|
|||
|
Description of the group:
Services and content:
Slide Show SperstadDiscussion.ppt Telecommunication technology towards 2020Geir E. Øien1, Nils Holte1, Steinar Andresen2, Torbjørn Svendsen1, and Mikael Hammer1 1Department of Electronics and Telecommunications, 2Department of Telematics Norwegian University of Science and Technology (NTNU)
1 Broadband for All and Access Everywhere – Two key challenges towards 2020Predicting the research focus and market development - over a 15-year period - of a field that changes as quickly as communication technology is not a task that can ever be executed with 100 % accuracy. Looking 15 years into the past, many of today’s most important and penetrating communication technologies (e.g., GSM, WWW) were not yet in public use. Taking into account the accelerating pace of improvements and cost reductions with regard to enabling technologies, as well as the ever-increasing impact that communication technology has on the human lifestyle, it is likely that the future may hold even greater surprises than did the past. However, some basic trends are emerging, and we will focus on these. We will concentrate on the technology dimension rather than the market dimension, since this is where our main competence lies, and it is also the dimension that we believe can be predicted with greatest accuracy. Many different groups and organisations have made predictions of future technological developments; for instance, the objective of the CELTIC initiative [CEL1] is almost the same as the scope of the present document. The two main communication technological challenges for the time period 2004 – 2020 can be summarised in the slogans Broadband for All and Access Everywhere. We should immediately note that this is not the same as Broadband Access Everywhere, which for several reasons we find not to be a realistic goal. In the present context we define broadband as full broadband, i.e., with enough bandwidth to implement a Full Service Access Network (FSAN) for domestic uses, including transmission of several high-quality TV signals. To achieve this, rates in the region 30 – 50 Mbit/s are needed. Contradictory, perhaps, to some predictions, it is our belief that broadband access according to this definition will mainly be wired (say, in 90 – 95 % of all cases). This is particularly true in urban and suburban regions, and is due to a lack of frequencies. While broadband fixed access has the potential for integrating all services in the same connection, there will in most cases probably be a wireless interface for the last few meters (< 50 m); hence the connection may still be perceived as wireless by the end user. The cost of long distance bulk data transport will be almost negligible; access, services, and content will be the cost-driving factors. The other main challenge, Access Everywhere, focuses on the need to make basic telecommunications services available in an (almost) ubiquitous and seamless fashion, even in areas with little or no infrastructure and for high-mobility nomadic users. Examples of such services are currently voice phone, e-mail, SMS/MMS, location and navigation services, emergency calls, banking transactions and information search via the Internet, etc. However, we emphasise again that these are not services requiring bandwidths that we have defined above as full broadband. Even when taking into account all anticipated technological advances and price reductions, there will still exist technological limitations, for instance due to cost and available frequency spectrum. There are fundamental trade-offs between accessible capacity, distance, degree of mobility, and cost per bit, which imply that full broadband access in sparsely populated areas with little or no infrastructure will always be expensive. A variety of wireless solutions, from satellite links to fixed radio access, will be important for implementing Access Everywhere. In the following, we look in more detail into some of the important basis technology trends within telecommunication technology, sorted by discipline under the headings Wired communications infrastructure, Wireless communications infrastructure, Networking technologies, and Services and content. Note that security aspects are mostly left out, since there exists a separate INFOSAM-2020 working group and a corresponding position paper on such issues. 2 Wired telecommunications infrastructureBoth from a technical and an economical viewpoint, the implementation of universal fixed broadband access is the most important challenge within wired communication for the next 15 years. Such access may provide TV distribution, video, teleconferencing, video telephony, high-speed Internet access, electronic mail, high-speed data transmission, telephony, and a variety of other services. It may carry broadcast, multicast, as well as unicast services. There are three major alternatives for wired broadband access: Cable (cable-TV networks), DSL (Digital Subscriber Lines), and optical fibres (Fibre To The Home, FTTH). FTTH will eventually remove all access rate bottlenecks, but until recently, the equipment costs of optical fibre systems have been too high for single users. Due to new technological developments like PON (Passive Optical Networks), optical transmission systems tend to become competitive, and further significant cost reductions are expected when mass production is started. However, installation is expensive if digging is necessary, in particular in urban areas. Full deployment of fibres will probably have to be undertaken during the normal cycle of renewing underground infrastructure, which takes 30 years or more. Consequently, only a part of the broadband access lines are expected to be optical fibres in 2020. DSL is more flexible than Cable, and according to recent market analyses [PTO1], DSL is expected to be the main technology for broadband access at least for the first 5 - 10 years. The success of current DSL systems (ADSL) is due to reuse of existing infrastructure, by exploiting the existing copper cables all the way from the central office to the subscriber. In order to increase the bit rate of DSL systems, the loop lengths have to be reduced, by moving the fibre/copper interface closer to the subscribers. A significant further improvement of DSL technology is expected. The introduction of DSM (Dynamic Spectral Management) [DSM1] is predicted to double the bit rates for a given loop length compared to current DSL technology. The principles of DSM are basically the same as those of the Multiple-Input Multiple-Output (MIMO) techniques that are applied to wireless communications. There will be a variety of DSL solutions for different classes of bit rates, and up to 100 Mbit/s two-way connections are feasible (up to 500 m). Hence, DSL will be able to meet all foreseen needs within the private market segment. Due to low cost and easy handling, twisted pair cable is today the main transmission medium for in-house communication, both in homes and offices. Typically, up to 100 Mbit/s Ethernet is provided over category 5 cables (or higher) for new installations. Complete cabling is usually installed as part of the building process so that changes will occur very slowly. Hence, twisted pair cables are still expected to dominate internal cabling except for some special high-speed applications. Optical fibre technology has been present only for the last three decades. There has been an overwhelming development of the performance, and numerous new techniques have been invented. The most important breakthrough has probably been Wavelength Division Multiplexing (WDM), a technique that quickly increased the bit rate of a single fibre by one to two decades. The ongoing evolution is expected to continue [OPT1]. Further reductions of cost, and bit rates of 10 Tbit/s in one single fibre are predicted within 2020 [OPT2]. Ten Tbit/s is equivalent to more than 100 million simultaneous telephone connections, or one million full quality TV channels, and means that the maximum capacity of optical fibres may be regarded as infinite for most practical purposes. Today, optical fibres are dominating almost all long distance communications; both national and international transport networks as well as metropolitan area networks. Undersea optical cables are already crossing underneath all oceans, and most of the world’s cities are connected by fibre. During the next decade, the capacity of the existing connections will be increased, and the fibre-optical network will be extended into a much finer grid, moving much closer to all subscribers. 3 Wireless telecommunications infrastructureWithin the next 10 - 15 years, wireless communications will become almost ubiquitous, and the ability to communicate seamlessly, nomadically, and with increased mobility will be evermore important. Relevant perspectives on the development of so-called 4th generation (4G) - or “Beyond 3rd Generation” (B3G) - wireless technology are described in [WWRF, 4GMF]. Also, the International Telecommunications Union (ITU) has developed a Recommendation - ITU-R M.1645 - for B3G [ITU1], and a corresponding list of key research areas [ITU2] providing guidance for researchers wishing to contribute to B3G development. B3G will represent a convergence between wireless access, wireless mobile, wireless LAN, and packet-division-multiplexed networks [4GMF], with one integrated terminal with one global personal number being able to access freely any wireless air interface. The radio transmission modules will be fully software-definable, re-configurable, and programmable (software radio, cognitive radio). The network will be heterogeneous with regard to available and interacting technologies, content, and services; re-configurable and adaptive; with different sub-networks interconnected and with different capacity and bandwidth requirements, channel characteristics, transmission technologies, processing capabilities, desired and available services, and sub-network architectures at the different levels. The technologies, subsystems, and bandwidths involved will depend on communication distance and will include Personal Area Networks (PANs) over ultra-short distances (< 1 m), broadband WLAN technology within small cells (< 50 m), a backbone cellular system for communication over medium ranges (< 1 km), all the way up to satellite links for providing ubiquitous access to more basic services in remote and sparsely populated areas. One of the most important issues governing future development of wireless technologies, services, and business models is the regulation of the available radio spectrum. In the US , there is currently a move towards regulatory reforms that may eventually free up enough radio frequency (RF) bandwidth to significantly influence the development of mobile telephony and wireless Internet services [SPEC]. As an example, 85 MHz bandwidth previously allocated for analogue UHF broadcasting is being released for use in future mobile communications services. Digital terrestrial broadcasting (DVB-T) exploits the bandwidth far more efficiently, facilitating both more TV programs and extra bandwidth for other services. Frequency reallocation, spectrum leases, and spectrum sharing will allow the use of several new frequency bands for future mobile communications. These frequencies will find different uses depending on range and capacity needs, with the highest frequencies being reserved for high-capacity short-range communications. The currently prevailing paradigm for the implementation of B3G is thus as follows: The core network will evolve toward a TCP/IP-based core network, serving a wireless Internet radio access based on packet switching for all services, including voice [NEW]. The major trend will be a move towards higher frequencies (above 5 GHz), leading to a nano- or pico-cell structure. This will make it difficult, if not impossible, to design the network on the basis of the standard cellular concept to provide continent-wide coverage. The network will evolve towards an ad-hoc wireless network, where base stations are installed where they are needed, and connected to each other in a self-configuring way to transfer TCP/IP traffic, similarly to present Internet wired architecture [NEW]. The overall structure may thus be one of multi-layered ad hoc networks rather than one rigid network structure. Distributed high-speed WLANs will serve local hot spots, inter-connected by the overlayed backbone cellular network and by a wired infrastructure. A multitude of wireless sensors will be embedded and integrated in the network for dynamic and intelligent interaction and communication between users and devices, as well as directly between devices. B3G will deliver much higher data rates and more diverse services than 2G and 3G systems, and the design of suitable radio interfaces has to take into account that the dominant load will be high-speed burst-type traffic; a great challenge for all existing radio interface technologies. The following technology components, foreseen to become central in B3G radio interfaces, will ultimately enable a potential increase in bandwidth efficiency by a factor of 10 to 100 compared to today’s wireless systems (GSM and UMTS):
In addition, the ability to accurately measure, model, characterise, and predict the properties of many different types of radio propagation environments will continue to increase in importance with regard to successful design and optimisation of novel and more advanced radio interfaces. Furthermore, the development and commercialisation of new equipment for RF communication will call for the development of hardware with improved performance and reduced component costs. In order to avoid specification of communication standards that are difficult or expensive to implement in hardware, it is important to maintain a useful dialogue between engineers working at different levels of the standardisation process. The impact of the 3G (UMTS) technology, which is currently under deployment, is still uncertain. B3G technology is still at the research stage and will not be available until after 2010. However, B3G has an enormous potential for increase of capacity, and is expected to have a very large impact on future mobile communications. 4 Networking technologiesAs the convergence of the 4 “C” (Computers, Communication, Consumer electronics, and Content) is happening, we witness a proliferation of terminals and access forms. This puts stress on standards (formal as well as industry initiatives) in order to sort things out. In the home area this has resulted in initiatives such as Multimedia Home Platform [ETSI], [MHP], which extends the existing, successful DVB open standards for broadcast and interactive services in all transmission networks including satellite, cable, terrestrial, and microwave systems; and the Home Audio Video Interoperability initiative [HAVI], joined by eight of the largest producers of consumer products in the field. We currently witness a massive transition from traditional circuit switched networking over to Internet based technology, this also concerns signalling and control. An SRI report [VOIP] recently termed Voice over IP, and the announcements of new hardware products complying with the SIP (Session Initiation Protocol) standard, a disruptive technology threatening traditional telecommunications. The key “disruptive effect” of SIP is that it puts the much of intelligence in the terminal instead of relying heavily on central resources, thereby undermining much of the importance of Intelligent Networks (IN) principles. It can be implemented both on a LAN connected device and on any DSL line. Meanwhile the traditional incumbents and other heavy players are working busy to define the brand new network of tomorrow, the Next Generation Network (NGN), which rely on the IP protocol for transport of all services and is a combination of the Internet, and the ISDN/PSTN of today. Different vendors, providers, and operators, trying to cooperate to create new business opportunities, seek ways and means to establish both the judicial and technological interfaces needed to enable flow of services, content, and accounting information. VoIP can also be introduced on top of wireless data solutions. However, this has so far not been done in the cellular/mobile business, with the given bit rate and latency limitations. The deployment of 3rd generation mobile systems, UMTS, has just started and it is to a large extent based on the IP protocol. Wireless LAN (WLAN) is a nomadic (session mobility) type of broadband systems that is already in extensive use. The IEEE 802.11 family of standards is still being extended, and it defines a number of different WLAN systems, which are all based on the IP protocol. There are many different initiatives working towards 4th generation fully mobile systems (B3G), and one of them is the IEEE 802.20 Mobile Broadband Wireless Access initiative, that is based on similar techniques as the IEEE 802.11 family. The complexity and cost of building networked applications can be alleviated by the use of a highly flexible, efficient, dependable, and secure middleware, which is system software that resides between applications and the operating networks. Middleware as a field of component-based systems will be further developed to cater for applications that can be efficiently distributed over a myriad of computing devices and communications networks. The self-organizing property of future applications will depend on ways of effectively managing distributed/networked objects or resources. Group communications mechanisms may be developed to meet these challenges. Middleware will also be developed not only to cater for request-response communications patterns, but also for event-based modes whereby complex patterns of events can be subscribed to, and trigger actions. Such solutions will help reducing notifications latency and save bandwidth. Application areas are surveillance services and wireless sensor networks, also on an ad-hoc basis [MIDD]. 5 Services and contentThe expectations with regard to use of text messaging (SMS) and Internet access (WAP) in the GSM network illustrate the difficulties in predicting the successful future services in the communications network. Still, some observations and predictions can be made. Not contesting the placement of the intelligence, future services surely will have more intelligence. ICT will be ubiquitous and pervasive. Traditional approaches to developing services offered by communications networks have largely been specific to a single network technology, such as ISDN. This has tended to create problems in enabling services to operate across multiple network technologies. New principles for managing user-services and their QoS parameters on an end-to-end basis across many provider domains and technologies have to be developed. This is quite a challenging task, when mobility and handoffs between heterogeneous systems are considered. There will also probably be many separate domains or a hierarchy of services, as sensors and intelligent agents will aid our everyday needs and tasks, and continuous information access will be available regardless of location. Access to information services will be through a variety of devices, ranging from desktop computers and interactive TV to palm-sized wireless terminals. Available bandwidth will vary from ultra-broadband to narrowband, depending on location. There will be a host of services yielding multimedia (video/audio/speech/data...), multimodal interfaces, entertainment, games, interactivity, browsing, streaming, location-based services, position estimation/navigation, personalization, mobility, nomadic communication, ambient intelligence, virtual reality, intelligent homes, IPR issues, and billing mechanisms. This gives rise to a growing demand for specialists producing intelligent middleware solutions that ensures interoperability and assured QoS. The authorities will increasingly provide information services through the Internet. The individual citizen will be required to return information, e.g. for taxation purposes, via online forms. Interactive online processing of e.g. applications to local (and maybe national) authorities, guided by automated decision support systems, will simplify straightforward application processing. Commercial information services will continue to grow. In addition to information services, the entertainment sector will dominate commercially, ranging from low-bandwidth applications like electronic books to more demanding pay-per-view live broadcasts and movies, music, interactive games, and other multimedia content. E-commerce with virtual showrooms and on-line secure shopping is another important area. The various services and content must handle variations in bandwidth and terminal properties (e.g. screen, keypad size and type) seamlessly, providing transparent services for different communications platforms. This poses particular challenges to information presentation and to the design and capabilities of the user interfaces. General basic design principles can improve overall quality of design, in addition to providing access for all with the purpose of minimising the digital divide. The different requirements and capabilities of e.g. a laptop with 100 Mb/s connection, and a palm-sized device with connection speed of 64 kb/s, will necessitate context-sensitive presentation and user input. The human-machine interaction will exploit multiple modalities such as speech and sound, touch-sensitive screens and pull-down menus, graphics and video, and of course typed and hand-written text. The preferred combination of modalities will depend on the situation, personal preferences, and terminal size and connection speed. Speech and language technology (e.g. speech recognition and synthesis, natural language understanding) will play an important role for enabling user friendly, multi-modal user interfaces [COX]. Localisation issues, e.g. presenting the information in the user’s language and understanding the written or spoken input, will be important for the universal citizen. References[CEL1] The CELTIC initiative – Cooperation for a European sustained Leadership in Telecommunications, "CELTIC Purple Book, Part two: Technical Scope of the CELTIC Initiative," http://www.celtic-initiative.org/Publications/purple-book.asp. [DSM1] K. B. Song, S. T. Chung, G. Ginis, and J.M. Cioffi, "Dynamic spectral management for next-generation DSL systems," Commun. Mag., vol. 40, pp.101-109, Oct. 2002. [MIDD] Gordon Blair, Andrew T. Compell, Douglas C. Schmidt, “ Middleware technologies for Future Communications Networks,” IEEE Network Magazine, Jan/Febr. 2004. [OPT1] OPTIMIST consortium, "Technology Trends (in Optical Communications)," http://www.ist-optimist.org/tech.asp. [OPT2] OPTIMIST consortium, "IST Roadmap for Optical Communications, May 2002," http://www.ist-optimist.org/tech.asp. [PTO1] Point-Topic, "Broadband Analysis," http://www.point-topic.com/home/press/dslanalysis.asp. [WWRF] Wireless World Research Forum (WWRF), http://www.wireless-world-research.org/. [4GMF] 4G Mobile Forum, http://www.delson.org/4gmobile/intro.htm. [ITU1] ITU Recommendation ITU-R M.1645, http://www.itu.int/rec/recommendation.asp?type=folders&lang=e&parent=R-REC-M.1645. [ITU2] ”General focus areas for research and further study for the future development of IMT-2000 and systems beyond IMT-2000,” http://www.itu.int/ITU-R/study-groups/rsg8/rwp8f/docs/focus-areas.doc. [SPEC] G. Staple and K. Werbach, “The end of spectrum scarcity,” IEEE Spectrum, March 2004. [NEW] Description of the pan-European Network of Excellence on Wireless COMmunications (NEWCOM), http://commgroup.polito.it/newcom/, February 2004. [ETSI] ETSI ES 201 812 V1.1.1. [MHP] http://www.hmp.org/. [HAVI] http://www.havi.org/. [VOIP] The Case for Voice over Internet Protocol, Dec. 2003, D. Klemitz and M. Gold, Digital Futures, SRI, Business Intelligence. [COX] R. V. Cox et al., “Speech and language processing for next-millenium communications services,” Proc. IEEE, vol. 88, no. 8, pp.1314 – 1337, Aug. 2000.
Members of the working group: Professor Geir E. Øien (Coordinator), Department of Electronics and Telecommunications |