Hydra Radio Access Network (H-RAN): Multi-Functional Communications and Sensing Networks, Collaboration-Based SRU Switching

The demand for an innovative cohesive network that integrates a variety of networks, services, and modern technologies is becoming increasingly urgent. This integration would facilitate more streamlined operations across diverse networks and applications, reducing the overhead associated with controlling and managing multiple, distinct networks. Such a unified infrastructure will support the growth of interconnected applications, which is essential in today's digital landscape. Hydra radio access networks (H-RANs) are engineered to achieve future research goals by seamlessly integrating communication and sensing functionalities into a cohesive network with extensive utilization of artificial intelligence/machine learning (AI/ML). Ultimately, this evolution towards a comprehensive integrated network is essential for meeting the multifaceted requirements of contemporary applications. By converging various technologies and services, H-RANs can develop more sophisticated systems that adapt to changing requirements, thereby ensuring a forward-looking approach in a highly dynamic digital landscape. In this paper, we propose a novel H-RAN cooperative multi-sparse input/multi-task learning-based federated learning (C-SMTL) paradigm that expands cooperation dimensions in H-RAN beyond traditional RAN aggregation. C-SMTL develops inter-component cooperation strategies between network components including data sharing, ML model sharing, computation sharing, decision-making sharing, etc. C-SMTL is driven by such comprehensive cooperation and fully utilizes open inter-faces among network components, servers, and infrastructure in contemporary edge-to-cloud architectures. For simplicity, among several promising advantages of C-SMTL cooperative strategies, we focus on a novel paradigm of collaborative-based sensor and radio units (SRUs) switching to provide alternative path connectivity. These cooperative approaches enable SRUs to work together synergistically, providing real-time insights into the network environment and identifying overlapping SRUs. These overlapping SRUs can serve as potential alternative communication paths through C-SMTL based on SRU switching in blockage scenarios. Our simulations validate that the proposed approach not only significantly enhances network performance but also achieves high spectral efficiency of remarkable 93 %, effectively reduces blockage probability, and maintains reliability under challenging conditions.