We proposed a new computing and communication architecture, reflex-tree, with massive parallel sensing, data processing, and control functions suitable for future smart cities, an emerging concept with market worth of $3.3 trillion by 2025. The central feature of the proposed reflex-tree architecture is inspired by a fundamental element of the human nervous system - reflex arcs or, neuromuscular reactions and instinctive motions of a part of body in response to urgent situations. While the brain is the central controller of body activities, a local neural reflex arc is able to cause an individual to pull their hand away from a source of discomfort without the need for direct control from the brain. Our reflex-tree architecture mimics such human neural circuits - a massive amount of intermediate computing nodes, edge devices, and sensors work in a parallel and cohesive fashion to gather, process, and, most importantly, to react to data concerning critical infrastructure elements. This processed information is ultimately passed up a 4-level hierarchy to a centralized cloud system, providing the intelligence and control necessary to successfully manage future smart cities.

Key outcomes of the proposed reflex-tree architecture include: 1) The creation of a novel, 4-level, large scale, and application-specific hierarchical computing and communication structure capable of carrying out sensor-based decision-making processes. The required computation and nodal computing power increases at each successive stage in the hierarchy, with the level-1 cloud performing the most complex tasks. 2) At each level of the tree, special parallel sensing methods and parallel machine learning algorithms have been developed targeting smart city applications. This novel fiber-optic sensing network provides distributed strain and temperature sensing with unprecedented spatial/temporal resolution (~mm, > 50 Hz) enabling massive, parallel, real-time sensing. 3) Novel, complementary machine intelligence algorithms were developed, which provide accurate control decisions via a multi-layer adaptive learning, spatial-temporal association, and complex system behavior analysis. 4) New hardware architectures and software run-time environments were developed that are specifically tailored to our new computing algorithms for the reflex-tree system. We investigated both intra-node and inter-nodes parallelisms across all 4 levels of the reflex-tree. The resulting system is a cross layer, high performance, and massively parallel computing platform providing a foundational sensing and computer architecture for future smart cities. 

This study has broad impacts on the sustainability, safety, and human welfare of urban areas as it represents a leap forward for future smart cities. The scientific foundation and engineering framework built by this project will pave the way for enhanced monitoring and management of critical smart city infrastructure, from gas/oil pipelines, water management, communication networks, and power grids, to public transportation and healthcare. Additionally, the interdisciplinary collaborative nature will also inspire broader participations in smart city research.  

Last Modified: 03/30/2021
Modified by: Tao Wei