1st International Workshop on:
Embedded Files
Multi-Agent Intelligence and Connectivity for
Context-Aware Communication
Collocated with:
IEEE International Symposium on Personal, Indoor and Mobile Radio Communications 2026
1 - 4 September 2026, Singapore
Scope
Scope
The dawn of 6G promises a transformative leap from connected things to collaborative intelligence. The future network will be an AI-native fabric, populated by a massive number of heterogeneous agents, swarms of drones, collaborative robots, autonomous systems, and pervasive IoT sensors, that must intelligently perceive, decide, and act upon a shared, dynamic context. This evolution demands a fundamental rethinking of the relationship between Multi-Agent System (MAS) intelligence and the communication network. The network must evolve from a passive data pipe into an active, cognitive enabler of collaboration, while the distributed intelligence of agents must be harnessed to optimize the network itself. This workshop aims to bridge the current disconnect between the MAS community, which often assumes ideal connectivity, and the communications community, which often treats agents as simple data sources. We seek to establish a foundational forum for the co-design of intelligence and connectivity, addressing the core tension between the theoretical performance of collaborative agents and the practical constraints of real-world, context-aware wireless networks. We invite original research contributions, as well as visionary position papers, on the following (and related) topics.
Important Dates
Important Dates
Paper Submission Deadline:
Paper Submission Deadline:
5 June 2026 15 June 2026
Paper acceptance notification:
Paper acceptance notification:
8 July 2026
Camera-ready:
Camera-ready:
1 August 2026
Topics of interest including, but are not limited to, the following:
Topics of interest including, but are not limited to, the following:
MAS for Communication (Intelligence as a Network Optimizer):
MAS for Communication (Intelligence as a Network Optimizer):
Decentralized Resource Orchestration: Utilizing distributed intelligence for autonomous spectrum sharing, interference mitigation, and dynamic slicing.
AI-Native Physical Layer Optimization: Intelligent beamforming, power control, and waveform design driven by collaborative agent frameworks.
Self-Organizing Network Management: MAS-based path planning and mobility management for ultra-dense and non-terrestrial networks (NTN).
LLM-based MAS: LLMs-driven agents for user intention translation, environment-aware channel estimation and decision making
Communication for MAS (Network as a Collaborative Enabler):
Communication for MAS (Network as a Collaborative Enabler):
Coordination-Efficient Protocols: Designing network layers that minimize the overhead required for large-scale agent synchronization.
Topology-Aware Networking: Dynamic optimization of communication graphs to maintain swarm stability under mobility and fading.
Emergent communication for MAS: Designing communication protocols that provides efficient information sharing among agents
Semantic and Compressed Communication: Communication primitives that prioritize mission-critical intent over raw data to reduce communication load.
Coordination among Heterogeneous Agents: Scheduling agents according to different task goals, and designing collaborative schemes for them
Cross-Layer Foundations and Technical Enablers:
Cross-Layer Foundations and Technical Enablers:
Multi-Agent Reinforcement Learning (MARL): Implementing MARL for decentralized decision-making and real-time network configuration.
Robustness and Security: MAS control logic resilient to imperfect CSI, latency, and adversarial threats.
Experimental Validation: Prototypes, SDR-based testbeds, and digital twins for benchmarking AI-native MAS implementations.
Submission Notificiation
Submission Notificiation
The page length limit for all initial submissions for review is SIX (6) printed pages (10-point font) and must be written in English. Initial submissions longer than SIX (6) pages will be rejected without review. Papers should be submitted via the submission link.
Page updated
Report abuse