WO2019093132A1 - Signal detecting system - Google Patents

Signal detecting system Download PDF

Info

Publication number
WO2019093132A1
WO2019093132A1 PCT/JP2018/039578 JP2018039578W WO2019093132A1 WO 2019093132 A1 WO2019093132 A1 WO 2019093132A1 JP 2018039578 W JP2018039578 W JP 2018039578W WO 2019093132 A1 WO2019093132 A1 WO 2019093132A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
noise
observation
stationary
stationary noise
Prior art date
Application number
PCT/JP2018/039578
Other languages
French (fr)
Japanese (ja)
Inventor
浩樹 松本
誠巳 小河
隆 石黒
啓一 小林
Original Assignee
太陽誘電株式会社
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 太陽誘電株式会社 filed Critical 太陽誘電株式会社
Publication of WO2019093132A1 publication Critical patent/WO2019093132A1/en

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
    • A61B5/0245—Detecting, measuring or recording pulse rate or heart rate by using sensing means generating electric signals, i.e. ECG signals
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60N—SEATS SPECIALLY ADAPTED FOR VEHICLES; VEHICLE PASSENGER ACCOMMODATION NOT OTHERWISE PROVIDED FOR
    • B60N2/00—Seats specially adapted for vehicles; Arrangement or mounting of seats in vehicles
    • B60N2/90—Details or parts not otherwise provided for
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
    • B60R16/00—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for
    • B60R16/02—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers

Definitions

  • the present invention relates to a signal detection system for separating a detection target signal desired to be detected from an observation signal obtained in an environment where various background noises exist, and in particular, it is one of biological information detected using a piezoelectric element.
  • the present invention relates to a signal detection system suitable for detecting a certain pulse wave signal.
  • Patent Document 2 shows a “vibration waveform sensor and waveform analysis device” which is used by being wound around a fingertip and capturing the vibration from the arterial blood vessel wall of the fingertip as a pulse wave.
  • the vibration waveform sensor using the above-mentioned piezoelectric body does not have to be directly attached to the human body. For example, if it is incorporated into a seat of an automobile and the pulse wave of the thigh is measured, the presence or absence of seating of the driver and health condition are detected. It can be extremely beneficial. However, when a piezoelectric sensor is incorporated into a car seat, various noises such as air conditioning noise, engine noise and road noise (running noise) are removed or separated to detect pulse wave signals well. There is a need.
  • BSS blind signal separation
  • ICA Independent Component Analysis
  • signal separation is achieved by adjusting parameters so as to establish statistical independence of observed signals, assuming that independent current signals are temporally and spatially mixed. ing.
  • Kullback Leibler Divergence is widely used.
  • ANC Active Noise Control
  • ANC uses a signal obtained from this sensor using a sensor that acquires only the signal of noise sources other than the desired signal, and generates a signal with the same amplitude and opposite phase as the noise to the point where noise is desired to be suppressed. It is a method to remove the noise at a specific point by making it interfere.
  • the present invention focuses on such a point, and an object thereof is to provide a signal detection system capable of appropriately detecting a detection target signal even when strong noise is mixed in an observation signal. Another object is to perform detection of pulse wave of a driver of a car well and to help prevent accidents.
  • the present invention is obtained by a plurality of observation signal acquisition means for acquiring observation signals including stationary noise and non-stationary noise in the detection target signal, noise acquisition means for acquiring the stationary noise, and the observation signal acquisition means And noise removing means for removing stationary noise obtained by the noise obtaining means from the observed signal.
  • Another invention is a plurality of observation signal acquisition means for acquiring observation signals in which stationary signals and non-stationary noise are included in a detection target signal, noise acquisition means for acquiring the stationary noise, and the observation signal acquisition means Noise removal means for removing stationary noise obtained by the noise acquisition means from the obtained observation signal; and signal separation means for separating the detection target signal from the signal after noise removal by the noise removal means. It is characterized by
  • a plurality of observation signal acquisition means for acquiring observation signals including stationary noise and non-stationary noise in a detection target signal, and the non-stationary noise from observation signals obtained by the observation signal acquisition means It comprises: signal separation means for separation; noise acquisition means for acquiring the stationary noise; and noise removal means for removing stationary noise obtained by the noise acquisition means from the signal obtained by separation by the signal separation means. It is characterized by Alternatively, a plurality of observation signal acquisition means for acquiring an observation signal in which non-stationary noise is included in the detection target signal, and signal separation means for separating the non-stationary noise from the observation signal obtained by the observation signal acquisition means It is characterized by having.
  • the detection target signal is a pulse wave of a person including a driver driving a vehicle.
  • the noise removal means removes noise by an ANC method, and the signal separation means separates signals by a BSS method.
  • a plurality of observation signals in which stationary noise and non-stationary noise are included in the detection target signal are acquired, stationary noise is separately acquired, and the stationary noise is removed from the observation signal to detect the detection target Since the signals are separated, even when strong noise is mixed in the observation signal, the detection target signal can be properly separated, and the pulse wave of the driver of the vehicle can be favorably detected.
  • FIG. 1 shows the basic configuration of the signal detection system in the present embodiment.
  • source signals G1, G2 and G3 output from the signal source are respectively input to the sensors S1 and S2 through various transmission paths.
  • the transfer characteristics of the transfer path in which the source signals G1, G2, G3 at this time reach the sensors S1, S2 are indicated by weighting elements H11 to H22, Z11, Z21.
  • the source signal G3 is also input to the sensor S3.
  • the source signal G1 transmitted by the weighting element H11, the source signal G2 transmitted by the weighting element H12, and the source signal G3 transmitted by the weighting element Z11 are added to become the input signal N1 of the sensor S1. Further, the source signal G1 transmitted by the weighting element H21, the source signal G2 transmitted by the weighting element H22, and the source signal G3 transmitted by the weighting element Z21 are added to become an input signal N2 of the sensor S2. On the other hand, the source signal G3 is the input signal N3 of the sensor S3. In the sensors S1, S2, and S3, the input signals N1, N2, and N3 are converted into electrical signals and output.
  • the output of the sensor S3 is input to the signal processing unit SP.
  • the signal processing unit SP performs processing for removing stationary noise from the input signal. Then, the output of the signal processing unit SP is transmitted by the weighting elements F11 and F21, and is added to the outputs of the sensors S1 and S2, respectively, to be input to the BSS input units 1 and 2.
  • Blind signal source separation represented by weighting elements W11 to W22 is performed on the inputs of the BSS input units 1 and 2, and separated signals T1 and T2 are obtained from the output units 1 and 2. That is, the BSS input 1 transmitted by the weighting element W11 and the BSS input 2 transmitted by the weighting element W21 are added to become the separated signal T1 of the output unit 1. Further, the BSS input 1 transmitted by the weighting element W12 and the BSS input 2 transmitted by the weighting element W22 are added to become the separated signal T2 of the output unit 2.
  • a pulse wave is assumed as the source signal G1
  • non-stationary noise body movement accompanying traveling or driving
  • stationary noise is mainly assumed as the source signal G3
  • road noise, engine sound, engine vibration, and other noises are applicable. Because it is a mixture of various noises, it can not be decided in general whether it is the noise derived from which.
  • the pulse wave of the driver in driving is detected.
  • the sensors S1 and S2 are installed on the driver's seat SH of the automobile CA and the pulse wave of the driver is detected, the pulse wave signal is transmitted through the driver's body and the sheet SH as indicated by arrows FA and FB.
  • the sensors S1 and S2 include body motion signals generated by the driver's body movement, noise from the engine EG, road noise from the tires TYF and TYB, air conditioning noise from the air conditioner, and various other stationary noises and non-noises.
  • the stationary noise also travels through the driver's body and the sheet SH, as represented by arrows FC, FD, FE, and FF.
  • the sensor S3 detects stationary noise excluding signals such as pulse wave and body movement of the driver.
  • stationary noise excluding signals such as pulse wave and body movement of the driver.
  • noise from the engine EG, road noise from the tires TYF, TYB, air conditioning noise from the air conditioner, and various other stationary noises reach the sensor S3, as represented by arrows FG and FH.
  • the pulse wave of the driver corresponds to the source signal G1 of FIG. 1
  • the body movement of the driver corresponds to the source signal G2 as non-stationary noise.
  • the vibrations of the engine EG and the road noise from the tires TYF and TYB correspond to the noise source G3 as stationary noise.
  • the characteristics of the transfer paths indicated by arrows FA, FC, and FE correspond to the weighting elements H11, H12, and Z11 of FIG. 1
  • the characteristics of the transfer paths indicated by arrows FB, FD, and FF correspond to the weighting elements H21, H22, It corresponds to Z21.
  • the outputs of the piezoelectric sensors 10 and 20 to which the input signals N1 and N2 are input are input to the noise removing units 12 and 22, respectively, together with the output of the piezoelectric sensor 30 to which the input signal N3 is input.
  • the outputs of the noise removal units 12 and 22 are input to the signal separation unit 40, and the signal separation unit 40 is configured to output pulse waves and nonstationary noise that are the separation signals T1 and T2.
  • the vibration waveform sensor of patent document 2 mentioned above is one of the suitable examples.
  • the noise removing units 12 and 22 for example, the ANC method is used. That is, by using a noise signal which is the input signal N3 detected by the piezoelectric sensor 30, a signal of the same amplitude and in reverse phase as the stationary noise is generated, and this is interfered with the output signals of the piezoelectric sensors 10 and 20. Is removed.
  • the signal separation unit 40 the method of BSS is used. That is, signal source separation processing is performed based on statistical features of pulse waves and non-stationary noise such as body movement. As shown in FIG. 3B, the order of the processing by the noise removing units 12 and 22 and the processing by the signal separating unit 40 may be reversed.
  • FIG. 6 shows an experimental example (simulation example) of noise removal and signal source separation according to this embodiment.
  • the horizontal axis in the figure is time, and the vertical axis is signal gain.
  • Graphs GA to GE are as follows.
  • the pulse wave signal GA obtained by removing the stationary noise from the observation signal and separating the non-stationary noise becomes a good signal waveform comparable to the pulse wave signal GE measured at the fingertip.
  • the following effects can be obtained. a. Even when various stationary and non-stationary noises are mixed in the detection signal, the stationary noise is separately detected and removed by the ANC method, and the signal separation is performed by the BSS method. Therefore, the original signal to be detected can be well separated and extracted. b. Since the pulse wave of the driver of the car can be detected well, the health condition of the driver while driving can be accurately grasped.
  • FIG. 4A an example in which the piezoelectric sensor 30 and the noise removing units 12 and 22 of the first embodiment described above are omitted, and the signal separating unit 40 is not used for the outputs of the piezoelectric sensors 10 and 20. It is an applied example.
  • the noise removal processing using the piezoelectric sensor 30 is not performed as compared with the first embodiment described above, for example, when the stationary noise component is small, signals of non-stationary noise such as pulse wave and body movement are also used in this method. Separation can be performed.
  • the example shown to the figure (B) is an example which abbreviate
  • the pulse wave can be detected also by this method.
  • FIG. 6C is an example in which MBPFs (multi-band pass filters) 14 and 24 are used instead of the noise removal units 12 and 22 of the first embodiment described above.
  • MBPFs multi-band pass filters
  • FIG. 5 is an example in which the MBPFs 14 and 24 are added to the example of FIG. 3 (A).
  • FIG. 5A shows an example in which the MBPFs 14 and 24 are connected between the noise removal units 12 and 22 and the signal separation unit 40 respectively
  • FIG. 5B shows the MBPFs 14 and 24 together with the piezoelectric sensors 10 and 20.
  • each of the noise removal units 12 and 22 is connected.
  • the MBPFs 14 and 24 can also be considered as performing noise removal processing because they take out the frequency band in which the pulse wave signal is included. That is, in the example of FIG. 5, it can be considered that noise removal by the noise removal units 12 and 22 and noise removal by the MBPFs 14 and 24 are performed.
  • the present invention is not limited to the embodiments described above, and various modifications can be made without departing from the scope of the present invention. For example, the following are also included.
  • the present invention is also applicable to drivers other than vehicles, such as railway cars, ships, and aircraft. It may be applied to pulse waves of animals.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mechanical Engineering (AREA)
  • Cardiology (AREA)
  • Physics & Mathematics (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Physiology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biophysics (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Transportation (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Signal Processing (AREA)
  • Automation & Control Theory (AREA)
  • Mathematical Physics (AREA)
  • Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
  • Seats For Vehicles (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)

Abstract

[Problem]To provide a signal detecting system capable of appropriately detecting a detection target signal even if intense noise is mixed into an observed signal. [Solution]Outputs from piezoelectric sensors 10, 20 into which input signals N1, N2 are respectively input are input respectively into noise removing units 12, 22 together with an output from a piezoelectric sensor 30 into which an input signal N3 is input. Outputs from the noise removing units 12, 22 are input into a signal separating unit 40, and separated signals T1, T2, which are pulse waves and nonstationary noise respectively, are output from the signal separating unit 40. Among the above, for the noise removing units 12, 22, an ANC technique is employed to generate a signal having the same amplitude and the opposite phase to stationary noise using a noise signal, which is the input signal N3 detected by the piezoelectric sensor 30, and the stationary noise is removed by causing said signal to interfere with the output signals from the piezoelectric sensors 10, 20. Next, for the signal separating unit 40, a BSS technique is employed to perform a signal source separating process based on statistical features of the nonstationary noise, such as body movements.

Description

信号検出システムSignal detection system
本発明は、多様な背景雑音が存在する環境において得た観測信号から検出したい検出対象信号を分離する信号検出システムに関するものであり、特に、圧電素子を利用して検出した生体情報の一つである脈波信号の検出に好適な信号検出システムに関するものである。 The present invention relates to a signal detection system for separating a detection target signal desired to be detected from an observation signal obtained in an environment where various background noises exist, and in particular, it is one of biological information detected using a piezoelectric element. The present invention relates to a signal detection system suitable for detecting a certain pulse wave signal.
脈波に関する背景技術としては、例えば、下記特許文献１記載の「動脈硬化評価装置」がある。これは、入射波と反射波とを正確に分離して、個体差による動脈硬化度を精度よく評価できるようにしたもので、被験者の頸部において、圧電トランスデューサで動脈を伝わる脈波を変位信号として検出するとともに、超音波診断装置のプローブで動脈の血流速度を測定する。そして超音波診断装置のプローブで得られた血流速度を変位信号に変換して入射波を得た後、圧電トランスデューサによって検出された変位信号から入射波を差し引いて反射波を得、更に、入射波と反射波の振幅強度から生体の血管機能を評価するようにしたものである。また、下記特許文献２には、指先に巻き付けて使用し、指先の動脈血管壁からの振動を脈波として捉えるようにした「振動波形センサ及び波形解析装置」が示されている。 As background art regarding pulse waves, for example, there is an "arteriosclerosis evaluation apparatus" described in Patent Document 1 below. This accurately separates the incident wave and the reflected wave so that the degree of arteriosclerosis due to individual differences can be evaluated accurately, and in the subject's neck, a pulse wave transmitted through the artery with a piezoelectric transducer is used as a displacement signal. As well as measuring the blood flow velocity of the artery with the probe of the ultrasonic diagnostic apparatus. Then, the blood flow velocity obtained by the probe of the ultrasonic diagnostic apparatus is converted into a displacement signal to obtain an incident wave, and then the incident wave is subtracted from the displacement signal detected by the piezoelectric transducer to obtain a reflected wave, and further incident The blood vessel function of the living body is evaluated from the amplitude intensity of the wave and the reflected wave. Further, Patent Document 2 below shows a “vibration waveform sensor and waveform analysis device” which is used by being wound around a fingertip and capturing the vibration from the arterial blood vessel wall of the fingertip as a pulse wave.
国際公開第２０１０／０２４４１７号パンフレットWO 2010/024417 pamphlet 国際公開第２０１６／１６７２０２号パンフレットInternational Publication No. 2016/167202 brochure
上述した圧電体を利用した振動波形センサは、人体に直接取り付けなくてもよく、例えば自動車のシートに組み込んで太腿の脈波を測定すれば、運転者の着座の有無や健康状態を検出することができ、極めて有益である。しかしながら、自動車のシートに圧電センサを組み込んだ場合、空調音，エンジン音，ロードノイズ（走行音）など、各種の雑音があることから、これらを除去ないし分離して脈波信号を良好に検出する必要がある。　 The vibration waveform sensor using the above-mentioned piezoelectric body does not have to be directly attached to the human body. For example, if it is incorporated into a seat of an automobile and the pulse wave of the thigh is measured, the presence or absence of seating of the driver and health condition are detected. It can be extremely beneficial. However, when a piezoelectric sensor is incorporated into a car seat, various noises such as air conditioning noise, engine noise and road noise (running noise) are removed or separated to detect pulse wave signals well. There is a need.
信号分離に関する技術としては、源信号の既知の統計的な特徴に基づいて分離するブラインド信号分離（ＢＳＳ：Blind signal separation）の手法が知られている。例えば、ＩＣＡ（Independent Component Analysis）では、それぞれ独立な現信号が時空間的に混合されたと仮定して、観測信号の統計的独立性を確立するようにパラメーターを調整することで信号分離を達成している。統計的独立性の基準となる尺度としては、ＫＬダイバージェンス（Kullback Leibler Divergence）が広く用いられている。　 As a technique related to signal separation, there is known a blind signal separation (BSS) method of separating based on known statistical features of a source signal. For example, in ICA (Independent Component Analysis), signal separation is achieved by adjusting parameters so as to establish statistical independence of observed signals, assuming that independent current signals are temporally and spatially mixed. ing. As a measure to be a standard of statistical independence, Kullback Leibler Divergence is widely used.
一方、空調音，エンジン音，ロードノイズなどの雑音を抑圧する手法として、ＡＮＣ（Active Noise Control）が知られている。ＡＮＣは、所望の信号以外の雑音源の信号のみを取得するセンサを用いて、このセンサから得た信号を用いて、雑音を抑圧したい地点に対して騒音と同振幅逆位相の信号を生成してこれを干渉させることで、特定の地点での騒音を除去する手法である。　 On the other hand, ANC (Active Noise Control) is known as a method for suppressing noise such as air conditioning noise, engine noise and road noise. ANC uses a signal obtained from this sensor using a sensor that acquires only the signal of noise sources other than the desired signal, and generates a signal with the same amplitude and opposite phase as the noise to the point where noise is desired to be suppressed. It is a method to remove the noise at a specific point by making it interfere.
しかしながら、信号分離を行う際に、観測信号を取得するセンサに強い雑音が混入してしまうと、分離の根拠となる源信号での統計的性質が成立しなくなってしまう場合がある。また、観測信号を取得するセンサのＳＮＲ（Signal to Noize Ratio）が低い環境下では、適切に信号分離が実行できない場合がある。　 However, when strong noise is mixed into a sensor that acquires an observation signal when performing signal separation, statistical properties may not be established in the source signal that is the basis of the separation. In addition, in an environment where the signal-to-noize ratio (SNR) of a sensor that acquires observation signals is low, signal separation may not be properly performed.
本発明は、かかる点に着目したもので、その目的は、観測信号に強い雑音が混入した場合でも検出対象信号を適切に検出することができる信号検出システムを提供することである。他の目的は、自動車の運転者の脈波の検出などを良好に行って、事故の防止などに役立てることである。 The present invention focuses on such a point, and an object thereof is to provide a signal detection system capable of appropriately detecting a detection target signal even when strong noise is mixed in an observation signal. Another object is to perform detection of pulse wave of a driver of a car well and to help prevent accidents.
本発明は、検出対象信号に定常雑音や非定常雑音が含まれている観測信号を取得する複数の観測信号取得手段と、前記定常雑音を取得する雑音取得手段と、前記観測信号取得手段で得た観測信号から、前記雑音取得手段で得た定常雑音を除去する雑音除去手段と、を備えたことを特徴とする。　 The present invention is obtained by a plurality of observation signal acquisition means for acquiring observation signals including stationary noise and non-stationary noise in the detection target signal, noise acquisition means for acquiring the stationary noise, and the observation signal acquisition means And noise removing means for removing stationary noise obtained by the noise obtaining means from the observed signal.
他の発明は、検出対象信号に定常雑音や非定常雑音が含まれている観測信号を取得する複数の観測信号取得手段と、前記定常雑音を取得する雑音取得手段と、前記観測信号取得手段で得た観測信号から、前記雑音取得手段で得た定常雑音を除去する雑音除去手段と、この雑音除去手段による雑音除去後の信号から前記検出対象信号を分離する信号分離手段と、を備えたことを特徴とする。　 Another invention is a plurality of observation signal acquisition means for acquiring observation signals in which stationary signals and non-stationary noise are included in a detection target signal, noise acquisition means for acquiring the stationary noise, and the observation signal acquisition means Noise removal means for removing stationary noise obtained by the noise acquisition means from the obtained observation signal; and signal separation means for separating the detection target signal from the signal after noise removal by the noise removal means. It is characterized by
更に他の発明は、検出対象信号に定常雑音や非定常雑音が含まれている観測信号を取得する複数の観測信号取得手段と、この観測信号取得手段で得た観測信号から前記非定常雑音を分離する信号分離手段と、前記定常雑音を取得する雑音取得手段と、前記信号分離手段で分離して得た信号から、前記雑音取得手段で得た定常雑音を除去する雑音除去手段と、を備えたことを特徴とする。あるいは、検出対象信号に非定常雑音が含まれている観測信号を取得する複数の観測信号取得手段と、この観測信号取得手段で得た観測信号から前記非定常雑音を分離する信号分離手段と、を備えたことを特徴とする。　 In still another invention, a plurality of observation signal acquisition means for acquiring observation signals including stationary noise and non-stationary noise in a detection target signal, and the non-stationary noise from observation signals obtained by the observation signal acquisition means It comprises: signal separation means for separation; noise acquisition means for acquiring the stationary noise; and noise removal means for removing stationary noise obtained by the noise acquisition means from the signal obtained by separation by the signal separation means. It is characterized by Alternatively, a plurality of observation signal acquisition means for acquiring an observation signal in which non-stationary noise is included in the detection target signal, and signal separation means for separating the non-stationary noise from the observation signal obtained by the observation signal acquisition means It is characterized by having.
主要な形態の一つによれば、前記検出対象信号が含まれている周波数帯域の信号を取り出すＭＢＰＦ（マルチバンドパスフィルタ）手段を含むことを特徴とする。他の形態によれば、前記検出対象信号が、乗り物を運転する運転者を含む人の脈波であることを特徴とする。更に他の形態によれば、前記雑音除去手段がＡＮＣの手法で雑音を除去し、前記信号分離手段がＢＳＳの手法で信号を分離することを特徴とする。本発明の前記及び他の目的，特徴，利点は、以下の詳細な説明及び添付図面から明瞭になろう。 According to one of the main modes, it is characterized in that it includes an MBPF (multi-band pass filter) means for extracting a signal of a frequency band in which the detection target signal is included. According to another aspect, the detection target signal is a pulse wave of a person including a driver driving a vehicle. According to still another aspect, the noise removal means removes noise by an ANC method, and the signal separation means separates signals by a BSS method. The above and other objects, features and advantages of the present invention will be apparent from the following detailed description and the accompanying drawings.
本発明によれば、検出対象信号に定常雑音や非定常雑音が含まれている観測信号を複数取得するとともに、定常雑音を別途取得し、前記観測信号から前記定常雑音を除去して前記検出対象信号を分離することとしたので、観測信号に強い雑音が混入した場合でも検出対象信号を適切に分離することができ、自動車の運転者の脈波の検出などを良好に行うことができる。 According to the present invention, a plurality of observation signals in which stationary noise and non-stationary noise are included in the detection target signal are acquired, stationary noise is separately acquired, and the stationary noise is removed from the observation signal to detect the detection target Since the signals are separated, even when strong noise is mixed in the observation signal, the detection target signal can be properly separated, and the pulse wave of the driver of the vehicle can be favorably detected.
本発明の一実施例の基本的な構成を示す図である。It is a figure which shows the basic composition of one Example of this invention. 本発明を自動車の運転者の脈波信号検出に適用する場合の様子を示す図である。It is a figure which shows a mode in the case of applying this invention to the pulse wave signal detection of the driver | operator of a motor vehicle. 前記図２に本発明を適用した場合のシステム構成を示す図である。It is a figure which shows the system configuration at the time of applying this invention to the said FIG. 本発明の他の実施例の構成を示す図である。It is a figure which shows the structure of the other Example of this invention. 本発明の他の実施例の構成を示す図である。It is a figure which shows the structure of the other Example of this invention. 前記実施例による雑音除去及び信号源分離の一例を示すグラフである。It is a graph which shows an example of the noise removal by the said Example, and source separation.
以下、本発明を実施するための最良の形態を、実施例に基づいて詳細に説明する。 Hereinafter, the best mode for carrying out the present invention will be described in detail based on examples.
最初に、図１～図３，図６を参照しながら、本発明の実施例１について説明する。図１には、本実施例における信号検出システムの基本構成が示されている。同図において、信号の発信元から出力された源信号Ｇ１，Ｇ２，Ｇ３は、多様な伝達経路を通じて、センサＳ１，Ｓ２にそれぞれ入力する。このときの源信号Ｇ１，Ｇ２，Ｇ３がセンサＳ１，Ｓ２に到達する伝達経路の伝達特性を、重み付け要素Ｈ11～Ｈ22，Ｚ11，Ｚ21で示している。一方、源信号Ｇ３は、センサＳ３にも入力する。　 First, Embodiment 1 of the present invention will be described with reference to FIGS. 1 to 3 and 6. FIG. 1 shows the basic configuration of the signal detection system in the present embodiment. In the figure, source signals G1, G2 and G3 output from the signal source are respectively input to the sensors S1 and S2 through various transmission paths. The transfer characteristics of the transfer path in which the source signals G1, G2, G3 at this time reach the sensors S1, S2 are indicated by weighting elements H11 to H22, Z11, Z21. On the other hand, the source signal G3 is also input to the sensor S3.
すなわち、重み付け要素Ｈ11で伝達した源信号Ｇ１と、重み付け要素Ｈ12で伝達した源信号Ｇ２と、重み付け要素Ｚ11で伝達した源信号Ｇ３とが加算されて、センサＳ１の入力信号Ｎ１となる。また、重み付け要素Ｈ21で伝達した源信号Ｇ１と、重み付け要素Ｈ22で伝達した源信号Ｇ２と、重み付け要素Ｚ21で伝達した源信号Ｇ３とが加算されて、センサＳ２の入力信号Ｎ２となる。一方、源信号Ｇ３は、センサＳ３の入力信号Ｎ３となる。センサＳ１，Ｓ２，Ｓ３では、入力信号Ｎ１，Ｎ２，Ｎ３が電気的信号に変換されて出力される。　 That is, the source signal G1 transmitted by the weighting element H11, the source signal G2 transmitted by the weighting element H12, and the source signal G3 transmitted by the weighting element Z11 are added to become the input signal N1 of the sensor S1. Further, the source signal G1 transmitted by the weighting element H21, the source signal G2 transmitted by the weighting element H22, and the source signal G3 transmitted by the weighting element Z21 are added to become an input signal N2 of the sensor S2. On the other hand, the source signal G3 is the input signal N3 of the sensor S3. In the sensors S1, S2, and S3, the input signals N1, N2, and N3 are converted into electrical signals and output.
センサＳ３の出力は信号処理部ＳＰに入力される。信号処理部ＳＰでは、入力信号に対して定常雑音を除去する処理が行われる。そして、信号処理部ＳＰの出力は、重み付け要素Ｆ11，Ｆ21で伝達し、上述したセンサＳ１，Ｓ２の出力とそれぞれ加算されて、ＢＳＳ入力部１，２の入力となる。　 The output of the sensor S3 is input to the signal processing unit SP. The signal processing unit SP performs processing for removing stationary noise from the input signal. Then, the output of the signal processing unit SP is transmitted by the weighting elements F11 and F21, and is added to the outputs of the sensors S1 and S2, respectively, to be input to the BSS input units 1 and 2.
ＢＳＳ入力部１，２の入力に対しては、重み付け要素Ｗ11～Ｗ22で表されるブラインド信号源分離が行われ、出力部１，２から、分離信号Ｔ１，Ｔ２が得られる。すなわち、重み付け要素Ｗ11で伝達したＢＳＳ入力１と、重み付け要素Ｗ21で伝達したＢＳＳ入力２とが加算されて、出力部１の分離信号Ｔ１となる。また、重み付け要素Ｗ12で伝達したＢＳＳ入力１と、重み付け要素Ｗ22で伝達したＢＳＳ入力２とが加算されて、出力部２の分離信号Ｔ２となる。　 Blind signal source separation represented by weighting elements W11 to W22 is performed on the inputs of the BSS input units 1 and 2, and separated signals T1 and T2 are obtained from the output units 1 and 2. That is, the BSS input 1 transmitted by the weighting element W11 and the BSS input 2 transmitted by the weighting element W21 are added to become the separated signal T1 of the output unit 1. Further, the BSS input 1 transmitted by the weighting element W12 and the BSS input 2 transmitted by the weighting element W22 are added to become the separated signal T2 of the output unit 2.
以上のうち、本実施例では、前記源信号Ｇ１として脈波を想定し、前記源信号Ｇ２として非定常雑音（走行や運転に伴う体動）を想定している。一方、源信号Ｇ３としては、主として定常雑音を想定しており、ロードノイズ，エンジン音，エンジンの振動，その他の雑音が該当する。色々な雑音の混合になるので、いずれから由来する雑音なのかは一概に決まらない。　 Among the above, in the present embodiment, a pulse wave is assumed as the source signal G1, and non-stationary noise (body movement accompanying traveling or driving) is assumed as the source signal G2. On the other hand, stationary noise is mainly assumed as the source signal G3, and road noise, engine sound, engine vibration, and other noises are applicable. Because it is a mixture of various noises, it can not be decided in general whether it is the noise derived from which.
センサＳ１，Ｓ２には、源信号Ｇ１の脈波と、源信号Ｇ２の非定常雑音が混合された信号とともに、源信号Ｇ３である定常雑音が混入している。しかし、センサＳ３をセンサＳ１，Ｓ２から隔離するようにすると、センサＳ３は源信号Ｇ３である定常雑音のみを取得し、他の源信号Ｇ１，Ｇ２は取得しない。従って、センサＳ３を利用すれば、センサＳ１，Ｓ２に混入した源信号Ｇ３の定常雑音を除去することが可能となる。　 In the sensors S1 and S2, stationary noise which is the source signal G3 is mixed together with a pulse wave of the source signal G1 and a signal in which non-stationary noise of the source signal G2 is mixed. However, when the sensor S3 is isolated from the sensors S1 and S2, the sensor S3 acquires only the stationary noise which is the source signal G3 and does not acquire the other source signals G1 and G2. Therefore, by using the sensor S3, it is possible to remove stationary noise of the source signal G3 mixed in the sensors S1 and S2.
例えば、図２に示すように、運転中の運転者の脈波を検出するような場合を想定する。自動車ＣＡの運転席シートＳＨに、センサＳ１，Ｓ２を設置し、運転者の脈波を検出する場合、脈波信号は、矢印ＦＡ，ＦＢで示すように、運転者の体内やシートＳＨを伝わって、センサＳ１，Ｓ２に到達する。一方、センサＳ１，Ｓ２には、運転者の体の動きによって生ずる体動信号，エンジンＥＧからの雑音，タイヤＴＹＦ，ＴＹＢからのロードノイズ，エアコンからの空調音，その他の各種の定常雑音・非定常雑音も、矢印ＦＣ，ＦＤ，ＦＥ，ＦＦで代表して示すように、運転者の体内やシートＳＨを伝わって到達する。　 For example, as shown in FIG. 2, it is assumed that the pulse wave of the driver in driving is detected. When the sensors S1 and S2 are installed on the driver's seat SH of the automobile CA and the pulse wave of the driver is detected, the pulse wave signal is transmitted through the driver's body and the sheet SH as indicated by arrows FA and FB. To reach the sensors S1 and S2. On the other hand, the sensors S1 and S2 include body motion signals generated by the driver's body movement, noise from the engine EG, road noise from the tires TYF and TYB, air conditioning noise from the air conditioner, and various other stationary noises and non-noises. The stationary noise also travels through the driver's body and the sheet SH, as represented by arrows FC, FD, FE, and FF.
これに対し、センサＳ３は、運転者の脈波や体動などの信号を除く定常雑音を検出する。例えば、エンジンＥＧからの雑音，タイヤＴＹＦ，ＴＹＢからのロードノイズ，エアコンからの空調音，その他の各種の定常雑音が、矢印ＦＧ，ＦＨで代表して示すように、センサＳ３に到達する。　 On the other hand, the sensor S3 detects stationary noise excluding signals such as pulse wave and body movement of the driver. For example, noise from the engine EG, road noise from the tires TYF, TYB, air conditioning noise from the air conditioner, and various other stationary noises reach the sensor S3, as represented by arrows FG and FH.
このような例のうち、運転者の脈波が図１の源信号Ｇ１に対応し、運転者の体動は非定常雑音として源信号Ｇ２に対応する。エンジンＥＧの振動やタイヤＴＹＦ，ＴＹＢからのロードノイズは、定常雑音として雑音源Ｇ３に対応する。また、矢印ＦＡ，ＦＣ，ＦＥで示す伝達経路の特性が、図１の重み付け要素Ｈ11，Ｈ12，Ｚ11に対応し、矢印ＦＢ，ＦＤ，ＦＦで示す伝達経路の特性が、重み付け要素Ｈ21，Ｈ22，Ｚ21に対応する。このような図２の例に、図１の手法を適用することで、分離信号Ｔ１として脈波信号が得られ、分離信号Ｔ２として体動信号が得られる。　 In such an example, the pulse wave of the driver corresponds to the source signal G1 of FIG. 1, and the body movement of the driver corresponds to the source signal G2 as non-stationary noise. The vibrations of the engine EG and the road noise from the tires TYF and TYB correspond to the noise source G3 as stationary noise. Also, the characteristics of the transfer paths indicated by arrows FA, FC, and FE correspond to the weighting elements H11, H12, and Z11 of FIG. 1, and the characteristics of the transfer paths indicated by arrows FB, FD, and FF correspond to the weighting elements H21, H22, It corresponds to Z21. By applying the method of FIG. 1 to such an example of FIG. 2, a pulse wave signal is obtained as the separated signal T1, and a body motion signal is obtained as the separated signal T2.
以上のような図１の手法を信号処理システムとして示すと、図３(A)に示すような構成となる。同図に示すように、入力信号Ｎ１，Ｎ２が入力される圧電センサ１０，２０の出力は、入力信号Ｎ３が入力される圧電センサ３０の出力とともに、雑音除去部１２，２２にそれぞれ入力されている。雑音除去部１２，２２の出力は信号分離部４０に入力されており、信号分離部４０から分離信号Ｔ１，Ｔ２である脈波及び非定常雑音が出力されるようになっている。なお、圧電センサ１０，２０，３０としては、上述した特許文献２に記載の振動波形センサが好適な例の一つである。　 When the method of FIG. 1 as described above is shown as a signal processing system, it has a configuration as shown in FIG. As shown in the figure, the outputs of the piezoelectric sensors 10 and 20 to which the input signals N1 and N2 are input are input to the noise removing units 12 and 22, respectively, together with the output of the piezoelectric sensor 30 to which the input signal N3 is input. There is. The outputs of the noise removal units 12 and 22 are input to the signal separation unit 40, and the signal separation unit 40 is configured to output pulse waves and nonstationary noise that are the separation signals T1 and T2. In addition, as piezoelectric sensor 10, 20, 30, the vibration waveform sensor of patent document 2 mentioned above is one of the suitable examples.
これらのうち、雑音除去部１２，２２としては、例えばＡＮＣの手法を用いる。すなわち、圧電センサ３０で検出された入力信号Ｎ３である雑音信号を用いて定常雑音と同振幅逆位相の信号を生成し、これを圧
電センサ１０，２０の出力信号に干渉させることで、定常雑音が除去される。次に、信号分離部４０としては、ＢＳＳの手法を用いる。すなわち、脈波と、体動などの非定常雑音の統計的な特徴に基づいた信号源分離処理が行われる。なお、図３(B)に示すように、雑音除去部１２，２２による処理と、信号分離部４０の処理の順序を逆に行うようにしてもよい。　 Among these, as the noise removing units 12 and 22, for example, the ANC method is used. That is, by using a noise signal which is the input signal N3 detected by the piezoelectric sensor 30, a signal of the same amplitude and in reverse phase as the stationary noise is generated, and this is interfered with the output signals of the piezoelectric sensors 10 and 20. Is removed. Next, as the signal separation unit 40, the method of BSS is used. That is, signal source separation processing is performed based on statistical features of pulse waves and non-stationary noise such as body movement. As shown in FIG. 3B, the order of the processing by the noise removing units 12 and 22 and the processing by the signal separating unit 40 may be reversed.
図６には、本実施例による雑音除去及び信号源分離の実験例（シミュレーション例）が示されている。同図の横軸は時間，縦軸は信号利得である。グラフＧＡ～ＧＥは、次のとおりである。ＧＡ：観測信号から定常雑音を除去し非定常雑音を分離した脈波信号ＧＢ：観測信号から定常雑音を除去した信号ＧＣ：ロードノイズなどの定常雑音ＧＤ：当初の観測信号ＧＥ：指先で計測した脈波信号　 FIG. 6 shows an experimental example (simulation example) of noise removal and signal source separation according to this embodiment. The horizontal axis in the figure is time, and the vertical axis is signal gain. Graphs GA to GE are as follows. GA: Pulse wave signal GB in which stationary noise is removed from observed signal and non-stationary noise is separated: signal GC in which stationary noise is removed from observed signal GC: stationary noise such as road noise GD: original observed signal GE: measured at fingertip Pulse wave signal
これらのグラフを比較すれば明らかなように、観測信号から定常雑音を除去し非定常雑音を分離した脈波信号ＧＡは、指先で計測した脈波信号ＧＥに匹敵する良好な信号波形となっている。　 As apparent from comparison of these graphs, the pulse wave signal GA obtained by removing the stationary noise from the observation signal and separating the non-stationary noise becomes a good signal waveform comparable to the pulse wave signal GE measured at the fingertip. There is.
以上のように、本実施例によれば、次のような効果が得られる。ａ，検出信号に各種の定常・非定常の雑音が混入している場合であっても、別途定常雑音を検出してＡＮＣの手法で除去するとともに、ＢＳＳの手法で信号分離を行うこととしたので、検出すべき本来の信号を良好に分離・抽出することができる。ｂ，自動車の運転者の脈波を良好に検出することができるので、運転中の運転者の健康状態を的確に把握することができる。 As described above, according to this embodiment, the following effects can be obtained. a. Even when various stationary and non-stationary noises are mixed in the detection signal, the stationary noise is separately detected and removed by the ANC method, and the signal separation is performed by the BSS method. Therefore, the original signal to be detected can be well separated and extracted. b. Since the pulse wave of the driver of the car can be detected well, the health condition of the driver while driving can be accurately grasped.
次に、図４～図５を参照しながら、本発明の他の実施例について説明する。まず、図４(A)に示す例は、上述した実施例１の圧電センサ３０と雑音除去部１２，２２を省略した例であり、圧電センサ１０，２０の出力に対して信号分離部４０を適用した例である。上述した実施例１と比較して、圧電センサ３０を利用した雑音除去の処理が行われないが、例えば定常雑音成分が少ないときは、この手法でも脈波と体動などの非定常雑音の信号分離を行うことができる。　 Next, another embodiment of the present invention will be described with reference to FIGS. 4 to 5. First, an example shown in FIG. 4A is an example in which the piezoelectric sensor 30 and the noise removing units 12 and 22 of the first embodiment described above are omitted, and the signal separating unit 40 is not used for the outputs of the piezoelectric sensors 10 and 20. It is an applied example. Although the noise removal processing using the piezoelectric sensor 30 is not performed as compared with the first embodiment described above, for example, when the stationary noise component is small, signals of non-stationary noise such as pulse wave and body movement are also used in this method. Separation can be performed.
同図(B)に示す例は、上述した実施例１の信号分離部４０を省略した例で、雑音除去部１２，２２の出力がそれぞれ脈波，非定常雑音となる。非定常雑音成分が少ないときは、この手法でも脈波を検出することができる。　 The example shown to the figure (B) is an example which abbreviate | omitted the signal separation part 40 of Example 1 mentioned above, and the output of the noise removal parts 12 and 22 turns into a pulse wave and non-stationary noise, respectively. When the non-stationary noise component is small, the pulse wave can be detected also by this method.
同図(C)に示す例は、上述した実施例１の雑音除去部１２，２２の代わりに、ＭＢＰＦ（マルチバンドパスフィルタ）１４，２４を使用する例である。ＭＢＰＦ１４，２４によって脈波信号が含まれている周波数帯域の信号を取り出すことで、良好に脈波信号を得ることができる。なお、ＭＢＰＦ１４，２４も、結果的に雑音除去を行っていると考えることができる。　 The example shown in FIG. 6C is an example in which MBPFs (multi-band pass filters) 14 and 24 are used instead of the noise removal units 12 and 22 of the first embodiment described above. By extracting the signal of the frequency band in which the pulse wave signal is included by the MBPFs 14 and 24, it is possible to obtain the pulse wave signal well. In addition, it can be considered that the MBPFs 14 and 24 also perform noise removal as a result.
図５は、図３(A)の例に、ＭＢＰＦ１４，２４を追加した例である。図５(A)は、ＭＢＰＦ１４，２４を雑音除去部１２，２２と信号分離部４０の間にそれぞれ接続した例であり、図５(B)は、ＭＢＰＦ１４，２４を、圧電センサ１０，２０と雑音除去部１２，２２の間にそれぞれ接続した例である。なお、図３(B)に示した雑音除去部１２，２２による処理と信号分離部４０の処理の順序を逆に行う場合にも、同様に適用可能である。これらの例のように、ＭＢＰＦ１４，２４を適用することで、より良好な信号分離を行って、脈波を検出することができる。なお、ＭＢＰＦ１４，２４も、脈波信号が含まれている周波数帯域を取り出すことから、雑音除去処理を行っていると考えることができる。すなわち、図５の例では、雑音除去部１２，２２による雑音除去と、ＭＢＰＦ１４，２４による雑音除去とが行われると考えることができる。　 FIG. 5 is an example in which the MBPFs 14 and 24 are added to the example of FIG. 3 (A). FIG. 5A shows an example in which the MBPFs 14 and 24 are connected between the noise removal units 12 and 22 and the signal separation unit 40 respectively, and FIG. 5B shows the MBPFs 14 and 24 together with the piezoelectric sensors 10 and 20. This is an example in which each of the noise removal units 12 and 22 is connected. The same applies to the case where the processing by the noise removal units 12 and 22 and the processing by the signal separation unit 40 shown in FIG. 3B are performed in the reverse order. By applying the MBPFs 14 and 24 as in these examples, better signal separation can be performed to detect a pulse wave. The MBPFs 14 and 24 can also be considered as performing noise removal processing because they take out the frequency band in which the pulse wave signal is included. That is, in the example of FIG. 5, it can be considered that noise removal by the noise removal units 12 and 22 and noise removal by the MBPFs 14 and 24 are performed.
なお、本発明は、上述した実施例に限定されるものではなく、本発明の要旨を逸脱しない範囲内において種々変更を加え得ることができる。例えば、以下のものも含まれる。(1)前記実施例では、自動車を運転する運転者の脈波を検出する場合を好適な例としたが、運転者の脈波に限定されるものではなく、脈波以外の信号に対しても適用可能である。また、自動車以外の乗り物，例えば鉄道車両，船舶，航空機などの運転者に対しても同様に適用可能である。動物の脈波に対して適用してもよい。(2)雑音除去部による雑音除去の手法としては、上述したＡＮＣに限定されるものではなく、各種の公知の手法を適用してよい。 The present invention is not limited to the embodiments described above, and various modifications can be made without departing from the scope of the present invention. For example, the following are also included. (1) In the above embodiment, although the case where the pulse wave of the driver driving the car is detected is a preferred example, the present invention is not limited to the pulse wave of the driver but for signals other than pulse waves. Is also applicable. The present invention is also applicable to drivers other than vehicles, such as railway cars, ships, and aircraft. It may be applied to pulse waves of animals. (2) As a method of the noise removal by a noise removal part, it is not limited to ANC mentioned above, You may apply various well-known methods.
本発明によれば、検出対象信号に定常雑音や非定常雑音が含まれている観測信号を複数取得するとともに、定常雑音を別途取得し、前記観測信号から前記定常雑音を除去して前記検出対象信号を分離することとしたので、観測信号に強い雑音が混入した場合でも検出対象信号を適切に分離することができ、自動車の運転者の脈波の検出などを良好に行うことができる。従って、運転者の健康状態を的確に把握して、事故の防止などに役立てることができる。 According to the present invention, a plurality of observation signals in which stationary noise and non-stationary noise are included in the detection target signal are acquired, stationary noise is separately acquired, and the stationary noise is removed from the observation signal to detect the detection target Since the signals are separated, even when strong noise is mixed in the observation signal, the detection target signal can be properly separated, and the pulse wave of the driver of the vehicle can be favorably detected. Therefore, the driver's health condition can be accurately grasped and it can be used to prevent an accident.
１０，２０，３０：圧電センサ１２，２２：雑音除去部１４，２４：ＭＢＰＦ４０：信号分離部ＣＡ：自動車ＥＧ：エンジンＧ１，Ｇ２，Ｇ３：源信号Ｈ11～Ｈ22，Ｚ11～Ｚ21，Ｆ11～Ｆ21，Ｗ11～Ｗ22：重み付け要素Ｎ１，Ｎ２：入力信号Ｓ１，Ｓ２，Ｓ３：センサＳＨ：シートＴ１，Ｔ２：分離信号ＴＹＦ，ＴＹＢ：タイヤ 10, 20, 30: Piezoelectric sensor 12, 22: noise removing unit 14, 24: MBPF 40: signal separation unit CA: automobile EG: engine G1, G2, G3: source signals H11 to H22, Z11 to Z21, F11 to F21, W11 to W22: weighting elements N1 and N2: input signals S1, S2 and S3: sensor SH: sheets T1 and T2: separation signals TYF and TYB: tires

Claims (8)

  1. 検出対象信号に定常雑音や非定常雑音が含まれている観測信号を取得する複数の観測信号取得手段と、　前記定常雑音を取得する雑音取得手段と、　前記観測信号取得手段で得た観測信号から、前記雑音取得手段で得た定常雑音を除去する雑音除去手段と、を備えたことを特徴とする信号検出システム。 A plurality of observation signal acquisition means for acquiring observation signals including stationary noise and non-stationary noise in the detection target signal, noise acquisition means for acquiring the stationary noise, and observation signals obtained by the observation signal acquisition means A signal detection system comprising: noise removing means for removing stationary noise obtained by the noise obtaining means.
  2. 検出対象信号に定常雑音や非定常雑音が含まれている観測信号を取得する複数の観測信号取得手段と、　前記定常雑音を取得する雑音取得手段と、　前記観測信号取得手段で得た観測信号から、前記雑音取得手段で得た定常雑音を除去する雑音除去手段と、　この雑音除去手段による雑音除去後の信号から前記検出対象信号を分離する信号分離手段と、を備えたことを特徴とする信号検出システム。 A plurality of observation signal acquisition means for acquiring observation signals including stationary noise and non-stationary noise in the detection target signal, noise acquisition means for acquiring the stationary noise, and observation signals obtained by the observation signal acquisition means A signal comprising: noise removing means for removing stationary noise obtained by the noise obtaining means; and signal separating means for separating the detection target signal from the signal after noise removal by the noise removing means. Detection system.
  3. 検出対象信号に定常雑音や非定常雑音が含まれている観測信号を取得する複数の観測信号取得手段と、　この観測信号取得手段で得た観測信号から前記非定常雑音を分離する信号分離手段と、　前記定常雑音を取得する雑音取得手段と、　前記信号分離手段で分離して得た信号から、前記雑音取得手段で得た定常雑音を除去する雑音除去手段と、を備えたことを特徴とする信号検出システム。 A plurality of observation signal acquisition means for acquiring an observation signal including stationary noise and non-stationary noise in a detection target signal; signal separation means for separating the non-stationary noise from the observation signal obtained by the observation signal acquisition means; A noise acquiring unit for acquiring the stationary noise; and a noise removing unit for removing the stationary noise obtained by the noise acquiring unit from the signal obtained by being separated by the signal separating unit. Signal detection system.
  4. 検出対象信号に非定常雑音が含まれている観測信号を取得する複数の観測信号取得手段と、　この観測信号取得手段で得た観測信号から前記非定常雑音を分離する信号分離手段と、を備えたことを特徴とする信号検出システム。 A plurality of observation signal acquisition means for acquiring an observation signal in which non-stationary noise is included in the detection target signal, and signal separation means for separating the non-stationary noise from the observation signal obtained by the observation signal acquisition means A signal detection system characterized by
  5. 前記検出対象信号が含まれている周波数帯域の信号を取り出すＭＢＰＦ（マルチバンドパスフィルタ）手段を含むことを特徴とする請求項１～４のいずれか一項に記載の信号検出システム。 The signal detection system according to any one of claims 1 to 4, further comprising an MBPF (multi-band pass filter) means for extracting a signal of a frequency band in which the detection target signal is included.
  6. 前記検出対象信号が、乗り物を運転する運転者を含む人の脈波であることを特徴とする請求項１～５のいずれか一項に記載の信号検出システム。 The signal detection system according to any one of claims 1 to 5, wherein the detection target signal is a pulse wave of a person including a driver driving a vehicle.
  7. 前記定常雑音が、空調音，エンジン音，ロードノイズまたはエンジンの振動を含む雑音であり、前記非定常雑音が、運転者の体の動きによって生ずる体動であることを特徴とする請求項１～６のいずれか一項に記載の信号検出システム。 The stationary noise is air conditioning noise, engine noise, road noise or noise including engine vibration, and the non-stationary noise is body movement caused by driver's body movement. 6. The signal detection system according to any one of 6.
  8. 前記雑音除去手段がＡＮＣ（Active Noise Control）の手法で雑音を除去し、前記信号分離手段がＢＳＳ（Blind signal separation）の手法で信号を分離することを特徴とする請求項１～７のいずれか一項に記載の信号検出システム。 The noise removal means removes noise by an ANC (Active Noise Control) method, and the signal separation means separates a signal by a BSS (Blind signal separation) method. A signal detection system according to one of the preceding claims.
PCT/JP2018/039578 2017-11-07 2018-10-25 Signal detecting system WO2019093132A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2017-214906 2017-11-07
JP2017214906A JP2019084069A (en) 2017-11-07 2017-11-07 Signal detection system

Publications (1)

Publication Number Publication Date
WO2019093132A1 true WO2019093132A1 (en) 2019-05-16

Family

ID=66437667

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2018/039578 WO2019093132A1 (en) 2017-11-07 2018-10-25 Signal detecting system

Country Status (2)

Country Link
JP (1) JP2019084069A (en)
WO (1) WO2019093132A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003235813A (en) * 2002-12-19 2003-08-26 Matsushita Electric Ind Co Ltd Monitoring device
JP2009538720A (en) * 2006-06-01 2009-11-12 ビアンカメッド リミテッド Apparatus, system, and method for monitoring physiological signs
JP2014151010A (en) * 2013-02-08 2014-08-25 Toyota Motor Corp Biological information acquisition apparatus and biological information acquisition method
JP2017093760A (en) * 2015-11-22 2017-06-01 国立大学法人埼玉大学 Device and method for measuring periodic variation interlocking with heart beat

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005110910A (en) * 2003-10-07 2005-04-28 Matsushita Electric Ind Co Ltd Vibration detector and toilet seat device
JP6602657B2 (en) * 2015-12-11 2019-11-06 公立大学法人首都大学東京 Biological information acquisition apparatus, biological information acquisition program, and biological information acquisition method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003235813A (en) * 2002-12-19 2003-08-26 Matsushita Electric Ind Co Ltd Monitoring device
JP2009538720A (en) * 2006-06-01 2009-11-12 ビアンカメッド リミテッド Apparatus, system, and method for monitoring physiological signs
JP2014151010A (en) * 2013-02-08 2014-08-25 Toyota Motor Corp Biological information acquisition apparatus and biological information acquisition method
JP2017093760A (en) * 2015-11-22 2017-06-01 国立大学法人埼玉大学 Device and method for measuring periodic variation interlocking with heart beat

Also Published As

Publication number Publication date
JP2019084069A (en) 2019-06-06

Similar Documents

Publication Publication Date Title
US8597196B2 (en) Cardiac signal processing apparatus and cardiac signal processing method
US9240176B2 (en) Active noise control system and method
JP2020531063A5 (en)
WO2018073939A1 (en) Measuring program, measuring method and measuring device
WO2016042697A1 (en) Obstacle detection device
US9398875B2 (en) Method and system for biological signal analysis
ATE312557T1 (en) METHOD FOR DETERMINING TISSUE DEFORMATION AND TISSUE VELOCITY VECTORS USING ULTRASONIC IMAGING
JP5919722B2 (en) Biological signal estimation apparatus and program
JP2011200271A (en) Pulse wave detector
WO2019093132A1 (en) Signal detecting system
JP2019201698A (en) Living body detection device
JP2016220816A (en) Biological information detection method for vehicle and device thereof
JP2018117740A (en) Biological information detection apparatus
CN107086035A (en) A kind of speech prompting device monitored based on internal car noise
CN111345799A (en) Vital sign measuring method and device
US10011176B2 (en) Method and device for recognising the condition of vehicle occupants
JP2020146235A (en) Biological information output device, biological information output method, biological information output program, and recording medium
CN117794452A (en) Method for diagnosing a disease, preferably a respiratory disease
WO2018167897A1 (en) Biometric information detection device and biometric information detection program
JP7226152B2 (en) Measurement system for calculating human RRI
KR20220104444A (en) Apparatus for recognizing passenger based on radar and method thereof
US20060251269A1 (en) Anatomy data-collection with low-frequency noise-cancellation capabililty
Sim et al. Advanced automobile crash detection by acoustic methods
KR20170079508A (en) Apparatus and method for acquiring biometric information of driver or passenger
WO2020195899A1 (en) Biological information detection device

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: 18876951

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18876951

Country of ref document: EP

Kind code of ref document: A1