When marine radar target detection meets pretrained large language models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hu, Qiying, Zhang, Linping, Wang, Xueqian, Li, Gang, Liu, Yu, Zhang, Xiao-Ping
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915495582629888
author Hu, Qiying
Zhang, Linping
Wang, Xueqian
Li, Gang
Liu, Yu
Zhang, Xiao-Ping
author_facet Hu, Qiying
Zhang, Linping
Wang, Xueqian
Li, Gang
Liu, Yu
Zhang, Xiao-Ping
contents Deep learning (DL) methods are widely used to extract high-dimensional patterns from the sequence features of radar echo signals. However, conventional DL algorithms face challenges such as redundant feature segments, and constraints from restricted model sizes. To address these issues, we propose a framework that integrates feature preprocessing with large language models (LLMs). Our preprocessing module tokenizes radar sequence features, applies a patch selection algorithm to filter out uninformative segments, and projects the selected patches into embeddings compatible with the feature space of pre-trained LLMs. Leveraging these refined embeddings, we incorporate a pre-trained LLM, fine-tuning only the normalization layers to reduce training burdens while enhancing performance. Experiments on measured datasets demonstrate that the proposed method significantly outperforms the state-of-the-art baselines on supervised learning tests.
format Preprint
id arxiv_https___arxiv_org_abs_2509_12110
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle When marine radar target detection meets pretrained large language models
Hu, Qiying
Zhang, Linping
Wang, Xueqian
Li, Gang
Liu, Yu
Zhang, Xiao-Ping
Signal Processing
Computation and Language
Machine Learning
Deep learning (DL) methods are widely used to extract high-dimensional patterns from the sequence features of radar echo signals. However, conventional DL algorithms face challenges such as redundant feature segments, and constraints from restricted model sizes. To address these issues, we propose a framework that integrates feature preprocessing with large language models (LLMs). Our preprocessing module tokenizes radar sequence features, applies a patch selection algorithm to filter out uninformative segments, and projects the selected patches into embeddings compatible with the feature space of pre-trained LLMs. Leveraging these refined embeddings, we incorporate a pre-trained LLM, fine-tuning only the normalization layers to reduce training burdens while enhancing performance. Experiments on measured datasets demonstrate that the proposed method significantly outperforms the state-of-the-art baselines on supervised learning tests.
title When marine radar target detection meets pretrained large language models
topic Signal Processing
Computation and Language
Machine Learning
url https://arxiv.org/abs/2509.12110