Moving Beyond 4G Wireless Systems Jim Martin, Rahul Amin School of Computing Clemson University Clemson, SC USA [email protected] Ahmed Eltawil, Amr Hussien Department of Electrical and Computer Engineering University of California Irvine, CA USA [email protected] Abstract—As broadband data further blends with cellular voice, mobile devices will become the dominant portals to the connected world. However current design practices still involve building independent networks that each make their own resource decisions. To move beyond 4G, wireless networks will need to cooperate to provide better performing and more robust wireless services. In this paper, we present a ‘clean-slate’ approach to wireless networking. The approach requires the following assumptions to hold: 1)Incentives are in place motivating independent, autonomous wireless systems (AWSs) to cooperate with each other to provide users with universal broadband wireless coverage; 2)In any given geographic area, a handset might have access to many independent, autonomous wireless networks; 3)Mobile nodes will support a small number of adaptive radios that are capable of operating in a number of supported communication modes and over a range of spectrum. With these liberating assumptions, we present an architecture for a heterogeneous wireless system that moves beyond 4G systems. We present preliminary results from a MATLAB-based simulation designed to demonstrate the essence of the ideas. We identify the key research challenges that must be addressed to move the idea from a concept to a viable network system. Keywords-heterogeneous wireless networks; reconfigurable radios; 4G wireless systems; cyberinfrastructure; I. INTRODUCTION While advances at both the network layer and the physical layer are prompting unprecedented mobile broadband growth, the means by which mobile terminals are offered services have created inefficiencies in both spectrum usage as well as mobile physical resources. The premise behind this issue is that while emerging devices support a multitude of wireless access methods, the current access methods require the user to select the active access network either by purchasing an appropriate handset (and service) or, in the case of multimodal smartphones, by manually selecting the access network. While this approach has worked fairly well for current 2G cellular networks, upcoming 3G and 4G systems will face significant challenges as wireless operators are held accountable for poor performance or inadequate coverage. It is well understood that individual systems that manage blocks of spectrum independently are inevitably operating at suboptimal performance. However, there are precipitous hurdles that the wireless community faces to move forward. Until recently, technology was the primary impediment to achieving universal, broadband wireless services that involve multiple radio access technologies. Today, the most significant impediments are the after-effects of antiquated government spectrum allocation policies and the resulting economic forces that drive the wireless industry. The effect is that in many geographic areas licensed spectrum is likely to be underutilized [1]. To move beyond current thinking, spectrum, economic, and technical issues must be addressed jointly. This reality has spurred academic interest across several relevant areas including dynamic spectrum management, cognitive radios, and heterogeneous networks spanning both the physical and the network layers. The physical layer and MAC layers (i.e., the radio) of a wireless node attempts to achieve the best performance within its own network, generally ignoring impacts of co-located wireless networks. This ‘selfish’ behavior will usually not lead to optimal resource usage. Techniques or paradigms such as cooperative communications, symbiotic networking, cognitive networking, and dynamic spectrum access attempt to improve spectral efficiency through cooperation at the radio level [2]-[22]. At the network level, architectures and frameworks to support hybrid or heterogeneous networks have been suggested [24-30]. Recent proposals have been based on the IEEE 802.21 standard which provides a framework to support vertical handoffs transferring a mobile user between two networks that are based on different radio access technologies[23]. Although emerging 4G networks embody many of these recent advances, current design practices still involve building independent networks. The wireless industry is at a crossroads. The economic forces that are driving the cellular industry are reducing the number of cellular providers but causing their wireless networks to become large, heterogeneous systems based on numerous cellular data technologies at various lifecycle stages. At the same time, users are requiring more bandwidth intensive services that might not exist in all coverage areas. On the other hand, 802.11 networks have proliferated to the point that they are now considered to be a part of the computing (cyber) infrastructure. To augment wireless connectivity, some organizations or agencies have deployed site-wide, campus-wide, or city-wide broadband wireless coverage. These trends and the subsequent impacts on end users are becoming commonplace. Using Clemson University as an example, depending on the specific location on campus, one can find multiple 802.11 networks (some operated by the University, some operated by local publicsafety, others operated by the adjoining city of Clemson or nearby businesses whose coverage spills out over areas of the campus), coverage from an experimental campus WiMAX network, and multiple 2G/3G networks from cellular providers including AT&T, Sprint, T-Mobile, and Verizon. Over the next year, the spectrum on campus will become further cluttered as cellular providers deploy 4G networks. While wireless technology is advancing rapidly, research that facilitate cooperation across independent networks is relatively untouched. In this paper we present