Secure and Energy-Efficient Wireless Agentic AI Networks

2026-02-16

AI Summary

This work addresses the challenge of jointly optimizing energy efficiency, latency, accuracy, and privacy in wireless multi-agent AI networks. To this end, we propose a novel framework that integrates friendly jamming with collaborative inference: a supervisory agent dynamically schedules a subset of cooperative agents to perform inference tasks, while the unselected agents act as friendly jammers to enhance communication security. The framework co-optimizes agent selection, base station beamforming, and transmit power to minimize total energy consumption. A key innovation is the incorporation of a large language model–based workflow optimizer for efficient resource allocation, with the resulting non-convex problem solved via a combination of ADMM, semidefinite relaxation (SDR), and successive convex approximation (SCA). Experiments demonstrate up to a 59.1% reduction in system energy consumption compared to baseline methods, while maintaining high inference accuracy across multiple public benchmarks on a real-world agent platform powered by Qwen.

Technology Category

• Multiagent Systems: Agent Communication
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Agentic search
• Security and Privacy: Security and privacy of machine learning and AI applications

Abstract

In this paper, we introduce a secure wireless agentic AI network comprising one supervisor AI agent and multiple other AI agents to provision quality of service (QoS) for users' reasoning tasks while ensuring confidentiality of private knowledge and reasoning outcomes. Specifically, the supervisor AI agent can dynamically assign other AI agents to participate in cooperative reasoning, while the unselected AI agents act as friendly jammers to degrade the eavesdropper's interception performance. To extend the service duration of AI agents, an energy minimization problem is formulated that jointly optimizes AI agent selection, base station (BS) beamforming, and AI agent transmission power, subject to latency and reasoning accuracy constraints. To address the formulated problem, we propose two resource allocation schemes, ASC and LAW, which first decompose it into three sub-problems. Specifically, ASC optimizes each sub-problem iteratively using the proposed alternating direction method of multipliers (ADMM)-based algorithm, semi-definite relaxation (SDR), and successive convex approximation (SCA), while LAW tackles each sub-problem using the proposed large language model (LLM) optimizer within an agentic workflow. The experimental results show that the proposed solutions can reduce network energy consumption by up to 59.1% compared to other benchmark schemes. Furthermore, the proposed schemes are validated using a practical agentic AI system based on Qwen, demonstrating satisfactory reasoning accuracy across various public benchmarks.

Problem

• secure wireless AI networks
• energy efficiency
• confidentiality
• quality of service
• cooperative reasoning

Innovation

• Wireless Agentic AI
• Secure Cooperative Reasoning
• Energy-Efficient Resource Allocation
• Friendly Jamming
• LLM-based Optimization