IoT-Based Smart Cradle System with Emotional Stabilization Function and Method for Managing Newborn Health Using the Same

Abstract

The present invention relates to an IoT-based smart cradle system including an emotional stability function and a method for managing newborn health using the same, comprising: a biometric information measurement unit that measures the biometric information of a newborn in real time; an emotional stability induction unit that analyzes the measured biometric information and outputs an emotional stability stimulus according to a determined emotional state; an intelligent control unit that receives biometric information, learns emotional response patterns for each newborn, and controls the emotional stability induction unit according to a determined emotional state; a wireless communication unit that transmits and receives output data from the intelligent control unit in real time to an external terminal or server; and a user interface unit that outputs the history of changes in biometric information and the status of stimulus execution according to the control of the intelligent control unit and receives user commands.

Description

Technology Field

[0001] The present invention relates to an IoT-based smart cradle system including an emotional stability function and a method for managing newborn health using the same. More specifically, the invention relates to an IoT-based smart cradle system including an emotional stability function configured to analyze the emotional state and feeding response based on the newborn's bio-information, and to manage the newborn's condition in an integrated manner through emotional stimulation and response to abnormal conditions, and a method for managing newborn health using the same. Background Technology

[0002] Recently, there has been an increasing demand for smart healthcare devices that monitor a newborn's health status in real time to determine the timing and amount of feeding. In particular, technologies are being developed to analyze a newborn's biological rhythms by linking them with satiety immediately after feeding, sleep patterns, and bowel movement intervals, and to provide appropriate feeding patterns.

[0003] Existing feeding management systems typically rely on manually recording data based on feeding times, amounts, or observations directly entered by guardians or medical staff, and provide feeding reminders based on schedules. This approach has limitations, as it fails to reflect the individual characteristics of newborns and makes it difficult to accurately measure actual feeding amounts or link physiological responses in real time.

[0004] Meanwhile, while weight-based feeding volume measurement technology can provide quantitative information within a certain range, it has drawbacks such as vulnerability to environmental vibrations and difficulty in obtaining reliable measurements due to interference from slings, bedding, etc. Additionally, if feeding status is interpreted solely based on weight changes before and after feeding, errors may occur due to various disturbance factors such as changes in posture and temporary load deviations; if these errors are not refined, the usability of the data decreases.

[0005] To address these issues, research is being conducted in some systems to precisely collect biological data by introducing advanced sensor modules or AI analysis techniques, and to predict the timing and amount of feeding based on this data. However, most technologies rely on fragmentary information based on a single sensor or lack calibration algorithms to improve analysis accuracy, often making stable application in real-world environments difficult.

[0006] In addition, while there are technologies that analyze only the timing and amount of feeding independently, technologies that quantitatively analyze feeding rhythms by linking them with status information such as a newborn's sleep cycle, bowel movements, and abdominal distension, or provide intuitively visualized reports to guardians and medical staff, have not yet been sufficiently implemented.

[0007] Accordingly, there is a need for technology capable of collecting various biological and environmental information in combination, refining errors, estimating actual feeding amounts based on this, and providing a comprehensive report on the feeding cycle and health status of newborns.

[0008] Accordingly, there is a need for an IoT-based smart cradle system equipped with an emotional stability function configured to analyze emotional states and feeding responses based on the newborn's biometric information, and to manage the newborn's condition in an integrated manner by responding to emotional stimuli and abnormal states, as well as a method for newborn health management using such a system. Prior art literature

[0009] Korean Patent Publication No. 10-2018-0095366 (Published Aug. 27, 2018) The problem to be solved

[0010] The present invention aims to solve the aforementioned problems by providing an IoT-based smart cradle system that includes an emotional stability function, which allows for the quantitative verification of a newborn's nutritional intake status without subjective judgment by the guardian, and a method for managing newborn health using the same, by continuously detecting biological information such as the newborn's weight, body temperature, and heart rate, and automatically calculating the amount of feeding based on weight changes before and after feeding.

[0011] In addition, the present invention aims to provide an IoT-based smart cradle system that includes an emotional stability function capable of inducing emotional stability through various means such as vibration, music, and white noise, and in addition to biological information, it collects various sensory data such as the newborn's body movement patterns, crying recognition, and skin contact, estimates the newborn's emotional state based on this data, and manages newborn health using the same.

[0012] In addition, the present invention aims to provide an IoT-based smart cradle system that includes an emotional stability function capable of long-term and structural monitoring of a newborn's health condition by accumulating and recording time-based data such as feeding times, sleep cycles, and weight change trends, and predicting the possibility of abnormal signs occurring, and a method for managing newborn health using the same.

[0013] In addition, the present invention aims to provide an IoT-based smart cradle system including an emotional stability function that enables remote monitoring-based care coordination and rapid intervention by linking with a guardian terminal or a medical staff server to transmit the newborn's vital signs and changes in emotional state in real time, and by providing immediate notifications in the event of an exception, and a method for managing the health of a newborn using the same.

[0014] The problem to be solved by this specification is not limited to what is described above and can be extended to various matters that can be derived from the embodiments of the invention described below. means of solving the problem

[0015] An IoT-based smart cradle system (100) including an emotional stability function according to one embodiment of the present invention may include: a biometric information measuring unit (110) that measures biometric information of a newborn in real time; an emotional stability induction unit (120) that analyzes biometric information collected from the biometric information measuring unit (110) and outputs an emotional stability stimulus according to a determined emotional state; an intelligent control unit (130) that receives biometric information collected from the biometric information measuring unit (110), learns an emotional response pattern for each newborn, controls the operation of the emotional stability induction unit (120) according to a determined emotional state, and determines a change in state based on biometric information in real time; a wireless communication unit (140) that transmits and receives biometric information and analysis results output from the intelligent control unit (130) in real time to an external terminal or server; and a user interface unit (150) that outputs the history of changes in biometric information and the status of the performance of emotional stability stimuli in visual or auditory form according to the control of the intelligent control unit (130), and receives commands from parents or medical staff.

[0016] According to one embodiment, the bio-information measuring unit (110) can continuously measure pre-feeding weight data and post-feeding weight data at regular time intervals, refine the measured weight change amount using a polynomial regression-based filter, cluster the cumulative distribution of the refined weight change amount by feeding event unit, analyze the trend of the feeding amount by time period, collect bio-response data such as body temperature, heart rate, and sleep transition point measured for a certain period after feeding, calculate a feeding sensitivity score based on the rate of change between the feeding response and the bio-response, and transmit the feeding sensitivity score to the intelligent control unit (130).

[0017] According to one embodiment, the intelligent control unit (130) arranges feeding history data, which consists of an estimated feeding amount and feeding interval information, into a time series, extracts a feeding rhythm pattern by frequency converting the feeding history data, predicts the next feeding timing based on the extracted feeding rhythm, classifies feeding response data, which consists of body temperature, heart rate, and sleep transition time measured immediately after feeding, into response types based on a multivariate state-space model, and adjusts the stimulation parameters of the emotional stability induction unit (120) according to the classified response types.

[0018] According to one embodiment, the wireless communication unit (140) can encrypt newborn status data and feeding history data with an AES-based symmetric key, generate a hash authentication token including current location data of the cradle, and transmit the encrypted data and the authentication token together when transmitting to a server.

[0019] A method for stabilizing the emotions of a newborn using an IoT-based smart cradle system including an emotional stabilization function according to another embodiment of the present invention may include: a step of measuring the newborn's biological information in real time through a biological information measurement unit; a step of analyzing the biological information provided from the biological information measurement unit through an emotional stabilization induction unit and outputting an emotional stabilization stimulus according to a determined emotional state; a step of receiving the biological information provided from the biological information measurement unit through an intelligent control unit, learning the emotional response pattern for each newborn, controlling the operation of the emotional stabilization induction unit according to a determined emotional state, and determining the state change based on the biological information in real time; a step of transmitting and receiving the biological information and analysis results output from the intelligent control unit in real time with an external terminal or server through a wireless communication unit; and a step of outputting the history of changes in the biological information and the status of the performance of the emotional stabilization stimulus in visual or auditory form according to the control of the intelligent control unit through a user interface unit, and receiving commands from parents or medical staff. Effects of the invention

[0020] According to one embodiment of the present invention, an IoT-based smart cradle system comprising a bio-information measuring unit, a motion detection unit, an emotional stability providing unit, and a health analysis unit has the advantage of realizing an emotion-based health management function by comprehensively detecting and analyzing not only the physiological state but also the emotional response of a newborn.

[0021] In addition, according to the present invention, the amount of breastfeeding intake can be automatically analyzed without separate user intervention through a high-precision 6-axis load cell-based bio-information measuring unit capable of detecting the difference in weight before and after breastfeeding in real time, thereby allowing for the quantitative assessment of the nutritional status of a newborn and the early detection of abnormalities in the periodic breastfeeding pattern.

[0022] Furthermore, according to the present invention, through a structure capable of automatically controlling the sound, vibration, and lighting outputs of an emotional stability provider based on data such as the newborn's movements, posture changes, and crying patterns collected from a motion detection unit, an autonomous stability induction system that responds in real time to the newborn's emotional responses can be implemented, thereby providing the advantage of maintaining an emotionally stable state without the intervention of a guardian.

[0023] In addition, according to the present invention, the health analysis unit comprehensively analyzes the health status of a newborn based on multiple factors such as feeding intake, weight change, crying frequency, and sleep cycle, and provides an immediate notification to a guardian's terminal upon detection of an abnormality, thereby enabling a preemptive response to abnormal conditions and consequently providing the advantage of ensuring real-time safety.

[0024] In addition, according to the present invention, since all collected biological information and response data can be stored and managed long-term through cloud integration, guardians can visually check the trends of the newborn's growth changes over time, and there is an advantage in that the data can be utilized as health history analysis data.

[0025] Furthermore, according to the present invention, the entire system is configured to be lightweight as an integrated IoT structure, and since a low-power sensor-based data collection and local control method is applied, it can operate stably under various usage conditions such as mobile devices, portable cradles, and home environments, thereby having the advantage of being widely applicable without location restrictions.

[0026] Furthermore, according to the present invention, since it is configured to automatically optimize customized emotional stability output values ​​(e.g., type of sound source, vibration period, etc.) in response to specific emotional responses and weight change patterns, it is possible to implement an adaptive emotional stability function that matches the individual characteristics of a newborn, and has the advantage of providing more precise health management services tailored to infants and toddlers.

[0027] It should be understood that the effects of this specification are not limited to the matters described above and can be extended to various contents that can be derived from the detailed description of the embodiments of the invention below. Brief explanation of the drawing

[0028] FIG. 1 is a diagram showing the overall configuration of an IoT-based smart cradle system (100) including an emotional stability function according to one embodiment of the present invention. FIG. 2 is a schematic diagram showing the overall shape of an IoT-based smart cradle system (100) including an emotional stability function according to one embodiment of the present invention. FIG. 3 is a diagram illustrating the flow of refining and analyzing changes in body weight before and after feeding and estimating the amount of feeding through a bio-information measuring unit (110) according to one embodiment of the present invention. FIG. 4 is a diagram illustrating the process of calculating a feeding sensitivity score based on biological response information according to one embodiment of the present invention and transmitting it to an intelligent control unit (130). FIG. 5 is a diagram showing a control flow in which an intelligent control unit (130) according to an embodiment of the present invention generates and adjusts the stimulation of an emotional stability induction unit (120) according to a determined emotional state. FIG. 6 is a diagram showing the process in which an intelligent control unit (130) according to one embodiment of the present invention analyzes a feeding rhythm based on feeding history data and then predicts the feeding timing. FIG. 7 is a diagram showing an external linkage flow including data encryption and authentication token generation process through a wireless communication unit (140) according to an embodiment of the present invention. FIG. 8 is a flowchart sequentially showing the entire procedure of a method for inducing emotional stability in newborns using an IoT-based smart cradle system according to one embodiment of the present invention. Specific details for implementing the invention

[0029] In describing the embodiments of this specification, if it is determined that a detailed description of known configurations or functions could obscure the essence of the embodiments of this specification, such detailed description is omitted. Additionally, parts of the drawings unrelated to the description of the embodiments of this specification have been omitted, and similar parts are denoted by similar reference numerals.

[0030] FIG. 1 is a diagram showing the overall configuration of an IoT-based smart cradle system (100) including an emotional stability function according to one embodiment of the present invention, and FIG. 2 is a diagram showing the overall shape of an IoT-based smart cradle system (100) including an emotional stability function according to one embodiment of the present invention.

[0031] Referring to FIGS. 1 and 2, an IoT-based smart cradle system (100) including an emotional stability function according to one embodiment of the present invention may be configured to include a bio-information measurement unit (110), an emotional stability induction unit (120), an intelligent control unit (130), a wireless communication unit (140), and a user interface unit (150).

[0032] The bio-information measurement unit (110) can perform the role of an input layer for detecting the physiological state of a newborn in real time through a plurality of sensor modules embedded inside a smart cradle according to one embodiment of the present invention and transmitting it to an intelligent control unit (130). The measurement targets may include the newborn's weight, body temperature, heart rate, respiratory rate, etc., and each item may be collected individually through different sensor modules and composed of a single integrated bio-information packet.

[0033] More specifically, in one embodiment of the present invention, the bio-information measuring unit (110) may be composed of a 6-axis load cell installed on the lower frame of the cradle, a contact-type temperature sensor fixed to the bed surface or side wall, and a PPG sensor attached to the side of the cradle in a non-contact manner. These sensors may operate by converting analog physical signals into digital data, pre-processing them through an internal MCU, and then transmitting them to an intelligent control unit (130).

[0034] According to one embodiment, the 6-axis load cell has a multi-axis structure capable of simultaneously detecting vertical loads as well as horizontal and rotational forces, and can comprehensively detect information such as changes in the newborn's weight, changes in posture, and torso shaking. This load cell is symmetrically positioned at the bottom of the cradle and can be mounted on an independent support frame to minimize interference caused by external vibrations or weight deviations.

[0035] The detected load signal is converted into a digital signal by an internal ADC and then transmitted to the MCU, which can perform moving average, upper / lower limit filtering, outlier correction, etc. through sampling at a period of approximately 100 Hz. For example, for detecting feeding events, only changes maintained for 5 seconds or longer can be determined as valid changes based on a resolution of ±10g or less, and the determined weight difference can be used as a basis value for quantitative analysis of feeding amount.

[0036] Additionally, in one embodiment, the skin surface temperature of the newborn can be collected through a contact-type temperature sensor. The contact-type temperature sensor is positioned on the backrest or the lower abdomen and is configured to be attached to the skin contact surface to directly measure the temperature. The sensor can be implemented as an NTC thermistor or RTD type and may include a silicone cap or protective layer with a waterproof coating to prevent interference from external contamination, moisture, body fluids, etc.

[0037] In addition, in one embodiment, temperature data is measured at 1-second intervals and can be refined into a value with a resolution of 0.1°C. At this time, if a rapid change of ±0.5°C or more is detected within a certain period of time, a temperature abnormality flag is set, and the value can be transmitted as a standard for estimating emotional instability and controlling emotional stability.

[0038] Additionally, heart rate and respiratory rate may be measured via a PPG (photoplethysmography) sensor that is optionally included. According to one embodiment, the sensor may be installed as an infrared transmitter / receiver pair on the headrest or side wall of the cradle and may detect periodic changes in blood flow and respiration based on IR signals reflected from the newborn's skin. Additionally, the transmitter may be composed of an infrared LED in the 880–940 nm band, and the receiver may be composed of a photodiode or PIN diode type receiver.

[0039] At this time, the received waveform is converted into heart rate cycles and respiratory rates through FFT-based frequency analysis, and Automatic Gain Control (AGC) may be applied if the signal's SNR drops below a threshold. Additionally, changes in nighttime illumination, newborn movements, and external light interference are removed using high-pass filters and time-series signal clustering algorithms to generate stabilized waveform analysis results.

[0040] Data collected from these sensor modules can be integrated and processed through preprocessing routines within the MCU. For example, changes in body weight are smoothed based on linear regression, temperature data is used to detect periods of rapid change based on differences between samples, and heart rate and respiration data can be analyzed for abnormal patterns through periodicity checks.

[0041] According to one embodiment, sensor data may be stored according to the logic of the intelligent control unit (130) or transmitted to a cloud server or guardian terminal via the wireless communication unit (140). At this time, symmetric key encryption based on AES-128 or AES-256 is applied during the transmission process, and authentication and integrity verification may be performed based on a TLS 1.3 security session.

[0042] If transmission failures are repeated or a sensor error is detected, the error log module of the MCU is activated, an error occurrence event is stored, and subsequently, the intelligent control unit (130) can analyze the error state and dynamically adjust the judgment criteria or control the output of an exception notification to the guardian.

[0043] The emotional stability induction unit (120) can be configured to generate complex stimuli based on the newborn's biological response and emotional state, and to convert the generated stimuli into a form of physical output stimuli that can be controlled in real time, thereby inducing the newborn's autonomic nervous system response to a state that can be stabilized. At this time, the types of output stimuli are classified into three series: auditory, tactile, and visual, and each stimulus can be generated independently or output in a combined form, and the intensity, frequency, and duration of the stimuli can be dynamically adjusted according to the emotional stability judgment result input from the intelligent control unit (130).

[0044] According to one embodiment, auditory stimulation may be provided through a full-range speaker unit, and various emotional stability sound sources, such as lullabies, white noise, maternal voices, and similar heartbeat rhythms, may be pre-registered and played. The sound source is output through a DAC-based playback circuit, the volume can be adjusted in increments of at least 1 dB, and frequency filtering can be implemented through a digital 3-band equalizer or an FIR-based waveform normalization circuit. The sound source being played may consist mainly of monotonous low-frequency white noise when the emotional stability state is severely deteriorated, and may switch to an emotionally stimulating high-frequency sound source during a stabilization phase.

[0045] In addition, tactile stimulation can be generated using a low-frequency vibration module of the ERM or LRA type and is installed on the lower part of the backrest or side support of the cradle to be transmitted to the newborn's body. The output frequency of the vibration can generally be adjusted in the range of 60 to 150 Hz, and the output intensity is adjusted based on a PWM control input or an RMS reference voltage.

[0046] In this case, the vibration intensity changes non-linearly according to the emotional response pattern, and the amplitude increase is applied in steps to minimize the stimulus-rebound response. At this time, a phase interference correction algorithm may be applied to ensure that the peak waveform of the vibration is not synchronized with the newborn's heart rate variability.

[0047] Visual stimulation can be configured using RGB LED modules, and the flashing cycle, color temperature, and brightness modulation of the illumination stimulation can be driven by a Look-Up Table (LUT)-based sequence for each emotional state. The color of the lighting defaults to a warm white hue and gradually transitions to a daylight hue if the emotional state is maintained for a certain period of time. The lighting flashing can be controlled independently of the heart rate cycle, and in one embodiment, in a special mode, it can be implemented to generate a lighting cycle synchronized with the newborn's stable respiratory rate pattern.

[0048] In order to comprehensively determine and control such multiple stimulus outputs, the emotional stability induction unit (120) can quantitatively integrate and analyze the newborn's biological response items through the following mathematical formula 1 and calculate an emotional stability control value that serves as a standard for stimulus outputs.

[0049] [Mathematical Formula 1]

[0050]

[0051] Here, is the biosignal item (i-th variable) at time s, and is a learned coefficient or device initial setting value, and is the second derivative of the biological response, which is a sensitivity value in the form of acceleration, and is the response validity weight over time as an exponential decay term.

[0052] Through this mathematical formula 1, the emotional stability induction unit (120) can perform quantitative optimization of the stimulus combination output. This prediction formula can be implemented as an internal quantification reference model of a non-linear neural network or LUT-based control algorithm of the intelligent control unit (130), and if stimulus inefficiency accumulates for more than a certain period of time, it can be determined to automatically switch the stimulus mode or enter a standby mode.

[0053] In addition, the entire log of stimulus output is recorded in chronological order in the system's internal storage and can be utilized for long-term learning of emotional stability patterns and adjustments based on user feedback. If configured to enable real-time monitoring via a guardian's terminal, output stimulus combinations, intensity, duration, and response results can be visualized and provided.

[0054] The intelligent control unit (130) can perform the role of analyzing the collected multiple biosignals and stimulus output history in real time, calculating an optimized stimulus output scenario, and transmitting it to the emotional stability induction unit (120).

[0055] According to one embodiment, the intelligent control unit (130) may be composed of a high-performance signal processing unit, a memory module, and a microcontroller based on a real-time operating system in terms of hardware, and may have a control algorithm embedded in software that includes a multi-signal filtering algorithm, a biological response prediction model, stimulation determination logic, and a system state judgment routine. Through this configuration, the intelligent control unit (130) can prevent false detection and excessive stimulation output caused by instantaneous noise or disturbances of the biological signal, and can finely adjust the stimulation timing and intensity by cross-referencing multiple signals.

[0056] More specifically, the intelligent control unit (130) can receive multiple bio-response signals in real time, such as the amount of weight change of the newborn, heart rate variability (HRV), skin temperature change, area sum (AUC) of the crying pattern, and gradient vector of posture change, collected from the bio-information measurement unit (110). These signals are time-synchronized and then compared with a reference database stored internally to determine whether the current state corresponds to a stable, unstable, or change boundary range.

[0057] At this time, the intelligent control unit (130) can perform calculations based on multidimensional conditions to determine which type of stimulus to output at what intensity in the current state can induce an emotional stability response by synthesizing the stimulus output history, the cumulative direction of previous state changes, and the stimulus results under the same biological response.

[0058] For example, if the sum of the areas of the crying pattern within the last 10 seconds rises above a certain threshold, the HRV drops sharply, and at the same time the change in posture remains stationary, this can be classified as a pattern indicating a reduced response to stimulation or deep discomfort, and accordingly, the stimulation output can be adjusted to output skin vibration stimulation at a low intensity instead of sound stimulation.

[0059] In addition, the intelligent control unit (130) can strengthen the basis for determining stimulation based on mutual pattern relationships between signals rather than simple numerical changes by also analyzing waveform characteristics such as time delay between multiple biological signals, pattern periodicity, and rapid peak occurrence. In this process, high-order signal processing algorithms such as wavelet transform, Kalman filter, and principal component analysis may be applied, thereby enabling stimulation determination that simultaneously considers the characteristics of individual signals and the integrated state.

[0060] Such stimulation determination is not limited to a single output but is repeatedly updated at regular time intervals, and feedback on the result of each stimulation output is reflected back into a bio-signal and subsequently reflected in a control loop, which can be implemented as a closed-loop control structure. Through this, the intelligent control unit (130) can learn stimulation conditions in real time according to the newborn's response state and perform an adaptive stimulation determination function that gradually improves stimulation efficiency in similar situations.

[0061] According to one embodiment, the intelligent control unit (130) may be configured to integrally reflect the correlation between multiple input signals and the characteristics of time change based on the biological response state collected from the newborn when performing a stimulus output determination for the emotional stability induction unit (120). To this end, the intelligent control unit (130) may apply a high-order non-linear prediction model that interprets the biological response state vector in the continuous time domain and calculates the stimulus output determination value by reflecting the past and present signal change rates, attenuation trends, response sensitivity, etc.

[0062] More specifically, the intelligent control unit (130) may have a prediction operation structure that can expand the time-cumulative response for the newborn’s biological response state vector x(t)=[x1(t), x2(t), … , xn(t)] into a higher-order tensor form and reflect the mutual cross-sensitivity within the time series and the non-linear responsiveness of the stimulus output. In one embodiment, a tensor-kernel-based stimulus output determination formula of the form of Equation 2 below may be applied.

[0063] [Mathematical Formula 2]

[0064]

[0065] Here, is a biological response state vector at time s, and is the second-order spatial derivative value resulting from this, and is a Frobenius game, and is the delay damping coefficient, and is a response sensitivity function defined by crying intensity and the rate of change in body temperature, etc., and is the sigmoid sensitivity coefficient.

[0066] Additionally, the intelligent control unit (130) may include a semi-learning-based parameter optimization structure to store the results of the interaction between the stimulus output and the biological response over a certain period of time and to reproduce the stimulus output under the same response conditions thereafter. For example, if patterns such as HRV, changes in body temperature, and crying cycles before and after sleep are accumulated and collected for the same newborn over a certain period of time, the intelligent control unit (130) may apply a customized algorithm to adjust the sensitivity of the stimulus output and the duration of the stimulus at each point in time based on this data.

[0067] In addition, in one embodiment, the intelligent control unit (130) may also perform an exception control function in which, under certain conditions, the biological response fluctuates in an excessively abnormal pattern or reaches a predefined threshold criterion (e.g., a sudden decrease in HRV, a sudden increase in body temperature, etc.), immediately stop the stimulation output of the emotional stability induction unit (120) and transmit an emergency notification signal to a guardian terminal or a hospital-linked system. At this time, the exception judgment criterion may be set as a multi-conditional judgment structure based on a combination of multiple biological signals, and may include a cool-down structure in which the resumption of stimulation output is restricted for a certain period of time after the notification is transmitted.

[0068] The wireless communication unit (140) can be configured to transmit and receive internal data, such as control signals generated by the intelligent control unit (130), bio-response data collected from the bio-information measurement unit (110), and output history of the emotional stability induction unit (120), to an external device in real time.

[0069] More specifically, the wireless communication unit (140) may include a wireless communication module based on BLE, Wi-Fi, or LoRa, and may be implemented with a multi-interface structure that can selectively activate one or more of the protocols depending on the system operating environment and the type of connected device.

[0070] According to one embodiment, the wireless communication unit (140) may branch the communication path according to the recipient, for example, when communicating 1:1 with a guardian terminal, it may use BLE first, and when connection with a hospital server or remote monitoring system is required, it may open a Wi-Fi-based TCP / IP session to transmit data. At this time, the wireless communication unit (140) may include a dynamic control structure that checks the operating status of each communication module in real time and can automatically re-select the communication path based on indicators such as packet loss rate, RSSI, and connection maintenance time.

[0071] Additionally, the wireless communication unit (140) can transmit various biosignals, such as changes in newborn weight, heart rate fluctuations, and crying duration, collected from the biosignal information measurement unit (110) as a real-time stream to a guardian mobile application or a hospital cloud system, and can be configured to transmit by applying a lightweight encryption algorithm (e.g., AES-GCM, an authentication structure based on Elliptic Curve Cryptography, etc.) for the protection of personal information during this process.

[0072] In addition, the wireless communication unit (140) may include metadata of the data block transmitted at each time of data transmission and reception, so that transmission can be performed in a structure that includes, for example, the time of collection of each biosignal, sensor ID, collection location, operation cycle, sensor calibration status, etc. Through this, the receiving side can perform reliable analysis based on contextual information beyond simple numerical values, and the hospital system can automatically classify and store the received biosignal data by newborn.

[0073] In addition, the wireless communication unit (140) may include a redundant transmission supplement structure that stores data for a certain period by allocating a temporary buffer space in internal memory in preparation for cases where communication is interrupted for a certain period or a connection with an external device fails, and then sequentially retransmits the missing data when the connection is restored. At this time, to ensure the sequentiality of the stored data, the wireless communication unit (140) may include a unique sequence number in each data frame, and may be linked so that the receiving side can determine whether a packet is missing based on this sequence.

[0074] In addition, in one embodiment, the wireless communication unit (140) may be configured to include a control algorithm capable of dynamically adjusting the transmission cycle, power consumption, packet size, etc., so as to automatically reduce the data transmission frequency during static periods when the newborn is sleeping or there is minimal change in biological response, and to set the transmission cycle to a short duration when crying or a rapid rise in body temperature is detected, thereby ensuring a rapid response flow.

[0075] The user interface section (150) can serve as an interface that allows a guardian or medical staff to intuitively check the operating status of the smart cradle system, the newborn's bio-response information, and the history of emotional stability induction, and to input manual control commands when necessary. This user interface section (150) can be implemented in the form of a local display device based on a touchscreen panel or an external mobile application, and can be operated in a dual interface structure that uses an internal display and an external terminal in parallel depending on the situation.

[0076] More specifically, the user interface section (150) may include a real-time system dashboard that summarizes the current status of the cradle system at the top of the screen. For example, the current operation status of the lullaby module, vibration induction intensity, light intensity change induction status, biosignal collection cycle, wireless communication connection status, etc., may be visually displayed along with real-time icons, and may be configured so that touching each icon allows entry into a detailed setting information or log checking screen.

[0077] According to one embodiment, the user interface unit (150) can graphically output the flow of the newborn’s biological response in a visualized form, and, for example, can provide the heart rate (HR), changes in body temperature (ΔT), the timing and duration of crying detection events, and trends in posture changes in the form of a time-axis-based graph. At this time, the data is processed based on a stream received in real time from the biological information measurement unit (110), and the history of stimulation induction linked to the judgment result of the intelligent control unit (130) can also be displayed on the timeline. Through this, the user can intuitively track which stimulation induction caused a specific change in biological response.

[0078] Additionally, the user interface section (150) may include a structure that allows a caregiver to manually input specific care events, such as breastfeeding, diaper changing, or sleep induction, and the input events may be stored as logs linked to changes in biological responses. For example, when a user presses the "Start breastfeeding" button, data on weight changes and posture changes for a certain period of time after that point are automatically tagged and stored, and can be used to distinguish between the pre- and post-breastfeeding states during the subsequent analysis process.

[0079] Additionally, the user interface unit (150) may be configured to include a notification system to output a warning message to a guardian when a specific abnormal condition is detected. For example, a warning may be issued in the form of a real-time pop-up notification or vibration alarm for situations such as when crying persists for a certain period of time, when body temperature rises above a threshold value, when biosignal collection is interrupted for a certain period of time, or when wireless communication is interrupted. This warning notification may be linked to a hospital server to automatically transmit logs and request a medical staff call.

[0080] In one embodiment, the user interface unit (150) may support a multilingual output structure, and an initial setup function may also be provided to allow the user to customize the language, time zone, newborn information (date of birth, weight, allergy information, etc.), notification sensitivity, etc. during initial setup. In addition, an integration mode may be supported through UI expansion in the form of a smartphone-based mobile application, which allows the guardian to remotely check the status of the cradle system or perform simple controls from outside.

[0081] Additionally, the user interface unit (150) may be linked with the control unit within the system to periodically store the user's input records and output visualization history, and may be configured to be automatically deleted or transmitted to a cloud server after a certain period, and may be equipped with a PIN authentication, biometric recognition, or guardian terminal-based authentication system in consideration of user privacy and data security.

[0082] Next, we will examine the flow of refining and analyzing weight changes before and after feeding through the bio-information measuring unit (110), and estimating the amount of feeding based on this.

[0083] FIG. 3 is a diagram illustrating the flow of refining and analyzing changes in body weight before and after feeding and estimating the amount of feeding through a bio-information measuring unit (110) according to one embodiment of the present invention.

[0084] Referring to FIG. 3, first, in step S301, the bio-information measuring unit (110) simultaneously detects the weight of the newborn in multiple axial directions, and load signals in each direction obtained through a 6-axis load cell can be collected in real time. At this time, since the measurement signal may contain noise caused by external vibrations, changes in posture, etc., the intelligent control unit (130) can be configured to determine abnormal signals based on multiple parameters such as changes in inclination between samples, short-term periodicity, spike occurrence period, and long-term flattening ratio, and to extract only refined weight signals below a reference deviation.

[0085] Next, in step S302, the time points before and after feeding can be distinguished from the time series flow of the refined weight signal. More specifically, the intelligent control unit (130) analyzes the weighted moving average of the weight data, the point of rapid increase or decrease, the inflection point based on the second derivative, etc., and the weight measurement interval can be divided based on a flat section that has lasted for a certain period of time or longer. At this time, to improve the reliability of the division, information such as past feeding patterns, cradle stay time, and sensor malfunction logs can be considered together.

[0086] Next, in step S303, a weight change amount ΔW can be calculated based on the difference between the average weight values ​​corresponding to the time before and after the divided feeding. At this time, to verify the validity of the weight change amount, the intelligent control unit (130) can evaluate the reliability of the weight change amount based on reference data such as the previous feeding record, feeding interval, average feeding amount of the previous day, and the range of variation per feeding session. If it exceeds the reference error range, the weight change amount can be set so that it is not considered a valid feeding amount. Through this, abnormal weight changes can be prevented.

[0087] Next, in step S304, if the calculated weight change amount deviates from the preset normal feeding amount range, the intelligent control unit (130) may designate a status flag based on an internal judgment condition and transmit data to the user interface unit (150). The user interface unit (150) may be configured to visually display the feeding amount status based on the input weight change amount value and to output a recent feeding history graph, a weight change trend by time period, etc., so that the user can compare it with the previous feeding flow. At this time, the notification information may be limited to visual output, and separate warning voice or external notification may be set not to operate.

[0088] Finally, in step S305, data such as weight change amount, weight value by interval, signal refinement indicator, splitting time, and status flag is stored in a structured format and can be transmitted to a local gateway or upper server via a wireless communication unit (140). The transmitted data can be configured to be utilized for chronological evaluation, iteration interval analysis, and deviation pattern detection in subsequent analysis steps, and can be linked to a merge operation and a statistics-based prediction algorithm based on accumulated data at regular time intervals.

[0089] Next, we will examine the process of calculating a feeding sensitivity score based on biological response information according to one embodiment of the present invention and transmitting it to an intelligent control unit (130).

[0090] FIG. 4 is a diagram illustrating the process of calculating a feeding sensitivity score based on biological response information according to one embodiment of the present invention and transmitting it to an intelligent control unit (130).

[0091] Referring to FIG. 4, first, in step S401, the bio-information measuring unit (110) can collect bio-response data of a newborn detected in real time. At this time, the items to be collected may consist of multiple items such as body temperature, heart rate, respiration rate, fine tremors of facial muscles, sucking rhythm immediately after feeding, and body movement patterns, and each bio-signal may be arranged in units of a certain time interval according to a designated sampling period. The bio-information measuring unit (110) can transmit the filtered value according to the preprocessing criteria to the intelligent control unit (130) after reflecting the sensor correction value to the raw signal of each item.

[0092] Next, in step S402, the intelligent control unit (130) can perform normalization and feature extraction processes based on the received bio-response data. According to one embodiment, the mean squared deviation of heart rate variability (HRV), the interval slope of the rate of change of body temperature (ΔT), the power ratio by frequency band of the facial electromyography signal, the periodicity and stability of the inhalation rhythm, and the peak distribution index of the body movement intensity can be calculated as key feature values. This feature extraction can be performed by considering the correlation between signals and the time delay effect together, and if an abnormal feature value is detected, the item can be set to be excluded from the calculation.

[0093] Next, in step S403, normalized feature values ​​are used as input values ​​for a neural network-based estimation model to calculate a feeding sensitivity score. At this time, the feeding sensitivity score may consist of continuous real values ​​between 0 and 1 and may be defined as a result that comprehensively reflects indicators such as emotional stability, physiological tolerance, and behavioral responsiveness. For example, if the similarity is high when comparing the biological response pattern immediately prior to past feeding with the current pattern, the corresponding feeding sensitivity score may be evaluated as a high value. This model may be implemented with a structure including a multi-layer perceptron, normalized regression coefficients, and a sigmoid activation function.

[0094] Next, in step S404, after the calculated feeding sensitivity score is compared with an internal threshold value, if the score exceeds a specific threshold, it may be flagged as 'high feeding probability'. Conversely, if it is below the threshold value, it is considered 'low feeding sensitivity' and may be excluded from the control input. In this case, the threshold value may be configured to be dynamically adjusted based on the newborn's individual history, average sleep interval, recent stress index, etc.

[0095] Finally, in step S405, structured data including a feeding sensitivity score, a feature value vector, and an evaluation result flag can be generated and transmitted to the intelligent control unit (130). This data can subsequently be reflected in the output judgment of the emotional stability induction unit (120), automatic control of the feeding time, and visualization judgment of the user interface unit (150), and in the long term, the time series flow of the sensitivity score can be accumulated and stored to be used for trend analysis.

[0096] Next, we will examine the control flow in which the intelligent control unit (130) according to one embodiment of the present invention generates and adjusts the stimulation of the emotional stability induction unit (120) according to the determined emotional state.

[0097] FIG. 5 is a diagram showing a control flow in which an intelligent control unit (130) according to an embodiment of the present invention generates and adjusts the stimulation of an emotional stability induction unit (120) according to a determined emotional state.

[0098] Referring to FIG. 5, first, in step S501, the intelligent control unit (130) can analyze the emotional state of the newborn based on real-time bio-response data received from the bio-information measurement unit (110). At this time, input values ​​may include heart rate variability (HRV), facial electromyography (EMG), abdominal respiration rate (RR), time-frequency characteristics of the body motion spectrum, and feeding sensitivity scores, and these items may be normalized into a single emotional state vector and configured for each time point. The emotional state analysis can be performed through a GRU-based sequence model with a time-series pattern recognition algorithm applied, along with rule-based conditional statements.

[0099] Next, in step S502, based on the classification result of the emotional state vector, the intelligent control unit (130) can map the newborn's current emotional state to multiple qualitative states such as 'tension', 'discomfort', 'anxiety', and 'relaxation'. For example, if the HRV value is lower than the standard and the high-frequency component of the facial EMG increases, it can be determined to be in an 'anxiety' state, and at this time, the state transition flow from a past point in time is also considered, and a state code such as 'acute tension transition' or 'chronic discomfort persistence' can be assigned together.

[0100] Next, in step S503, a stimulus type corresponding to the emotional state may be selected. The intelligent control unit (130) may determine a stimulus combination corresponding to each emotional state by referring to a predefined stimulus mapping table, and this stimulus combination may include 'sleep-inducing vibration stimulation', 'micro-sound rhythm stimulation', 'thermal skin stimulation', 'light color temperature change stimulation', etc. For example, if classified as an 'anxiety' state, a stimulus combination combining low-frequency vibration and a red-colored light intensity change may be set.

[0101] Next, in step S504, the individual stimulus parameters of the selected stimulus combination can be quantified. The stimulus parameters for each stimulus channel may consist of 'intensity', 'period', 'duration', 'phase offset', etc., and these can be determined by considering the stimulus history and response effect indicators from the previous time point, along with the urgency index of the current emotional state. In particular, when the same stimulus is applied repeatedly, a random number-based phase deviation value may be added to prevent stimulus adaptation effects.

[0102] Finally, in step S505, the calculated stimulation parameters are transmitted to the emotional stability induction unit (120) and can be distributed to individual stimulation modules of the induction unit for execution. At this time, vibration stimulation can be controlled by a servo-controlled linear motor, and sound wave stimulation can be output to a speaker module through an amplification circuit using a DSP-processed waveform. Additionally, brightness control can be achieved through a PWM-based LED driver circuit, and thermal stimulation can be performed using a PID temperature control method based on thermistor feedback.

[0103] Next, we will examine the process in which an intelligent control unit (130) according to one embodiment of the present invention analyzes a feeding rhythm based on feeding history data and then predicts the feeding timing.

[0104] FIG. 6 is a diagram showing the process in which an intelligent control unit (130) according to one embodiment of the present invention analyzes a feeding rhythm based on feeding history data and then predicts the feeding timing.

[0105] Referring to FIG. 6, first, in step S601, an intelligent control unit (130) can collect accumulated feeding history data through a wireless communication unit (140) and a biometric information measurement unit (110). This feeding history data may include items such as the feeding start time, feeding end time, feeding amount, weight change before and after feeding, and biometric response vectors (heart rate, skin temperature, crying characteristics, etc.) at the time of feeding, and each item may be recorded on a time-series basis along with a UTC timestamp. The collected data may be preprocessed into an accumulated feeding history dataset through normalization and outlier correction processes.

[0106] Next, in step S602, the intelligent control unit (130) can analyze the feeding rhythm pattern based on the feeding interval, changes in feeding amount, night-day periodicity, and fatigue accumulation indicators. At this time, periodic characteristics can be extracted through FFT-based frequency analysis and Autocorrelation-based autocorrelation series analysis, and the current feeding rhythm state of the newborn can be classified as 'stable', 'irregular', 'high frequency', or 'low frequency' by considering the average daily variation of the feeding amount, standard deviation, and outlier ratio together.

[0107] Next, in step S603, based on the results of the feeding rhythm analysis, the intelligent control unit (130) can predict future feeding timing. The prediction model may be composed of a time series prediction neural network based on LSTM (Long Short-Term Memory).

[0108] Next, in step S604, the intelligent control unit (130) can preemptively adjust the stimulation schedule of the emotional stability induction unit (120) according to the predicted feeding timing. Specifically, based on a specific time prior to the predicted time of feeding start (e.g., 15 minutes prior), sedative induction stimuli (low-frequency vibration, gentle sound stimulation, etc.) can be prepared sequentially, and conversely, during the period immediately after feeding, stimulation can be minimized to maintain a resting rhythm. Such adjustment can be applied only under conditions where the error between the real-time state and the predicted value does not exceed a certain threshold.

[0109] Finally, in step S605, the feeding prediction results and stimulation adjustment history are transmitted to the user interface unit (150) via the wireless communication unit (140) so that they can be visually checked on the guardian terminal or management system. At this time, information such as the prediction time range, cumulative accuracy, and recommended response actions is displayed together, and the user can refer to this to select whether to prepare for feeding or adjust the rest environment.

[0110] Next, we will examine the external linkage flow including the process of data encryption and authentication token generation through the wireless communication unit (140) according to one embodiment of the present invention.

[0111] FIG. 7 is a diagram showing an external linkage flow including data encryption and authentication token generation process through a wireless communication unit (140) according to an embodiment of the present invention.

[0112] Referring to FIG. 7, first, in step S701, the intelligent control unit (130) can transmit core data generated within the device, such as feeding history, bio-response information, and emotional stimulation schedule, to the wireless communication unit (140). At this time, the data to be transmitted may be composed of a structured packet with metadata attached for each item, and, for example, a packet format in the form of data type, timestamp, target module ID, importance grade, and payload may be applied.

[0113] Next, in step S702, the wireless communication unit (140) may perform encryption preparation internally. More specifically, a unique session key is generated for the data payload, which can be derived based on random number exchange using the ECDH method. The session key can be used for symmetric encryption in AES-256 CBC mode, and the encryption target may be limited to identification information and biometric sensitive information items rather than the entire payload. Additionally, the CBC initialization vector (IV) is updated for each request and can be calculated based on terminal UUID and timestamp hash.

[0114] Next, in step S703, the wireless communication unit (140) can generate an authentication token to enable the transmission of the encrypted data. The authentication token may be formed, for example, in a JSON Web Token (JWT) structure, and the token may include information such as a device unique identifier, a hash of user authentication information, a time of issuance, a validity period, and an authorization scope. At this time, the signature may be generated using an RSA-SHA256-based asymmetric key method, and the secret key may be stored in an HSM storage area within the intelligent control unit (130).

[0115] Next, in step S704, a JWT token can be transmitted to an external network along with encrypted data. The destination of transmission may be a guardian's smartphone app, a cloud-based breastfeeding management server, a hospital remote monitoring server, etc., and the transmission path may be configured through an HTTPS channel encrypted with a security protocol of TLS 1.3 or higher. Additionally, mutual authentication based on a CA certificate may be performed during the handshake process with the target server prior to message transmission.

[0116] Finally, in step S705, the receiving server verifies the signature of the received JWT token and can restore the encrypted data using a decryption key only if the token is valid. The restored data is then stored in the server's database, and subsequent processing, such as real-time analysis, notification transmission, and response command generation, can be performed based on user requests or trigger conditions of the analysis model.

[0117] Next, we will sequentially examine the entire procedure of a method for inducing emotional stability in newborns using an IoT-based smart cradle system according to one embodiment of the present invention.

[0118] FIG. 8 is a flowchart sequentially showing the entire procedure of a method for inducing emotional stability in newborns using an IoT-based smart cradle system according to one embodiment of the present invention.

[0119] Referring to FIG. 8, first, in step S801, the biometric information of a newborn can be measured in real time through the biometric information measurement unit (110). At this time, the measurement items may include changes in body weight, heart rate variability (HRV), skin temperature, crying intensity, and changes in posture, and each item may be preprocessed based on signals obtained from an internal 6-axis load cell, infrared sensor, microphone array, inertial measurement unit (IMU), etc., and then normalized into a vector format with a timestamp and output.

[0120] Next, in step S802, the emotional stability induction unit (120) determines the emotional state based on the bio-information measured by the bio-information measurement unit (110), and stimulation output can be performed accordingly. More specifically, the input bio-information is transmitted to a built-in neural network-based response classifier and can be classified into states such as stable, anxious, or irritable, and depending on the classification result, audiovisual stimulation, vibration stimulation, temperature stimulation, etc., can be output in combination. At this time, the type, intensity, and duration of the stimulation can be adjusted according to a set stimulation table or control command.

[0121] Next, in step S803, the intelligent control unit (130) receives real-time data provided by the bio-information measurement unit (110) and can analyze or learn the emotional response pattern of the newborn based on this. More specifically, a high-order non-linear analysis model based on time-accumulated bio-response data may be applied, and the model can quantify changes in the emotional state by comparing the similarity between the current response state and the past history. The result of this analysis is then transmitted to the emotional stability induction unit (120) and can be used as a command to control the stimulus output.

[0122] Next, in step S804, the biometric information and emotional state analysis results output from the intelligent control unit (130) can be transmitted and received in real time to an external terminal or remote server via the wireless communication unit (140). For example, it can be linked with a guardian's smartphone app, a hospital cloud system, a monitoring console for medical staff, etc., and the transmitted data may include real-time biometric values, stimulus history, emotional sensitivity scores, etc. The communication can be performed securely through TLS-based encryption and a JWT-based authentication structure.

[0123] Finally, in step S805, the history of changes in bio-information and the history of stimulation performance transmitted from the intelligent control unit (130) through the user interface unit (150) may be output in a visual or voice-based manner, and an interface may be provided that allows a guardian or medical staff to directly input necessary commands. More specifically, a touchscreen-based UI, a voice command receiving microphone, a notification indicator LED, a vibration feedback module, etc. may be included, and the current feeding status, emotional stability score, stimulation history graph, etc. may be displayed on the screen in real time. In addition, a function to modify the stimulation sensitivity threshold, notification conditions, user registration information, etc. may be provided in the guardian settings menu.

[0124] As described above, the present invention can be configured to measure bio-responses such as changes in weight, heart rate fluctuations, and posture of a newborn in real time through a bio-information measuring unit, and to output various types of stimuli according to the emotional state determined through an emotional stability induction unit.

[0125] In addition, the present invention can be configured to learn emotional response patterns based on collected bio-information and to implement a response flow suitable for the condition of an individual newborn through the analysis of feeding rhythms and the generation of stimulus control commands.

[0126] Through this, the present invention has the advantage of quantifying emotional state and feeding sensitivity based on biological responses before and after feeding, predicting newborn-specific feeding timing based on accumulated response data, and achieving emotional stability and feeding cycle optimization.

[0127] Meanwhile, since the description of the technology disclosed in this specification is merely an example for structural or functional explanation, the scope of the disclosed technology should not be interpreted as being limited by the examples described in the text. That is, since the examples are subject to various modifications and may take various forms, the scope of the disclosed technology should be understood to include equivalents capable of realizing the technical concept. Furthermore, since the purposes or effects presented in the disclosed technology do not imply that a specific example must include all of them or only such effects, the scope of the disclosed technology should not be understood as being limited by them.

[0128] Furthermore, when it is stated that one component is “connected” to another component, it should be understood that while it may be directly connected to that other component, there may also be other components in between. Conversely, when it is stated that one component is “directly connected” to another component, it should be understood that there are no other components in between. Meanwhile, other expressions describing the relationships between components, such as “between” and “between” or “adjacent to” and “directly adjacent to,” should be interpreted in the same way.

[0129] A singular expression should be understood to include a plural expression unless the context clearly indicates otherwise, and terms such as “include” or “have” are intended to specify the existence of the set-up features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood not to preclude the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. Explanation of the symbols

[0130] 100: IoT-based smart cradle system with emotional stability features 110: Biometric Information Measurement Unit 120: Emotional Stability Induction Unit 130: Intelligent control unit 140: Wireless communication unit 150: User Interface Section

Claims

Claim 1 A biometric information measuring unit (110) that measures the biometric information of a newborn in real time; an emotional stability induction unit (120) that analyzes the biometric information collected from the biometric information measuring unit (110) and outputs an emotional stability stimulus according to the determined emotional state; an intelligent control unit (130) that receives the biometric information collected from the biometric information measuring unit (110), learns the emotional response pattern for each newborn, controls the operation of the emotional stability induction unit (120) according to the determined emotional state, and determines the state change based on the biometric information in real time; and a wireless communication unit (140) that transmits and receives the biometric information and analysis results output from the intelligent control unit (130) in real time to an external terminal or server. An IoT-based smart cradle system including an emotional stability function, comprising a user interface unit (150) that outputs the history of changes in biological information and the status of performing emotional stability stimulation in visual or auditory form according to the control of the intelligent control unit (130), and receives commands from parents or medical staff; wherein the intelligent control unit (130) arranges feeding history data composed of an estimated feeding amount and feeding interval information in a time series, extracts a feeding rhythm pattern by frequency converting the feeding history data, predicts the next feeding timing based on the extracted feeding rhythm, classifies feeding response data composed of body temperature, heart rate, and sleep transition time measured immediately after feeding into response types based on a multivariate state-space model, and adjusts the stimulation parameters of the emotional stability induction unit (120) according to the classified response types. Claim 2 An IoT-based smart cradle system including an emotional stability function, wherein, in claim 1, the intelligent control unit (130) divides the bio-information vector measured immediately after feeding into a reference threshold area, determines an abnormal state when the bio-information vector deviates from the reference area, prioritizes the operation of the emotional stability induction unit (120), and transmits warning data to a user terminal through the wireless communication unit (140). Claim 3 An IoT-based smart cradle system including an emotional stability function, wherein the wireless communication unit (140) encrypts newborn status data and feeding history data with an AES-based symmetric key, generates a hash authentication token including current location data of the cradle, and transmits the encrypted data and the authentication token together when transmitting to a server. Claim 4 A method for stabilizing the emotions of a newborn using an IoT-based smart cradle system including an emotional stabilization function according to any one of claims 1 to 3, wherein the method for stabilizing the emotions of a newborn includes: a step of measuring the newborn's biometric information in real time through a biometric information measuring unit; a step of analyzing the biometric information provided from the biometric information measuring unit through an emotional stabilization induction unit and outputting an emotional stabilization stimulus according to a determined emotional state; a step of receiving the biometric information provided from the biometric information measuring unit through an intelligent control unit, learning the emotional response pattern for each newborn, controlling the operation of the emotional stabilization induction unit according to a determined emotional state, and determining the state change based on the biometric information in real time; a step of transmitting and receiving the biometric information and analysis results output from the intelligent control unit in real time with an external terminal or server through a wireless communication unit; and a step of outputting the history of changes in the biometric information and the status of the performance of the emotional stabilization stimulus in visual or auditory form according to the control of the intelligent control unit through a user interface unit, and receiving commands from parents or medical staff. Claim 5 A method for emotional stability of a newborn using an IoT-based smart cradle system including an emotional stability function, wherein, in claim 4, the step of measuring the above-mentioned bio-information in real time includes: a step of measuring weight data before and after feeding at regular time intervals, respectively; and a step of refining the measured weight change amount using a polynomial regression-based filter and clustering the cumulative distribution of the refined weight change amount by feeding event units to analyze the time-dependent trend of the feeding amount.

Images