WO2017117910A1 - 基于无线网络信号传输的室内探测火情及报警的方法及系统 - Google Patents

基于无线网络信号传输的室内探测火情及报警的方法及系统 Download PDF

Info

Publication number
WO2017117910A1
WO2017117910A1 PCT/CN2016/084192 CN2016084192W WO2017117910A1 WO 2017117910 A1 WO2017117910 A1 WO 2017117910A1 CN 2016084192 W CN2016084192 W CN 2016084192W WO 2017117910 A1 WO2017117910 A1 WO 2017117910A1
Authority
WO
WIPO (PCT)
Prior art keywords
fire
channel state
wireless
signal
state information
Prior art date
Application number
PCT/CN2016/084192
Other languages
English (en)
French (fr)
Inventor
伍楷舜
钟舒馨
王璐
Original Assignee
深圳大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳大学 filed Critical 深圳大学
Publication of WO2017117910A1 publication Critical patent/WO2017117910A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G08SIGNALLING
    • G08BSIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B17/00Fire alarms; Alarms responsive to explosion
    • G08B17/10Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means
    • GPHYSICS
    • G08SIGNALLING
    • G08BSIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B17/00Fire alarms; Alarms responsive to explosion
    • GPHYSICS
    • G08SIGNALLING
    • G08BSIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B25/00Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems
    • G08B25/01Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium
    • G08B25/10Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium using wireless transmission systems

Definitions

  • the invention relates to wireless information processing technology, in particular to an indoor fire detection and alarm system based on wireless signals
  • indoor fire detection mainly depends on the installation of various types of detectors, such as suction detectors, infrared flame detectors, smoke detectors, etc., all of which must be installed in a fire room beforehand, but in fact, very few Ordinary households will install the above detectors. Therefore, it will cause obstacles to the detection of indoor fire.
  • detectors such as suction detectors, infrared flame detectors, smoke detectors, etc.
  • the present invention utilizes wireless communication technology to monitor indoor fires and alert.
  • every household has basically installed a wireless LAN, which is WIFI, and can detect the fire with existing equipment without additional cost.
  • the present invention provides a low-cost, high-precision method and system for indoor fire detection and alarm based on wireless signals.
  • a specific indoor environment by using existing wireless networks and equipment, effective detection of indoor fires is achieved, and timely alarm and feedback functions are achieved.
  • An indoor detection fire and alarm method based on wireless network signal transmission the steps of which include:
  • the wireless receiving end receives the wireless signal from the wireless transmitting end, evaluates CSI (channel state information), and uses data filtering technology (Butterworth filter) to denoise the data;
  • the number of the wireless transmitting ends is one, and the data of the wireless receiving end is one.
  • the step S1 evaluating channel state information includes:
  • the step S2 uses the anomaly detection algorithm to identify that the abnormality of the channel state information change is a time series anomaly detection algorithm based on the local anomaly factor.
  • step S2 comprises:
  • the subsequence is output as an abnormal mode
  • the step S3 includes:
  • step S32 The abnormal mode calculation feature value outputted in step S22 is compared to whether the feature value of the pre-stored signal is matched;
  • the invention also provides an indoor detection fire and alarm system based on wireless network signal transmission, including:
  • a CSI acquisition module for receiving, by the wireless receiving end, a wireless signal from the wireless transmitting end, and evaluating CSI (channel state information), and using data filtering technology (Butterworth filter) to perform white noise removal processing on the data;
  • an abnormality detecting module for identifying an abnormality of channel state information change by using an abnormality detecting algorithm (sequence abnormality);
  • the fire condition judgment module uses a random forest algorithm to select five methods for extracting features (for example, standard deviation, interquartile range, etc.) to store the signal information of the fire in the room, if the abnormal mode signal characteristics and implementation If the stored signal characteristics match, it is judged that there is a fire in the room, and then a fire alarm signal is issued;
  • features for example, standard deviation, interquartile range, etc.
  • the alarm module is used to generate a fire alarm signal when a fire occurs indoors.
  • the CSI acquisition module includes:
  • a sensing unit used for initial channel state data, to obtain a matrix of the number of links ⁇ the number of subcarriers
  • a data processing unit configured to obtain, for each spatial stream, CSI values of 30 consecutive subcarriers at the same time point as channel state information
  • the filtering unit uses the Butterworth algorithm to remove the influence of white noise on the channel state information.
  • the abnormality detecting module is based on a time series abnormality detecting algorithm of a local anomaly factor, and includes:
  • an LOF calculation unit configured to perform data segmentation on the CSI data to obtain a subsequence, and obtain a local anomaly factor of each subsequence
  • An abnormal output unit configured to output the sub-sequence as an abnormal mode when the local abnormality factor is greater than or equal to a preset threshold.
  • the fire judgment module includes:
  • the fire identification unit calculates the feature value from the abnormal mode output by the abnormal output unit, and uses a random forest algorithm to see whether it matches the previously stored signal characteristics.
  • the invention has the beneficial effects that no additional detecting device or sensor is needed in the detected environment, and can be used in any environment, and only the existing WIFI router can be used for signal collection and processing, and we can use CSI channel state information and The indoor fire is linked.
  • the CSI data of the mobile phone is abnormal or not, it matches the target action mode obtained in advance, and if a fire occurs, it will promptly report the alarm. Therefore, it has great popularity.
  • 1 is a schematic diagram of hardware for transmitting and receiving data of an indoor fire detection and alarm system based on wireless network signal transmission;
  • FIG. 2 is a schematic diagram of evaluating CSI status information
  • FIG. 3 is a schematic diagram of an embodiment of an indoor fire detection and alarm method based on wireless network signal transmission.
  • An indoor detection fire and alarm method based on wireless network signal transmission the steps of which include:
  • the wireless receiving end receives the wireless signal from the wireless transmitting end, evaluates CSI (channel state information), and uses data filtering technology (Butterworth filter) to denoise the data;
  • the wireless receiving end is a wireless network card
  • the wireless transmitter is a wireless router.
  • the method is based on a radio propagation mechanism in an indoor environment, and establishes a relationship between a wireless signal and a fire, and only needs to use a home existing
  • the wireless network device that is, can analyze the change of the wireless signal caused by the change of the fire environment, and determine whether there is a fire in the room and an alarm is issued.
  • the number of the wireless transmitting ends is one, the number of the wireless receiving ends is one, and the wireless signal is transmitted and received by the multiple antennas in the system; the wireless network card used by the system can receive the channel state information. As shown in FIG.
  • CSI channel state information abbreviation, channel state information, in the field of wireless communication
  • CSI is communication
  • the present invention uses the channel state information CSI of the wireless network as an indicator.
  • CSI can describe the propagation path of a signal under the combined influence of time delay, amplitude attenuation and phase shift.
  • the present invention establishes a link between CSI and indoor fire conditions.
  • the propagation of CSI data is a major path and multiple scattering paths (such as ceiling, ground, room furnishings).
  • OFDM Orthogonal Frequency Division Multiplex
  • the evaluating the CSI status information includes:
  • the wireless receiving end receives the wireless signal from the wireless transmitting end, evaluates CSI (channel state information), and uses data filtering technology (Butterworth filter) to denoise the data;
  • the wireless transmitting end transmits a wireless signal
  • the computer equipped with the wireless network card as the wireless receiving end collects CSI data as initial channel state information.
  • the AP we used the AP with three antennas and one antenna at the computer end, thus forming three signal links, each link has 30 subcarriers, so we can get a 1 ⁇ 3 ⁇ 30 matrix.
  • the step S2 is a time series anomaly detection algorithm based on a local factor that uses the anomaly detection algorithm to identify the abnormality of the channel state information change.
  • Step S2 includes:
  • the subsequence is output as an abnormal mode
  • the local density of a point using LOF is compared to its neighbors. If the former is significantly lower than the latter (there is a LOF value greater than 1), the point is in a sparse area, and for its neighbors, this indicates that the point is an outlier and the exception pattern is obtained.
  • the channel effects will output corresponding abnormal modes due to various reasons (for example, human activities). These abnormal modes will calculate the eigenvalues separately and in the random forest algorithm and the previously stored signal characteristics. For comparison, if it matches, it is judged to be a fire, and an alarm is issued.
  • step S3 includes:
  • step S32 Calculate the feature value by using the abnormal mode outputted in step S22, and use a random forest algorithm to see whether it matches the previously stored signal feature;
  • the system is mainly divided into data processing, abnormality detection, and action classification.
  • the process of detecting fire and alarm in the room is realized.
  • the wireless transmitting end transmits a wireless signal
  • the wireless receiving end receives the wireless signal and collects CSI (Channel State Information) data
  • the invention is an indoor detection fire and alarm system based on wireless network signal transmission, and the module included therein is as shown in FIG. 5, comprising:
  • a CSI acquiring module configured to receive a wireless signal, and evaluate CSI channel state information
  • An abnormality detecting module configured to identify an abnormal change of channel state information
  • the fire condition judging module is configured to judge whether the feature of the abnormal mode is consistent with the information feature stored in advance (random forest), thereby determining whether there is a fire in the room;
  • the alarm module is used to send an alarm signal when a fire occurs indoors.
  • the CSI obtaining module includes:
  • a sensing unit configured to initial channel state data, to obtain a matrix of the number of links ⁇ the number of subcarriers
  • a data processing unit configured to obtain, for each spatial stream, CSI channel state information of 30 consecutive subcarriers at the same time point;
  • the filtering unit uses the Butterworth algorithm to remove the white noise from the channel state information.
  • the abnormality detecting module is based on a time series abnormality detecting algorithm of a local anomaly factor, and includes:
  • the LOF calculation unit is configured to perform data segmentation on the CSI data to obtain a subsequence, and obtain a local anomaly factor of each subsequence;
  • the abnormal output unit is configured to output the sub-sequence as an abnormal mode when the local abnormality factor is greater than or equal to a preset threshold.
  • the fire judgment module includes:
  • the fire identification unit calculates the feature value from the abnormal mode outputted by the abnormal output unit, and uses a random forest algorithm to see whether it matches the previously stored signal characteristics.
  • the alarm module is used to issue an alarm (for example, texting the homeowner, etc.) in the event of a fire.

Landscapes

  • Business, Economics & Management (AREA)
  • Emergency Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Alarm Systems (AREA)
  • Fire Alarms (AREA)

Abstract

本发明涉及无线信息技术处理领域,其公开了一种基于无线信号的室内火情探测及报警的系统,实现所述系统的步骤主要有:无线接收端接收来自无线发射端的无线电磁信号,评估CSI(channel state information信道状态信息),并采用数据过滤技术(巴特沃斯滤波器)对数据进行去噪处理;利用异常检测算法(序列异常)来识别信道状态信息变化的异常;采用随机森林算法,选择五种提取特征的方法(例如,标准差,四分位距等),将室内出现火情的信号信息实现储存,如果异常模式信号特征与实现存储的信号特征相匹配,则判断室内出现火情,继而发出失火警报信号。本发明的有益效果是:主要利用了现有的无线网络及设备,房间内无需额外地安装其他检测或者传感设备,不需要额外的铺设代价,能够实时监控火情,具有极高的普及性。

Description

基于无线网络信号传输的室内探测火情及报警的方法及系统 技术领域
本发明涉及无线信息处理技术,尤其涉及一种基于无线信号的室内火情探测及报警系统
背景技术
虽然烟花鞭炮已经被明文禁止了,但仍然有些市民会在自己的小区中玩鞭炮烟花,这时候就会造成安全隐患。如果家中人刚好都外出旅游过节了,此时冷不丁飞进来一个火苗,造成的后果那是不堪设想。每年这个举国欢庆的日子,这种悲剧都频频发生。再例如,震惊世界的天津爆炸事件,如果在室内刚有些小火苗时,就能被及时地发现并得到有效地控制,就不会有后来接二连三的爆炸以及人员伤亡了。因此,我们迫切地想寻找一种能方便,有效且不用购买额外设备,能自动探测室内火情并且在发现火情后能及时报警的方法。现在,室内探测火情主要依赖于安装形形色色的探测器,例如吸气探测器,红外火焰探测器,烟雾探测器等,上述探测器都必须事先安装在失火的房间内,但事实上,极少普通家庭会安装上述探测器。因此,会对室内火情的探测造成了障碍。
随着无线通信技术的发展,越来越多的无线设备被应用在人们的生活中。因此,本发明利用无线通信技术来监测室内火情并报警。如今,家家户户基本上都安装了无线局域网,也就是WIFI,利用现有的设备就能探测火情,而无需额外增加成本。
发明内容
为了克服上述所指设备在使用上普及度不够的不足,本发明提供一种低代价,高精度的基于无线信号的室内火情探测及报警的方法及系统。在特定的室内环境中,通过利用现有的无线网络及设备,实现对室内火情的有效检测,并达到及时报警和反馈的功能。
本发明是通过以下技术方案实现的:
一种基于无线网络信号传输的室内探测火情及报警方法,其步骤包括:
S1.无线接收端接收来自无线发射端的无线信号,评估CSI(channel state information信道状态信息),并采用数据过滤技术(巴特沃斯滤波器)对数据进行去噪处理;
S2.利用异常检测算法(序列异常)来识别信道状态信息变化的异常;
S3.采用随机森林算法,选择五种提取特征的方法(例如,标准差,四分位距等),将室内出现火情的信号信息实现储存,如果异常模式信号特征与实现存储的信号特征相匹配,则判断室内出现火情,继而发出失火警报信号。
作为本发明的进一步改进:所述无线发射端的数目为一个,所述无线接收端的数据为一个。
作为本发明的进一步改进:所述步骤S1评估信道状态信息包括:
S11.采集初始信道状态数据,得到一个链路数×子载波数的矩阵;
S12.对于每一个空间流,求取在同一个时间点上的30个连续子载波的CSI值作为信道状态信息;
S13.利用巴特沃斯算法对信道状态信息进行去除白噪声的影响。
作为本发明的进一步改进:所述步骤S2利用异常检测算法识别信道状态信息变化的异常是基于局部异常因子的时间序列异常检测算法。
作为本发明的进一步改进:所述步骤S2包括:
S21.对CSI数据进行数据分割得到子序列,并求出各个子序列的局部异常因子;
S22.当所述局部异常因子大于或等于预设阀值时,则将子序列作为异常模式输出;
作为本发明的进一步改进:所述步骤S3包括:
S31.事先进行实验,得到由于室内失火导致CSI(信道状态信息)变化的异常模式,提取特征,储存特征值;
S32.将步骤S22输出的异常模式计算特征值看是否与事前存储信号特征相匹配;
S33.判断是否发生火情,若是,发出火情报警信号。
本发明还提供了基于无线网络信号传输的室内探测火情及报警系统,包括:
1)CSI获取模块,用于无线接收端接受来自无线发射端的无线信号,并评估CSI(信道状态信息),并利用数据过滤技术(巴特沃斯过滤器)对数据进行去除白噪音处理;
2)异常探测模块,用于利用异常检测算法(序列异常)来识别信道状态信息变化的异常;
3)火情判断模块,采用随机森林算法,选择五种提取特征的方法(例如,标准差,四分位距等),将室内出现火情的信号信息实现储存,如果异常模式信号特征与实现存储的信号特征相匹配,则判断室内出现火情,继而发出失火警报信号;
4)警报模块,用于当室内发生火情时,发出火警信号。
作为本发明的进一步改进:所述CSI获取模块包括:
1)感应单元,用于初始信道状态数据,得到一个链路数×子载波数的矩阵;
2)数据处理单元,用于对每一个空间流,求取在同一时间点上的30个连续子载波的CSI值作为信道状态信息;
3)过滤单元,利用巴特沃斯算法对信道状态信息进行去除白噪声的影响。
作为本发明的进一步改进:所述异常检测模块基于局部异常因子的时间序列异常检测算法,包括:
1)LOF计算单元,用于对CSI数据进行数据分割得到子序列,并求出各个子序列的局部异常因子;
2)异常输出单元,用于当所述局部异常因子大于或者等于预设阀值时,将子序列作为异常模式输出。
作为本发明的进一步改进:所述火情判断模块包括:
1)建立模型单元,用于事先进行实验,得到由于室内失火导致CSI(信道状态信息)变化的异常模式,提取特征,存储特征值;
2)火情识别单元,将异常输出单元所输出的异常模式计算特征值,利用随机森林的算法,看是否与事先存储的信号特征相匹配。
本发明的有益效果:被检测的环境中并不需要额外安装探测设备或者传感器,可以在任何环境中使用,仅仅利用现有的WIFI路由器,就能进行信号收集处理,我们将CSI信道状态信息与室内失火联系起来,通过判断手机的CSI数据是否有异常,与预先得到的目标动作模式进行匹配,若发生了火情,就及时报警。因此,具有极大的普及性。
附图说明
图1为基于无线网络信号传输的室内探测火情及报警系统发送和接收数据硬件的示意图;
图2为评估CSI状态信息的示意图;
图3为基于无线网络信号传输的室内探测火情及报警方法的一实施例示意图。
具体实施方式
为了便于本领域技术人员的理解,下面结合附图和实施例对本发明作进一步的描述。
一种基于无线网络信号传输的室内探测火情及报警方法,其步骤包括:
S1.无线接收端接收来自无线发射端的无线信号,评估CSI(channel state information信道状态信息),并采用数据过滤技术(巴特沃斯滤波器)对数据进行去噪处理;
S2.利用异常检测算法(序列异常)来识别信道状态信息变化的异常;
S3.采用随机森林算法,选择五种提取特征的方法(例如,标准差,四分位距等),将室内出现火情的信号信息实现储存,如果异常模式信号特征与实现存储的信号特征相匹配,则判断室内出现火情,继而发出失火警报信号。
在实际应用中,所述无线接收端为无线网卡,所述无线发射器为无线路由器,该方法基于室内环境下的无线电传播机制,建立无线信号和失火之间的关系,只需要使用家庭现有的无线网络设备,即能够通过失火这一环境的变化而造成的无线信号的改变进行分析,判断出室内是否发生失火并进行报警。在本发明 中,所说无线发射端的数目为一个,所述无线接收端的数目为一个,系统中分别由多根天线来发送和接收无线信号;系统所使用的无线网卡可以接收信道状态信息。如附图1所述,被检测环境中存在一对无线发射端和无线接收端,接收端接收来自发射器的CSI(Channel State Information的缩写,即信道状态信息,在无线通信领域,CSI就是通信链路的信道属性,描述了信号在每条传输路径中的衰减因子)。在被检测的环境中,房间内无需安装额外的设备,就用现成的WIFI路由器就可以了。系统将利用无线接收端所接收的CSI来对室内的环境进行检测,从而对室内是否有失火现象进行判断。
为了建立无线信号和室内火情之间的联系,本发明采用无线网络的信道状态信息CSI作为指示物。CSI能够描述出在时间延迟,振幅衰减和相位转移的共同影响下,一个信号的传播途径。基于室内环境下的无线传播模型,本发明建立了CSI和室内火情之间的联系。在一个特定的室内环境中,CSI数据的传播就是一条主要路径和多条散射路径(例如天花板,地面,房间摆设),当所有东西都静止时,房内失火这一动作的变化就会使得接收端收集到的数据发生变化。CSI是利用正交频分载波复用(Orthogonal Frequency Division Multiplex,OFDM)来传输数据的,到了接收端再分解为30个子载波,通过对链路数×子载波组数据进行分析,就会对室内情况进行比较准确的判断。
具体地,所述步骤S1中,评估CSI状态信息包括:
S1.无线接收端接收来自无线发射端的无线信号,评估CSI(channel state information信道状态信息),并采用数据过滤技术(巴特沃斯滤波器)对数据进行去噪处理;
S2.利用异常检测算法(序列异常)来识别信道状态信息变化的异常;
S3.采用随机森林算法,选择五种提取特征的方法(例如,标准差,四分位距等),将室内出现火情的信号信息实现储存,如果异常模式信号特征与实现存储的信号特征相匹配,则判断室内出现火情,继而发出失火警报信号。
当本发明的系统开始工作时,无线发射端(AP)发射无线信号,装有无线网卡的电脑作为无线接收端会收集CSI数据作为初始信道状态信息。实验中,我们是采用AP有三根天线,电脑端有一根天线,从而形成了三条信号链路,每个链路会有30个子载波,因此我们可以得到一个1×3×30的矩阵。为了消除白噪音的影响,我们采用巴特沃斯算法对信号进行去噪处理。
所述步骤S2是利用异常检测算法识别信道状态信息变化的异常是基于局部因子的时间序列异常检测算法。步骤S2包括:
S21.对CSI数据进行数据分割得到子序列,并求出各个子序列的局部异常因子;
S22.当所述局部异常因子大于或等于预设阀值时,则将子序列作为异常模式输出;
使用LOF一个点的局部密度就会与它的邻居进行比较。如果前者明显低于后者(有一个大于1的LOF值),该点则位于一个稀疏区域,对它的邻居而言,这就表明,该点是一个异常值,得到异常模式。用上述方法,我们会将由于室内失火这一环境因素导致的信道状态信息变化的到的异常模式计算的特征值在随机森林算法中作为目标特征。
在完成异常检测模块以后,由于各种原因(例如:人体活动)造成信道影响都会输出相对应的的异常模式,这些异常模式都会分别计算特征值,并在随机森林算法中与事先存储的信号特征进行对比,如果相吻合,则判定为失火,就进行报警。
进一步地,所述步骤S3包括:
S31.事先进行实验,得到由于室内失火导致CSI(信道状态信息)变化的异常模式,提取特征,存储特征值;
S32.将步骤S22输出的异常模式计算特征值,利用随机森林算法看是否与事先存储的信号特征匹配;
S33.判断是否发生火情,若是,发出火情报警信号。
具体地,如附图2所示,系统主要分为数据处理,异常检测以及动作分类。
具体地,如附图3所示,实现室内探测火情及报警的流程。
1)无线发射端(AP)发送无线信号,无线接收端(带无线网卡的电脑)接收无线信号,采集CSI(信道状态信息)数据;
2)取每一条信号链路的30个子载波的值作为该时刻的信道状态信息;
3)用巴特沃斯算法对收集的信号做去噪处理;
4)利用异常检测算法计算异常因子;
5)输出异常模式,并提取特征;
6)利用随机森林进行分类;
7)若为目标动作类(室内失火所导致的信道状态信息的异常模式)则发出报警信号;
8)若否,则继续收集信号,不做任何处理。
本发明是一种基于无线网络信号传输的室内探测火情及报警系统,其中包含的模块如附图5所示,包括:
CSI获取模块,用于接收无线信号,并评估CSI信道状态信息;
异常检测模块,用于识别信道状态信息的异常变化;
火情判断模块,用于判断异常模式的特征是否与事先存储的信息特征相吻合(随机森林),进而判断房间内是否有火情;
报警模块,用于在室内有火情发生时发出报警信号。
进一步地,所述CSI获取模块包括:
感应单元,用于初始信道状态数据,得到一个链路数×子载波数的矩阵;
数据处理单元,用于对每一个空间流,求取在同一时间点上的30个连续子载波的CSI信道状态信息;
过滤单元,利用巴特沃斯算法对信道状态信息进行去除白噪声的影响。
进一步地,所述异常检测模块基于局部异常因子的时间序列异常检测算法,包括:
LOF计算单元,用于对CSI数据进行数据分割得到子序列,并求出各个子序列的局部异常因子;
异常输出单元,用于当所述局部异常因子大于或者等于预设阀值时,将子序列作为异常模式输出。
进一步地,所述火情判断模块包括:
建立模型单元,建立模型单元,用于事先进行实验,得到由于室内失火导致CSI(信道状态信息)变化的异常模式,提取特征,存储特征值;
火情识别单元,将异常输出单元所输出的异常模式计算特征值,利用随机森林的算法,看是否与事先存储的信号特征相匹配。
最后利用报警模块在火情发生的情况下,发出警报(例如,给房主发短信等)。
以上内容是结合具体实现方式对本发明做的进一步阐述,不应认定本发明的具体实现只局限于以上说明。对于本技术领域的技术人员而言,在不脱离本发明构思的前提下,还可以做出若干简单推演或替换,均应视为有本发明所提交的权利要求确定的保护范围之内。

Claims (9)

  1. 一种基于无线网络信号传输的室内探测火情及报警方法,其特征在于:包括如下步骤:
    (S1)无线接收端接收来自无线发射端的无线信号,评估信道状态信息CSI,并采用数据过滤技术对数据进行去噪处理;
    (S2)利用异常检测算法来识别信道状态信息变化的异常;
    (S3)采用随机森林算法,选择提取特征的方法将室内出现火情的信号信息实现储存,如果异常模式信号特征与实现存储的信号特征相匹配,则判断室内出现火情,继而发出失火警报信号。
  2. 根据权利要求1所述的基于无线网络信号传输的室内探测火情及报警方法,其特征在于:所述无线发射端的数目为一个,所述无线接收端的数据为一个。
  3. 根据权利要求1所述的基于无线网络信号传输的室内探测火情及报警方法,其特征在于,所述步骤(S1)评估信道状态信息包括:
    (S11)采集初始信道状态数据,得到一个链路数×子载波数的矩阵;
    (S12)对于每一个空间流,求取在同一个时间点上的30个连续子载波的CSI值作为信道状态信息;
    (S13)利用巴特沃斯算法对信道状态信息进行去除白噪声的影响。
  4. 根据权利要求1所述的基于无线网络信号传输的室内探测火情及报警方法,其特征在于:所述步骤(S2)是利用异常检测算法识别信道状态信息变化的异常是基于局部因子的时间序列异常检测算法;所述步骤(S2)进一步包括:
    (S21)对CSI数据进行数据分割得到子序列,并求出各个子序列的局部异常因子;
    (S22)当所述局部异常因子大于或等于预设阀值时,则将子序列作为异常模式输出。
  5. 根据权利要求1所述的基于无线网络信号传输的室内探测火情及报警方法,其特征在于:所述步骤(S3)具体为:
    (S31)事先进行实验,得到由于室内失火导致CSI变化的异常模式,提取特征,并储存;
    (S32)将步骤(S2)输出的异常模式提取特征后用随机森林算法计算,看是否与之前存储的信息特征匹配;
    (S33)判断是否发生火情,若是,发出火情报警信号。
  6. 一种基于无线网络信号传输的室内探测火情及报警系统,其特征在于,包括:
    信道状态信息CSI获取模块,用于无线接收端接受来自无线发射端的无线信号,并评估CSI,并利用数据过滤技术对数据进行去除白噪音处理;
    异常探测模块,用于利用异常检测算法(序列异常)来识别信道状态信息变化的异常;
    火情判断模块,采用随机森林算法,选择提取特征的方法将室内出现火情的信号信息实现储存,如果异常模式信号特征与实现存储的信号特征相匹配,则判断室内出现火情,继而发出失火警报信号;
    警报模块,用于当室内发生火情时,发出火警信号。
  7. 根据权利要求6所述的基于无线网络信号传输的室内探测火情及报警系统,其特征在于,所述的CSI获取模块包括:
    感应单元,用于初始信道状态数据,得到一个链路数×子载波数的矩阵;
    数据处理单元,用于对每一个空间流,求取在同一时间点上的30个连续子载波的CSI值作为信道状态信息;
    过滤单元,利用巴特沃斯算法对信道状态信息进行去除白噪声的影响。
  8. 根据权利要求6所述的基于无线网络信号传输的室内探测火情及报警系统,其特征在于:所述异常检测模块基于局部异常因子的时间序列异常检测算法,包括:
    LOF计算单元,用于对CSI数据进行数据分割得到子序列,并求出各个子序列的局部异常因子;
    异常输出单元,用于当所述局部异常因子大于或者等于预设阀值时,将子序列作为异常模式输出。
  9. 根据权利要求6所述的基于无线网络信号传输的室内探测火情及报警系统,其特征在于:所述火情判断模块包括:
    建立模型单元,用于事先进行实验,得到由于室内失火导致CSI变化的异常模式,特征提取,储存特征值;
    火情识别单元,用于将异常输出单元所输出的异常模式特征利用随机森林算法,看是否与实现存储的特征相匹配。
PCT/CN2016/084192 2016-01-06 2016-05-31 基于无线网络信号传输的室内探测火情及报警的方法及系统 WO2017117910A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610002723.4 2016-01-06
CN201610002723.4A CN105761407B (zh) 2016-01-06 2016-01-06 基于无线网络信号传输的室内探测火情及报警方法及系统

Publications (1)

Publication Number Publication Date
WO2017117910A1 true WO2017117910A1 (zh) 2017-07-13

Family

ID=56342314

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/084192 WO2017117910A1 (zh) 2016-01-06 2016-05-31 基于无线网络信号传输的室内探测火情及报警的方法及系统

Country Status (2)

Country Link
CN (1) CN105761407B (zh)
WO (1) WO2017117910A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112163540A (zh) * 2020-10-09 2021-01-01 上海第二工业大学 一种基于WiFi的姿态识别方法
CN113176523A (zh) * 2021-04-27 2021-07-27 深圳供电局有限公司 一种基于多传感器的电力设备火灾预警方法及系统
WO2023049962A1 (en) * 2021-10-01 2023-04-06 Newsouth Innovations Pty Limited System and method of detecting environmental changes using wireless sensing

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106368813B (zh) * 2016-08-30 2018-09-25 北京协同创新智能电网技术有限公司 一种基于多元时间序列的异常报警数据检测方法
CN107025751B (zh) * 2017-03-10 2018-05-08 深圳大学 基于无线信号传输的室内火情探测和报警的方法及其系统
CN108416974B (zh) * 2018-01-23 2020-07-07 中国矿业大学 基于无线信道状态信息的自动报警装置及方法
CN110059567A (zh) * 2019-03-21 2019-07-26 浙江工业大学 一种基于WiFi的火灾烟雾检测方法
CN111221018B (zh) * 2020-03-12 2022-04-08 南京航空航天大学 一种用于抑制海上多路径的gnss多源信息融合导航方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103023589A (zh) * 2012-12-06 2013-04-03 中山大学 一种室内无源被动式移动检测方法及其检测装置
CN103606248A (zh) * 2013-09-30 2014-02-26 广州市香港科大霍英东研究院 一种人体摔倒自动检测方法及系统
CN104766427A (zh) * 2015-04-27 2015-07-08 太原理工大学 一种基于Wi-Fi的房屋非法入侵检测方法
CN104796204A (zh) * 2015-03-16 2015-07-22 深圳大学 一种基于无线网络的金属携带检测方法
CN104951757A (zh) * 2015-06-10 2015-09-30 南京大学 一种基于无线信号的动作检测和识别的方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2514103B (en) * 2013-05-12 2017-08-02 Microsense Solutions Ltd Bidirectional bistatic radar perimeter intrusion detection system
CN104751591B (zh) * 2015-04-01 2017-03-22 大连希尔德安全技术有限公司 一种基于半透明特征的早期火灾烟雾检测方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103023589A (zh) * 2012-12-06 2013-04-03 中山大学 一种室内无源被动式移动检测方法及其检测装置
CN103606248A (zh) * 2013-09-30 2014-02-26 广州市香港科大霍英东研究院 一种人体摔倒自动检测方法及系统
CN104796204A (zh) * 2015-03-16 2015-07-22 深圳大学 一种基于无线网络的金属携带检测方法
CN104766427A (zh) * 2015-04-27 2015-07-08 太原理工大学 一种基于Wi-Fi的房屋非法入侵检测方法
CN104951757A (zh) * 2015-06-10 2015-09-30 南京大学 一种基于无线信号的动作检测和识别的方法

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112163540A (zh) * 2020-10-09 2021-01-01 上海第二工业大学 一种基于WiFi的姿态识别方法
CN112163540B (zh) * 2020-10-09 2024-01-19 上海第二工业大学 一种基于WiFi的姿态识别方法
CN113176523A (zh) * 2021-04-27 2021-07-27 深圳供电局有限公司 一种基于多传感器的电力设备火灾预警方法及系统
CN113176523B (zh) * 2021-04-27 2022-07-26 深圳供电局有限公司 一种基于多传感器的电力设备火灾预警方法及系统
WO2023049962A1 (en) * 2021-10-01 2023-04-06 Newsouth Innovations Pty Limited System and method of detecting environmental changes using wireless sensing

Also Published As

Publication number Publication date
CN105761407A (zh) 2016-07-13
CN105761407B (zh) 2019-05-10

Similar Documents

Publication Publication Date Title
WO2017117910A1 (zh) 基于无线网络信号传输的室内探测火情及报警的方法及系统
US11138854B2 (en) Digital fingerprint tracking
US20220225101A1 (en) Ai cybersecurity system monitoring wireless data transmissions
CN204288415U (zh) 智能预警灭火机器人
WO2018161433A1 (zh) 基于无线信号传输的室内火情探测和报警的方法及其系统
CN104299351A (zh) 智能预警灭火机器人
Islam et al. Fire detection system with indoor localization using ZigBee based wireless sensor network
CN104978829A (zh) 一种室内吸烟监控控制方法及系统
CN110910604B (zh) 用于火灾报警的监控方法、系统以及装置
KR20150114373A (ko) 음장 스펙트럼의 상관계수의 변화 패턴을 이용한 보안 감시 장치 및 그것의 보안 감시 방법
CN105575034A (zh) 一种双波段森林防火智能监控软件图像处理分析方法
CN105405260A (zh) 一种基于无线信号的防盗系统及防盗方法
Zhong et al. Wi-fire: Device-free fire detection using WiFi networks
WO2018006798A1 (zh) 基于无线感知的湿度检测的方法及系统
CN110070692A (zh) 一种智能视觉消防监视系统及方法
CN102842199B (zh) 火灾识别方法和系统
CN106502115A (zh) 一种基于gps技术的智能家居控制系统及方法
CN110059567A (zh) 一种基于WiFi的火灾烟雾检测方法
US11424843B2 (en) Cellular signal degradation based detection of events
Saida et al. LORA Based Forest Fire Monitoring System
Islam et al. Measuring trustworthiness of IoT image sensor data using other sensors’ complementary multimodal data
US8977514B2 (en) Thermal activity detection and response
Sampath et al. Sensor network based device free trespasser detection and localisation
CN116168499A (zh) 基于csi幅值平稳性分析的室内人员被动感知与入侵检测方法
WO2021108361A1 (en) Nonintrusive detection of concealed suspicious objects using devices with wi-fi transceiver clusters

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16883051

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 19.10.18)

122 Ep: pct application non-entry in european phase

Ref document number: 16883051

Country of ref document: EP

Kind code of ref document: A1