There are a number of significant distinguishing features associated with wireless and mobile communication systems and networks which serve to differentiate the needs and requirements for basic research in this area from that conducted for fixed wireline communication systems and networks. While there clearly remain important similarities which can benefit from the results of a considerable body of research available on the design, deployment and operation of fixed wireline networks, it is important to identify and characterize those features of wireless mobile communication systems and networks that require new research approaches with high payoff potential. This facilitates the development of appropriate focused research priorities which can provide input to DNCRI in supporting basic research activities in this area. Furthermore, it is important that if this research is to be truly effective in addressing some of the distinguishing features and properties of wireless mobile communication systems and networks it should be posed and conducted within an appropriate systems context. This is not to suggest that only systems-oriented research should be encouraged and supported, but rather that important and basic focused research in this area can and should be directly related to the particular overall systems issues addressed.
The most prominent and distinguishing characteristics associated with wireless mobile communication systems and networks are the extraordinary premium placed on bandwidth and power efficiency, the use of inherently unreliable transmission channels and the need to provide services for mobile and nomadic users which are in some sense comparable to those accessed on fixed wireline networks. Research addressing these basic distinguishing characteristics has tended to proceed along two different, and often distinct, directions. At the one extreme we have discipline-oriented and sometimes narrowly focused communications research activities. This research is generally either focused on lower-layer issues, such as modulation and coding, or specific applications-related issues, such as audio and video compression. Typically this work makes use of signal processing concepts and employs information theory as an integrative conceptual tool to guide system design and provide performance benchmarks. At the other extreme we have more applications-oriented networking research activities which tend to focus on broader systems considerations and are generally concerned with higher-layer issues. There is much less of a discipline-oriented flavor for these activities although there is a strong emphasis on protocol design issues and a reliance on queuing-theoretic concepts. A significant characteristic feature of this work is the absence of an integrative conceptual tool analogous to the role played by information theory for communications research activities. A major conclusion of the workshop was that many of the major problems influencing the design, development and deployment of future wireless mobile communication systems and networks will require a judicious combination of discipline-oriented communications research and more applications-oriented networking research to provide truly effective solutions which address many of the distinguishing characteristics of the wireless mobile environment.
It was felt that effective research in this area should address these issues in an integrated fashion across multiple networking layers. Existing protocols for wireline networks are very limited in their ability to accomplish this; they are generally designed and implemented to provide specific and narrowly-defined services with little ability to adapt to the highly time-varying conditions associated with wireless mobile networks. What is required is an appropriate suite of adaptive, event-driven protocols that pass state information across layers in an effort to cope with this variability. Furthermore, these protocols must interact in a relatively seamless fashion with their counterparts on the fixed wireline backbone networks. To accomplish this will require a closer coupling between communications and networking research than presently exists.