A cross-layer architecture for next-generation wireless systems
Carnegie Mellon University's CROSSMobile architecture is enabled by controlled cross-layer information exchange between radio, network and application layers (both on-device and in-cloud), coupled with information-owner–based privacy and security controls. We are actively seeking partners to join us in the creation of the next-generation wireless test bed.
- Wireless network system performance suffers, because phones and mobile apps conceal important information from each other
Commercial mobile network architectures emerged from their wired counterparts in an evolutionary manner. Market pressures fueled upgrades in bandwidth and functionality, but decades later, these networks maintain historical artifacts from wired networks. The artifacts work against the fundamental needs of mobile systems.
- Share information across network layers in a controlled way
- Apply machine learning to optimize system performance
- Embed virtualized computing in the network
- Prototype network functional — Fall 2013
With commonly available hardware, open-source software and software-defined radio, we’ve constructed, at lab-scale, a network supporting cellular calls, text messages, data transmission and an application-optimized virtual computing framework. To optimize performance, a central server collects data across network layers and, using machine learning algorithms, can be used to intelligently manage computing and communication resources in real time.
"CROSSMobile: A Cross-Layer Architecture for Next-Generation Wireless Systems" by Bob Iannucci, Patrick Tague, Ole Mengshoel, and Jason Lohn
Silicon Valley Technical Report Series CMU-SV-14-001 — Download PDF