SiteEvo: An LLM Agent for Autonomous Discovery of Site-Adaptive Massive MIMO Detectors
Abstract
Designing a massive multiple-input multiple-output (MIMO) detector that is tailored to each specific deployment site can improve detection performance, yet most existing detectors, including deep-unfolding methods, adapt mainly within human-prescribed algorithmic structures. This letter proposes SiteEvo, a site-adaptive agentic discovery framework that extends adaptation from parameter tuning to detector-algorithm discovery. Rather than selecting or recombining predefined architectures, the large language model (LLM) agent synthesizes complete executable detector programs under a common design contract by combining its pretrained communication knowledge, site-specific wireless evidence, and experience accumulated during the search. Each proposal defines its own processing structure and trainable parameterization, and the parameters are optimized by a communication-system backend. The resulting high-performance detector is further structurally refined to reduce block-level computational cost under a bit error rate (BER)-preservation constraint and is finally frozen for deployment without online LLM inference. Simulations across three sites demonstrate BER reductions by factors of 2.6 to 6.0 over the best-performing competing detector at each site at 6 dB, while structural refinement reduces block-level computational cost by 16.7-34.1% under a validation-BER constraint.
Related
- The Blind Spot Paradox: When Adaptive Classifiers Defeat Drift Detectors
- GRAFT-ATHENA: Self-Improving Agentic Teams for Autonomous Discovery and Evolutionary Numerical Algorithms
- Autonomous Structural Discovery Of Bioprocess Dynamics And Kinetics Via LLM-Driven Agents
- Fronthaul-Efficient Cell-Free Massive MIMO via Stream-Adaptive Resolution Control
- Large-Scale Autonomous Discovery of Kissing Number Constructions
BibTeX
@misc{guo2026siteevo,
title = {{SiteEvo: An LLM Agent for Autonomous Discovery of Site-Adaptive Massive MIMO Detectors}},
author = {Jiajia Guo and Chao-Kai Wen and Shi Jin and Jun Zhang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.15490},
eprint = {2610.15490},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.15490},
}
RIS
TY - UNPB
TI - SiteEvo: An LLM Agent for Autonomous Discovery of Site-Adaptive Massive MIMO Detectors
AU - Jiajia Guo
AU - Chao-Kai Wen
AU - Shi Jin
AU - Jun Zhang
PY - 2026
DA - 2026/10/11
PB - arXiv
DO - 10.48550/arxiv.2610.15490
UR - https://arxiv.org/abs/2610.15490
AB - Designing a massive multiple-input multiple-output (MIMO) detector that is tailored to each specific deployment site can improve detection performance, yet most existing detectors, including deep-unfolding methods, adapt mainly within human-prescribed algorithmic structures. This letter proposes SiteEvo, a site-adaptive agentic discovery framework that extends adaptation from parameter tuning to detector-algorithm discovery. Rather than selecting or recombining predefined architectures, the large language model (LLM) agent synthesizes complete executable detector programs under a common design contract by combining its pretrained communication knowledge, site-specific wireless evidence, and experience accumulated during the search. Each proposal defines its own processing structure and trainable parameterization, and the parameters are optimized by a communication-system backend. The resulting high-performance detector is further structurally refined to reduce block-level computational cost under a bit error rate (BER)-preservation constraint and is finally frozen for deployment without online LLM inference. Simulations across three sites demonstrate BER reductions by factors of 2.6 to 6.0 over the best-performing competing detector at each site at 6 dB, while structural refinement reduces block-level computational cost by 16.7-34.1% under a validation-BER constraint.
ER -