CN102008291A - Single-channel UWB-based radar type life detection instrument for multi-target detection - Google Patents

Single-channel UWB-based radar type life detection instrument for multi-target detection Download PDF

Info

Publication number
CN102008291A
CN102008291A CN 201010502319 CN201010502319A CN102008291A CN 102008291 A CN102008291 A CN 102008291A CN 201010502319 CN201010502319 CN 201010502319 CN 201010502319 A CN201010502319 A CN 201010502319A CN 102008291 A CN102008291 A CN 102008291A
Authority
CN
China
Prior art keywords
signal
target
pulse
digital
single
Prior art date
Application number
CN 201010502319
Other languages
Chinese (zh)
Other versions
CN102008291B (en
Inventor
于霄
吕昊
张杨
李岩峰
李钊
王健琪
荆西京
Original Assignee
中国人民解放军第四军医大学
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 中国人民解放军第四军医大学 filed Critical 中国人民解放军第四军医大学
Priority to CN2010105023196A priority Critical patent/CN102008291B/en
Publication of CN102008291A publication Critical patent/CN102008291A/en
Application granted granted Critical
Publication of CN102008291B publication Critical patent/CN102008291B/en

Links

Abstract

The invention discloses a single-channel ultra-wide bandwidth (UWB)-based radar type life detection instrument for multi-target detection. The life detection instrument comprises a UWB biologic radar front end and a calculating unit, wherein the UWB biologic radar front end comprises a transmitting antenna, a receiving antenna, a pulse oscillator, an electromagnetic pulse generator and a sampling integrator; the pulse oscillator generates a pulse signal; the signal triggers the electromagnetic pulse generator to generate a narrow pulse and radiates the narrow pulse out through the transmitting antenna; a reflected signal is transmitted to the sampling integrator through the receiving antenna; a pulse signal generated by the pulse oscillator simultaneously generates a distance gate through a delay circuit and a distance gate generator to select a received signal; the signal passes through a sampling integration circuit; a weak signal is detected after accumulation, amplified and filtered by an amplifier and a filter, sampled by a high-speed analogue/digital (A/D) acquisition card and transmitted to the calculating unit; and the acquired signal is analyzed by the calculating unit to extract life information and each target distance of multiple human targets.

Description

一种可用于多目标探测的单通道基于UWB的雷达式生命探 Radar Life probe which can be used single-channel multi-target detection of UWB-based

测仪 Tester

技术领域 FIELD

[0001] 本发明涉及属于非接触生命参数探测技术领域,特别涉及一种可用于多目标探测的单通道基于UWB的雷达式生命探测仪。 [0001] The present invention relates belongs to the technical field of life parameters of non-contact detection, and more particularly to radar life detector for single channel A multiple target detection based on the UWB.

背景技术 Background technique

[0002] 雷达式生命探测仪是一种融合雷达技术和生物医学工程技术可穿透非金属介质(砖墙、废墟等)非接触、远距离地探测人类生命体(呼吸、心跳、体动等)的一种新兴特殊雷达。 [0002] radar life detector radar technology and is a fusion of biomedical engineering technology penetrate non-metallic medium (brick, rubble, etc.) a non-contact, distance detected human beings (respiration, heart rate, body movement, etc. ) is an emerging special radar. 而雷达式生命探测仪技术则是以生命体为探测目标的一项新兴技术,是国际科技界公认的一个非常重要的前沿技术领域。 The radar life detector technology is based on a living body to detect targets an emerging technology that is recognized by the international scientific community in a very important field of cutting-edge technology. 由于该技术对被测量对象无任何约束,无需接触性电极、传感器、电缆等的连接,而且可以隔一定的距离、穿透一定的介质(如衣服、纱布、砖墙、废墟等)对人体进行识别探测,所以可广泛用于灾害被埋人员搜救、反恐斗争中隔墙监控及战场侦察等领域,特别是在应急救援、反恐等领域具有不可替代的优势。 Since no constraint on the technical object to be measured without contact electrodes, sensors, etc. connected to cable, and can be separated a certain distance, to penetrate a certain medium (such as clothes, gauze, brick, rubble, etc.) on the human body identification detection, it can be used in disaster search and rescue personnel were buried, and monitor the fight against terrorism partition battlefield reconnaissance and other fields, in particular, has irreplaceable advantages in the field of emergency rescue and counter-terrorism.

[0003] 目标识别能力和距离、角度分辨力是当今雷达式生命探测仪领域研究的两个重点,也是本文需要突破的关键问题。 [0003] ability to identify and target distance, angle resolution is focused on two of today's radar life detector field, the key issue is the need to break this article. 目前,较为成熟的基于连续波雷达体制的雷达式生命探测仪系统只能给出有人无人的结果,而无法给出目标的距离和角度信息等,穿透能力也有待进一步提高。 Currently, more mature only give life detector radar system based on continuous wave radar system was no result, but can not give information about the target's distance and angle, etc., penetration also be further improved. 鉴于超宽谱雷达所具有的优势,我们采用了目前国际上先进的超宽谱技术,将其与非接触生命探测技术相结合,研究基于超宽谱的非接触探人雷达技术。 Given the wide spectrum of radar has the advantage, we have adopted a wide spectrum of current advanced technology in the world, its life and non-contact detection technology, research Detection Radar technology of non-contact wide spectrum.

[0004] 现行的雷达式生命探测技术以对单目标的探测识别为主,对多目标的探测和定位也仅限于运动目标。 [0004] The current life detector radar technology to detect and identify targets based on a single, multi-detection and location of targets is limited to moving targets. 到目前为止,该领域尚未解决多个静止人体目标的识别和定位问题。 So far, the field has not been resolved to identify and locate multiple targets still human. 多静目标探测识别定位技术是国际生命探测领域的一个新的研究方向和难点,该技术是雷达式生命探测仪的关键技术,它制约着雷达式生命探测仪的广泛应用。 Multi-target detection and recognition static positioning technology is a new and difficult international research in the field of life detection, the technology is a key technology life style radar detector, which restricts the wide application of radar life detector. 多静目标探测识别定位难题的解决可以极大地提高非接触生命探测中的探测效率,满足实际工作中对多目标快速探测定位的需求。 Multi-target detection and recognition static positioning problem solving can greatly improve the detection efficiency non-contact detection of life, practical work to meet the multi-target rapid detection and location of demand right.

发明内容 SUMMARY

[0005] 本发明所要解决的技术问题是针对现有技术的不足,提供一种单通道的可实现多目标探测的基于UWB的雷达式生命探测仪,解决多个静止人体目标的探测和定位问题。 [0005] The present invention solves the technical problem is lack of the prior art, to provide a single channel can be multi-target detection based UWB radar life detector, resolve to detect and locate multiple stationary targets human problem .

[0006] 一种可用于多目标探测的单通道基于UWB的雷达式生命探测仪,包括UWB生物雷达前端和计算单元,所述UWB生物雷达前端包括发射天线、接收天线、脉冲振荡器、电磁脉冲产生器、取样积分器;脉冲振荡器产生脉冲信号,该信号触发电磁脉冲产生器产生窄脉冲,并通过发射天线辐射出去;反射信号经过接收天线送到取样积分器, 由脉冲振荡器产生的脉冲信号同时经过延时电路和距离门产生器产生距离门,对接收信号进行选择,信号通过取样积分电路,经过积累后微弱信号被检测出来,并经由放大器和滤波器进行放大、滤波,再经高速A/D采集卡采样后送入计算单元,由计算单元对采集到的信号进行分析处理,最终提取多个人体目标生命信息和各目标距离。 [0006] A single channel for multiple target detection based UWB radar life detector, including biological UWB radar front-end and calculating means, said UWB radar front end includes biological transmitting antenna, receiving antenna, an oscillator pulse, electromagnetic pulse generator, sampling integrator; pulse oscillator generates a pulse signal which triggers electromagnetic pulse generator generates a narrow pulse, and radiated through the transmitting antenna; reflected signal to the receiving antenna via an integrator sampling pulses generated by the pulse oscillator through the delay circuit and the signal at the same range gate generator generates range gate, the received signal is selected, the sampling signal integration circuit, after the accumulation of a weak signal is detected, and amplifies, filters and amplifiers via a filter, then by high-speed after the a / D acquisition card samples into the computing unit, analyze and process the signal collected by the calculation unit, the final extract a plurality of targets of human life, and each object distance information.

[0007] 所述的单通道UWB雷达式生命探测仪,所述滤波器采用增益为1、通带为0.08-5000HZ硬件滤波电路。 [0007] The single-channel UWB radar life detector, said filter using the gain is 1, the pass band of the filter circuit 0.08-5000HZ hardware.

[0008] 所述的单通道UWB雷达式生命探测仪,所述计算单元包括信号积分模块、信号分解重构模块、数字滤波模块和数字微分模块、空间频率分析模块,所述信号积分模块在距离上对信号进行积分,所述信号分解重构模块将信号打散进行分解、重构,合成目标回波信号和距离信号,所述数字滤波和数字微分模块对目标回波信号进行数字滤波和数字微分,所述空间频率分析模块用于根据数字滤波和数字微分后的目标回波信号以及距离信号进行空间频率分析,得到目标一维距离。 [0008] The single-channel UWB radar life detector, the calculation unit includes a signal integrator module, the signal decomposition and reconstruction module, digital filtering and digital modules differential module, spatial frequency analyzing module, the signal from the integrator module integrating the signal on the signal decomposition and reconstruction module will break the signal decomposition, reconstruction, synthesis target echo signal and a distance signal, said digital filtering and digital differential module target echo signal digital filtering and digital differential, the spatial frequency analysis means for performing spatial frequency analysis according to the target signal and the echo signal from the digital filtering and digital differential, to give a certain dimension range.

[0009] 所述的单通道UWB雷达式生命探测仪,所述数字滤波模块采用160阶、截止频率为0.5Hz的海明窗FIR数字滤波器。 [0009] The single-channel detector life UWB radar, the digital filter module 160 uses the order, the cutoff frequency of the digital FIR Hamming filter is 0.5Hz.

[0010] 所述的单通道UWB雷达式生命探测仪,所述数字微分模块采用60阶数字微分 [0010] The single-channel UWB radar life detector, the digital differentiator block 60 using a digital differential stage

ο ο

[0011] 所述的单通道UWB雷达式生命探测仪,还包括波峰判别模块,用于根据空间频率分析结果和系统预设置阈值判断波峰位置是否有人体目标存在。 [0011] The single-channel detector UWB radar life, further comprising a peak discriminating means for frequency analysis results and the system determines a pre-set threshold value if the peak position according to the presence of certain human body space.

[0012] 本发明的创新之处在于: [0012] The innovation of the present invention:

[0013] (1)提出了先实现对静止人体目标微弱生命特征的增强、人体识别和一维距离区分,再进行多目标二维定位的新方法,为雷达式生命探测仪的多个静止人体目标定位开辟新的途径。 [0013] (1) A new method to achieve enhanced stationary target human weak vital signs, and body identification to distinguish from a one-dimensional, two-dimensional multi-target then positioning a plurality of static radar life detector body targeting open up new avenues.

[0014] (2)采用以改形的时频分析方法一空频分析(空间、频率)为主的一维距离区分算法对单通道超宽谱雷达式生命探测仪系统采集的回波信号进行拆分、重组和有关处理,可望为生命探测中多个静目标的一维距离区分提供新的方法。 [0014] (2) Analysis using a space-frequency echo signal (spatial frequency) mainly distinguished from a one-dimensional single-channel algorithm wide spectrum radar life detector system to collect the time-frequency analysis reshaping method demolition points, restructuring and related processing, are expected to provide a new method to distinguish multiple one-dimensional distance motionless object detection in life.

附图说明 BRIEF DESCRIPTION

[0015] [0015]

[0016] [0016]

[0017] [0017]

[0018] [0018]

[0019] [0019]

[0020] [0021] [0022] [0020] [0021] [0022]

[0023] [0023]

[0024] [0024]

[0025] [0025]

[0026] [0026]

图1为单通道超宽谱雷达式生命探测仪系统原理框图; FIG 1 is a single-channel radar UWS life detector system block diagram;

图2为单通道超宽谱雷达式生命探测仪计算单元结构示意图; FIG 2 is a single-channel radar UWS life detector schematic structural calculation unit;

图3为超宽谱雷达参数设置示意图; 3 is a schematic view of wide spectrum radar parameters;

图4为目标回波信号和距离信号; FIG. 4 is a target echo signal and a distance signal;

图5为硬件滤波电路组成方框图; FIG 5 is a block diagram showing a hardware filtering circuit;

图6为微分算法前后的信号波形比较(30秒数据); 6 is a signal waveform before and after differential comparison algorithm (30 seconds of data);

图7为波峰判别方法对双目标数据进行判别的结果; FIG 7 is a double peak discriminating method discriminating result of the object data;

图8为自由空间无目标的时频图(无目标); 8 is the free space in FIG non-target frequency (non-targeted);

图9为自由空间单目标的时频图(目标实际位置6.5m); FIG frequency (actual target position 6.5m) Figure 9 is a free space of a single target;

图10为自由空间双目标的时频图(目标实际位置2.5m和7.5m); FIG frequency (target and actual position 2.5m 7.5m) Figure 10 is a free space of the double goal;

图11为穿30cm砖墙单目标的时频图(目标实际位置6.0m); FIG 11 FIG frequency (actual target position 6.0m) is a single brick wear 30cm target;

图12为穿30cm砖墙双目标的时频图(目标实际位置3.0m禾Π 6.0m)。 Figure 12 is a dual-brick wear 30cm FIG target frequency (the actual position of the target 3.0m Wo Π 6.0m). 具体实施方式 Detailed ways

[0027] 以下结合附图和具体实施例,对本发明进行详细说明。 [0027] The following embodiments in conjunction with accompanying drawings and specific embodiments, the present invention will be described in detail.

[0028] 实施例1 [0028] Example 1

[0029] 本实施例提供一种单通道超宽谱雷达式生命探测仪,图1为单通道超宽谱雷达式生命探测仪系统原理框图。 [0029] The present embodiment provides a wide spectrum of single-channel radar life detector, FIG. 1 is a single-channel radar UWS life detector system block diagram. 首先脉冲振荡器产生脉冲信号,该信号触发电磁脉冲产生器产生窄脉冲,并通过发射天线辐射出去。 First pulse oscillator generates a pulse signal which triggers electromagnetic pulse generator generates a narrow pulse, and radiated through the transmitting antenna. 反射信号经过接收天线送到取样积分器,由脉冲振荡器产生的信号经过延时电路产生距离门,对接收信号进行选择,信号通过取样积分电路,经过成千上万个脉冲的积累后微弱信号被检测出来,并进行放大滤波,再经高速采集卡采样后送入计算单元,由计算单元对采集到的信号进行分析处理并识别,最后计算目标距离。 After the reflected signal to the receiving antenna sampler integrator, an oscillator signal generated by the pulse generated from the gate through the delay circuit, selection of the received signal, by sampling a signal integration circuit, after the accumulation of thousands of weak signal pulse It is detected, amplified and filtered, and then the high-speed acquisition card samples into the computing unit, analyze and process the signal collected by the calculation unit and the identification of the last calculated target distance.

[0030] 如图1所示,虚线框内为雷达前端,系统的中心频率和带宽同为500MHz,波束覆盖角度为60°。 [0030] 1, within the dashed box for the center frequency and bandwidth radar front-end system for the same 500MHz, beam coverage angle of 60 °. 计算单元控制距离门产生器来获得探测区域内不同距离段的回波信号。 Calculating a distance gate unit controls the generator to obtain the echo signal segments from the different detection area.

[0031] 计算机可控制的参数为:起始距离、探测范围、采样频率和天线增益。 [0031] computer-controlled parameters are: start distance, the detection range, the sampling frequency and antenna gain. 如图3 所示,天线穿透砖墙后,探测区域为一扇形,通过设置起始距离和探测范围,可以实现图中阴影部分的扇形区域的扫描探测,如果回波信号经分析后显示有目标信息,就可以判定此扇形区域内有目标。 Shown, the antenna penetrate walls, the detection region is a sector 3, and by setting a starting distance detection range, scanning may be effectuated in the detection sector area shaded portion in FIG. If the display has an echo signal analysis target information, it can be determined within this target sector area. 通过不断的调整起始距离,可以实现一定区域内的断层扫描。 By constantly adjusting the initial distance can be achieved within a certain area tomography. 而调整探测范围(天线的接收点数不变),则可以调整探测系统的灵敏度,改变系统的目标距离分辨力,实现一定区域内的粗扫和细扫。 Adjusted detection range (constant reception antenna points), you can adjust the sensitivity of the detection system, the system changes the target range resolution, within a certain coarse-scan and fine scan region.

[0032] 例如,起始距离设置为6m(40纳秒),探测范围设置为3m(20纳秒),本系统的回波信号为2048个点组成的序列,那么当前雷达有效探测区域为天线正前方6m〜9m, 角度为60°的扇形区域,回波信号只反映纵向上3m的信息,且将3m的范围平均等分为2048份,即每次采样得到2048个数据,我们称之为2048个点,第η点代表的距离为: [0032] For example, the initial distance is set to 6m (40 ns), the detection range is set to 3m (20 ns), the echo signal system is composed of a sequence of 2048 points, the effective detection region of radar antenna front 6m~9m, an angle of 60 ° sector area, only the echo signal is reflected on the information longitudinal 3m, 3m and the average range divided into 2048 parts, i.e., 2048 data obtained for each sample, we call 2048 points, from the point of η is represented by:

[0033] s = 6 + -^—x3(<m) ......(1) [0033] s = 6 + - ^ - x3 (<m) ...... (1)

2048 2048

[0034] 公式(1)中:η为点序数。 [0034] Equation (1): η is the number of the dot-sequential.

[0035] 根据奈奎斯特采样定理,采样频率必须大于信号最高频率的两倍,我们设定A/ D采样频率为64Hz。 [0035] According to the Nyquist sampling theorem, the sampling frequency must be greater than twice the highest frequency of the signal, we set the A / D sampling frequency of 64Hz.

[0036] 图2为本发明单通道超宽谱雷达式生命探测仪计算单元结构示意图;所述计算单元包括信号积分模块、信号分解重构模块、数字滤波模块和数字微分模块、空间频率分析模块,所述信号积分模块在距离上对信号进行积分,所述信号分解重构模块将信号打散进行分解、重构,合成目标回波信号和距离信号,所述数字滤波和数字微分模块对目标回波信号进行数字滤波和数字微分,所述空间频率分析模块用于根据数字滤波和数字微分后的目标回波信号以及距离信号进行空间频率分析,得到目标一维距离。 [0036] FIG. 2 is a schematic diagram of a single-channel radar wide spectrum calculation unit life detector construction of the invention; the calculating unit comprises a signal integrator module, the signal decomposition and reconstruction module, digital filtering and digital modules differential module, spatial frequency analyzing module the signal integration module integrates the signal over a distance, the signal into the signal reconstruction module scattered decomposition, reconstruction, synthesis target echo signal and a distance signal, said digital filtering and digital differentiation of the target module digitally filtering the echo signals and digital differential, the spatial frequency analysis means for performing spatial frequency analysis according to the target signal and the echo signal from the digital filtering and digital differential, to give a certain dimension range.

[0037] 实施例2 [0037] Example 2

[0038] 本实施例提供单通道超宽谱雷达生命探测仪参数指标: [0038] The present embodiment provides a wide spectrum of single-channel radar life detector parameter index:

[0039] 系统的天线、收发系统相关参数指标如下: [0039] The antenna system, a transceiver system parameters indicators are as follows:

[0040] (1)天线为:介质耦合屏蔽型; [0040] (1) antenna: medium is coupled shielded;

[0041] (2)天线个数:1个发射天线、1个接收天线;[0042] (3)收发天线中心频率:500MHz ; [0041] (2) the number of antennas: a transmitting antenna, a receiving antenna; [0042] (3) and receiving antenna center frequency: 500 MHz;

[0043] (4)带宽:500MHz ; [0043] (4) Bandwidth: 500MHz;

[0044] (5)时窗:4 〜5000ns 可调; [0044] (5) When the window: 4 ~5000ns adjustable;

[0045] 实施例3 [0045] Example 3

[0046] 静目标微弱信号增强: [0046] static target weak signal enhancement:

[0047] 要实现静止人体目标的识别,首先应该对静止人体的微弱生命信号进行增强。 [0047] To achieve the goal of identifying a stationary body, first of all should enhance the weak signal of a still life of the human body. 在本实施例中,针对UWB雷达回波信号的特点,采用微弱生物医学信号处理方法对经过高速采样后信号进行处理和有用信号的增强,来提高信噪比,实现对人体目标的基本识别。 In the present embodiment, the characteristics of UWB radar echo signals, using a weak biomedical signal processing method for processing and enhancing the useful signal after the high-speed sampling the signal to noise ratio, identification of human basic goals.

[0048] 采用8点间4点积分法在距离上对信号进行积分;再将信号打散进行分解、重构,合成目标回波信号和距离信号;对目标回波信号进行数字滤波和数字微分,以实现微弱有用信号的增强。 [0048] The Points between 8:04 in the method the signal from the integrator; then decomposed scattered signal, reconstructed, and the synthetic echo signal from the target signal; target echo signal is digitally filtered and digitized differential in order to achieve a weak useful signal enhancement.

[0049] 3.1信号的积分 [0049] 3.1 signal integration

[0050] 本实施例中采用的高速采集卡采样率为64Hz,则经AD采样后的数据量大,不利于实时运算;数据量减少太多又会导致回波信号缺少足够的距离信息。 [0050] This high speed data acquisition card sampling rate employed in the embodiment of 64Hz, the amount of data by the AD sampling, is not conducive to real-time operation; much reduced amount of data will lead to an echo signal from a lack of adequate information. 所以在确保拥有足够的距离分辨力的情况下本实施例选取8点间4积分法对采样后信号进行分段积分。 Therefore, in the present embodiment, selected to ensure that have between 8:04 integration of the sampled signal segment integrating case where a sufficient range resolution.

[0051] 8点间4积分法就是把数据每8点相加取平均,每两次积分间隔4点(0〜7, 4〜11,8〜15,后面类推),使采样后信号数据量经过距离上的积分变成了原信号的四分之一,在不损失信号特征的情况下减少了信号的序列长度,降低了运算量,加快了运算速度。 [0051] The integration between 8:04 summed data is taken every 8-point average, twice per integration interval 4:00 (0 to 7, 4~11,8~15, later on), so that the amount of data sampled signal after integration into a quarter of the distance of the original signal is reduced without loss of signal characteristic length of the signal sequence, the reduced amount of computation, computing speed faster.

[0052] 3.2信号的分解重构 [0052] The signal decomposition and reconstruction 3.2

[0053] 将积分后信号按时间和空间两个域进行分解、重构,合成含有时间信息的目标回波信号X(t)和含有空间信息的距离信号y(d),其中t为时间变量,d为距离变量。 [0053] After the integrated signal decomposition time and space domains, reconstruction, the target echo signal X (t) and the time information contained synthetic signal from the spatial information contained y (D), where t is the time variable , d is the distance variable. 目标回波信号反映的是对应距离点上的信号幅值随时间变化的情况,目标回波信号的横坐标是时间;而距离信号则为同一时刻不同距离上的各点的幅值组成的序列,距离信号的横坐标为距离。 Target echo signal is reflected on the amplitude of the signal point distance of the corresponding time-varying, the abscissa is the time of the target echo signal; and the sequence compared with the same timing signal from the amplitude of each point of the composition at different distances , as the distance from the abscissa signal. 图4为随机选择的一路目标回波信号(1600点,25秒)和距离信号(60ns, 9m)波形图。 4 all the way to the target echo signal is randomly selected (1600, 25 seconds) and the distance signal waveform diagrams (60ns, 9m) FIG.

[0054] 目标回波信号提高了信噪比,更有利于生命特征信号的提取,距离信号在大大降低运算量的同时,又保证了合适的距离分辨力。 [0054] The echo signal is to improve the signal to noise ratio, is more conducive to extract vital signal from the signal at the same time greatly reducing the amount of computation, but also ensure the proper range resolution.

[0055] 3.3滤波器的选择 Select [0055] 3.3 filter

[0056] 3.3.1硬件滤波器 [0056] 3.3.1 hardware filters

[0057] 在本实施例中将硬件滤波电路接入高速AD采集卡之前,滤波器带宽为可调,在前期预实验中,先后试验了带宽为0.08-10Hz、0.08-100Hz、0.08-1000Hz、 0.08-2000Hz、0.08-3000Hz、0.08_4000Hz、0.08_5000Hz 的几种滤波器,通过效果对比,最终选定了0.08-5000HZ作为硬件滤波电路的通带,增益分为两档:增益为1时,放大倍数为1倍,增益为2时,放大倍数为2倍。 [0057] Example hardware filter circuit in the present embodiment before accessing high-speed AD acquisition card, adjustable filter bandwidth, early preliminary experiment, the test has a bandwidth 0.08-10Hz, 0.08-100Hz, 0.08-1000Hz, 0.08-2000Hz, 0.08-3000Hz, 0.08_4000Hz, 0.08_5000Hz several filter, by effect of contrast, the final selection of a hardware 0.08-5000HZ passband filter circuit, the gain is divided into two steps: 1 gain amplifying 1 multiple times, a gain of 2, 2-fold magnification.

[0058] 分别采用不加硬件滤波电路的单通道UWB系统、加硬件滤波电路(增益为1)的单通道UWB系统和加硬件滤波电路(增益为2)的单通道UWB系统随机采集数据各16 组(无目标、单目标数据),合计共48组数据。 [0058] UWB systems are single channel without hardware filter circuit, the filter circuit stiffener (gain 1) single channel UWB systems and processing hardware filter circuit (a gain of 2) were collected from single-channel data for each UWB systems 16 (no target, single-target data), a total sum of 48 sets of data. 对这48组数据分别采用计算单元包含的算法进行处理和判别,统计识别正确率,统计结果如下表1所示。 This set of data 48 respectively included in the arithmetic computing unit for processing and determination, the correct recognition rate statistics, statistical results are shown in Table 1.

[0059] 表1增减硬件滤波器时的识别正确率情况(48组数据) Hardware identification when increasing or decreasing the filter [0059] Table 1 where the correct ratio (48 sets of data)

[0060] [0060]

Figure CN102008291AD00071

[0061] 在使用增益为1的硬件滤波电路的识别正确率最高,为62%。 [0061] correctly identified using the highest gain of the filter circuit of the hardware, and 62%.

[0062] 通过比较发现,采用增益为1、通带为0.08-5000HZ硬件滤波电路的UWB系统的探测效果最好。 [0062] By comparison, using a gain of one to detect the effect of the pass band UWB system hardware filtering circuit 0.08-5000HZ best.

[0063] 3.3.2数字滤波器 [0063] 3.3.2 digital filter

[0064] 由于微弱生命体征信号中相位信息对静目标检测非常重要,且静目标识别和一维区分技术对算法稳定性和后续数字信号处理的要求较高,所以在本实施例中采用有限脉冲响应(FIR)滤波器来去除高频干扰,提取出呼吸等有用信号。 [0064] Since the phase information signal is weak vital signs is important for the static target detection, target recognition, and high static and a distinction between the technical requirements of dimensional stability and subsequent algorithm for digital signal processing, so the use of finite impulse in the present embodiment, response (FIR) filter to remove high frequency interference, the useful signal is extracted respiration. FIR滤波器的系统函数为: FIR filter as a function of the system:

[0065] [0065]

Figure CN102008291AD00072

[0066] 差分方程为: [0066] The difference equation:

[0067] [0067]

Figure CN102008291AD00073

[0068] 滤波器阶数的选择直接关系到了其幅频特性,阶数越高,幅频特性越好,滤波效果越佳。 Select [0068] The filter order is directly related to its amplitude-frequency characteristic, the higher the order, the better the amplitude-frequency characteristic, the better filtering effect. 但无限制地增加阶数也带来了一些负面影响,如增加了系统运算量,延长了滤波输出的延迟时间等。 However, the order of increasing unrestricted brought some negative effects, such as increasing the amount of operation of the system, extending the filtered output delay time and the like. 综合以上两方面考虑,在系统运算能力允许的情况下,我们选用160阶FIR滤波器进行了实验。 Based on the above two considerations, in the case of computing capability allows the system, we use a 160-order FIR filter experiment.

[0069] 在滤波器的设计上采用窗函数法,通过对比几种窗函数低通滤波器的幅频特性,最终采用了海明窗。 [0069] The filter employed in the design of the window function method, pass filter amplitude-frequency characteristic window function by comparing several low, eventually using Hamming window.

[0070] 正常状态下人的呼吸率为每分钟15〜20次,考虑非正常状态其频率一般也不会超过0.4Hz。 [0070] Human normal breathing rate of 15 ~ 20 times per minute, which is considered non-normal state generally does not exceed the frequency 0.4Hz. 所以我们采用的数字滤波器,其截止频率也主要参考0.4Hz这一指标,即以截止频率不低于0.4Hz的低通滤波器对目标回波信号进行滤波,以比较各滤波器的性能。 Therefore, we use the digital filter, whose cutoff frequency is 0.4Hz the main reference indicator, that is not lower than the cutoff frequency of the low pass filter 0.4Hz target echo signal is filtered in order to compare the performance of each filter. 在本文中我们对截止频率分别为0.4Hz、0.5Hz、0.6Hz、0.7Hz、0.8Hz的低通数字滤波器进行了实验。 In this paper we have cutoff frequencies of 0.4Hz, 0.5Hz, 0.6Hz, 0.7Hz, 0.8Hz low pass digital filter experimentally.

[0071] 随机选取探测范围(即雷达时窗)为20纳秒(3m)和60纳秒(9m)的数据各48 组,共计96组,这些信号均为单通道UWB系统的采样后信号(含无目标、单目标的数据),对这96组数据采用计算单元包含的算法进行判别。 [0071] randomly selected detection range (i.e., when the radar window) of 20 nanoseconds (3m) and 60 ns (9m) of the data of each group 48, a total of 96 groups, these signals are sampled signals are single UWB system ( containing no target, data of a single object), this set of data 96 using an algorithm contained in the determination computing unit. 在对比实验中,只改变滤波器截止频率,其他各软硬件参数不变,统计判别结果的正确率,其判别正确率如表2所示。 In comparative experiments, the only changes the filter cutoff frequency, various other hardware and software parameters constant, the statistical accuracy of the determination result, which determines the correct rate as shown in Table 2.

[0072] 表2改变滤波器截止频率对判别正确率的影响 [0072] Table 2 Effect of changing filter cutoff frequency to determine the correct rate

[0073] [0073]

Figure CN102008291AD00081

[0074] 根据以上实验结果,通过综合比较,最终选取160阶、截止频率方0.5Hz的海明窗FIR数字滤波器来对目标回波信号滤除高频干扰,保留呼吸等生命特征信号。 [0074] From the above experimental results, through the comprehensive comparison, the final selection step 160, the cutoff frequency of the square 0.5Hz Hamming window FIR digital filter to filter out high frequency interference on the target echo signals, and other vital breathing retention signal.

[0075] 3.4微分器的选择 [0075] 3.4 differentiator choice

[0076] 由于直流分量和基线漂移现象的存在,目标回波信号中往往包含能量很大的极低频成分,使得信号严重偏离基线,对微弱的生命信号识别产生很大的影响。 [0076] Due to a DC component and a baseline drift, the target echo signals often contain significant energy extremely low frequency component, so that serious deviation from the baseline signal, a huge impact on the life of the weak signal identification. 本实施例提出采用数字微分的方法来在时间上滤除直流分量和极低频干扰,使有用信号围绕零基线上下波动,以达到增强呼吸等生命特征信号的目的。 The present embodiment proposes a method of digital differential DC component and to filter out very low frequency interference on time, that the useful signal fluctuations around zero baseline, for the purpose of enhancing the vital respiration signal. 微分器的计算过程如公式(4)所示: Differentiator calculation process shown in Equation (4):

Figure CN102008291AD00082

[0078] 公式中:y为输出信号,χ为输入信号,m为阶数,η为点的序号。 [0078] equation: y is the output signal, χ input signal, m is the order number, η is the number of points.

[0079] 随机选取探测范围(雷达时窗)为20纳秒(3m)和60纳秒(9m)的数据各48组, 共计96组,这些信号均为单通道UWB系统的采样后信号(含无目标、单目标的数据), 对这些数据分别采用20阶、40阶、60阶、80阶、100阶、120阶、140阶、160阶、180 阶的数字微分器进行微分处理后进行识别和距离计算,其判别正确率如表3所示: [0079] randomly selected detection range (radar window) of 20 nanoseconds (3m) and 60 (9m) ns data of each group 48, a total of 96 groups, these signals are single UWB signal after sampling system (including identification no target, data of a single object), these data were used 20 bands, 40 bands, 60 bands, 80 bands, 100 step 120 step 140 step 160 step 180 step digital differentiator differentiates treatment and distance calculation that determines the correct ratio as shown in table 3:

[0080] 表3改变微分器阶数对判别正确率的影响 [0080] Table 3 the effect of changing the order of the differentiator to determine the correct rate

[0081] [0081]

Figure CN102008291AD00091

[0082] [0082]

[0083] 可以看出,60阶数字微分器处理后的信号的判别正确率最高,其96组数据的总正确率为60.78%。 [0083] As can be seen, determines the correct rate of a signal after the 60 highest order digital differentiator process, which total 96 sets of data accuracy was 60.78%. 通过比较,本实施例选用了60阶数字微分器来去除直流分量和极低频干扰,增强信号的生命特征。 By comparison, the present embodiment selected order digital differentiator 60 to remove the DC component and the very low frequency interference, enhance the life characteristics of the signal.

[0084] 从图6的比较可以看出,经过对目标回波信号在时间上进行60阶微分后,信号回到了基线附近并紧紧围绕基线上下波动,直流分量和极低频成分得到了抑制,有用信号得到了增强。 [0084] As can be seen from a comparison of FIG. 6, after the target echo signal 60 in the time-order differentiation, and the signal returned to baseline near focus baseline fluctuations, the DC component, and very low frequency component is suppressed, the useful signal has been enhanced.

[0085] 实施例4 [0085] Example 4

[0086] 空间频率分析法进行一维距离区分: Distinguished from the one-dimensional [0086] spatial frequency analysis:

[0087] 完成静目标微弱信号增强后,就要对人体目标在距离上进行区分。 After [0087] completion of the static target weak signal enhancement, it is necessary to distinguish human target distance. 因为距离信号为反映目标距离信息的超低频信号,所以针对距离信号含有空间信息和非平稳性等特点,本实施例构建了空间、频率的联合分布函数,采用空间频率联合分析方法(改形的时频分析)对距离信号进行分析,描述信号在不同距离、频率上的能量密度和强度,从而给出各个人体目标的距离信息。 Because the distance signal to reflect certain ultra-low frequency signal from the information, the containing for a distance signal space information and the non-stationary characteristics of the present embodiment constructed in a joint distribution function space, frequency, spatial frequency conjoint analysis (reshaping frequency analysis) of the signal is analyzed from describe different distances, the energy density and strength of the frequency signal, to give the respective information from the target body.

[0088] 时频分析表示的是信号频谱在时间轴上的变化情况,当将时间变量变成距离变量以后,时频分析结果表示的就是频谱在空间上的变化情况,所以利用时频分析的这一特点来对不同距离上的目标进行频谱分析,进而获得人体判别结果和目标一维距离信息,这样就形成了空间、频率联合分析这一时频分析新的应用形式,它的实质仍然是时频分析。 [0088] indicates the time-frequency analysis of changes in the signal spectrum on the time axis, the time variable becomes When distance variable, time-frequency analysis on the change in the spectrum space is indicated by the results, the time-frequency analysis this feature is to target at different distances spectrum analysis, and then get the result of human judgment and a goals-dimensional distance information, thus forming a space, frequency analysis of this joint time-frequency analysis of the new application form, its essence is still time frequency analysis.

[0089] 在本实施例中,将时频分析中的时间变量变成空间(距离)变量,构建空间、频率联合函数,使其能够同时利用空间、频率信息描述输入信号的能量密度,使这种方法具备了空间、频率的“定位”功能,从而为我们提供了一个很好的非平稳信号在某一距离范围内频率估计的方法。 [0089] In the present embodiment, the time-frequency analysis of time variable into the space (distance) variables, to build the spatial frequency of joint function, it is possible to simultaneously use of space, frequency information describing the energy density of the input signal, so that the method comprising the spatial frequency of "positioning" feature, which provides a good method of non-stationary signals in the frequency estimation within a certain distance range for us.

[0090] 时频变换包括单线性变换如短时傅立叶变换,双线性变换如维纳_维尔分布, 小波变换等。 Frequency conversion comprises a single linear transform such as short time Fourier transform, such as a bilinear transformation Wiener _ Virgin distribution, wavelet transform, etc. [0090] time. 本实施例中的空间频率变换是将时频分析中的时间变量替换成空间(距离) 变量而来的,其距离分辨力为事先确定,并不需要通过变化窗宽来改变;而实验目标为静止人体,其呼吸信号为在长时间内较为稳定,属于局部平稳而长度大的非平稳信号, 对于这类信号适合用短时傅立叶变换来进行分析,所以在本实施例中选取了单线性变换的短时傅立叶变换来进行空间频率分析,并对处理结果进行了分析和比较。 Spatial frequency conversion embodiment of the present embodiment is replaced by a space (distance) Analysis of time variables from variable frequency, which is the range resolution determined in advance, does not need to be changed by changing the window width; while the experimental target a stationary body, which is a more stable respiratory signal over a long time, a large part of the length of local smooth and non-stationary signals, such signals to be analyzed for a short time Fourier transform, the selected single linear transformation in the present embodiment, the short time Fourier transform to analyze the spatial frequency, and the processing results are analyzed and compared.

[0091] 在本实施例中将时窗为60ns(对应9m探测范围,起始距离为Ins)的微分后距离信号在距离上均勻分成26段,对应的距离分辨力约为0.36m。 After the time window in the differential Example [0091] embodiment is 60ns (9m detection range corresponding to the starting distance Ins) present on the distance from the signal 26 is divided into uniform sections, the corresponding range resolution is approximately 0.36m. 因为天线为收发一体天线,收发天线距离较近,受天线直达波的影响,距离信号靠近接收天线部分有一段信号幅值较大,影响判断,所以将此距离信号前4段截去,不予考虑,然后将第5至第26段的每一段上的20点(第26段为11点)的幅值作段内相加,得到的和作为本段的值,从而形成只有22个数值组成的新的距离信号,这22个数值对应的是从0.36X4 = 1.44m开始至9m结束,均勻分布的各距离上的点的目标回波信号。 Since the antenna is an antenna transceiver, transmitting and receiving antennas are close, the antenna is affected by the direct wave from the signal close to the receiving antenna section some signal amplitude, affecting the determination, so the first four paragraphs of this truncated from the signal, not consider, then 20 points on each segment 26 to the fifth paragraph as the magnitude of paragraph (para. 26 to 11) are added, and as the value of this section, so as to form only 22 values ​​obtained composition a new distance signal 22 which corresponds to the target value of the echo signal points in each of the distance from the beginning to the 0.36X4 = 1.44m 9m end, uniformly distributed.

[0092] 分段后,根据实际探测所需要的定位结果刷新率,每隔10秒钟取出该时间段内的所有新距离信号,共计64X10 = 640个新距离信号;将每个距离信号拆分成点(22 点),再将各点按时间先后顺序将序列重组,形成含时间信息的新目标回波信号;将新的各目标回波信号按距离天线由近及远的顺序首尾相连,构成空间频率分析的输入信号。 [0092] After segmentation, the positioning result according to the refresh rate actually detected need, every 10 seconds the signal is taken from all the new time period, a total of 64X10 = 640 new distance signal; signal splitting each distance to a point (22 points), then the respective points in chronological sequence recombination, forming a new target echo signal containing time information; each new target echo signal from an antenna connected by a sequence of near and far end to end, constituting the spatial frequency of the input signal analysis.

[0093] 对合成的输入信号作空间频率分析,即作短时傅立叶变换,其中窗宽对应于新目标回波信号的长度,定为64X10 = 640,窗每次滑动距离对应于距离信号的距离分辨力,根据变换点数不小于窗宽的原则和判别结果对频率分辨率的需求选择傅立叶变换点数为1024点。 [0093] The synthesized spatial frequency of the input signal for analysis, i.e., as short time Fourier transform, wherein the window width corresponds to the length of the new target echo signal set to 64X10 = 640, each sliding window a distance corresponding to the distance from the signal resolution, according to the transformation point of not less than the window width selection principles and the discrimination result of the demand frequency resolution Fourier transform points to 1024 points. 确定以上参数以后,对输入信号进行短时傅立叶变换,并绘出结果图。 After determining the above parameters, the input signal STFT, and the results plotted in FIG. 短时傅立叶变换公式如式(6)所示: The short time Fourier transform equation of formula (6):

[0094] STFT (t, w) = / S(x) γ (τ-OeiwMx ......(6) [0094] STFT (t, w) = / S (x) γ (τ-OeiwMx ...... (6)

[0095] 其中S( τ )为输入信号,Y (t)为窗函数。 [0095] where S (τ) is the input signal, Y (t) is the window function.

[0096] 在窗函数长短的选择上,为了提高短时傅立叶变换的时间分辨率,常常要求选择的窗函数时间宽度尽可能短。 [0096] In the selection of the length of the window function, in order to improve the time resolution of the short time Fourier transform is often required to select the window function time width as short as possible. 另一方面,短时傅立叶变换要想得到高的频率分辨率, 则要求选择的窗函数时间宽度尽可能长,因此时间分辨率的提高与频率分辨率的提高相矛盾。 On the other hand, short-time Fourier transform in order to obtain a high frequency resolution is required to select the window function time width as long as possible, thus improving the temporal resolution and frequency resolution increase contradict. 实际中,选择的窗函数γ ω的宽度应该与信号的局域平稳长度相适应。 In practice, the width of the window function γ ω selected should be compatible with the length of the stationary local signal. 在本实验中,探测对象人体的正常呼吸频率为每分钟15-20次,即3-4秒完成一次呼吸运动,为了减小人体呼吸偶然因素、个体差异的影响以及保证频率分辨率,我们选取的窗函数的时间宽度为10秒,对应到空间频率分析中的窗宽为640。 In this experiment, normal human body detection target respiratory rate is 15-20 times per minute, i.e. 3-4 seconds to complete a breathing exercise, in order to reduce human respiratory causal factors of individual differences and to ensure the frequency resolution, we select window function time width of 10 seconds, corresponding to a spatial frequency analysis window width is 640.

[0097] 实施例5 [0097] Example 5

[0098] 波峰判别方法及阈值的设定: [0098] The method of determining the peak threshold value and setting:

[0099] 空间、频率分析的结果是一个3维(空间、频率、能量)对应关系,两条坐标轴分别为距离和频率,而能量强度是由颜色的深浅来对应的。 [0099] Results spatial, frequency analysis is a 3-dimensional (spatial, frequency, energy) corresponding relationship, two axes distance and frequency, respectively, and the energy intensity is corresponding to the color depth. 通过适当的方式并设置合适的人体生命特征判定阈值,即可以实现单通道对多个静目标的距离区分及距离计算。 By appropriate means and set the appropriate human life characteristic determination threshold value, i.e., a single channel can be realized to calculate the distance to distinguish between a plurality of stationary targets and distance. 如果在某一距离上,信号能量大,谱峰集中,明显高于相邻距离上的信号能量,且符合判定阈值,则认为在该接收天线探测范围内的相应距离上有静止人体目标(一维距离确定);如果多个距离上有大能量的信号出现,且符合阈值,则认为在多个距离上有静止人体目标存在,通过算法记录这些目标的一维距离值,即为各个目标到天线的距离。 If at a certain distance, the signal energy, peak concentration was significantly higher than the signal energy of the adjacent distance, and in accordance with the determination threshold value, it is still considered a target body (a corresponding distance within the detection range of the receiving antenna dimensional distance determination); if there is a large energy from the plurality of signal occurs, and in accordance with the threshold value, it is certain that there still exist on the human body from a plurality of one-dimensional distance value by the algorithm records these goals, namely to the respective target distance from the antenna.

[0100] 多目标的判别和距离计算具体步骤如下: [0100] Multi-target distance is calculated and the determination steps are as follows:

[0101] 找出22段中能量值最大的12段,并找出这12段中的所有的能量波峰,按能量大小分别记为Epeakl, Epeak2, Epeak3...波峰是这样规定的:即本段的能量值大于相邻两段能量值的,则本段为波峰。 [0101] Find the maximum energy value segment 22 segment 12, and find all the energy peaks in segment 12 which, according to the amount of energy denoted as Epeakl, Epeak2, Epeak3 ... peak is defined: the present energy value is greater than the adjacent section of the two energy values, the peak of this paragraph. 找出能量波峰以后,记录波峰所在段的段序号,用于后面计算目标距离。 After find the energy peaks, where the segment number recorded segments peaks for calculating a target distance behind.

[0102] 计算22段中能量值最小的8段的平均能量值记为:Ε·η,利用波峰能量和最小8段的平均能量值作比较来确定目标的个数。 [0102] calculating the minimum energy value of the average energy of 8 segment 22 referred to as a segment value: Ε · η, peak energy and average energy use of the minimum value comparing section 8 determines the target number. 比较阈值如下: Comparison threshold as follows:

[0103] (1)如果能量波峰1的能量Epeakl大于4倍的最小平均能量值Emean,即Epeakl > 4Emean,则认为波峰1位置有目标存在,目标的距离由波峰1的序号计算得出; [0103] (1) if the minimum average energy of the peak energy in the energy Epeakl 1 is more than 4 times the value Emean, i.e. Epeakl> 4Emean, the peak is considered a target position exists, from the calculated target number is derived from the peak 1;

[0104] (2)如果能量波峰2的能量Epeak2大于3倍的最小平均能量值Emean,即Epeak2 > 3Emean,则认为波峰2位置有目标存在,目标的距离由波峰2的序号计算得出; [0104] (2) If the minimum average energy of the peak energy of the energy Epeak2 2 is greater than 3 times the value Emean, i.e. Epeak2> 3Emean, is considered the peak position of the target 2 is present, the distance of the target derived from the calculated peak number 2;

[0105] (3)如果能量波峰3的能量Epeak2大于2.5倍的最小平均能量值Emean,即Epeak3 > 2.5Emean,则认为波峰3位置有目标存在,目标的距离由波峰3的序号计算得出。 [0105] (3) If the minimum average energy of the peak energy of the energy Epeak2 3 is greater than 2.5 times the value Emean, i.e. Epeak3> 2.5Emean, is considered the peak position of the target 3 is present, the distance of the target derived from the calculated peak number 3.

[0106] 图7是根据波峰判别方法及所定阈值对一个双目标数据进行的判别的结果。 [0106] FIG. 7 is a result of the discrimination method of discriminating peaks according to a predetermined threshold value, and one pair of object data.

[0107] 可以看出,22段信号存在两个波峰,这两个波峰的能量经过与各自的阈值比较,得出两个波峰均为目标,通过计算两个目标的距离为:13X0.36 = 4.68m,21X0.36 =7.56m。 [0107] As can be seen, there are two peaks 22 of the signal, these two energy peaks by comparison with respective threshold values, two peaks are obtained target, two targets by calculating the distance: 13X0.36 = 4.68m, 21X0.36 = 7.56m.

[0108] 至此,我们采用多目标距离区分算法完成了对不同距离上的多个静止目标的识别和各目标距离的计算,即有了多目标的距离信息以后,再对有目标位置上的信号幅值进行归一化处理,形成二维平面在各通道上的投影信号。 [0108] At this point, we have a multi-target distance discrimination algorithm finishes the calculation of a plurality of stationary targets on the recognition of the different distances for each target distance, i.e., with the future multi-target distance information, and then the target position of the signal on the normalized amplitude values, a two-dimensional planar projection signals on each channel.

[0109] 实施例6 [0109] Example 6

[0110] 验证实验: [0110] Verification experiment:

[0111] 因为雷达式生命探测仪在实际应用中常常是隔墙探测的,所以其穿墙探测性能是评价雷达式生命探测仪优劣的一项重要指标。 [0111] Because the radar life detector in practical applications often partition detection, detection through the wall so its performance is an important index of radar life detector merits. 在本部分的实验中,先采用自由空间采集的数据来验证多静目标识别及距离区分算法的可行性,再利用穿墙采集的数据来对算法的性能进行评价,穿墙数据分为无目标、单目标、双目标、三目标四种情况来分别评价。 In the experimental section of this first free space using data collected to verify the feasibility of multi-static target recognition and distinction from the algorithm, then use the data collected through the wall to evaluate the performance of the algorithm, the data is divided into non-target through the wall single goal, two goals, three goals four cases were evaluated.

[0112] 首先,在实验中采用单通道UWB雷达式生命探测仪系统对10名志愿者分别进行了自由空间状态下的探测,并采集、存储了所有的实验数据,这些数据包含探测范围内无目标的数据,单目标的数据和双目标的数据。 [0112] First, in the experiment using single channel UWB radar life detector system 10 to detect the volunteers were free space state, respectively, and collecting, storing all of the experimental data, which contain no detection range data object, the data object and the data of a single dual goals. 从这些数据中对无目标、单目标、双目标数据随机各选了一组,采用本实验所叙述的方法对选取的数据进行了处理和计算, 其结果如图8-图10所示。 From these data for non-target, a single objective, bis target data selected from each of a set of random, using the method described in this experiment were selected for data processing and calculation, the result shown in FIG. 8-10. 其中图8为自由空间无目标情况的数据的时频分析图,判别计算结果为:该探测区域无人体目标存在;图9为自由空间单目标数据的时频分析图, 判别计算结果为:该探测区域内有一个静止人体目标存在,其位置在6.48m处(目标实际位置6.5m);图10为自由空间双目标情况的数据的时频分析结果,判别计算结果为:该探测区域内有两个静止人体目标存在,其位置分别在2.88m和7.56m处(目标实际位置为2.5m 禾口7.5m)。 Time-frequency analysis diagram in which FIG. 8 is a free space without the target condition data determination calculation result is: the detection area without human object exists; FIG. 9 is a frequency analysis view of the free space single target data determination calculation result is: The a stationary object is present within the detection area of ​​the human body, its location in the 6.48m (actual target position 6.5m); frequency analysis results of FIG. 10 is a case where the target data, two free space, it is determined as the calculation result: within the detection region two stationary target human presence, position, and respectively at 2.88m 7.56m (the actual position of the target opening Wo 2.5m 7.5m).

[0113] 从自由空间各种情况的数据的时频分析结果可以看出,多静目标识别及距离区分算法可以在自由空间很好地区分探测范围内是无目标、单目标还是双目标情况,并能较为准确地计算出每个静止目标的距离,从而验证了此方法的可行性。 [0113] in each case the free space data from the frequency analysis results can be seen, the static multi-target recognition and discrimination algorithms distance in free space is well within the partial detection range untargeted area, single or dual target certain circumstances, and can accurately calculate the distance of each still object, which proves the feasibility of the method.

[0114] 在进行了算法的可行性验证以后,又在实验中采用单通道UWB雷达式生命探测仪系统对10名志愿者分别进行了穿墙(30cm厚的砖墙)状态下的探测,并采集、存储了所有的实验数据,这些数据包含探测范围内无目标的数据,单目标的数据,双目标的数据以及三目标的数据,采用本实验所叙述的方法对所有的数据进行了处理和计算,随机选择其中两组数据,其时频分析结果如图11-图12所示。 [0114] The feasibility of performing verification algorithm later, and single-channel UWB radar life detector system of experiments were carried out in 10 volunteers through walls (30cm thick brick wall) detecting a state, and collecting, storing all of the experimental data, which contain no target data to the detection range data of a single object, data double object and the data three-goal, the method according to the present experiment described in all the data processed and calculating, randomly selected sets of data, frequency analysis results are shown in FIG 11- FIG 12 the time. 其中图11为穿30cm砖墙单目标数据的时频分析图,判别计算结果为:该探测区域内有一个静止人体目标存在,其位置在6.48m处(目标实际位置6.0m);图12为穿30cm砖墙双目标数据的时频分析结果, 判别计算结果为:该探测区域内有两个静止人体目标存在,其位置分别在3.24m和6.48m 处(目标实际位置为3.0m禾Π 6.0m)。 Wherein the frequency analysis of FIG. 11 through FIG 30cm brick is a single target data determination result is calculated as: presence of a stationary body within the detection target area, which is at a position 6.48m (6.0m actual target position); FIG. 12 is a frequency analysis target data through 30cm double brick result, the calculation result is determined as follows: there are two stationary target exists within the human detection area, the position at 6.48m and 3.24m respectively (the actual position of the target is 3.0m Wo Π 6.0 m).

[0115] 规定目标判别结果分为正确判别、漏判、误判、错判。 [0115] provisions goal discrimination results into proper identification, Missing, miscarriage of justice, wrongful convictions. 目标的个数、距离均判别无误的为正确判别,有目标而判别为无目标为漏判,无目标而判别为有目标为误判, 目标位置判别错误为错判。 The number of goals from both determine and correct the correct judgment, determines that there is no objective and goal is Missing, and determines that there is no target goal for the miscarriage of justice, the target location identification error for the miscarriage of justice. 考虑到距离分辨力对误差的影响和人体呼吸时胸腔位置的移动情况,规定判别位置和目标实际位置的误差小于0.5m均为目标距离判别正确。 Considering the impact movement when the distance resolution of the error position and the chest during breathing human body, predetermined error determined actual position and the target position is less than 0.5m from the target are determined correctly. 按照以上的分类方式对所采数据的处理判别结果进行统计。 To deal with the discrimination result of the data collected for statistical accordance with the above classification.

[0116] 其中,穿墙的无目标数据共采集了23组。 [0116] wherein, the target data without through the wall of the 23 groups were collected. 按照上述分类方式对所采数据的处理结果进行统计,因为所有的数据均为无目标数据,所以不存在漏判和错判的情况。 The results of processing the collected data in accordance with the above classification statistics, because all the data are no objective data, so the situation and Missing miscarriage of justice does not exist. 统计结果如表4所示。 Statistics The results shown in Table 4.

[0117] 表4穿30cm砖墙无目标数据的识别正确率情况 [0117] Table 4 through a brick wall 30cm without certain cases to identify the correct data rate

Figure CN102008291AD00131

[0119] 穿墙的单目标数据共采集27组(目标位置随机分布在0_9m的探测范围内)。 [0119] through the wall 27 a single target group were collected data (target position randomly distributed within the detection range 0_9m). 同样按照上述分类方式对所采数据的处理结果进行统计,统计结果如表5所示。 Similarly to the processing result collected statistical data, the statistical results shown in Table 5 in accordance with the above classification.

[0120] 表5穿30cm砖墙时单目标数据的识别正确率情况 Identifying single target data wear 30cm brick [0120] Table 5 where accuracy

Figure CN102008291AD00132

[0123] 穿墙数据中的双目标数据共采集了59组(目标位置随机分布在0_9m的探测范围内,两个目标之间的距离间隔从Im到6m不等,且两个目标在正对探测天线的方向上错开一个肩宽的距离)。 [0123] bis entrapment target data were collected data set 59 (target position randomly distributed within the detection range 0_9m ranging from Im to 6m distance between two target interval and facing the two objectives a shoulder displaced from the direction detection antenna). 同样按照上述分类方式对所采数据的处理结果进行统计,统计结果如表6所示。 Similarly to the processing result collected statistical data is shown, statistics are shown in Table 6 in the above classification.

[0124] 表6穿30cm砖墙双目标数据的识别正确率情况 [0124] Table 6 through recognition rate 30cm double brick case where the target data

Figure CN102008291AD00133

[0126] 穿墙数据中的三目标数据共采集了24组(目标位置随机分布在0_9m的探测范围内,任意两个目标之间的距离间隔从Im到6m不等,且任意两个目标在正对探测天线的方向上均错开一个肩宽的距离)。 [0126] the target data through the wall three data sets were collected 24 (target position randomly distributed within the detection range 0_9m ranging from Im to 6m distance between any two of the target interval, and any two objectives the positive direction of the probe antenna are shifted by a distance of the shoulder). 同样按照上述分类方式对所采数据的处理结果进行统计,因为所有的数据均为三目标数据,在算法中也仅考虑了对最多三个目标的识别和区分,所以不存在误判的情况。 Similarly to the processing result collected statistical data in accordance with the above classification, because all the data are three object data, the algorithm also takes into account only up to identify and distinguish the three objectives, the absence of false positives. 统计结果如表7所示。 Table 7 shows the statistical results.

[0127] 表7穿30cm砖墙三目标数据的识别正确率情况 [0127] Table 7 through the recognition rate of object data 30cm brick three cases

Figure CN102008291AD00134

[0129] 由以上对不同目标数的各种数据进行处理后的结果统计来看,多静目标识别及距离区分算法对23组无目标数据判别正确率为91%,对27组单目标数据判别正确率为56%,对59组双目标数据判别正确率为73%,对24组三目标数据判别正确率为46%。 [0129] The results after various data processed by the above target number of different statistical point of view, from the plurality of static target recognition and discrimination algorithms of the Group 23 no data to determine the correct target was 91%, the data set is determined for a single target 27 correct 56% of the target data 59 to determine the correct set of two 73%, three sets of 24 data to determine the correct target was 46%. 该方法对无目标数据的识别正确率最高,对三目标的识别正确率最低。 The method identifies the correct target data for non-highest, lowest recognition rate for three-goal.

[0130] 应当理解的是,对本领域普通技术人员来说,可以根据上述说明加以改进或变 [0130] It should be understood that those of ordinary skill in the art, can be modified or changed in accordance with the above description

换,而所有这些改进和变换都应属于本发明所附权利要求的保护范围。 In other words, and all such modifications and variations shall fall within the scope of the appended claims of the invention.

Claims (6)

1. 一种可用于多目标探测的单通道基于UWB的雷达式生命探测仪,其特征在于,包括UWB生物雷达前端和计算单元,所述UWB生物雷达前端包括发射天线、接收天线、 脉冲振荡器、电磁脉冲产生器、取样积分器;脉冲振荡器产生脉冲信号,该信号触发电磁脉冲产生器产生窄脉冲,并通过发射天线辐射出去;反射信号经过接收天线送到取样积分器,由脉冲振荡器产生的脉冲信号同时经过延时电路和距离门产生器产生距离门, 对接收信号进行选择,信号通过取样积分电路,经过积累后微弱信号被检测出来,并经由放大器和滤波器进行放大、滤波,再经高速A/D采集卡采样后送入计算单元,由计算单元对采集到的信号进行分析处理,最终提取多个人体目标生命信息和各目标距离。 A single channel may be used for multiple target detection based UWB radar life detector, wherein the biological UWB radar front-end and comprising a computing unit, said UWB radar front end includes biological transmitting antenna, receiving antenna, the pulse oscillator electromagnetic pulse generator, a sampling integrator; pulse oscillator generates a pulse signal which triggers electromagnetic pulse generator generates a narrow pulse, and radiated through the transmitting antenna; reflected signal via an antenna to a sampled integrator receiving, by the pulse oscillator pulse signal generated while passing through the delay circuit and generating a range gate range gate generator, the received signal is selected, the sampling signal integration circuit, after the accumulation of a weak signal is detected, and amplifies, filters and amplifiers via a filter, and then the calculation means into the high-speed a / D acquisition card sampling, analysis and processing of the collected signal by the calculation unit, the final extract a plurality of targets of human life, and each object distance information.
2.根据权利要求1所述的单通道UWB雷达式生命探测仪,其特征在于,所述滤波器采用增益为1、通带为0.08-5000HZ硬件滤波电路。 The single-channel UWB radar life detector according to claim 1, wherein said filter uses a gain of 1, band pass filter circuit 0.08-5000HZ hardware.
3.根据权利要求1所述的单通道UWB雷达式生命探测仪,其特征在于,所述计算单元包括信号积分模块、信号分解重构模块、数字滤波模块和数字微分模块、空间频率分析模块,所述信号积分模块在距离上对信号进行积分,所述信号分解重构模块将信号打散进行分解、重构,合成目标回波信号和距离信号,所述数字滤波和数字微分模块对目标回波信号进行数字滤波和数字微分,所述空间频率分析模块用于根据数字滤波和数字微分后的目标回波信号以及距离信号进行空间频率分析,得到目标一维距离。 The single-channel UWB radar life detector according to claim 1, wherein said calculating means includes a signal integrator module, the signal decomposition and reconstruction module, digital filtering and digital modules differential module, spatial frequency analyzing module, the integrator module signal on the signal from the integrator, the signal decomposition and reconstruction module will break the signal decomposition, reconstruction, synthesis target echo signal and a distance signal, said digital filtering and digital differential module back to the target wave signal and digital differential digital filtering, the spatial frequency analysis means for performing spatial frequency analysis according to the target signal and the echo signal from the digital filtering and digital differential, to give a certain dimension range.
4.根据权利要求3所述的单通道UWB雷达式生命探测仪,其特征在于,所述数字滤波模块采用160阶、截止频率为0.5Hz的海明窗FIR数字滤波器。 The single-channel UWB radar life detector according to claim 3, wherein the digital filtering modules 160 bands, the cutoff frequency of the digital FIR Hamming filter is 0.5Hz.
5.根据权利要求3所述的单通道UWB雷达式生命探测仪,其特征在于,所述数字微分模块采用60阶数字微分器。 According to claim UWB radar life single channel detector of claim 3, wherein the digital differentiator block 60 using a digital differentiator stage.
6.根据权利要求3所述的单通道UWB雷达式生命探测仪,其特征在于,还包括波峰判别模块,用于根据空间频率分析结果和系统预设置阈值判断波峰位置是否有人体目标存在。 According to claim UWB radar life single channel detector of claim 3, wherein the determining module further comprises a peak, and a frequency analysis system according to the space pre-set threshold value determines whether a peak position of the human object exists.
CN2010105023196A 2010-10-11 2010-10-11 Single-channel UWB-based radar type life detection instrument for multi-target detection CN102008291B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2010105023196A CN102008291B (en) 2010-10-11 2010-10-11 Single-channel UWB-based radar type life detection instrument for multi-target detection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2010105023196A CN102008291B (en) 2010-10-11 2010-10-11 Single-channel UWB-based radar type life detection instrument for multi-target detection

Publications (2)

Publication Number Publication Date
CN102008291A true CN102008291A (en) 2011-04-13
CN102008291B CN102008291B (en) 2012-11-14

Family

ID=43838784

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2010105023196A CN102008291B (en) 2010-10-11 2010-10-11 Single-channel UWB-based radar type life detection instrument for multi-target detection

Country Status (1)

Country Link
CN (1) CN102008291B (en)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012055148A1 (en) * 2010-10-27 2012-05-03 中国人民解放军第四军医大学 Multichannel ultra wide band based (uwb-based) radar life detector and positioning method thereof
CN102488506A (en) * 2011-10-27 2012-06-13 中国人民解放军第四军医大学 Excitation source circuit module for micro radar
CN102512170A (en) * 2011-10-27 2012-06-27 中国人民解放军第四军医大学 Carrying type small radar life detection instrument
CN103027670A (en) * 2012-12-13 2013-04-10 中国人民解放军第四军医大学 Micropower impact-type biological radar front end
CN103082995A (en) * 2012-11-29 2013-05-08 中国人民解放军第四军医大学 Tension pneumothorax detection system based on ultra wide spectrum biological radar
CN103116159A (en) * 2013-01-18 2013-05-22 湖南华诺星空电子技术有限公司 Multi-mode self-positioning networking radar life detection method and device
CN103308899A (en) * 2013-05-23 2013-09-18 中国人民解放军第四军医大学 Biological radar human body target identification method based on zero crossing point technology
CN103454691A (en) * 2013-08-22 2013-12-18 中国人民解放军第四军医大学 Scanning probing method and system based on UWB biological radar
CN103908228A (en) * 2014-02-17 2014-07-09 天津大学 Two-channel tumor ultra-wide band signal extraction method
CN104274184A (en) * 2013-07-01 2015-01-14 西门子公司 Radar system for medical use
CN105387768A (en) * 2015-10-30 2016-03-09 北京艾克利特光电科技有限公司 Electronic sighting device capable of detecting surrounding life entity
CN106019254A (en) * 2016-05-20 2016-10-12 中国人民解放军第四军医大学 Separating and identifying method for multiple human body objects in distance direction of UWB impact biological radar
CN106054156A (en) * 2016-06-22 2016-10-26 中国人民解放军第四军医大学 Static human target recognizing and positioning method based on UWB (Ultra Wideband) MIMO (Multiple-Input Multiple-Output) bio-radar
CN106950544A (en) * 2017-03-06 2017-07-14 哈尔滨工程大学 DSP-based large time-width signal segmented identification implementation method

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103616729B (en) * 2013-11-06 2016-05-18 中国人民解放军第四军医大学 One of a plurality of human target estimation method and system for radar-based biological uwb

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2529267Y (en) * 2002-03-15 2003-01-01 中国人民解放军第四军医大学 Non-contact life parameters radar detection means
US7567200B1 (en) * 2006-04-27 2009-07-28 Josef Osterweil Method and apparatus for body position monitor and fall detect ion using radar
CN201402322Y (en) * 2009-04-30 2010-02-10 薛亚明 Radar life detection instrument

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2529267Y (en) * 2002-03-15 2003-01-01 中国人民解放军第四军医大学 Non-contact life parameters radar detection means
US7567200B1 (en) * 2006-04-27 2009-07-28 Josef Osterweil Method and apparatus for body position monitor and fall detect ion using radar
CN201402322Y (en) * 2009-04-30 2010-02-10 薛亚明 Radar life detection instrument

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 20090215 夏林林 基于RQA方法的生物雷达中人体数量识别技术的研究 第12-13页,第3.1.2节,3.2.2节,3.3.2节 1-4 , *
《信息化研究》 20100831 张杨等 超宽谱雷达人体目标识别的新方法 摘要、第1部分,第2部分、图1-2 1-2 第36卷, 第8期 2 *
《第八届全国信号与信息处理联合学术会议论文集》 20091231 张杨等 生物雷达多静目标检测识别技术初步研究 摘要、第2部分、图1-2 1-6 , *
《第八届全国信号与信息处理联合学术会议论文集》 20091231 张杨等 生物雷达多静目标检测识别技术初步研究 摘要、第2部分、图1-2 1-6 , 2 *

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012055148A1 (en) * 2010-10-27 2012-05-03 中国人民解放军第四军医大学 Multichannel ultra wide band based (uwb-based) radar life detector and positioning method thereof
CN102488506A (en) * 2011-10-27 2012-06-13 中国人民解放军第四军医大学 Excitation source circuit module for micro radar
CN102512170A (en) * 2011-10-27 2012-06-27 中国人民解放军第四军医大学 Carrying type small radar life detection instrument
CN102488506B (en) 2011-10-27 2014-07-16 中国人民解放军第四军医大学 Excitation source circuit module for micro radar
CN103082995A (en) * 2012-11-29 2013-05-08 中国人民解放军第四军医大学 Tension pneumothorax detection system based on ultra wide spectrum biological radar
CN103027670A (en) * 2012-12-13 2013-04-10 中国人民解放军第四军医大学 Micropower impact-type biological radar front end
CN103116159A (en) * 2013-01-18 2013-05-22 湖南华诺星空电子技术有限公司 Multi-mode self-positioning networking radar life detection method and device
CN103308899A (en) * 2013-05-23 2013-09-18 中国人民解放军第四军医大学 Biological radar human body target identification method based on zero crossing point technology
CN104274184A (en) * 2013-07-01 2015-01-14 西门子公司 Radar system for medical use
CN103454691A (en) * 2013-08-22 2013-12-18 中国人民解放军第四军医大学 Scanning probing method and system based on UWB biological radar
CN103454691B (en) * 2013-08-22 2017-03-01 中国人民解放军第四军医大学 A scanning-based method and system for detection of biological radar uwb
CN103908228A (en) * 2014-02-17 2014-07-09 天津大学 Two-channel tumor ultra-wide band signal extraction method
CN103908228B (en) * 2014-02-17 2015-08-05 天津大学 A dual-channel ultra-wideband signal extraction tumor
CN105387768A (en) * 2015-10-30 2016-03-09 北京艾克利特光电科技有限公司 Electronic sighting device capable of detecting surrounding life entity
CN106019254A (en) * 2016-05-20 2016-10-12 中国人民解放军第四军医大学 Separating and identifying method for multiple human body objects in distance direction of UWB impact biological radar
CN106019254B (en) * 2016-05-20 2018-03-20 中国人民解放军第四军医大学 Uwb more than one kind of impact human biological radar target identification method to the separation distance
CN106054156A (en) * 2016-06-22 2016-10-26 中国人民解放军第四军医大学 Static human target recognizing and positioning method based on UWB (Ultra Wideband) MIMO (Multiple-Input Multiple-Output) bio-radar
CN106054156B (en) * 2016-06-22 2018-05-04 中国人民解放军第四军医大学 A static body target identification and location MIMO based UWB radar biological
CN106950544A (en) * 2017-03-06 2017-07-14 哈尔滨工程大学 DSP-based large time-width signal segmented identification implementation method

Also Published As

Publication number Publication date
CN102008291B (en) 2012-11-14

Similar Documents

Publication Publication Date Title
Zidelmal et al. QRS detection based on wavelet coefficients
CN101065058B (en) Monitoring physiological activity using partial state space reconstruction
US7725150B2 (en) System and method for extracting physiological data using ultra-wideband radar and improved signal processing techniques
US20070219453A1 (en) Automated analysis of a cardiac signal based on dynamical characteristics of the cardiac signal
US20030229289A1 (en) Method and system for generating a likelihood of cardiovascular disease, analyzing cardiovascular sound signals remotely from the location of cardiovascular sound signal acquisition, and determining time and phase information from cardiovascular sound signals
Kim et al. Automatic spike detection based on adaptive template matching for extracellular neural recordings
Pal et al. Empirical mode decomposition based ECG enhancement and QRS detection
US5609158A (en) Apparatus and method for predicting cardiac arrhythmia by detection of micropotentials and analysis of all ECG segments and intervals
Sayeed et al. Optimal detection using bilinear time-frequency and time-scale representations
Derode et al. Influence of correlations between scatterers on the attenuation of the coherent wave in a random medium
US7652617B2 (en) Radar microsensor for detection, tracking, and classification
Saniie et al. Quantitative grain size evaluation using ultrasonic backscattered echoes
Moukadem et al. A robust heart sounds segmentation module based on S-transform
Madeiro et al. An innovative approach of QRS segmentation based on first-derivative, Hilbert and Wavelet Transforms
Chandra et al. An approach to remove the clutter and detect the target for ultra-wideband through-wall imaging
Arif et al. Detection and localization of myocardial infarction using k-nearest neighbor classifier
Wang et al. Analysis of low-order autoregressive models for ultrasonic grain signal characterization
de Chazal et al. Automatic classification of ECG beats using waveform shape and heart beat interval features
Martinez et al. Evaluation of a wavelet-based ECG waveform detector on the QT database
JP2010515878A (en) System and method for detecting dangerous objects and object
CN103267932B (en) Gis one kind of partial discharge detection system and method
JP6072005B2 (en) Averaging of the signal
Shahid et al. A new spike detection algorithm for extracellular neural recordings
CN1951320A (en) Abnormal electrocardiogram recognition method based on ultra-complete characteristics
US8112147B2 (en) Method and apparatus for generating determination indexes for identifying ECG interfering signals

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C14 Grant of patent or utility model
EXPY Termination of patent right or utility model