[Research Overview] [Department Overview] [List of References]

The Information and Communication Research Program in EPP is probably the strongest research group in the country working on the policy aspects of advanced telecommunications and information technologies. New computer and information technologies are turning traditional telecommunications on its head. In the past, communications networks were optimized to carry voice. Using modems, data were made to look like voice and sent over the same networks. Today we are in the midst of a dramatic transformation. Future networks will be optimized for data. Voice will be treated as just another data stream, somewhat time critical, although nowhere near as demanding as real-time video and similar broadband services. This transformation poses wrenching technical, strategic, and economic challenges for carriers. It also presents fundamental regulatory challenges to international, national and state regulatory structures, designed for an earlier era. Research in EPP conducted by Marvin Sirbu, Jon Peha, Alex Hills and others is addressing a variety of these issues. How should various services be classified in the future? What services, if any, should be regulated and how? At the same time, wireless service is growing rapidly. How should technologies and standards be chosen? How should spectrum be managed in this new environment? The Internet is rapidly becoming an important vehicle for commerce. However, most commerce-related laws and regulations, such as state sales taxes, were not designed with this in mind. How should internet commerce be regulated? Should it be taxed, and, if so, how? New information technologies raise fundamental questions of reliability, security and privacy. It is much easier to deal with such issues in the early design phases of such systems, but competitive markets provide few incentives for private parties to address such social externalities.

A particular concern of current telecommunication research in EPP involves the future of the telephone. The Internet is essentially unregulated. In contrast, local and long distance telephone service is extensively regulated. As voice moves to the Internet, regulatory arrangements built on concepts such as the "subscriber line" will no longer make sense. In a system optimized for data, there will no longer be individual voice lines. An individual home with a broadband connection could easily have several "virtual" voice lines, depending upon the needs of the moment. Traditional economic arrangements, such as cross subsidies to support universal service in sparsely populated regions, will also have to undergo major transformations. Marvin Sirbu and Jon Peha are both concerned with the economics of gigabit networks. They have explored cost-based schemes for resource allocation and scheduling in packet networks. Other related work includes studies of the regulation, pricing, and market penetration of new networked services, work on spectrum allocation and competition in satellite-based mobile communication services, Personal Communications Networks, and the economics of replacing existing local telephone lines with integrated access to networks over twisted pair, cable or via wireless. Marvin Sirbu and his students have conducted extensive engineering/economic assessments of the use of fiber optics to provide both video and telephony services to residences. Developments in fiber technology substantially undermine the premises of current regulatory policies governing telephone and cable television companies, and raise important questions to be answered by public policy makers. Jon Peha has conducted research on the promise and pitfalls of spectrum sharing. In recent years, regulators in several countries have turned to auctions, in which a license for exclusive access to spectrum goes to the highest bidder. Simultaneously there have been more opportunities to share spectrum without a license. Professor Peha has evaluated these and other proposals. As computers evolve from desktop to laptop to personal digital assistants, demand grows for wireless data communications services to interconnect these devices regardless of their locations. Alex Hills has lead a university effort to build a campus-wide wireless test bed called "Wireless Andrew," which, in addition to serving the needs of the Carnegie Mellon community, is being used to study technical, economic, and policy issues relevant to wireless networks and their applications. Wireless Andrew is the largest installation of its type anywhere. Started as a research network in 1994 to support Carnegie Mellon's wireless research initiatives, it has been dramatically expanded into a full-production network, providing highly reliable service to the entire campus community of 10,000 faculty, students and staff. The significant challenges in building such a large network have largely been solved through Carnegie Mellon research efforts. They included designing the network so that coverage is continuous throughout the campus and so that adequate capacity is provided to handle the traffic load generated by the campus community. Further, network management must be provided so that the network provides highly reliable service. Interoperability standards in computers and communications have long been a focus of research interest in EPP. In markets where standards are important, they alter normal patterns of product innovation and introduction; indeed, they may be critical to the successful establishment of new telecommunications services, such as HDTV. Faculty and students have examined the technical, economic, and social dimensions of standards development, and the appropriate role for government in the standards development process. Developments in computers, networking, and storage technology may finally render feasible the dream of the electronic library and the electronic dissemination of research publications. Working with doctoral students in EPP, as well as students and staff from Computer Science, the Information Networking Institute, and the Carnegie Mellon libraries, the program has examined the provision of scholarly journals directly to desktop workstations. The realization of the electronic library raises important questions of industry organization and copyright policies. It also raises interesting questions of when to bundle and unbundle the pricing of information services, and economies of scale and scope in the electronic delivery of information. Electronic commerce provides a rich environment for exploring a number of critical issues in highly survivable, secure, and available distributed systems. Under the leadership of Marvin Sirbu, the NetBill project, conducted as a joint effort of EPP, Computer Science, and the Information Networking Institute during the mid 1990's examined these issues through research, development, and implementation of an operating test system called NetBill. In the NetBill system design, NetBill acts as a third party to provide the authentication, account management, transaction processing, billing, and reporting services for network-based clients and users. With a NetBill account and client software, users can buy information, software, CPU cycles, or other services from NetBill-authorized service providers, under a variety of payment schemes. NetBill acts like an electronic debit card service to provide financial services in support of electronic commerce. To meet the requirements of flexibility and scalability, and so that independent parties could easily be involved in commerce transactions, NetBill was designed as a system of systems. It depends on an infrastructure of authentication, certificate management, internet access (including DNS lookup), databases, real-time customer service and dispute resolution servers, etc. Issues of establishing trust among independently operated systems, making operations robust against network and system failures, and survivability in the event of compromise of participating systems were all explicitly addressed. High fault-tolerance was demonstrated through the use of strong ACID (atomic, consistent, isolated, and durable) transactions. NetBill has now been licensed to CyberCash. Additional details can be found at http://www.netbill.com. Many policies that have evolved to govern face to face commercial transactions are not consistent with the new world of electronic commerce. Jon Peha and his colleagues have explored methods that can make taxation of electronic commerce possible without sacrificing privacy. Several faculty in EPP have worked on issues of telecommunications in the industrializing world. Jon Peha recently spent a year on leave to help establish a new US government program that assists developing countries on information infrastructure. He has advised around the world on telephone, internet, and wireless technology and corresponding policy issues, such as privatization, policies to serve rural areas, and spectrum management. He has worked extensively in Africa, Asia, Latin America and the Caribbean. V. S. Arunachalum is leading an effort to build a high-speed internet backbone among leading universities and research institutes in India. This project involves an inventive new approach in which the government of India is a minority partner in a private sector enterprise. Alex Hills has helped to link research organizations in the Russian Far East with the University of Alaska. He has also developed recommendations for the delivery of communications via wireless technologies in underserved cities in the industrializing world, as well as addressing the problems of providing telecommunications service in rural and remote areas. While this area of research in EPP began with a strong focus on telecommunications, recent years have seen a shift to a much broader set of issues in information technology and public policy. Several recent Ph.D. theses have addressed issues in data system security and reliability. Over the next few years, the Department plans to expand its computer- and information-system related research to include work on issues such as: the social dimensions of ubiquitous computing; strategies to infuse social concerns (privacy, security, interoperability, etc.) into the early specification and design stages of software development; software, institutional and regulatory issues related to interoperability, reliability, security, and privacy; studies of the social, regulatory, and legal issues related to automated agents; software issues in international security (including export control); and a number of issues in technology and R&D policy related to computer and information technology.