Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Secure and Energy-Efficient Wireless Agentic AI Networks

Yuanyan Song, Kezhi Wang, Xinmian Xu

TL;DR

This work tackles energy-efficient QoS provisioning for cooperative reasoning in secure wireless agentic AI networks. It formulates a joint energy minimization problem over AI agent selection, BS beamforming, and AI agent transmission powers under latency and accuracy constraints, and then presents two solution frameworks: ASC (ADMM-SDR-SCA) and LAW (LLM-optimizer-based). ASC decomposes the problem into three subproblems and solves them iteratively, while LAW uses a supervisor-driven agentic workflow with an LLM as the optimizer. Experimental results show up to a $59.1\%$ reduction in network energy consumption and satisfactory reasoning accuracy on public benchmarks, validating both approaches and their practical viability in Qwen-based deployments.

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.

Secure and Energy-Efficient Wireless Agentic AI Networks

TL;DR

This work tackles energy-efficient QoS provisioning for cooperative reasoning in secure wireless agentic AI networks. It formulates a joint energy minimization problem over AI agent selection, BS beamforming, and AI agent transmission powers under latency and accuracy constraints, and then presents two solution frameworks: ASC (ADMM-SDR-SCA) and LAW (LLM-optimizer-based). ASC decomposes the problem into three subproblems and solves them iteratively, while LAW uses a supervisor-driven agentic workflow with an LLM as the optimizer. Experimental results show up to a reduction in network energy consumption and satisfactory reasoning accuracy on public benchmarks, validating both approaches and their practical viability in Qwen-based deployments.

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.
Paper Structure (22 sections, 57 equations, 5 figures, 5 tables, 2 algorithms)

This paper contains 22 sections, 57 equations, 5 figures, 5 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Works
  3. MD-deployed LLMs
  4. Agentic AI-empowered reasoning
  5. Open challenges of wireless agentic AI networks
  6. System Model and Problem Formulation
  7. Secure communication model
  8. Reasoning model
  9. The overall process
  10. Problem formulation
  11. The proposed solutions
  12. ASC
  13. AI agent Selection
  14. Beamforming design
  15. Transmission power optimization
  16. ...and 7 more sections

Figures (5)

  • Figure 1: A typical workflow of agentic AI to execute a specific reasoning task.
  • Figure 2: The framework of a secure wireless agentic AI network.
  • Figure 3: The framework of LAW.
  • Figure 4: The execution energy consumption versus different numbers of AI agents.
  • Figure 5: The transmission energy consumption versus different numbers of AI agents.