Oxygen cabin intelligent control system based on GaN MQW diode array

Abstract

The invention discloses an oxygen cabin intelligent control system based on a GaN MQW diode array, which relates to the technical field of photoelectron and intelligent control, and comprises an external control module and an internal control module, a plurality of GaN MQW diodes are deployed on an oxygen cabin to form a diode array, the diode array comprises a lighting mode, a detection mode and a time division multiplexing mode, and when an oxygen cabin door is in a closed state, an external control module is used for configuring the diode array into the detection mode, whether the cabin door is opened or not is judged through a preset password sequence, and an internal control module is triggered; the internal control module is used for configuring the diode array to be in a time division multiplexing mode, and when a user displays different gesture actions, different internal control functions of the oxygen cabin are triggered. According to the invention, traditional oxygen cabin contact type operation is abandoned, multiple functions of cabin door control, illumination, internal control and the like of the oxygen cabin are realized through the GaN MQW diode array, the integration level and the safety of the oxygen cabin system can be effectively improved, and the equipment deployment and maintenance cost is greatly reduced.

Description

Technical Field

[0001] This invention relates to the field of optoelectronics and intelligent control technology, specifically an intelligent control system for oxygen chambers based on a GaN MQW diode array. Background Technology

[0002] The existing control systems of oxygen chambers mainly rely on mechanical buttons or touch screens, which have the following shortcomings: 1. Contact operation: inconvenient for patients or the elderly to use and not conducive to the control of a sterile environment; 2. High hardware complexity: external access requires additional RFID, fingerprint or password modules, and internal control requires independent switches; 3. Insufficient security: electromagnetic signal transmission is prone to failure in a closed oxygen chamber.

[0003] GaN MQW (gallium nitride quantum well) diodes have the characteristic of coexisting light emission and detection, serving as both a light source and a detector. This makes them suitable for the special environment of oxygen chambers. By using GaN MQW diodes in oxygen chamber control and through reasonable software scheduling, dual functions of lighting and sensing can be achieved in different time slices. This enables multiple functions such as access control, lighting, and internal control, improving the integration and safety of the oxygen chamber system and significantly reducing equipment deployment and maintenance costs. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent control system for oxygen chambers based on GaN MQW diode arrays to solve the problems raised in the prior art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: An intelligent control system for oxygen chambers based on GaN MQW diode arrays, comprising an external control module and an internal control module; External control module: Several GaN MQW diodes are deployed on the oxygen chamber to form a diode array. The diode array includes an illumination mode, a detection mode, and a time-division multiplexing mode that alternates between illumination mode and detection mode. When the oxygen chamber door is closed, the external control module is used to configure the diode array to detection mode. When the diode array receives an coded light signal, it determines whether to open the door through a preset password sequence. When it is determined that the door is to be opened, the internal control module is triggered. Traditional oxygen chambers typically use a contact-based opening mechanism, requiring users to open the door via a handle. This is not only inconvenient but also detrimental to aseptic environment control. Therefore, this solution leverages the characteristics of GaN MQW diodes to enable intelligent door opening, which helps reduce bacterial transmission and creates a clean and comfortable environment for the oxygen chamber.

[0006] Internal control module: Used to configure the diode array in time-division multiplexing mode. In illumination mode, the diode array actively emits modulated light and in detection mode, it receives reflected light signals. When the user displays different gestures, the reflected light signals received by the diode array will produce different differences, thereby triggering different internal control functions of the oxygen chamber.

[0007] Since traditional oxygen chambers require additional independent switches for internal control, and various internal control functions require multiple buttons, this solution leverages the time-division multiplexing mode of GaN MQW diodes, which have the characteristics of emitting modulated light and receiving reflected light signals, to realize different internal control functions, improve the integration and safety of the oxygen chamber system, and significantly reduce equipment deployment and maintenance costs.

[0008] Preferably, the steps for determining whether to open the hatch using a preset password sequence are as follows: Monitoring equipment is deployed on the oxygen chamber to enable real-time monitoring of the surrounding environment. The diode array is preset to collect the coded light signal at a frequency. Users use a lighting device to emit the coded light signal into the oxygen chamber. When the diode array operates at the collection frequency and receives the coded light signal, it retrieves the monitoring video from a period of time before receiving the signal, captures and analyzes the moving objects in the monitoring video, and determines whether the coded light signal is reliable. When personnel want to open the oxygen chamber, they need to face the oxygen chamber and hold a light device. Therefore, the confidence condition is set here to exclude those who accidentally touch it. The judgment is based on local area because facing the oxygen chamber is required to open it, and the judgment is based on histogram similarity because holding a light device is required to open it. Both are indispensable. If the coded optical signal is deemed reliable, it is converted into a digital sequence through sampling, quantization, and decoding. The converted digital sequence is then compared with a pre-stored set of cipher sequences to determine whether to open the hatch.

[0009] Preferably, the steps for capturing and analyzing moving objects in surveillance video and determining the reliability of the encoded optical signal are as follows: Using object detection algorithms, all moving objects in the surveillance video are captured, and all local regions of each moving object are obtained, including the eyes, ears, nose and mouth of the face; Acquire a standard light source image free from environmental interference, convert the standard light source image into a grayscale image, and establish a baseline histogram based on the grayscale value of each pixel in the grayscale image; acquire an ROI region image of the moving object, convert the ROI region image into a grayscale image, and establish a target histogram based on the grayscale value of each pixel in the grayscale image. If the number of local regions of any moving object exceeds a preset threshold, and the histogram similarity between the corresponding baseline histogram and the target histogram exceeds a preset similarity threshold, the encoded optical signal is deemed reliable.

[0010] Preferably, the steps for comparing the converted digital sequence with a pre-stored set of password sequences to determine whether to open the hatch are as follows: Obtain the set of password sequences for opening the hatch. Each password sequence in the set has a length of N. Set password update rules to obtain the updated password sequence set. If the length of the converted number sequence is N, obtain the updated number sequence according to the password update rules. Match the updated number sequence with each element in the updated password sequence set. If the match is successful, determine to open the hatch; if the match fails, keep the hatch closed.

[0011] Preferably, the steps for triggering different internal control functions of the oxygen chamber are as follows: Different hand gestures correspond to different control functions of the oxygen chamber. Sample data of different hand gestures made by several users are collected, and target data is extracted from the sample data. The sample data includes the type of hand gesture, the time when each diode receives the reflected light signal, and the oxygen chamber control function. A federated network model is established, with each user object as a source domain client. The local models obtained by parallel training of each source domain client and the global model obtained by aggregating each local model are obtained from the federated network model. Based on the target data collected from each source domain client, the local models and global models corresponding to each source domain client are trained to achieve federated network model training. The federated network model is a distributed collaborative training system under the federated learning architecture, consisting of a central node (global model) and multiple client nodes (local models). It has the advantages of balancing data privacy and global model performance. In this scheme, the time when each diode in the target data receives the reflected light signal is used as input, and the oxygen chamber control function is used as output to train the local model and the global model, thereby realizing the training of the federated network model. The trained federated network model is used to obtain the corresponding oxygen chamber control function based on the time when each diode receives the reflected light signal. Based on changes in the oxygen chamber environment, the duty cycle of the alternation between detection mode and illumination mode in time-sharing multiplexing mode is adjusted in real time. Based on the current time when each diode receives the reflected light signal, and based on the trained federated network model, different internal control functions of the oxygen chamber are triggered.

[0012] Preferably, the steps for implementing real-time adjustment of the duty cycle of the switching time between the detection mode and the illumination mode in the time-sharing multiplexing mode are as follows: Light sensors are deployed at the gesture recognition point in the oxygen chamber to monitor the light intensity at the gesture recognition point in real time when users make different gestures. Get the current illumination intensity X at the gesture recognition location; The user's gestures were collected at different duty cycles. The number of sample data and the number of target data were extracted when the illumination intensity was X and the duty cycle was K, so as to obtain the success rate when the illumination intensity was X and the duty cycle was K. Establish a scatter plot of the successful recognition rate as a function of duty cycle corresponding to light intensity X, and adjust the duty cycle in time-sharing multiplexing mode in real time based on the scatter plot.

[0013] Preferably, the steps for extracting target data from sample data are as follows: identify several user objects with different physical characteristics, including height, weight, gender, and age, and have each user object make different hand gestures; use the sample data of the oxygen chamber triggering the correct control function after the user object makes the corresponding hand gesture as the target data of the source domain client.

[0014] Preferably, the steps to achieve real-time adjustment of the duty cycle in the time-division multiplexing mode are as follows: the duty cycle is the proportion of the time the diode array is in the illumination mode to the total cycle time within a complete working cycle of the time-division multiplexing mode, and the value is between 0 and 1; the minimum duty cycle is set to K0, all scatter plots with a duty cycle greater than K0 are obtained, and the duty cycle corresponding to the scatter plot with the highest success rate is taken as the current duty cycle of the diode array in the illumination mode.

[0015] Preferably, the gestures include hovering, moving forward, moving backward, moving left, moving right, moving up, and moving down.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention provides an intelligent control system for an oxygen chamber based on a GaN MQW diode array, including: an external control module and an internal control module; several GaN MQW diodes are deployed on the oxygen chamber to form a diode array, the diode array including a lighting mode, a detection mode and a time-division multiplexing mode; when the oxygen chamber door is closed, the external control module is used to configure the diode array in detection mode, determine whether to open the door through a preset password sequence, and trigger the internal control module; the internal control module is used to configure the diode array in time-division multiplexing mode, and trigger different internal control functions of the oxygen chamber when the user displays different gestures. This invention abandons the traditional contact-based operation of oxygen chambers, using GaN MQW diode arrays for both external and internal control of the oxygen chamber, offering the following advantages: High device integration: Access control, lighting, and control can be achieved using a small number of GaN MQW diodes; Strong dual-function capability: Time multiplexing switching is achieved through MCU software, solving the problem of diodes not being able to emit light / detect simultaneously; Convenient operation: External support for mobile phone optical password unlocking and internal support for contactless gesture control; Privacy and security: Avoids privacy leaks and is unaffected by electromagnetic shielding interference; Scalability: The number of internal diode arrays can be increased as needed, and functions can be software-defined to adapt to various scenarios. Attached Figure Description

[0017] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the structure of an intelligent control system for an oxygen chamber based on a GaN MQW diode array according to the present invention; Figure 2 This is a schematic diagram of the external control module of the present invention. Detailed Implementation

[0019] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0020] Example: Figure 1As shown, the present invention provides a technical solution for an intelligent control system for an oxygen chamber based on a GaN MQW diode array, including an external control module and an internal control module; It should be noted that the hardware components of the oxygen chamber in this solution mainly include: a diode array (customized InGaN / GaN multi-MQW structure diodes): divided into an internal array and an external array, with the internal array inside the oxygen chamber and the external array on the outer door of the oxygen chamber; and an MCU unit (STM32F103C8T6 or other equivalent models (Cortex-M3 with 12-bit ADC), or other microprocessors with corresponding functions such as the STC series and Arduino series): controlling the switching of the diodes between lighting / detection modes, used for controlling the lighting inside the chamber and receiving changes in external light signals to enable lighting and control the oxygen chamber to perform corresponding actions. The control section power module is used for power supply. The drive circuit is connected to the diode array. The diode array is divided into an internal array and an external array. The internal array is inside the oxygen chamber, and the external array is on the outer door of the oxygen chamber. It is used to receive and emit light signals. The drive circuit is connected to the microprocessor unit. The microprocessor unit is connected to the oxygen chamber execution circuit, drive circuit, acquisition circuit and microprocessor peripheral circuit. The acquisition circuit is connected to the microprocessor unit.

[0021] External control module: such as Figure 2 As shown, several GaN MQW diodes are deployed on the oxygen chamber to form a diode array. The diode array includes an illumination mode, a detection mode, and a time-division multiplexing mode that alternates between the illumination mode and the detection mode. When the oxygen chamber door is closed, the external control module is used to configure the diode array to the detection mode. Monitoring equipment is deployed on the oxygen chamber to realize real-time monitoring of the surrounding environment of the oxygen chamber. The GaN MQW diode has the function of transmitting / receiving optical signals. In illumination mode, it transmits optical signals; in detection mode, it receives optical signals; and in time-division multiplexing mode, it can realize the dual function of transmitting / receiving optical signals. When the oxygen chamber door is closed, the diode is preset to collect the coded optical signal at a certain frequency. The user uses a lighting device to transmit the coded optical signal into the oxygen chamber. In this embodiment, it can be a mobile phone flash. When the diode array operates according to the collection frequency and receives the coded optical signal, it retrieves the monitoring video from a previous period of time. Using object detection algorithms, all moving objects in the surveillance video are captured, and all local regions of each moving object are obtained, including the eyes, ears, nose and mouth of the face; Acquire a standard light source image free from environmental interference, convert the standard light source image into a grayscale image, and establish a baseline histogram based on the grayscale value of each pixel in the grayscale image; acquire an ROI region image of the moving object, convert the ROI region image into a grayscale image, and establish a target histogram based on the grayscale value of each pixel in the grayscale image. Since the user needs to hold the light device, the ROI region in this embodiment is the hand region. The process of establishing the baseline histogram is as follows: First, the gray value of each pixel in the grayscale image converted from the standard light source image is obtained. Since the gray value ranges from 0 to 255, grayscale downgrading is performed in this scheme. The gray values ​​are divided into 8 levels according to 0-31, 32-63, ..., 233-255. Pixels belonging to the same grayscale level are collected to obtain the ratio of the number of pixels of each grayscale level to the total number of pixels in the grayscale image. Based on the grayscale level and the corresponding ratio, the baseline histogram corresponding to the grayscale image is established. The method of establishing the target histogram can also refer to this method.

[0022] If the number of local regions of any moving object exceeds a preset threshold, and the histogram similarity between the corresponding baseline histogram and the target histogram exceeds a preset similarity threshold, the encoded optical signal is deemed reliable.

[0023] When personnel wish to activate the oxygen chamber, they need to face the chamber and hold a light device. Therefore, this confidence condition is set to exclude those who accidentally activate it. The judgment is based on a local area because facing the chamber is required to activate it, and on histogram similarity because holding a light device is necessary; both are indispensable. The histogram similarity is calculated using cosine similarity, which is existing technology and will not be elaborated upon here.

[0024] If the encoded optical signal is deemed reliable, high-frequency pulses are set to represent encoded bit 1 and low-frequency pulses to represent encoded bit 0. Through sampling, quantization, and decoding, the encoded optical signal is quantized by ADC, converting the received encoded optical signal into a binary digital sequence with 1s and 0s. Obtain the set of password sequences for opening the hatch. Each password sequence in the set has the same length, N. Set password update rules. The password update rules can be based on the last digit of the sequence code. Add a code different from the last digit of the sequence code to the end. For example, if a password sequence is a 4-digit "1001", the updated password sequence is "10010". If a password sequence is "1010", the updated password sequence is "10101". Then, based on each updated password sequence, obtain the updated password sequence set. The updated password sequence set obtained here is designed to protect password security, and different password update rules can be set according to the actual situation.

[0025] If the length of the converted number sequence is N, an updated number sequence is obtained according to the password update rules. This updated number sequence is then matched against each element in the updated password sequence set. If a match is found, the hatch is opened; otherwise, the hatch remains closed. When the hatch is determined to be open, the internal control module is triggered.

[0026] Traditional oxygen chambers typically use a contact-based opening mechanism, requiring users to open the door via a handle. This is not only inconvenient but also detrimental to aseptic environment control. Therefore, this solution leverages the characteristics of GaN MQW diodes to enable intelligent door opening, which helps reduce bacterial transmission and creates a clean and comfortable environment for the oxygen chamber.

[0027] Internal control module: Used to configure the diode array in time-division multiplexing mode. In illumination mode, the diode array actively emits modulated light and in detection mode, it receives reflected light signals. When the user displays different gestures, the reflected light signals received by the diode array will produce different differences, thereby triggering different internal control functions of the oxygen chamber.

[0028] Different hand gestures correspond to different control functions of the oxygen chamber. Sample data of different hand gestures from several users were collected. The sample data includes the type of hand gesture, the time when each diode receives the reflected light signal, and the control function of the oxygen chamber. The hand gestures include hovering, moving forward, moving backward, moving left, moving right, moving up, and moving down. For example, moving left decreases the oxygen concentration, moving right increases the oxygen concentration, moving forward increases the illumination, moving backward decreases the illumination, and so on.

[0029] Identify several user subjects with different physical characteristics, including height, weight, gender, and age, and have each user subject perform different hand gestures; use the sample data of the oxygen chamber triggering the correct control function after the user subject performs the corresponding hand gesture as the target data of the source domain client. A federated network model is established, with each user object as a source domain client. The local models obtained by parallel training of each source domain client and the global model obtained by aggregating each local model are obtained from the federated network model. Based on the target data collected from each source domain client, the local models and global models corresponding to each source domain client are trained to achieve federated network model training. The federated network model is a distributed collaborative training system under the federated learning architecture, consisting of a central node (global model) and multiple client nodes (local models). It has the advantages of balancing data privacy and global model performance. In this scheme, the time when each diode in the target data receives the reflected light signal is used as input, and the oxygen chamber control function is used as output to train the local model and the global model, thereby realizing the training of the federated network model. The trained federated network model is used to obtain the corresponding oxygen chamber control function based on the time when each diode receives the reflected light signal. Since the federated network model is an existing technology, it will not be described in detail here.

[0030] A light sensor is deployed at the gesture recognition point in the oxygen chamber to monitor the light intensity at the gesture recognition point in real time when the user makes different gestures; the current light intensity X at the gesture recognition point is obtained. The user's gestures were collected at different duty cycles. The number of sample data and the number of target data were extracted when the illumination intensity was X and the duty cycle was K, so as to obtain the success rate when the illumination intensity was X and the duty cycle was K. Establish a scatter plot of the successful recognition rate as a function of duty cycle corresponding to light intensity X, and adjust the duty cycle in time-sharing multiplexing mode in real time based on the scatter plot.

[0031] In this scheme, the duty cycle is the proportion of the time the diode array is in illumination mode within a complete working cycle of the time-division multiplexing mode, ranging from 0 to 1. The closer it is to 1, the longer the illumination mode time. The rule for setting the duty cycle is as follows: set the minimum duty cycle to K0, obtain all scatter plot data with a duty cycle greater than K0, and take the duty cycle corresponding to the scatter plot with the highest success rate as the current duty cycle of the diode array in illumination mode.

[0032] For example, when the minimum duty cycle is K0=0.6, first extract all scattered data with a duty cycle greater than 0.6, and take the duty cycle corresponding to the scattered point with the highest success rate as the current duty cycle of the diode array in the illumination mode. Here, both illumination and success rate can be guaranteed, so as to reduce the false recognition rate and improve user satisfaction.

[0033] Gesture recognition relies on the signal of actively modulated light emitted by a diode reflected by the palm. Ambient light inside the oxygen chamber, such as interior lighting and direct sunlight, will generate background noise at the diode detection end. Therefore, this solution automatically adjusts the duty cycle of the emission / detection time multiplexing based on the real-time measured illuminance inside the chamber to extend the detection time as much as possible while meeting illuminance requirements, thereby improving the robustness of gesture recognition. The specific adjustment can be made according to actual conditions.

[0034] Based on the timing of the reflected light signals received by each diode, and using the trained federated network model, the internal control functions of the oxygen chamber are output to achieve oxygen chamber control.

[0035] Since traditional oxygen chambers require additional independent switches for internal control, and various internal control functions require multiple buttons, this solution leverages the time-division multiplexing mode of GaN MQW diodes, which have the characteristics of emitting modulated light and receiving reflected light signals, to realize different internal control functions, improve the integration and safety of the oxygen chamber system, and significantly reduce equipment deployment and maintenance costs.

[0036] Compared with traditional oxygen chambers, this solution offers at least the following benefits: High device integration: Multiple functions such as access control, lighting, and control can be achieved using a small number of GaN MQW diodes; Strong dual-function capability: Time multiplexing switching is achieved through MCU software, solving the problem that diodes cannot emit light / detect simultaneously; Convenient operation: External support for mobile phone optical password unlocking, and internal support for contactless gesture control; Privacy and security: Avoids privacy leaks and is not affected by electromagnetic shielding interference; Scalability: The number of internal diode arrays can be increased as needed, and functions can be software defined to adapt to various scenarios.

[0037] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0038] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0039] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent control system for oxygen chambers based on GaN MQW diode arrays, characterized in that, The system includes an external control module and an internal control module; External control module: Several GaN MQW diodes are deployed on the oxygen chamber to form a diode array. The diode array includes an illumination mode, a detection mode, and a time-division multiplexing mode that alternates between the illumination mode and the detection mode. When the oxygen chamber door is closed, the external control module is used to configure the diode array to the detection mode. When the diode array receives an encoded optical signal, it determines whether to open the door through a preset password sequence. When it is determined that the door is to be opened, the internal control module is triggered. Internal control module: Used to configure the diode array in time-division multiplexing mode. In illumination mode, the diode array actively emits modulated light and in detection mode, it receives reflected light signals. When the user displays different gestures, the reflected light signals received by the diode array will produce different differences, thereby triggering different internal control functions of the oxygen chamber.

2. The intelligent control system for an oxygen chamber based on a GaN MQW diode array according to claim 1, characterized in that, The steps to determine whether to open the hatch using a preset password sequence are as follows: Monitoring equipment is deployed on the oxygen chamber to enable real-time monitoring of the surrounding environment. The diode array is preset to collect the coded light signal at a frequency. Users use a lighting device to emit the coded light signal into the oxygen chamber. When the diode array operates at the collection frequency and receives the coded light signal, it retrieves the monitoring video from a period of time before receiving the signal, captures and analyzes the moving objects in the monitoring video, and determines whether the coded light signal is reliable. If the coded optical signal is deemed reliable, it is converted into a digital sequence through sampling, quantization, and decoding. The converted digital sequence is then compared with a pre-stored set of cipher sequences to determine whether to open the hatch.

3. The intelligent control system for an oxygen chamber based on a GaN MQW diode array according to claim 2, characterized in that, The steps to capture and analyze moving objects in surveillance video and determine the reliability of encoded optical signals are as follows: Using object detection algorithms, all moving objects in the surveillance video are captured, and all local regions of each moving object are obtained. The local regions include the eyes, ears, nose and mouth of the face. Acquire a standard light source image free from environmental interference, convert the standard light source image into a grayscale image, and establish a baseline histogram based on the grayscale value of each pixel in the grayscale image; Obtain the ROI region image of the moving object, convert the ROI region image into a grayscale image, and build a target histogram based on the grayscale value of each pixel in the grayscale image. If the number of local regions of any moving object exceeds a preset threshold, and the histogram similarity between the corresponding baseline histogram and the target histogram exceeds a preset similarity threshold, the encoded optical signal is deemed reliable.

4. The intelligent control system for an oxygen chamber based on a GaN MQW diode array according to claim 2, characterized in that, The steps to compare the converted digital sequence with the pre-stored password sequence set to determine whether to open the hatch are as follows: Obtain the set of password sequences for opening the hatch, where each password sequence in the set has a length of N. Set password update rules to obtain an updated password sequence set. If the length of the converted number sequence is N, obtain an updated number sequence according to the password update rules. Match the updated number sequence with each element in the updated password sequence set. If the match is successful, determine that the hatch is open. If the match fails, keep the hatch closed.

5. The intelligent control system for an oxygen chamber based on a GaN MQW diode array according to claim 1, characterized in that, The steps to trigger different internal control functions of the oxygen chamber are as follows: Different hand gestures correspond to different control functions of the oxygen chamber. Sample data of different hand gestures made by several users are collected, and target data is extracted from the sample data. The sample data includes the type of hand gesture, the time when each diode receives the reflected light signal, and the oxygen chamber control function. A federated network model is established, with each user object as a source domain client. The local models obtained by parallel training of each source domain client and the global model obtained by aggregating each local model are obtained from the federated network model. Based on the target data collected from each source domain client, the local models corresponding to each source domain client and the global model are trained to realize the training of the federated network model. Based on changes in the oxygen chamber environment, the duty cycle of the switching time between the detection mode and the lighting mode in the time-sharing multiplexing mode is adjusted in real time. Based on the current time when each diode receives the reflected light signal, and based on the trained federated network model, different internal control functions of the oxygen chamber are triggered.

6. The intelligent control system for an oxygen chamber based on a GaN MQW diode array according to claim 5, characterized in that, The steps to achieve real-time adjustment of the duty cycle of the switching time between detection mode and illumination mode in time-sharing multiplexing mode are as follows: Light sensors are deployed at the gesture recognition point in the oxygen chamber to monitor the light intensity at the gesture recognition point in real time when users make different gestures. Get the current illumination intensity X at the gesture recognition location; The user's gestures were collected at different duty cycles. The number of sample data and the number of target data were extracted when the illumination intensity was X and the duty cycle was K, so as to obtain the success rate when the illumination intensity was X and the duty cycle was K. A scatter plot is created to show the success rate as a function of duty cycle, corresponding to light intensity X. Based on the scatter plot, the duty cycle in the time-sharing multiplexing mode is adjusted in real time.

7. The intelligent control system for an oxygen chamber based on a GaN MQW diode array according to claim 5, characterized in that, The steps to extract target data from sample data are as follows: identify several user objects with different physical characteristics, including height, weight, gender and age, and have each user object make a different gesture; use the sample data of the oxygen chamber triggering the correct control function after the user object makes the corresponding gesture as the target data of the source domain client.

8. The intelligent control system for an oxygen chamber based on a GaN MQW diode array according to claim 6, characterized in that, The steps to achieve real-time adjustment of the duty cycle in time-division multiplexing mode are as follows: The duty cycle is the proportion of the time the diode array is in illumination mode to the total cycle time within a complete working cycle of time-division multiplexing mode, with a value of 0 to 1; the minimum duty cycle is set to K0, all scatter plots with a duty cycle greater than K0 are obtained, and the duty cycle corresponding to the scatter plot with the highest success rate is taken as the current duty cycle of the diode array in illumination mode.

9. The intelligent control system for an oxygen chamber based on a GaN MQW diode array according to claim 5, characterized in that, Gestures include hovering, moving forward, moving backward, moving left, moving right, moving up, and moving down.

Images