Passive devices have been deployed at an unprecedented scale across logistics, retail, industrial automation, and smart environments. However, because ambient energy is both extremely limited and intermittent, and the hardware is highly simplified, these devices cannot support conventional complex communication protocols and therefore lack unified mechanisms for network-level connectivity and management. To address these fundamental challenges, the IEEE 802.11 has established Task Group bp (TGbp) to standardize ambient- power (AMP) radio as part of the IEEE 802.11 family. This article provides a comprehensive overview of IEEE 802.11bp, covering its deployment scenarios, medium access control layer operations, frame structure, physical layer design, and wireless power transfer mechanisms. This article aims to bridge the gap between standardization and academic research, and to support the development of scalable, energy-efficient, and sustainable next-generation internet of things.
To empower the low-altitude economy, wireless networks must undergo a paradigm shift from connectivity-oriented infrastructures to safety-critical perception systems. Integrated sensing and communication (ISAC) offers a promising foundation, yet most existing studies focus on terrestrial or single-node scenarios and overlook the unique characteristics of low-altitude wireless networks (LAWNs), primarily three-dimensional (3D) geometric blindness, fragmented coverage, high mobility, and diverse non-cooperative threats. To overcome these limitations, this article explores cooperative ISAC for LAWNs and proposes a safety- driven layered hybrid architecture. We introduce a hierarchical centralized-distributed primary-secondary topology that employs adaptive mechanisms to integrate local feature-level fusion and global decision-level fusion for scalable wide-area tracking. Furthermore, we present key enabling technologies, notably the multiple cellular neighbor- assisted coverage paradigm, for achieving continuous perception of low-altitude targets. A comprehensive field case study in a dense urban environment validates that the proposed architecture and schemes effectively reconstruct fragmented trajectories and enable active target recognition, establishing a blueprint for trustworthy LAWNs. Finally, we outline open research issues and future directions toward intelligent, cooperative ISAC-enabled low-altitude networks.
Integrated sensing and communication (ISAC) is a cornerstone of 6G, poised to transform the network into a multi-modal sensor system, providing massive data for wireless and physical environmental sensing. However, existing multi-modal fusion approaches often struggle with fragmented optimization, lacking the strict temporal semantics and unified architectural framework necessary to translate raw data into actionable knowledge. To bridge this gap, we propose a novel multi-layer ISAC environmental knowledge graph (ISAC-KG) that unifies service, data, and knowledge into a dynamic relational network. We detail the semantic modeling and construction methodology of the ISAC-KG, emphasizing how its topological reasoning and temporal attributes overcome inconsistent modeling and asynchronous multi-dimensional data. Subsequently, we introduce utilization mechanisms that leverage this graph-based semantic structure for a holistic system optimization loop, supporting context-aware and time-sensitive decision-making. Finally, we present a multi-modal ISAC testbed as a comprehensive proof of concept, demonstrating the framework's scalability and topological resilience in orchestrating concurrent ISAC applications across diverse environments.
Network Digital Twins (NDTs) provide a realtime digital representation of an Original Network (ON), enabling monitoring, automation, optimization, flexibility, and scalability in Beyond 5G (B5G) and 6G networks. NDTs are typically established through a one-to-one (1:1) model, in which a unique NDT replica is derived from an ON. The main limitation of current NDT systems is the inability to support concurrent and isolated real-time operation among multiple tenants. To address this, an extended NDT model, designated as one-to-many (1:N), is presented, with multiple NDTs derived from a single ON based on an NDT Federation Framework. Federation mechanisms are supported by a centralized Digital Twin Manager (DTM), enabling coordination, synchronization, isolation, and consistency across multiple NDT instances. The 1:N NDT architecture is technology-agnostic; this article examines the role of extended reality (XR) in customizing each deployed NDT instance to facilitate immersive experiences, allowing users to interact intuitively with digital environments. A Proof-of-Concept (PoC) implementation is introduced to validate and evaluate the coordination scalability of a federated NDT system by analyzing event drop ratio and DTM throughput under controlled conditions. The results indicate that, with the proposed architecture, the offered load increases linearly with the number of NDT instances. They also highlight the importance of determining the maximum number of supported allowed federated NDTs based on processing capacity constraints to ensure proper system operation. This supports the feasibility of federated NDT operation for collaborative network management scenarios in future networks.
Automated inspection of critical infrastructure in extreme environments faces significant challenges, including limited connectivity and the need for real-time decision-making, which make conventional raw-data collection and centralized processing increasingly inadequate. This paper proposes a shift from isolated data-collection UAVs to collaborative swarms driven by Agentic AI, where each UAV operates as an autonomous agent that closes the perception, cognition, and decision-making loop onboard. Instead of transmitting raw data, agents prioritize task-relevant features, performing adaptive, integrated perception, communication, and computation across nodes. We analyze the challenges encountered by Agentic AI-driven swarms and present solutions such as intent-driven distributed perception-communication-computation-control optimization, semantic-driven feature transmission, and robust spatiotemporal fusion. Two application cases validate the framework at the agent and swarm levels: 1) agent-level collaborative perception mitigates occlusions and sparse observations, achieving detection accuracy comparable to early fusion while reducing transmission cost by 96.8%; and 2) swarm-level integration of perception, communication, and computation enables online sensing, transmission, and processing, improving final inspection coverage by 6.7% over the best baseline.
