MAC协议在ISM频段中使用机器学习方法进行分类 - 论文摘要与研究

Aug 5, 2026

Radio spectrum shortages arise as wireless networks multiply, making spectrum awareness and protection critical. Detecting and classifying MAC-layer protocols helps Cognitive Radio users exploit idle spectrum while limiting interference. This study classifies common ISM-band Wi‑Fi and Bluetooth MAC protocols using machine learning and deep learning. Support Vector Machine and K‑Nearest Neighbors are trained with features derived from USRP N210 SDR time-frequency measurements, including frame width, silence gap, and PAPR, then evaluated under varying transmitter/receiver conditions and added Gaussian noise.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages18

Reading Time45 min

AccessFree to download

Abstract

Introduction

Spectrum utilization and cognitive radio motivation

Importance of MAC protocol identification in crowded ISM bands

Related work on ML/DL for signal and protocol classification

MAC protocol classification in the ISM band using machine learning methods Hanieh Rashidpour, Hossein Bahramgiri [hanierp@mut.ac.ir](hanierp@mut.ac.ir) [Bahramgiri@mut.ac.ir](Bahramgiri@mut.ac.ir) Abstract With the emergence of new technologies and a growing number of wireless networks, we face the problem of radio spectrum shortages. As a result, identifying the wireless channel spectrum with the goal of exploiting the channel's idle state while also boosting network security is pivotal issue. Detecting and classifying protocols in the MAC sublayer enables Cognitive Radio (CR) users to improve spectrum utilization and minimize potential interference. In this paper, we classify the Wi-Fi and Bluetooth protocols, which are the most widely used MAC sublayer protocols in the ISM radio band. With the advent of various wireless technologies, especially in the 2.4 GHz frequency band, the ISM frequency spectrum has become a crowded and high-traffic band, which faces a lack of spectrum resources and user interference. Therefore, identifying and classifying protocols is an effective and useful method. Leveraging machine learning (ML) and deep learning (DL) techniques, known for their advanced classification capabilities, we apply Support Vector Machine (SVM) and K-Nearest Neighbors (KNN) algorithms, which are machine learning algorithms, to classify protocols into three classes: Wi-Fi, Wi-Fi Beacon, and Bluetooth. To capture the signals, we use the USRP N210 Software Defined Radio (SDR) device and sample the real data in the indoor environment in different conditions of the presence and absence of transmitters and receivers for these two protocols. By assembling this dataset and studying the time and frequency features of the protocols, we extract the frame width and the silence gap between the two frames as time features and the PAPR of each frame as a power feature. By comparing the output of the protocols classification in different conditions and also adding Gaussian noise, it was found that the samples in the nonlinear SVM method with RBF and KNN functions have the best performance, with 97.83% and 98.12% classification accuracy, respectively. Keywords: MAC Sublayer Protocols, Cognitive Radio, ISM Band, USRP N210, Machine Learning, SVMand KNN Methods. I. Introduction With the rapid expansion of new wireless communication systems, including wireless personal area networks (WPANs), wireless local area networks (WLANs), and wireless metropolitan area networks (WMANs), the demand for radio spectrum is increasing day by day, so spectrum awareness and allocation become a more important subject. However, measurements in this area indicate that the wide range of spectrum allocated to users (primary networks) is underutilized. As a result, wireless channel spectrum identification with the goal of exploiting the channel's idle state and its optimal management, as well as network security, is a critical issue. Users can adjust their transmission parameters in Cognitive Radio (CR) systems by sensing the current state of the external radio environment [1], thus improving spectrum utilization and effectively reducing the problem of spectrum resource scarcity. Spectrum utilization can be improved if the network is aware of channel parameters such as empty capacity, interference, signal modulation, media access control protocols (MAC), power levels, transmission schemes, and so on [2] . In particular, CR users can identify present protocols in any transmission by identifying and classifying MAC sublayer characteristics. This issue is more critical in heterogeneous spectrums such as the ISM (Industrial, Scientific, and Medical) bands. The 2.4 GHz ISM band is unlicensed and free, and several important technologies like Wi-Fi, ZigBee, and Bluetooth share it. As a result, it has become a crowded and high-traffic band, facing user and network interferences. These technologies compete for resources and strive to coexist. On the other

FAQs

为什么需要对ISM频段的MAC协议进行检测与分类？

ISM频段网络密集且存在干扰风险，且频谱资源常被低效利用。识别空闲状态并提高安全性依赖于对MAC子层协议的判别与分类。

本文使用哪些机器学习方法对协议进行分类？

使用支持向量机（SVM）与K近邻（KNN）两类机器学习算法，并提到SVM的RBF等核函数与KNN函数的对比评估。

训练与评估的数据特征是如何提取的？

基于USRP N210软件无线电采样的真实室内信号，从时频特征提取帧宽与帧间静默间隔，并将每帧的PAPR作为功率特征；在不同发射机/接收机条件以及加入高斯噪声下比较分类表现。

Aug 5, 2026

Upload by Genevieve