Internet of Things and application development comprehensive experiment box based on integration algorithm

Abstract

The invention relates to the field of Internet of Things, and particularly discloses an Internet of Things and application development comprehensive experiment box based on an integration algorithm. Comprising an experiment box structure module, an experiment box hardware module, an experiment box system module, an equipment communication module, a communication transmission module, a voice recognition module and an automatic control module. According to the Internet of Things and application development comprehensive experiment box based on the integration algorithm, the sensor hardware integrated in the experiment box is combined with the Internet of Things sensing and communication technology, a built-in microphone array assists developers in application development, and experiment teaching requirements are met.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things, and in particular to a comprehensive experiment box for Internet of Things and application development based on integrated algorithms. Background Art

[0002] With the development and popularization of Internet of Things (IoT) technology, IoT devices are increasingly used in various fields. From smart homes to smart cities, from industrial automation to agricultural monitoring, IoT is gradually changing our way of life and production. IoT builds an intelligent world by organically combining sensors, actuators, networks and application software. However, with the continuous deepening and diversification of IoT applications, the development, debugging and deployment of IoT systems are facing more and more challenges.

[0003] First, IoT systems are usually composed of many different types of hardware devices and software systems. These devices and systems need to work together efficiently to meet the needs of specific applications. The data collected by sensors needs to be processed and transmitted in real time, control instructions need to be accurately delivered to actuators, and communication between various devices needs to be efficient and reliable. However, compatibility and interoperability issues between different devices and systems make the development and debugging of IoT systems complicated and time-consuming.

[0004] Secondly, the development and testing of IoT systems requires a stable and flexible experimental environment. Traditional laboratories usually need to purchase and deploy a large number of hardware devices, such as sensors, actuators, network devices, etc., and these devices often need to be manually connected and configured. This not only increases the cost of development and testing, but also increases the difficulty of system failure and debugging. In addition, with the rapid development of IoT technology, new sensors and communication technologies are constantly emerging. How to quickly integrate and test these new technologies in an experimental environment has become a problem that needs to be solved.

[0005] Furthermore, the data transmission and processing efficiency of the IoT system directly affects the system performance and user experience. Traditional static channel allocation methods often fail to fully utilize network resources in the face of dynamically changing network environments, resulting in low communication efficiency. At the same time, with the increase in the number of IoT devices, network congestion and data packet loss are becoming increasingly serious.

[0006] In addition, with the development of artificial intelligence technology, speech recognition has gradually become an important means of human-computer interaction. In the Internet of Things system, speech recognition technology can control various smart devices through voice commands, greatly improving the user experience. However, the existing speech recognition system has a low recognition accuracy in noisy environments and cannot meet the application requirements in complex environments. Summary of the invention

[0007] The purpose of the present invention is to provide an integrated experiment box for Internet of Things and application development based on an integrated algorithm, aiming to solve the problems mentioned in the above background. The present invention provides an integrated, modular and intelligent experimental environment by integrating multiple functional modules, aiming to solve many problems in the development and application of existing Internet of Things systems. The comprehensive experiment box integrates sensors, processors, communication modules and control modules, which can realize the rapid development, debugging and deployment of Internet of Things devices. At the same time, it adopts the advanced Hongmeng operating system, provides good compatibility and scalability, and supports a variety of applications and development tools. The experiment box can not only be used for the research and development of Internet of Things technology, but also can be used as a teaching and training tool to provide strong support for the cultivation of Internet of Things professionals.

[0008] In order to achieve the above-mentioned purpose, the present invention proposes an integrated experiment box for Internet of Things and application development based on an integrated algorithm, including an experiment box structure module, an experiment box hardware module, an experiment box system module, a device communication module, a communication transmission module, a speech recognition module, and an automatic control module; in the experiment box structure module, all the experiment box hardware is integrated inside the experiment box; in the experiment box hardware module, the Internet sensor is integrated inside the comprehensive experiment box; in the experiment box system module, the comprehensive experiment box adopts the Hongmeng operating system; in the device communication module, the comprehensive experiment box is connected and communicated with external devices through wired and wireless connections; in the communication transmission module, a dynamic channel allocation algorithm is used to divide the channel into sub-channels for dynamic channel allocation; in the speech recognition module, an integrated processing algorithm is proposed to construct a distributed learner, form an integrated learner, and perform speech signal enhancement processing; in the automatic control module, based on the results of speech recognition analysis, the comprehensive experiment box transmits control instructions to control the corresponding equipment.

[0009] Furthermore, in the experimental box structure module, there is a device opening on the front of the experimental box body, a hardware installation position is set inside the experimental box body, and there is a device interface on the outside of the experimental box, and the device interface includes an RJ-45 network interface, a USB-A interface, and a USB-C interface.

[0010] Furthermore, in the hardware module of the experimental box, a variety of sensor hardware are integrated and installed inside, and the sensor hardware includes a temperature sensor, a light sensor, an infrared sensor, an ultrasonic sensor, a voice sensor, an LED dot matrix unit, a network communication unit, a display unit, and a keyboard control unit.

[0011] Furthermore, in the experiment box system module, the experiment box operating system adopts the OpenHarmony operating system.

[0012] Furthermore, in the device communication module, a connection is made with an external device through an interface to perform device communication connection and signal data transmission.

[0013] Furthermore, the dynamic channel allocation algorithm first defines the receiving rules of the receiving layer, then divides the channel into sub-channels, calculates the probability of sub-channel interference signals, and based on the principle of minimization of allocation, obtains the channel allocation rules for signal data transmission and performs dynamic channel allocation.

[0014] Furthermore, the channels are dynamically allocated, and the detailed process is as follows: During the signal data transmission between the integrated experimental box and the external equipment, the interference signal generated affects the signal data transmission process. The signal data sending layer sends signal data, and the signal data receiving layer receives data. The signal data receiving rules of the signal data receiving layer are expressed as follows: , in, Indicates the signal data received by the signal receiving layer. represents the transmission attenuation coefficient, Indicates the signal data of the signal data transmission layer, Indicates the sending layer interference signal, Indicates the interference signal of the receiving layer, Represents the noise interference signal. The signal data transmission channel is divided into sub-channels. The signal data is modulated into the sub-channel. The noise interference signal appears in the transmission process of the signal data in the sub-channel. The occurrence probability of the noise interference signal in the sub-channel is calculated. The calculation formula is as follows: , middle, represents the probability of noise interfering with the signal, represents the variance, Indicates the peak value of the noise interference signal, represents the noise sampling time, Indicates the peak signal sampling time. Based on the probability of noise interfering with the signal, the sub-channel modulation coefficient is calculated. The calculation formula is as follows: , in, Indicates Sub-channel modulation coefficients, Indicates Iteration cycles, Indicates Iteration cycle sampling frequency, represents the amount of noise interference signal, Indicates The sub-channel width is used, and the sub-channel modulation coefficient is used. Based on the principle of minimization allocation, the channel allocation rule for signal data transmission is obtained. The allocation rule is as follows: , Based on the allocation rules, channel allocation is performed during signal data transmission. The present invention proposes a dynamic channel allocation algorithm, defines the receiving rules of the receiving layer, divides the channel into sub-channels, calculates the probability of sub-channel interference signals, and obtains the channel allocation rules for signal data transmission based on the principle of minimization of allocation. The channel allocation process proposed by the present invention improves the signal transmission quality during the use of the comprehensive experimental box.

[0015] Furthermore, the integrated processing algorithm first pre-processes the speech signal, then constructs a distributed learner, uses the distributed learner to train speech signal processing, and based on the distributed learner, forms an integrated learner to perform speech signal enhancement processing.

[0016] Furthermore, based on the distributed learner, an integrated learner is formed to perform speech signal enhancement processing. The detailed process is as follows: The comprehensive experiment box is connected to the microphone to collect the user's voice signal. For the collected voice signal, the comprehensive experiment box first performs a voice signal preprocessing operation. The preprocessing process is shown as follows: , in, represents the speech signal at k sampling time, represents the speech signal at k-1 sampling time, represents the speech signal at time k after preprocessing, represents the length of the time domain, , They represent the weighting coefficient at the k sampling moment and the weighting coefficient at the k-1 sampling moment respectively. Based on the preprocessed speech signal, an integrated learner is constructed. The integrated learner is composed of multiple distributed learners. Each distributed learner performs speech signal recognition and speech signal enhancement respectively. The distributed learner fuses signals of different phases to obtain multi-level time-frequency features, enhances the amplitude spectrum and the frequency domain spectrum, and performs speech signal waveform enhancement. During the processing, a loss function is constructed for the training of the distributed learner. The training set data is used to train the distributed learner. The training set data includes training data and verification data. The loss function of the distributed learner is expressed as follows: , in, represents the loss function, Indicates A distributed learner with iterative learning. Indicates A distributed learner, Indicates The iteration coefficient of the initial distributed learner, Indicates Distributed Learners in Iterative Learning The error between the training part of the data and the validation part of the data at the t moment is based on minimizing the loss function, training the distributed learner, and forming an integrated learner through the combination of distributed learners. Different distributed learners have different degrees of advantage in the integrated learner. The advantage coefficient of the distributed learner is calculated as follows: , in, Represents a distributed learner The favorable coefficient of Represents a distributed learner The loss function is based on the favorable coefficients and the distributed learners are combined. The combination process is as follows: , in, Indicates After the iteration, the integrated learner processes the speech signal through the integrated learner. The integrated processing algorithm proposed in the present invention first pre-processes the speech signal, performs preliminary signal enhancement, then constructs a distributed learner, uses training data and verification data to train the distributed learner for speech signal processing, constructs a loss function, performs iterative optimization of the distributed learner, and then calculates the favorable coefficient of the distributed learner to form an integrated learner to perform speech signal enhancement processing. The integrated processing algorithm proposed in the present invention can enhance the speech signal and improve the speech recognition stability and robustness of the comprehensive experimental box.

[0017] Furthermore, in the automatic control module, based on the results of speech recognition analysis, control instructions are transmitted to the controlled equipment to perform intelligent automatic control of the equipment, thereby realizing artificial intelligence practical training in experimental teaching.

[0018] The beneficial effects of the present invention are as follows: 1. The present invention proposes a dynamic channel allocation algorithm, defines the receiving rules of the receiving layer, divides the channel into sub-channels, calculates the probability of sub-channel interference signals, obtains the channel allocation rules for signal data transmission based on the principle of minimization allocation, and performs dynamic channel allocation. The channel allocation process proposed by the present invention improves the signal transmission quality during the use of the comprehensive experimental box; 2. The integrated processing algorithm proposed in the present invention first pre-processes the speech signal, performs preliminary signal enhancement, then constructs a distributed learner, uses training data and verification data to train the distributed learner for speech signal processing, constructs a loss function, performs iterative optimization of the distributed learner, and then calculates the favorable coefficient of the distributed learner to form an integrated learner to perform speech signal enhancement processing. The integrated processing algorithm proposed in the present invention can enhance the speech signal and improve the speech recognition stability and robustness of the comprehensive experimental box; 3. The comprehensive experiment box for Internet of Things and application development based on integrated algorithm proposed in the present invention integrates various hardware and software modules required for Internet of Things development in a compact box, reducing the hardware procurement and wiring costs required for traditional laboratories. At the same time, the integrated design makes hardware maintenance and upgrading more convenient, and users can replace or add new sensors and equipment at any time as needed; the experiment box adopts Hongmeng operating system, providing a stable, compatible and scalable system environment. Hongmeng operating system supports multi-task parallel processing and rich application program interfaces, and users can easily develop and test applications. Good compatibility ensures that the experiment box can support a variety of different sensors and peripheral devices; it is suitable for the research and development of Internet of Things technology, and can also be widely used in teaching and training. The experiment box provides a complete Internet of Things experimental environment, and students and researchers can use the experiment box to learn and experiment with Internet of Things technology, and quickly master the development and application skills of Internet of Things systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings in the following description are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 is a system schematic diagram of the present invention; DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] See also Figure 1In order to achieve the above-mentioned purpose, the present invention proposes an integrated experiment box for Internet of Things and application development based on an integrated algorithm, including an experiment box structure module, an experiment box hardware module, an experiment box system module, a device communication module, a communication transmission module, a speech recognition module, and an automatic control module; in the experiment box structure module, all the experiment box hardware is integrated inside the experiment box; in the experiment box hardware module, the comprehensive experiment box integrates Internet sensors inside; in the experiment box system module, the comprehensive experiment box adopts the Hongmeng operating system; in the device communication module, the comprehensive experiment box is connected and communicated with external devices through wired and wireless connections; in the communication transmission module, a dynamic channel allocation algorithm is used to divide the channel into sub-channels and dynamically allocate channels; in the speech recognition module, an integrated processing algorithm is proposed to construct a distributed learner, form an integrated learner, and perform speech signal enhancement processing; in the automatic control module, based on the results of speech recognition analysis, the comprehensive experiment box transmits control instructions to control the corresponding equipment.

[0023] See also Figure 1 Specifically, in the experimental box structure module, there is a device opening on the front of the experimental box body, a hardware installation position is set inside the experimental box body, and there is a device interface on the outside of the experimental box, and the device interface includes an RJ-45 network interface, a USB-A interface, and a USB-C interface.

[0024] See also Figure 1 Specifically, in the hardware module of the experimental box, a variety of sensor hardware are integrated and installed inside, and the sensor hardware includes a temperature sensor, a light sensor, an infrared sensor, an ultrasonic sensor, a voice sensor, an LED dot matrix unit, a network communication unit, a display unit, and a keyboard control unit.

[0025] See also Figure 1 ,Specifically, in the experimental box system module, the experimental box operating system adopts the ,OpenHarmony operating system.

[0026] See also Figure 1 Specifically, in the device communication module, it is connected with the external device through the interface to perform device communication connection and signal data transmission.

[0027] See also Figure 1 ,Specifically, the dynamic channel allocation algorithm first defines the receiving ,rules of the receiving layer, then divides the channel into sub-channels, ,calculates the probability of sub-channel interference signals, and obtains the ,channel allocation rules for signal data transmission based on the principle of ,minimizing allocation, and performs dynamic channel allocation. ,The detailed process is as follows: During the signal data transmission between the integrated experimental box and the external equipment, the interference signal generated affects the signal data transmission process. The signal data sending layer sends signal data, and the signal data receiving layer receives data. The signal data receiving rules of the signal data receiving layer are expressed as follows: , in, Indicates the signal data received by the signal receiving layer. represents the transmission attenuation coefficient, Indicates the signal data of the signal data transmission layer, Indicates the sending layer interference signal, Indicates the interference signal of the receiving layer, Represents the noise interference signal. The signal data transmission channel is divided into sub-channels. The signal data is modulated into the sub-channel. The noise interference signal appears in the transmission process of the signal data in the sub-channel. The occurrence probability of the noise interference signal in the sub-channel is calculated. The calculation formula is as follows: , in, represents the probability of noise interfering with the signal, represents the variance, Indicates the peak value of the noise interference signal, represents the noise sampling time, Indicates the peak signal sampling time. Based on the probability of noise interfering with the signal, the sub-channel modulation coefficient is calculated. The calculation formula is as follows: , in, Indicates Sub-channel modulation coefficients, Indicates Iteration cycles, Indicates Iteration cycle sampling frequency, represents the amount of noise interference signal, Indicates The sub-channel width is used, and the sub-channel modulation coefficient is used. Based on the principle of minimization allocation, the channel allocation rule for signal data transmission is obtained. The allocation rule is as follows: , Based on the allocation rules, channel allocation is performed during signal data transmission.

[0028] See also Figure 1 ,Specifically, the integrated processing algorithm first pre-processes the speech signal, and then constructs a distributed learner, and uses the distributed learner to train speech signal ,processing. Based on the distributed learner, an integrated learner is formed to perform ,speech signal enhancement processing.

[0029] See also Figure 1 Specifically, based on the distributed learner, an integrated learner is formed to perform speech signal enhancement processing. The detailed process is as follows: The comprehensive experiment box is connected to the microphone to collect the user's voice signal. For the collected voice signal, the comprehensive experiment box first performs a voice signal preprocessing operation. The preprocessing process is shown as follows: , in, represents the speech signal at k sampling time, represents the speech signal at k-1 sampling time, represents the speech signal at time k after preprocessing, represents the length of the time domain, , They represent the weighting coefficient at the k sampling moment and the weighting coefficient at the k-1 sampling moment respectively. Based on the preprocessed speech signal, an integrated learner is constructed. The integrated learner is composed of multiple distributed learners. Each distributed learner performs speech signal recognition and speech signal enhancement respectively. The distributed learner fuses signals of different phases to obtain multi-level time-frequency features, enhances the amplitude spectrum and the frequency domain spectrum, and performs speech signal waveform enhancement. During the processing, a loss function is constructed for the training of the distributed learner. The training set data is used to train the distributed learner. The training set data includes training data and verification data. The loss function of the distributed learner is expressed as follows: , in, represents the loss function, Indicates A distributed learner with iterative learning. Indicates A distributed learner, Indicates The iteration coefficient of the initial distributed learner, Indicates Distributed Learners in Iterative Learning The error between the training part of the data and the validation part of the data at the t moment is based on minimizing the loss function, training the distributed learner, and forming an integrated learner through the combination of distributed learners. Different distributed learners have different degrees of advantage in the integrated learner. The advantage coefficient of the distributed learner is calculated as follows: , in, Represents a distributed learner The favorable coefficient of Represents a distributed learner The loss function is based on the favorable coefficients and the distributed learners are combined. The combination process is as follows: , in, Indicates After the iteration, the integrated learner processes the speech signal through the integrated learner.

[0030] See also Figure 1 ,Specifically, in the automatic control module, based on the results of speech recognition ,analysis, control instructions are transmitted to the controlled equipment, ,and intelligent automatic control of the equipment is performed, realizing artificial intelligence ,practical training teaching in experimental teaching.

[0031] In a specific embodiment, when the user uses the comprehensive experiment box, he first understands the specific structure and composition of the comprehensive experiment box. The hardware of the experiment box is all integrated inside the experiment box, including temperature sensors, light sensors, infrared sensors, ultrasonic sensors, voice sensors, LED dot matrix units, network communication units, display units, and keyboard control units. The comprehensive experiment box adopts the Hongmeng operating system. The comprehensive experiment box is connected and communicates with external devices through wired and wireless connections. When the user uses the comprehensive experiment box, the comprehensive experiment box is connected to the external Internet of Things device and the dynamic channel allocation algorithm is used to dynamically allocate channels. The user outputs voice commands and the integrated processing algorithm and the integrated learner are used to perform voice signal enhancement processing. The analyzed user voice commands are transmitted to the corresponding devices through the comprehensive experiment box to control the devices and realize artificial intelligence teaching.

[0032] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0033] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0034] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A comprehensive experiment box for Internet of Things and application development based on integrated algorithms, characterized in that: It includes an experimental box structure module, an experimental box hardware module, an experimental box system module, a device communication module, a communication transmission module, a speech recognition module, and an automatic control module; in the experimental box structure module, all the experimental box hardware is integrated inside the experimental box; in the experimental box hardware module, the Internet of Things sensor is integrated inside the comprehensive experimental box; in the experimental box system module, the comprehensive experimental box adopts the Hongmeng operating system; in the device communication module, the comprehensive experimental box is connected and communicated with external devices through wired and wireless connections; in the communication transmission module, a dynamic channel allocation algorithm is used to divide the channel into sub-channels and dynamically allocate channels; in the speech recognition module, an integrated processing algorithm is proposed, a distributed learner is constructed, and an integrated learner is composed to perform speech signal enhancement processing; In the automatic control module, based on the results of the speech recognition analysis, the integrated experimental box transmits control instructions to control the corresponding equipment.

2. According to claim 1, a comprehensive experiment box for Internet of Things and application development based on integrated algorithm, characterized in that: The experimental box structural module has an equipment opening at the front of the experimental box body, a hardware installation position is set inside the experimental box body, and an equipment interface is provided outside the experimental box. The equipment interface includes an RJ-45 network interface, a USB-A interface, and a USB-C interface.

3. The Internet of Things and application development comprehensive experiment box based on integrated algorithm according to claim 1 is characterized in that: The experimental box hardware module has a variety of sensor hardware integrated and installed inside, and the sensor hardware includes a temperature sensor, a light sensor, an infrared sensor, an ultrasonic sensor, a voice sensor, an LED dot matrix unit, a network communication unit, a display unit, and a keyboard control unit.

4. The Internet of Things and application development comprehensive experiment box based on integrated algorithm according to claim 1 is characterized in that: The experimental box system module and the experimental box operating system adopt the OpenHarmony operating system.

5. The Internet of Things and application development comprehensive experiment box based on integrated algorithm according to claim 1 is characterized in that: The device communication module is connected to an external device through an interface to perform device communication connection and signal data transmission.

6. The Internet of Things and application development comprehensive experiment box based on integrated algorithm according to claim 1 is characterized in that: The dynamic channel allocation algorithm first defines the receiving layer receiving rules, then divides the channel into sub-channels, calculates the probability of sub-channel interference signals, obtains the channel allocation rules for signal data transmission based on the minimization allocation principle, and performs dynamic channel allocation.

7. The Internet of Things and application development comprehensive experiment box based on integrated algorithm according to claim 6 is characterized in that: The detailed process of dynamic channel allocation is as follows: During the signal data transmission between the integrated experimental box and the external equipment, the interference signal generated affects the signal data transmission process. The signal data sending layer sends signal data, and the signal data receiving layer receives data. The signal data receiving rules of the signal data receiving layer are expressed as follows: ， in, Indicates the signal data received by the signal receiving layer. represents the transmission attenuation coefficient, Indicates the signal data of the signal data transmission layer, Indicates the sending layer interference signal, Indicates the interference signal of the receiving layer, Represents the noise interference signal. The signal data transmission channel is divided into sub-channels. The signal data is modulated into the sub-channel. The noise interference signal appears in the transmission process of the signal data in the sub-channel. The occurrence probability of the noise interference signal in the sub-channel is calculated. The calculation formula is as follows: ， in, represents the probability of noise interfering with the signal, represents the variance, Indicates the peak value of the noise interference signal, represents the noise sampling time, Indicates the peak signal sampling time. Based on the probability of noise interfering with the signal, the sub-channel modulation coefficient is calculated. The calculation formula is as follows: ， in, Indicates Sub-channel modulation coefficients, Indicates Iteration cycles, Indicates Iteration cycle sampling frequency, represents the amount of noise interference signal, Indicates Sub-channel widths, using sub-channel modulation coefficients, based on the principle of minimization allocation, the channel allocation rules for signal data transmission are obtained. The allocation rules are as follows: ， Based on the allocation rules, channel allocation is performed during signal data transmission.

8. The Internet of Things and application development comprehensive experiment box based on integrated algorithm according to claim 1 is characterized in that: The integrated processing algorithm first pre-processes the speech signal, then constructs a distributed learner, uses the distributed learner to train speech signal processing, and based on the distributed learner, forms an integrated learner to perform speech signal enhancement processing.

9. The Internet of Things and application development comprehensive experiment box based on integrated algorithm according to claim 8 is characterized in that: The distributed learner is used to form an integrated learner to perform speech signal enhancement processing. The detailed process is as follows: The comprehensive experiment box is connected to the microphone to collect the user's voice signal. For the collected voice signal, the comprehensive experiment box first performs a voice signal preprocessing operation. The preprocessing process is shown as follows: ， in, represents the speech signal at k sampling time, represents the speech signal at k-1 sampling time, represents the speech signal at time k after preprocessing, represents the length of the time domain, , They represent the weighting coefficient at the k sampling moment and the weighting coefficient at the k-1 sampling moment respectively. Based on the preprocessed speech signal, an integrated learner is constructed. The integrated learner is composed of multiple distributed learners. Each distributed learner performs speech signal recognition and speech signal enhancement respectively. The distributed learner fuses signals of different phases to obtain multi-level time-frequency features, enhances the amplitude spectrum and the frequency domain spectrum, and performs speech signal waveform enhancement. During the processing, a loss function is constructed for the training of the distributed learner. The training set data is used to train the distributed learner. The training set data includes training data and verification data. The loss function of the distributed learner is expressed as follows: ， in, represents the loss function, Indicates A distributed learner with iterative learning. Indicates A distributed learner, Indicates The iteration coefficient of the initial distributed learner, Indicates Distributed Learners in Iterative Learning The error between the training part of the data and the validation part of the data at the t moment is based on minimizing the loss function, training the distributed learner, and forming an integrated learner through the combination of distributed learners. Different distributed learners have different degrees of advantage in the integrated learner. The advantage coefficient of the distributed learner is calculated as follows: ， in, Represents a distributed learner The favorable coefficient of Represents a distributed learner The loss function is based on the favorable coefficients and the distributed learners are combined. The combination process is as follows: ， in, Indicates After the iteration, the integrated learner processes the speech signal through the integrated learner.

10. The Internet of Things and application development comprehensive experiment box based on integrated algorithm according to claim 1 is characterized in that: The automatic control module transmits control instructions to the controlled device based on the results of speech recognition analysis, performs intelligent automatic control of the device, and realizes artificial intelligence practical training in experimental teaching.

Images