United States Patent [19]
`Aoshima et al.
`
`US006099478A
`[11] Patent Number:
`[45] Date of Patent:
`
`6,099,478
`Aug. 8, 2000
`
`[54] PULSE COUNTER AND PULSE DISPLAY
`METHOD
`
`5,795,300
`5,807,267
`
`.... .. 600/500
`8/1998 Bryars etal. ..
`9/1998 Bryars et al. ......................... .. 600/500
`
`[75] Inventors: Ichiro Aoshima; Tsukasa Kosuda,
`both of SuWa’Japan
`
`[73] Assignees: Seiko Epson Corporation, Tokyo;
`SelkO Instruments, IIlC-, Chlba, DOlh Of
`Japan
`[21] Appl' NO‘:
`09/180,727
`
`FOREIGN PATENT DOCUMENTS
`60-259239 12/1985 Japan.
`4-75503
`7/1992 Japan.
`6_245912 9/1994 Japan _
`7_148127 6/1995 Japan _
`7-227383
`8/1995 Japan .
`8-289876 11/1996 Japan .
`
`[22] PCT Flled:
`[86] PCT NO;
`
`Mar‘ 18’ 1998
`PCT/JP98/01142
`
`§ 371 Date:
`
`Nov. 12, 1998
`
`§ 102(e) Date: Nov. 12, 1998
`
`[87] PCT Pub N05 W098/41142
`PCT Pub Date, Sep_ 24’ 1998
`
`Foreign Application Priority Data
`[30]
`Mar. 18, 1997
`[JP]
`Japan .................................. .. 9-064991
`
`[51]
`
`Int. Cl.7 .............................................. .. A61B 5/02
`
`[52] US. Cl. . . . . . . . . . . . . . . . .
`. . . .. 600/500; 600/502
`[58] Field of Search ................................... .. 600/500, 502,
`600/503, 501, 481
`
`[56]
`
`References Cited
`
`U.S. PATENT DOCUMENTS
`
`Primary Examiner—Cary O’Connor
`Assistant Examiner—Navin Natnithithadha
`Attorney, Agent, or Firm—Mark P. Watson
`
`[57]
`
`ABSTRACT
`
`Apulse counter is provided in Which the value displayed for
`the detected value is highly reliable. An SN condition
`detecting means detects the SN condition of a pulse Wave
`signal (step S201). The SN condition detecting means then
`determines Whether or not the detected SN condition is good
`based on a Speci?c threshold Value (Step S202)‘ When a
`determination is made that the SN condition is good, a
`display control signal to display the pulse rate on a display
`
`means is Output to a display method Switching means (step
`$203) Conversely, When a determination is made in step
`S202 that the SN condition is not good, the pulse rate is not
`displayed on the display means, but rather a display control
`signal indicating that no information at all be displayed is
`output to the display method switching means (step S204).
`
`5,697,374 12/1997 Odagiri et al. ........................ .. 600/500
`
`16 Claims, 10 Drawing Sheets
`
`jIOI
`
`jIOS
`
`5105
`
`£108
`
`PULSE WAVE
`DETECTING
`MEANS
`
`FIRST
`‘ CALCULATING
`MEANS
`
`PULSE WAVE
`> EXTRACT I NC
`J~>
`MEANS
`
`DISPLAY
`MEANS
`
`>
`
`jlOZ
`
`BODY MOTION
`DETECTING
`MEANS
`
`£104
`
`SECOND
`> CALCULATING
`MEANS
`
`5106
`
`> SN CONDITION
`‘ DETECTING
`MEANS
`
`jIO7
`
`DISPLAY METHOD
`> SWITCHING
`MEANS
`
`TomTom Exhibit 1007, Page 1 of 19
`
`

`

`U.S. Patent
`
`Aug. 8,2000
`
`Sheet 1 0f 10
`
`6,099,478
`
`TomTom Exhibit 1007, Page 2 of 19
`
`

`

`U.S. Patent
`
`Aug. 8,2000
`
`Sheet 2 0f 10
`
`6,099,478
`
`mow
`
`A
`
`mom
`
`A
`
`mom
`
`A
`
`N .@E
`
`mom
`
`A
`
`m GI
`
`Pow
`
`w
`
`Em
`
`W
`
`TomTom Exhibit 1007, Page 3 of 19
`
`

`

`U.S. Patent
`
`Aug. 8,2000
`
`Sheet 3 0f 10
`
`6,099,478
`
`FIG. 4
`
`TomTom Exhibit 1007, Page 4 of 19
`
`

`

`U.S. Patent
`
`Aug. 8,2000
`
`Sheet 4 0f 10
`
`6,099,478
`
`FIG. 5
`
`I
`
`S201
`
`SN CONDITION OF PULSE ;
`WAVE SIGNAL DETECTED
`
`SN CONDITION GOOD?
`
`‘ YES
`
`DISPLAY CONTROL
`SIGNAL SET TO
`"DISPLAY PULSE"
`
`S204
`
`DISPLAY CONTROL
`SIGNAL SET T0
`"N0 DISPLAY OF' PULSE"
`
`FIG. 6
`S301
`
`E
`
`CONDITION OF I
`ODY MOTION
`SIGNAL DETECTED
`
`SN CONDITION GOOD’?
`
`S304
`
`YES
`
`S303
`
`I
`DISPLAY CONTROL
`SIGNAL SET TO
`TCH"
`"DISPLAY PI
`
`ERMINED
`PITCH
`E NO
`TO
`PERI ODICI TY
`I
`AY CON
`ET
`OF P I TCH"
`
`DI
`S
`"NO D I SP
`
`S305
`
`TomTom Exhibit 1007, Page 5 of 19
`
`

`

`U.S. Patent
`
`Aug. 8,2000
`
`Sheet 5 0f 10
`
`6,099,478
`
`FIG. 7
`
`FIG. 8
`
`
`
`m2: mw<m Lo mwmzsz
`
`96 120160 240 PITCH
`
`TomTom Exhibit 1007, Page 6 of 19
`
`

`

`U.S. Patent
`
`Aug. 8,2000
`
`Sheet 6 0f 10
`
`6,099,478
`
`FIG. 9
`
`PITCH
`
`~96
`
`97~12O 121 ~16O 160~240
`
`x
`
`16
`
`13
`
`1O
`
`7
`
`FIG. 10
`
`Pmax
`2 \1
`
`7 5
`
`L“ \
`
`TomTom Exhibit 1007, Page 7 of 19
`
`

`

`U.S. Patent
`
`Aug. 8,2000
`
`Sheet 7 0f 10
`
`6,099,478
`
`FIG. 1 1
`
`I
`
`S401
`
`1 S402
`
`PULSE WAVE, BODY /(
`MOTION DETECTION
`I
`A/D CONVERSION
`I
`S403
`FFT PROCESSING //
`+
`5 S404
`SN CONDITION DETECTED
`
`SN CONDITION GOOD?
`
`S405
`
`N0
`8406
`
`YES 1
`
`PULSE WAVE COMPONENT
`EXTRACTED FROM RESULTS
`OF FFT PROCESSING OF
`PULSE WAVE AND BODY
`MOTION
`
`L
`
`S409
`
`j
`
`PULSE WAVE COMPONENT
`EXTRACTED FROM RES
`.
`0F FFT PROC
`NG
`PULSE WAV
`LY
`
`I
`
`S407
`
`S409
`
`PULSE RATE CALCULATED
`
`"
`
`PULSE RATE DISPLAYED
`
`@3
`
`TomTom Exhibit 1007, Page 8 of 19
`
`

`

`U.S. Patent
`
`Aug. 8,2000
`FIG. 12
`
`Sheet 8 0f 10
`
`6,099,478
`
`1001
`
`\K
`
`1002
`
`\K
`
`PULSE SENSOR
`
`FREQUENCY ANALYZER
`
`1003 \K
`BODY MOTIONQSENSOR
`
`F|G.13A M81 FREQUENCY
`
`FIG. 135
`B.
`
`/\
`
`————> FREQUENCY
`
`n
`
`AZ
`
`TomTom Exhibit 1007, Page 9 of 19
`
`

`

`U.S. Patent
`
`Aug. 8,2000
`
`Sheet 9 0f 10
`
`6,099,478
`
`W
`
`W
`
`Q5 Q5“
`
`
`
`Magi; *1 KEEEOO Q2
`
`
`
`Alllll wzzéw 528%; Al
`
`25M; 22 52 E8 2262 E3. 2252 E3
`, 526 Al
`
`
`. m5 26 SSW 6 25% NONFW
`
`
`
`555d , ?ling,‘ i755 mowzww
`
`NEW :SW
`
`Al
`
`3‘ .@_u_
`
`
`
`82 wow 3 SN V g 3 SN V
`
`
`
`
`E1 Li 52%; M53 Al M22 M23 . w><>> m2:
`
`
`
`
`:32; wzzéw “WEE? 322m “82%
`
`
`
`
`
`:32; @2555 102358
`
`TomTom Exhibit 1007, Page 10 of 19
`
`

`

`U.S. Patent
`
`Sheet 10 0f 10
`
`6,099,478
`
`Aug. 8,2000
`FIG. 1 5
`
`@
`
`BODY MOT I ON DETECTION
`
`S1302
`
`BODY MOTION
`PRESENT?
`
`PULSE WAVE DETECTION
`
`S1308
`
`S1303
`5
`
`BODY MOTION
`IGNAL PRESENT’?
`J81 309
`
`NO
`
`PULSE WAVE
`RECTANGULAR SHAPING
`
`PULSE WAVE DETECTION,
`A/D CONVERSION
`+
`
`BODY MOT I ON DETECTION,
`A/D CONVERSION
`+
`
`1
`$1304
`$1305
`
`FFT PROCESSING
`S1306
`I
`PULSE WAVE COMPONENT }
`EXTRACTION
`
`I
`
`I
`
`PULSE RATE CALCULATION
`
`§SI3IO
`
`25131 1
`
`I
`PULSE RATE DISPLAY
`
`@
`
`TomTom Exhibit 1007, Page 11 of 19
`
`

`

`6,099,478
`
`1
`PULSE COUNTER AND PULSE DISPLAY
`METHOD
`
`TECHNICAL FIELD
`
`The present invention relates to a pulse counter for
`monitoring the amount of exercise performed by a user and
`the user’s physical condition, and more speci?cally, to a
`pulse counter Which measures the pulse rate With a high
`degree of accuracy regardless of Whether the user is at rest
`or exercising.
`
`BACKGROUND OF THE INVENTION
`
`2
`Waveform shaping circuit 1206 shapes the pulse Wave
`signal, and outputs it to CPU 1208. Body motion Waveform
`shaping circuit 1207 shapes the body motion signal and
`outputs it to CPU 1208.
`FIG. 15 is a How chart shoWing the operation of the pulse
`counter shoWn in FIG. 14. As shoWn in the How chart, the
`presence or absence of the body motion signal is con?rmed,
`the pulse Wave calculating method is sWitched, and the pulse
`rate is calculated and displayed. In FIGS. 14 and 15, CPU
`1208 con?rms Whether or not a body motion signal is
`present based on the signal output from body motion Wave
`form shaping circuit 1207, and sWitches the calculating
`method (step S1302). During the time that the body motion
`signal is being con?rmed, the pulse Wave signal and the
`body motion signal, Which Were converted from analog to
`digital signals (step S1303 and step S1304, respectively), are
`subjected to a fast Fourier transform (FFT hereinafter) (step
`S1305), and the pulse Wave frequency component is
`extracted (step S1306).
`When a body motion signal cannot be con?rmed, the
`pulse Wave is detected (step S1307), and the pulse Waveform
`is subjected to rectangular Wave conversion processing (step
`S1309). During this interval, CPU 1208 once again con?rms
`Whether or not body motion Was present (step S1308). When
`body motion is not present, then the pulse rate is calculated
`from the rectangular Wave Without modi?cation (step
`S1310). Because A/D conversion of the pulse Wave and
`body motion are not necessary in this case, operation of A/D
`converting circuit 1205 is halted, as is the operation of
`multiplier 1210, Which is required for FFT processing.
`Processing in CPU 1208 necessary for extracting the pulse
`Wave is also halted. Thus, total consumption of electrical
`poWer can be reduced.
`When body motion is present in step S1308, frequency
`analysis is performed using FFT processing (step S1305),
`and the pulse rate is calculated from the extracted pulse
`Wave component (step S1310).
`An exercise pitch measurer can also be formed having the
`same structure as the pulse counter shoWn in FIGS. 14 and
`15. The frequency component of the exercise pitch is
`speci?ed by body motion Waveform shaping circuit 1207.
`This exercise pitch measurer can be used to inform the user
`of his running pitch, Which is useful information for a
`runner. In addition, the distance of running can also be
`obtained from the running pitch and the stride length. An
`exercise pitch measurer and a pulse counter such as shoWn
`in FIGS. 14 and 15 have been disclosed in Japanese Patent
`Application First Publication No. Hei 7-227383, for
`example.
`HoWever, in the above-described conventional pulse
`counters, When a body motion signal is present such as
`shoWn in FIGS. 14 and 15 (steps S1302, S1308), then body
`motion detection is carried out all the time, and pulse Wave
`component extraction processing (step S1306) to exclude
`the body motion spectrum from the pulse Wave spectrum,
`and body motion pitch display (step S1407) are carried out.
`As a result, a body motion pitch is displayed even during
`exercise Which does not have a periodicity, such as in the
`case of gymnastics. Moreover, conventional pulse counters
`perform pulse Wave component extraction processing (step
`S1306) and display the body motion pitch (step S1407)
`regardless of Whether or not noise is present in the detected
`body motion signal. As a result, an incorrect value may be
`detected for the body motion pitch, so that the body motion
`pitch and pulse rate displayed are less reliable.
`In addition, conventional pulse counters perform the
`display of the pulse rate based on the pulse Wave signal
`
`If an individual is able to measure his pulse rate While
`jogging or running a marathon, for example, then he is able
`to monitor the amount of exercise performed as Well as his
`physical condition (i.e., to avoid placing himself at risk).
`Accordingly, for this purpose, a portable pulse counter
`Which can measure the user’s pulse rate by being af?xed to
`his arm is proposed. This type of portable pulse counter
`employs a photoelectric sensor to measure the pulse signal.
`The pulse rate is then determined by extracting the signal
`corresponding to the pulse from the pulse Wave signal.
`HoWever, a pulse Wave signal obtained While the user is
`jogging includes a signal component generated by the user’s
`body motion. Accordingly, it is not possible to extract the
`signal corresponding to the pulse Without some correction
`for the body motion signal.
`FIG. 12 is a block diagram shoWing an example of a
`conventional pulse counter disclosed in Japanese Patent
`Application First Publication No. Sho 60-259239. In addi
`tion to having a pulse Wave detecting sensor 101, this pulse
`counter has a body motion detecting sensor 1003. The
`signals obtained from both of these sensors are subjected to
`frequency analysis at frequency analyZer 1002.
`As shoWn in FIG. 13A, frequency analyZer 1002 converts
`the pulse Wave signal detected by pulse Wave detecting
`sensor 1001 to the spectrum expressed by Waveform In As
`shoWn in FIG. 13B, frequency analyZer 1002 converts the
`body motion signal detected by body motion detecting
`sensor 1003 to the spectrum expressed by Waveform n. Here,
`Waveform n is the result obtained after frequency analysis of
`the signal detected by body motion detecting sensor 1003.
`Accordingly, peak value B‘ expressing the fundamental
`Wave component thereof represents the fundamental fre
`quency of the body’s oscillation. Thus, if the frequency of
`peak B‘ and the frequency of peak B in Waveform m
`coincide, then the peak value B in Waveform m is deemed
`to be the Waveform due to the body’s oscillation. The peak
`obtained by excluding peak value B from Waveform m, i.e.,
`peak A, the Waveform corresponding to the pulse Wave, can
`be read out.
`FIG. 14 is a functional block diagram shoWing the struc
`ture of another conventional pulse counter. Pulse Wave
`sensor 1201 detects the pulse Wave in the body, and outputs
`the detected pulse Wave signal to pulse Wave signal ampli
`fying circuit 1203. Body motion sensor 1202 detects body
`motion, and outputs the detected body motion signal to body
`motion signal amplifying circuit 1204.
`Pulse Wave signal amplifying circuit 1203 ampli?es the
`pulse Wave signal, and outputs it to A/D converter 1205 and
`pulse Waveform shaping circuit 1206. Body motion signal
`amplifying circuit 1204 ampli?es the body motion signal,
`and outputs it to A/D converter 1205 and body motion
`Waveform shaping circuit 1207. A/D converter 1205 con
`verts the pulse Wave signal and body motion signal from
`analog to digital and outputs the result to CPU 1208. Pulse
`
`15
`
`25
`
`35
`
`45
`
`55
`
`65
`
`TomTom Exhibit 1007, Page 12 of 19
`
`

`

`6,099,478
`
`3
`Which is being detected at all times, irrespective of Whether
`or not noise is present in the detected pulse Wave signal (step
`S1303). Accordingly, When the user performs an irregular
`action, such as sudden action of the hand, the noise com
`ponent in the spectrum of the pulse Wave detection signal
`increases, increasing the probability of incorrect detection of
`the pulse rate. Thus, the reliability of the displayed value for
`the pulse rate falls.
`
`4
`FIG. 12 is a block diagram shoWing an example of a
`conventional pulse counter.
`FIG. 13 is an explanatory ?gure shoWing an overvieW of
`the operation for extracting the pulse Wave.
`FIG. 14 is a block diagram shoWing the structure of
`another conventional pulse counter.
`FIG. 15 is a How chart shoWing the operation of the pulse
`counter shoWn in FIG. 14.
`
`SUMMARY OF THE INVENTION
`
`The present invention Was conceived in consideration of
`the aforementioned circumstances, and has as its objective
`the provision of a pulse counter in Which the displayed value
`for detected results is highly reliable.
`In order to resolve the aforementioned problems, the
`present invention is provided With a ?rst calculating means
`for carrying out frequency analysis of the pulse Wave signal
`detected by the pulse Wave detecting means; a second
`calculating means for carrying out frequency analysis of the
`body motion signal detected by the body motion detecting
`means; a pulse Wave extracting means for calculating the
`pulse rate by extracting the frequency of the pulse based on
`the results of frequency analysis by the ?rst and second
`calculating means; a display means for displaying various
`information including at least the output of the pulse Wave
`extracting means; a SN condition detecting means for deter
`mining Whether or not a noise component exceeding a
`speci?c value is included in at least one of the results
`obtained from frequency analysis by the ?rst calculating
`means and frequency analysis by the second calculating
`means; and a display method sWitching means for sWitching
`the display details in the display means in response to results
`detected by the SN condition detecting means.
`
`BRIEF DESCRIPTION OF DRAWINGS
`
`FIG. 1 is a block diagram shoWing the structure of the
`pulse counter according to one embodiment of the present
`invention.
`FIG. 2 is a block diagram shoWing the details of the ?rst
`calculating means in FIG. 1.
`FIG. 3 is a block diagram shoWing the details of the
`second calculating means in FIG. 1.
`FIG. 4 is an explanatory ?gure shoWing an example of the
`mechanical structure of the pulse counter according to an
`embodiment of the present invention.
`FIG. 5 is a How chart shoWing the operation of the SN
`condition detecting means in FIG. 1.
`FIG. 6 is a How chart shoWing another operation of the SN
`condition detecting means in FIG. 1.
`FIG. 7 is an explanatory ?gure shoWing the method for
`specifying the noise base line spectrum N from the body
`motion signal spectrum.
`FIG. 8 is an explanatory ?gure shoWing the relationship
`betWeen the body motion pitch number and the number of
`the base line spectrum in the body motion Waveform spec
`trum thereof.
`FIG. 9 is an explanatory ?gure shoWing a speci?c
`example of the noise base line spectrum in response to the
`pitch number.
`FIG. 10 is an explanatory ?gure shoWing an alternative
`method for specifying the noise base line spectrum N from
`the body motion signal spectrum.
`FIG. 11 is a How chart shoWing the operation of the pulse
`counter according to a modi?cation of the present invention.
`
`10
`
`15
`
`25
`
`35
`
`45
`
`55
`
`65
`
`BEST MODE FOR CARRYING OUT THE
`INVENTION
`
`Preferred embodiments of the present invention Will noW
`be explained With reference to the ?gures.
`A: Structure of the Embodiment
`FIG. 1 is a functional block diagram shoWing one
`example of a representative structure of the present inven
`tion. The pulse Wave detecting means 101 detects the pulse
`Wave in the user’s body, and outputs the detected pulse Wave
`signal to the ?rst calculating means 103. A pieZoelectric
`microphone, photoelectric sensor or the like may be
`employed as pulse Wave detecting means 101. When a pulse
`Wave detecting means 101 comprising an LED or photo
`transistor is employed, for example, light radiated by the
`LED is re?ected by the blood in the user’s blood vessels, and
`the re?ected light is then received by the photo transistor.
`Because the hemoglobin in blood absorbs light, the amount
`of light re?ected is a function of the blood capacity ?oWing
`through the user’s blood vessels.
`Thus, the pulse Wave is detected by detecting the change
`in the amount of light re?ected by phototransistor. Body
`motion detecting means 102 detects body motion, and
`outputs the detected body motion signal to second calculat
`ing means 104. An acceleration sensor or the like may be
`employed for the body motion detecting means, for
`example. Body motion is detected by attaching this accel
`eration sensor to the user’s arm, for example. First calcu
`lating means 103 performs frequency analysis on the signal
`output from pulse Wave detecting means 101, and second
`calculating means 104 performs frequency analysis on the
`signal output from body motion detecting means 102. The
`frequency analysis by ?rst and second calculating means
`103,104 may employ FFT, for example.
`Pulse Wave extracting means 105 speci?es the pulse Wave
`component corresponding to the frequency of the pulse from
`the output of ?rst calculating means 103 and second calcu
`lating means 104, i.e., from the results of frequency analysis
`of the respective outputs from pulse Wave detecting means
`101 and body motion detecting means 102. In addition,
`pulse Wave extracting means 105 calculates the pulse rate
`per one-minute time interval from the speci?ed pulse Wave
`component. Display means 108 displays the pulse rate
`calculated by pulse Wave extracting means 105.
`Based on the output of ?rst calculating means 103 and the
`results of frequency analysis by second calculating means
`104, SN (Signal/Noise) condition detecting means 106 out
`puts a display control signal to display method sWitching
`means 107 for sWitching the contents of the display shoWn
`on display means 108. Display method sWitching means 107
`sWitches the contents of the display on display means 108
`based on the display control signal output by SN condition
`detecting means 106. For example, SN condition detecting
`means 106 detects the SN condition of the body motion
`signal and the pulse Wave signal detected from the result of
`frequency analysis by ?rst and second calculating means
`103,104. When the SN condition is poor, then the pulse rate
`
`TomTom Exhibit 1007, Page 13 of 19
`
`

`

`6,099,478
`
`5
`and body motion pitch are not displayed on display means
`108. Conversely, When the SN condition is good, then SN
`condition detecting means 106 directs the display of the
`body motion pitch and pulse rate on display means 108.
`FIG. 2 is a block diagram shoWing the details of ?rst
`calculating means 103 shoWn in FIG. 1. Pulse Wave signal
`converting means 201 converts the body’s pulse Wave
`analogue voltage signal detected by pulse Wave detecting
`means 101 to a digital signal, and outputs this result to pulse
`Wave signal storing means 202. Pulse Wave signal storing
`means 202 stores the digitally converted pulse Wave signal.
`Pulse Wave signal calculating means 203 sequentially reads
`out the pulse Wave signal stored in pulse Wave signal storing
`means 202, performs frequency analysis of the pulse Wave
`signal, and outputs these results to pulse Wave extracting
`means 105 and SN condition detecting means 106.
`FIG. 3 is a block diagram shoWing the details of the
`second calculating means 104 shoWn in FIG. 1. Body
`motion signal converting means 301 converts the analogue
`voltage signal, Which Was compared to the siZe of the body
`motion detected by body motion detecting means 102, to a
`digital signal, and outputs this result to body motion signal
`storing means 302. Body motion signal storing means 302
`stores the digitally converted body motion signal. Body
`motion signal calculating means 303 sequentially reads out
`body motion signals stored in body motion signal storing
`means 302, performs frequency analysis on the body motion
`signal, and outputs these results to pulse Wave extracting
`means 105 and SN condition detecting means 106.
`B: Operation of the Embodiment
`(1) Overall Operation
`FIG. 5 is a How chart shoWing the operation of SN
`condition detecting means 106, Which is one characteristic
`feature of the present invention. First, SN condition detect
`ing means 106 detects the SN condition of the pulse Wave
`signal detected by pulse Wave detecting means 101 based on
`the output from ?rst calculating means 103 (step S201). “SN
`condition” as employed here means the amount of noise
`included in the detected signal. The method for detecting the
`SN condition Will noW be explained in detail.
`FIG. 7 is an example of the spectrum for the body motion
`signal detected by body motion detecting means 102, Which
`is obtained When second calculating means 104 performs a
`FFT on the pulse Wave signal. For example, from among the
`various base line spectrums, the base line spectrum N Which
`is the tenth largest spectrum as counted from maximum base
`line spectrum Pmax is designated to represent the base line
`spectrum of the noise component. The reason for selecting
`the tenth largest base line spectrum N to represent the base
`line spectrum of the noise component is that in experiments
`conducted thus far, the tenth base line spectrum has had the
`highest probability of being noise. Further, a determination
`can be made as to Whether the SN condition is good or poor
`by comparing the siZe of maximum base line spectrum Pmax
`and the siZe of base line spectrum N. For example, When the
`value of N/Pmax is larger than a speci?c threshold value,
`then a determination is made that the SN condition is poor.
`Here, N indicates the siZe of base line spectrum N, While
`Pmax indicates the siZe of base line spectrum Pmax. Note
`that While the base line spectrum N Which is tenth largest in
`siZe as counted from maximum base line spectrum Pmax
`Was designated to represent the base line spectrum of the
`noise component in the preceding example, the present
`invention is not limited thereto. For example, other base line
`spectrums, i.e., the seventh base line spectrum for example,
`may be employed to represent the noise component.
`
`10
`
`15
`
`25
`
`35
`
`45
`
`55
`
`65
`
`6
`Thus, in this embodiment, the base line spectrums are
`aligned in order of siZe, and the base line spectrum at a
`speci?c position x from the maximum base line spectrum is
`determined to be the base line spectrum N of the noise
`component. The SN condition can be judged by comparing
`the siZe of the maximum base line spectrum (Which is
`regarded as the signal component) and the siZe of the base
`line spectrum of the noise component. As a result, it is
`possible to quickly and easily detect the SN condition.
`SN condition detecting means 106 determines Whether or
`not the detected SN condition is good based on a speci?c
`threshold value (step S202). When a determination is made
`that the SN condition is good, the display control signal for
`displaying the pulse rate on display means 108 is output to
`display method sWitching means 107 (step S203).
`Conversely, When a determination is made in step S202
`that the SN condition is not good, then the pulse rate is not
`displayed on display means 108, and a display control signal
`indicating that no information at all should be displayed is
`output to display method sWitching means 107 (step S204).
`Note that in step S204, it is also acceptable that SN condition
`detecting means 106 output a display control signal that
`directs a blinking display of the pulse rate extracted and
`calculated by pulse Wave extracting means 105. In addition,
`in step S204, it is also acceptable for SN condition detecting
`means 106 to output a display control signal for displaying
`the pulse rate extracted and calculated by pulse Wave
`extracting means 105, and for displaying an indication that
`the likelihood of error in that display is high. The reason Why
`the pulse rate is not displayed constantly When the SN
`condition is poor is because it is not possible for pulse Wave
`extracting means 105 to accurately extract the pulse Wave
`component in the presence of a poor SN condition, making
`the calculation and display of an accurate pulse rate dif?cult.
`Accordingly, in the pulse counter of the present invention,
`When the SN condition of the detected signal of the pulse
`Wave is poor, i.e., When a great deal of noise is included in
`the detected signal, then the pulse rate is not constantly
`displayed. Conversely, When there is a little noise in the
`signal, the pulse rate is displayed at all times. As a result, the
`reliability of the displayed value for the pulse rate is
`improved.
`C: Other Methods for Detecting SN Condition
`(1) First Detecting Method
`Modi?cations of the above-described method for detect
`ing the SN condition Will noW be explained. FIG. 8 is an
`explanatory ?gure shoWing the relationship betWeen the
`body motion pitch number and the number of the base line
`spectrum of the body motion Waveform spectrum thereof, in
`the case Where there is no noise included in the body motion
`Waveform detected by body motion detecting means 102,
`and provided that, in addition to a side lobe component,
`three base lines are present for the fundamental Wave
`component of body motion, or for one high frequency
`component. FIG. 7 is an explanatory ?gure shoWing the
`relationship betWeen the body motion pitch number and the
`xth largest base line spectrum Which has been designated to
`represent the noise component.
`In this embodiment, the base line spectrum N of the noise
`component in FIG. 7 is speci?ed in response to the body
`motion pitch number calculated by second calculating
`means 104. In FIG. 8, the body motion pitch number is
`plotted along the horiZontal axis, and the number of the base
`line spectrum in the body motion Waveform spectrum When
`noise is not included in the body motion Waveform is shoWn
`along the vertical axis. As shoWn in this ?gure, 15 base line
`
`TomTom Exhibit 1007, Page 14 of 19
`
`

`

`6,099,478
`
`7
`spectrums are generated When the body motion pitch num
`ber is 96 times or less. When the body motion pitch number
`is greater than 96 times but less than 120 times, then 12 base
`line spectrums are generated.
`When the number of base line spectrums included in the
`body motion spectrum obtained by FFT processing by
`second calculating means 104 is greater than the number of
`the base line spectrum derived from FIG. 8, then a deter
`mination is made that noise in an amount corresponding to
`the additional number of base line spectrums is present. For
`example, When the body motion pitch number is 96 times or
`less, then the ?fteen largest base line spectrums are deemed
`to the base line spectrums of the body motion component,
`While base line spectrum from the sixteenth largest and
`beyond are determined to be the noise component.
`Further, as shoWn in FIG. 9, the value obtained by adding
`1 to the number of the base line spectrum obtained from the
`relationship betWeen the number of the base line spectrum of
`the body motion component and the pitch number shoWn in
`FIG. 8 is de?ned as X. Then, the base line spectrums are
`aligned in order of siZe, and the base line spectrum at a
`speci?c position X from the maximum base line spectrum is
`determined to be the base line spectrum N of the noise
`component. The SN condition can be judged by comparing
`the siZe of the maximum base line spectrum Pmax and the
`siZe of the base line spectrum N. For example, When the
`value of N/Pmax is larger than a speci?c threshold value, a
`determination is made that the SN condition is poor.
`Because the base line spectrum N representing the noise
`component in response to the pitch number of the detected
`body motion changes in this SN condition detecting method,
`it is possible to more accurately specify the siZe of the noise
`component. Accordingly, the reliability of the pulse rate and
`body motion pitch displays on display means 108 is
`improved.
`FIG. 6 is a How chart shoWing another operation of SN
`condition detecting means 106, Which is a characteristic
`feature of this embodiment. First, SN condition detecting
`means 106 detects the SN condition of the body motion
`signal detected by body motion detecting means 102 based
`on the output from second calculating means 104 (step
`S301). Then, SN condition detecting means 106 determines
`Whether or not the detected SN condition is good based on
`a speci?c threshold value (step S302). When a determination
`is made that the SN condition is good, a display control
`signal to display the body motion pitch on display means
`108 is output to display method sWitching means 107 (step
`S303). Here, “body motion pitch” indicates the number of
`regular body movements that are generated While the user is
`running, etc. during a one minute interval of time.
`Conversely, When a determination is made that the SN
`condition is not good in step S302, then a determination is
`made that the detected body motion signal does not have a
`periodicity (step S304), the body motion pitch is not dis
`played on the display means 108, and a display control
`signal directing that no information be displayed is output to
`display method sWitching means 107 (step S305). Note that
`it is also acceptable for SN condition detecting means 106 to
`output a display control signal for a blinking display of the
`body motion pitch calculated by pulse Wave extracting
`means 105 in step S305. SN condition detecting means 106
`may also output a display control signal for displaying the
`pulse rate calculated by pulse Wave extracting means 105,
`and for displaying an indication that the probability of an
`error in that display is high. The reason for not displaying the
`body motion pitch constantly When the SN condition is poor
`
`10
`
`15
`
`25
`
`35
`
`45
`
`55
`
`65
`
`8
`is that it is not possible for pulse Wave extracting means 105
`to accurately calculate the body motion pitch When a poor
`SN condition is present, so that the display of an accurate
`body motion pitch is dif?cult.
`As a result, in this embodiment, When the SN condition of
`the detected signal for body motion is poor, i.e., When a great
`deal of noise is included in the detected signal, then the body
`motion pitch is not displayed. Conversely, When there is a
`little noise in the signal, the body motion pitch is displayed
`at all times. As a result, the reliability of the value displayed
`for the body motion pitch is improved.
`(2) Second Detecting Method
`Next, another SN condition detecting method for use in
`SN condition detecting means 106 Will be explained. In this
`detecting method, the method for specifying the base line
`spectrum N Which is regarded to represent the noise com
`ponent differs from the above-described ?rst detecting
`method.
`FIG. 10 shoWs an example of the body motion signal
`spectrum detected by body motion detecting means 102.
`This body motion signal spectrum is obtained by FFT
`processing of the pulse Wave signal by second calculating
`means 104. For example, the various base line spectrums are
`distinguished betWeen so that the position of the maximum
`base line spectrum Pmax is designated as the standard, the
`base line spectrum positioned next to base line spectrum
`Pmax is designated as the second base line spectrum, and the
`base line spectrum positioned ne