CN113485257B - Industrial protocol analysis built-in program optimization method - Google Patents
Industrial protocol analysis built-in program optimization method Download PDFInfo
- Publication number
- CN113485257B CN113485257B CN202110612738.3A CN202110612738A CN113485257B CN 113485257 B CN113485257 B CN 113485257B CN 202110612738 A CN202110612738 A CN 202110612738A CN 113485257 B CN113485257 B CN 113485257B
- Authority
- CN
- China
- Prior art keywords
- scheme
- protocol analysis
- built
- program
- analysis program
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM]
- G05B19/41885—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS], computer integrated manufacturing [CIM] characterised by modeling, simulation of the manufacturing system
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32339—Object oriented modeling, design, analysis, implementation, simulation language
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Communication Control (AREA)
- Computer And Data Communications (AREA)
Abstract
The invention discloses an optimization method of an industrial protocol analysis built-in program, which comprises the following steps: s1, establishing a side cloud cooperative framework; s2, carrying out built-in scheme coding on the industrial protocol; s3, optimizing according to historical data; s4, pre-deploying an optimal program group on the edge equipment; and S5, updating the built-in program as required. The method is based on the edge cloud cooperative architecture. According to the invention, historical use data of the program is analyzed according to the industrial protocol, an optimal program built-in scheme is obtained by using an optimization algorithm, and a program group corresponding to the optimal scheme is built into factory edge equipment. And after the edge equipment runs, dynamically updating the built-in protocol analysis program according to actual requirements.
Description
Technical Field
The invention belongs to the technical field of industrial Internet of things, and relates to an optimization method for an industrial protocol analysis built-in program.
Background
With the rapid development of the industrial internet of things technology, massive heterogeneous industrial protocol analysis programs are generated, which brings huge challenges to the deployment of edge devices. The edge device is customized to embed a plurality of different industrial protocol analysis programs according to different industrial scenes. In the conventional method, all protocol analysis programs are often built in the edge device and are selectively used according to requirements in an actual industrial scene. Currently, there is also a method using a cloud edge coordination mode to implement a protocol parser for dynamically configuring an edge device. And the edge equipment sends a downloading request to the cloud end through the network according to the actual requirement of the industrial environment, and downloads the corresponding industrial protocol analysis program. However, these methods face a complex and varied industrial environment with the following challenges:
1) In the traditional method, a large amount of useless analysis programs are arranged in the edge device, so that a large amount of storage space of the edge device is wasted, and the capability of storing business related data is reduced.
2) Although the dynamic deployment of the protocol analysis program can be realized by a method based on cloud edge cooperation, the method can cause a large amount of consumption of network bandwidth and cause network congestion due to the use of network transmission in the case of facing massive edge devices.
Disclosure of Invention
In order to solve the problems, the invention provides an industrial protocol analysis built-in program optimization method, which comprises the following steps:
s1, establishing a side cloud cooperative framework;
s2, carrying out built-in scheme coding on the industrial protocol;
s3, optimizing according to historical data;
s4, pre-deploying an optimal program group on the edge equipment;
and S5, updating the built-in program as required.
Preferably, the establishing of the edge cloud collaborative architecture comprises the step that an industrial protocol analysis program library of a cloud side manages an industrial protocol analysis program; the sending and receiving request program of the edge terminal updates the built-in protocol analysis program; and running a protocol analysis program at the edge terminal.
Preferably, the encoding of the built-in schemes for the industrial protocols includes regarding all protocol analysis programs on each edge device as a set of built-in schemes, and encoding each set of built-in schemes according to a preset encoding mode.
Preferably, the optimizing according to the historical data comprises the following steps:
s31, randomly generating an initialization scheme;
s32, calculating the fitness of each scheme according to a predefined fitness function, and storing the optimal scheme as a global optimal scheme;
s33, calculating the updating step length of each scheme by a step length updating method, and updating each scheme;
s34, selecting a scheme randomly according to a preset elimination probability for elimination;
s35, calculating the fitness of the updated and eliminated scheme, comparing the fitness with the fitness of the global optimal scheme, and replacing the global optimal scheme if the fitness is better;
s36, judging whether the optimal scheme meets an expected target or reaches the maximum iteration times, and if not, returning to S33;
and S37, if yes, obtaining an optimal scheme.
Preferably, the step-size updating method comprises a priority-based binary cuckoo algorithm.
Preferably, the optimal program group is pre-deployed, the built-in protocol analysis program is updated as required, the optimal built-in protocol analysis program group is obtained by optimizing according to historical data, and the optimal protocol analysis program group is pre-deployed in the edge device. After the edge device runs, the built-in program of the edge device is updated according to the requirement.
The beneficial effects of the invention at least comprise: the method is based on a side cloud cooperative architecture, an optimal protocol analysis program built-in scheme is obtained through collected historical use data of the industrial protocol analysis program by using an optimization algorithm, and a program group corresponding to the scheme is placed in factory edge equipment. During actual operation, the edge device dynamically updates the built-in program through the cloud protocol analysis program library according to the requirement. The method avoids the waste of storage space caused by the built-in of all protocol analysis programs, and also reduces the network bandwidth pressure caused by requesting to download the programs from the cloud under the high concurrency state of massive edge devices.
Drawings
FIG. 1 is a flowchart illustrating steps of a method for optimizing an industrial protocol parsing embedded program according to an embodiment of the present invention;
fig. 2 is a diagram of a cloud-side coordination architecture of an industrial protocol parsing built-in program optimization method according to an embodiment of the present invention;
FIG. 3 is a flowchart of the step S3 of the method for optimizing the industrial protocol analysis built-in program according to the embodiment of the present invention;
fig. 4 is a data protocol parsing process diagram of the industrial protocol parsing built-in program optimization method according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
On the contrary, the invention is intended to cover alternatives, modifications, equivalents and alternatives which may be included within the spirit and scope of the invention as defined by the appended claims. Furthermore, in the following detailed description of the present invention, certain specific details are set forth in order to provide a better understanding of the present invention. It will be apparent to one skilled in the art that the present invention may be practiced without these specific details.
Referring to fig. 1, a flowchart of an optimization method for analyzing a built-in program for an industrial protocol according to a technical solution of the present invention is shown, and includes the following steps:
s1, establishing a side cloud cooperative framework;
s2, carrying out built-in scheme coding on the industrial protocol;
s3, optimizing according to historical data;
s4, pre-deploying an optimal program group on the edge equipment;
and S5, updating the built-in program as required.
S1, establishing a side cloud cooperative framework, wherein an industrial protocol analysis program library comprising a cloud end manages an industrial protocol analysis program; the sending and receiving request program of the edge terminal updates the built-in protocol analysis program; and running a protocol analysis program at the edge terminal.
And S2, carrying out built-in scheme coding on the industrial protocol, wherein all protocol analysis programs on each edge device are regarded as a set of built-in schemes, and each set of built-in schemes are coded according to a preset coding mode.
S3, optimizing according to historical data, and comprising the following steps:
s31, randomly generating an initialization scheme;
s32, calculating the fitness of each scheme according to a predefined fitness function, and storing the optimal scheme as a global optimal scheme;
s33, calculating the updating step length of each scheme by a step length updating method, and updating each scheme;
s34, selecting a scheme randomly according to a preset elimination probability for elimination;
s35, calculating the fitness of the updated and eliminated scheme, comparing the fitness with the fitness of the global optimal scheme, and replacing the global optimal scheme if the fitness is better;
s36, judging whether the optimal scheme meets an expected target or reaches the maximum iteration times, and if not, returning to S33;
and S37, if yes, obtaining an optimal scheme.
The step size updating method comprises a binary cuckoo algorithm based on priority.
Pre-deploying an optimal program group, updating the built-in protocol analysis program according to needs, obtaining the optimal built-in protocol analysis program group according to historical data optimization, and pre-deploying the optimal built-in program group on the edge equipment. After the edge device runs, the built-in program of the edge device is updated according to the requirement.
The invention is based on a cloud edge collaboration architecture, and referring to fig. 2, the invention has a cloud management platform and a plurality of edge devices. The cloud platform assists the edge device to dynamically update the built-in protocol analysis program, and the edge device realizes the deployment and operation of different protocol analysis programs according to different industrial environments. The cloud platform can be deployed in public clouds such as the Ali cloud and the like, and can also be deployed in private clouds. An industrial protocol analysis program library is deployed on the cloud platform and is responsible for daily maintenance of various protocol analysis programs, responding to an update request from the edge equipment and providing downloading service of the protocol analysis programs for the edge equipment. The edge device can be any embedded device supporting the operation of the industrial protocol parser, and the optional scheme is raspberry pi 4B. The edge and the cloud interact in a Modbus UDP and FTP mode. The request information of the protocol analysis program is interacted through a Modbus UDP protocol, and the industrial protocol analysis program is transmitted through the FTP.
The built-in scheme optimization flow of the invention is shown in figure 3. And S31, randomly generating a group of initialization schemes, wherein the coding of the schemes adopts a preset coding mode, and a 01 coding mode is adopted in the embodiment of the invention. All protocol resolvers are encoded as a set of 01 arrays, with 0 indicating that the protocol resolver represented by the bit is not adopted by the schema and 1 indicating that the protocol resolver represented by the bit is adopted by the schema. And S32, calculating the fitness of each scheme according to a predefined fitness function, and storing the scheme with the optimal fitness as a global optimal scheme. And S33, calculating the updating step length of each scheme by adopting a problem-adaptive step length updating method, and updating each scheme. An alternative in an embodiment of the invention is to use a priority-based binary cuckoo algorithm as the update method for the step size.
In order to adapt to discrete coding, the binary cuckoo algorithm based on priority adopts two methods to carry out binary updating on the step length of the cuckoo algorithm. The mathematical expression of the method is shown as formulas 1, 2 and 3. Wherein i represents a dimension number, step is a Step length calculated based on the flight of the Levy, and PS i The rand is a random number, W, subject to uniform distribution for the probability value obtained after conversion of the flight step length of the Levis i For the binary value, W, of the protocol analysis program corresponding to the ith dimension in the iteration i ' is the binary value of the protocol parser corresponding to the ith dimension in the previous iteration. The first method is formula 1, and probability value PS corresponding to each dimension step length is calculated through sigmoid function i . PS for random number rand generated by each protocol and corresponding dimension i The comparison updates the protocol parser array W. The second method is equation 2 and equation 3. When Step i <If not, adopting formula 2 to update the protocol analysis program array W, and when Step i >If =0, the protocol analysis program array W is updated using equation 3. The second method is to calculate the probability value corresponding to each dimension step size through the deformation of the sigmoid function. If only formula 1 is used for updating, the global diversity is strong, and almost no convergence exists. And only the formula 2,3 is used for updating, the convergence is strong, but the global diversity is weak. Therefore, the present invention is similar to the method of setting the control coefficient prTwo methods are used to solve the problem. pr is between 0 and 1. A random number is generated for each step calculated from the levey flight. If the random number is less than pr, the code of the corresponding dimension is updated using equation 1. If the random number is greater than pr, updating the code of the corresponding dimension using formula 2 and formula 3.
And S34, randomly selecting a scheme for elimination according to a preset elimination probability, and updating the eliminated scheme by adopting a proper method. In the priority-based binary cuckoo algorithm, two protocol analysis program arrays are selected from a group to perform multipoint intersection in a roulette mode. And selecting the offspring with better fitness to update the abandoned scheme. And judging whether the updated scheme meets the constraint condition, if not, setting the dimensionality with the array value of 1 as 0 in a roulette mode according to the preset priority of each protocol analysis program until the constraint condition is met.
And S35, calculating the fitness of the updated and eliminated scheme, comparing the fitness with the global optimal scheme, and replacing the global optimal scheme if the fitness is better. And S36, returning to S33 to continue iteration until the global optimal scheme meets the expected target or the maximum iteration number is reached. And S37, obtaining the finally obtained global optimal scheme as the optimal scheme.
The execution process of the dynamic data protocol analysis method of the invention is shown in figure 4 and mainly comprises two parts: the initial protocol analysis program is internally provided with and dynamically updates the protocol analysis program. In the initial protocol analysis program built-in stage, firstly, finding an optimal program scheme through an optimization algorithm according to historical use data of the protocol analysis program; and secondly, embedding a corresponding protocol analysis program into the edge device according to the optimal scheme. In the stage of dynamically updating the protocol analysis program, the edge device deletes the unused program in the built-in program according to the actual industrial scene, then sends a request for downloading a new program to the cloud protocol analysis program library, and after receiving the request, the cloud allows the edge device to download the corresponding protocol analysis program from the protocol analysis program library.
In particular, a typical configuration table of historical usage data of the protocol parser of the present invention is shown in table 1 below. The configuration table contains the edge device ID, as well as the ID, type, size, and most recent usage time of the protocol parser for each protocol parser used by the edge device. The device ID and the program ID are unique, and the unit of the program size is KB.
Table 1 protocol parser configuration table
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (1)
1. An industrial protocol analysis built-in program optimization method is characterized by comprising the following steps:
s1, establishing a side cloud cooperative framework;
s2, carrying out built-in scheme coding on the industrial protocol;
s3, optimizing according to historical data;
s4, pre-deploying an optimal program group on the edge equipment;
s5, updating the built-in program as required;
the establishment side cloud cooperative architecture comprises an industrial protocol analysis program library of a cloud end, and the industrial protocol analysis program is managed by the industrial protocol analysis program library; the sending and receiving request program of the edge terminal updates the built-in protocol analysis program; running a protocol analysis program at an edge end;
the method for coding the built-in scheme of the industrial protocol comprises the steps that all protocol analysis programs on each edge device are regarded as a set of built-in scheme, and each set of built-in scheme is coded according to a preset coding mode;
the optimizing according to the historical data comprises the following steps:
s31, randomly generating an initialization scheme;
s32, calculating the fitness of each scheme according to a predefined fitness function, and storing the optimal scheme as a global optimal scheme;
s33, calculating the updating step length of each scheme by a step length updating method, and updating each scheme;
s34, randomly selecting a scheme for elimination according to a preset elimination probability;
s35, calculating the fitness of the updated and eliminated scheme, comparing the fitness with the fitness of the global optimal scheme, and if the fitness is better, replacing the global optimal scheme;
s36, judging whether the optimal scheme meets an expected target or reaches the maximum iteration number, and if not, returning to S33;
s37, if yes, obtaining an optimal scheme;
the step updating method comprises a priority-based binary cuckoo algorithm;
the method comprises the steps that an optimal program group is pre-deployed, a built-in protocol analysis program is updated as required, the optimal built-in protocol analysis program group is obtained through optimization according to historical data, the optimal protocol analysis program group is pre-deployed in edge equipment, and the built-in program of the edge equipment is updated as required after the edge equipment runs;
and S33, calculating the updating step length of each scheme by adopting a problem-adaptive step length updating method, and updating each scheme, wherein the method specifically comprises the following steps of adopting a priority-based binary cuckoo algorithm as the updating method of the step length:
priority-based binary to accommodate discrete codingThe cuckoo algorithm adopts two methods to carry out binary updating on the Step length of the cuckoo algorithm, the mathematical expressions are formulas (1), (2) and (3), wherein i represents a dimension number, step is the Step length obtained based on Levy flight calculation, and PS is the Step length obtained based on Levy flight calculation i The rand is a random number, W, subject to uniform distribution for the probability value obtained after conversion of the flight step length of the Levis i For the binary value, W, of the protocol analysis program corresponding to the ith dimension in the iteration i ' is the binary value of the protocol analysis program corresponding to the ith dimension in the previous iteration; the first method is formula (1), and probability value PS corresponding to each dimension step length is calculated through sigmoid function i The random number rand generated by each protocol and the PS of the corresponding dimension i Comparing and updating the protocol analysis program array W; the second method is formula (2) and formula (3), when Step i <If =0, the protocol analysis program array W is updated by the formula (2), and when Step i >If =0, updating the protocol analysis program array W by using the formula (3); the second method is to calculate the probability value corresponding to each dimension step length through the deformation of a sigmoid function, and simultaneously adopt the two methods through the mode of setting a control coefficient pr, wherein the value of pr ranges from 0 to 1, a random number is generated for each step length calculated by Lai dimensional flight, if the random number is less than pr, the code of the corresponding dimension is updated by using a formula (1), and if the random number is more than pr, the code of the corresponding dimension is updated by using a formula (2) and a formula (3);
s34, selecting a random selection scheme according to a preset elimination probability for elimination, wherein in a binary cuckoo algorithm based on priority, selecting two protocol analysis program arrays from a group in a roulette mode for multipoint intersection; and selecting the offspring with the maximum fitness to update the abandoned scheme, judging whether the updated scheme meets the preset constraint condition, if not, setting the dimensionality with the array value of 1 as 0 in a roulette mode according to the preset priority of each protocol analysis program until the constraint condition is met.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110612738.3A CN113485257B (en) | 2021-06-02 | 2021-06-02 | Industrial protocol analysis built-in program optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110612738.3A CN113485257B (en) | 2021-06-02 | 2021-06-02 | Industrial protocol analysis built-in program optimization method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113485257A CN113485257A (en) | 2021-10-08 |
| CN113485257B true CN113485257B (en) | 2023-03-03 |
Family
ID=77934029
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110612738.3A Active CN113485257B (en) | 2021-06-02 | 2021-06-02 | Industrial protocol analysis built-in program optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113485257B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114153468A (en) * | 2021-11-04 | 2022-03-08 | 云镝智慧科技有限公司 | Industrial protocol adaptation method and device, edge device and cloud server |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018051108A1 (en) * | 2016-09-14 | 2018-03-22 | flexciton Limited | Automated supervisory control system |
| CN109325580A (en) * | 2018-09-05 | 2019-02-12 | 南京邮电大学 | A kind of adaptive cuckoo searching method for Services Composition global optimization |
| CN111340456A (en) * | 2020-03-06 | 2020-06-26 | 国网冀北电力有限公司 | Edge cloud collaborative data processing method, device and equipment based on Internet of things |
| CN111427664A (en) * | 2020-03-25 | 2020-07-17 | 杭州电子科技大学 | Container-based protocol analysis method, device, system and storage medium |
| CN112600891A (en) * | 2020-12-07 | 2021-04-02 | 中科蓝智(武汉)科技有限公司 | Edge cloud cooperation system based on information physical fusion and working method |
| CN112702405A (en) * | 2020-12-18 | 2021-04-23 | 太原理工大学 | Internet of things equipment identification method based on multi-protocol detection |
| CN113015942A (en) * | 2018-12-28 | 2021-06-22 | 西门子股份公司 | Data analysis method, device and system |
-
2021
- 2021-06-02 CN CN202110612738.3A patent/CN113485257B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018051108A1 (en) * | 2016-09-14 | 2018-03-22 | flexciton Limited | Automated supervisory control system |
| CN109325580A (en) * | 2018-09-05 | 2019-02-12 | 南京邮电大学 | A kind of adaptive cuckoo searching method for Services Composition global optimization |
| CN113015942A (en) * | 2018-12-28 | 2021-06-22 | 西门子股份公司 | Data analysis method, device and system |
| CN111340456A (en) * | 2020-03-06 | 2020-06-26 | 国网冀北电力有限公司 | Edge cloud collaborative data processing method, device and equipment based on Internet of things |
| CN111427664A (en) * | 2020-03-25 | 2020-07-17 | 杭州电子科技大学 | Container-based protocol analysis method, device, system and storage medium |
| CN112600891A (en) * | 2020-12-07 | 2021-04-02 | 中科蓝智(武汉)科技有限公司 | Edge cloud cooperation system based on information physical fusion and working method |
| CN112702405A (en) * | 2020-12-18 | 2021-04-23 | 太原理工大学 | Internet of things equipment identification method based on multi-protocol detection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113485257A (en) | 2021-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110191148B (en) | Statistical function distributed execution method and system for edge calculation | |
| EP3494522A1 (en) | Communication efficient federated learning | |
| CN111447083A (en) | Federal learning framework under dynamic bandwidth and unreliable network and compression algorithm thereof | |
| CN113114756A (en) | Video cache updating method for self-adaptive code rate selection in mobile edge calculation | |
| CN113485257B (en) | Industrial protocol analysis built-in program optimization method | |
| CN103095785B (en) | Remote procedure calling (PRC) method and system, client and server | |
| CN102035669A (en) | Function calling system and method | |
| CN111935025B (en) | Control method, device, equipment and medium for TCP transmission performance | |
| CN115796271A (en) | Federal learning method based on client selection and gradient compression | |
| CN103714184A (en) | Quick parallel processing method and device of remote-sensing images | |
| Idrees et al. | Energy-efficient Data Processing Protocol in edge-based IoT networks | |
| CN102325367A (en) | Data packet synchronizing device and method for client application | |
| Panem et al. | Data reduction and recovery in wireless communication system: An extensive experimental evaluation using PSK and QAM modulations | |
| Deligiannakis et al. | Dissemination of compressed historical information in sensor networks | |
| CN102244665B (en) | Bitmap management and node collection inquiry method, node and index server | |
| CN103259824A (en) | Method and device for synchronizing data route in distributed data storage system | |
| CN116546053A (en) | Edge computing service placement system in resource-constrained Internet of things scene | |
| US20210006639A1 (en) | Data acquisition method, service provider, service consumer and network functional entity | |
| CN109543123A (en) | A kind of short network address generation method, device, system, equipment and readable storage medium storing program for executing | |
| CN110381540B (en) | Dynamic cache updating method for responding popularity of time-varying file in real time based on DNN | |
| CN112001495B (en) | Neural network optimization method, system, device and readable storage medium | |
| US8547996B2 (en) | Self learning performance optimized data transfer via one or several communication channels between systems | |
| KR20200104959A (en) | Apparatus and method for compressing data in distributed deep-learning environment | |
| CN105515988A (en) | Distributed route synchronization method based on probability quantized broadcast gossip algorithm | |
| Abolhassani et al. | Optimal Push and Pull-Based Edge Caching For Dynamic Content |
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 |