A Hybrid Computing Framework Enabled by AI and Blockchain in Fiber-Wireless Networks

Abstract

Centralized cloud storage, computing, and processing can not fulfill the quality of service (QoS), e.g., high bandwidth and low latency, at reduced costs to emerging applications, such as virtual reality, smart transportation, etc. This paper proposes a hybrid computing framework that can use the best of cloud, fog, and edge capabilities and can be tailored to the needs of applications. Note that fog and edge computing infrastructures, designed for proximity to users, have emerged as promising complements to traditional cloud computing. While fiber access networks can offer QoS to end users connected over fixed devices, an integrated fiber-wireless access network can fulfill mobile end-users requirements. Our framework integrates artificial intelligence techniques for resource optimization, control, and monitoring, among other tasks. Furthermore, we envision that edge devices and sensors, together with fog and cloud, can utilize a distributed ledger technology, such as Blockchain, to establish a trust-based relationship. Finally, we present a use case of hybrid computing over a distributed fiber-wireless network enabled by AI and Blockchain. We show that hybrid computing is essential in balancing a tradeoff between communication latency to generate alert signals and minimizing the power consumption of a system.