According to the mainstream of the Sixth Generation (6G) communication paradigm, the incoming IoT big data era requires efficient and resource-saving mining of large sets of distributed data through the seamless integration of heterogeneous terrestrial and aerial (e.g., drone supported) wireless subnetworks. To this end, a viable approach could be to attempt to resort to the synergic exploitation of two emerging paradigms, namely, Deep Neural Networks (DNNs) with early exits and Fog Computing (FC). From an IoT perspective, the most appealing feature of DNNs is their hierarchical layer-wise organization, potentially matching the hierarchical organization of the underlying Fog-supported execution platform. However, both the training and inference phases of current DNNs are too cumbersome to be executed by IoT devices, and their remote and centralized implementation induces large communication delays and is bandwidth-wasting. Motivated by these considerations, this proposal plans the activity for the SECOND year of the (two-year) "DeepFog" project. The project aims at developing the integration of the emerging paradigms of the Conditional DNNs (CDNNs) with early exits and FC. For this purpose, the exploitation of the powerful and recent paradigm of Federated Learning (FL) is also exploited for the distributed and asynchronous Fog-supported (and, possibly, drone-aided) training of the underlying CDNN under resource and fairness constraints. Overall, according to the 6G vision, main goals of the DeepFog-2 project are to complete the design and validate in SW the related algorithms already developed during the FIRST research year, by focusing on the:
- formalization of the main building blocks and functionalities of the overall DeepFog technological platform;
- design of a novel asynchronous fast-adaptive FL protocol for the training of CDNNs for Fog supported IoT stream applications;
- design and implementation of the Flying Federated Learning (FFL) toolbox.
5) EXPECTED INNOVATIVE PROJECT OUTPUTS AND THEIR POTENTIAL TECHNO-SOCIAL IMPACTS
The DeepFog technological platform is expected to quickly foster the development of Fog-supported Future Internet Applications [11], that are considered a world-wide huge business opportunity (Fig.3). In particular, it is foreseen that both DL applications and Fog related outputs could move from the research cycle to prototype cycle in the next years, since in between five years we will expect six times the software revenue with respect today (Fig.3a).
The main services that are expected to be fostered by the resulting IoT-CDNN-FC converged platform promoted by the DeepFog paradigm are summarized in Table 2, together with some possible supported applications. Specifically, it is expected that the IoT-CDNN-FC convergence highlights three benefits: i) the demand for real-time analytics instead of batch processing at remote Cloud data centers; ii) allowing the acquisition and joint mining of data generated by spatially scattered IoT devices under both real-time and fairness requirements; and, iii) the use of resource-limited IoT devices supporting the planned services of Table 2. All these benefits well match with the native features of both CDNN and FC paradigms, namely, their local processing capability and their inherently distributed and scalable nature.
Expected Technological Impact: The proposed DeepFog project is expected to have a significant positive impact on the development of future low-latency and energy-aware intelligent applications done feasible by the convergence of FC and DL paradigms. This is expected to have a relevant technological impact on many novel families of technologies, such as Drone-supported Ground-Aerial Networks, Smart Objects of IoT and Internet of Flying Things Overall, it is expected that the proposed DeepFog project allows to enter a new phase where real-world problems emerging from complex real-time IoT applications are addressed.
Expected Social-Economical Impact: It is expected that intelligent FC platforms are able to provide more robust, energy-efficient, lower latency and higher QoS applications and, for this reason, are expected to increase the trust people have in machine-related and real-time critical applications. Furthermore, since the novel DeepFog platform relies on FL protocols that does not need to share local data, it is expected that it solves data security and related privacy issues. Overall, it is expected that the proposed DeepFog paradigm will provide a feasible technological platform for the actual support and real-world implementation of a large overall spectrum of Fog-supported DL-aided IoT-oriented stream applications.
6) PROJECT PLAN AND PROJECT TARGETS
The overall two-year research activity of the overall DeepFog workprogramme is organized into two Workpackages, coordinated Prof. M. Scarpiniti and by Prof. E. Baccarelli, respectively. As detailed in Table III, each WP comprises two tasks.
The current DeepFog-2 proposal covers the SECOND year of the overall DeepFog project. It engages 4 researchers (3 promoters supported by 1 PhD student) and is coordinated by Prof. Michele Scarpiniti. The multiple scientific sectors involved by the DeepFog-2 proposal (ING-INF/03 and ING-IND/31) confirm, indeed, its inter-disciplinary nature. Table IV sketches the tasks completed during the FIRST year and the tasks scheduled for the SECOND year, i.e, the tasks covered by the current DeepFog-2 proposal.
From Table 4, we argue that main targets to be covered by the DeepFog-2 proposal during the SECOND year of the overall DeepFog project concern to:
a) complete the formalization of the main building blocks and functionalities of the proposed DeepFog technological platform
b) design of a novel asynchronous fast-adaptive FL protocol for the training of CDNNs for Fog supported IoT stream applications
c) design and implement the Flying Federated Learning (FFL) toolbox, which allows the optimized design and energy-vs.-training delay performance evaluation of an overall FL ecosystem in which the central Fog server is mounted atop a drone, whose 3D flying trajectory may be optimally co-designed together with the corresponding computing-plus-communication resource scheduling (see Fig. 2).