JP2000316832A - Sleep monitoring device and awakening device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an awakening device allowing for the scheduled time of rising and allowing the user to wake up exactly during REM sleep for a pleasant feeling of rising. SOLUTION: When the number of body movements detected decreases and becomes much smaller than the number of body movements in an awakened condition, this awakening device determines that the user has shifted into a sleeping condition, and when the number of body movements detected after the shift into the sleeping condition increases and becomes much greater than the number of body movements during sleep, the device determines that the user has shifted into REM sleep. According to the time of shift into REM sleep and the general interval and period of REM sleep, the device then determines whether the scheduled time of rising is within or out of the period of REM sleep (step s51), and when it is within that period, the time to awaken the user is set as the scheduled time of rising (step s52); when it is out of the period, REM sleep located before the preset scheduled time of rising is specified (step s53), so that the time to awaken the user is equal to the start time of the specified REM sleep plus less than fifteen minutes (step s55).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sleep monitoring device for determining whether or not a user is in REM sleep, and a wake-up device capable of providing a comfortable wake-up feeling.

[0002]

2. Description of the Related Art When a person enters a sleep state, a human repeatedly repeats non-REM sleep with a small number of body movements and REM sleep with a large number of body movements. It is said that if you wake up in REM sleep, you will not sleepy and will have a comfortable wake-up feeling. Therefore, the following technology has been conventionally proposed.

The alarm clock described in Japanese Patent Application Laid-Open No. 59-23284 (conventional example 1) uses a vibration sensor and determines REM sleep from the output value. In addition, Japanese Unexamined Patent Publication
In the wake-up device described in Japanese Patent No. 245148 (conventional example 2), biological information of body movement, heartbeat, and respiration of a sleeping person is used as data of a neural network, and the sleeping feeling of the sleeping person is input.
The good sleep feeling of the sleeping person is obtained by performing the collation learning.

[0004]

However, the above-mentioned conventional technique has the following disadvantages. Determining whether or not REM sleep is detected by detecting the body motion component pattern in the vibration component is accurate because it is easily affected by disturbances such as bad sleep and standing in the toilet, as well as large individual differences in body motion at bedtime Poor properties (conventional example 1).

[0005] It is necessary to input very vague information that is easily influenced by the sleeping time and environment of the sleeping person, which is the sleeping feeling of the sleeping person, and the sleep wake-up is inaccurate at an incorrect timing (conventional example 2).

A first object of the present invention is to provide a sleep monitoring device capable of accurately determining whether a sleeping user is in REM sleep. A second object of the present invention relates to a wake-up device capable of accurately waking up during REM sleep while taking into account the scheduled wake-up time, and providing a comfortable wake-up feeling.

[0007]

[Means for Solving the Problems] A plurality of load sensors installed at a predetermined distribution under, inside, or on the surface of the bedding output a load signal of a level corresponding to the applied load. The load distribution calculating means calculates a characteristic amount of the load distribution from the plurality of load signals. The storage unit stores the characteristic amount of each load distribution when the user is in each sleeping state.

The sleeping form detecting means compares the characteristic amount of the load distribution of each sleeping form stored in the storage means with the characteristic amount of the load distribution calculated by the load distribution calculating means to determine the current sleeping state of the user. Detect the figure. The position specifying means specifies the position of each part of the body based on the detected sleeping posture. The body movement detecting means counts the number of body movements in which the specified position of each body part has changed for a predetermined time. The REM sleep determining means determines whether or not the user is in REM sleep from the counted number of body movements.

The sleep monitoring apparatus is configured to count the number of body movements of each part of the user's body during sleep for a predetermined period of time and determine whether or not the subject is in REM sleep from the counted number of body movements. Therefore, it can be accurately determined whether or not the sleeping user is in REM sleep. This allows another person to wake up the user accurately during REM sleep.

[0010] [Claim 2] A plurality of load sensors installed at a predetermined distribution under, inside, or on the surface of the bedding outputs a load signal at a level corresponding to the applied load.
The load distribution calculating means calculates a characteristic amount of the load distribution from the plurality of load signals. The characteristic amount storage means stores the characteristic amount of each load distribution when the user is in each sleeping state.

The sleeping form detecting means compares the characteristic amount of the load distribution of each sleeping form stored in the characteristic amount storing means with the characteristic amount of the load distribution calculated by the load distribution calculating means to determine the current user's current state. Detects the sleeping shape of. The position specifying means specifies the position of each part of the body based on the detected sleeping posture.

The body movement detecting means counts the number of body movements in which the position of the specified body part has changed for a predetermined time. The body movement frequency storage unit stores the number of body movements counted by the body movement detection unit during awakening and during falling asleep.

The wake-up time setting means sets an expected wake-up time to be awakened. When the sleeping state is detected by the sleeping state detecting means after the level change of the load signal, the sleeping state determining means determines that the user has gone to bed.

If the number of body movements detected by the sleeping form detecting means is significantly reduced from the number of body movements during awakening after it is determined that the user has fallen asleep, it is determined that the user has fallen asleep. Judge.

If the number of body movements detected by the body movement detecting means greatly increases after the user has determined that the user has transitioned to a sleep state, the number of body movements during sleep has increased. The determining means determines.

After determining that the user has shifted to REM sleep, the wake-up means determines whether the set scheduled wake-up time is within or outside the REM sleep period based on the general REM sleep interval and period. Is determined, and if it is on, the annunciator is activated at the scheduled wake-up time. In this case, the user can wake up at the set scheduled wake-up time.

If the set scheduled wake-up time is out of the REM sleep period, the wake-up means identifies the REM sleep located before the set scheduled wake-up time and significantly increases the number of body movements. The alarm is activated after a set time has elapsed since the specific REM sleep is detected. In this case, the user can wake up at the end of REM sleep near the set scheduled wake-up time before the scheduled wake-up time. In any case, since it is possible to wake up accurately during REM sleep, a comfortable wake-up feeling is obtained.

[Claim 3] Each time it is determined that the sleep mode has shifted to REM sleep, the wake-up means sets the set wake-up time based on the appearance pattern of REM sleep determined up to the present time to the REM sleep period. This is a configuration in which it is determined whether to enter or leave, and if it does, REM sleep located before the set scheduled wake-up time is specified. For this reason, it is possible to accurately determine whether the scheduled wake-up time is within the REM sleep period, and specify the REM sleep located before the scheduled wake-up time.

When the set scheduled wake-up time deviates from the REM sleep period, the wake-up means identifies the REM sleep positioned before the set scheduled wake-up time for each REM sleep determination and significantly increases the number of times of body movement. The alarm is activated after a set time has elapsed since the specific REM sleep is detected due to the increase. In this case, the user can wake up at the end of REM sleep near the set scheduled wake-up time before the scheduled wake-up time.

When the set scheduled wake-up time is within the REM sleep period, the wake-up means activates the alarm when the scheduled wake-up time is reached. In this case, the user can wake up at the set scheduled wake-up time.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention (Claim 1,
2) will be described with reference to FIGS. As shown in FIG. 1, the wake-up device A includes a sensor sheet 2 inserted below a mattress 10 (bedding) installed on a bed 1, a controller 3, and a wake-up time setting for setting an expected wake-up time. And an alarm device 5.

The bed 1 has a placing portion 11 on which the mattress 10 is placed, and a back plate portion 12 standing upright from an end of the placing portion 11. The sensor sheet 2 includes a pressure-sensitive element 21 whose electric resistance value changes in accordance with the load applied to the element.
(Interlink FSR sensor)
Four are arranged and attached to the lower part of the mattress 10.

As shown in FIG. 2, the controller 3 comprises a multiplexer circuit 301, an A / D converter 302, a parallel I / O
303, a memory 304, and an ECU 305.
The multiplexer circuit 301 processes the load signal from each pressure-sensitive element 21 based on the switching signal sent from the parallel I / O 303. The multiplexer circuit 301 converts a parallel signal into a serial analog voltage signal and sends the signal to an A / D converter 302. A / D
The converter 302 converts a serial analog voltage signal into a load value and inputs the load value to the ECU 305.

The memory 304 is a storage device for storing a characteristic amount of each load distribution when the user is in each sleeping state, and the number of body movements counted while awake and falling asleep. The ECU 305 is a microcomputer, and performs a feature amount storage flow, a load distribution calculation flow, a sleeping form detection flow, a position identification flow, a body movement detection flow, a body movement frequency storage flow, a sleeping state determination flow, and a wake-up flow, which will be described later. Do.

The feature amount storage flow is an operation performed before using the alarm device A. In the state where the user is in each sleeping state, the feature of each load distribution calculated by the load distribution calculation flow described later is used. This is an operation of storing the amount in the memory 304 (see FIG. 3).

The load distribution calculation flow is an operation of comparing the threshold value and the load value output by the A / D converter 302 to calculate the binarized load distribution characteristic amount (see FIG. 3). . The sleeping shape detection flow compares the feature amount of the load distribution of each sleeping shape of the user stored in the memory 304 with the feature amount of the load distribution calculated by the load distribution calculating flow, and compares the feature amount of the load distribution calculated by the load distribution calculating flow. This is an operation of detecting a sleeping state (see FIG. 3).

The position specifying flow is an operation of specifying each position of each part of the body (head, waist, feet, both hands) based on the sleeping shape detected by the sleeping shape detecting flow and the skeleton model.

The body movement detection flow is an operation for counting the number of body movements in which the position of each body part calculated by the position specifying flow has changed for a predetermined time (set between 1 minute and 10 minutes). The body movement count storage flow is an operation of storing the number of body movements counted by the body movement detection flow in the memory 304 while awake and falling asleep. The sleeping state determination flow is an operation of determining which sleeping state (awake, sleeping, or REM sleep) the user is currently in. The wake-up flow determines when to wake up the user,
This is an operation for operating the alarm 5 at that time.

Next, the operation of the alarm device A will be described with reference to FIGS.
This will be described based on the flowchart shown in FIG. In addition,
The feature amount of each load distribution in the state where the user takes each sleeping state is already stored in the memory 304.

FIG. 4 shows the flow from going to bed to getting up. The user (or another person) sets the scheduled wake-up time using the wake-up time setting device 4 (step s1). At step s2, a floor presence determination is made, and it is determined that the user is present (YE).
S) and proceeds to step s3, and if it is determined that the user is not occupied (NO), step s2 is maintained.

The details of the in-bed determination in step s2 are shown in steps s21 to s27 of FIG. At step s21, A
It is determined whether or not there is a change in the load value output from the / D converter 302. If there is a change (YES), the process proceeds to step s22. If there is no change (NO), the process proceeds to step s22.
Maintain 21.

In step s22, a sleeping posture detection flow is executed to determine whether or not the person is on the floor (whether or not a sleeping posture is detected). Specifically, it is determined whether or not the characteristic amount of the load distribution calculated by the load distribution flow matches any of the characteristic amounts of the load distribution of each sleeping form stored in the memory 304.

If it is determined that the user is on the floor, a position specifying flow is performed in step s23, and a body movement detecting flow is subsequently performed in step s24. In particular,
Based on the sleeping posture determined in step s22, each position of each part of the body (head, waist, feet, both hands) is determined in step s23.
To specify. If the position of each part of the body (head, waist, feet, both hands) specified in step s23 has changed, counting is performed in step s24.

In step s25, it is determined whether or not a predetermined time (several minutes) has elapsed. If the predetermined time has not elapsed (NO), the flow proceeds to step s26, and the sleeping posture detection flow is performed again. I do. Further, a predetermined time elapses (YE
S), and proceeds to step s27. In step s27, the body movement frequency storage flow is performed, and the counted number is stored in the memory 304 as the body movement frequency during awakening.

At step s3, a fall asleep is determined, and if it is determined that the user is asleep (YES), the process proceeds to step s4, and if it is determined that the user is not asleep (NO), step s3 is maintained. I do.

Details of the sleep onset determination in step s3 are shown in steps s31 to s36 in FIG. In step s31, a sleeping form detection flow is performed. Specifically, the current sleeping state of the user is detected by comparing the characteristic amount of the load distribution calculated by the load distribution flow with the characteristic amount of the load distribution of each sleeping state stored in the memory 304.

In step s32, a position specifying flow is performed, and subsequently, in step s33, a body movement detecting flow is performed. Specifically, each position of each part of the body (head, waist, feet, both hands) is specified in step s33 based on the sleeping form determined in step s32. If the position of each part of the body (head, waist, feet, both hands) specified in step s32 has changed, counting is performed in step s33.

In step s34, it is determined whether or not a predetermined time (several minutes) has elapsed. If the predetermined time has not elapsed (NO), the flow returns to step s31, and the sleeping shape detection flow is performed again. I do. Further, a predetermined time elapses (YE
S), and the process proceeds to step s35.

In step s35, the counted number of body movements is compared with the number of body movements at the time of awakening stored in the memory 304, and the counted number of body movements is significantly reduced from the number of body movements at the time of awakening. Is determined. If it has decreased significantly (YES), the process proceeds to step s36. If it has not decreased significantly (NO), the process returns to step s31.

At step s36, it is determined that the apparatus has shifted to the sleep state, and the counted number of body movements is stored in the memory 304 as the number of body movements during falling asleep.

In step s4, REM sleep determination is performed. If it is determined that the user is in REM sleep (YES), the process proceeds to step s5, and it is determined that the user is not in REM sleep (NO). Step s4 is maintained.

FIG. 7 shows details of the REM sleep determination in step s4.
Are shown in steps s41 to s46. In step s41, a sleeping form detection flow is performed. Specifically, the current sleeping state of the user is detected by comparing the characteristic amount of the load distribution calculated by the load distribution flow with the characteristic amount of the load distribution of each sleeping state stored in the memory 304.

In step s42, a position specifying flow is performed, and subsequently, in step s43, a body movement detecting flow is performed. Specifically, each position of each part of the body (head, waist, feet, both hands) is specified in step s43 based on the sleeping posture determined in step s42. If the position of each part of the body (head, waist, feet, both hands) specified in step s42 has changed, counting is performed in step s43.

In step s44, it is determined whether or not a predetermined time (several minutes) has elapsed. If the predetermined time has not elapsed (NO), the flow returns to step s41, and the sleeping shape detection flow is performed again. I do. Further, a predetermined time elapses (YE
S) and proceed to step s45.

In step s45, the counted number of body movements is compared with the number of body movements during sleep stored in the memory 304, and the counted number of body movements is significantly increased from the number of body movements during sleep. Is determined. If it has increased significantly (YES), the process moves to step s46, and if it has not increased significantly (NO), the process returns to step s41. It is determined in step s46 that the state has shifted to the REM sleep state.

The details of the determination of the time at which step s5 occurs are shown in steps s51 to s55 of FIG. 8 and FIG. In step s5, the ECU 305 performs a wake-up flow.

Based on the time determined to have shifted to REM sleep in step s46, the general REM sleep interval (90 minutes) and the general REM sleep duration (15 minutes), the wake-up time is set. Step s5: whether the scheduled wake-up time set by the device 4 is in or out of the REM sleep period
It is determined in step 1 and if it is included (YES), step s
Go to step 52, if not (NO), step s
Go to 53.

In step s52, the wake-up time is set as the scheduled wake-up time. In step s53, based on the general REM sleep interval (90 minutes) and the general REM sleep duration (15 minutes), the REM sleep located before the set scheduled wake-up time is specified. .

In step s54, the start timing of the specific REM sleep is detected. Specifically, when the scheduled start time of the specific REM sleep approaches, the following operation is performed.

The sleeping shape is detected by the load distribution flow. Each part of the body (head, waist, feet,
Specify each position of both hands). The number of body movements in which the position of each body part (head, waist, foot, both hands) whose position is specified has changed is counted for a predetermined time.

When the predetermined time has elapsed, the counted number of body movements is compared with the number of body movements during sleep stored in the memory 304, and the counted number of body movements is significantly greater than the number of body movements during sleep. If it has increased, it is determined that the user has entered the specific REM sleep state. Steps
At 55, the time to wake up is the start time of specific REM sleep +15
Less than a minute. In step s6, the buzzer, which is the alarm 5, is activated at the time of waking up.

The alarm device A of this embodiment has the following advantages. [A] The user can wake up during the REM sleep period by operating the alarm 5, so that a comfortable wake-up feeling can be obtained. If the set scheduled wake-up time is within the REM sleep period, the user can wake up at the scheduled wake-up time. In addition, when the scheduled wake-up time is out of the REM sleep period, the user can wake up at the end of REM sleep close to the set scheduled wake-up time before the scheduled wake-up time, so that the user does not oversleep. , It is possible to secure a sufficient sleeping time.

[A] The alarm device A determines that if the number of body movements detected by the body movement detection flow is significantly greater than the number of body movements during falling asleep after determining that the user has transitioned to a sleep state, Since the configuration is such that it has transitioned to REM sleep, it is possible to accurately detect the transition to REM sleep.

Next, a second embodiment of the present invention (Claim 1,
3 corresponding to FIG. 10 (FIGS. 1, 2, 3, 4, 5, 6,
It will be described based on (7, 8, 9). The alarm device of this embodiment determines whether the scheduled wake-up time is in or out of the REM sleep period and specifies the specific REM sleep based on the REM sleep appearance pattern determined up to the present time.
5 is different from the wake-up device A.

Next, the operation of the wake-up device of this embodiment will be described with reference to the flowchart shown in FIG. 10 (see FIGS. 4, 5, 6, and 7). The characteristic amounts of the respective load distributions in a state where the user is in each sleeping state are already stored in the memory 30.
4 is stored.

The user (or another person) sets the expected wake-up time using the wake-up time setting device 4 (step s1). At step s2, the presence determination is performed. If it is determined that the user is present (YES), the process proceeds to step s3, and if it is determined that the user is not present (NO), step s2 is maintained. I do.

Details of the in-bed determination in step s2 are shown in steps s21 to s27 of FIG. At step s3, a fall asleep determination is made, and it is determined that the user is asleep (YE).
S) and proceed to step s4, and if it is determined that the user has not fallen asleep (NO), step s3 is maintained.

Details of the sleep onset determination in step s3 are shown in steps s31 to s36 in FIG. In step s4, REM sleep determination is performed. If it is determined that the user is in REM sleep (YES), the process proceeds to step s7, and if it is determined that the user is not in REM sleep (NO), step s4 is performed. maintain.

FIG. 7 shows details of the REM sleep determination in step s4.
Are shown in steps s41 to s46. Based on the time determined to have shifted to REM sleep in step s46 and the appearance pattern of REM sleep determined up to the present time, the scheduled wake-up time set by the wake-up time setting device 4 enters the REM sleep period. It is determined in step s7 whether or not it has been removed. If it has been entered (YES), the process proceeds to step s8. If it has not been entered (NO), the process proceeds to step s10.

In step s8, it is determined whether or not the current REM sleep is the last REM sleep. If the current REM sleep is the last REM sleep (YES), the process proceeds to step s9, and if the current REM sleep is not the last REM sleep (NO). Returns to step s4. In step s9, the wake-up time is set as the scheduled wake-up time, and step s6
Proceed to.

At step s10, the REM sleep located before the set scheduled wake-up time is specified based on the REM sleep appearance patterns (interval and duration) determined up to the present time. In step s11, it is determined whether or not the current REM sleep is the specific REM sleep. If the current REM sleep is the specific REM sleep (YES), the process proceeds to step s12. If the current REM sleep is not the specific REM sleep (NO), the process proceeds to step s4. Return.

In step s52, the wake-up time is set to be less than the start time of the specific REM sleep + 15 minutes. Step s6
Then, the buzzer, which is the alarm 5, is activated at the time of waking up.

The wake-up device according to the present embodiment includes the above [A],
It has the following advantages in addition to the advantages according to [A]. [C] Every time it is determined that the system has shifted to REM sleep, the ECU
305, based on the appearance pattern of the REM sleep determined up to the present time, determines whether the set scheduled wake-up time enters or goes out of the REM sleep period, and if not, before the set scheduled wake-up time This is a configuration for identifying the located REM sleep. For this reason, it is possible to accurately determine whether the scheduled wake-up time is within the REM sleep period, and specify the REM sleep located before the scheduled wake-up time.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of a wake-up device according to a first embodiment of the present invention.

FIG. 2 is a block diagram of the alarm device.

FIG. 3 is an explanatory diagram when the wake-up device detects a sleeping state.

FIG. 4 is a flowchart showing the operation of the alarm device.

FIG. 5 is a flowchart showing an operation of a part related to the presence determination of the alarm device.

FIG. 6 is a flowchart showing the operation of a part related to sleep onset determination of the wake-up device.

FIG. 7 is a flowchart showing an operation of a part relating to REM sleep determination of the alarm device.

FIG. 8 is a flowchart showing the operation of a part for determining the time of the wake-up device.

FIG. 9 is a graph showing the relationship between elapsed time and the number of body movements.

FIG. 10 is a flowchart showing the operation of the wake-up device according to the second embodiment of the present invention.

[Explanation of symbols]

A Wake-up device 4 Wake-up time setting means (wake-up time setting device) 5 Alarm 10 Mattress (bedding) 21 Pressure-sensitive element (load sensor) 304 Memory (storage means, feature quantity storage means, body movement frequency storage means)

Continued on the front page (72) Inventor Masahiko Ito 1-1-1 Showa-cho, Kariya-shi, Aichi F-term in DENSO Corporation (reference) 4C038 VA15 VB35 VC20

Claims (3)

[Claims] 1. A plurality of load sensors installed at a predetermined distribution under, inside, or on the surface of a bedding and outputting a load signal at a level corresponding to an applied load; A load distribution calculating means for calculating the characteristic amount of each load distribution in a state where the user is in each sleeping state; and a characteristic amount of the load distribution of each sleeping state stored in the storage means. And a sleeping form detecting means for detecting the current sleeping state of the user by comparing the characteristic amount of the load distribution calculated by the load distribution calculating means, and a position of each part of the body based on the detected sleeping state. Position specifying means for specifying the position, body movement detecting means for counting the number of body movements at which the position of the specified body part has changed for a predetermined time, and the user is in REM sleep from the counted number of body movements REM sleep judgment to determine whether or not Sleep monitoring device comprising: 2. A plurality of load sensors installed under, under, or on the bedding with a predetermined distribution and outputting a load signal at a level corresponding to an applied load, and a characteristic amount of a load distribution based on the plurality of load signals. A load distribution calculating means for calculating a characteristic amount of each load distribution in a state where the user is in each sleeping state; and a load of each sleeping form stored in the characteristic amount storing means. Based on the detected sleeping shape, the distribution feature amount is compared with the load distribution feature amount calculated by the load distribution calculating device to detect the current sleeping shape of the user. Position specifying means for specifying the position of each body part, body movement detecting means for counting the number of body movements in which the specified position of each body part has changed for a predetermined period of time, and Record the number of movements counted Means for storing the number of body movements to be remembered; sleeping state determining means for determining what kind of sleeping state the user is in; wake-up time setting means for setting a scheduled wake-up time; Awakening means for activating an alarm at that time, setting a wake-up expected time desired to be awakened by the wake-up time setting means, and after the level of the load signal changes, the sleeping state is detected by the sleeping state detecting means. The sleeping state determination means determines that the user has fallen asleep on the bedding, and after the sleep determination, if the number of body movements detected by the determination of the body movement detecting means is significantly reduced from the number of body movements during awakening. The sleeping state determination means determines that the user has transitioned to the sleep state, and after the transition determination to the sleep state, the number of body movements detected by the body movement detection means indicates the body movement during sleep. When the number significantly increases, the sleeping state determination means determines that the user has shifted to REM sleep, and after the determination to shift to REM sleep, is set based on a general REM sleep interval and period. The wake-up means determines whether the scheduled wake-up time is in or out of the REM sleep period, and if so, the wake-up means activates the annunciator when the wake-up scheduled time comes, and the wake-up time is out. The wake-up means specifies the REM sleep located before the set scheduled wake-up time, and activates the alarm after a lapse of a set time from the detection of the specific REM sleep due to a large increase in the number of body movements. Alarm device. 3. A plurality of load sensors installed under, under, or on the bedding with a predetermined distribution and outputting a load signal at a level corresponding to an applied load, and a characteristic amount of a load distribution based on the plurality of load signals. Load characteristic calculating means for calculating the characteristic amount of each load distribution in a state where the user is in each sleeping state, and the load of each sleeping form stored in the characteristic amount storing means. Based on the detected sleeping state, the distribution characteristic amount is compared with the load distribution characteristic amount calculated by the load distribution calculating means to detect a current sleeping state of the user. Position specifying means for specifying the position of each part of the body, body movement detecting means for counting the number of body movements in which the position of the specified body part has changed for a predetermined time, and the body movement detecting means during awakening and falling asleep Record the number of movements counted Means for storing the number of body movements to be remembered; sleeping state determining means for determining what kind of sleeping state the user is in; wake-up time setting means for setting a scheduled wake-up time; Awakening means for activating an alarm at that time, setting a wake-up expected time to be awakened by the wake-up time setting means, and after the level of the load signal changes, the sleeping state is detected by the sleeping state detecting means. The sleeping state determination means determines that the user has fallen asleep on the bedding, and after the sleep determination, if the number of body movements detected by the determination of the body movement detecting means is significantly reduced from the number of body movements during awakening. The sleeping state determination means determines that the user has transitioned to the sleep state, and after the transition determination to the sleep state, the number of body movements detected by the body movement detection means indicates the body movement during sleep. When the number of times is significantly increased, the sleeping state determination means determines that the user has shifted to REM sleep, and the set scheduled wake-up time is set to REM sleep based on the appearance pattern of REM sleep determined up to the present time. The wake-up means determines whether to enter or exit the period for each REM sleep determination, and if it deviates, the wake-up means identifies the REM sleep located before the set scheduled wake-up time for each REM sleep determination. An alarm device that activates the alarm after a lapse of a set time after a specific REM sleep is detected due to a large increase in the number of body movements, and the alarm device activates the alarm when the scheduled wake-up time is reached.