Machine Learning Methods for Inferring the Number of UAV Emitters via Massive MIMO Receive Array

Aug 4, 2026

Upload by Logic

The work develops a complete DOA preprocessing system that infers the number of UAV emitters using a massive MIMO receive array, aiming to supply prior knowledge for future wireless networks. Noise signals are suppressed with two high-precision detectors, SR-MME and GM, which jointly provide high detection probability and extremely low false alarms. Once emitter existence is detected, a feature extraction strategy over eigenvalue sequences and a multi-layer neural network are used, alongside SVM and naïve Bayes. Results show neural methods maintain classification accuracy above 70% with massive MIMO.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Abstract

Introduction

UAV communication challenges

Need for DOA prior knowledge

Massive MIMO for array processing

Machine Learning Methods for Inferring the Number of UAV Emitters via Massive MIMO Receive Array Yifan Li, Feng Shu, Member, IEEE, Jinsong Hu, Shihao Yan, Haiwei Song, Weiqiang Zhu, Da Tian, Yaoliang Song, Senior Member, IEEE, and Jiangzhou Wang, Fellow, IEEE arXiv :2203 .00917v3 [ ee ss . SP] 10 Mar 2023 Abstract—To provide important prior knowledge for the DOA estimation of UAV emitters in future wireless networks, we present a complete DOA preprocessing system for inferring the number of emitters via massive MIMO receive array. Firstly, in order to eliminate the noise signals, two high-precision signal detectors, square root of maximum eigenvalue times minimum eigenvalue (SR-MME) and geometric mean (GM), are proposed. Compared to other detectors, SR-MME and GM can achieve a high detection probability while maintaining extremely low false alarm probability. Secondly, if the existence of emitters is determined by detectors, we need to further conﬁrm their number. Therefore, we perform feature extraction on the the eigenvalue sequence of sample covariance matrix to construct feature vector and innovatively propose a multi-layer neural network (ML-NN). Additionally, the support vector machine (SVM), and naive Bayesian classiﬁer (NBC) are also designed. The simulation results show that the machine learning-based methods can achieve good results in signal classiﬁcation, especially neural networks, which can always maintain the classiﬁcation accuracy above 70% with massive MIMO receive array. Finally, we analyze the classical signal classiﬁcation methods, Akaike (AIC) and Minimum description length (MDL). It is concluded that the two methods are not suitable for scenarios with massive MIMO arrays, and they also have much worse performance than machine learning-based classiﬁers. Index Terms—unmanned aerial vehicle (UAV), massive MIMO, threshold detection, emitter number detection, machine learning, information criterion. I. INTRODUCTION With the advantages of high mobility and low cost, unmanned aerial vehicles (UAVs) are always supposed to play important roles in wireless networks for implementing the tasks like weather monitoring, trafﬁc control, emergency search, communication relaying, etc. [1] . However, different from the traditional ground-to-ground (G2G) communications, Y. Li, and Y. Song are with the School of Electronic and Optical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China (e-mail: [liyifan97@foxmail.com](liyifan97@foxmail.com)). F. Shu is with the School of Information and Communication Engineering, Hainan University, Haikou 570228, China (e-mail: [shufeng0101@163.com](shufeng0101@163.com)). Jinsong Hu is with the College of Physics and Information Engineering, Fuzhou University, Fuzhou, Fujian, 350116, China (e-mail: jin[song.hu@fzu.edu.cn](song.hu@fzu.edu.cn)). S. Yan is with the School of Science and Security Research Institute, Edith Cowan University, Perth, WA 6027, Australia (e-mail: [s.yan@ecu.edu.au](s.yan@ecu.edu.au)). H. Song, W. Zhu and D. Tian are with the 8511 Research Institute, China Aerospace Science and Industry Corporation, Nanjing 210007, China (e-mail: [hw8511@126.com](hw8511@126.com)) . J. Wang is with the School of Engineering, University of Kent, Canterbury CT2 7NT, U.K (e-mail: [j.z.wang@kent.ac.uk](j.z.wang@kent.ac.uk)). UAV communications have some special characteristics and challenges, e.g., the high mobility will lead to the UAV communication channels change much faster, the high ﬂight altitude requiring the ground base stations to provide larger 3D signal coverage for UAVs, the line of sight (LoS) paths between UAVs and base stations are vulnerable to interference from ground users over the same frequency [2] . Obviously, 4G wireless networks are difﬁcult to meet the requirements for UAV communications. But as is known to us, massive multiple-input multiple-output (MIMO) is a key technology in 5G or future 6G systems [3], [4]

FAQs

What problem does the document address?

It addresses how to infer the number of UAV emitters for DOA estimation using a massive MIMO receive array when the emitter count is unknown.

How does the system detect whether emitters exist?

It proposes two high-precision detectors, SR-MME and GM, designed to eliminate noise signals while achieving high detection probability and extremely low false alarm probability.

How is the number of emitters determined after detection?

After existence is confirmed, the method extracts features from the eigenvalue sequence of the sample covariance matrix and uses a multi-layer neural network, with SVM and naïve Bayes also implemented for comparison.

Aug 4, 2026

Upload by Logic