Distributed Treansfer Learning in B5G Industrial Networks

IEEE Transactions on Industrial Informatics

In this paper, we propose a lightweight blockchain inspired framework - Magnum - as a magazine of transfer learning models in blocks. We propose the storage of these blocks on proximal fog nodes to simplify access to pre-trained base models by industrial plants to tune them before deployment. We design Magnum for B5G-enabled scenarios to reduce the block transfer time. We formulate a demand-centric distribution scheme to further reduce search and access time by adopting a Nonlinear Program model and solving it using the branch and bound method. Through extensive experiments and comparison with state-of-the-art solutions, we show that Magnum retains the accuracy of the models and present its feasibility with a maximum CPU and memory usage of 80% and 6%, respectively. Additionally, while Magnum requires a maximum of 10 seconds for writing models as large as 17 Mb on the blocks, it requires 16 micro-seconds for fetching the same.

Read More PDF

Beyond 5G/6G Networks

IEEE Communications Magazine

In this work, we put forth a philosophical treatise on a novel paradigm for an intangible, on-demand, and power-free aerial communication relaying, for which we coin the term “Phantom Networks”. The proposed Phantom Network aims to temporarily bridge communication gaps between two or more out-of-range communicating devices or infrastructures using aerial nanonodes suspended in atomized aqueous mists. This proposed on-demand shoot-and-scoot communication enabler is of particular interest to scenarios where there is a need for rapid deployment of temporary communication or relaying infrastructures for enabling higher data volume communication networks, primarily associated with military operations. As the conceptualized aerial network is primarily aqueous, its coverage and lifetime are highly dependent on ensuing atmospheric conditions in the deployment area. The possible future use of this proposed paradigm includes establishing temporary infrastructure free networks during disaster management and battlefield communications. We discuss the design and implementation challenges posed by the intended on-demand aerial nano-network based communication relaying paradigm, its application horizon, and scope for future works.

Read More PDF