The Intelligent Wireless Web

Chapter 10.0

Progress in Developing

the Intelligent Wireless Web

In this chapter, we take the components developed in earlier chapters and lay out a plausible framework for building the Intelligent Wireless Web including our evaluation of the compatibility, integration and synergy issues facing the five merging technology areas that will build it:

User Interface – from click to speech
Personal Space – from tangled wires to multifunction wireless devices
Networks – from wired infrastructure to integrated wired/wireless
Protocols – from IP to Mobile IP
Web Architecture – from dumb and static to intelligent and dynamic.

Finally, we present strategic planning guidelines and the conclusions you could reach as a result of this book.

We began this book by describing what we meant by the “Intelligent Wireless Web and presenting an overview of the framework for plausibly constructing it. Our concept of an Intelligent Wireless Web weaves together several important concepts related to intelligence (the ability to learn), “wirelessness” (mobility and convenience) and its advances in telecommunications and information technology that together promised to deliver increasingly capable information services to mobile users anytime and anywhere.

We suggested putting these concepts together to form the “Intelligent Wireless Web.” We stated that it was certainly possible to develop intelligent applications for the Internet without media (audio/video) Web features or wireless capability. But, it was our suggestion that Web media, such as, audio could lead to improved user interfaces using speech and that small wireless devices widely distributed could lead to easier access to large portions of the worlds population. The end result could be, not just an intelligent Internet but a widely available, easily accessible, user friendly, Intelligent Wireless Web.

Fundamentally, our vision for an Intelligent Wireless Web is very simple - it is a network that provides anytime, anywhere access through efficient user interfaces to applications that learn. Notwithstanding the difficulty of defining intelligence (in humans or machines), we recognized that terms such as “artificial intelligence”, “intelligent agents”, “smart machines” and the like, refer to the performance of functions that mimic those associated with human intelligence.

The full range of information services is the next logical step along with the introduction of a variety of different portable user devices (e.g., pagers, PDAs, web-enabled cell phones, small portable computers) that have wireless connectivity. The results will be wireless technology as an extension of the present evolutionary trend in information technology. In addition, Artificial Intelligence and intelligent software applications will also make their way onto the wireless Web and that a performance Index or measure should be developed to evaluate the progress of Web smarts.

In the following sections, we will bring together the components of the Intelligent Wireless Web and how it is being constructed. But building it will be a broad and far-reaching task involving more technology integration and synthesis than revolutionary inventions.

Future Wireless Communication Process

Ideally, the future wireless communication process should start with a user interface based on speech recognition where we merely talk to a personal mobile device that recognizes our identity, words and commands. The personal mobile device would connect seamlessly to embedded and fixed devices in the immediate environment. The message would be relayed to a server residing on a network with the necessary processing power and software to analyzed the contents of the message. The server could then draw necessary supplemental knowledge and services from around the world through the Internet. Finally, the synthesized messages would be delivered to the appropriate parties in their own language on their own personal mobile device.

To build this ideal future wireless communication process we must connect the following inherent technologies of communications along with their essential components:

Connecting People to Devices – the user interface. Currently we rely on the mouse, keyboard and video display; speech recognition and understanding deployed for mobile devices is a key component for the future.

Connecting Devices to Devices. Currently hard-wired connections between devices limit mobility and constrain the design of networks. In the future, the merging of wired and wireless communication infrastructure require the establishment of wireless protocols and standards for the connection between devices; future smart applications require the development and improvement of intelligence services. Also needed is a method to measure the performance and/or intelligence of the Internet so that we can assess advancements.

Connecting Devices to People. To deliver useful information to the globally mobile user, future systems require advances in speech synthesis and language translation.

So lets start connecting the necessary technologies to fulfill the vision of an Intelligent Wireless Web.

The physical components and software necessary to construct and implement the Intelligent Wireless Web require compatibility, integration and synergy of five merging technology areas:

< >User Interface – to transition from the mouse click to speech as the primary method of communication between people and devices;Personal Space – to transition from connection of devices by tangled wires to multifunction wireless devices;Networks – to transition from a mostly wired infrastructure to an integrated wired/wireless system of interconnections;Protocols – transition from the original Internet protocol (IP) to the new Mobile IP; andWeb Architecture – to transition from dumb and static applications to new applications that are intelligent, dynamic and constantly learning.

FIGURE 10-1 Building the Intelligent Wireless Web

User Interface – from Click to Speech

We have evaluated communication between humans and their machines and found the problem of how to obtain speech recognition functionality in a handheld or embedded device to be challenging; however efforts currently underway look favorable for solutions in the relatively near term. While we may expect speech interfaces to permeate society steadily, we anticipate that successful traditional interfaces, such as, mouse and touch screen, will continue to be in operation for a very long time. Particularly, for such high power applications as selecting events on detailed graphical representations.

Certainly, it is not a difficult problem for a handheld device (such as a cell phone) to perform limited speech recognition activities (such as voice activated dialing). But since the demands for speech functionality increase with the greater complexity of the speech recognition tasks, it becomes more and more difficult to provide these capabilities on a small mobile wireless device with limited capabilities. Therefore, the problem becomes one of distributing the capability for speech recognition and understanding between the local wireless device and remote processing resources to which it is connected.

This problem is being currently addressed in far-reaching research at several places, but most notably at the MIT AI Laboratory and at Microsoft Research. The Microsoft effort is directed at technology projects supporting and leading to the vision of a fully speech enabled computer. The Microsoft concept Dr. Who, uses continuous speech recognition and spoken language understanding. Dr. Who is designed to power a voice-based pocket PC with a web browser, email, and cellular telephone capabilities.

The highly promising initiative know as, Project Oxygen, is ongoing at MIT’s AI Laboratory. This visionary effort is developing a comprehensive system to achieve the objective of anytime anywhere computing. In this concept, a user carries a wireless interface device that is continuously connected to a network of computing devices in a manner similar to the way cell phone communications maintain continuous connection to a communications network. The local device is speech enabled, and much of the speech recognition capability is embedded in the remote system of high-capability computers.

Systems for conversational interface are also being developed that are capable of communicating in several languages. These systems can answer queries in real-time with a distributed architecture that can retrieve data from several different domains of knowledge to answer a query. Such systems have five main functions: speech recognition, language understanding, information retrieval, language generation and speech synthesis.

Speech recognition may be an ideal interface for the handheld devices being developed as part of the Oxygen project, but the Oxygen project will need far more advanced speech-recognition systems than are currently available to achieve its ultimate objective of enabling interactive conversation with full understanding. Figure 10-2 identifies the main requirements for an effective speech-based user interface and identifies the current status of each. To meet the needs of the Intelligent Wireless Web, the ultimate desired result is that speech recognition, understanding, translation and synthesis become practical for routine use on handheld, wearable and embedded devices.

USER INTERFACE – from click to speech

REQUIREMENTS STATUS

Speech Recognition

Speech Understanding

Text to Speech

Translation

Speech Synthesis

Speech Synthesis Markup Language

Advanced

Continuing

Advanced

Continuing

Lagging

Speech recognition, understanding, translation and synthesis become practical for use on handheld, wearable and embedded devices.

RESULTS

FIGURE 10-2 Building the User Interface

Personal Space – from Wired to Wireless

We imagined living our life within the confines of our own Personal Space - without wires, but with devices to “connect” us wherever we travel. Implementation of a Wireless Personal Area Network (WPAN), composed of the personal devices around us as well as our immediate environment is one solution. In the office, devices improve work productivity by enabling access to data, text, and images relating to performing our jobs and by providing for analysis, access to software applications and communications as needed. Creating a WPAN of our immediately available devices will enable a future where a lifetime of knowledge may be accessed through gateways worn on the body or placed within the immediate environment (including our home, auto, office, school, library, ect).

WPAN will also allow devices to work together and share each other's information and services. For example, a web page can be called up on a small screen, and can then be wirelessly sent to a printer for full size printing. A mobile WPAN can even be created in a vehicle via interface devices such as wireless headsets, microphones and speakers for communications.

As envisioned, WPAN will allow the user to customize his or her communications capabilities permitting everyday devices to become smart, tetherless devices that spontaneously communicate whenever they are in close proximity. Figure 10-3 summarizes the requirements and their status for this element of the Intelligent Wireless Web; the objective is the achieve the ability for handheld, wearable, and embedded devices to connect easily without wires and share software applications, as needed, producing office, home and mobile Wireless Personal Area Networks.

PERSONAL SPACE – from wired to wireless

REQUIREMENTS STATUS

Advanced

Continuing

Lagging

Adaptable wireless devices

Wireless protocol

Wireless small screen applications

“Nomadic” or mobile software for devices

Handheld, wearable, and embedded devices connect easily without wires and share software applications, as needed, producing office, home and mobile Wireless Personal Area Networks.

RESULTS

FIGURE 10 - 3 Building the Your Personal Space

Networks – from Wired to Integrated Wired/Wireless

The earliest computers were stand-alone, unconnected machines. During the 1980’s, mergers, takeovers and downsizing have led to a need to consolidate company data in fast, seamless, integrated database have for all corporate information. With these driving forces, Intranets and local networks began to increase in size, and this required ways to interface with each other. Over the past decade, enterprise models and architectures, as well as, their corresponding implementation in actual business practices have changed to take advantage of new technologies.

The big lure to wireless is the potential for big money in implementing wireless architectures that can send information packets from people with small personal devices, such as cell phones, to the a company’s Web site and there to conduct transactions. The number of wireless subscribers is expected to grow globally from the current few million to more than 400 million by 2005.

The vast system of interconnecting networks that comprise the Internet is composed of several different types of transmission media, dominated by wired media but including:

< >WiredFiber opticTwisted pairs (copper)Coaxial cable < >WirelessMicrowaveInfraredLaser

NETWORKS – from wired to integrated wired/wireless

REQUIREMENTS STATUS

Wireless LAN

Wireless WAN

Satellites

Wired Interface

Advanced

Continuing

Networks continue migration to optical fiber for long haul while last mile is met by both fiber, mobile wireless, and fixed wireless (LMDS & MDDS)

RESULTS

FIGURE 10-4 Building Integrated Networks

Protocols – from IP to Mobile IP

To achieve the mobility requirements of the Intelligent Wireless Web, the Wireless Appliance Protocol, WAP, provides a global standard for data-oriented services to mobile devices thereby enabling anywhere and anytime access. In so doing, access will be provided to far more end-users than can be reached by using the personal computer as a fixed end point. Figure 10-5 provides an overview of the needed changes to support the Intelligent Wireless Web. The anticipated result is to provide intelligent networking software for routing and tracking that leads to general changes in IP networking protocols toward mobile IP. Sitting on top of the entire layer infrastructure will be a new control-plan for applications that smooth routing.

PROTOCOLS - from IP to Mobile IP

Continuing

IPv6

Mobile IP standard

REQUIREMENTS STATUS

Intelligent networking software for routing and tracking that leads to general changes in IP networking protocols toward mobile IP. Sitting on top of the entire layer infrastructure will be a new control-plan for applications that smooth routing.

RESULTS

FIGURE 10 - 5 Building the Mobile Internet Protocols

Web Architecture - Dumb & Static to Intelligent & Dynamic

Ideally, the wireless communication process should start with the user talking to a personal, or embedded, device that recognizes his identity, words and commands. It will connect seamlessly to the correct transmission device, drawing on whatever resources are required from around the Web. In one case, only database search sorting and retrieval might be required. Or in another case, a specialized Web Service application program might be required. In any case, the information will be evaluated, and the content of the message will be augmented with the appropriate supporting data to fill in the ‘blanks’. If there is appropriate supplementary audio, or video, it will be included for reference. Finally, the results will be delivered to the appropriate parties in their own language through their own different and varied connection devices.

For the Web to learn how to conduct this type of intelligent processing requires a mechanism for the adapting and self-organizing on a hypertext network. In addition, it needs to develop Learning Algorithms that would allow it to autonomously change its structure and organize the knowledge it contains, by "learning" the ideas and preferences of its users.

The World Wide Web Consortium (W3C) suggests the use of better semantic information as part of web documents, and of the use of next generation Web languages Figure 10-6 provides a summary of the semantic web architecture needed to support the Intelligent Wireless Web. Intelligent applications running directly over the Web, as well as, AI Web Services served from AI service providers will progressively increase the tasking performed with adaptive, dynamic intelligent products. In addition, a Web performance Index will provide some useful measures of Web progress.

WEB ARCHITECTURE – from dumb and static to intelligent and dynamic

REQUIREMENTS STATUS

XMLschema

RDF schema & Topic Maps

Logic Layer

Dynamic Languages

Adaptive Applications

Distributed AI

AI Web Sevices

Registration and Validation of Information

Intelligent applications running directly over the Web, as well as, AI Web Services supported from AI service providers progressively increasing the percent of applications performed with adaptive, dynamic intelligent products. An overall increase can be expected in the total percentage of learning algorithms operating on the Web.

RESULTS

FIGURE 10- 6 Building AI Servers with the Semantic

Strategic Planning Guidelines

Strategic planning is the determination of the course of action and allocation of resources necessary to achieve selected long-term goals. But charting strategic direction for wireless communications networks in a diverse and competitive landscape is complicated by an economy that has introduced dynamic rules for success. Both the rate of technology change and the speed at which new technologies become available have increased. The shorter product life cycles resulting from this rapid diffusion of new technologies places a competitive premium on being able to quickly introduce new goods and services into the marketplace. In order to develop guidelines for strategic planning, we must consider enterprise goals.

Traditionally driven by technology, network planning has evolved and now faces new challenges. But the network planning process itself includes two "discordant" requirements: first, to optimize of the network’s long-term investment while second, optimizing of the time to market for each new product. Finding the right balance is not easy. However, opportunities for developers and service providers will exist if they can reach all mobile users by developing infrastructure to support:

< >any wireless carrierany wireless network (TDMA, CDMA, etc.)any wireless device (pager, digital cell phone, PDA)any wireless applicationsany Web format (XML, HTML, etc.)any wireless technology (WAP, SMS, pager, etc.)any medium (text, audio, text-to-speech, voice recognition or video)balancing innovations in software (e.g. adaptive software, nomadic software) against innovations in hardware (e.g. chip designs), balancing proprietary standards (motivating competition) against open standards (offering universal access), and balancing local(centralized) Web innovations (e.g. Web Services) against global(distributed) Web architectural evolution (e.g. the Semantic Web).A vendor dominates a market and sets a de facto standard (for example; POTS telephony from AT&T, or PC operating systems from Microsoft).Standards organizations establish standards (for example; HTML).Vendor and market collaboration that is not clearly attributable to any one organization (for example; TCP/IP or VCR formats).

FIGURE 10-7 Possible Technology Timeline

Conclusion

In this chapter, we presented the components developed in earlier chapters and outlined a feasible framework for building the Intelligent Wireless Web, including our evaluation of the compatibility, integration and synergy issues facing the five merging technology areas: User Interface, Personal Space, Networks, Protocols, and Web Architecture.

Ten conclusions you could reach from this book about building the Intelligent Wireless Web include:

- User Interface -

< >Speech recognition and speech synthesis offer attractive solutions to overcome the input and output limitations of small mobile devices, if they can overcome their own limitation of memory and processing power through the right balance for the client-server relationship between the small device and nearby embedded resources. The essential components for achieving this balance are new chip designs coupled with open adaptive nomadic software. The new chips may provide hardware for small devices that is small, light weight, and consumes little power while having the ability to perform applications by downloading adaptive software as needed.- Personal Space -

< >Handheld, wearable and embedded devices are upgrading many existing office and home locations making computing access more universal through Wireless Personal Area Networks.Competition between the wireless networking standards Bluetooth and IEEE 802.11b, as well as general networking software, Jini and UpnP, will continue for several years as each finds strong points to exploit before a final winner emerges. MIT’s Project OXYGEN may introduce some innovative protocol alternatives within several years.

- Networks -

< >Wired and wireless networks will continue to merge and improve backbone performance to greater than the 10 Tera-bps range as well as produce improved interoperability. < >Over time, there will be a migration of core networks to optical fiber simply because photons carry a lot more information more efficiently and at less expense than electrons. By 2003, ultra-long haul (> 4000 km) high bandwidth optical transport will be deployed in the US. The quest for the last mile will be met with a combination of fiber and wireless. In dense metropolitan areas free-space optical networks will provide 622Mbps of bandwidth to buildings without digging the streets. Second generation LMDS and MDDS fixed wireless will be deployed to buildings requiring less bandwidth.- Internet Protocols -

< >Intelligent networking software for routing and tracking will lead to general changes in IP networking protocols to include IPv6 and mobile IP. Sitting on top of the entire layer infrastructure may be a number of new control-plane software applications that may add intelligence to the network for smooth integration of routing (layer 3) and wavelength switching. - Web Architecture -

< >Intelligent agents, intelligent software application and Artificial Intelligence applications from AI Servers

Providers may make their way onto the Web in greater numbers as adaptive software, dynamic programming languages and Learning Algorithms are introduced into Web Services (including both .NET and J2EE architectures).The evolution of Web Architecture may allow intelligent applications to run directly on the Web by introducing XML, RDF/Topic Maps and a Logic Layer.A Web performance Index, or measure, may be developed to evaluate the progress of Internet progress in performing intelligent tasks utilizing learning algorithms.The Intelligent Wireless Web’s significant potential for rapidly completing information transactions may become an important contribution to global worker productivity.

1 [1] Bogdanowicz, K.D., Scapolo, F., Leijten, J., and Burgelman, J-C., “Scenarios for Ambient Intelligence in 2010,” ISTAG Report, European Commission, Feb. 2001.