Challenges: A Radically New Architecture For Next Generation Mobile Ad Hoc Networks

Abstract

Despite decades of research and development, mobile ad hoc networks (MANETs) continue to lag behind wireline networks in terms of latency, capacity and robustness. We contend that a key reason for this is the way MANETs are thought about and architected today. We propose a radically new architecture that we believe will elevate MANETs to a performance plane on par with wireline networks. Our design concept for next generation MANETs is based on several revolutionary ideas - 1) a relay-oriented physical layer that selectively switches incoming packets "on-the-fly" without the intervention of the MAC or network layers, 2) a path-centric medium access mechanism that acquires the floor for not just the next, but several hops toward the destination, 3) cooperative transport of packets in stages using diversity combining. Our vision is to enable emerging and future very-low-latency, very-high-bandwidth applications to work seamlessly over large MANETs. Realizing our vision requires solving a number of challenging problems. We enumerate and briefly discuss these exciting new research areas.