteristics of background visual information and characteristics received at the head mounted displays into account). This transparency problem had to be solved as the visual information could not be projected in just any part of the field of view [10].

4.3 Compression.
Most compression methods perform well on a transmission channel with a very low bit error rate, but many of them break down on a wireless channel where errors are more rule than exception. Very-low bit rate compression should not be the prime objective, since the emphasis is on sufficient quality in order to perform tasks which depend strongly on the context. Therefore that reducing the bit rate is important, but not the most important objective in wireless communication. It is more important to make the compression (in collaboration with the transmission system) as robust as possible against the errors that are typical for the wireless channel, possibly with the help of the videoanalysis tools that are also being used for MPEG-4 multimedia applications. Research has been carried out in two areas for source coding and two techniques have been proposed: H.263 For mobile video communication. This technique proposed in the MMC project is based on the high compression recommendation by the International Telecommunication Union (ITU) H.263. The basic configuration of the video source coding consists of two parts: the reference frame, with respect to which the differential frames are coded, and the possibility to force intra coding of the macro blocks that have been detected as corrupted [11]. Compression of the Shapes of Video Objects. Errors will occur during the transmission, so the compression has to be error-robust. A technique for contour coding combining polar coordinates and Discrete Cosine Transform (DCT) has been proposed [12].

provide a service to them. Transmission errors can be countered by Forward Error Correction (FEC) and Automatic Repeat reQuest (ARQ) at the link layer, but this has consequences for the delay and throughput performance that is experienced by the higher layers. The answer found in the MMC project is a hierarchical protocol structure that provides different Qualities of Service (QoS) to the various traffic streams in mobile multimedia communication. The hierarchy is realised with a hybrid TDM/FDM (Time Division Multiplexing / Frequency Division Multiplexing) technique in which frames (the largest unit of data) are subsequently composed of packets, fragments and Radio Data Units (RDUs). The hierarchical approach allows for easy future expansion, both in terms of the allocated bandwidth and technological advances in modulation and coding [13]. Further steps have been taken to complete the system for communication by developing a method to combine data from multiple receivers in a virtual cellular network (VCN). A virtual cellular network uses a virtual cell architecture, defined as an area of a predetermined size and formed virtually around a mobile station. The basic idea is that one mobile is heard by multiple base stations and viceversa, multiple base stations can send data to one mobile. It is shown that virtual cell concept greatly enhances the quality of transmission [14]. The MMC-project proposes a virtual cellular network for the links between base stations and mobile users. This implies that at every instant each active mobile station is in contact with multiple base stations. A testbed for high speed cellular networks is being developed. The testbed consists of an RF front-end that operates in the 2.4GHz band, and a low-power processing unit. Data rates are up to 25Mbit/s, with higher rates attainable by placing multiple units in parallel.

4.5. Broadband Radio Transmission.

Future demands for mobile communication systems will be dominated by the heterogeneity of broadband services which are to be supplied in indoors and outdoor environments, simultaneously, with varying degrees of mobility. Because these heterogeneous telecommunication services get more sophisticated and because the introduction of multimedia, the amount of information being exchanged increases. More information means higher data rates in order to meet the challenges of the information society. In the context of mobile services one of the constraints in realising the needs of the information society is the availability of frequency spectrum. The scarcity of spectrum and the new technical possibilities in recent years have drawn attention all over the world to the millimetre band. In the MMC project, transmission is located at the V band (from 40 to 75 GHz), centered at 60 GHz. Measurements have been developed, new techniques for measurements

4.4. Transmission Protocols.

In all OSI layers measures are taken to deal with the adverse conditions of the wireless channel. It is hard to maintain the OSI ideal according to which the higher layers need not know by which method the lower layers

have been proposed, measurements have been carried out and channel simulators have been built [15]. The Chosen modulation scheme has been Orthogonal Frequency Division Multiplex (OFDM), this modulation is specifically able to cope with the problems of the multipath reception. Current research on OFDM covers the development of advanced synchronisation algorithms and the influence of non-linear power amplification.

[5] [6] [7]

http://www.homerf.org http://www.bluetooth.com http://www.comnets.rwthaachen.de/report96/node211.html# Mobile Broadband Systems, Final Report, ERO July 1.997. www.ero.dk F.W.G. van den Anker & R.A. Lichtveld. "Early evaluation of new technologies. The case of mobile multimedia communications for emergency medicine". To appear in: C. Vincent & B. de Mol (Eds.): Risk and safety in medicine. From human error to organizational learning. To be published in 1999. O.1 F.H. Oosterhoff: Exploring binocular rivalry in transparency situations Supplement Perception, volume 27, page 114a, London, UK, August 1998. Spaan, F., Lagendijk, R.L., Biemond, J. Error Robust Video Coding based on H.263 Proceedings International Conference on Image Processing Chicago, U.S.A., October 1998 98 Spaan, F., Lagendijk, R.L., Biemond, J. ErrorRobust Contour-Coding Using Polar Coordinates Proceedings Advanced School for Computing and Imaging Conference. 98 Lommel, Belgium, June 1998 J. Bakker, F. C. Schoute and R. Prasad. "An Air Interface for High Bandwidth Cellular Digital Communications on Microwave Frequencies". IEEE Vehicular Technology Conference (VTC) '98, Ottawa, Canada, 18-21 May 1.998. Bakker, J.D., R. Prasad, "An uplink protocol implementation in a virtual cellular network", Proceedings of the First Symposium on Wireless Personal Multimedia Communications, Yokosuka, Japan, November 1998. D. M. D. Matic, H.Harada and R. Prasad, "Indoor and outdoor frequency measurements for mm- waves in the range of 60 GHz". IEEE vehicular Technology Conference (VTC) '98, Ottawa, Canada, 18-21 May 1.998.

[8]

[9]

5. Conclusions.
Mobile communications are evolving fast. While data rate is growing, different services are being integrated and multimedia based services are shown to be the next step to be taken. For fourth generation mobile telecommunication systems, innovative concepts are needed that may not be found by evolutionary enhancements to existing systems. The approach taken at the Mobile Multimedia Communication project of the Delft University of Technology is to form a multidisciplinary team in which user aspects get as much attention as the technological challenges. Within the project, an experimentation platform has been developed, and multimedia services has been tested in working and emergency situations. To provide information to the user, different interfaces are being analysed. A Head Mounted Display is the solution proposed by the MMC members. Regarding Video compression has in consideration that frequency errors will happen due to the wireless channel where the transmission takes place. Virtual Cell architecture and protocols specially designed to cope with the mobile environment are presented in combination with Orthogonal Frequency Division Multiplexing as modulation technique.

[10]

[11]

[12]

[13]

References.
[1] [2] [3] www.gsmdata.com http://www.cs.hut.fi/~hhk/GPRS "Vision for the evolution from GSM to UMTS" GSM MoU Association 1998, TG.13 http://www.gsmworld.com/3g/index.html GSM MoU Association Permanent Reference Document, TG 32, April 98. "UMTS Radio Requirements" at : http://www.gsmdata.com/artholley.htm

[14]

[15]

[4]