CN113329358A - RISC-V instruction set-based AIOT multi-system edge gateway communication system and equipment - Google Patents
RISC-V instruction set-based AIOT multi-system edge gateway communication system and equipment Download PDFInfo
- Publication number
- CN113329358A CN113329358A CN202110476307.9A CN202110476307A CN113329358A CN 113329358 A CN113329358 A CN 113329358A CN 202110476307 A CN202110476307 A CN 202110476307A CN 113329358 A CN113329358 A CN 113329358A
- Authority
- CN
- China
- Prior art keywords
- module
- layer
- data
- management
- submodule
- 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.)
- Granted
Links
- 238000004891 communication Methods 0.000 title claims abstract description 69
- 238000007726 management method Methods 0.000 claims abstract description 100
- 230000005540 biological transmission Effects 0.000 claims abstract description 97
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 36
- 238000013473 artificial intelligence Methods 0.000 claims abstract description 19
- 230000003993 interaction Effects 0.000 claims abstract description 15
- 230000006978 adaptation Effects 0.000 claims abstract description 12
- 238000006243 chemical reaction Methods 0.000 claims abstract description 7
- 238000013523 data management Methods 0.000 claims abstract description 4
- 230000015654 memory Effects 0.000 claims description 28
- 238000012549 training Methods 0.000 claims description 28
- 238000000034 method Methods 0.000 claims description 26
- 230000006870 function Effects 0.000 claims description 23
- 238000004364 calculation method Methods 0.000 claims description 19
- 238000011156 evaluation Methods 0.000 claims description 19
- 238000013138 pruning Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 11
- 230000001133 acceleration Effects 0.000 claims description 10
- 238000007906 compression Methods 0.000 claims description 10
- 230000006835 compression Effects 0.000 claims description 10
- 238000011002 quantification Methods 0.000 claims description 9
- 238000011158 quantitative evaluation Methods 0.000 claims description 9
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000013475 authorization Methods 0.000 claims description 4
- 230000001960 triggered effect Effects 0.000 claims description 4
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000007418 data mining Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000004806 packaging method and process Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 26
- 238000005457 optimization Methods 0.000 description 10
- 238000001514 detection method Methods 0.000 description 7
- 230000001360 synchronised effect Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 238000013139 quantization Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000010267 cellular communication Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 241000272878 Apodiformes Species 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 239000004303 calcium sorbate Substances 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 238000013478 data encryption standard Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/04—Network architectures or network communication protocols for network security for providing a confidential data exchange among entities communicating through data packet networks
- H04L63/0428—Network architectures or network communication protocols for network security for providing a confidential data exchange among entities communicating through data packet networks wherein the data content is protected, e.g. by encrypting or encapsulating the payload
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Security & Cryptography (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides an AIOT multi-system edge gateway communication system and equipment based on a RISC-V instruction set. The scheme comprises a cloud cooperation platform, an intelligent algorithm layer, a security encryption layer, a management layer, a gateway adaptation layer, a protocol stack layer, an equipment driving layer, an embedded operation layer and an instruction set processor; the cloud collaboration platform performs data interaction with the intelligent algorithm layer, the secure encryption layer, the management layer, the gateway adaptation layer and the protocol stack layer perform secure encryption, data management, protocol conversion and network transmission on interacted data respectively, the embedded operation layer is provided with an instruction set processor, the embedded operation layer is provided with an equipment driving layer, and the equipment driving layer performs function interaction with external transmission equipment. According to the scheme, function interaction of an edge gateway and an equipment end is formed by combining an artificial intelligence data processing algorithm and multiple standard Internet of things communication protocols through open source instruction set architecture design, and multidimensional modeling, so that the working efficiency and the data compatibility are improved.
Description
Technical Field
The invention relates to the technical field of network communication, in particular to an AIOT multi-system edge gateway communication system and equipment based on a RISC-V instruction set.
Background
The internet of things is an important component of a new generation of information industry. With the emergence of artificial intelligence and the emergence of artificial intelligence hardware accelerators, a new field AIOT is derived, wherein the AIOT is the technical field of the combination of informatization and artificial intelligence. The AIOT can more comprehensively provide great upgrade for human production and living services, and has a huge application prospect.
With the development of the internet of things technology, a data center on the traditional cloud side performs data interaction with thousands of internet of things nodes, so that the computing resource cost is huge, network congestion exists, and the data interaction is more obvious in peak hours. Therefore, various industries are exploring novel' cloud removal and terminal AI (edge terminal) calculation, so that the safety and the performance of an Internet of things system are provided.
At present, in the existing internet of things technology, a 5G edge computing gateway is mainly placed at an edge close to data acquisition equipment to perform local edge computing, a main control processing module communicates with a 5G communication module through a USB and is configured to be connected with a sound pickup, a loudspeaker and a camera through the 5G communication module to perform voice acquisition, voice playing, image acquisition and data operation processing; the camera is used for acquiring images in real time and transmitting the images to the neural network processor for convolution neural network convolution calculation, so that vehicle information acquisition, detection and recognition, face recognition and detection and human body posture detection are realized. On one hand, on the other hand, only a compiled and added third-party library is provided for calling, and the algorithm is closed, so that the development of a user-defined operator and the optimization of a model are not facilitated in an AIOT application scene; on the other hand, the compatibility is poor, and only the 5G base station can be accessed, but the access edge node cannot be realized.
Disclosure of Invention
In view of the above problems, the invention provides an AIOT multi-system edge gateway communication system and equipment based on a RISC-V instruction set, and through design based on an open source instruction set architecture, combination with an artificial intelligence data processing algorithm, and utilization of an Internet of things communication protocol of multiple systems and multi-dimensional module, functional interaction of an edge gateway and an equipment end is formed, and working efficiency and data compatibility are improved.
According to a first aspect of the embodiments of the present invention, an AIOT multi-mode edge gateway communication system based on RISC-V instruction set is provided.
The AIOT multi-system edge gateway communication system based on RISC-V instruction set specifically comprises:
the system comprises a cloud cooperation platform, an intelligent algorithm layer, a security encryption layer, a management layer, a gateway adaptation layer, a protocol stack layer, a device driving layer, an embedded operation layer and an instruction set processor; the cloud collaboration platform performs data interaction with the intelligent algorithm layer, the secure encryption layer, the management layer, the gateway adaptation layer and the protocol stack layer perform secure encryption, data management, protocol conversion and network transmission on interacted data respectively, the embedded operation layer is provided with the instruction set processor, the embedded operation layer is configured with the device driving layer, and the device driving layer performs function interaction with external transmission equipment.
Specifically, the RISC-V Instruction Set is called Reduced Instruction Set Computing V in its entirety, and is called a fifth-generation RISC Reduced Instruction Set computer in Chinese.
In one or more embodiments, preferably, the secure encryption layer specifically includes a DTLA sub-module, an AES sub-module, a CA sub-module, an SSL sub-module, a SHA265 sub-module, a HASH sub-module, and an MD5 sub-module; the DTLA submodule is used for digital transmission authorization management, the AES submodule is used for high-level encryption, the CA submodule is used for managing issuing security certificates and encryption information security keys, the SSL submodule is used for providing a secure transmission protocol, and the SHA265 submodule is used for performing secure encryption by adopting cryptographic hashing; the HASH submodule is used for mapping an input with any length into a HASH value with a fixed length through a HASH algorithm; the MD5 submodule is used to generate a 16 byte hash value.
In one or more embodiments, preferably, the protocol stack layer includes an MQTT transmission sub-module, a CoAP transmission protocol sub-module, a TCP/IP transmission protocol sub-module, a UDP transmission protocol sub-module, an ARP transmission protocol sub-module, a DHCP transmission protocol sub-module, an IPv4 transmission protocol sub-module, an IPv6 transmission protocol sub-module, an NVME transmission protocol sub-module, and a BLE transmission protocol sub-module; the MQTT transmission submodule is used for issuing transmission data according to a message queue telemetry transmission standard; the CoAP transmission protocol sub-module is used for forwarding data by a CoAP transmission protocol; the TCP/IP transmission protocol sub-module is used for forwarding transmission data between networks by a TCP/IP transmission protocol, and the IPv4 transmission protocol sub-module and the IPv6 transmission protocol sub-module are respectively used for generating IP addresses of different levels; the NVME transmission protocol sub-module is used for carrying out open collection on information in mobile equipment and a data center through a nonvolatile memory; the UDP transmission protocol sub-module is used for forwarding data by a UDP transmission protocol; the ARP transmission protocol submodule is used for forwarding the data by an ARP transmission protocol, and the DHCP transmission protocol submodule is used for forwarding the data in a local area network by a DHCP transmission protocol.
In one or more embodiments, preferably, the management layer includes a Radio management submodule, a Powr management submodule, a Memory management submodule, a Time management submodule, a Client management submodule, a KeyManager management submodule, and an algorithm management submodule; the Radio management submodule is used for managing an audio module in the embedded system; the Powr management submodule is used for managing a power supply management module in the embedded system; the Memory management submodule is used for managing a management module of a Memory in the embedded system; the Time management submodule is used for managing a timer and a Time-triggered task function module in the embedded system; the Client management submodule is used for a user group management module in the embedded system; the KeyManager management submodule is used for managing a user key management module in the embedded system; the algorithm management submodule is used for managing the adopted AI intelligent algorithm, wherein the AI intelligent algorithm comprises YOLOV3 and MobileNet.
In one or more embodiments, preferably, the gateway adaptation layer includes access to various network interfaces, specifically including WAN, LAN, RS485, Wi-Fi, Zigbee, 4G/5G, LTE-V, BLE 4.1.1, NB-IoT, Sigfox, LoRa, 6 loppan.
In one or more embodiments, preferably, the cloud collaboration platform includes edge end nodes and a cloud center, the edge end nodes include a management module, a data operation module, an application service module, and a book storage module, and the cloud center includes an edge intelligent management suite and an artificial intelligence application service, wherein the artificial intelligence application service cooperates with the intelligent algorithm layer to perform image processing, function operation, network training, streaming computation, and big data mining; the edge intelligence management suite includes program, resource, and application management.
According to a second aspect of the embodiments of the present invention, an AIOT multi-mode edge gateway communication method based on RISC-V instruction set is provided.
The AIOT multi-system edge gateway communication method based on RISC-V instruction set specifically comprises the following steps:
performing convolution operation through an embedded operation layer and an instruction set processor configured on the embedded operation layer to generate processing data, wherein the processing data comprises Internet of things node data, image information and sound information;
performing data security check on all data acquired through the device driver layer, dividing the data into training data and verification data after check and preprocessing, compressing the training data, performing heterogeneous operation, generating and storing a data analysis model;
performing parameter quantitative evaluation according to edge hardware computing equipment to obtain an evaluation accuracy, and optimizing and compressing again when the evaluation accuracy is reduced;
and adjusting the speed of the completed calculation to enable different hardware acceleration chips to complete different edge operations and output support codes.
In one or more embodiments, preferably, the performing parameter quantitative evaluation according to the edge hardware computing device to obtain an evaluation accuracy, and when the evaluation accuracy decreases, performing optimization and compression again includes:
inputting a pre-training model;
determining pruning and quantification parameters according to hardware parameters and the pre-training model;
adjusting the pre-training model according to the pruning and quantification parameters;
quantizing the pre-training model to generate model accuracy;
when the model accuracy is lower than a preset accuracy threshold value, re-determining pruning and quantification parameters;
and saving the fine-tuned pre-training model as a compression model.
In one or more embodiments, preferably, the speed adjusting the completed computation to enable different hardware acceleration chips to complete different edge operations and output a support code specifically includes:
packaging an edge calculation model;
acquiring current scheduling configuration information, and calculating scheduling configuration through a CPU (central processing unit);
calculating scheduling configuration by using hardware acceleration, and generating a C + + reasoning code and a bmode file;
and generating a computation scheduling subgraph, and storing the scheduling configuration, the bmode file, the C + + inference code and the scheduling subgraph together as an edge computation model.
According to a third aspect of embodiments of the present invention, there is provided an electronic device comprising a memory and a processor, the memory being configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the steps as described in the first aspect of embodiments of the present invention.
The technical scheme provided by the embodiment of the invention can have the following beneficial effects:
1) the scheme adopts an open source instruction set architecture, and utilizes artificial intelligence to process data, so that online access, analysis, processing and operation of a large amount of node data can be realized, the efficiency is high, and the compatibility is strong;
2) the gateway function of all the nodes of the Internet of things is integrated, and the data sensing and data processing speed is improved through multi-system, multi-protocol and multifunctional equipment;
3) according to the scheme, a large number of modularized designs are adopted, so that an external hardware module can be quickly accessed into a system, and convenient and fast application of the Internet of things is realized;
4) according to the scheme, through multiple application scenes such as load vehicle-road coordination and the like, the functions of data sensing operation, data fusion and node broadcasting under different scenes are realized, and the operation performance requirement under a high-load system is met.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a block diagram of an AIOT multi-mode edge gateway communication system based on RISC-V instruction set according to an embodiment of the present invention.
Fig. 2 is a block diagram of a secure encryption layer in an AIOT multi-mode edge gateway communication system based on RISC-V instruction set according to an embodiment of the present invention.
Fig. 3 is a diagram of the structure of the protocol stack layers in the AIOT multi-system edge gateway communication system based on the RISC-V instruction set according to an embodiment of the present invention.
Fig. 4 is a block diagram of a management layer in an AIOT multi-mode edge gateway communication system based on a RISC-V instruction set according to an embodiment of the present invention.
Fig. 5 is a structural diagram of a cloud collaboration platform in an AIOT multi-mode edge gateway communication system based on a RISC-V instruction set according to an embodiment of the present invention.
Fig. 6 is a schematic diagram of an AIOT multi-mode edge gateway communication system based on RISC-V instruction set according to an embodiment of the present invention.
Fig. 7 is a flowchart of an AIOT multi-mode edge gateway communication method based on RISC-V instruction set according to an embodiment of the present invention.
Fig. 8 is a flowchart of performing parameter quantitative evaluation to obtain an evaluation accuracy according to an edge hardware computing device in an AIOT multi-mode edge gateway communication method based on a RISC-V instruction set according to an embodiment of the present invention, and performing optimization and compression again when the evaluation accuracy is reduced.
Fig. 9 is a schematic diagram of performing parameter quantitative evaluation to obtain an evaluation accuracy according to an edge hardware computing device in an AIOT multi-mode edge gateway communication method based on a RISC-V instruction set according to an embodiment of the present invention, and performing optimization and compression again when the evaluation accuracy is reduced.
Fig. 10 is a flowchart of performing speed adjustment on completed calculations in an AIOT multi-mode edge gateway communication method based on a RISC-V instruction set, so that different hardware accelerator chips complete different edge operations and output support codes according to an embodiment of the present invention.
Fig. 11 is a schematic diagram of performing speed adjustment on completed calculation in an AIOT multi-mode edge gateway communication method based on a RISC-V instruction set, so that different hardware accelerator chips complete different edge operations and output support codes according to an embodiment of the present invention.
Fig. 12 is a block diagram of an electronic device in one embodiment of the invention.
Detailed Description
In some of the flows described in the present specification and claims and in the above figures, a number of operations are included that occur in a particular order, but it should be clearly understood that these operations may be performed out of order or in parallel as they occur herein, with the order of the operations being indicated as 101, 102, etc. merely to distinguish between the various operations, and the order of the operations by themselves does not represent any order of performance. Additionally, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first", "second", etc. in this document are used for distinguishing different messages, devices, modules, etc., and do not represent a sequential order, nor limit the types of "first" and "second" to be different.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The internet of things is an important component of a new generation of information industry. With the emergence of artificial intelligence and the emergence of artificial intelligence hardware accelerators, a new field AIOT is derived, wherein the AIOT is the technical field of the combination of informatization and artificial intelligence. The AIOT can more comprehensively provide great upgrade for human production and living services, and has a huge application prospect.
With the development of the internet of things technology, a data center on the traditional cloud side performs data interaction with thousands of internet of things nodes, so that the computing resource cost is huge, network congestion exists, and the data interaction is more obvious in peak hours. Therefore, various industries are exploring novel' cloud removal and terminal AI (edge terminal) calculation, so that the safety and the performance of an Internet of things system are provided.
At present, in the existing internet of things technology, a 5G edge computing gateway is mainly placed at an edge close to data acquisition equipment to perform local edge computing, a main control processing module communicates with a 5G communication module through a USB and is configured to be connected with a sound pickup, a loudspeaker and a camera through the 5G communication module to perform voice acquisition, voice playing, image acquisition and data operation processing; the camera is used for acquiring images in real time and transmitting the images to the neural network processor for convolution neural network convolution calculation, so that vehicle information acquisition, detection and recognition, face recognition and detection and human body posture detection are realized. On one hand, on the other hand, only a compiled and added third-party library is provided for calling, and the algorithm is closed, so that the development of a user-defined operator and the optimization of a model are not facilitated in an AIOT application scene; on the other hand, the compatibility is poor, and only the 5G base station can be accessed, but the access edge node cannot be realized.
The embodiment of the invention provides an AIOT multi-system edge gateway communication system and equipment based on a RISC-V instruction set. According to the scheme, function interaction between the edge gateway and the equipment end is formed by designing based on an open source instruction set architecture, combining an artificial intelligence data processing algorithm and utilizing an Internet of things communication protocol of various systems and multi-dimensional modeling, and the working efficiency and the data compatibility are improved.
According to a first aspect of the embodiments of the present invention, an AIOT multi-mode edge gateway communication system based on RISC-V instruction set is provided.
Fig. 1 is a block diagram of an AIOT multi-mode edge gateway communication system based on RISC-V instruction set according to an embodiment of the present invention.
As shown in fig. 1, the AIOT multi-mode edge gateway communication system based on RISC-V instruction set specifically includes:
the system comprises a cloud collaboration platform 101, an intelligent algorithm layer 102, a security encryption layer 103, a management layer 104, a gateway adaptation layer 105, a protocol stack layer 106, a device driver layer 107, an embedded operation layer 108 and an instruction set processor 109; the cloud collaboration platform 101 performs data interaction with the intelligent algorithm layer 102, the secure encryption layer 103, the management layer 104, the gateway adaptation layer 105, and the protocol stack layer 106 perform secure encryption, data management, protocol conversion, and network transmission on interacted data, respectively, the instruction set processor 109 is disposed on the embedded operation layer 108, the device driver layer 107 is configured on the embedded operation layer 108, and the device driver layer 107 performs function interaction with an external transmission device.
Specifically, the Instruction Set refers to RISC-V, which is called Reduced Instruction Set Computing V in english as RISC of the fifth generation (Reduced Instruction Set computer); IOT/AIOT is called Internet of Things/AI + Internet of Things in English, and corresponding to Chinese name of Internet of Things/artificial intelligence Internet of Things.
In the embodiment of the invention, on one hand, artificial intelligence processing data is introduced into the whole data processing architecture through a multi-layer and multi-type module, and meanwhile, an open source instruction set architecture is utilized to analyze, process and operate a large amount of node data; on the other hand, the modularized design is adopted, so that different modules execute different types of functions, the flow of data processing is simplified, and the possibility and efficiency of accessing the external hardware into the system are increased.
Fig. 2 is a block diagram of a secure encryption layer in an AIOT multi-mode edge gateway communication system based on RISC-V instruction set according to an embodiment of the present invention.
As shown in fig. 2, in one or more embodiments, preferably, the secure encryption layer 103 specifically includes a DTLA submodule 201, an AES submodule 202, a CA submodule 203, an SSL submodule 204, an SHA265 submodule 205, a HASH submodule 206, and an MD5 submodule 207; the DTLA submodule 201 is configured to perform digital transmission authorization management, the AES submodule 202 is configured to perform high-level encryption, the CA submodule 203 is configured to manage issuing security credentials and an encryption information security key, the SSL submodule 204 is configured to provide a secure transmission protocol, and the SHA265 submodule 205 is configured to perform secure encryption by using cryptographic hash; the HASH submodule 206 is configured to map an input with an arbitrary length to a HASH value with a fixed length through a HASH algorithm; the MD5 submodule 207 is used to generate a 16 byte hash value.
In the embodiment of the invention, the multi-type data encryption submodule is used for processing the data and the information with any length and encrypting the data and the information into the transmission data meeting the data transmission protocol and the data encryption standard, so that the subsequent processing of the data is convenient and the safe operation of the whole system is ensured.
Fig. 3 is a diagram of the structure of the protocol stack layers in the AIOT multi-system edge gateway communication system based on the RISC-V instruction set according to an embodiment of the present invention.
As shown in fig. 3, in one or more embodiments, preferably, the protocol stack layer 106 includes an MQTT transmission sub-module 301, a CoAP transmission protocol sub-module 302, a TCP/IP transmission protocol sub-module 303, a UDP transmission protocol sub-module 304, an ARP transmission protocol sub-module 305, a DHCP transmission protocol sub-module 306, an IPv4 transmission protocol sub-module 307, an IPv6 transmission protocol sub-module 308, an NVME transmission protocol sub-module 309, and a BLE transmission protocol sub-module 310; the MQTT transmission sub-module 301 is configured to issue transmission data according to a message queue telemetry transmission standard; the CoAP transmission protocol sub-module 302 is configured to forward data according to a CoAP transmission protocol; the TCP/IP transport protocol sub-module 303 is configured to forward transport data between networks by using a TCP/IP transport protocol, and the IPv4 transport protocol sub-module 307 and the IPv6 transport protocol sub-module 308 are respectively configured to generate IP addresses at different levels; the NVME transport protocol sub-module 309 is configured to perform open collection on information in the mobile device and the data center through the nonvolatile memory; the UDP transport protocol sub-module 304 is configured to forward the data in a UDP transport protocol; the ARP transport protocol sub-module 305 is configured to forward data in an ARP transport protocol, and the DHCP transport protocol sub-module 306 is configured to forward data in a DHCP transport protocol within the lan.
In the embodiment of the invention, the actually received multi-type data is subjected to data forwarding of different types of protocols through the multi-type data protocols, the compatibility of the multi-type data in the Internet of things can be realized through the method, when the corresponding type of protocol data exists, the data is directly converted according to the corresponding module, and the data is sent and transferred according to the configuration of the different types of data forwarding sub-modules and the specified standard of transmission data.
Fig. 4 is a block diagram of a management layer in an AIOT multi-mode edge gateway communication system based on a RISC-V instruction set according to an embodiment of the present invention.
As shown in fig. 4, in one or more embodiments, preferably, the management layer 104 includes a Radio management submodule 401, a Powr management submodule 402, a Memory management submodule 403, a Time management submodule 404, a Client management submodule 405, a KeyManager management submodule 406, and an algorithm management submodule 407; the Radio management submodule 401 is configured to manage an audio module in an embedded system; the Powr management submodule 402 is used for managing a power management module in the embedded system; the Memory management submodule 403 is a management module for managing a Memory in the embedded system; the Time management submodule 404 is configured to manage a timer and a Time-triggered task function module in the embedded system; the Client management submodule 405 is used for a user group management module in the embedded system; the KeyManager management submodule 406 is configured to manage a user key management module in the embedded system; the algorithm management submodule 407 is configured to manage the adopted AI intelligent algorithms, where the AI intelligent algorithms include YOLOV3 and MobileNet.
In the embodiment of the invention, the management layer is the core of the whole multi-system communication, power is supplied to each submodule through the management layer, data storage, audio data storage and time recording and triggering are carried out, communication with a user is carried out through the management layer, and the KeyManager management submodule can also carry out data encryption through the security encryption layer and carry out management on a user key, so that the system security is ensured.
In the embodiment of the invention, the management of an AI intelligent algorithm is correspondingly included, and the rapid identification characteristics of different types of input data are realized by setting the intelligent algorithms at least comprising YOLOV3 and MobileNet, and the training of corresponding capability is carried out according to the characteristics.
The AIOT multi-system edge gateway communication system based on the RISC-V instruction set specifically further comprises a device driver, wherein the device driver comprises an Ethernet module, a synchronous transceiver module, an asynchronous transceiver module, a digital-to-analog conversion module, a safety input and output module, a USB interface module, a GPIO interface, an I2C interface and an SPI interface; the ethernet module is configured to transmit data in an ethernet manner, the synchronous transceiver module, the asynchronous transceiver module, the USB interface module, the GPIO interface, the I2C interface, and the SPI interface are all configured to serve as interfaces for data transmission, the secure input/output module is configured to perform information transmission between memory data and the synchronous transceiver module, the asynchronous transceiver module, the USB interface module, the GPIO interface, the I2C interface, and the SPI interface, and the digital-to-analog conversion module is configured to perform conversion between an analog quantity and a digital quantity on the collected data.
In the embodiment of the invention, the cooperation of the protocol stack layer and the hardware layer on the software layer on the hardware layer is realized through the cooperation between the interfaces of various types and the converted hardware module of the interface data type, thereby realizing the compatible transmission of the data of the interfaces of various types.
In one or more embodiments, preferably, the gateway adaptation layer includes access to various network interfaces, specifically including WAN, LAN, RS485, Wi-Fi, Zigbee, 4G/5G, LTE-V, BLE 4.1.1, NB-IoT, Sigfox, LoRa, 6 loppan.
In the embodiment of the invention, the access of multi-system network data has many advantages for designing a multi-system communication protocol compatible architecture, including mass access technology which can meet various communication systems such as long distance, short distance, wired and wireless, and the like, and can be compatible with data transmission of various communication systems such as WAN, LAN, RS485, Wi-Fi, Zigbee, 4G/5G, LTE-V, BLE 4.1.1, NB-IoT, Sigfox, LoRa, 6LoPWAN, and the like.
Specifically, the WAN is fully called Wide Area Network; LAN is called Local Area Network LAN; 5G is called 5th generation mobile networks 5th generation cellular communication network; zigbee is a novel wireless communication technology and is suitable for a series of electronic component devices with short transmission range and low data transmission rate; the 4G is called 4th generation mobile networks 4th generation cellular communication network; LTE-V is specifically long term evolution technology-vehicle communication; BLE is called Bluetooth Low Energy totally, or Bluetooth LE, BLE Low power consumption Bluetooth; NB-IoT is known as Narrow Band Internet of Things, and represents the narrowband Internet of Things; the Sigfox refers to a low-power-consumption wide area network deployed by a Sigfox company on the whole world and provides Internet of things connection service, and the user equipment integrates a radio frequency module or chip supporting a Sigfox protocol and can be connected to the Sigfox network after the connection service is opened; the LoRa is called Long Range, and is specifically a Long-distance transmission narrow-band Internet of things; the 6LoPWAN is all called IPv6over Low-Power Wireless Personal Area Networks, in particular to a private Wireless local Area network based on IPv 6.
Fig. 5 is a structural diagram of a cloud collaboration platform in an AIOT multi-mode edge gateway communication system based on a RISC-V instruction set according to an embodiment of the present invention.
As shown in fig. 5, in one or more embodiments, preferably, the cloud collaboration platform includes edge end nodes and a cloud center, the edge end nodes include a management group module, a data operation group module, an application service group module, and a book storage group module, and the cloud center includes an edge intelligent management suite and an artificial intelligence application service, wherein the artificial intelligence application service cooperates with the intelligent algorithm layer to perform image processing, function operation, network training, streaming computation, and big data mining; the edge intelligence management suite includes program, resource, and application management.
In the embodiment of the invention, intelligent operations in the fields of visual processing, face recognition, object detection, classification and the like of neural network models based on a terminal cloud cooperative intelligent computing model, such as YOLO V2, YOLO V3, YOLO V2-tiny, YOLO V3-tiny, MobileNet V2, SqueezeNet, GoogleNet, ShuffleNet, Xception and the like, are realized, and on the basis, optimization of the neural network terminal cloud cooperative inference algorithm model and a terminal cloud computing cooperative dynamic task division strategy are realized, so that intelligent computing of edge equipment and cloud cooperation is realized, and the application service efficiency and reliability are improved. Among them, YOLO, all called youonly Look one, is a neural network model.
Fig. 6 is a schematic diagram of an AIOT multi-mode edge gateway communication system based on RISC-V instruction set according to an embodiment of the present invention. As shown in fig. 6, a specific embodiment is shown, wherein the AI algorithm framework includes: YOLOV3, MobileNet, etc. The IOT gateway adaptation layer includes: zigbee, LORA and other Internet of things communication modes; a security encryption layer: DTLA (digital transmission authorization management), AES (Advanced Encryption Standard), CA (Certificate Authority) is a network organization that manages and issues Security credentials and Encryption information Security keys, SSL (Secure Sockets Layer), TLS (Transport Layer Security), which is a Security protocol that provides Security and data integrity for network communications, SHA265 (Secure Hash Algorithm (SHA) is a family of cryptographic Hash functions), Hash (Hash, which is an input of arbitrary length, also called pre-mapping) is transformed into an output of fixed length by a Hash Algorithm, MD5 (5 information Digest Algorithm (english: MD 26 Message-Digest 5), which is a widely used cryptographic Hash function that can generate a 128-bit Hash value, for ensuring the complete and consistent information transmission.
Protocol stack layer: MQTT (message queue telemetry transport) is a message Protocol based on a publish/subscribe paradigm under The ISO standard, CoAP (The managed Application Protocol, CoAP is an Application layer Protocol in a 6LowPAN Protocol stack), TCP/IP (Transmission Control Protocol/Internet Protocol), which is a Protocol cluster capable of implementing information Transmission among a plurality of different networks, UDP (User Datagram Protocol), which is a User Datagram Protocol, which is collectively called User Datagram Protocol, ARP (Address Resolution Protocol), which is a TCP/IP Protocol that acquires a physical Address according to an IP Address, DHCP (dynamic host configuration Protocol), which is a network Protocol of a local area network, which refers to a range of IP addresses controlled by a server, and which can automatically acquire an IP Address and a mask allocated by The server, IPv4 (IPv version 4 (english: Protocol 4), IPv4), also known as the fourth version of the internet communication protocol, is the fourth revision in the development of the internet protocol, which is also the first widely deployed version of this protocol. ) IPv6 (english full name Internet Protocol Version 6, chinese name: internet protocol version 6) is the next generation IP protocol designed by the internet engineering task force to replace IPv4), NVME (full name: NVM Express is an open collection of standards and information to fully demonstrate the advantages of non-volatile memory in all types of computing environments from mobile devices to data centers.
And (3) a management layer: radio refers to an audio module for managing an embedded system; powr refers to a power management module for managing the embedded system; memory refers to a management module for managing Memory in an embedded system; the Time refers to a timer and a Time-triggered task function module used for managing the embedded system; the Client refers to a user group management module used in the embedded system; the KeyManager refers to a user key management module for managing the embedded system;
radio management submodule, Powr management submodule, Memory management submodule, Time management submodule, Client management submodule and KeyManager management submodule
Device driver layer: ethernet refers to an Ethernet module on a circuit board; UARTs refers to synchronous and asynchronous transceiver modules on a circuit board, also known as serial ports; the ADC is an analog-to-digital converter on the circuit board and is used for converting an analog circuit into a digital circuit; SDIO (secure Digital Input and output) Chinese name: the safe digital input and output refers to an external interface on the circuit board and is used for realizing a communication interface bus with the memory card; USB, a universal serial bus, a common communication bus, a bus used to implement high-speed communication between peripherals and a master controller; GPIO (general purpose input/output) is a general programmable input/output control port and is used for realizing programmable control on a digital circuit; I2C is a simple, bidirectional two-wire synchronous serial bus. It requires only two wires to transfer information between devices connected to the bus. SPI is a full-duplex synchronous serial bus developed by motorola, a synchronous serial port for communication between a Microprocessing Control Unit (MCU) and peripheral devices; AI Uint is an AI specific circuit.
According to a second aspect of the embodiments of the present invention, an AIOT multi-mode edge gateway communication method based on RISC-V instruction set is provided.
Fig. 7 is a flowchart of an AIOT multi-mode edge gateway communication method based on RISC-V instruction set according to an embodiment of the present invention.
As shown in fig. 7, the AIOT multi-mode edge gateway communication method based on RISC-V instruction set specifically includes:
s701, performing convolution operation through an embedded operation layer and an instruction set processor configured on the embedded operation layer to generate processing data, wherein the processing data comprises Internet of things node data, image information and sound information;
s702, performing data security check on all data acquired through the device driver layer, dividing the data into training data and verification data after check and preprocessing, compressing the training data, performing heterogeneous operation, generating and storing a data analysis model;
s703, carrying out parameter quantitative evaluation according to the edge hardware computing equipment to obtain an evaluation accuracy, and optimizing and compressing again when the evaluation accuracy is reduced;
s704, speed adjustment is carried out on the completed calculation, so that different hardware acceleration chips complete different edge operations, and support codes are output.
In the embodiment of the invention, a series of work such as data processing, model training analysis, hardware equipment configuration, operation speed optimization and the like is carried out by combining the embedded operation layer with a specific instruction set processor.
Fig. 8 is a flowchart of performing parameter quantitative evaluation to obtain an evaluation accuracy according to an edge hardware computing device in an AIOT multi-mode edge gateway communication method based on a RISC-V instruction set according to an embodiment of the present invention, and performing optimization and compression again when the evaluation accuracy is reduced.
As shown in fig. 8, in one or more embodiments, preferably, the performing quantitative evaluation on the parameter according to the edge hardware computing device to obtain an evaluation accuracy, and when the evaluation accuracy decreases, performing optimization and compression again specifically includes:
s801, inputting a pre-training model;
s802, pruning and quantifying parameters are determined according to hardware parameters and the pre-training model;
s803, adjusting the pre-training model according to the pruning and quantification parameters;
s804, quantizing the pre-training model to generate model accuracy;
s805, when the accuracy of the model is lower than a preset accuracy threshold value, pruning and quantification parameters are determined again;
s806, storing the fine-tuned pre-training model as a compression model.
In the embodiment of the invention, when the model is used, the accuracy of the pre-training model needs to be evaluated to a certain extent, the specific evaluation mainly depends on the determination of pruning and quantization parameters, the model accuracy is generated on the basis, and after the model accuracy is obtained, fine adjustment is carried out in a mode of comparing with a threshold value, so that the online self-adaptive adjustment of the pre-training model is realized.
Fig. 9 is a schematic diagram of performing parameter quantitative evaluation to obtain an evaluation accuracy according to an edge hardware computing device in an AIOT multi-mode edge gateway communication method based on a RISC-V instruction set according to an embodiment of the present invention, and performing optimization and compression again when the evaluation accuracy is reduced. As shown in fig. 9, the current model is optimized and compressed according to the deployment hardware condition, and the compression ratio and quantization parameters of pruning are determined according to the edge hardware computing device. The model is then fine-tuned and quantified, and the accuracy and size are evaluated. If the accuracy is greatly reduced after pruning and quantization, the pruning and quantization parameters are adjusted again for recompression.
Fig. 10 is a flowchart of performing speed adjustment on completed calculations in an AIOT multi-mode edge gateway communication method based on a RISC-V instruction set, so that different hardware accelerator chips complete different edge operations and output support codes according to an embodiment of the present invention.
In one or more embodiments, preferably, the speed adjusting the completed computation to enable different hardware acceleration chips to complete different edge operations and output a support code specifically includes:
s1001, packaging an edge calculation model;
s1002, acquiring current scheduling configuration information, and calculating scheduling configuration through a CPU (central processing unit);
s1003, calculating scheduling configuration by using hardware acceleration, and generating a C + + reasoning code and a bmode file;
s1004, generating a computation scheduling subgraph, and storing the scheduling configuration, the bmode file, the C + + inference code and the scheduling subgraph as an edge computation model.
In the embodiment of the present invention, before performing edge calculation, the edge calculation model needs to be encapsulated, so that a total edge calculation model is generated for each function type in this way, and is encapsulated.
Fig. 11 is a schematic diagram of performing speed adjustment on completed calculation in an AIOT multi-mode edge gateway communication method based on a RISC-V instruction set, so that different hardware accelerator chips complete different edge operations and output support codes according to an embodiment of the present invention. As shown in fig. 11, for the selected computation scheduling configuration, the selected computation scheduling configuration is decomposed into a CPU computation scheduling configuration and a hardware acceleration chip computation scheduling configuration, and for the kangzhi K210, the green wave GAP8, the computationally abundant BM1880, the crystal core N22, and the hummingbird E203, which are all manufacturers and chip models of RISC-V chips, the specific rear-end computation support codes or models are respectively generated and packed with the subgraph description to the edge computation model.
According to a third aspect of the embodiments of the present invention, there is provided an electronic apparatus. Fig. 12 is a block diagram of an electronic device in one embodiment of the invention. The electronic device shown in fig. 12 is a general multi-mode edge gateway communication apparatus, which includes a general computer hardware structure, which includes at least a processor 1201 and a memory 1202. The processor 1201 and the memory 1202 are connected by a bus 1203. The memory 1202 is adapted to store instructions or programs executable by the processor 1201. The processor 1201 may be a stand-alone microprocessor or a collection of one or more microprocessors. Thus, the processor 1201 implements the processing of data and the control of other devices by executing instructions stored by the memory 1202 to perform the method flows of embodiments of the present invention as described above. The bus 1203 connects the above components together, as well as connecting the above components to a display controller 1204 and a display device and input/output (I/O) device 1205. Input/output (I/O) devices 1205 may be a mouse, keyboard, modem, network interface, touch input device, motion-sensing input device, printer, and other devices known in the art. Typically, the input/output devices 1205 are connected to the system through input/output (I/O) controllers 1206.
The technical scheme provided by the embodiment of the invention is based on a RISCV instruction set and an edge calculation controller of an integrated CNN hardware accelerator, wherein the CNN hardware accelerator is used for realizing convolution processing on data in specified memories such as a ROM, a RAM and the like, the problems of low convolution operation speed, heavy load, high cost and the like of the CNN at the edge end are solved, and the specific beneficial effects can comprise:
1) the scheme adopts an open source instruction set architecture, and utilizes artificial intelligence to process data, so that online access, analysis, processing and operation of a large amount of node data can be realized, the efficiency is high, and the compatibility is strong;
2) the gateway function of all the nodes of the Internet of things is integrated, and the data sensing and data processing speed is improved through multi-system, multi-protocol and multifunctional equipment;
3) according to the scheme, a large number of modularized designs are adopted, so that an external hardware module can be quickly accessed into a system, and convenient and fast application of the Internet of things is realized;
4) according to the scheme, through multiple application scenes such as load vehicle-road coordination and the like, the functions of data sensing operation, data fusion and node broadcasting under different scenes are realized, and the operation performance requirement under a high-load system is met.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. An AIOT multi-system edge gateway communication system based on RISC-V instruction set, characterized by comprising: the system comprises a cloud cooperation platform, an intelligent algorithm layer, a security encryption layer, a management layer, a gateway adaptation layer, a protocol stack layer, a device driving layer, an embedded operation layer and an instruction set processor; the cloud collaboration platform performs data interaction with the intelligent algorithm layer, the secure encryption layer, the management layer, the gateway adaptation layer and the protocol stack layer perform secure encryption, data management, protocol conversion and network transmission on interacted data respectively, the embedded operation layer is provided with the instruction set processor, the embedded operation layer is configured with the device driving layer, and the device driving layer performs function interaction with external transmission equipment.
2. The AIOT multi-standard edge gateway communication system based on RISC-V instruction set as claimed in claim 1, wherein said secure encryption layer comprises DTLA sub-module, AES sub-module, CA sub-module, SSL sub-module, SHA265 sub-module, HASH sub-module, MD5 sub-module; the DTLA submodule is used for digital transmission authorization management, the AES submodule is used for high-level encryption, the CA submodule is used for managing issuing security certificates and encryption information security keys, the SSL submodule is used for providing a secure transmission protocol, and the SHA265 submodule is used for performing secure encryption by adopting cryptographic hashing; the HASH submodule is used for mapping an input with any length into a HASH value with a fixed length through a HASH algorithm; the MD5 submodule is used to generate a 16 byte hash value.
3. The RISC-V instruction set-based AIOT multi-system edge gateway communication system of claim 1, wherein the protocol stack layer includes an MQTT transmission sub-module, a CoAP transmission protocol sub-module, a TCP/IP transmission protocol sub-module, a UDP transmission protocol sub-module, an ARP transmission protocol sub-module, a DHCP transmission protocol sub-module, an IPv4 transmission protocol sub-module, an IPv6 transmission protocol sub-module, an NVME transmission protocol sub-module, a BLE transmission protocol sub-module; the MQTT transmission submodule is used for issuing transmission data according to a message queue telemetry transmission standard; the CoAP transmission protocol sub-module is used for forwarding data by a CoAP transmission protocol; the TCP/IP transmission protocol sub-module is used for forwarding transmission data between networks by a TCP/IP transmission protocol, and the IPv4 transmission protocol sub-module and the IPv6 transmission protocol sub-module are respectively used for generating IP addresses of different levels; the NVME transmission protocol sub-module is used for carrying out open collection on information in mobile equipment and a data center through a nonvolatile memory; the UDP transmission protocol sub-module is used for forwarding data by a UDP transmission protocol; the ARP transmission protocol submodule is used for forwarding the data by an ARP transmission protocol, and the DHCP transmission protocol submodule is used for forwarding the data in a local area network by a DHCP transmission protocol.
4. The AIOT multi-system edge gateway communication system based on RISC-V instruction set as claimed in claim 1, wherein said management layer comprises Radio management submodule, Powr management submodule, Memory management submodule, Time management submodule, Client management submodule, KeyManager management submodule, algorithm management submodule; the Radio management submodule is used for managing an audio module in the embedded system; the Powr management submodule is used for managing a power supply management module in the embedded system; the Memory management submodule is used for managing a management module of a Memory in the embedded system; the Time management submodule is used for managing a timer and a Time-triggered task function module in the embedded system; the Client management submodule is used for a user group management module in the embedded system; the KeyManager management submodule is used for managing a user key management module in the embedded system; the algorithm management submodule is used for managing the adopted AI intelligent algorithm, wherein the AI intelligent algorithm comprises YOLOV3 and MobileNet.
5. The AIOT multi-system edge gateway communication system based on RISC-V instruction set as claimed in claim 1, wherein the gateway adaptation layer comprises access to various network interfaces, specifically WAN, LAN, RS485, Wi-Fi, Zigbee, 4G/5G, LTE-V, BLE 4.1.1, NB-IoT, Sigfox, LoRa, 6 LoPWAN.
6. The AIOT multi-mode edge gateway communication system based on RISC-V instruction set as claimed in claim 1 wherein the cloud collaboration platform comprises edge end nodes and a cloud center, the edge end nodes comprise a management module, a data operation module, an application service module, and a book storage module, the cloud center comprises an edge intelligence management suite and an artificial intelligence application service, wherein the artificial intelligence application service cooperates with the intelligence algorithm layer to perform image processing, function operation, network training, streaming computation, big data mining; the edge intelligence management suite includes program, resource, and application management.
7. An AIOT multi-system edge gateway communication method based on RISC-V instruction set is characterized in that the method specifically comprises the following steps:
performing convolution operation through an embedded operation layer and an instruction set processor configured on the embedded operation layer to generate processing data, wherein the processing data comprises Internet of things node data, image information and sound information;
performing data security check on all data acquired through the device driver layer, dividing the data into training data and verification data after check and preprocessing, compressing the training data, performing heterogeneous operation, generating and storing a data analysis model;
performing parameter quantitative evaluation according to edge hardware computing equipment to obtain an evaluation accuracy, and optimizing and compressing again when the evaluation accuracy is reduced;
and adjusting the speed of the completed calculation to enable different hardware acceleration chips to complete different edge operations and output support codes.
8. The AIOT multi-mode edge gateway communication method based on RISC-V instruction set as claimed in claim 7, wherein said performing parameter quantification assessment to obtain assessment accuracy according to edge hardware computing device, and when the assessment accuracy is reduced, re-optimizing and compressing, specifically comprises:
inputting a pre-training model;
determining pruning and quantification parameters according to hardware parameters and the pre-training model;
adjusting the pre-training model according to the pruning and quantification parameters;
quantizing the pre-training model to generate model accuracy;
when the model accuracy is lower than a preset accuracy threshold value, re-determining pruning and quantification parameters;
and saving the fine-tuned pre-training model as a compression model.
9. The AIOT multi-mode edge gateway communication method based on RISC-V instruction set as claimed in claim 7, wherein said speed adjusting the completed calculation, making different hardware acceleration chips complete different edge operations, and outputting support codes, specifically comprising:
packaging an edge calculation model;
acquiring current scheduling configuration information, and calculating scheduling configuration through a CPU (central processing unit);
calculating scheduling configuration by using hardware acceleration, and generating a C + + reasoning code and a bmode file;
and generating a computation scheduling subgraph, and storing the scheduling configuration, the bmode file, the C + + inference code and the scheduling subgraph together as an edge computation model.
10. An electronic device comprising a memory and a processor, wherein the memory is configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the steps of any of claims 7-9.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110476307.9A CN113329358B (en) | 2021-04-29 | 2021-04-29 | RISC-V instruction set-based AIOT multi-system edge gateway communication system and equipment |
| PCT/CN2021/092328 WO2022227106A1 (en) | 2021-04-29 | 2021-05-08 | Aiot multi-standard edge gateway communication system based on risc-v instruction set, and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110476307.9A CN113329358B (en) | 2021-04-29 | 2021-04-29 | RISC-V instruction set-based AIOT multi-system edge gateway communication system and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113329358A true CN113329358A (en) | 2021-08-31 |
| CN113329358B CN113329358B (en) | 2022-10-18 |
Family
ID=77414096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110476307.9A Active CN113329358B (en) | 2021-04-29 | 2021-04-29 | RISC-V instruction set-based AIOT multi-system edge gateway communication system and equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113329358B (en) |
| WO (1) | WO2022227106A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114338281A (en) * | 2021-11-11 | 2022-04-12 | 广州软件应用技术研究院 | Model distribution application design method and system based on edge computing gateway |
| CN115270668A (en) * | 2022-07-18 | 2022-11-01 | 北京师范大学 | Chip special for open source hardware of information technology education |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102271124A (en) * | 2010-06-01 | 2011-12-07 | 富士通株式会社 | Data processing equipment and data processing method |
| JP2017163612A (en) * | 2017-06-12 | 2017-09-14 | パナソニックＩｐマネジメント株式会社 | Terminal authentication system, server device, and terminal authentication method |
| CN109635936A (en) * | 2018-12-29 | 2019-04-16 | 杭州国芯科技股份有限公司 | A kind of neural networks pruning quantization method based on retraining |
| CN109862011A (en) * | 2019-02-01 | 2019-06-07 | 华南理工大学 | A kind of environment of internet of things real-time monitoring system calculated based on mist |
| CN110007961A (en) * | 2019-02-01 | 2019-07-12 | 中山大学 | A kind of edge calculations hardware structure based on RISC-V |
| CN110633247A (en) * | 2018-06-25 | 2019-12-31 | 国际商业机器公司 | Cognitive computing system and service using an internet of things environment |
| CN110913391A (en) * | 2019-10-25 | 2020-03-24 | 国网陕西省电力公司信息通信公司 | Multi-data-source communication encryption and decryption system and method |
| CN111324383A (en) * | 2020-02-28 | 2020-06-23 | 西安微电子技术研究所 | RISC-V instruction extension-based security coprocessor structure |
| WO2020143094A1 (en) * | 2019-01-09 | 2020-07-16 | 网宿科技股份有限公司 | Intelligent management method and system based on edge computing |
| US10742421B1 (en) * | 2019-03-08 | 2020-08-11 | Ares Technologies, Inc. | Methods and systems for anonymous hardware attestation |
| CN112199322A (en) * | 2020-09-30 | 2021-01-08 | 中国电力科学研究院有限公司 | Electric power intelligent control terminal and SOC electric power chip architecture based on RISC-V |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11509654B2 (en) * | 2017-02-06 | 2022-11-22 | Pcms Holdings, Inc. | Securing communication of devices in the internet of things |
| CN108092884A (en) * | 2017-11-23 | 2018-05-29 | 南京邮电大学 | A kind of wireless access gateway system and application process |
| CN109889592B (en) * | 2019-02-25 | 2020-05-01 | 北京邮电大学 | Intelligent manufacturing method and device based on edge calculation |
| CN111163002B (en) * | 2019-12-31 | 2022-06-10 | 广州智光电气股份有限公司 | Container-based edge gateway system and energy data processing method |
| CN112260944A (en) * | 2020-10-26 | 2021-01-22 | 浙江大学 | Embedded intelligent edge computing gateway based on ARM-Linux system |
-
2021
- 2021-04-29 CN CN202110476307.9A patent/CN113329358B/en active Active
- 2021-05-08 WO PCT/CN2021/092328 patent/WO2022227106A1/en active Application Filing
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102271124A (en) * | 2010-06-01 | 2011-12-07 | 富士通株式会社 | Data processing equipment and data processing method |
| JP2017163612A (en) * | 2017-06-12 | 2017-09-14 | パナソニックＩｐマネジメント株式会社 | Terminal authentication system, server device, and terminal authentication method |
| CN110633247A (en) * | 2018-06-25 | 2019-12-31 | 国际商业机器公司 | Cognitive computing system and service using an internet of things environment |
| CN109635936A (en) * | 2018-12-29 | 2019-04-16 | 杭州国芯科技股份有限公司 | A kind of neural networks pruning quantization method based on retraining |
| WO2020143094A1 (en) * | 2019-01-09 | 2020-07-16 | 网宿科技股份有限公司 | Intelligent management method and system based on edge computing |
| CN109862011A (en) * | 2019-02-01 | 2019-06-07 | 华南理工大学 | A kind of environment of internet of things real-time monitoring system calculated based on mist |
| CN110007961A (en) * | 2019-02-01 | 2019-07-12 | 中山大学 | A kind of edge calculations hardware structure based on RISC-V |
| US10742421B1 (en) * | 2019-03-08 | 2020-08-11 | Ares Technologies, Inc. | Methods and systems for anonymous hardware attestation |
| CN110913391A (en) * | 2019-10-25 | 2020-03-24 | 国网陕西省电力公司信息通信公司 | Multi-data-source communication encryption and decryption system and method |
| CN111324383A (en) * | 2020-02-28 | 2020-06-23 | 西安微电子技术研究所 | RISC-V instruction extension-based security coprocessor structure |
| CN112199322A (en) * | 2020-09-30 | 2021-01-08 | 中国电力科学研究院有限公司 | Electric power intelligent control terminal and SOC electric power chip architecture based on RISC-V |
Non-Patent Citations (1)
| Title |
|---|
| 谢小瑜: "深度学习在泛在电力物联网中的应用与挑战", 《电力自动化设备》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114338281A (en) * | 2021-11-11 | 2022-04-12 | 广州软件应用技术研究院 | Model distribution application design method and system based on edge computing gateway |
| CN114338281B (en) * | 2021-11-11 | 2023-04-07 | 广州软件应用技术研究院 | Model distribution application design method and system based on edge computing gateway |
| CN115270668A (en) * | 2022-07-18 | 2022-11-01 | 北京师范大学 | Chip special for open source hardware of information technology education |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022227106A1 (en) | 2022-11-03 |
| CN113329358B (en) | 2022-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113329358B (en) | RISC-V instruction set-based AIOT multi-system edge gateway communication system and equipment | |
| JP6587330B2 (en) | Random forest model training method, electronic apparatus, and storage medium | |
| CN109426574A (en) | Distributed computing system, data transmission method and device in distributed computing system | |
| CN110012002A (en) | Message parsing method, device and server, computer readable storage medium | |
| CN102143218B (en) | Web access cloud architecture and access method | |
| CN104052759A (en) | System for realizing add-and-play technology of internet of things | |
| CN103957237A (en) | Architecture of elastic cloud | |
| CN110928694B (en) | Computer system | |
| CN109598250A (en) | Feature extracting method, device, electronic equipment and computer-readable medium | |
| CN110515889A (en) | Embedded FPGA swarm intelligence computing platform hardware frame | |
| CN204231387U (en) | A kind of intelligent micro-grid control system realizing high-speed communication based on FPGA | |
| CN117196014B (en) | Model training method and device based on federal learning, computer equipment and medium | |
| CN111597411B (en) | Method and system for distinguishing and identifying power specification data frames | |
| CN106713411A (en) | Communication apparatus and method based on Ethernet virtual serial port technology and LTE wireless network | |
| CN103731344A (en) | 6LOWPAN Internet-of-things gateway based on linux application layer and achieving method | |
| CN110198533A (en) | A kind of method and BLE bluetooth equipment of long-range control BLE bluetooth equipment | |
| CN112416844A (en) | Spike signal detection and classification device based on FPGA and GPU | |
| WO2023221644A1 (en) | Aiot device reuse modeling method based on cloud-edge collaboration system | |
| Wang et al. | An Edge Computing Based Gateway for WIA-PA Networks | |
| CN114003238B (en) | Container deployment method, device, equipment and storage medium based on transcoding card | |
| CN109753359A (en) | It is a kind of for constructing FPGA board, server and the system of resource pool | |
| CN115858672A (en) | Power terminal management method and device, electronic equipment and storage medium | |
| CN103795686B (en) | Internet of things service system and the exchange method for realizing virtual information | |
| CN106487694A (en) | A kind of data flow processing method and device | |
| CN106600754A (en) | Cloud calculation iris identification access control apparatus |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |