WiFi CSI of indoor intrusion

This dataset comprises Channel State Information (CSI) data collected from WiFi signals in six indoor environments, specifically designed for research in indoor intrusion detection. The dataset captures fine-grained variations in wireless signals caused by human, which are indicative of potential intrusions. CSI data, extracted from commercial WiFi chipsets, provides detailed amplitude and phase information across subcarriers, enabling robust detection of subtle environmental changes. The dataset includes labeled samples for various intrusions, such as walking, running, and sneaking, as well as intrusion scenarios like unauthorized entry or loitering. It is collected in diverse indoor settings with varying layouts, furniture, and interference levels to ensure generalizability. This dataset is intended to support the development and evaluation of machine learning and signal processing algorithms for accurate, real-time indoor intrusion detection systems. It is particularly valuable for researchers working on WiFi-based sensing, human activity recognition, and security applications.本数据集包含从6种室内环境的WiFi信号中采集到的信道状态信息（Channel State Information, CSI）数据，专为室内入侵检测相关研究设计。该数据集捕捉到由人体活动引发的无线信号细粒度变化，这类变化可作为潜在入侵行为的指示特征。从商用WiFi芯片中提取的CSI数据能够提供各子载波上的详细幅度与相位信息，可实现对细微环境变化的可靠检测。本数据集包含各类入侵行为的标注样本，如行走、奔跑、潜行等，同时涵盖未经授权进入、徘徊等入侵场景。数据集采集自布局、家具摆放与干扰水平各不相同的多样化室内环境，以确保所开发算法具备良好泛化能力。本数据集旨在支撑面向精准实时室内入侵检测系统的机器学习与信号处理算法的开发与评估工作。对于从事WiFi感知、人体活动识别及安防应用研究的科研人员而言，该数据集具有极高的学术与应用价值。

搜集汇总

背景与挑战

背景概述

该数据集包含从六个室内环境收集的WiFi信道状态信息（CSI）数据，专门用于室内入侵检测研究。数据集中提供了标注的入侵行为样本，如行走、奔跑和潜行，并涵盖多种室内布局和干扰条件，以支持机器学习算法的开发和评估。