Internet of Things monitoring method for monitoring unattended machine room

Abstract

The invention relates to the technical field of machine room monitoring, in particular to an Internet of Things monitoring method for unattended machine room monitoring, comprising an environment monitoring module, a comprehensive gateway, a radar module, a sound collection module and a machine room camera group, the environment monitoring module is used for monitoring machine room environment data; the sound collection module is used for collecting sound data in the machine room in different time periods and transmitting the sound data to the sound analysis module; the machine room camera group is used for acquiring video data in a machine room and monitoring intruders; the radar module is used for multi-target tracking and positioning; the comprehensive gateway comprises a video analysis module and a sound analysis module, and is used for generating an alarm signal according to the machine room environment data, the sound data and a video analysis result. According to the method, video monitoring, millimeter wave radar detection and sound recognition technologies are comprehensively applied, comprehensive monitoring of the internal environment of the machine room is achieved, the detection precision of abnormal events and the accuracy of fault diagnosis are improved, and meanwhile the visual effect of environment monitoring is enhanced.

Description

Technical Field

[0001] The invention relates to the technical field of computer room monitoring, and in particular to an Internet of Things monitoring method for unattended computer room monitoring. Background Art

[0002] As important infrastructure supporting the modern information society, the security and stability of rail transit and urban (suburban) railway computer rooms are increasingly valued. A large number of servers, network equipment and storage devices are densely deployed in the computer room, and the normal operation of these devices is crucial to ensure the operational safety of rail transit and urban (suburban) railways. Given that the computer rooms of important equipment at domestic rail transit and urban (suburban) railway stations generally adopt unattended mode, security precautions such as cameras, environmental monitoring facilities, and access control systems are usually installed in the computer room.

[0003] For example, the Chinese utility model patent with announcement number CN204102280U discloses a computer room monitoring system based on the Internet of Things, which consists of a computer room monitor and a computer room monitoring center. The computer room monitor includes a Zigbee communication module, a single-chip microcomputer, a temperature and humidity sensor, a wireless water leakage sensor, a smoke sensor, a water leakage alarm, a fire alarm, an air conditioner remote control and a power supply. The computer room monitoring center includes a wireless gateway, a remote server and a personal mobile communication terminal.

[0004] However, existing technologies are still insufficient in anomaly detection, fault diagnosis, and environmental monitoring. Although relevant safety precautions can provide the basic function of real-time computer room monitoring, their effectiveness is relatively limited in the current technical environment. Administrators can view the real-time images of the computer room at any time through cameras scattered in key locations in the computer room to grasp the latest developments in the computer room. However, although this real-time monitoring capability can theoretically enable administrators to respond quickly to various emergencies, there are still many delays and blind spots in operation, making the possibility of preventing and reducing safety accidents not as high as expected. Although computer room environmental monitoring sensors can monitor single parameters such as temperature and humidity, they lack comprehensive environmental assessment capabilities, and the collection and processing of sensor data may not be real-time enough, resulting in environmental anomalies that are difficult to detect and handle in a timely manner.

[0005] In general, although the current security precautions set up in the computer rooms of domestic rail transit and urban (suburban) railways mainly assist relevant staff in understanding the overall security situation of the computer room, they do play a certain role in computer room inspections, post-event investigations, and video linkage with the computer room environment monitoring system. However, due to its passive nature, lack of initiative, prevention, and intelligence, the role of traditional security precautions in current and future computer room monitoring is relatively limited, and it is difficult to assume the responsibility of computer room security protection alone. This situation highlights the urgency of improving monitoring technology, updating equipment, and enhancing system intelligence. Summary of the invention

[0006] The system of the present invention proposes an Internet of Things monitoring method for unattended computer room monitoring by integrating technologies such as video monitoring, millimeter-wave radar detection and sound recognition, and constructs an all-round and high-precision computer room monitoring system, which is particularly suitable for unattended computer room monitoring scenarios of key infrastructure such as urban rail transit, urban (suburban) railways, etc.

[0007] In order to achieve the above objectives, the following technical solutions are proposed: An Internet of Things monitoring method for unattended computer room monitoring comprises the following steps: S1: Build an IoT monitoring system for unattended computer room monitoring; S2: The system includes an environment monitoring module, an integrated gateway, a radar module, a sound collection module and a computer room camera group. The environment monitoring module includes a temperature sensor, a humidity sensor and a water immersion sensor for monitoring the computer room environment data; the sound collection module is used to collect sound data in the computer room at different time periods and transmit it to the sound analysis module; the computer room camera group includes several cameras for collecting video data in the computer room and monitoring intruders; the radar module is used for multi-target tracking and positioning, and when the movement of objects in the computer room is detected, the computer room camera group recording and video analysis module are started to analyze the video data; the integrated gateway includes a video analysis module and a sound analysis module, which are used to generate alarm signals according to the computer room environment data, sound data and video analysis results.

[0008] Preferably, the integrated gateway also includes a voiceprint library and a face library, the data in the voiceprint library is used as a basis for determining whether the sound data meets the alarm conditions; the data in the face library is used as a basis for determining whether the video analysis results meet the alarm conditions.

[0009] Preferably, the steps of the sound analysis module performing sound recognition and alarm include: denoising the sound data, comparing the denoised sound data with the data in the voiceprint library, and alarming when inconsistency with the normal scene voiceprint library or typical fault voiceprint is found.

[0010] Preferably, the denoising process uses IIR and FIR filters, and the sound data after denoising is converted into a structured vector using a deep learning model, and the structured vector is used for voiceprint comparison.

[0011] Preferably, the analysis functions of the video analysis module include: Compare the face recognized by the camera with the faces in the face database, and if it is not a face in the face database, issue an alarm; The video data collected by the camera is analyzed, and an alarm is issued if there is any illegal behavior or someone enters the restricted area of ​​the computer room.

[0012] Preferably, the analysis function of the video analysis module also includes smoke and fire detection, which performs real-time analysis on video data through a smoke and fire recognition algorithm, and generates an alarm linkage when the system detects smoke or flame.

[0013] Preferably, the radar module determines the physical information of the object by sending and receiving millimeter wave signals, and uses machine learning to perform dynamic and static analysis of targets in the environment to achieve multi-target tracking and positioning.

[0014] Preferably, the radar module includes several millimeter wave radars.

[0015] Preferably, the alarm signal is displayed in the video image in a character superimposed manner.

[0016] Beneficial effects of the present invention: The system of the present invention can realize comprehensive monitoring of the internal environment of the computer room by comprehensively using video monitoring, millimeter wave radar detection and sound recognition technology, improve the detection accuracy of abnormal events and the accuracy of fault diagnosis, and enhance the visualization effect of environmental monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a structural diagram of an Internet of Things monitoring method for unattended computer room monitoring in Example 1; Figure 2 It is a schematic diagram of the overall architecture of the Internet of Things monitoring method for unattended computer room monitoring in Example 1; Figure 3 It is a schematic diagram of the working principle of the Internet of Things monitoring method for unattended computer room monitoring in Example 1. DETAILED DESCRIPTION

[0018] The present invention is further described in detail below in conjunction with test examples and specific implementation methods. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention.

[0019] Example 1 An Internet of Things monitoring method for unattended computer room monitoring, such as Figure 1 As shown, the following steps are included: S1: Build an IoT monitoring system for unattended computer room monitoring; S2: The system includes an environment monitoring module, an integrated gateway, a radar module, a sound collection module and a computer room camera group. The environment monitoring module includes a temperature sensor, a humidity sensor and a water immersion sensor for monitoring the computer room environment data; the sound collection module is used to collect sound data in the computer room at different time periods and transmit it to the sound analysis module; the computer room camera group includes several cameras for collecting video data in the computer room and monitoring intruders; the radar module is used for multi-target tracking and positioning, and when the movement of objects in the computer room is detected, the computer room camera group recording and video analysis module are started to analyze the video data; the integrated gateway includes a video analysis module and a sound analysis module, which are used to generate alarm signals according to the computer room environment data, sound data and video analysis results.

[0020] The core technical solution adopted by the present invention is as follows: 1. Introduction of integrated gateway: As the hub of the entire monitoring system, it is responsible for integrating and uniformly managing all cameras, millimeter-wave radars, microphones and other equipment in the computer room, ensuring seamless collaboration among all devices, and has analysis functions (including video analysis and sound analysis), while being able to access signals from the computer room environment monitoring.

[0021] 2. Application of millimeter wave radar technology: This invention seamlessly integrates millimeter wave radar technology into the monitoring system. By using the motion detection capability and wide coverage of millimeter wave radar, it can keenly capture abnormal phenomena such as human activities, small animals, and equipment movement in the computer room. Once such an event is detected, the system immediately activates the camera for detailed recording and in-depth video analysis to ensure that no possible safety hazards are missed.

[0022] 3. Integration of sound recognition technology: By integrating advanced sound signal acquisition and processing modules, the present invention realizes real-time monitoring of the operating sound of equipment in the computer room. The system builds a set of efficient sound comparison mechanisms by learning and establishing the normal sound model and fault model of the equipment. This enables the system to analyze the operating status of the equipment in real time, and quickly issue an early warning when a potential fault is detected, and superimpose text on the video image to improve the timeliness and accuracy of fault diagnosis.

[0023] 4. Visual display of environmental monitoring: The integrated gateway of the present invention reserves a sensor signal interface for easy access to the existing computer room environment monitoring system. By superimposing environmental parameters such as temperature, humidity, and water immersion on the video image, the monitoring personnel can clearly understand the real-time environmental conditions of the computer room. This intuitive visual monitoring greatly enhances the effect of environmental monitoring.

[0024] like Figure 2As shown, the present invention is a comprehensive monitoring method based on an integrated gateway, combining video and Internet of Things technologies such as cameras, millimeter-wave radars, and microphones, and having analysis functions, and its analysis functions include video analysis functions and sound analysis functions.

[0025] 1) Overall functionality The present invention is based on an integrated gateway and comprehensively manages equipment such as computer room cameras, millimeter-wave radars, and microphones. It can be used to meet various intelligent monitoring and management needs of the computer room, including personnel entry and exit monitoring, illegal behavior monitoring, fireworks monitoring, intrusion monitoring, equipment operation status monitoring, etc. At the same time, the integrated gateway can be interconnected with the computer room environment monitoring sensor equipment to superimpose real-time environmental parameters such as temperature, humidity, and water immersion on the video monitoring image.

[0026] 2) Millimeter wave radar function: In the present invention, millimeter wave radar technology not only serves as a trigger condition to start video analysis and video storage, but also exhibits significant auxiliary functions in personnel positioning and enhancing video surveillance effects.

[0027] In the scenario of an unmanned computer room, the monitoring screen is usually static and there is no need to implement 24-hour uninterrupted video recording and analysis.

[0028] The present invention uses the millimeter wave radar's sensitive detection capability for moving objects to start the video storage and analysis process only when the object is detected to be moving. This intelligent trigger mechanism effectively saves video storage space, while also optimizing the computing power allocation for video analysis, improving the operating efficiency and economy of the entire monitoring system.

[0029] Furthermore, the application of millimeter-wave radar in personnel positioning solves a major limitation of traditional video surveillance. Due to the limitations of the camera installation position and illumination angle in the machine room, it is sometimes difficult to accurately identify the specific equipment operated by the personnel directly from the video image. However, through millimeter-wave radar positioning technology, the coordinate position of the personnel can be captured in real time and accurately. By matching this information with the coordinates of the equipment in the machine room, it is possible to clearly know the specific equipment operated by the personnel, and then monitor whether the personnel's operating behavior complies with regulations through video images.

[0030] Specifically, the features related to the target are extracted from the millimeter wave signal data received by the radar module. For example, for dynamic targets, the position change, speed, acceleration and other features of the target at different time points can be extracted; for static targets, the shape, size, reflection intensity and other features can be extracted. These features will be used as inputs for the machine learning model.

[0031] The machine learning model is trained using a labeled dataset containing dynamic and static information of known targets and corresponding mmWave signal features. The training process aims to adjust the model’s parameters so that it can accurately predict the target’s state (dynamic or static) and other properties based on the input features.

[0032] As a specific embodiment, dynamic analysis: In the computer room, the movement of people or equipment is a dynamic target. Suppose a technician is inspecting in the computer room, and the data reflected by the millimeter wave signal sent by the radar module is collected. The machine learning model can analyze the technician's movement trajectory, speed changes, etc. based on continuous signal data. For example, when a technician walks between server cabinets, the model accurately determines his walking route and speed by analyzing the changes in signal characteristics. If the speed suddenly decreases or stops, it may mean that the technician is operating in a certain cabinet.

[0033] Static analysis: For static objects such as large server cabinets in the computer room, the machine learning model can analyze whether their shape and position have changed based on the millimeter wave signal. If the cabinet has changed state for some reason (such as being hit or displaced), the model can identify the change in the state of the static object by comparing the current signal characteristics with the initial signal characteristics.

[0034] 3) Video analysis function: In the field of video analysis, the present invention has four core functions: face recognition, behavior analysis, intrusion monitoring, and fireworks monitoring.

[0035] First of all, facial recognition technology plays a vital role in the security management of the computer room. Through high-precision facial recognition algorithms, the system can identify and verify the identity of people entering the computer room in real time. Once an illegal person attempts to enter, the system will immediately issue an alarm and push relevant information to the responsible manager. In addition, facial recognition can also realize the attendance record of patrol personnel to ensure the standardization and traceability of computer room management.

[0036] Secondly, the behavior analysis function further ensures the safety and order inside the computer room.

[0037] This function can accurately identify and warn of illegal behaviors, such as smoking, illegal operation of equipment, mobile devices, etc., to ensure the safety and stability of the computer room environment. Combined with advanced millimeter-wave radar positioning technology, the system can also accurately record special behaviors such as cabinet door opening, providing strong support for subsequent safety assessment and safety tracing.

[0038] In terms of intrusion monitoring, by combining the millimeter-wave radar positioning function and the camera intrusion detection function technology, through the precise demarcation of the restricted area of ​​the computer room, once any unauthorized person enters, the system will immediately trigger an alarm and notify the relevant management personnel, thereby effectively preventing potential security threats.

[0039] Finally, the fire and smoke monitoring function can quickly identify the presence of flames and smoke by analyzing video images in real time. Once signs of fire and smoke are detected, the system will complete the fire detection and alarm process within seconds, greatly shortening the fire alarm time. At the same time, the system can also push on-site images in real time, providing managers with intuitive and accurate fire information, so as to more efficiently command and dispatch firefighting work.

[0040] 4) Sound pickup and comparison function: In the present invention, the application of sound pickup and comparison technology is mainly focused on fault monitoring and early warning of electromechanical systems. This technology plays a pivotal role in the production and operation of urban rail transit and urban (suburban) railway projects. The stability and reliability of electromechanical systems are crucial to ensure passenger safety. However, in real application scenarios, the monitoring of electromechanical systems is often insufficient and usually requires the judgment of professionals, which makes the identification of potential risks particularly difficult.

[0041] In order to solve this problem, the present invention uses sound pickup technology to establish a sound model database of the computer room under normal scenarios, and combines the sound data of various electromechanical equipment under fault or abnormal operation and the on-site alarm sound data. The present invention can perform data processing and pattern recognition through sound comparison technology. This process can not only accurately identify potential faults, but also realize real-time early warning.

[0042] When the system detects an abnormal sound, it will immediately display the corresponding warning information in the video surveillance image to remind relevant managers to conduct further inspection and confirmation in time. This intelligent monitoring method not only improves the efficiency and accuracy of fault identification, but also reduces the dependence on professionals, providing a more reliable guarantee for the stable operation of the electromechanical system.

[0043] like Figure 3 As shown, as a specific embodiment: Logic 1 Sound recognition technology function logic description 1.1 Establish a normal scene voiceprint library, select high-quality recording equipment, and sample the equipment room sound multiple times at different time periods to fully understand the sound characteristics of the equipment under different load conditions. De-noise the recorded sound and convert the processed sound file into a structured vector using a deep learning model (such as x-vector).

[0044] 1.2 The microphone continuously collects sound data in the room and transmits it to the sound analysis module.

[0045] 1.3 De-noise the collected speech signal through IIR+FIR filter.

[0046] 1.4 By regularly collecting device sounds and performing voiceprint comparison, you can monitor the operating status of the device and detect abnormal device operating conditions or alarm sounds in a timely manner.

[0047] 1.5 Link alarm upload and display the alarm in the video image in the form of character overlay.

[0048] Logic 2 Functional logic description of millimeter wave radar technology and video analysis technology 2.1 Millimeter-wave radar is based on the propagation and reflection characteristics of electromagnetic wave signals. It determines the speed and coordinates of an object by sending and receiving millimeter-wave signals, and uses key algorithms such as the stationary object elimination algorithm, multipath interference elimination algorithm, group target tracking algorithm, and human and non-human object classification algorithm to solve the false targets caused by indoor multipath, effectively classify moving and static objects, and achieve multi-target tracking and positioning in indoor environments.

[0049] 2.2 When a moving object is detected, the camera will be linked to record and the video analysis function will be enabled. Video analysis mainly includes the following three aspects: 1) Face recognition analysis: The camera captures the face image and compares it with the face database of the backstage staff to confirm whether the person’s identity is legal and whether he has the authority to operate the equipment. When it is confirmed to be an illegal person, the linkage alarm is uploaded and the alarm is displayed in the video image in the form of character overlay.

[0050] 2) Illegal behavior analysis: Through video analysis, whether there are illegal behaviors of personnel, such as smoking, illegal operation of equipment, mobile devices, etc., the illegal operation of equipment must be combined with the millimeter wave radar positioning function and the camera illegal behavior analysis function to accurately determine whether the equipment is illegally operated. When it is confirmed to be an illegal behavior, the linkage alarm is uploaded and the alarm is displayed in the video image in the form of character superposition.

[0051] 3) Intrusion monitoring: Determine the restricted area of ​​the computer room, combine the millimeter-wave radar positioning function and the camera intrusion detection function to determine whether a person has entered the restricted area, and combine the millimeter-wave radar positioning function and the camera to analyze whether there are small animals (such as mice) invading the computer room. When the intrusion behavior is confirmed, the linkage alarm is uploaded and the alarm is displayed in the video image in the form of character overlay.

[0052] 4) Fireworks monitoring: The firework recognition algorithm is used to analyze the video surveillance images in real time to detect the firework that appears in the image. When the system detects smoke or fire, an alarm is uploaded and displayed in the video image as a character overlay.

[0053] Logic 3 Environmental Monitoring Function Logic Description The temperature sensor, humidity sensor, and water immersion sensor signals in the computer room are connected to this system, and the real-time temperature, humidity, water immersion and other environmental parameters are superimposed on the video surveillance image.

[0054] The significant beneficial effects brought about by the present invention include: 1. Double improvement in comprehensiveness and accuracy: Through the collaborative operation of video surveillance and millimeter-wave radar, our system can efficiently detect and accurately distinguish various abnormal events, significantly improving the comprehensiveness and accuracy of computer room monitoring.

[0055] 2. Enhanced fault diagnosis capabilities: With the help of sound recognition technology, the system can monitor the operating status of equipment in real time, promptly detect and warn of potential faults, thereby effectively reducing losses caused by equipment failures.

[0056] 3. Visual improvement of environmental monitoring: By combining environmental parameters with video images, monitoring personnel can understand the environmental conditions of the computer room more intuitively. This improvement in visual monitoring greatly facilitates the daily work of monitoring personnel.

[0057] 4. Significant improvement in the level of intelligence: By integrating a variety of sensing technologies and analysis algorithms, our system has achieved intelligent monitoring and automatic response to the internal environment of the computer room and the operating status of equipment, significantly improving the intelligence level of the monitoring system.

[0058] In summary, the present invention effectively makes up for the deficiencies of the prior art in computer room monitoring through a series of innovative technical solutions, greatly improves the overall performance and reliability of the monitoring system, and provides a solid guarantee for the safe and stable operation of the computer room.

[0059] Finally, it should be noted that the embodiments described in detail above are only the best practices of the invention and cannot be used to limit the scope of rights of the invention. Equivalent replacement of the technical solutions recorded in the aforementioned embodiments does not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the invention, and they should all be included in the scope of the claims and description of the invention.

Claims

1. An Internet of Things monitoring method for unattended computer room monitoring, characterized in that: The following steps are involved: S1: Build an IoT monitoring system for unattended computer room monitoring; S2: The system includes an environment monitoring module, an integrated gateway, a radar module, a sound collection module and a computer room camera group. The environment monitoring module includes a temperature sensor, a humidity sensor and a water immersion sensor for monitoring the computer room environment data; the sound collection module is used to collect sound data in the computer room at different time periods and transmit it to the sound analysis module; the computer room camera group includes several cameras for collecting video data in the computer room and monitoring intruders; the radar module is used for multi-target tracking and positioning, and when the movement of objects in the computer room is detected, the computer room camera group recording and video analysis module are started to analyze the video data; the integrated gateway includes a video analysis module and a sound analysis module, which are used to generate alarm signals according to the computer room environment data, sound data and video analysis results.

2. The Internet of Things monitoring method for unattended computer room monitoring according to claim 1, characterized in that: The integrated gateway also includes a voiceprint library and a face library. The data in the voiceprint library is used as a basis for determining whether the sound data meets the alarm conditions; the data in the face library is used as a basis for determining whether the video analysis results meet the alarm conditions.

3. The Internet of Things monitoring method for unattended computer room monitoring according to claim 2, characterized in that: The steps of the sound analysis module performing sound recognition and alarm include: denoising the sound data, comparing the denoised sound data with the data in the voiceprint library, and alarming when inconsistency with the normal scene voiceprint library or typical fault voiceprint is found.

4. The Internet of Things monitoring method for unattended computer room monitoring according to claim 3, characterized in that: The denoising process uses IIR and FIR filters, and the sound data after denoising is converted into a structured vector using a deep learning model, and the structured vector is used for voiceprint comparison.

5. The Internet of Things monitoring method for unattended computer room monitoring according to claim 1, characterized in that: The analysis functions of the video analysis module include: Compare the face recognized by the camera with the faces in the face database, and if it is not a face in the face database, issue an alarm; The video data collected by the camera is analyzed, and an alarm is issued if there is any illegal behavior or someone enters the restricted area of ​​the computer room.

6. The Internet of Things monitoring method for unattended computer room monitoring according to claim 5, characterized in that: The analysis function of the video analysis module also includes smoke and fire detection. Smoke and fire detection uses a smoke and fire recognition algorithm to analyze video data in real time, and an alarm is triggered when the system detects smoke or flames.

7. The Internet of Things monitoring method for unattended computer room monitoring according to claim 1, characterized in that: The radar module determines the physical information of an object by sending and receiving millimeter wave signals, and uses machine learning to perform dynamic and static analysis of targets in the environment to achieve multi-target tracking and positioning.

8. The Internet of Things monitoring method for unattended computer room monitoring according to claim 1, characterized in that: The radar module includes several millimeter wave radars.

9. An Internet of Things monitoring method for unattended computer room monitoring according to any one of claims 1 to 8, characterized in that: The alarm signal is displayed in the video image in a character superimposed manner.

Images