JP5249476B1 - Pedometer - Google Patents

Abstract

In a pedometer that provides useful walking-related information such as walking speed, exercise intensity, or calories burned together with the number of steps, the user experienced a problem that the walking-related information did not match known numerical contents. In view of such problems, the present invention relates to a pedometer that has a simple structure and is economically inexpensive, in which provided walking-related information has at least practical accuracy and can further improve accuracy. The issue is to provide
In order to solve the problems, the present invention captures a change in the ratio of the stride height to the walking pitch with a piecewise linear function, and expresses the relationship of the exercise intensity to the walking speed with a piecewise linear function based on known numerical values. Based on the technology that simplifies the walking speed and exercise intensity derived from the above by regression analysis in divided sections. In order to improve accuracy, a technology that easily reflects the walking characteristics of pedometer users in the pedometer is also provided.
[Selection] Figure 3

Description

The present invention relates to a pedometer that measures the number of steps during walking and provides walking-related information such as walking speed, walking distance, exercise intensity, calorie consumption, or fat burning amount together with the number of steps.

The pedometer has a pendulum type or acceleration sensor type as the principle of pedometer measurement, and has many functions such as the basic step count measurement or the calculation of walking related information together with the step count measurement. ing.

Regarding the number of steps measured by the pedometer, an error is defined (± 3%) in JIS (Japanese Industrial Standards JIS S 7200), so an accurate step value can be obtained within the error.

Among the walking-related information provided by the pedometer, the calorie consumption or the amount of fat burning is useful information for users who use the pedometer for health management. However, no standard is defined for the walking related information like the number of steps. And since the pedometer manufacturer's original calculation method is adopted, the accuracy cannot be predicted at present.

The inventor of the present application has been walking (hereinafter, walking is used as a synonym) for daily health improvement. Carry your pedometer with basic data on walking distance and walking time (hereinafter these two quantities are called primitive data), walking-related information such as the number of steps, calories burned, and fat burning provided by the pedometer. Has been recorded as. I set some walking courses in advance and measured the distance on a map. The course distance measured in advance and the time measured during walking were recorded as primitive data.

The inventor of the present application had an opportunity to participate in a briefing session for health promotion held by the city. In response to the explanation of “Exercise Guidelines 2006 for Health Promotion <Exercise Guide 2006>” reported as part of the Ministry of Health, Labor and Welfare's exercise measures, I gained knowledge such as the relationship between quantity and the relationship between walking speed and exercise intensity. Using this knowledge, we calculated calories burned during walking from the original data of walking records, and compared with the calorie consumption display value of the pedometer for analysis. Since the relationship between the above quantities and the relationship between walking speed and exercise intensity are deeply related to the present invention, they will be described in detail before describing the analysis results.

In the <Exercise Guide 2006> reported as part of the Ministry of Health, Labor and Welfare's exercise measures, in addition to “exercise” performed for the purpose of maintaining and improving physical fitness, “life activity” performed in daily life is also regarded as physical activity. The intensity of physical activity, ie, exercise intensity, is expressed in units of mets, and this value corresponds to a multiple of rest (1 mets). For example, normal walking is 3 mets. The exercise (Ex) is a unit representing the amount of physical activity, and is obtained by multiplying the exercise intensity (Mets) of the physical activity by the execution time (hour) of the physical activity. Exercise for 3 minutes of physical activity at 3Mets is 1.5Ex.

Regarding the calculation of calorie consumption, a formula for calculating the amount of energy consumption corresponding to the amount of physical activity is described in <Exercise Guide 2006>. It is proportional to the weight of the person who is performing physical activity, and is calculated by Equation 1 including the rest.

1.05 in Formula 1 is the energy consumption per hour / kg calculated from the oxygen intake at rest (3.5 ml / kg / min) and the energy consumption per 1000 ml of oxygen = 5 kcal. The portion of 1.05 × weight corresponds to the basal metabolic rate per hour. In some cases, 1.05 is roughly 1. In this case, exercise intensity × execution time is directly multiplied by weight. Moreover, the basal metabolic rate per hour calculated | required by the well-known calculation formula from height, weight, sex, and age may be used. Below, it demonstrates using Numerical formula 1 multiplied by 1.05. Moreover, the unit of energy consumption is kcal, and is expressed as calorie consumption below.

Regarding the fat burning amount, the calorie consumption necessary for burning 1 g of fat is 7.2 kcal (may be approximately 7 kcal), and the fat burning amount can be calculated from the calorie consumption.

Next, the relationship between walking speed and exercise intensity will be described. In general, it is said that the walking speed of a person is about 4 km / hour. In <Exercise Guide 2006>, the relationship between walking speed and exercise intensity is shown in stages as follows.
Slow walking: Flat ground, 54 m / min (3.2 km / hr) 2.5 Mets ordinary walking: Flat ground, 67 m / min (4.0 km / hr) 3 Mets walking: Flat ground, 81 m / min (4.8 km / hr) ) 3.3 Mets slightly fast walking: flat ground, 94 m / min (5.6 km / hour) 3.8 Mets rapid walking: flat ground, 95-100 m / min (about 6 km / hour) 4 metres quite fast walking: flat ground, 107 m / min ( 6.4km / h) 5 mets

To calculate calorie consumption from the original data of walking records, calculate the walking speed from the walking distance and walking time of the original data, find the exercise intensity corresponding to the walking speed from the above relationship, and use Equation 1 to consume. Calculate calories. Since <Exercise Guide 2006> only shows the exercise intensity with respect to the walking speed for each stage, the walking speed during that time is calculated by linear interpolation.

It is relatively easy to analyze the accumulated walking record data with a personal computer because a convenient software tool is provided. The inventor of the present application also made use of this to compare and analyze the calorie consumption calculated from Formula 1 with the formula 1 and the calorie consumption display value of the pedometer. As a result of the analysis, compared to the calorie consumption calculated from the original data, it is noticed that the calorie consumption display value of the pedometer is excessive (about 1.5 times or more depending on walking speed), and wonders why. I came to have it.

Here, I will touch on the current status of commercially available pedometers. Pedometers are broadly divided into pendulum type and acceleration sensor type, and provided walking related information is limited to walking distance and calorie consumption, walking distance, walking time, walking speed, exercise intensity, exercise, calorie consumption, Or abundant fat burning amount, etc. Furthermore, as an additional function, memory of walking data (for example, for one week) is possible, walking data display is diverse with numerical display or graph display, with many types of data communication functions, etc. Is commercially available. The more walking-related information or additional functions, the more complicated the structure and the higher the price. A simple structure pedometer with no additional functions and limited walking-related information is inexpensive. In addition, activity meters for daily activities based on pedometer functions (deskwork and other household tasks) and mobile devices such as mobile phones with built-in pedometer functions are beginning to become popular. .

As described above, an error is defined by JIS for the step count measurement value of the pedometer. Because it is a low-priced product, there are no errors and lack of accuracy. However, there is no compliant standard for walking-related information, and the accuracy of the calorie consumption that is of particular interest is completely unpredictable.

Therefore, the manufacturer was inquired about the above-mentioned question about calorie consumption, but the answer was that it was not possible to disclose because of the unique calculation method inside the pedometer. For this reason, the inventor of the present application diligently investigated patent documents.

The main function of the pedometer is to measure the number of steps more accurately, and it does not directly detect the walking speed. It can be easily understood that the walking speed increases as the number of steps measured per unit time increases, and decreases as the number of steps measured decreases. However, it seems that it is not easy to find the walking speed accurately. In fact, as the survey progressed, it turned out that how to accurately determine the walking speed, which is the basis for calculating calories burned, was an important technique. Further, as a result of the investigation, it has been found that there are some problems in the techniques disclosed in Patent Documents 1 to 3 shown below.

Japanese Patent Application Laid-Open No. 2004-133826 calculates a step length during walking from a set height and a walking pitch that is a step count measurement value per unit time, and uses this step length to calculate a walking speed by a step length × walking pitch. It is disclosed. The calorie consumption calculation is based on a kinetic energy formula known in physics, and a technique is disclosed in which walking speed is applied to the speed term of this formula.

In the above-mentioned stride calculation, a regression line based on a regression analysis that has been found by research is used. However, no consideration is given to the pitch range in which the change in the stride is less than the change in the walking pitch described later. As a result, if the walking pitch increases, the step length increases as the walking pitch increases, and as the walking pitch decreases, the step length decreases as it decreases. The reason why it is not taken into account is that there is no description or suggestion that the coefficient calculated in the regression analysis is changed in the walking pitch range.

Regarding calorie consumption calculation, there is a problem that a large difference occurs compared to the value of the calculation formula described in <Exercise Guide 2006>. A numerical calculation example showing this will be shown below. Let's calculate the calorie consumption for 1 hour, assuming the step length: 0.8 m, walking pitch: 125 steps / minute, weight: 65 kg, walking time: 1 hour. The walking speed is 0.8 × 125 × 60 = 6 km / hour, 100 m / minute, and 1.67 m / second. In <Exercise Guide 2006>, when the walking speed is about 6 km / hour, the exercise intensity is 4 mets. Use the 4 Mets, and subtract 1 Met at rest.
Calculation result by Patent Document 1 Calorie consumption (kcal)
= 1/2 * 65 * 1.67 * 1.67 * 3600 / 4.2 / 1000
= 77.7
<Calculation result by the calculation formula described in <Exercise Guide 2006> Calorie consumption (kcal)
= (4-1) × 1 × 1.05 × 65 = 205

In Patent Document 2, there is a method of calculating the walking speed from the walking pitch and height (equation (3) of the same document), and a method of calculating the exercise load from the walking speed and weight (equation (4) of the same document). It is disclosed. The former walking speed calculation method does not take into account the pitch range in which the change in the stride is small with respect to the change in the walking pitch, which will be described later. As a result, there is a problem that an accurate walking speed is not required. The reason for not being considered is that the calculation formula of walking speed is given by the same calculation formula throughout the walking pitch range, and that the correction coefficient is also changed and calculated in the walking pitch range. None of them.

The amount of exercise load based on the latter kinetic energy is considered to be an amount corresponding to calorie consumption. Since it is an amount calculated from the walking speed lacking the accuracy described above, the exercise load also has a problem of lacking accuracy.

Patent Document 3 describes two examples for obtaining exercise intensity, which is an existing technique. Let us examine these two examples of walking. If the step length is 85 cm and the walking pitch is 125 steps / minute, the walking speed is 6.4 km / hour, and the exercise intensity described in <Exercise Guide 2006> is 5 mets. Subtracting 1 Met at rest is 4 Mets.
Calculation result in the first example: −7.065 + 0.105 × 125 = 6.06
Calculation result in second example: 0.015 × 125 × 0.85 + 1.599 = 3.19
The Met value of the calculation result differs depending on the calculation method, and does not match the Met value described in <Exercise Guide 2006>, and it can be said that there is a problem with the existing technology.

Japanese Patent No. 3734429 (Claim 1, Claim 2) Japanese Patent No. 3916228 (Formula (3), Formula (4)) JP 2009-279239 A (paragraphs 0023 and 0024)

New Exercise Standards and Exercise Guidelines “METs Table of Physical Activities” (Germany) Updated by National Institute of Health and Nutrition 2007 Revised “Met's Table of Physical Activities” (National Institute of Health and Nutrition) Revised in 2012

The present invention has been made in view of the conventional problems described above. The subject of the present invention is the walking speed, walking distance, exercise intensity calculated from the measured number of steps and the height or weight of the set pedometer user regardless of the principle of the pendulum type or acceleration sensor type pedometer Alternatively, it is an object of the present invention to provide a technique relating to a pedometer, in which walking-related information such as calorie consumption has at least practical accuracy, can be further improved, and is simple in structure and economical. The practical accuracy will be described later.

In order to solve the above-mentioned problems, the pedometer according to the present invention adopts the following characteristic configuration for the calculation of walking speed and exercise intensity, and the walking distance and consumption based on the calculated walking speed or exercise intensity. Calories and the like are calculated and displayed together with the number of steps.

The walking speed calculation configuration according to claim 3 is measured by the setting means for setting the height of the pedometer user, the step count measuring means for measuring the number of steps when the pedometer user walks, and the step count measuring means. The number of steps per unit time is calculated as a walking pitch, and the result obtained by substituting the walking pitch into a correction function formula defined as a piecewise linear function related to the walking pitch and the height set in the setting means Computing means for calculating the walking speed. Then, the correction function with respect Characteristics indicator quadratic function obtained by multiplying the walking pitch characteristics piecewise linear function for walking pitch defining the gait characteristics, comprising at least one of the plurality of sections of the characteristic piecewise linear function section Is a regression line for each subdivision section obtained by performing regression analysis for each subdivision section divided into a plurality of sections, and the characteristic piecewise linear function is adapted to change the ratio of the stride height to the walking pitch according to the change. In other words, each segment is a piecewise linear function represented by a linear function related to the walking pitch.

Thereby, the calculation formula of the walking speed by the characteristic classification quadratic function can be simplified as required. If the typical value of the walking characteristic is fixed and reflected in the characteristic piecewise linear function related to the walking pitch that determines the walking characteristic, the walking speed is the accuracy of the request unless the difference from the walking characteristic of the pedometer user is large. Thus, it can be calculated relatively easily.

With regard to the configuration of the walking speed calculation according to claim 2, the setting means for setting the height of the pedometer user, the step count measuring means for measuring the number of steps when the pedometer user walks, and the step count measuring means The number of steps per unit time is calculated as a walking pitch, and the result obtained by substituting the walking pitch into a correction function formula defined as a piecewise linear function related to the walking pitch and the height set in the setting means And calculating means for calculating the walking speed. The straight line of slope and intercept of each section defining the characteristic of the correction function, as well as division points between the adjacent sections can be further set in said setting means. Then, the straight line of slope and intercept of each of the sections, as well as division point between adjacent intervals, Characteristics indicator secondary obtained by multiplying the walking pitch characteristics piecewise linear function for walking pitch defining the gait characteristic of the pedometer user For the function, the slope and intercept of the regression line for each subdivision section obtained by performing regression analysis for each subdivision section obtained by dividing a section including at least one of the plurality of sections of the characteristic piecewise linear function, In addition, the characteristic categorizing linear function is a division point with respect to the walking pitch that determines the walking characteristics of the pedometer user for each section that is divided into a plurality of sections according to the change. It is a piecewise linear function represented by a linear function related to the walking pitch.

As a result, the correction function relating to the calculation of the walking speed is determined as a relatively simple piecewise linear function, so that the walking speed can be calculated relatively easily. Furthermore, the walking characteristic value of the pedometer user can be set and reflected. Therefore, there is no error due to the deviation of the walking characteristics, and the walking speed can be calculated relatively easily with the required accuracy within the error due to the simplification of the subdivision of the arithmetic expression.

With regard to the structure of the walking speed calculation according to claim 1, the setting means for setting the height of the pedometer user, the step count measuring means for measuring the number of steps when the pedometer user walks, and the step count measuring means Walking based on the result obtained by calculating the number of steps per unit time as a walking pitch and substituting the walking pitch into a correction function formula defined as a straight line related to the walking pitch and the height set in the setting means Computing means for calculating the speed. A slope and an intercept that define the characteristics of the correction function can be further set in the setting means. Then, the slope and intercept, compared Characteristics indicator quadratic function obtained by multiplying the walking pitch characteristics piecewise linear function for walking pitch defining the gait characteristic of the pedometer users, plurality of sections of the characteristic piecewise linear function Is a slope and intercept of one regression line obtained by performing a regression analysis in a section including at least one of the above, and the characteristic piecewise linear function is a step height with respect to a walking pitch that defines a walking characteristic of the pedometer user . It is a piecewise linear function expressed by a linear function related to a walking pitch for each section in which a change in the ratio is divided into a plurality of sections according to the change.

As a result, the correction function related to the calculation of the walking speed is determined as the simplest straight line, so that the walking speed can be easily calculated. Furthermore, the walking characteristic value of the pedometer user can be set and reflected. Therefore, there is no error due to the deviation of the walking characteristics, and the walking speed can be calculated accurately and easily within the error due to the simplification of the calculation formula.

Regarding the configuration of the exercise intensity calculation according to claim 4, the configuration of the walking speed calculation according to any one of claims 1 to 3 further includes the following characteristic configuration. That is, the calculation means further calculates a result obtained by substituting the walking speed into an expression of an exercise intensity function relating to the walking speed as an exercise intensity. The exercise intensity function is a function obtained by performing the first operation or the second operation on the current piecewise linear function in which the relationship of the exercise intensity to the walking speed is expressed, and the first operation is , Replacing the current piecewise linear function with a new new piecewise linear function composed of a plurality of regression lines obtained by performing regression analysis for each section divided into a plurality of sections including all sections of the current piecewise linear function , and the second operation, said current piece-wise linear function or a continuous section of the new piecewise linear function, wherein when the inclination is gradually nondecreasing linear function of a continuous section, the linear function of the continuous section replaced with a new one section to the minimum value function for selecting the minimum of the function values when the replaced new one section to the maximum value function to elect the maximum value of the function value, gradual nonincreasing, the Either Time, wherein the subtracting from the continuous sections one section is an operation of replacement can eliminate repeated until the second operation.

Thereby, with respect to the calculation formula of the exercise intensity representing the relationship between the walking speed and the exercise intensity, the calculation expression by the first operation is simplified or the calculation process by the second operation is simplified as required. I can do it. If accuracy is required, it is desirable to limit the first operation. Further, no error occurs in the second operation. The exercise intensity is calculated by applying the obtained walking speed to the exercise intensity function replaced by the first or second operation. Since the calculation result of exercise intensity includes a calculation error of walking speed, it is necessary to verify the total error. Details will be described later.

With regard to the configuration according to claim 5 for calculating the walking-related information such as walking speed, walking distance, exercise intensity, or calorie consumption and displaying it together with the number of steps, the following configuration is further added to the configuration according to claim 4. Prepare. That is, the computing means further calculates the walking distance and exercise by multiplying the walking speed and the exercise intensity by walking time, calculates the calorie consumption by multiplying the exercise by the basal metabolic rate, and the step counting means Display means for displaying the walking speed, walking distance, exercise intensity, exercise, or calorie consumption calculated by the computing means together with the number of steps measured in step (b). Then, the basal metabolic rate is set in addition to the height set in the setting means, the value of the weight of the pedometer user set, the value obtained by multiplying the weight by a factor of 1.05, or the setting means In addition, the pedometer user is set to any one of values calculated from the sex and age of the pedometer user and the height and the weight.

Thus, walking-related information such as walking speed, walking distance, exercise intensity, or calorie consumption is calculated by calculating walking distance, calorie consumption, etc. from the two amounts of walking speed and exercise intensity determined according to the accuracy of the request. Can be displayed along with the number of steps. The accuracy of these two quantities is directly reflected in the value of the walking related information.

The means for solving the problem has been described above. By taking this measure, the walking speed, exercise intensity, and walking-related information can be obtained with desired accuracy as follows. As will be described later, attention was paid to the fact that the walking characteristics in the intermediate pitch interval, the lower pitch interval and the upper pitch interval in which the step change is reduced can be expressed by a piecewise linear function. The walking speed can be obtained more easily by calculating the walking speed calculated based on the piecewise linear function with a regression analysis while maintaining practical accuracy. In addition, the relationship of exercise intensity to walking speed is expressed as a piecewise linear function based on known numerical values, and this is simplified by maintaining the practical accuracy by regression analysis. I can do it. Simplification can be made depending on the degree of accuracy required. Furthermore, the configuration in which the walking characteristics can be set has a greater accuracy because the walking characteristics of the pedometer user are directly reflected than the configuration in which the walking characteristics are fixed to the representative value. By taking the above measures, the problems of the present invention are solved, and walking-related information such as walking speed, walking distance, exercise intensity, or calorie consumption provided with the number of steps has at least practical accuracy. Further, it is possible to provide a pedometer that can further improve accuracy and that is simple in structure and economical.

It is the data and graph showing the relationship between walking speed and exercise intensity. It is the data and graph showing the result of the walk experiment performed in order to confirm the relationship between a walk pitch and a step. It is a block diagram showing the structure of the pedometer in the Example concerning this invention.

How to determine the walking speed from the measured number of steps, the exercise intensity that matches the <Exercise Guide 2006> or other known values from the walking speed with practical accuracy, and how to determine the accuracy Increasing is an important technique of the present invention, and will be described in detail from the latter exercise intensity. Before we elaborate, we refer to practical accuracy.

The error in the number of steps measured by the pedometer is defined as ± 3% in the JIS standard. Within the response range of the pedometer (described later), an accurate step value within the standard error can be obtained without being substantially affected by the walking characteristics (how to walk) of the pedometer user. On the other hand, since the walking related information is greatly influenced by the walking characteristics related to the body data such as the height or weight of the pedometer user, the stride or the walking pitch, it is not easy to define an error in the standard. However, it is problematic that walking-related information that does not conform to standards such as calories consumed provided with the number of steps is a calculation method originally developed by the pedometer manufacturer, and is far from known numbers. It is. Pedometer users should believe and use gait-related information to generally match known values. Therefore, the accuracy of walking-related information alone according to the calculation method of the present invention that does not include a step count measurement error (± 3%) is set to an error of about ± 5 to 10%. The reference value for evaluating this error is a value based on a function derived by the present invention based on a known numerical value. In the present invention, this error is referred to as practical error or practical accuracy.

<Exercise Guide 2006> does not show the relationship between the walking speed and the exercise intensity related to walking exceeding “much fast walking” of 107 m / min. Non-Patent Document 1 shows exercise intensity, that is, Met's value for various physical activities. The following data are shown for walking motion. This mets value is a value including one mets at rest.
53 m / min (3.2 km / h) flat, slow 2.5 mets 67 m / min (4.0 km / h) Hard and stable flat 3.0 mets 80 m / min (4.8 km / h) flat, moderate speed 3.3 mets 93m / min (5.6km / h) flat ground, speed as a motion 3.8 mets 107m / min (6.4km / hr) flat, refreshing speed 5.0 mets 120m / min (7.2km) / Hour) Flat land, very crisp speed 6.3 mets 133m / min (8.0km / hr) 8.0 mets

The above data fits well with the data from the <Exercise Guide 2006> and a lot of relevant information on the Internet. Considering that <Exercise Guide 2006> is being revised and that Non-Patent Document 1 was recently revised, the following data (revised) of Non-Patent Document 2, which is a revised version of Non-Patent Document 1. (Referred to as data) is the basis for the relationship between walking speed and exercise intensity in the present invention. The data of <Exercise Guide 2006>, the data of Non-Patent Document 1, and the data of Non-Patent Document 2 are well-known data published by an authoritative public institution and are sufficient for the basis. The Mets value of the revised data is also a value including one Mets at rest.
53m / min (3.2km / h) Slow, flat and hard ground 2.8Met 67m / min (4.0km / h) Flat and hard ground 3.0Met 80m / min (4.8km / h) Speed, flat and hard ground 3.5 mets 93m / min (5.6km / h) fast, flat and hard ground, walking for exercise purpose 4.3 mets 107m / min (6.4km / h) flat and hard ground, Very fast 5.0 metres 120m / min (7.2km / h) flat and hard ground, very fast 7.0 mets 133m / min (8.0km / hr) flat and hard ground 8.3 mets

The Mets value of the revised data has been revised upward by 0.3 / 0 / 0.2 / 0.5 / 0 / 0.7 / 0.3 Mets in order from the slower speed from the Mets value before the revision. . The Mets value increases by a maximum of 0.5 / 3.8≈13% with respect to the walking speed. However, due to the increase in the Met's value due to this upward revision, the problem of the excessive display of the calorie consumption of the pedometer and the problem of the prior art found in the patent literature survey described above based on the contents of the <Exercise Guide 2006> Will not be resolved.

FIG. 1 is a so-called piecewise linear function in which the horizontal axis is walking speed (km / hour), the vertical axis is exercise intensity (mets), and the revised data is plotted and connected by a straight line. The relationship between walking speed and exercise intensity is considered to be close to a quadratic function piecewise. However, determining this function and determining the exercise intensity is too complicated. Therefore, this continuous piecewise linear function is used as a basic (reference) function representing the relationship between walking speed and exercise intensity according to the present invention. As can be seen from FIG. 1, the number of sections is not small, and it is considered that there is no problem in practical use of this piecewise linear function. The reason why “continuous” is given above is to distinguish it from a discontinuous piecewise linear function described later.

It is natural that the exercise intensity can be accurately obtained by substituting the walking speed into the piecewise linear function formula. Specifically, it is first determined which of the six sections the walking speed is in, and the exercise intensity is obtained by substituting the walking speed into an equation of a straight line (linear function) in the section. This ordinary calculation procedure requires the same number of six straight lines as a section and a maximum of three section determinations. Obviously, the procedure becomes simpler if the number of intervals is reduced. The problem of the present invention is that it has at least practical accuracy, can further improve accuracy, and has a simple structure and is economically inexpensive. Simplification of the procedure, in other words, simplification of the operation structure is an effective means for solving the latter problem. Before explaining the method of simplifying by reducing the number of sections, the concept of reducing the number of section determinations will be described.

The section determination is a complicated process in which the branched path is reached to the target section while comparing the walking speed. The process is simplified by reducing the number of interval judgments. A continuous piecewise linear function shows a concave characteristic in a continuous section if the slope of the straight line for each section is gradually non-decreasing, and shows a convex characteristic if the slope is non-increasing gradually. Then, in the calculation of the function value of the piecewise linear function in the continuous interval, if the concave characteristic is in the continuous interval, a function that selects the maximum value of the linear expression value for each interval (in the present invention, the maximum value function and And a function that selects the minimum value of the linear expression for each interval (referred to as a minimum value function in the present invention) can be used. Therefore, it is not necessary to perform further section determination in continuous sections. This is likely to be easily understood and will not be described in detail.

The six speed sections in FIG. 1 are referred to as a first section, a second section,. The piecewise linear function of FIG. 1 has a concave characteristic from the first section to the third section, a concave characteristic from the fourth section to the fifth section, and the sixth section may be treated as a concave characteristic or a convex characteristic for convenience. good. Alternatively, the first section to the second section are concave characteristics, the third section to the fourth section are convex characteristics, and the fifth section to the sixth section are convex characteristics. In either case, the number of section determinations is two, and the maximum value function or the minimum value function can be used depending on the characteristics. This calculation for determining the exercise intensity is simpler than a normal calculation procedure without error. Next, a method for further simplifying the procedure will be described although an error occurs by reducing the number of sections.

In FIG. 1, the slopes of the two straight lines in the third and fourth consecutive sections are slightly different, but it seems that there is no problem in terms of error even if these two sections are combined into one section and one straight line. is there. In FIG. 1, when the first section and the second section, or the fifth section and the sixth section are combined into one, an error seems to occur to some extent. However, if this is a practical error, the procedure can be further simplified by combining them into one.

To reduce the number of sections in order to simplify the procedure, a method of combining two or more consecutive sections into one section to make a new section, or walking speed regardless of the original six sections. There is also a method of creating a new section by dividing a part or all of the section into several sections. In order to define one straight line in a new section, a method of obtaining a regression line by performing a regression analysis by a known least square method is simple and clear. A specific method will be described in detail below.

For a function defined in a certain section, a regression line can be obtained analytically by regression analysis using the least squares method in that section. For the sake of simplification, a description will be given of two consecutive equally spaced intervals (v1, v2) and (v2, v3). Let the slope and intercept of the straight line in each section be (a1, b1) and (a2, b2). At this time, the slope and intercept (α, β) of one regression line obtained by performing the regression analysis in a single section (v1, v3) should be obtained as a solution of the following binary simultaneous linear equation. Is shown. Omit proof.
2 · (A + B) · α + 3 · (C + D) · β = 2 · (a1 · A + a2 · B) +
3 ・ (b1 ・ C + b2 ・ D)
(C + D) · α + 4.β = (a1 · C + a2 · D) +
2 ・ (b1 + b2)
Here, A = v1 · v1 + v1 · v2 + v2 · v2, C = v1 + v2
B = v2 / v2 + v2 / v3 + v3 / v3, D = v2 + v3

The section put together into 1 is expressed as the 1st 2nd section, for example. When the above analysis results are applied to the third and fourth intervals in FIG. 1, the slope and intercept of the regression line in the third to fourth intervals are (0.9375, −0.975). When this approximate value (0.94, -0.98) is used and compared with the value of the original straight line equation, it can be confirmed that the error is less than 1%, and there is no problem as a practical error.

Strictly speaking, the regression line in the third to fourth section becomes discontinuous at the boundary between the adjacent second and fifth sections, and the five straight lines in the five sections become discontinuous piecewise linear functions. Since the error at the discontinuity is less than 1%, it can be regarded as a piecewise linear function continuous in five sections. Furthermore, since the slopes of the straight lines in the first section, the second section, the third ∨4 section, and the fifth section are gradually increasing (gradual non-decreasing), the function is regarded as a concave characteristic in these four sections. I can do it. Then, using the concave characteristic, it is possible to use the maximum value function without further section determination in these four sections. Specifically, it is determined once whether or not the walking speed is in the sixth section, and if it is in the sixth section, the exercise speed is obtained by substituting the walking speed into the straight line expression of the sixth section. Otherwise, the walking speed is substituted into the straight line expressions of the first section, the second section, the third base 4 section, and the fifth section, and the maximum value thereof is obtained as the exercise intensity. Of course, it is also possible to use a method in which branch determination is performed faithfully for these four sections, and substitution calculation is performed using the linear expression of the section that has been reached. The determination of which method to select will be encountered during program design, for example.

The above method is a simplified calculation method with an error of less than 1% compared to the ordinary procedure described above. Although the error slightly increases, a simplified calculation method will be described below. This method is a calculation method in which the number of sections is further reduced and section determination is unnecessary.

The first section and the second section in FIG. 1 are combined into one, and the fifth section and the sixth section are combined into one. Apply the above binary simultaneous linear equations of α and β. The slope and intercept of the regression line in section 1∨2 are (0.4375, 1.325) ≈ (0.44, 1.33), and the slope and intercept of the regression line in section 5∨6 are (2.0625). , −8.025) ≈ (2.06, −8.03). As described above, the slope and intercept of the regression line in the third to fourth section are (0.9375, −0.975) ≈ (0.94, −0.98). The adjacent regression line becomes discontinuous at the boundary of the new three sections, but the original exercise intensity (3.5 mets) at the discontinuous point (walking speed 4.8 km / h, 6.4 km / h). , 5.0 mets) are (3.532-3.442) /3.5≈2.6%, (5.154-5.036) /5.0≈2.4%, The error is slight. In addition, the slopes of the three regression lines gradually increase. Therefore, a piecewise linear function composed of three regression lines in these three sections is continuous, and is treated as a concave characteristic in the three sections. By doing this, it is possible to obtain the exercise intensity M (Mets) by the following formula 2 using the maximum value function relating to the walking speed V (km / hour) without performing zone determination. When targeting physical activity, subtract 1 Met at rest.

An error between the value of the piecewise linear function of the six sections as a reference and the calculated value according to Equation 2 is confirmed. As a result of the numerical verification, the maximum error is about 3.1% at walking speed V = 6.4 km / hour, and it can be said that there is no problem as a practical error.

The above-described simplified calculation method for obtaining exercise intensity from walking speed is derived from a piecewise linear function based on the present invention based on the revised data of Non-Patent Document 2. This calculation method is a simplified method with practical accuracy, using one or both of the following characteristic processes. This greatly contributes to solving the problems of the present invention. The characteristic process is a process using a maximum value function or a minimum value function according to the characteristics without performing further section determination in the continuous section if it is either a concave characteristic or a convex characteristic in the continuous section. This is a process in which continuous sections are combined into one and replaced with one regression line obtained by regression analysis. The calculation method may be determined by combining these processes in consideration of the tolerance in product design. It goes without saying that this characteristic process can also be applied to exercise intensity calculations such as jogging and running other than walking even if the shape of the piecewise linear function changes.

After obtaining the walking speed, for example, the exercise intensity can be calculated using Equation 2. Furthermore, the exercise intensity is multiplied by the measured walking time, the calorie consumption is multiplied by the basal metabolic rate of 1.05 × body weight, and this is divided by 7.2 to calculate the fat burning amount. Walking related information can be calculated so that it continues. Therefore, since walking speed is the basis of all walking-related information, it is understood that the key is how to determine walking speed with practical accuracy and how to increase accuracy. The The method for calculating the walking speed will be described in detail below.

The conclusion is shown first. The following Equation 3 is an arithmetic expression for walking speed. N is the number of steps per unit time (for example, 1 minute), that is, the walking pitch. Hereinafter, it is also simply referred to as a pitch. L is height, and the walking stride is 0.45 times the known height. 0.45 L · N is a walking distance per unit time, that is, a walking speed. However, the stride is not always constant, it becomes wider when the pitch goes up and becomes narrower when it goes down. If the stride is set to 0.45 · L, the walking speed cannot be obtained accurately. If the stride is not corrected by any method, the walking related information lacks accuracy. Therefore, the stride correction according to the pitch N is performed by multiplying by the correction function f (N), and the walking speed is calculated by Equation 3. Formula 3 does not use 0.45 as it is, but is a general formula using a stride factor S3 representing a magnification with respect to height. The stride factor S3 will be described later.

Formula 3 above shows only the form of the walking speed calculation formula without aligning the units, and is different from the actual calculation formula described later in the presence or absence of coefficients. In addition, although the form of Formula 3 itself is simple, the derivation of the correction function f (N) is not so easy, and the inventor of the present application has derived it from the investigation of related information including patent documents and the results of repeated walking experiments. Is.

There is walking as an exercise to increase the speed consciously to promote or improve health, and walking in daily activities such as going out for shopping or walking. The pitch and stride of both are naturally different, and it is easy to see that the former is larger than the latter. Therefore, how the relationship between the pitch and the stride can be grasped will be described first.

A fast walk of about 6 km / h corresponds to a pitch of 120 to 130 steps / minute and a stride of a known walking stride of 0.45 · L (L is height). On the other hand, the normal walking at a speed of about 4 km / h including the walking of daily activities shown as an example above has a pitch of 90 to 100 steps / minute and a step length of about 0.40 · L.

Next, changes in the stride with respect to the pitch will be described. When the pitch is in the range of 90 to 130 steps / minute, changes in pitch and stride are considered to be linear. In this respect, the method using the regression line in Patent Document 1 is appropriate, and a substantially linear change was confirmed also from the walking experiment conducted by the present inventor. This walking experiment will be described later.

However, the question is the change in stride when the pitch is in the range of 90 steps / min or less and in the range of 130 steps / min or more. It is not necessary to extend the straight line to this range. If it is extended, a large error occurs in the step length calculation from the following points, and it becomes impossible to obtain an accurate walking speed.

When the pitch is in the range of 90 to 130 steps / minute, the stride increases as the pitch is increased, and the walking speed increases in proportion to both the pitch and the stride. When the pitch is in the range of 130 steps / minute or more, even if the pitch is increased, the change in the stride is small and the increase is considered to be slight, and the walking speed is mainly increased in proportion to the pitch. This is because it is not easy to keep walking while maintaining a balance between the time required for one walk with a wider stride and a fast pitch. There are the following opinions of experts. It can't be said that no one walks at 50% of the stature, but the gait would be an unnatural gait or an exceptional person trained with muscles in the lower body. . (Http://www2.wainet.ne.jp/-tukasa/U_6.html)

When the pitch is further increased, the stride becomes narrower and turns into an unnatural sprint for walking, and eventually the stride is recovered and the run begins. Therefore, as long as natural gait as exercise is targeted, it is necessary to set an upper limit of pitch (for example, a value in the range of 150 to 160 steps / minute) as a response range of the pedometer.

On the other hand, in the range where the pitch is 90 steps / minute or less, there is a margin for one walking time, and therefore it is possible to widen or narrow the stride. However, since it is necessary to support the body with the other leg until the raised one leg is landed, there is a limit to widening the stride from the standpoint of maintaining the pitch and the balance of the body, and it also leads to unnatural walking . Conversely, narrowing the stride does not cause a problem of body balance, but making it too narrow results in unnatural walking. From the above, it can be considered that when the pitch is 90 steps / minute or less, there is little change in the stride during natural walking, and the decrease in the stride is slight. This is also confirmed by the inventor's walking experiment described later. If the pitch goes down too much, it will be difficult to say that it is natural walking as an exercise including daily activities. Accordingly, as described above, it may be necessary to set the lower limit of the pitch (for example, a value in the range of 70 to 80 steps / minute) as the response range of the pedometer.

The relationship between pitch and stride has been described in detail quantitatively and qualitatively. In order to verify this relationship, the present inventor conducted a walking experiment. A certain section (distance 550 m) is walked in accordance with the transmitted sound of the electronic metronome carried to maintain the walking pitch, and the walking time and the number of steps of the pedometer are recorded and the data is analyzed. FIG. 2 shows the experimental results and graphs the relationship between the calculated walking pitch and step length. A range with a small increase in stride appears from above about 125 steps / minute, and a range with a small decrease in stride appears from below about 95 steps / minute. The experimental results agree well with the above. From the height of the present inventor's height of 172 cm, the walking stride = 172 × 0.45 = 77.4 cm and the normal walking stride = 172 × 0.40 = 68.8 cm, which are also in good agreement with the experimental results.

Now, the stride correction function f (N) will be described below. The meanings of the symbols used in the explanation are as follows. N2, N3, S2, S3, K1, and K4 can be regarded as parameters that determine walking characteristics.
N1 (steps / minute): lower limit pitch, for example 70 ≦ N1 ≦ 80 constant N2 (steps / minute): normal walking pitch, for example, 90 ≦ N2 ≦ 100 constant N3 (steps / minute): fast walking pitch, for example 120 ≦ N3 ≦ 130 Constant N4 (steps / minute): upper limit pitch, for example, constant S2 of 150 ≦ N4 ≦ 160: normal stride factor, constant of 0 ≦ S2 ≦ 1, for example, 0.40
S3: Rapid walking stride coefficient 0 ≦ S3 ≦ 1 constant For example, 0.45
K1: Downward slope coefficient 0 ≦ K1 ≦ 1 constant, for example, 0.50
K4: Constant of upward inclination coefficient 0 ≦ K4 ≦ 1 For example, 0.50
L (cm): The height symbols N1 and N4 are the lower and upper pitches of the response range of the pedometer, and are not directly related to the walking characteristics. S2 and S3 are magnifications with respect to the height of the stride, and K1 and K4 are coefficients relating to the slope of the correction function represented by a straight line for each section, which are multiples of the slope of the straight line in the intermediate pitch section described below.

The number of steps measured per unit time, that is, the walking pitch is N (steps / minute). Based on the above detailed description, the stride correction function f (N) can be given by Equation 4, Equation 5, and Equation 6 by providing three walking pitch sections. This correction function constitutes a continuous piecewise linear function having three intervals. Note that f (N) is normalized by the rapid walking stride S3 · L.

Normal walking pitch N2 to fast walking pitch N3 (hereinafter referred to as “intermediate pitch section”)

Lower limit pitch N1 to normal walking pitch N2 (hereinafter referred to as “downward pitch section”)

Rapid walking pitch N3 to upper limit pitch N4 (hereinafter referred to as “upper pitch section”)

The walking speed can be accurately obtained by applying the correction value f (N) calculated by using the stride correction function of any one of the above formulas 4, 5, or 6 to the formula 3.

However, according to the present invention, the walking-related information provided together with the number of steps has at least practical accuracy, can further improve accuracy, and has a simple structure and is economically inexpensive. It is an issue. A method for solving this problem will be described in detail below by simplifying the calculation method of the walking speed according to Equation 4, Equation 5, Equation 6, and Equation 3.

First, formula 4, formula 5, and formula 6 are substituted into formula 3, and the formula for calculating the walking speed is specifically developed. For intermediate pitch sections N2 to N3, Formula 4 is substituted into Formula 3.
Walking speed V = f (N) / S3 / L / N
= ((1-S2 / S3) / (N3-N2). (N-N3) +1) .S3.L.N
= (S3-S2) / (N3-N2)
(N + (S2, N3-S3, N2) / (S3-S2)), N, L

For the lower pitch sections N1 to N2, Formula 5 is substituted into Formula 3.
Walking speed V = f (N) / S3 / L / N
= (K1. (1-S2 / S3) / (N3-N2). (N-N2) + S2 / S3).
S3 ・ L ・ N
= (S3-S2) / (N3-N2)
(K1 * N + S2 * (N3-N2) / (S3-S2) -K1 * N2) * N * L

For the upper pitch sections N3 to N4, Expression 6 is substituted into Expression 3.
Walking speed V = f (N) / S3 / L / N
= (K4. (1-S2 / S3) / (N3-N2). (N-N3) +1).
S3 ・ L ・ N
= (S3-S2) / (N3-N2)
(K4 * N + S3 * (N3-N2) / (S3-S2) -K4 * N3) * N * L

The form of the expression obtained by developing the walking speed calculation expression is walking speed V = k · (ki · N + ci) · N · L. Here, N and L are the walking pitch (steps / minute) and height (cm), the coefficient k is a coefficient common to the straight lines of the three walking pitch sections, and the coefficient ki and the constant ci are the straight lines of the three walking pitch sections. It corresponds to the slope and intercept. The walking speed is obtained by multiplying the above equation by the conversion factor h = 60/100/1000 for walking speed. Then, h · k · ki is considered as a new coefficient ki, and h · k · ci is considered as a new constant ci. In this way, the calculation formula of the walking speed is given by the following formula 7. In Formula 7, it goes without saying that a straight line represented by a linear expression of N determined for every three pitch sections constitutes a continuous piecewise linear function. Further, a quadratic expression obtained by multiplying the linear expression by N also becomes a quadratic function in a piecewise manner continuous in three pitch sections. In the present invention, this is called a piecewise quadratic function. This piecewise quadratic function is a standard for error evaluation of various regression equations described later.

The method for obtaining the walking speed using Equation 7 first determines the pitch interval. Since there are three sections, a maximum of two determinations are required. Then, a linear expression defined for the pitch section is selected and substitution calculation is performed. At this time, there is multiplication with a coefficient. Next, this result is multiplied by N and L. This method is a normal calculation procedure through a proper process of calculating the walking speed accurately. In the previous exercise intensity calculation, a method for reducing the number of section determinations for a continuous piecewise linear function has been described. In this method, the maximum value function or the minimum value function is used according to the concave characteristic or the convex characteristic in the continuous section. By this method, the number of section determinations can be set to one. However, this alone has little effect on the simplification of the procedure.

In Equation 7, a regression analysis is performed on a quadratic part obtained by multiplying a primary expression of N by N. If the obtained regression line is replaced with a linear expression, the number of multiplications can be reduced by one. Although the number of multiplications is reduced by one at most, if the approximation error is within a practical error, repeating such a device simplifies the structure of the calculation, and as a result, the problem of the present invention is solved. This will be specifically described below.

The quadratic expression (ki · N + ci) · N is subjected to regression analysis by the method of least squares in the interval P to Q. The regression line is obtained analytically, and its slope and intercept are given by the following general formula. Omit proof. Needless to say, the slope and intercept of the regression line are coefficients and constants of the regression equation.
Inclination of regression line: ki · (P + Q) + ci
Regression line intercept: −ki · (P · P + 4 · P · Q + Q · Q) / 6
The quadratic expressions ki and ci and P and Q of the section may be considered positive values from the examination content. The difference value obtained by subtracting the quadratic equation from the regression equation is maximum (positive value) in the middle of the interval, and is minimum (negative value, twice the maximum value) at both ends of the interval. . Accordingly, since the value of the quadratic expression is increased in the section, the magnitude of the error with respect to the quadratic expression of the difference is the maximum at the lower end (P) of the section and is calculated by the following expression.
Maximum value of absolute difference: ki · (PQ) · (PQ) / 6
Maximum error absolute value: ki · (PQ) · (PQ) / 6 / ((ki · P + ci) · P)

Check using specific values. Use representative values: N1 = 70, N2 = 95, N3 = 125, N4 = 160, S2 = 0.40, S3 = 0.45, K1 = 0.5, and K4 = 0.5. The quadratic equation, regression equation, and maximum miscalculation are as follows for each of the three pitch sections. In the calculation, the coefficient ki and the constant ci of the three linear expressions (ki · N + ci) are the same value (S3−S2) / (N3−N2) = 1/600, and the hourly speed conversion coefficient h = It is divided by the value multiplied by 60/100/1000.

Lower pitch section (70-95 steps / min)
Quadratic and regression: (N / 2 + 385/2) · N≈275 · N-3377
Maximum error: 0.00327 ≒ 0.33%

Intermediate pitch section (95 to 125 steps / minute)
Quadratic and regression: (N + 145) · N≈365 · N−12025
Maximum error: 0.00657 ≒ 0.66%

Upper pitch section (125-160 steps / minute)
Quadratic and regression equations: (N / 2 + 415/2) · N≈350 · N−10102
Maximum error: 0.00302 ≒ 0.30%

In any pitch interval, the error is less than 1%, and there is no problem even if the quadratic equation is replaced with a regression equation. By replacing it, the number of multiplications can be reduced by one. It goes without saying that when the above common number is multiplied to return to the walking speed calculation formula, the coefficients and constants of the three regression equations change, but the error does not change. Further, the three regression lines constitute a piecewise linear function with three pitch sections. Strictly speaking, it is discontinuous at the boundary of the interval, but the error of the discontinuous point with respect to the original quadratic value is less than 1%, and is regarded as a continuous piecewise linear function, and in addition, from the slope of the regression line Also, it can be regarded as a concave characteristic in the lower pitch section and the intermediate pitch section, or as a convex characteristic in the intermediate pitch section and the upper pitch section. Therefore, it is possible to combine a method that makes it unnecessary to determine the section in the concave characteristic or convex characteristic section by using the maximum value function or the minimum value function.

The above walking speed calculation method requires the determination of the pitch interval, but the quadratic expression is simplified by the regression equation, and the miscalculation is less than 1%. This is an effective means for solving the problems of the present invention. The method to be described below is a more simplified and excellent walking speed calculation method that does not require the determination of the pitch section, although the error slightly increases.

As described above, the quadratic part of Equation 7 is a piecewise quadratic function continuous in three pitch sections. In this method, the piecewise quadratic function is subjected to regression analysis by the least square method through the pitch interval and is replaced with one regression line. This regression line can also be obtained analytically. In this method, the square value of the difference between the slope and the regression line having the intercept as a variable is integrated in the target section, and the variable that minimizes the integral value is determined. The regression line obtained earlier is also obtained by the same method. This time, it will be a little complicated, but the results are shown below. Omit proof.

The symbols N1, N2, N3, N4, K1, and K4 that have been used above are used. Note that N1 ≦ N2 ≦ N3 ≦ N4. A quadratic expression of the intermediate pitch section is represented by (k2 · N + c2) · N. At this time, since the primary expression of the lower pitch section and the upper pitch section is continuous, the secondary expression is expressed by the following expression.
Secondary equation of the lower pitch section: (K1 · k2 · (N−N2) + k2 · N2 + c2) · N
Quadratic expression of upper pitch section: (K4 · k2 · (N−N3) + k2 · N3 + c2) · N
Also, the equation of the regression line to be obtained is the following equation.
Regression line formula: γ · N + δ
The square value of the difference between the quadratic expression and the regression line expression in the pitch sections N1 to N4 is integrated in this section, and γ and δ giving the minimum values of the integrated values are determined.

The slope γ and intercept δ of the regression line, that is, the coefficient γ and the constant δ of the regression equation are given as solutions of the following binary simultaneous linear equations.
(2/3) · (G3 (N4, N1)) · γ + (G2 (N4, N1)) · δ = SGM1
(G2 (N4, N1)) · γ + 2 · (G1 (N4, N1)) · δ = SGM2
Here, the functions G1 (X, Y), G2 (X, Y), G3 (X, Y), and G4 (X, Y), and SGM1 and SGM2 are as follows.
G1 (X, Y) = XY
G2 (X, Y) = X.X-Y.Y
G3 (X, Y) = X, X, XY, Y, Y
G4 (X, Y) = X, X, X, X-Y, Y, Y, Y
SGM1 = (2/4) · k2 · (G4 (N4, N1))
+ (2/3) · c2 · (G3 (N4, N1))
-2 ・ (1-K1) ・ k2 ・ (1/4) ・ (G4 (N2, N1))
-2, (1-K4), k2, (1/4), (G4 (N4, N3))
+2 ・ (1-K1) ・ k2 ・ (1/3) ・ N2 ・ (G3 (N2, N1))
+2 ・ (1-K4) ・ k2 ・ (1/3) ・ N3 ・ (G3 (N4, N3))
SGM2 = (2/3) · k2 · (G3 (N4, N1))
+ (2/2) · c2 · (G2 (N4, N1))
-2 ・ (1-K1) ・ k2 ・ (1/3) ・ (G3 (N2, N1))
-2, (1-K4), k2, (1/3), (G3 (N4, N3))
+2 ・ (1-K1) ・ k2 ・ (1/2) ・ N2 ・ (G2 (N2, N1))
+2 ・ (1-K4) ・ k2 ・ (1/2) ・ N3 ・ (G2 (N4, N3))

It is quite complicated to solve γ and δ, which are solutions of the above binary simultaneous linear equations, in the form of general formulas. It is also complicated to derive the difference error between the original piecewise quadratic function and the obtained regression line in the form of a general formula. However, if the coefficients and constants of the binary simultaneous linear equations are determined by specific numerical values, they can be solved easily and the regression line can be determined. And the error can also be calculated. Therefore, we will check with specific numerical values. Representative numerical values used earlier, N1 = 70, N2 = 95, N3 = 125, N4 = 160, S2 = 0.40, S3 = 0.45, K1 = 0.5, K4 = 0.5, and Further, a quadratic expression (N + 145) · N of the intermediate pitch section is used. At this time, k2 = 1 and c2 = 145. The regression equation obtained by solving the binary simultaneous linear equations is as follows. The reason for the regression analysis in the two pitch sections is related to the response range of the pedometer in addition to the confirmation of the error tendency. These will be described later.
Regression equation regressed through pitch sections 70 to 160: 342 · N-9174
Regression equation regressed through pitch sections 80 to 150: 349 N-10094

If the above regression equation is used, it is not necessary to determine the pitch interval, and only one arithmetic equation is required, and the arithmetic structure is simplified. Before confirming whether or not the error is within a practical range, a calculation method immediately before this method will be described. It seems likely that there will be fewer errors than the simplest method. In the previous method, the three pitch sections are divided, for example, in the middle and are set to two pitch sections. Two regression lines are obtained by regression analysis of the quadratic expression in each pitch interval. And it is the method of omitting area determination using the maximum value function or the minimum value function.

The pitch to be divided into two is assumed to be an intermediate 115 (steps / minute). N1 = 70 (or 80), N2 = 95, N3 = eg 105, N4 = 115, S2 = 0.4, S3 = 0.45, K1 = 0.5, K4 = 1.0 By solving the binary simultaneous linear equations relating to γ and δ as N + 145) · N, the regression line in the pitch sections 70 (80) to 115 can be determined. Also, if N1 = 115, N2 = 120, N3 = 125, N4 = 160 (or 150), S2 = 0.4, S3 = 0.45, K1 = 1.0, K4 = 0.5, A regression line in the pitch sections 115 to 160 (150) can be determined. The determined regression equation is as follows.
Pitch section 70-160 is divided into two at pitch 115. Regression equation for regression analysis with pitch section 70-115: 309 N-6069
Regression equation regressed in pitch sections 115 to 160: 352 · N-10459
Regression formula of regression analysis of pitch sections 80 to 150 with pitch 115 and two divided pitch sections 80 to 115: 323 · N-7656
Regression equation regressed in pitch sections 115 to 150: 352 · N-10361

Evaluate the error between the piecewise quadratic function that is the standard for error evaluation and the regression line obtained. The standard secondary equation is as shown above. The regression equation to compare is as shown immediately above. One division means a regression analysis through the pitch sections 70 to 160 or 80 to 150. Table 1 shows the error calculation results. The error is expressed as% by subtracting the value of the quadratic expression from the value of the regression equation and dividing the result by the value of the quadratic expression.
Secondary expression of pitch section 70-95: (N / 2 + 385/2) · N
Secondary expression of pitch section 95-125: (N + 145) · N
Secondary expression of pitch sections 125 to 160: (N / 2 + 415/2) · N
Regression analysis with pitch interval 70-160 1 division: 342 N-9174
Divided into two: MAX (309, N-6069, 352, N-10459)
Regression analysis with pitch interval 80-150 1 division: 349 N-10094
Divided into two: MAX (323, N-7656, 352, N-10361)

As shown in Table 1, an error of about 5% or more is a case where a one-part regression equation is used in the pitch sections 70 to 160. It is only a part of the section below the mid 70 pitches. Whichever regression equation is used, it is about 4% or less from the pitch 80 to the middle of the 80 units, about 2% or less at the middle of the pitch 80 or more, and about 1% or less at the middle of the 110 units or more. . In 2 divisions, it is about 2% or less throughout the pitch section, as expected. From this error result, it can be said that the above-described method that can eliminate the interval determination and can further simplify the calculation structure is sufficiently effective as means for solving the problems of the present invention.

Mention the response range of the pedometer. The inventor of the present application has experienced the following with respect to the response performance to the walking pitch of the pendulum and acceleration sensor built in the pedometer. In the previous walking experiment, when walking at a pitch of 80 or less, or a pitch of 150 or more, there are rare cases where the pedometer has a strange number of steps, so that it can be clearly noticed. That is, it is considered that the sensor is not responding correctly. Further, considering the walking pitch when walking as a natural physical activity, it can be said that the error generated when the arithmetic expression is simplified may be evaluated in the pitch sections 80 to 150.

If Table 1 is reviewed from the above points, the regression equation for one division in the pitch sections 70 to 160 is sufficiently practical compared to the other because the error is about 2% or less at a pitch of 80 or more. . It is a design tradeoff to make the error smaller, in other words, to pursue more accuracy, or to simplify the calculation structure and make it economically cheaper. Therefore, it can be said that the method of simplifying the calculation structure described above has a great effect of expanding the selection range when the calculation method is examined in product design.

The calculation method for obtaining the walking speed from the walking pitch has been described in detail above. The operation using one regression line obtained by regression analysis of a piecewise quadratic function through the pitch interval is the simplest method. A specific calculation formula using a regression formula obtained by regression analysis in the pitch sections 80 to 150 is represented by the following formula 8.

The walking speed V is obtained by Equation 8, and this is applied to Equation 2 to obtain the exercise intensity M. Expressions 2 and 8 are arithmetic expressions that are simplified while maintaining practical errors. This is a simplified arithmetic expression based on two piecewise linear functions derived by the inventors of the present invention through well-known numerical values and sufficient examination. The two piecewise linear functions are the functions that are the basis of the present invention. One defines the relationship between walking speed and exercise intensity, and the other defines the relationship between walking pitch and stride (specifically, stature height ratio).

The former piecewise linear function can be applied not only to a specific pedometer user but also to an unspecified number of users. On the other hand, the latter piecewise linear function has several parameters that determine the walking characteristics of the pedometer user in the intermediate pitch section and in the lower and upper pitch sections where the step change is small. By determining the values of these parameters, the shape of the latter piecewise linear function is specifically determined. And the stride according to the walking characteristic can be calculated accurately. As a result, the walking speed can be accurately calculated. Accordingly, in addition to setting the height and weight, accurate walking related information can be obtained by allowing the pedometer to set a walking characteristic parameter that matches the walking characteristic of the pedometer user.

However, since the walking characteristic parameter is not a numerical value that is handled daily like height and weight, the setting is complicated and the structure is complicated. When designing a general-purpose pedometer that has at least practical accuracy, can improve accuracy, and has a simple structure and is economically inexpensive, the structure for setting the walking characteristic parameters is disadvantageous. . In order to solve this problem, Formula 8 is an arithmetic expression obtained by fixing characteristic parameters to representative numerical values and greatly simplifying the arithmetic structure. This representative value is the following fixed value derived based on commonly used numerical values or the results of walking experiments conducted by the inventors of the present application. These numbers were often used in the above explanations.
Lower limit pitch N1 (steps / minute): 70
Normal walking pitch N2 (steps / minute): 95
Rapid walking pitch N3 (steps / minute): 125
Maximum pitch N4 (steps / minute): 160
Normal stride factor S2: 0.40
Rapid walking stride coefficient S3: 0.45
Downward slope coefficient K1: 0.50
Upward tilt coefficient K4: 0.50

When the pedometer user has a walking characteristic away from the above fixed value, the error generated in the exercise intensity obtained by applying the walking speed calculated by Equation 8 to Equation 2 is a simplification (approximation) of the equation. And total error including both errors due to fixation of walking characteristic parameters. That is, it is an error in the structure of the pedometer when the simplified arithmetic expression is adopted, and the accuracy of the walking related information provided by the pedometer, that is, the practical performance is determined. Therefore, verification of this total error is essential.

Therefore, the walking speed of the pedometer user, that is, the difference in walking characteristics, is simulated by changing the walking characteristics parameter, and the walking speed V (km / hour) obtained by Formula 8 using a fixed value is calculated. The error from the exercise intensity M (mets) obtained by applying to 2 is evaluated. The reference value for error evaluation is a function value of a piecewise linear function used as a reference in the present invention. The reference value of the exercise intensity is based on the piecewise linear function shown in FIG. The reference value of the walking speed is an hourly speed converted value based on calculations using Equation 4, Equation 5, Equation 6, and Equation 7 when the walking characteristic parameter is changed. The change from the fixed value of the walking characteristic parameter is as follows. Height is L = 170 cm.
(N2, N3): (95,125) → (95,135), (95,115)
Change width: ± 10 (85,125), (105,125)
(S2, S3): (40%, 45%) → (40, 47), (40, 43)
Change width: ± 2% (38, 45), (42, 45)
(Downward and upward): (50%, 50%) → (50, 100), (50, 0)
Change width: ± 50% (0, 50), (100, 50)

For convenience of explanation, three types of characteristic parameters (N2, N3), (S2, S3), and (downward tilt, upward tilt) are abbreviated as (N), (S), and (tilt). Here, the change width of the characteristic parameter will be described. 50% of the fixed value determined by (tilt) was set to an average value of 0% and 100% because a known numerical value for this characteristic parameter was not found. For this reason, the range of change is deliberately set to 50% (100% in rate). 10 of (N) is about 10% at 10/110 (= average value of 95 and 125). 2% of the variation width of (S) is about 5% at 2 / 42.5 (= average value of 40 and 45). It is never a small value. At the change width of 2%, the height L = 170 cm is 45% stride: 76.5 cm and 47% stride: 79.9 cm. It is not a small change width from the walking experience of the inventor of this application whose height is 172 cm.

When changing one type of characteristic parameter, the remaining two types were fixed to fixed values, and numerical verification of the error was performed. Table 2 shows the verification results when (N), Table 3 (S), and Table 4 (inclination) are changed from fixed values to the above four sets of numerical values. These tables show the calculated value of walking speed and the error of this calculated value, and the calculated value of exercise intensity and the error of this calculated value. The error is expressed in% by subtracting the reference value from the calculated value of Equation 8 (or Equation 8 and Equation 2) and dividing the result by the reference value. In addition, the first decimal place is rounded off in order to capture the error as an approximate number. In addition, the calculated value of exercise intensity includes 1 Mets at rest.

From Table 2, Table 3, and Table 4, the error of the walking speed calculation value and the error of the exercise intensity calculation value are summarized as follows.
Section within ± 5% Section within ± 6% Error in section 80-150 (N) Walking speed V 85-150 Within 80-150 ± 6% Exercise intensity M 80-130 80-130 Within ± 12% (S) Walking speed V 105-140 90-150 Within ± 10% Exercise intensity M 105-120 80-130 Within ± 17% (tilt) Walking speed V 85-150 85-150 Within ± 7% Exercise intensity M 80-135 80- Within 135 ± 10%

From this aggregated result, the section with error within ± 5% almost covers the intermediate pitch section fixed with normal walking pitch N2 = 95 and fast walking pitch N3 = 125, and it is completely covered with error ± 6%. doing. Further, in the sections 80 to 150, the walking speed error is within ± 10%, and the exercise intensity error is within ± 12% except for some sections 140 to 150 as shown in Table 3. The reason why the error of the exercise intensity is enlarged in this section is considered as follows. As shown in Table 3, the walking speed calculation error that occurs when the parameter S3 changes from 45% to 47% or 43% is as small as 5% to 6%. However, as can be seen from FIG. 1, when the calculated value of the walking speed is 6.4 to 7.2 km / hour, the exercise intensity is calculated using a linear equation having the maximum inclination (20/8 = 2.5). Therefore, the error of the exercise intensity is expanded by 2.5 times. This is considered to be the cause of the error expansion of exercise intensity.

The above verification is a case where only one type of characteristic parameter is changed. The error when three types are changed at the same time and two parameters of each type are simultaneously changed is also verified. At this time, since it is a simultaneous change, the change range was set to the above ½, and the same numerical verification was performed with 4 × 4 × 4− = 64 combinations. About the error of walking speed, the result almost the same as the above-mentioned aggregation result was obtained. Regarding the error of the exercise intensity, 36 combinations have an error of ± 10% or less in the sections 80 to 150, and the remaining 28 patterns have an error exceeding ± 10% in the partial sections 140 to 150. The breakdown of 28 patterns was 14 patterns with errors exceeding ± 10% and ± 12% or less, 8 patterns exceeding ± 12% and ± 15% or less, and 6 combinations exceeding ± 15%.

Six combinations with an error exceeding ± 15% are shown in the order of (N), (S), (tilt), and error.
(100,120), (39,46), (25 or 75,75) within ± 23% (100,120), (41,46), (25 or 75,75) within ± 18% (90,120 ), (39, 46), (25 or 75, 75) within ± 17% The reason for the error expansion is that in the group of 23%, the intermediate pitch section becomes narrow (30 → 20), and the step correction function of the same section The slope increases (5/30 → 7/20), and the slope in the upper pitch section also increases (5/30 × 50% → 7/20 × 75%). As a result, the deviation from the fixed value is large. This is the main cause. The error expansion due to the linear inclination of the exercise intensity calculation also overlaps. The lower group has the same reason. These are combinations under numerical verification. In these combinations, the increase in the stride in the upper pitch section is not normal, and the walking speed is a calculation result exceeding 7 km / h, and it is difficult to say that it is natural walking. Therefore, it is considered that these combinations may be excluded from the error evaluation target.

As described above, the walking speed calculated by Expression 8 of the present invention and the error (total error) included in the value of exercise intensity calculated by applying the result to Expression 2 were numerically verified in detail. There are two types of verification: only one parameter is changed, and three are changed simultaneously. Comprehensive evaluation of two types of verification results. The following conclusion is obtained except for the six extreme combinations shown above.

As can be seen from Tables 2, 3 and 4, there are of course sections within ± 5% and ± 10% defined as practical errors in the present invention. When the conclusion is expressed in the greatest common divisor, the error in walking speed and exercise intensity is within ± 6% in the sections 90 to 130, and within ± 12% in most combinations in the sections 80 to 150. What exceeds ± 12% is only an error in the calculated value of the exercise intensity, which occurs in the sections 140 to 150. A walking pitch of 140 or more is a considerably fast pitch. It is not easy to continue walking at this pitch, and the majority of pedometer users are considered walking below this pitch. As described above, the combination of the simplified formula 8 or the formula 8 and the formula 2 not only greatly simplifies the calculation structure, but also the practical error about walking related information mentioned above is ± 5 to 10%. It will generally agree with the degree. Therefore, it is confirmed that it has at least practical accuracy, and it can be concluded that it is sufficiently practical.

The walking distance of exercise related information, exercise, calorie consumption in question, fat burning amount, etc. are the walking speed, exercise intensity, and the measured walking time (the pedometer that provides walking related information generally has a built-in clock function. , Accurate time information with negligible error is obtained), weight set values, constants and other values that are error-free. Therefore, it goes without saying that the error of the walking related information is not different from the above.

At the end of the numerical verification using Equation 8 and Equation 2, another interesting numerical verification result is shown. This is a result obtained by determining the walking characteristic parameter from the walking experiment data of FIG. 2 of the present inventor and performing the same error verification as described above. The walking characteristic parameters N2, N3, S2, S3, K1, and K4 are determined using the data of the walking pitches 80 to 150 in FIG. In the case of the present inventor, it can be read from FIG. 2 that N2 = 100 and N3 = 125. Assuming that N1 = 80 and N4 = 150, three straight lines are determined from the experimental data. This obtained the slope and intercept using the calculation software of regression analysis. From these straight lines, S2 = 40%, S3 = 45%, K1 = 33%, and K4 = 4% were obtained. The numerical verification results using this parameter show that the walking speed is 80 to 140 and the error is ± 2% or less, 80 to 150 is ± 4% or less, and the exercise intensity is 80 to 135, ± 4% or less, 80 It was ± 8% or less at -150. The height was changed to L = 172 cm of the present inventor, and numerical verification was performed.

The inventor's way of walking (walking characteristics) is by no means an extremely biased way of walking. I'm not going to say it's a typical way to walk, but, like many others, it's a natural gait. If there is a distribution of walking characteristic parameters, we think that it is not so far from the center of the distribution. Such an error when using the walking characteristic parameter of the present inventor has no problem as described above. From the numerical verification results shown at the end, it is considered that the practicality of the calculation method based on Formula 2 and Formula 8 fixed to the representative value has been sufficiently proved.

It is clear that more accurate walking-related information can be provided by allowing the pedometer to set a walking characteristic parameter that matches the walking characteristic of the pedometer user. When designing a pedometer with more accurate walking-related information and abundant additional functions than economically inexpensive, an arithmetic circuit built in the pedometer will have a high capability. In this case, the walking speed and exercise intensity are calculated using the normal calculation procedure without simplifying the piecewise linear function determined by the set walking characteristic parameters and without further simplifying the piecewise linear function of exercise intensity. do it.

However, when designing a general-purpose pedometer that has a simple structure and is economically inexpensive, it is difficult to set a walking characteristic parameter and use it for calculation. This is because not only the capacity of the setting circuit is increased, but it is also burdensome to cause a general-purpose arithmetic circuit to calculate the walking speed and exercise intensity each time using the same normal arithmetic procedure as described above. It is also difficult to use the set walking characteristic parameter only at the initial stage of calculation, calculate the slope and intercept of the regression line, which is a simplified calculation formula, and use it during normal calculation. This is because, in order to determine the regression line, it is necessary to solve the binary simultaneous linear equation shown above, and there is a concern that the capacity of the memory circuit is insufficient. Therefore, the setting of the walking characteristic parameter is not performed. For example, it may be considered to adopt Formula 8 which is simplified by fixing the walking speed with a representative value, and Formula 2 which is simplified for exercise intensity. .

It is necessary to collect walking data in advance, such as a walking experiment conducted by the inventor of the present application, in order to determine the walking characteristic parameter, which is certainly complicated. However, in terms of improving the error, it is best that the pedometer user's walking characteristics can be reflected in some way. In particular, if the walking characteristic of the pedometer user can be reflected with a small burden on an economically inexpensive general-purpose pedometer that employs a simplified arithmetic expression with the walking characteristic fixed to the representative value as in Equation 8. For example, it is very attractive and worthwhile. This is because, in principle, there is no error due to the difference in walking characteristics, and only an error due to simplification of the arithmetic expression. In other words, it is possible to provide an ideal pedometer that is inexpensive and has few errors.

Let us look at the errors in the walking speed calculation values and the exercise intensity calculation values in Tables 2 to 4. This first stage error is helpful. This error is only due to the simplification of the arithmetic expression with no deviation in walking characteristics. The error is small and very good. A method for reflecting the walking characteristics will be described below.

The method is as follows. For example, a walking characteristic parameter determined by walking data obtained by a walking experiment or the like is input to a personal computer or the like in which a program is previously incorporated. Then, the slope and intercept of one regression line are output. These are values that replace the coefficients and constants of the linear expression of N in Equation 8. This value is set as a set value for the pedometer. By doing so, the pedometer reflects the walking characteristics of the pedometer user with a minimum burden. The default value of the set value is set to, for example, the slope and intercept value of the linear expression of N in Formula 8 obtained by fixing to a representative value. And what is necessary is just to enable it to set according to a request | requirement of a pedometer user. It is not always necessary to collect walking data and determine walking characteristic parameters in advance, and the pedometer user does not have to worry about using this pedometer.

For example, the program incorporated in a personal computer or the like includes the above-described pedometer user's walking characteristic parameters N2 (normal walking pitch), N3 (fast walking pitch), S2 (normal walking step coefficient), S3 (fast walking step coefficient). , K1 (downward slope coefficient), K4 (upward slope coefficient), and predetermined N1 (lower limit pitch) and N4 (upper limit pitch) as inputs, and the binary of γ and δ described above Calculate the coefficients and constants of simultaneous linear equations. Then, two solutions of this equation are output. These are the slope and intercept of one regression line.

Here, in order to reflect the walking characteristics of the pedometer user, a structure provided in the pedometer will be described. Simplified by replacing the quadratic part of N in Equation 7 with a regression equation of a regression line obtained by regression analysis for each divided section. Formula 8 is the most simplified walking speed calculation formula of one division. The pedometer has a structure in which a coefficient and a constant of a linear expression of N in Expression 8 can be set, and must have a structure in which a set value can be read and calculated. Formula 9 shows the structure of the arithmetic expression provided in the pedometer. This is the simplest form of multiplying the expression of the linear function by the height L.

A coefficient k and a constant c, which are parameters that determine the characteristics of the linear expression N in Expression 9, are set values for the pedometer. This set value is the slope and intercept of a regression line calculated by a program relating to a binary simultaneous linear equation of γ and δ, which is incorporated in the personal computer or the like, based on the walking characteristic parameter of the pedometer user. Enter the above inventor's walking characteristic parameters, N2 = 100, N3 = 125, S2 = 40%, S3 = 45%, K1 = 33%, and K4 = 4%, lower limit N1 = 80, upper limit N4 = 150 Then, k = 0.000334 and c = −0.008813 are output. Of course, the handling of k and c, such as the number of significant digits and the sign, is a matter to be considered in the design.

The structure of the arithmetic expression provided in the pedometer when the segmented quadratic function is divided into two or three will be described. First, the case of two divisions is shown. The inventor's walking characteristic parameter is used again. An example in which the pitch sections 80 to 150 are divided into two at the pitch 125 is shown. The two regression equations are slopes k1 = 0.000337 and intercept c1 = −0.009165 in sections 80 to 125, slopes k2 = 0.000277 and intercepts c2 = −0.000905 in sections 125 to 150. . This can be calculated using the above program. Since the slope as a straight line does not gradually increase, the minimum value function (MIN) is used without determining the section. The arithmetic expression simplified by the regression equation is as follows.
Walking speed V (km / h)
= (MIN (0.000337 · N-0.009165,
0.000277 ・ N-0.000905)) ・ L
The structure of the arithmetic expression provided in the pedometer is one of the following. Since the slope does not necessarily increase gradually, the maximum value function (MAX) is used when the inclination does not decrease gradually. The slopes and intercepts (k1, c1) and (k2, c2) of the two regression lines, and any of MAX and MIN are parameters that determine the characteristics of the walking speed calculation formula. These are set. Of course, the following three-part arithmetic expression and setting structure may be used.
When the slope gradually decreases, walking speed V (km / h) = (MAX (k1 · N + c1, k2 · N + c2)) · L
When the slope does not gradually increase Walking speed V (km / h) = (MIN (k1 · N + c1, k2 · N + c2)) · L

Next, a general arithmetic expression in the case of three divisions will be described. Three regression lines are obtained by regression analysis of the piecewise quadratic function. The slopes and intercepts of these regression lines can also be obtained by utilizing the program relating to the above-mentioned binary simultaneous linear equations of γ and δ. The slopes of the regression lines are not necessarily gradual non -decrease or gradual non-increase in the three sections, and there are several combinations. The calculation method is not as simple as two divisions, and the method of section determination described below will be simple. The division pitches that are the division points of the three divisions are defined as p1 and p2 from below. Let the slope and intercept of the regression line be (k1, c1), (k2, c2), (k3, c3) from below. The setting procedure for the pedometer will be, for example, (k1, c1) → p1 → (k2, c2) → p2 → (k3, c3). The arithmetic expression is shown in the following flow.
Entrance Step 1: If N ≦ p1, go to Step 3, otherwise go to Step 2 Step 2: If N ≦ p2, go to Step 4, otherwise go to Step 5 Step 3: Walking Speed V (km / hour) = (K1 · N + c1) · L Exit step 4: walking speed V (km / hour) = (k2 · N + c2) · L Exit step 5: walking speed V (km / hour) = (k3 · N + c3) · L exit

A regression analysis is performed on the segmented quadratic function in a plurality of divided sections to obtain a plurality of regression lines. These constitute a new piecewise linear function. Then, the walking speed is calculated based on this piecewise linear function. This calculation method can easily generalize the calculation flow in the case of the above three divisions regardless of whether the calculation flow is two divisions or more than three divisions. For example, (k1, c1), p1, (k2, c2), p2, (k3, c3), p3, (k4) are set in order from the lower side, such as the slopes and intercepts of a plurality of regression lines, and the division pitches. , C4), p4... Can uniquely determine the new piecewise linear function. Therefore, these set values are parameters that determine the calculation formula of the walking speed, including the setting order. They are regression line parameters and segment point parameters. The definition of the arithmetic expression by this setting method can be applied to two divisions or one division. An impossible pitch, for example 500, is used. In two divisions, p2 = 500, and in one division, p1 = 500. This 500 can also be used to finish reading parameters. The above method is a uniform setting method for all divisions, and can uniformly determine the calculation structure of walking speed. This unified calculation method will be referred to as a step calculation method for the convenience of later explanation. Increasing the number of divisions naturally reduces the error, but it leads to complicated settings, an increase in calculation flow and storage capacity. The characteristic of the stature height ratio with respect to the walking pitch is a characteristic originally defined in three walking pitch sections. Therefore, the division will be about the same as the number of sections.

The method of reflecting the pedometer user's walking characteristics in the pedometer was explained in detail. According to this method, the computational burden is small due to the structure of the pedometer. When the walking characteristic is not set, the regression line parameter calculated with the walking characteristic parameter as a representative value is the default value. Practical accuracy by this value has been confirmed from several numerical verification results obtained in one division (Equation 8), which is considered to have an error larger than that in a plurality of divisions. When the walking characteristic is set, in principle, the error due to the deviation of the walking characteristic is eliminated, and only the error due to the simplification of the arithmetic expression. Thus, the error is greatly improved. In this way, the problems of the present invention are solved, and a pedometer that has at least practical accuracy, can be further improved, and has a simple structure and is economically inexpensive can be created.

In the above, the calculation method of the walking speed and the exercise intensity in the pedometer has been described in detail. This calculation method is derived based on two reference piecewise linear functions. Even if the number of sections of these piecewise linear functions and the linear expression in the sections change, it is not limited to pedometers, but also to portable devices such as activity meters and mobile phones, and also to jogging and running, not limited to walking exercises. Needless to say, the calculation method of the present invention can be applied or applied.

Hereinafter, an embodiment of the present invention will be described with reference to FIG. FIG. 3 is a block diagram showing the configuration of the pedometer using the simplified formula 9 and formula 2 described at the end of the present invention. Even in the case where the simplified equations 9 and 2 are used, even in the case where an ordinary calculation procedure is used without simplifying the two piecewise linear functions used as the basis of the present invention, or a part thereof is simplified. Even in this case, an arithmetic circuit corresponding to the required performance is selected. However, in either case, the configuration of FIG. 3 does not change significantly.

First, the unit time for calculating the walking pitch and the calculation cycle for making a round of various calculations will be described. Since the unit of pitch is usually steps / minute, the unit time is also set to 1 minute in the embodiment. There is no problem at other times.

In recent years, MPU (microprocessor) is mounted on many consumer products incorporating electronic circuits. The performance and functions of the MPU are diverse, and even if an MPU having a low calculation speed is selected with emphasis on economy, the unit time of 1 minute is sufficient for the MPU.

Even if the walking pitch is obtained from the latest step value and the step value before unit time and various kinds of walking related information are calculated, the required time will be about 100 ms and 200 ms. Therefore, waiting for the unit time of 1 minute when the next step value is obtained causes the MPU to play. Therefore, for example, 1/6 of 10 seconds is set as a calculation cycle, and a predetermined amount of memory is stored for the step value for each calculation cycle in such a manner that the oldest is discarded and the latest is left. Then, the step value before unit time is subtracted from the latest step value in the memory to obtain the step value of unit time in the calculation cycle, that is, the walking pitch. By adopting such a method, the MPU can be used efficiently.

MP U and peripheral functions, such as ROM for (read only memory) and RAM (Random Access Memory) memory unit, such as input and output unit, which is a technique pedometer and power unit according to the present invention is not directly related, Figure 3 is not shown. The explanation will proceed with the calculation cycle, memory, etc. in mind. For the sake of explanation, the calculation cycle is 10 seconds. One second would be possible. Although there is little change in the number of steps in 1 second, there is a disadvantage that the amount of memory increases. However, the responsiveness of the display unit (interface unit with the pedometer user) is poor in a 10-second cycle. Therefore, for example, the first 1 second of 10 seconds will be calculated, and the remaining 9 seconds will be processed by the display unit.

The blocks denoted by the numerals 1 to 8 in FIG. 3 will be described below. Before that, a personal computer (program) without the reference numerals in FIG. 3 will be described. In order to reflect the walking characteristics of the pedometer user in the pedometer, the walking characteristics parameters are input to the personal computer. The personal computer incorporates a program for outputting k and c, which are regression line parameters corresponding to the slope and intercept of the linear function of Equation 9, (the program relating to the above-mentioned binary simultaneous linear equations of γ and δ). . Then, the obtained regression line parameters k and c are input to the pedometer as set values. This minimizes the burden on the pedometer and reflects the walking characteristics of the pedometer user. This personal computer is necessary only when the regression line parameter is obtained, and it is desirable to obtain the personal computer prior to the start of use of the pedometer of the embodiment. The walking speed calculation uses the unified step calculation method just described. Accordingly, the set values are (k1 = k, c1 = c) and p1 = 500.

In the set value storage unit 1, the walking speed and the height L (cm) of the pedometer user necessary for calculating the calorie consumption, the regression line parameters k1 = k and c1 = c reflecting the walking characteristics, and the division point parameter p1 = 500 and weight W (kg) are stored as set values. In order to obtain regression line parameters, it is necessary to prepare in advance, such as collecting gait data and determining gait characteristic parameters, and installing a program on a personal computer. Considering the effort of this preparation, it is desirable that the regression line parameter can be set according to the request of the pedometer user. If not set, for example, the regression line parameter of Equation 8 determined from the representative value of the walking characteristic parameter described above may be used as the default value. This default value is stored in a ROM (not shown) and transferred to the set value storage unit 1 when the pedometer is initialized. And when it is set, it is overwritten. The set value in the set value storage unit 1 is protected by the battery of the pedometer and does not disappear.

The step count measurement unit 2 is a step count measurement unit based on walking exercise, which is a basic function of a pedometer. Any step counting means such as a pendulum type or an acceleration sensor type may be used. Using existing technology, it is possible to configure a measurement unit with high accuracy within standard errors.

The timer 3 transmits an activation signal every 10 seconds. This activation signal is given to the walking pitch calculation unit 4, and the calculation cycle starts. Various calculation processes shown below are calculation processes in the calculation cycle.

Based on the activation signal, the walking pitch calculation unit 4 reads the latest step count data of the step count measurement unit 2 and updates the step count data in such a manner that the oldest step count data is discarded and the latest step count data is left. To do. Then, the data 6 data before (unit time before) is subtracted from the latest data to obtain the unit time step value, that is, the walking pitch N (steps / minute) in the calculation cycle.

The walking pitch N that is the output of the walking pitch calculation unit 4 is the walking speed calculation unit 5 together with the height setting value L (cm) of the pedometer user set in the setting value storage unit 1, the regression line parameter, and the segment point parameter. Is read. Following the regression line parameters k1 and c1, the segment point parameter p1 is read, and p1 = 500 is recognized, and the parameter reading ends. Then, the walking speed V (km / hour) is calculated according to the step calculation method. If it demonstrates concretely, the walking pitch N read by the part 5 will be compared with p1. Since N ≦ p1 = 500, the process immediately proceeds to the calculation step, N and k1 are multiplied, c1 is added to this, and the obtained result is multiplied by the input height setting value L in the same part 5 to walk. The speed V is obtained and the calculation is finished.

The walking speed V output from the walking speed calculator 5 is input to the exercise intensity calculator 6, and the exercise intensity M (Mets) is calculated according to Equation 2. This will be specifically described. The slopes and intercepts of the linear equations 0.44 · V + 1.33, 0.94 · V−0.98, and 2.06 · V−8.03 for the three walking speeds V are stored in the ROM. The value of each primary expression with respect to the walking speed V input to the same unit 6 is calculated. Then, the maximum value of the obtained results is obtained as the exercise intensity M. Subtract 1 from the maximum value for physical activity only.

The exercise intensity M, which is the output of the exercise intensity calculation unit 6, is input to the calorie consumption calculation unit 7 together with the weight setting value W (kg) of the pedometer user set in the set value storage unit 1. According to Equation 1, calorie consumption K (kcal) equivalent to one hour is calculated with an implementation time of 1 (hour). More specifically, the exercise intensity M input to the part 7 is multiplied by the basal metabolic rate obtained by multiplying the weight setting value W input to the part 7 by a coefficient 1.05. This is the calorie consumption equivalent to one hour. The reason for setting the value to 1 hour is that it is not necessary to change the calculation in the same part 7 even if the calculation period is changed in design, as will be described below. In place of the above basal metabolic rate, the factor of 1.05 is an approximate number, or sex and age can be set in addition to height and weight. Basal metabolism can also be used.

The walking speed V in the walking speed calculation unit 5 and the calorie consumption K in the calorie consumption calculation unit 7 obtained in the calculation cycle are input to the calculation display unit 8. The calorie consumption K is a value corresponding to one hour. Since the walking speed V is also in units of hourly speed, the value of the walking speed V is a distance corresponding to one hour. By multiplying both by the calculation cycle time 1/360 (10 seconds), it is converted into a value in the calculation cycle. Further, in the same section 8, in order to display a walking distance and a cumulative value of calorie consumption, a cumulative calculation is also performed. These conversion calculation and total calculation are performed in common for V and K in the same section 8. The structure is simpler than when the walking speed calculation unit 5 and the calorie consumption calculation unit 7 individually perform conversion calculation and total calculation. In addition, even if the calculation cycle time is changed in design, there is a merit because only the calculation change in the part 8 is required.

The walking speed V and the calorie consumption K input to the calculation display unit 8 are updated in such a manner that they are added to each cumulative value before one calculation cycle. The cumulative value of the walking speed V and the calorie consumption K is multiplied by 1/360 to be converted into display data.

These display data can be confirmed by the pedometer user via the calculation display unit 8 together with the number of steps as walking related information. The setting value in the setting value storage unit 1 is also input to the same unit 8, and the setting value is confirmed or changed via the same unit 8.

Finally, the contents indicated by the dotted line in FIG. 3 will be described. The output of the exercise intensity calculation unit 6 is exercise intensity M (Mets). This value is the amount of physical activity per hour, that is, the value of exercise (Ex) itself. By multiplying this by the calculation cycle time 1/360 (10 seconds), it can be converted into a value in the calculation cycle. Further, the display format is a cumulative value. Therefore, if the output of the same unit 6 is directly input to the calculation display unit 8 (dotted arrow next to the figure) and the cumulative calculation and the conversion calculation are performed in the same manner as the above walking distance and calorie consumption, exercise is also related to walking. Can be provided as information. Note that the walking speed V and the exercise intensity M input to the part 8 may be handled not only as a cumulative value but also as an original immediate value. For example, the average value or the maximum value can be processed by the same unit 8 and displayed from the same unit 8 without significantly changing the structure.

The vertical dotted arrow in FIG. 3 will be described. As explained first, the amount of fat burning can be obtained by simply dividing the calorie consumption by the predetermined value 7.2. Therefore, when displaying this as walking related information, the value of the calorie consumption displayed may be divided by 7.2 and displayed.

Walking is an easy and effective exercise for health promotion and prevention of lifestyle-related diseases. And the importance of walking is stressed in many scenes. For this reason, it is considered that the number of users who use pedometers for health management increases year by year. The inventor of the present application also started to walk from the same motive and became interested in the pedometer. However, it has been found that the conventional pedometer technology has problems in terms of the accuracy of walking related information. In the present invention, the solution to this problem is theoretically explained in detail and provided as a specific technique that can be used immediately. Furthermore, a plurality of technologies are provided so that the technology can be selected according to the design goal in product design. From the above, it can be said that the industrial applicability of the present invention is extremely large.

DESCRIPTION OF SYMBOLS 1 Set value memory | storage part 2 Step count measuring part 3 Timer 4 Walking pitch calculating part 5 Walking speed calculating part 6 Exercise intensity calculating part 7 Consumption calorie calculating part 8 Calculation display part

Claims (5)

A setting means for setting the height of the pedometer user, a step count measuring means for measuring the number of steps when the pedometer user walks, and a step count per unit time measured by the step count measuring means is calculated as a walking pitch. A pedometer comprising a calculation means for calculating a walking speed based on a result obtained by substituting the walking pitch into an expression of a correction function determined as a straight line related to the walking pitch and a height set in the setting means. An inclination and an intercept that define the characteristics of the correction function can be further set in the setting means, and the slope and the intercept can be converted into a characteristic piecewise linear function relating to the walking pitch that defines the walking characteristics of the pedometer user. respect characteristics indicator quadratic function obtained by multiplying the characteristic classification plurality of at least one one obtained by performing regression analysis in section including the regression section of a linear function The slope and intercept of the line, the characteristics piecewise linear function, walking pitch changes in ratio height stride for walking pitch defining the gait characteristic of the pedometer user in each section was divided into a plurality in accordance with the change A pedometer characterized by being a piecewise linear function represented by a linear function. A setting means for setting the height of the pedometer user, a step count measuring means for measuring the number of steps when the pedometer user walks, and a step count per unit time measured by the step count measuring means is calculated as a walking pitch. The number of steps provided with a computing means for calculating a walking speed based on a result obtained by substituting the walking pitch into an expression of a correction function defined as a piecewise linear function related to the walking pitch and a height set in the setting means a total slope of the line and intercept of each section defining the characteristics of the correction function, and also can set the division point between the adjacent segment is the setting unit, and a straight line of slope and intercept of each of the sections, and division point between the adjacent segment, compared characteristics indicator quadratic function obtained by multiplying the walking pitch characteristics piecewise linear function for walking pitch defining the gait characteristic of the pedometer user, before Characteristics piecewise linear function of a plurality of slope and intercept of the regression line for each of the subdivision segments obtained by performing a regression analysis for each re-divided sections obtained by dividing into a plurality of sections including at least one of the sections, as well as an adjacent segment The characteristic segmentation linear function is a linear function related to the walking pitch for each section in which a change in the ratio of the stride height to the walking pitch that defines the walking characteristics of the pedometer user is divided into a plurality according to the change. A pedometer characterized by the piecewise linear function expressed by A setting means for setting the height of the pedometer user, a step count measuring means for measuring the number of steps during walking of the pedometer user, and calculating the number of steps per unit time measured by the step count measuring means as a walking pitch, A pedometer comprising a calculation means for calculating a walking speed based on a result obtained by substituting the walking pitch into an expression of a correction function defined as a piecewise linear function related to the walking pitch and a height set in the setting means. a is, the correction function with respect characteristics indicator quadratic function obtained by multiplying the walking pitch characteristics piecewise linear function for walking pitch defining the gait characteristics, at least one of the plurality of sections of the characteristic piecewise linear function said containing section for each re-divided sections obtained by dividing into a plurality obtained by performing a regression analysis is a regression line for each re-divided sections, the characteristics piecewise linear function, for walking pitch Pedometer characterized in that the change in the ratio height width is piecewise linear function expressed by a linear function relating walking pitch for each zone which is divided into a plurality in accordance with the change. The computing means is further a pedometer for calculating, as exercise intensity, a result obtained by substituting the walking speed into an expression of an exercise intensity function relating to the walking speed , wherein the exercise intensity function is a relationship of exercise intensity to walking speed. first operation to the current piece-wise linear function which is expressed, or a function obtained by performing the second operation, the first operation, the relative current piece-wise linear function, of the current piece-wise linear function It is an operation to replace with a new new piecewise linear function consisting of a plurality of regression lines obtained by performing regression analysis for each section divided into a plurality of sections including all sections , and the second operation is the current piecewise linear function or the continuous section of the new piecewise linear function, wherein the slope of the linear function of a continuous interval when gradual nondecreasing, a new value with a maximum value function of selecting the maximum value of the continuous interval function value of the linear function 1 Replaced by sections, replacing the gradual when the non-increasing one new to the minimum value function for selecting the minimum value of the function value interval, and if not its either subtracts one segment from the continuous interval, replaces pedometer according to any one of claims 1 to 3, characterized in that the operation of repeating the second operation until the possible eliminated. The computing means further calculates the walking distance and exercise by multiplying the walking speed and the exercise intensity by the walking time, calculates the calorie consumption by multiplying the exercise by the basal metabolic rate, and is measured by the step count measuring means. A pedometer comprising a display means for displaying the walking speed, walking distance, exercise intensity, exercise, or calorie consumption calculated by the computing means together with the number of steps taken , wherein the basal metabolic rate is the setting A weight value set in addition to the height set in the means, a value obtained by multiplying the weight by a factor of 1.05, or a gender and age set in addition to the setting means, and the height and the weight The pedometer according to claim 4, wherein the pedometer is one of values calculated from:

Images