Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Large Language Models as Bidding Agents in Repeated HetNet Auction

Ismail Lotfi, Ali Ghrayeb, Samson Lasaulce, Merouane Debbah

TL;DR

Simulation results reveal that the LLM-empowered UE consistently achieves higher channel access frequency and improved budget efficiency compared to benchmarks, and pave the way for lightweight, edge-deployable LLMs to support intelligent resource allocation in next-generation HetNets.

Abstract

This paper investigates the integration of large language models (LLMs) as reasoning agents in repeated spectrum auctions within heterogeneous networks (HetNets). While auction-based mechanisms have been widely employed for efficient resource allocation, most prior works assume one-shot auctions, static bidder behavior, and idealized conditions. In contrast to traditional formulations where base station (BS) association and power allocation are centrally optimized, we propose a distributed auction-based framework in which each BS independently conducts its own multi-channel auction, and user equipments (UEs) strategically decide both their association and bid values. Within this setting, UEs operate under budget constraints and repeated interactions, transforming resource allocation into a long-term economic decision rather than a one-shot optimization problem. The proposed framework enables the evaluation of diverse bidding behaviors -from classical myopic and greedy policies to LLM-based agents capable of reasoning over historical outcomes, anticipating competition, and adapting their bidding strategy across episodes. Simulation results reveal that the LLM-empowered UE consistently achieves higher channel access frequency and improved budget efficiency compared to benchmarks. These findings highlight the potential of reasoning-enabled agents in future decentralized wireless networks markets and pave the way for lightweight, edge-deployable LLMs to support intelligent resource allocation in next-generation HetNets.

Large Language Models as Bidding Agents in Repeated HetNet Auction

TL;DR

Simulation results reveal that the LLM-empowered UE consistently achieves higher channel access frequency and improved budget efficiency compared to benchmarks, and pave the way for lightweight, edge-deployable LLMs to support intelligent resource allocation in next-generation HetNets.

Abstract

This paper investigates the integration of large language models (LLMs) as reasoning agents in repeated spectrum auctions within heterogeneous networks (HetNets). While auction-based mechanisms have been widely employed for efficient resource allocation, most prior works assume one-shot auctions, static bidder behavior, and idealized conditions. In contrast to traditional formulations where base station (BS) association and power allocation are centrally optimized, we propose a distributed auction-based framework in which each BS independently conducts its own multi-channel auction, and user equipments (UEs) strategically decide both their association and bid values. Within this setting, UEs operate under budget constraints and repeated interactions, transforming resource allocation into a long-term economic decision rather than a one-shot optimization problem. The proposed framework enables the evaluation of diverse bidding behaviors -from classical myopic and greedy policies to LLM-based agents capable of reasoning over historical outcomes, anticipating competition, and adapting their bidding strategy across episodes. Simulation results reveal that the LLM-empowered UE consistently achieves higher channel access frequency and improved budget efficiency compared to benchmarks. These findings highlight the potential of reasoning-enabled agents in future decentralized wireless networks markets and pave the way for lightweight, edge-deployable LLMs to support intelligent resource allocation in next-generation HetNets.
Paper Structure (22 sections, 14 equations, 5 figures)

This paper contains 22 sections, 14 equations, 5 figures.

Table of Contents

  1. Introduction
  2. Background
  3. Related Works
  4. Contributions
  5. System Model
  6. System Description
  7. Evaluation Metrics: Data Rate and QoS
  8. Utility Functions
  9. BS Utility
  10. UE Utility
  11. Problem Formulation: Auction-based Distributed Channel Allocation
  12. BS Selection Procedure (UE-Side Decision)
  13. Myopic Truthful Strategy
  14. Greedy Bayesian Strategy
  15. Winning probability and expected payment
  16. ...and 7 more sections

Figures (5)

  • Figure 1: An example of a two-tier HetNet including one macrocell BS and four small BSs.
  • Figure 2: Template of the LLM prompt and expected response structure used for strategic BS selection and bidding.
  • Figure 3: Average utility for different bidding strategies.
  • Figure 4: (a) average number of channel access and (b) bid precision for different bidding strategies.
  • Figure 5: (a) utility variability and (b) average number of channel access for the LLM UE and the other greedy UEs in greedy majority population.