CN117421019A - Model upgrading method and system for edge-end artificial intelligent chip - Google Patents

Model upgrading method and system for edge-end artificial intelligent chip Download PDF

Info

Publication number
CN117421019A
CN117421019A CN202311338529.XA CN202311338529A CN117421019A CN 117421019 A CN117421019 A CN 117421019A CN 202311338529 A CN202311338529 A CN 202311338529A CN 117421019 A CN117421019 A CN 117421019A
Authority
CN
China
Prior art keywords
edge
equipment
data
model
upgrade
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202311338529.XA
Other languages
Chinese (zh)
Inventor
王景
姜凯
赵鑫鑫
李锐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Inspur Science Research Institute Co Ltd
Original Assignee
Shandong Inspur Science Research Institute Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shandong Inspur Science Research Institute Co Ltd filed Critical Shandong Inspur Science Research Institute Co Ltd
Priority to CN202311338529.XA priority Critical patent/CN117421019A/en
Publication of CN117421019A publication Critical patent/CN117421019A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F8/00—Arrangements for software engineering
    • G06F8/60—Software deployment
    • G06F8/65—Updates
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/602—Providing cryptographic facilities or services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/03—Protecting confidentiality, e.g. by encryption
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/04—Key management, e.g. using generic bootstrapping architecture [GBA]
    • H04W12/041—Key generation or derivation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/10—Integrity
    • H04W12/106—Packet or message integrity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02—Power saving arrangements
    • H04W52/0209—Power saving arrangements in terminal devices
    • H04W52/0225—Power saving arrangements in terminal devices using monitoring of external events, e.g. the presence of a signal
    • H04W52/0248—Power saving arrangements in terminal devices using monitoring of external events, e.g. the presence of a signal dependent on the time of the day, e.g. according to expected transmission activity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02—Power saving arrangements
    • H04W52/0209—Power saving arrangements in terminal devices
    • H04W52/0261—Power saving arrangements in terminal devices managing power supply demand, e.g. depending on battery level
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W8/00—Network data management
    • H04W8/22—Processing or transfer of terminal data, e.g. status or physical capabilities
    • H04W8/24—Transfer of terminal data
    • H04W8/245—Transfer of terminal data from a network towards a terminal

Abstract

The invention discloses a model upgrading method and a system of an edge-end artificial intelligent chip, which belong to the fields of artificial intelligence, chip design and edge calculation, and the invention aims to solve the technical problem of how to conveniently and rapidly realize the model upgrading of the edge-end artificial intelligent chip while ensuring the data security in UWB wireless communication, and adopts the following technical scheme: model data preparation: the special upgrade equipment obtains the model data of the new version through the USB interface and stores the model data into the internal FLASH storage; activating the edge device: placing special upgrading equipment in an area where the edge equipment is located, selecting model data to be upgraded through a display screen, clicking to start an upgrading task, and activating a UWB wireless communication module of the edge equipment into a normal working state; generating key data D; packaging the data encryption machine; data transmission; data receiving and checking; decrypting the data; and upgrading the model.

Description

Model upgrading method and system for edge-end artificial intelligent chip
Technical Field
The invention relates to the fields of artificial intelligence, chip design and edge computing, in particular to a model upgrading method and system of an edge-end artificial intelligence chip.
Background
In recent years, along with the acceleration fusion of artificial intelligence and the Internet of things, the demand of edge calculation is increasingly large; the deep learning algorithm represented by the convolutional neural network is deployed on an edge processor more and more, but the model updating speed of the existing artificial intelligent chip is high, model parameters need to be updated frequently along with the iterative change of model training, and for the equipment at the edge, the equipment is limited by the installation position, the product volume, the power consumption and the requirements of three proofings, and many of the equipment do not have the condition of updating the model parameters in a wired networking mode, and are updated in a wireless network communication mode, if the equipment is used for data transmission in a plaintext, certain data safety hidden hazards exist. Meanwhile, if the model data is encrypted by encryption algorithms such as DES, AES and the like, the problem of safe transmission of the secret key must be considered.
Therefore, how to conveniently and rapidly realize the model upgrade of the edge artificial intelligent chip while ensuring the data security in UWB wireless communication is a technical problem to be solved at present.
Disclosure of Invention
The technical task of the invention is to provide a model upgrading method and a system for an edge-side artificial intelligent chip, which are used for solving the problem of how to conveniently and rapidly realize the model upgrading of the edge-side artificial intelligent chip while ensuring the data security in UWB wireless communication.
The technical task of the invention is realized in the following way, and the method for upgrading the model of the edge-end artificial intelligent chip is as follows:
model data preparation: the special upgrade equipment obtains the model data of the new version through the USB interface and stores the model data into the internal FLASH storage;
activating the edge device: placing special upgrading equipment in an area where the edge equipment is located, selecting model data to be upgraded through a display screen, clicking to start an upgrading task, and activating a UWB wireless communication module of the edge equipment into a normal working state;
generating key data D: after the UWB wireless communication module of the edge equipment is activated and converted into a normal working state, the UWB wireless communication module of the edge equipment initiates positioning communication with the special upgrading equipment, and the two pieces of equipment calculate the accurate position distance D between the edge equipment and the special upgrading equipment through a DS-TWR algorithm;
and (3) packaging a data encryption machine: the special upgrade equipment encrypts the model data by taking the calculated data D as an encryption key of an AES algorithm; after encryption, the data are packaged and checked according to a agreed communication protocol;
and (3) data transmission: the special upgrade device sends the encrypted model data to the edge terminal device through the special upgrade UWB wireless communication module;
data receiving and checking: the edge terminal equipment receives the analysis model data according to the agreed communication protocol, and checks the model data according to the agreed check rule so as to ensure the integrity of the data packet;
decrypting data: if the data packet is checked to be correct, decrypting the model data packet through the key data D;
upgrading model: updating the obtained model data to an access area corresponding to the FLASH, and reconfiguring the acceleration unit by using the new model data, and after the model upgrading of the chip is completed, setting the edge-side UWB wireless communication module into a dormant mode by the processor again, and completing the upgrading.
Preferably, when the edge terminal equipment is activated, after an upgrade task is started, the special upgrade equipment calls the special upgrade UWB wireless communication module and sends an upgrade signal in a broadcasting mode; the edge device sniffs whether an upgrade signal exists or not in the wake-up period of the self edge UWB device; and once the edge equipment discovers the upgrade signal, the edge UWB wireless communication module is activated, and the operation mode is changed into a normal operation mode.
More preferably, the accurate position distance D between the edge terminal equipment and the special upgrade equipment has uniqueness for the two equipment, meanwhile, the data D is obtained through solving by a DS-TWR algorithm and cannot be captured by other equipment, and then the data D is used as an encryption key of an AES algorithm.
More preferably, the special upgrade device comprises a processor, a special upgrade UWB wireless communication module, a Flash chip, a display screen, a USB OTG interface and a lithium battery power supply module, wherein the special upgrade device is connected with a server through the USB OTG interface in the form of a storage device, after the server trains to obtain new model parameter data, the data is written into a memory corresponding to the special upgrade device or the new model parameter data obtained by the server trains is copied into the mobile device of the U disk for storage; and accessing the USB Flash disk to a USB interface of the special upgrade equipment, taking the special upgrade equipment as a host, reading new model data from the USB Flash disk, and writing the new model data into a Flash chip of the USB Flash disk for storage.
More preferably, the edge hardware equipment comprises an artificial intelligent processor, a FLASH memory chip, an edge UWB wireless communication module, a lithium battery power supply module and other peripheral sensing equipment; when the equipment leaves the factory, the related parameter data of the model required by the AI accelerating unit in the artificial intelligent processor is stored in the on-chip FLASH memory chip or FLASH memory chip as a read-only data segment according to the actual size, and the CPU configures the AI accelerating unit by reading the data in the FLASH during program running.
The system comprises edge equipment and special upgrading equipment, wherein the special upgrading equipment comprises a processor, a special upgrading UWB wireless communication module, a Flash chip, a display screen, a USB OTG interface and a lithium battery power supply module, the special upgrading equipment is connected with a server through the USB OTG interface in the form of storage equipment as a slave machine, after the server is trained to obtain new model parameter data, the data is written into a memory corresponding to the special upgrading equipment or the new model parameter data obtained by the server is copied into mobile equipment of a USB Flash disk for storage; accessing the USB Flash disk to a USB interface of special upgrade equipment, taking the special upgrade equipment as a host, reading new model data from the USB Flash disk, and writing the new model data into a Flash chip storage of the USB Flash disk;
the edge end hardware equipment comprises an artificial intelligent processor, a FLASH memory chip, an edge end UWB wireless communication module, a lithium battery power supply module and other peripheral sensing equipment; when the equipment leaves the factory, the related parameter data of the model required by the AI accelerating unit in the artificial intelligent processor is stored in the on-chip FLASH memory chip or FLASH memory chip as a read-only data segment according to the actual size, and the CPU configures the AI accelerating unit by reading the data in the FLASH during program running.
Preferably, the working process of the system is as follows:
when new model parameters are obtained through training at the server side, the special upgrading device transmits the model data to the local storage of the special upgrading device through the USB interface;
when the edge equipment needs to be upgraded by the model, the special upgrade equipment is placed in an area where the edge equipment is located, the special upgrade equipment is started, a model version needing to be upgraded is selected through a display screen, after the special upgrade equipment initiates upgrade operation, the special upgrade UWB wireless communication module sends upgrade signals in a broadcast mode, and when the edge equipment receives the upgrade scanning signals in a wake-up period, the processor executes an activation program to activate the edge UWB wireless communication module into a normal working mode, and communication connection is actively initiated to the special upgrade equipment;
after the two parties establish connection, calculating the accurate distance D between the two parties through a DS-TWR algorithm, encrypting the model data to be upgraded by taking the data D as a secret key at a special upgrading equipment end, and transmitting the encrypted model data to an edge end device;
the edge terminal equipment firstly checks the received model data to ensure the data integrity; if the verification is successful, decrypting the model data by taking the data D as a secret key, and writing the model data into a Flash memory; after the parameter updating is completed, the processor sets the edge UWB wireless communication module to a sleep mode again, and the upgrading is completed.
More preferably, when the edge terminal equipment is activated, after an upgrade task is started, the special upgrade equipment calls the special upgrade UWB wireless communication module and sends an upgrade signal in a broadcasting mode; the edge device sniffs whether an upgrade signal exists or not in the wake-up period of the self edge UWB device; once the edge equipment finds the upgrade signal, the edge UWB wireless communication module is activated and is converted into a normal working mode;
the accurate position distance D between the edge terminal equipment and the special upgrading equipment has uniqueness for the two pieces of equipment, meanwhile, the data D is obtained through solving by a DS-TWR algorithm and cannot be captured by other equipment, and the data D is further used as an encryption key of an AES algorithm.
An electronic device, comprising: a memory and at least one processor;
wherein the memory has a computer program stored thereon;
the at least one processor executes the computer program stored by the memory, so that the at least one processor executes the model upgrading method of the edge-side artificial intelligent chip.
A computer readable storage medium having a computer program stored therein, the computer program being executable by a processor to implement a model upgrade method for a marginal side artificial intelligence chip as described above.
The method and the system for upgrading the model of the edge-end artificial intelligent chip have the following advantages:
the edge end hardware equipment of the invention needs to be provided with an artificial intelligent processor, a FLASH memory chip, a UWB wireless communication module, a lithium battery power supply module, other peripheral sensing equipment and other components; when the equipment leaves the factory, the related parameter data of the model required by the AI accelerating unit in the processor is stored in the on-chip Flash or the off-chip Flash as a read-only data segment according to the actual size of the data segment, and the CPU configures the AI accelerating unit by reading the data in the Flash when the program runs;
secondly, in order to reduce the overall power consumption of the equipment, the UWB wireless communication module is configured into a low-power sleep mode through the CPU when the equipment works normally, and the sleep wakeup time of the UWB wireless communication module is agreed, so that the equipment can normally receive a model upgrade signal transmitted by the special upgrade equipment through UWB wireless communication;
the method and the device can solve the problem of difficult upgrading of the edge equipment model while ensuring the data security in wireless communication;
the invention can be applied to the projects related to artificial intelligence and the Internet of things.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a flow chart diagram of a model upgrade method of an edge-side artificial intelligent chip;
FIG. 2 is a schematic diagram of the structure of an edge device;
FIG. 3 is a schematic diagram of a dedicated upgrade apparatus;
fig. 4 is a schematic diagram of generating key data D.
Detailed Description
The model upgrading method and system of the edge-side artificial intelligent chip of the invention are described in detail below with reference to the accompanying drawings and specific embodiments.
Example 1:
as shown in fig. 1, the embodiment provides a method for upgrading a model of an edge-side artificial intelligent chip, which specifically includes the following steps:
s1, preparing model data: the special upgrade equipment obtains the model data of the new version through the USB interface and stores the model data into the internal FLASH storage;
s2, activating edge equipment: placing special upgrading equipment in an area where the edge equipment is located, selecting model data to be upgraded through a display screen, clicking to start an upgrading task, and further activating a UWB wireless communication module of the edge equipment to be converted into a normal working state;
s3, generating key data D: after the UWB wireless communication module of the edge equipment is activated and converted into a normal working state, the UWB wireless communication module initiates positioning communication with the special upgrading equipment, and the two pieces of equipment calculate the accurate position distance D between the edge equipment and the special upgrading equipment through a DS-TWR algorithm, as shown in figure 4;
s4, packaging a data encryption machine: the special upgrade equipment encrypts the model data by taking the calculated data D as an encryption key of an AES algorithm; after encryption, the data are packaged and checked according to a agreed communication protocol;
s5, data transmission: the special upgrade device sends the encrypted model data to the edge terminal device through the special upgrade UWB wireless communication module;
s6, data receiving and checking: the edge terminal equipment receives the analysis model data according to the agreed communication protocol, and checks the model data according to the agreed check rule so as to ensure the integrity of the data packet;
s7, decrypting data: if the data packet is checked to be correct, decrypting the model data packet through the key data D;
s8, upgrading a model: updating the obtained model data to an access area corresponding to the FLASH, and reconfiguring the acceleration unit by using the new model data, and after the model upgrading of the chip is completed, setting the edge-side UWB wireless communication module into a dormant mode by the processor again, and completing the upgrading.
When the edge terminal equipment is activated in the step S2 of the embodiment, after an upgrade task is started, the special upgrade equipment calls the special upgrade UWB wireless communication module and sends an upgrade signal in a broadcasting mode; the edge device sniffs whether an upgrade signal exists or not in the wake-up period of the self edge UWB device; and once the edge equipment discovers the upgrade signal, the edge UWB wireless communication module is activated, and the operation mode is changed into a normal operation mode.
In step S3 of this embodiment, the accurate position distance D between the edge device and the dedicated upgrade device is unique to both devices, and meanwhile, the data D is obtained by resolving through the DS-TWR algorithm, and is not captured by other devices, so that D is used as an encryption key of the AES algorithm.
As shown in fig. 3, the special upgrade device in this embodiment includes a processor, a special upgrade UWB wireless communication module, a Flash chip, a display screen, a USB OTG interface, and a lithium battery power supply module, where the special upgrade device is connected with a server through the USB OTG interface in the form of a storage device, and after the server trains to obtain new model parameter data, the server writes the data into a memory corresponding to the special upgrade device or copies the new model parameter data obtained by the server training to a mobile device of the USB disk for storage; and accessing the USB Flash disk to a USB interface of the special upgrade equipment, taking the special upgrade equipment as a host, reading new model data from the USB Flash disk, and writing the new model data into a Flash chip of the USB Flash disk for storage.
As shown in fig. 3, the edge hardware device in this embodiment includes an artificial intelligent processor, a FLASH memory chip, an edge UWB wireless communication module, a lithium battery power supply module, and other peripheral sensing devices; when the equipment leaves the factory, the related parameter data of the model required by the AI accelerating unit in the artificial intelligent processor is stored in the on-chip FLASH memory chip or FLASH memory chip as a read-only data segment according to the actual size, and the CPU configures the AI accelerating unit by reading the data in the FLASH during program running.
Example 2:
as shown in fig. 1 and 2, the embodiment provides a model upgrading system of an edge artificial intelligent chip, which comprises edge equipment and special upgrading equipment, wherein the special upgrading equipment comprises a processor, a special upgrading UWB wireless communication module, a Flash chip, a display screen, a USB OTG interface and a lithium battery power supply module, the special upgrading equipment is connected with a server through the USB OTG interface by taking a storage device form as a slave machine, and after the server is trained to obtain new model parameter data, the data is written into a memory corresponding to the special upgrading equipment or the new model parameter data obtained by the server is copied into a mobile device of a USB disk for storage; accessing the USB Flash disk to a USB interface of special upgrade equipment, taking the special upgrade equipment as a host, reading new model data from the USB Flash disk, and writing the new model data into a Flash chip storage of the USB Flash disk;
the edge end hardware equipment comprises an artificial intelligent processor, a FLASH memory chip, an edge end UWB wireless communication module, a lithium battery power supply module and other peripheral sensing equipment; when the equipment leaves the factory, the related parameter data of the model required by the AI accelerating unit in the artificial intelligent processor is stored in the on-chip FLASH memory chip or FLASH memory chip as a read-only data segment according to the actual size, and the CPU configures the AI accelerating unit by reading the data in the FLASH during program running.
The working process of the system is specifically as follows:
when new model parameters are obtained through training at the server side, the special upgrading device transmits the model data to the local storage of the special upgrading device through the USB interface;
when the edge end equipment needs to be upgraded by a model, placing special upgrade equipment in an area where the edge end equipment is located, starting the special upgrade equipment, selecting a model version needing to be upgraded through a display screen, after the special upgrade equipment initiates upgrade operation, sending an upgrade signal by a special upgrade UWB wireless communication module in a broadcast mode, and after the edge end equipment receives the upgrade scanning signal during the wakeup period of the UWB wireless communication module, executing an activation program by a processor, activating the edge end UWB wireless communication module into a normal working mode, and actively initiating communication connection to the special upgrade equipment;
after the two parties establish connection, calculating the accurate distance D between the two parties through a DS-TWR algorithm, encrypting the model data to be upgraded by taking the data D as a secret key at a special upgrading equipment end, and transmitting the encrypted model data to an edge end device;
the edge terminal equipment firstly checks the received model data to ensure the data integrity; if the verification is successful, decrypting the model data by taking the data D as a secret key, and writing the model data into a Flash memory; after the parameter updating is completed, the processor sets the edge UWB wireless communication module to a sleep mode again, and the upgrading is completed.
In this embodiment, when the edge device is activated, after an upgrade task is started, the dedicated upgrade device invokes the dedicated upgrade UWB wireless communication module, and broadcasts an upgrade signal; the edge device sniffs whether an upgrade signal exists or not in the wake-up period of the self edge UWB device; once the edge equipment finds the upgrade signal, the edge UWB wireless communication module is activated and is converted into a normal working mode;
in this embodiment, the precise location distance D between the edge device and the dedicated upgrade device has uniqueness for both devices, and meanwhile, the data D is obtained by resolving through the DS-TWR algorithm, and is not captured by other devices, so that D is used as an encryption key of the AES algorithm.
Example 3:
the embodiment also provides an electronic device, including: a memory and a processor;
wherein the memory stores computer-executable instructions;
and the processor executes the computer-executed instructions stored in the memory, so that the processor executes the model upgrading method of the edge-side artificial intelligent chip in any embodiment of the invention.
The processor may be a Central Processing Unit (CPU), but may also be other general purpose processors, digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may be used to store computer programs and/or modules, and the processor implements various functions of the electronic device by running or executing the computer programs and/or modules stored in the memory, and invoking data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the terminal, etc. The memory may also include high-speed random access memory, but may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, memory card only (SMC), secure Digital (SD) card, flash memory card, at least one disk storage period, flash memory device, or other volatile solid state memory device.
Example 4:
the embodiment also provides a computer readable storage medium, in which a plurality of instructions are stored, the instructions being loaded by a processor, to cause the processor to execute the model upgrading method of the edge-side artificial intelligent chip in any embodiment of the invention. Specifically, a system or apparatus provided with a storage medium on which a software program code realizing the functions of any of the above embodiments is stored, and a computer (or CPU or MPU) of the system or apparatus may be caused to read out and execute the program code stored in the storage medium.
In this case, the program code itself read from the storage medium may realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code form part of the present invention.
Examples of storage media for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (e.g., CD-ROMs, CD-R, CD-RWs, DVD-ROMs, DVD-RYM, DVD-RWs, DVD+RWs), magnetic tapes, nonvolatile memory cards, and ROMs. Alternatively, the program code may be downloaded from a server computer by a communication network.
Further, it should be apparent that the functions of any of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform part or all of the actual operations based on the instructions of the program code.
Further, it is understood that the program code read out by the storage medium is written into a memory provided in an expansion board inserted into a computer or into a memory provided in an expansion unit connected to the computer, and then a CPU or the like mounted on the expansion board or the expansion unit is caused to perform part and all of actual operations based on instructions of the program code, thereby realizing the functions of any of the above embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.

Claims (10)

1. The method for upgrading the model of the edge-end artificial intelligent chip is characterized by comprising the following steps of:
model data preparation: the special upgrade equipment obtains the model data of the new version through the USB interface and stores the model data into the internal FLASH storage;
activating the edge device: placing special upgrading equipment in an area where the edge equipment is located, selecting model data to be upgraded through a display screen, clicking to start an upgrading task, and activating a UWB wireless communication module of the edge equipment into a normal working state;
generating key data D: after the UWB wireless communication module of the edge equipment is activated and converted into a normal working state, the UWB wireless communication module of the edge equipment initiates positioning communication with the special upgrading equipment, and the two pieces of equipment calculate the accurate position distance D between the edge equipment and the special upgrading equipment through a DS-TWR algorithm;
and (3) packaging a data encryption machine: the special upgrade equipment encrypts the model data by taking the calculated data D as an encryption key of an AES algorithm; after encryption, the data are packaged and checked according to a agreed communication protocol;
and (3) data transmission: the special upgrade device sends the encrypted model data to the edge terminal device through the special upgrade UWB wireless communication module;
data receiving and checking: the edge terminal equipment receives the analysis model data according to the agreed communication protocol, and checks the model data according to the agreed check rule so as to ensure the integrity of the data packet;
decrypting data: if the data packet is checked to be correct, decrypting the model data packet through the key data D;
upgrading model: updating the obtained model data to an access area corresponding to the FLASH, and reconfiguring the acceleration unit by using the new model data, and after the model upgrading of the chip is completed, setting the edge-side UWB wireless communication module into a dormant mode by the processor again, and completing the upgrading.
2. The method for upgrading the model of the edge-side artificial intelligent chip according to claim 1, wherein when the edge-side equipment is activated, after an upgrading task is started, the special upgrading equipment calls the special upgrading UWB wireless communication module and sends an upgrading signal in a broadcasting mode; the edge device sniffs whether an upgrade signal exists or not in the wake-up period of the self edge UWB device; and once the edge equipment discovers the upgrade signal, the edge UWB wireless communication module is activated, and the operation mode is changed into a normal operation mode.
3. The method for upgrading the model of the edge-side artificial intelligent chip according to claim 1 or 2, wherein the accurate position distance D between the edge-side device and the dedicated upgrading device is unique to both devices, and the data D is obtained by solving through a DS-TWR algorithm, and further D is used as an encryption key of an AES algorithm.
4. The method for upgrading the model of the edge-side artificial intelligent chip according to claim 3, wherein the special upgrading device comprises a processor, a special upgrading UWB wireless communication module, a Flash chip, a display screen, a USB OTG interface and a lithium battery power supply module, the special upgrading device is connected with a server through the USB OTG interface in the form of a storage device, after the server is trained to obtain new model parameter data, the data are written into a memory corresponding to the special upgrading device or the new model parameter data obtained by the server is copied into a mobile device of a USB Flash disk for storage; and accessing the USB Flash disk to a USB interface of the special upgrade equipment, taking the special upgrade equipment as a host, reading new model data from the USB Flash disk, and writing the new model data into a Flash chip of the USB Flash disk for storage.
5. The method for upgrading a model of an edge-side artificial intelligent chip according to claim 4, wherein the edge-side hardware device comprises an artificial intelligent processor, a FLASH memory chip, an edge-side UWB wireless communication module, a lithium battery power module and other peripheral sensing devices; when the equipment leaves the factory, the related parameter data of the model required by the AI accelerating unit in the artificial intelligent processor is stored in the on-chip FLASH memory chip or FLASH memory chip as a read-only data segment according to the actual size, and the CPU configures the AI accelerating unit by reading the data in the FLASH during program running.
6. The system is characterized by comprising edge equipment and special upgrade equipment, wherein the special upgrade equipment comprises a processor, a special upgrade UWB wireless communication module, a Flash chip, a display screen, a USB OTG interface and a lithium battery power supply module, the special upgrade equipment is connected with a server through the USB OTG interface in the form of storage equipment as a slave machine, after the server is trained to obtain new model parameter data, the data is written into a memory corresponding to the special upgrade equipment or the new model parameter data obtained by the server is copied into mobile equipment of a U disk for storage; accessing the USB Flash disk to a USB interface of special upgrade equipment, taking the special upgrade equipment as a host, reading new model data from the USB Flash disk, and writing the new model data into a Flash chip storage of the USB Flash disk;
the edge end hardware equipment comprises an artificial intelligent processor, a FLASH memory chip, an edge end UWB wireless communication module, a lithium battery power supply module and other peripheral sensing equipment; when the equipment leaves the factory, the related parameter data of the model required by the AI accelerating unit in the artificial intelligent processor is stored in the on-chip FLASH memory chip or FLASH memory chip as a read-only data segment according to the actual size, and the CPU configures the AI accelerating unit by reading the data in the FLASH during program running.
7. The model upgrade system of the edge-side artificial intelligent chip according to claim 6, wherein the working process of the system is as follows:
when new model parameters are obtained through training at the server side, the special upgrading device transmits the model data to the local storage of the special upgrading device through the USB interface;
when the edge end equipment needs to be upgraded by the model, the special upgrade equipment is placed in an area where the edge end equipment is located, the special upgrade equipment is started, a model version needing to be upgraded is selected through a display screen, after the special upgrade equipment initiates upgrade operation, the special upgrade UWB wireless communication module sends upgrade signals in a broadcast mode, when the edge end UWB wireless communication module receives the upgrade scanning signals in a wake-up period, the processor executes an activation program, the edge end UWB wireless communication module is activated into a normal working mode, and communication connection is actively initiated to the special upgrade equipment;
after the two parties establish connection, calculating the accurate distance D between the two parties through a DS-TWR algorithm, encrypting the model data to be upgraded by taking the data D as a secret key at a special upgrading equipment end, and transmitting the encrypted model data to an edge end device;
the edge terminal equipment firstly checks the received model data to ensure the data integrity; if the verification is successful, decrypting the model data by taking the data D as a secret key, and writing the model data into a Flash memory; after the parameter updating is completed, the processor sets the edge UWB wireless communication module to a sleep mode again, and the upgrading is completed.
8. The model upgrade system of the edge-side artificial intelligent chip according to claim 7, wherein when the edge-side equipment is activated, after the upgrade task is started, the special upgrade equipment calls the special upgrade UWB wireless communication module and sends an upgrade signal in a broadcasting manner; the edge device sniffs whether an upgrade signal exists or not in the wake-up period of the self edge UWB device; once the edge equipment finds the upgrade signal, the edge UWB wireless communication module is activated and is converted into a normal working mode;
the accurate position distance D between the edge terminal equipment and the special upgrading equipment has uniqueness for the two pieces of equipment, meanwhile, the data D is obtained through solving by a DS-TWR algorithm, and then the data D is used as an encryption key of an AES algorithm.
9. An electronic device, comprising: a memory and at least one processor;
wherein the memory has a computer program stored thereon;
the at least one processor executing the computer program stored by the memory causes the at least one processor to perform the model upgrade method of the edge-side artificial intelligence chip of any one of claims 1 to 5.
10. A computer readable storage medium having stored therein a computer program executable by a processor to implement the model upgrade method of the edge-side artificial intelligence chip of any one of claims 1 to 5.
CN202311338529.XA 2023-10-17 2023-10-17 Model upgrading method and system for edge-end artificial intelligent chip Pending CN117421019A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311338529.XA CN117421019A (en) 2023-10-17 2023-10-17 Model upgrading method and system for edge-end artificial intelligent chip

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311338529.XA CN117421019A (en) 2023-10-17 2023-10-17 Model upgrading method and system for edge-end artificial intelligent chip

Publications (1)

Publication Number Publication Date
CN117421019A true CN117421019A (en) 2024-01-19

Family

ID=89530826

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311338529.XA Pending CN117421019A (en) 2023-10-17 2023-10-17 Model upgrading method and system for edge-end artificial intelligent chip

Country Status (1)

Country Link
CN (1) CN117421019A (en)

Similar Documents

Publication Publication Date Title
US20180089434A1 (en) Preserving trust data during operating system updates of a secure element of an electronic device
EP2988470B1 (en) Automatic purposed-application creation
US7366916B2 (en) Method and apparatus for an encrypting keyboard
CN102624699B (en) Method and system for protecting data
US11921860B2 (en) Rollback resistant security
CN101578609B (en) Secure booting a computing device
CN101984575A (en) Method and device for protecting mobile terminal software
TW201539242A (en) On-board applet migration
CN101373440B (en) Method and device for processing firmware upgrading data
CN106447855A (en) Shake-based unlocking method and terminal equipment
Wouters et al. My other car is your car: compromising the Tesla Model X keyless entry system
CN104320250A (en) Anti-counterfeiting authentication method based on NFC chip
CN112788150A (en) Registration method, terminal device, block chain management screen platform and storage medium
CN101888627B (en) Mobile terminal and system data protection method thereof
CN108573181B (en) RFID reader-writer, RFID system and block chain network
CN102982265B (en) Authentication method for storing basic input and output system (BIOS) setting
EP3493092A1 (en) Secure firmware provisioning and device binding mechanism
CN109088872A (en) Application method, device, electronic equipment and the medium of cloud platform with service life
KR20170009541A (en) Display driver integrated circuit for certifying application processor and mobile apparatus having the same
CN113127844A (en) Variable access method, device, system, equipment and medium
EP3769183B1 (en) Devices with removable smart cards
CN117421019A (en) Model upgrading method and system for edge-end artificial intelligent chip
EP4261713A1 (en) License file management method and apparatus, and device
CN111386513A (en) Data processing method, device and system chip
JP2016103217A (en) Information processing terminal and control program of information processing terminal

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination