A mainboard circuit, an artificial intelligence component and an electronic device
By introducing detachable artificial intelligence components into the motherboard circuit and utilizing the main control processor to detect level changes, the problem of inflexible deployment of artificial intelligence components is solved, achieving greater flexibility and selectivity, and improving the intelligence and operability of electronic devices.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- HANGZHOU HIKSTORAGE TECHNOLOGY CO LTD
- Filing Date
- 2025-05-22
- Publication Date
- 2026-06-19
AI Technical Summary
The existing artificial intelligence components are integrated with the motherboard circuitry in a non-removable manner, resulting in a lack of flexibility and choice in deployment.
By introducing detachable artificial intelligence components into the motherboard circuitry, and utilizing the expansion interface pins of the main control processor to detect level changes to determine the electrical connection status, the artificial intelligence components can be flexibly deployed and upgraded.
It improves the deployment flexibility and selectivity of artificial intelligence components, enhances the intelligence and operability of electronic devices, and supports flexible function upgrades and expansions.
Smart Images
Figure CN224383659U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the field of artificial intelligence, and in particular, to a motherboard circuit, an artificial intelligence component, and an electronic device. Background Technology
[0002] With the development of artificial intelligence technology, more and more AI models are being deployed in electronic devices. Existing AI components used to implement AI functions are deployed on motherboard circuits that include at least a main control unit, and are usually integrated with the motherboard circuitry as a non-removable unit. This results in a lack of flexibility and selectivity in the deployment of AI models. Summary of the Invention
[0003] This invention provides a motherboard circuit, an artificial intelligence component, and an electronic device, enabling selective deployment of the artificial intelligence component.
[0004] The first aspect of this utility model provides a motherboard circuit, the motherboard circuit comprising:
[0005] The main control processor is used to control system functions.
[0006] in,
[0007] The main control processor detects the current electrical connection status of the detachable AI component used to implement AI functions based on the level changes of its expansion interface pins.
[0008] Electrical connection status includes: online status where the AI component and the main control processor's expansion interface pins are electrically connected, and offline status where the AI component and the main control processor's expansion interface pins are not electrically connected.
[0009] As one possible implementation, the motherboard circuitry further includes any of the following circuits located on the motherboard:
[0010] A speech processing circuit is used for speech recognition and speech synthesis. This speech processing circuit is electrically connected to the peripheral interface pins of the main control processor.
[0011] A storage circuit for data storage, which is electrically connected to the interface pins of the main control processor for connecting the processor and the storage circuit.
[0012] A communication circuit for network access, which is electrically connected to the data communication interface pins of the main control processor.
[0013] A first power management circuit is used to manage the power supply of the motherboard, and this communication circuit is electrically connected to the power interface pins of the main control processor.
[0014] As one possible implementation, the main control processor, upon detecting that the artificial intelligence component is offline, connects to the server storing the artificial intelligence model via a communication circuit.
[0015] The speech processing circuit includes: a sound pickup for picking up sound signals, a speech processing unit for processing the sound information from the sound pickup, and a speaker for outputting sound signals. The speech processing unit preprocesses the speech signals collected by the sound pickup, converts the preprocessed speech signals into text, analyzes the converted text and generates natural language or generates control commands, synthesizes the processing results corresponding to the generated natural language or control commands into speech and outputs it through the speaker.
[0016] As one possible implementation, the peripheral interface is a Serial Peripheral Interface (SPI).
[0017] As one possible implementation, the interface for connecting the processor and storage circuitry is a Serial High Technology Configuration Interface (SATA).
[0018] As one possible implementation, the data communication interface is a simplified gigabit media independent interface (RGMII).
[0019] As one possible implementation, the artificial intelligence component includes at least:
[0020] As one possible implementation, one or more AI processors for performing AI task calculations are provided, wherein the expansion interface pins of the AI processors are electrically connected to the expansion interface pins of the main control processor when the AI components are online.
[0021] As one possible implementation, the artificial intelligence component further includes:
[0022] Memory used to store data from the artificial intelligence processor, and / or,
[0023] Heat dissipation devices for heat dissipation, and / or,
[0024] A second power management circuit for providing operating power to artificial intelligence components;
[0025] The expansion interface is a high-speed serial computer expansion bus interface (PCIE).
[0026] A second aspect of this application provides an artificial intelligence component that is detachably electrically connected to the expansion interface pins of the main control processor in the motherboard circuit, which is used to control system functions.
[0027] in,
[0028] The main control processor detects the current electrical connection status of the artificial intelligence component based on the level changes of its expansion interface pins.
[0029] Electrical connection status includes: online status where the AI component and the main control processor's expansion interface pins are electrically connected, and offline status where the AI component and the main control processor's expansion interface pins are not electrically connected.
[0030] As one possible implementation, the artificial intelligence component further includes at least:
[0031] One or more AI processors for performing AI task calculations, wherein the AI processors are electrically connected to the expansion interface pins of the main control processor when the AI components are online.
[0032] As one possible implementation, the artificial intelligence component further includes:
[0033] Heat dissipation devices for heat dissipation, and / or
[0034] A second power management circuit for providing operating power to artificial intelligence components, and / or
[0035] Memory used to store data from the artificial intelligence processor;
[0036] The expansion interface is a high-speed serial computer expansion bus interface (PCIE).
[0037] A third aspect of this application provides an electronic device including any of the described motherboard circuits and / or any of the described artificial intelligence components.
[0038] As one possible implementation, the electronic device is a network-attached storage device (NAS).
[0039] The motherboard circuit provided by this utility model connects the main control processor's expansion interface pins to a detachable artificial intelligence component via electrical connection. It detects the current electrical connection status of the artificial intelligence component based on the level changes of the main control processor's expansion interface pins, enabling the artificial intelligence component to be flexibly deployed locally and facilitating its upgrade or update. Attached Figure Description
[0040] Figure 1 This is a schematic diagram of the electrical connection between the motherboard circuit and the artificial intelligence component in an embodiment of this application.
[0041] Figure 2 This is a schematic diagram of the NAS device motherboard circuit in this embodiment.
[0042] Figure 3 This is a schematic diagram of the voice processing circuit in this embodiment.
[0043] Figure 4 This is a schematic diagram of an artificial intelligence component in this embodiment.
[0044] Figure 5 This is a schematic diagram illustrating the voltage level changes of the PCIe interface pins in this embodiment. Detailed Implementation
[0045] To make the objectives, technical means, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings.
[0046] To enable the AI component to be flexibly deployed in the motherboard circuit, this application embodiment connects the AI component to the motherboard circuit in a detachable manner.
[0047] See Figure 1 As shown, Figure 1 This is a schematic diagram illustrating the electrical connection between a motherboard circuit and an artificial intelligence component according to an embodiment of this application. The expansion interface pins of the main control processor in the motherboard circuit, which controls system functions, are electrically connected to the detachable artificial intelligence component.
[0048] The main control processor detects the current electrical connection status of the detachable AI component used to implement AI functions based on the level changes of its expansion interface pins.
[0049] The electrical connection status includes: an online state where the AI component is electrically connected to the expansion interface pins of the main control processor, and an offline state where the AI component is not electrically connected to the expansion interface pins of the main control processor. That is, when the AI component is electrically connected to the expansion interface pins of the main control processor, the AI component is in an unremoved state relative to the motherboard circuit and is deployed in the motherboard circuit. When the AI component is not electrically connected to the expansion interface pins of the main control processor, the AI component is in a detached state relative to the motherboard circuit, so that the AI component is not deployed in the motherboard circuit.
[0050] As an example, the detachable method can be a direct plug-and-play method, and the detachable AI components can be located in the motherboard circuitry or outside the motherboard circuitry.
[0051] As an example, the main control processor is a digital signal processor (DSP), and the artificial intelligence component is a hardware module for implementing artificial intelligence functions, including:
[0052] An AI processor for performing AI task processing, wherein the expansion interface pins of the AI processor are electrically connected to the expansion interface pins of the main control processor when the AI component is online, and
[0053] Memory used to store data from the artificial intelligence processor.
[0054] The artificial intelligence processor described in this embodiment can be any one of a graphics processing unit (GPU), a tensor processing unit (TPU), a neural network processing unit (NPU), a deep learning processing unit (DPU), an accelerated processing unit (APU), and a general-purpose computing on graphics processing unit (GPGPU), and this application does not limit it.
[0055] As an example, the artificial intelligence component also includes: a heat dissipation device for heat dissipation, and / or a second power management circuit for providing operating power to the artificial intelligence component.
[0056] As an example, the motherboard circuitry also includes any of the following circuits located on the motherboard:
[0057] A speech processing circuit is used for speech recognition and speech synthesis. This speech processing circuit is electrically connected to the peripheral interface pins of the main control processor.
[0058] A storage circuit for data storage, which is electrically connected to the interface pins of the main control processor for connecting the processor and the storage circuit.
[0059] A communication circuit for network access, which is electrically connected to the data communication interface pins of the main control processor.
[0060] A first power management circuit is used to manage the power supply of the motherboard, and this communication circuit is electrically connected to the power interface pins of the main control processor.
[0061] As an example, when the main control processor detects that the artificial intelligence component is offline, it connects to a server or cloud that stores the artificial intelligence model via a communication circuit.
[0062] The peripheral interface is a serial peripheral interface (SPI).
[0063] The interface used to connect the processor and storage circuitry is a Serial Advanced Technology Attachment (SATA) interface.
[0064] The data communication interface is a Reduced Gigabit Media Independent Interface (RGMII).
[0065] The expansion interface is a high-speed serial computer expansion bus interface (PCIE, peripheral component interconnect express).
[0066] This application embodiment electrically connects a detachable artificial intelligence component to the expansion interface of the main control processor located in the motherboard circuit. The main control processor detects the current electrical connection status of the artificial intelligence component based on the level change of the expansion interface pin of the main control processor, thereby enabling the artificial intelligence component to be selectively deployed, improving the flexibility and selectivity of the deployment of the artificial intelligence component.
[0067] To facilitate understanding of the embodiments of this application, the following description uses a network attached storage device (NAS) as an example. It should be understood that the embodiments of this application are not limited to NAS, and any other electronic device may also be applied.
[0068] The main function of current NAS devices is to transmit and store private data over a network. Operation mainly relies on online mobile applications, while offline operation is achieved through traditional physical buttons or touch screens. The device functions are relatively limited and can only be used as a storage device.
[0069] To enrich the functionality of the NAS device, this embodiment adds a detachable intelligent component deployment function and a voice interaction function to the NAS device.
[0070] See Figure 2 As shown, Figure 2 This is a schematic diagram of the motherboard circuitry of the NAS device in this embodiment. The device includes: a motherboard circuitry, which includes:
[0071] The main control processor is used to control various functional logics, including power management and task scheduling.
[0072] Storage circuit unit, used for data storage, supports multiple file formats;
[0073] A communication circuit unit is used to access a network and establish communication links with servers, terminals, etc.
[0074] The first power management circuit unit is used to provide operating power to the main control processor unit, storage circuit unit, and communication circuit unit to ensure the operation of the NAS device.
[0075] To enable voice interaction, the main control circuit also includes a voice processing circuit unit. This unit integrates voice recognition and synthesis functions, allowing users to interact with and control the NAS device via voice.
[0076] As an example, see Figure 3 As shown, Figure 3 This is a schematic diagram of the voice processing circuit in this embodiment. The voice processing circuit unit includes: a sound pickup for picking up sound signals, a voice processing unit for processing the sound information from the sound pickup, and a speaker for outputting sound signals. For example, the sound pickup, such as a microphone, picks up the user's voice input in real time. The voice processing unit performs preprocessing on the picked-up voice signal, including noise reduction and echo removal, converts the preprocessed voice signal into text, analyzes the converted text, generates natural language or controls the NAS device, synthesizes the generated natural language or the control result of the NAS device into speech, and outputs it through the speaker.
[0077] in,
[0078] The SPI interface pin of the main control processor is electrically connected to the voice processing circuit unit, the RGMII interface pin is electrically connected to the communication circuit unit, and the SATA interface pin is electrically connected to the storage circuit unit.
[0079] All of the aforementioned units, as well as the main control processor, may be located entirely or partially within the motherboard, depending on the size of the motherboard. This application does not impose any restrictions on this.
[0080] As an example, the main control processor can use the RTD1296 chip, the communication circuit unit can use the RTL8152B chip, and the voice processing unit can use the WT3000A chip.
[0081] To allow for the selective deployment of AI components, the main controller's PCIe interface pins are electrically connected to the detachable AI components.
[0082] As an example, see Figure 4 As shown, Figure 4This is a schematic diagram of an artificial intelligence (AI) component. The AI component employs a modular design with detachable structures, such as pin and / or slot structures, for easy replacement or upgrades as needed. The AI component includes one or more AI processors, for example, equipped with a high-performance GPU processor to ensure rapid response and processing power, while enabling local deployment of AI intelligent models for high intelligence. The AI component may also include memory, such as LPDDR4 chips, to support complex AI algorithms and large amounts of data storage. The AI component may also include a second power management circuit, which employs a low-power design to ensure stable power supply to the AI component during long-term operation, reducing device power consumption and heat generation. The PCIe interface pins of each AI processor are electrically connected to the PCIe interface pins of the main control processor.
[0083] All circuits in the artificial intelligence component may be located entirely on the motherboard, or some may be located on the motherboard; this application does not impose any restrictions on this.
[0084] The main control processor performs GPIO (General Purpose Input / Output) detection on the PCIe interface pins. When the AI component is not electrically connected, the main control processor's IO interface is in a pull-down state. When the AI component is electrically connected, the voltage of the GPIO port is pulled up through the pull-up resistor on the PCIe interface pin of the AI processor in the AI component, thereby changing the level of the main control processor's IO interface. In this way, the main control processor can determine whether the AI component is electrically connected.
[0085] See Figure 5 As shown, Figure 5 This diagram illustrates a change in the voltage level of a PCIe interface pin. The left image shows a GPIO circuit for the main controller's PCIe interface pin, and the right image shows a GPIO circuit for the AI processor's PCIe interface pin. When the AI processor is electrically connected to the main controller's PCIe interface pin, the GPIO 1 level of the main controller's PCIe interface pin 1 changes from low to high.
[0086] As an example, the voice processing circuit unit in the main control circuit receives voice commands from the outside, processes them, and outputs them as digital signals to the main control processor. The main control processor analyzes the input digital signals. When it is necessary to call the artificial intelligence component for processing, it detects whether there is a level change on the PCIe interface pin of the main control processor, for example, from low level to high level. When there is a change from low level to high level on the PCIe interface pin, it is determined that the artificial intelligence component is online, that is, the artificial intelligence component is electrically connected to the main control processor, and the artificial intelligence component is called to analyze the digital signals. When there is no change from low level to high level on the PCIe interface pin, it is determined that the artificial intelligence component is offline, that is, the artificial intelligence component is not electrically connected to the main control processor, and the main control processor connects to the network through the communication circuit to establish a communication link with the server or cloud, and calls the artificial intelligence model on the server side or cloud side to analyze the digital signals.
[0087] To facilitate the deployment of detachable AI components, corresponding space is reserved in the NAS device. When the AI components are not installed, they can be hidden by the outer cover. When installation is required, the cover can be opened, the AI components can be inserted, and electrical connection with the NAS device motherboard can be achieved through the PCIe high-speed interface. The AI components can be fixed to the NAS device with fasteners such as clips or screws. Closing the cover completes the installation of the AI components.
[0088] The NAS device in this embodiment achieves voice interaction functionality through a voice processing circuit unit, eliminating the need for manual operation. Control of the NAS device can be achieved within a reasonable distance, overcoming spatial limitations. The deployment of detachable artificial intelligence components enables AI functionality, increasing the device's selectivity and intelligence. This fundamentally changes the NAS device's positioning and attributes, transforming it from a mere personal storage product into a personal management tool. It offers voice interaction, intelligent features, flexible configuration options to meet different user needs, and support for future upgrades. Furthermore, it enhances operability and user experience, increasing the frequency of operation and significantly boosting customer loyalty.
[0089] For the device / network-side equipment / storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and relevant parts can be referred to in the description of the method embodiments.
[0090] In this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, without necessarily requiring or implying any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0091] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A motherboard circuit, characterized in that, The motherboard circuitry includes: The main control processor is used to control system functions. in, The main control processor detects the current electrical connection status of the detachable AI component used to implement AI functions based on the level changes of its expansion interface pins. Electrical connection status includes: online status where the AI component and the main control processor's expansion interface pins are electrically connected, and offline status where the AI component and the main control processor's expansion interface pins are not electrically connected.
2. The motherboard circuit as described in claim 1, characterized in that, The motherboard circuit also includes any of the following circuits located on the motherboard: A speech processing circuit is used for speech recognition and speech synthesis. This speech processing circuit is electrically connected to the peripheral interface pins of the main control processor. A storage circuit for data storage, which is electrically connected to the interface pins of the main control processor for connecting the processor and the storage circuit. A communication circuit for network access, which is electrically connected to the data communication interface pins of the main control processor. A first power management circuit is used to manage the power supply of the motherboard, and this communication circuit is electrically connected to the power interface pins of the main control processor.
3. The motherboard circuit as described in claim 2, characterized in that, When the main control processor detects that the artificial intelligence component is offline, it connects to the server storing the artificial intelligence model through the communication circuit. The speech processing circuit includes: a sound pickup for picking up sound signals, a speech processing unit for processing the sound information from the sound pickup, and a speaker for outputting sound signals. The speech processing unit preprocesses the speech signals collected by the sound pickup, converts the preprocessed speech signals into text, analyzes the converted text and generates natural language or generates control commands, synthesizes the processing results corresponding to the generated natural language or control commands into speech and outputs it through the speaker. The peripheral interface is a serial peripheral interface (SPI). The interface used to connect the processor and storage circuitry is a Serial High Technology Configuration Interface (SATA). The data communication interface is a simplified gigabit media independent interface (RGMII).
4. The motherboard circuit as described in any one of claims 1 to 3, characterized in that, The artificial intelligence components include at least: One or more AI processors for performing AI task calculations, wherein the expansion interface pins of the AI processors are electrically connected to the expansion interface pins of the main control processor when the AI components are online.
5. The motherboard circuit as described in claim 4, characterized in that, The artificial intelligence components also include: Memory used to store data from the artificial intelligence processor, and / or, Heat dissipation devices for heat dissipation, and / or, A second power management circuit for providing operating power to artificial intelligence components; The expansion interface is a high-speed serial computer expansion bus interface (PCIE).
6. An artificial intelligence component, characterized in that, The artificial intelligence component is detachably electrically connected to the expansion interface pins of the main control processor on the motherboard circuit, which is used to control system functions. in, The main control processor detects the current electrical connection status of the artificial intelligence component based on the level changes of its expansion interface pins. Electrical connection status includes: online status where the AI component and the main control processor's expansion interface pins are electrically connected, and offline status where the AI component and the main control processor's expansion interface pins are not electrically connected.
7. The artificial intelligence component as described in claim 6, characterized in that, The artificial intelligence components also include at least: One or more AI processors for performing AI task calculations, wherein the AI processors are electrically connected to the expansion interface pins of the main control processor when the AI components are online.
8. The artificial intelligence component as described in claim 7, characterized in that, The artificial intelligence components also include: Heat dissipation devices for heat dissipation, and / or A second power management circuit for providing operating power to artificial intelligence components, and / or Memory used to store data from the artificial intelligence processor; The expansion interface is a high-speed serial computer expansion bus interface (PCIE).
9. An electronic device, characterized in that, Includes the motherboard circuitry as described in any one of claims 1 to 5 and / or the artificial intelligence components as described in any one of claims 6 to 8.
10. The electronic device as claimed in claim 9, characterized in that, The electronic device is a network-attached storage device (NAS).