Edge Computing and Communications for NextG: Fundamental Limits and Applications

In this talk, we review recent work on edge computing and communications as a key enabler for future wireless networks. In particular, we present samples of our work on proactive caching, decentralized coded caching, cache-aided MIMO networks and federated learning over blockchain networks. In the first part of the talk, we shed light on a sample of our work on edge content caching.

We present performance limits for decentralized coded caching in Fog Radio Access Networks with focus on the normalized delivery time (NDT) performance metric. Afterwards, we extend our discussion to cache-aided MIMO networks where we generalize prior studies to the case of arbitrary number of transmitters, receivers and antennas.

We establish fundamental limits characterizing the relative contributions of the spatial multiplexing gain vs. the coded caching gain, with respect to reducing the delivery latency. In the second part of the talk, we shift our focus to edge intelligence, in particular, federated learning over blockchain networks.

Motivated by the wide prevalence of decentralized data, we propose a novel federated learning framework with no parameter service, yet, relying on the distributed ledger security infrastructure of blockchains. This is achieved by employing “validators” who evaluate the quality of the “trainer” model updates through a validation dataset and byzantine fault tolerance.

This work opens ample room for future research through revealing interesting insights and fundamental trade-offs. Co-sponsored by: UTEC Prof. org/m/566697 \n