High-performance silicon optical interconnection system applied to AI computing power integration of mobile terminal

Abstract

The invention relates to the technical field of mobile computing and optical communication, and discloses a high-performance silicon optical interconnection system applied to mobile terminal AI computing power integration, which comprises three main components, namely a silicon optical interconnection module, a mobile terminal processing terminal and a distributed computing network, all the parts are connected through a high-speed optical network to realize efficient data transmission and cooperative computing functions. High-speed conversion and transmission of electric signals and optical signals are achieved through the silicon optical interconnection module, the bandwidth and the transmission rate are remarkably improved, and the requirements of mobile terminal AI calculation for real-time performance and large-capacity data interaction are met. The integrated environment adaptation unit is combined with real-time analysis of the control chip, transmission parameters are automatically adjusted or early warning is triggered, and stable operation of the system in a complex environment is guaranteed. The miniaturized design of the silicon optical interconnection module adapts to mobile terminal equipment, and the overall power consumption of the system is remarkably reduced by combining the adaptive modulation technology.

Description

Technical Field

[0001] The present invention relates to the field of mobile computing and optical communication technology, and specifically to a high-performance silicon photonic interconnection system for integrating mobile AI computing power. Background Art

[0002] In the field of mobile devices, with the continuous expansion of artificial intelligence (AI) applications, the demand for computing power is increasing. Existing mobile devices usually rely on traditional electronic interconnection systems to achieve data transmission and processing, but these systems have significant deficiencies in performance, power consumption and integration. First, the transmission rate of traditional electronic interconnection systems is limited and cannot meet the needs of high-speed AI computing, resulting in data transmission bottlenecks. Secondly, the power consumption of traditional systems is high, which is not conducive to the long-term operation of mobile devices, especially in high-performance computing scenarios. In addition, the integration of existing interconnection systems is low, resulting in low space utilization, which limits the design flexibility and miniaturization of mobile devices. In view of these problems, it is particularly important to develop a high-performance, low-power, and highly integrated interconnection system. Silicon photonic interconnection systems, with their advantages of high bandwidth, low power consumption and high integration, have become an ideal choice to solve the deficiencies of existing technologies. Therefore, the present invention aims to propose a high-performance silicon photonic interconnection system for mobile AI computing power integration to improve the computing power and overall performance of mobile devices. Summary of the invention

[0003] In view of the shortcomings of the existing technology, the present invention aims to provide a high-performance silicon photonic interconnection system for integrating mobile AI computing power, so as to achieve efficient data transmission and processing capabilities, significantly improve the performance of mobile devices in complex computing tasks, reduce energy consumption, and support flexible expansion and optimized configuration in multiple scenarios.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: A high-performance silicon photonic interconnection system for mobile AI computing power integration, comprising:

[0005] Silicon photonic interconnect module: It includes an optical signal modulation unit, an optoelectronic converter and a control chip. The optical signal modulation unit is used to convert electrical signals into optical signals and transmit them at high speed. The optoelectronic converter is used to receive optical signals and restore them to electrical signals. The control chip is electrically connected to the optical signal modulation unit and the optoelectronic converter to coordinate the modulation, transmission and demodulation processes of the signal and dynamically adjust the transmission parameters according to system requirements.

[0006] Mobile processing terminal: It is equipped with an AI computing power scheduling platform, which has a user interface through which users can set computing power allocation strategies, including task priority division, data flow path planning, and energy consumption management instructions; the mobile processing terminal communicates with external computing resources at high speed through a silicon photonic interconnect module, and is used to offload local tasks to the cloud or edge nodes for collaborative computing;

[0007] Distributed computing network: It is connected to the mobile processing terminal and silicon photonic interconnect module through a high-speed optical network, and is used to store historical data of computing tasks, optimize algorithm models, and perform real-time task scheduling; the distributed computing network also has intelligent analysis capabilities, which can model user usage habits and task characteristics, and provide users with personalized computing power allocation recommendations.

[0008] Preferably, the silicon photonic interconnect module further includes an environmental adaptation unit, which includes a temperature sensor, a vibration sensor and an electromagnetic interference detector. The environmental adaptation unit is electrically connected to the control chip and is used to collect data of the operating environment of the silicon photonic interconnect module in real time and transmit the environmental data to the control chip; the control chip automatically adjusts the transmission power and modulation mode of the optical signal according to the received environmental data and a preset safety threshold; the control chip analyzes the received environmental adaptation unit data and performs an early warning and corresponding operation when specific conditions are met, and the operation is as follows:

[0009] If the detected ambient temperature T exceeds the set maximum temperature threshold t_max, and the vibration intensity V is greater than the set maximum vibration threshold v_max, and the electromagnetic interference intensity E is higher than the set interference threshold e_max, that is, when (T>t_max)∧(V>v_max)∧(E>e_max) is satisfied, the control chip sends a prompt message to the mobile processing terminal through the distributed computing network, informing the user to take cooling or interference shielding measures;

[0010] When the system predicts through environmental adaptation unit data and data analysis that extreme operating conditions are about to occur, including but not limited to high temperature, strong vibration or high electromagnetic interference, the distributed computing network pushes an alarm message to the mobile processing terminal and asks the user whether to start the emergency protection mode. The user can choose whether to allow automatic processing or manual confirmation on the platform; if manual is selected, the user must click to confirm before the platform will send the corresponding control command to the silicon photonic interconnect module to continue the action process.

[0011] Preferably, the AI ​​computing power scheduling platform also has an intelligent scene mode setting function. Users can pre-configure computing power allocation strategies according to different application scenarios, including but not limited to game mode, video rendering mode and low power consumption mode; when the corresponding scene is triggered, the platform automatically sends corresponding control instructions to the silicon photonic interconnect module; the control chip can also automatically adjust the transmission parameters of the optical signal to optimize performance based on the received environmental data and the intelligent scene mode set by the user.

[0012] Preferably, the communication between the mobile processing terminal and the silicon photonic interconnect module adopts a quantum encryption protocol to ensure the security of data transmission; the quantum encryption protocol combines quantum key distribution and classical encryption algorithm to generate a one-time key for encryption before data transmission, and uses the same key for decryption at the receiving end; the distributed computing network has data redundant storage and rapid recovery functions, and regularly performs distributed backup of stored user task data and historical transmission records; when data is lost or damaged, it can be quickly restored through redundant copies to ensure the normal operation of the system.

[0013] Preferably, the silicon photonic interconnect module also includes a voice interaction unit, which is electrically connected to the control chip and is used to receive voice commands from the user and convert the voice commands into control signals for transmission to the control chip; the control chip adjusts the transmission parameters of the optical signal according to the received voice control signal to achieve control of data transmission rate and direction through voice.

[0014] Preferably, the optical signal modulation unit adopts adaptive modulation technology, monitors the quality of the optical signal in real time through a built-in optical feedback loop, and the control chip adjusts the modulation depth and frequency according to the feedback signal quality to ensure the stability and accuracy of data transmission; during the data transmission process, if abnormal signal attenuation is detected, the control chip automatically reduces the transmission rate and sends an alarm message to the user.

[0015] Preferably, the AI ​​computing power scheduling platform has a remote collaboration function, and users can share access rights to computing resources with other users, and the shared users can perform collaborative computing through the platform on their mobile processing terminals; when sharing permissions, the access time range and resource usage limit can be set.

[0016] Preferably, the AI ​​computing power scheduling platform has an interface integrated with IoT devices and can be linked and controlled with other smart devices; for example, when the smart home system detects that the user enters a resting state, it automatically sends instructions to the silicon photonic interconnect module to reduce the data transmission rate to save energy.

[0017] In this invention, in order to achieve efficient computing power scheduling and optimization, a task allocation algorithm based on dynamic weight is proposed, and its formula is as follows:

[0018]

[0019] in, Indicates The comprehensive weight of each task, represents the priority score of the task, represents the energy consumption evaluation value of the task, Indicates the estimated completion time of the task; and are the weight coefficients of priority, energy consumption and time respectively, and satisfy Through this algorithm, the system can dynamically allocate computing resources according to the comprehensive weight of the task, thereby achieving the best balance between performance and energy consumption.

[0020] The present invention provides a high-performance silicon photonic interconnection system for integrating mobile AI computing power. It has the following beneficial effects:

[0021] 1. The present invention realizes high-speed conversion and transmission of electrical and optical signals through silicon photonic interconnect modules, significantly improving bandwidth and transmission rate, and meeting the requirements of mobile AI computing for real-time and large-capacity data interaction. The integrated environmental adaptation unit, combined with real-time analysis of the control chip, automatically adjusts transmission parameters or triggers early warning to ensure stable operation of the system in complex environments.

[0022] 2. The AI ​​computing power scheduling platform of the present invention supports multiple scenario modes, realizes dynamic allocation of computing power resources through dynamic weight task allocation algorithm, and balances performance and energy consumption. It adopts quantum encryption protocol to ensure data transmission security, and combines data redundant storage and rapid recovery mechanism of distributed computing network to prevent data leakage or loss.

[0023] 3. The present invention supports voice control and remote collaboration functions, improving user experience and system flexibility. It provides a linkage interface with smart home devices, such as automatically adjusting the transmission strategy according to the user status to achieve intelligent scene adaptation.

[0024] 4. The miniaturized design of the silicon photonic interconnect module of the present invention is suitable for mobile devices and combined with adaptive modulation technology, it significantly reduces the overall power consumption of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the overall system architecture of the present invention;

[0026] Figure 2 This is a block diagram of the internal structure of the silicon photonic interconnect module in the present invention;

[0027] Figure 3 This is a schematic diagram of the user interface of the AI ​​computing power scheduling platform in the present invention;

[0028] Figure 4 It is a flow chart of the task allocation algorithm based on dynamic weight in the present invention;

[0029] Figure 5 This is a flowchart of the collaboration between the voice interaction unit and the control chip in the present invention;

[0030] Figure 6 Schematic diagram of data redundant storage and rapid recovery mechanism of the distributed computing network in the present invention. DETAILED DESCRIPTION

[0031] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only 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.

[0032] The present invention provides a high-performance silicon optical interconnection system for mobile AI computing power integration. Figure 1 To Attachment Figure 6 The system includes three main components: silicon photonic interconnect module, mobile processing terminal and distributed computing network. Each component is connected through a high-speed optical network to achieve efficient data transmission and collaborative computing functions. The following will describe the specific implementation of the present invention in detail from the aspects of system architecture, module functions, operating principles and practical application scenarios.

[0033] As attached Figure 1 As shown in the figure, the overall system architecture consists of a silicon photonic interconnect module, a mobile processing terminal, and a distributed computing network. As the core data transmission unit, the silicon photonic interconnect module is responsible for converting electrical signals into optical signals and transmitting them at high speed, while receiving optical signals and restoring them to electrical signals. The mobile processing terminal is equipped with an AI computing power scheduling platform, through which users can set task priority divisions, data flow path planning, and energy consumption management instructions, and communicate with external computing resources through the silicon photonic interconnect module. The distributed computing network stores historical data, optimizes algorithm models, and performs real-time task scheduling. It also has intelligent analysis capabilities and can model according to user usage habits and task characteristics, thereby providing personalized computing power allocation recommendations. The direction of data flow between the three is in the attached Figure 1 Clear markings are made in the system to ensure efficient and coordinated operation of the system.

[0034] The internal structure of the silicon photonic interconnect module is shown in the attached figure. Figure 2As shown, it mainly includes an optical signal modulation unit, an optoelectronic converter, a control chip and an environmental adaptation unit. The optical signal modulation unit adopts adaptive modulation technology and monitors the quality of the optical signal in real time through a built-in optical feedback loop. When abnormal signal attenuation is detected, the control chip will automatically reduce the transmission rate and send an alarm message to the user. The optoelectronic converter is used to receive the optical signal and restore it to an electrical signal for subsequent processing. The control chip is electrically connected to the optical signal modulation unit and the optoelectronic converter to coordinate the modulation, transmission and demodulation process of the signal, and dynamically adjust the transmission parameters according to system requirements. The environmental adaptation unit includes a temperature sensor, a vibration sensor and an electromagnetic interference detector, which are used to collect data on the operating environment of the silicon optical interconnect module in real time and transmit the data to the control chip. Based on the received environmental data and combined with the preset safety threshold, the control chip automatically adjusts the transmission power and modulation mode of the optical signal. For example, when it is detected that the ambient temperature T exceeds the set maximum temperature threshold t_max, and the vibration intensity V is greater than the set maximum vibration threshold v_max, and the electromagnetic interference intensity E is higher than the set interference threshold e_max, that is, (T>t_max)∧(V>v_max)∧(E>e_max), the control chip sends a prompt message to the mobile processing terminal through the distributed computing network, informing the user to take cooling or interference shielding measures. In addition, when the system predicts that extreme operating conditions are about to occur, the distributed computing network will push an alarm message to the mobile processing terminal and ask the user whether to start the emergency protection mode. The user can choose automatic processing or manual confirmation on the platform. If manual is selected, the platform will send the corresponding control instruction to the silicon photonic interconnect module after clicking confirm.

[0035] The core of the mobile processing terminal is the AI ​​computing power scheduling platform, and its user interface is as shown in the attached Figure 3As shown. Users can use this interface to set task priority division, data flow path planning, and energy management instructions. The platform also has an intelligent scene mode setting function, and users can pre-configure computing power allocation strategies according to different application scenarios. For example, in game mode, the platform will prioritize high computing power resources to ensure a smooth gaming experience; in video rendering mode, the platform will optimize the data transmission rate to support high-quality video processing; and in low power mode, the platform will reduce the data transmission rate to save energy. When the corresponding scene is triggered, the platform automatically sends the corresponding control instructions to the silicon photonic interconnect module. The control chip can also automatically adjust the transmission parameters of the optical signal to optimize performance based on the received environmental data and the intelligent scene mode set by the user. In addition, the communication between the mobile processing terminal and the silicon photonic interconnect module adopts the quantum encryption protocol to ensure the security of data transmission. The quantum encryption protocol combines quantum key distribution and classical encryption algorithms to generate a one-time key for encryption before data transmission, and use the same key for decryption at the receiving end. This encryption method effectively prevents data from being stolen or tampered with during transmission.

[0036] The functions of distributed computing networks are as follows: Figure 6 As shown in the figure, it not only stores historical data and optimization algorithm models of computing tasks, but also has data redundancy storage and fast recovery functions. The distributed computing network regularly performs distributed backup of stored user task data and historical transmission records. When data is lost or damaged, it can quickly restore data through redundant copies to ensure the normal operation of the system. In addition, the distributed computing network also has intelligent analysis functions, which can model user usage habits and task characteristics and provide users with personalized computing power allocation recommendations. For example, when users frequently perform video rendering tasks, the distributed computing network will automatically optimize the relevant algorithm models to improve the efficiency of task completion.

[0037] In order to achieve efficient computing power scheduling and optimization, the present invention proposes a task allocation algorithm based on dynamic weights, the formula of which is as follows:

[0038] .

[0039] in, Indicates The comprehensive weight of each task, represents the priority score of the task, represents the energy consumption evaluation value of the task, Indicates the estimated completion time of the task; and are the weight coefficients of priority, energy consumption and time respectively, and satisfy Through this algorithm, the system can dynamically allocate computing resources according to the comprehensive weight of the task, thereby achieving the best balance between performance and energy consumption. For example, when a user needs to process multiple tasks at the same time, the system will calculate the comprehensive weight of each task and prioritize resources to tasks with higher weights, thereby ensuring the smooth completion of key tasks.

[0040] The silicon photonic interconnect module also includes a voice interaction unit, which is electrically connected to the control chip and is used to receive the user's voice commands and convert the voice commands into control signals for transmission to the control chip. The control chip adjusts the transmission parameters of the optical signal according to the received voice control signal to achieve voice control of the data transmission rate and direction. The collaboration process between the voice interaction unit and the control chip is shown in the attached figure. Figure 5 , which shows the specific steps of receiving, converting and adjusting transmission parameters of voice commands. For example, a user can increase the transmission speed of an optical signal by using the voice command "increase transmission rate", or reduce the transmission power of an optical signal by using the voice command "reduce energy consumption".

[0041] The AI ​​computing power scheduling platform also has a remote collaboration function, and users can share access rights to computing resources with other users. The shared users can perform collaborative computing through the platform on their mobile processing terminals. When sharing permissions, users can set the access time range and resource usage limit. For example, user A can share access rights to computing resources with user B, and set the access time to 2 hours and the resource usage limit to 50%. In addition, the AI ​​computing power scheduling platform also has an interface for integration with IoT devices, which can be linked and controlled with other smart devices. For example, when the smart home system detects that the user is in a resting state, it automatically sends instructions to the silicon photonic interconnect module to reduce the data transmission rate to save energy.

[0042] In summary, the present invention realizes efficient data transmission and processing capabilities through the collaborative work of silicon photonic interconnect modules, mobile processing terminals and distributed computing networks, significantly improves the performance of mobile devices in complex computing tasks, reduces energy consumption, and supports flexible expansion and optimized configuration in multiple scenarios. The operating principle and actual application scenarios of the system have been fully verified, and it has broad application prospects and technical value.

[0043] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A high-performance silicon photonic interconnect system for mobile AI computing power integration, characterized in that: include: Silicon photonic interconnect module: It includes an optical signal modulation unit, an optoelectronic converter and a control chip. The optical signal modulation unit is used to convert electrical signals into optical signals and transmit them at high speed. The optoelectronic converter is used to receive optical signals and restore them to electrical signals. The control chip is electrically connected to the optical signal modulation unit and the optoelectronic converter to coordinate the modulation, transmission and demodulation processes of the signal and dynamically adjust the transmission parameters according to system requirements. Mobile processing terminal: It is equipped with an AI computing power scheduling platform, which has a user interface through which users can set computing power allocation strategies, including task priority division, data flow path planning, and energy consumption management instructions; the mobile processing terminal communicates with external computing resources at high speed through a silicon photonic interconnect module, and is used to offload local tasks to the cloud or edge nodes for collaborative computing; Distributed computing network: It is connected to the mobile processing terminal and silicon photonic interconnect module through a high-speed optical network, and is used to store historical data of computing tasks, optimize algorithm models, and perform real-time task scheduling; the distributed computing network also has intelligent analysis capabilities, which can model user usage habits and task characteristics, and provide users with personalized computing power allocation recommendations.

2. According to claim 1, a high-performance silicon photonic interconnection system for integrating mobile AI computing power is characterized in that: The silicon photonic interconnect module further includes an environmental adaptation unit, which includes a temperature sensor, a vibration sensor, and an electromagnetic interference detector. The environmental adaptation unit is electrically connected to the control chip and is used to collect data of the operating environment of the silicon photonic interconnect module in real time and transmit the environmental data to the control chip; the control chip automatically adjusts the transmission power and modulation mode of the optical signal according to the received environmental data and a preset safety threshold; the control chip analyzes the received environmental adaptation unit data and performs an early warning and corresponding operation when specific conditions are met, and the operation is as follows: If the detected ambient temperature T exceeds the set maximum temperature threshold t_max, and the vibration intensity V is greater than the set maximum vibration threshold v_max, and the electromagnetic interference intensity E is higher than the set interference threshold e_max, that is, when (T>t_max)∧(V>v_max)∧(E>e_max) is satisfied, the control chip sends a prompt message to the mobile processing terminal through the distributed computing network, informing the user to take cooling or interference shielding measures; When the system predicts through environmental adaptation unit data and data analysis that extreme operating conditions are about to occur, including but not limited to high temperature, strong vibration or high electromagnetic interference, the distributed computing network pushes an alarm message to the mobile processing terminal and asks the user whether to start the emergency protection mode. The user can choose whether to allow automatic processing or manual confirmation on the platform; if manual is selected, the user must click to confirm before the platform will send the corresponding control command to the silicon photonic interconnect module to continue the action process.

3. According to claim 1, a high-performance silicon photonic interconnection system for mobile AI computing power integration is characterized in that: The AI ​​computing power scheduling platform also has an intelligent scene mode setting function. Users can pre-configure computing power allocation strategies according to different application scenarios, including but not limited to game mode, video rendering mode and low power consumption mode. When the corresponding scene is triggered, the platform automatically sends corresponding control instructions to the silicon photonic interconnect module. The control chip can also automatically adjust the transmission parameters of the optical signal to optimize performance based on the received environmental data and the intelligent scene mode set by the user.

4. According to claim 1, a high-performance silicon photonic interconnection system for mobile AI computing power integration is characterized in that: The communication between the mobile processing terminal and the silicon photonic interconnect module adopts a quantum encryption protocol to ensure the security of data transmission; the quantum encryption protocol combines quantum key distribution and classical encryption algorithm to generate a one-time key for encryption before data transmission, and uses the same key for decryption at the receiving end; the distributed computing network has data redundant storage and rapid recovery functions, and regularly performs distributed backup of stored user task data and historical transmission records; when data is lost or damaged, it can be quickly restored through redundant copies to ensure the normal operation of the system.

5. According to claim 1, a high-performance silicon photonic interconnection system for mobile AI computing power integration is characterized in that: The silicon photonic interconnect module also includes a voice interaction unit, which is electrically connected to the control chip and is used to receive user voice commands and convert the voice commands into control signals for transmission to the control chip; the control chip adjusts the transmission parameters of the optical signal according to the received voice control signal to achieve data transmission rate and direction control through voice.

6. A high-performance silicon photonic interconnection system for mobile AI computing power integration according to claim 1, characterized in that: The optical signal modulation unit adopts adaptive modulation technology and monitors the quality of the optical signal in real time through a built-in optical feedback loop. The control chip adjusts the modulation depth and frequency according to the feedback signal quality to ensure the stability and accuracy of data transmission. During the data transmission process, if abnormal signal attenuation is detected, the control chip automatically reduces the transmission rate and sends an alarm message to the user.

7. A high-performance silicon photonic interconnection system for mobile AI computing power integration according to claim 1, characterized in that: The AI ​​computing power scheduling platform has a remote collaboration function. Users can share access rights to computing resources with other users, and the shared users can perform collaborative computing through the platform on their mobile processing terminals. When sharing permissions, the access time range and resource usage limit can be set.

8. The high-performance silicon photonic interconnection system for mobile AI computing power integration according to claim 1, characterized in that: The AI ​​computing power scheduling platform has an interface integrated with IoT devices and can be linked and controlled with other smart devices. When the smart home system detects that the user is in a resting state, it automatically sends instructions to the silicon photonic interconnect module to reduce the data transmission rate to save energy.

Images