CN115243303B - Deployment method, system and medium for edge computing device for spectrum monitoring - Google Patents
Deployment method, system and medium for edge computing device for spectrum monitoring Download PDFInfo
- Publication number
- CN115243303B CN115243303B CN202210878917.6A CN202210878917A CN115243303B CN 115243303 B CN115243303 B CN 115243303B CN 202210878917 A CN202210878917 A CN 202210878917A CN 115243303 B CN115243303 B CN 115243303B
- Authority
- CN
- China
- Prior art keywords
- edge computing
- remote monitoring
- monitoring station
- deployment
- computing device
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 316
- 238000001228 spectrum Methods 0.000 title claims abstract description 229
- 238000000034 method Methods 0.000 title claims abstract description 90
- 230000005540 biological transmission Effects 0.000 claims abstract description 113
- 230000008569 process Effects 0.000 claims abstract description 21
- 230000003595 spectral effect Effects 0.000 claims description 43
- 238000003860 storage Methods 0.000 claims description 23
- 238000004891 communication Methods 0.000 claims description 13
- 238000004088 simulation Methods 0.000 claims description 10
- 238000000411 transmission spectrum Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 description 6
- 238000013507 mapping Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 101100001674 Emericella variicolor andI gene Proteins 0.000 description 1
- 241000540325 Prays epsilon Species 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000035755 proliferation Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/18—Network planning tools
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/22—Traffic simulation tools or models
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Monitoring And Testing Of Transmission In General (AREA)
Abstract
The invention relates to a deployment method, a system and a medium of edge computing equipment for spectrum monitoring. When the edge computing equipment is deployed, according to different deployment schemes, the deployment cost generated by deploying all the edge computing equipment and the transmission cost for transmitting all the spectrum data are calculated, and finally, the lowest total cost is calculated according to the deployment cost and the transmission cost, and the optimal edge computing equipment deployment scheme is selected. According to the invention, the problem of time delay caused by uploading the spectrum data to the cloud computing center is solved at low cost by deploying the edge computing equipment to analyze and process the spectrum data.
Description
Technical Field
The application relates to the field of spectrum monitoring, in particular to a deployment method, a deployment system and a deployment medium of edge computing equipment for spectrum monitoring.
Background
With the rapid progress of communication and computer technology and the proliferation of wireless services, the number of various frequency-consuming devices has been exponentially increased, and the frequency spectrum resource crowding phenomenon caused by the unprecedented scale of the corresponding generated frequency spectrum data is also more serious in every corner of the society. In the prior art, as shown in fig. 1, a spectrum data transmission scene network includes a spectrum acquisition device 20, a remote monitoring station 10 and a cloud computing center 30, the remote monitoring station 10 is respectively in communication connection with the spectrum acquisition device 20 and the cloud computing center 30, the remote monitoring station 10 is closer to the spectrum acquisition device 20, the spectrum acquisition devices 20 distributed at each corner are used for acquiring spectrum data, the acquired spectrum data are transmitted to the remote monitoring station 10, the remote monitoring station 10 uploads the spectrum data acquired by the spectrum acquisition device 20 to the cloud computing center 30 after receiving the spectrum data, and the cloud computing center 30 analyzes and computes the spectrum data.
However, as the spectrum monitoring network construction scale increases, the number of spectrum acquisition devices increases exponentially, the amount of spectrum data acquired increases dramatically, and for large-scale spectrum data, because the cloud computing center 30 is far from the remote monitoring station 10, for remote communications, the cloud computing center 30 has network delay in data processing and high demands for bandwidth resources, so that real-time monitoring and analysis of large-scale spectrum data cannot be performed, and the limited computing and storage capabilities of the remote monitoring station 10 cannot provide powerful support.
Therefore, it is critical to solve the problems of network delay and occupied mass bandwidth resources caused by the data analysis processing mode using the cloud computing center 30 as the processing center in the face of the real-time requirements of large-scale data and spectrum monitoring generated by the terminal.
Disclosure of Invention
The application aims to solve the problems of network delay and occupied massive bandwidth resources caused by a data analysis processing mode taking a cloud computing center as a processing center in the prior art facing large-scale spectrum data, and provides a deployment method, a deployment system and a deployment medium of edge computing equipment for spectrum monitoring.
In order to solve the technical problems, the application provides a deployment method of edge computing equipment for spectrum monitoring, which comprises a remote monitoring station and spectrum acquisition equipment, wherein the remote monitoring station is in communication connection with the spectrum acquisition equipment, the edge computing equipment is deployed at least one remote monitoring station, the edge computing equipment is connected with the remote monitoring station, the remote monitoring station which is not deployed with the edge computing equipment transmits spectrum data acquired by the spectrum acquisition equipment connected with the edge computing equipment to the remote monitoring station which is deployed with the edge computing equipment, so that the edge computing equipment processes the spectrum data; the deployment method further comprises the following steps: acquiring the total number M and the position information of the remote monitoring stations, and obtaining the total amount Md i of the spectrum data capacity collected by all the remote monitoring stations by the maximum spectrum data capacity d i collected by each remote monitoring station; determining the total number N of edge computing devices and the storage capacity c k corresponding to each edge computing device according to the total amount Md i of the spectrum data, wherein the transmission distance of the edge computing devices capable of receiving the spectrum data is r; calculating deployment costs generated by deploying all edge computing devices; calculating transmission cost for transmitting all the spectrum data; calculating total cost according to deployment cost and transmission cost; wherein i=1, 2..m, k=1, 2..m, N, M is greater than or equal to N, M is the total number of remote monitoring stations and N is the total number of edge computing devices.
Preferably, the step of computing deployment costs for deploying all edge computing devices comprises: disposing the kth edge computing device e k at the jth monitoring station s j, denoted as z jk, and z jk = 1, the ith remote monitoring station s i may transmit spectral data to the edge computing device e k disposed at the jth remote monitoring station s j, if z jk = 0, the kth edge computing device e k not disposed at the jth remote monitoring station s j; determining a required deployment expense w jk of the kth edge computing device e k according to the deployment of the kth edge computing device e k at the jth remote monitoring station s j; determining deployment costs to deploy all edge computing devicesWhere j=1, 2,..m.
Preferably, the step of the ith remote monitoring station s i transmitting spectral data to the edge computing device e k deployed at the jth remote monitoring station s j may be preceded by: calculating the Euclidean distance h ij between the ith remote monitoring station s i and the jth remote monitoring station s j; comparing the Euclidean distance h ij with the transmission distance r; if h ij is less than or equal to r, the spectrum data can be transmitted; if h ij > r, spectral data cannot be transmitted.
Preferably, the step of calculating transmission costs for transmitting the entire spectrum data includes: according to the ith remote monitoring station s i, the spectrum data can be transmitted to the edge computing device e k deployed at the jth remote monitoring station s j, and the cost of transmitting the unit spectrum data is determined to be t ij; according to the ith remote monitoring station s i, transmitting spectrum data to edge computing equipment deployed at the jth remote monitoring station s j, setting the transmission proportion as y ij, and obtaining the transmission data capacity of the ith remote monitoring station s i for transmitting the spectrum data as d iyij, wherein y ij is more than or equal to 0 and less than or equal to 1; determining the transmission cost of transmitting all the spectrum data as
Preferably, the step of calculating the total cost from the deployment cost and the transmission cost further comprises: setting the weight occupied by the deployment expense in the total expense as beta, and setting the weight occupied by the transmission expense in the total expense as lambda, wherein beta+lambda=1; according to the deployment cost, the transmission cost and the weights occupied by the deployment cost and the transmission cost in the total cost, the objective function is obtained as follows: Wherein z= { z jk, 1.ltoreq.j.ltoreq.M, 1.ltoreq.k.ltoreq.N } represents a deployment scheme corresponding to all edge computing devices, y= { y ij, 1.ltoreq.i.ltoreq.M, 1.ltoreq.j.ltoreq.M } represents a spectrum data transmission allocation proportion among all remote monitoring stations; and solving the objective function by utilizing a local search method according to the objective function and the constraint conditions thereof.
Preferably, the constraint includes: transmitting spectral data when the euclidean distance h ij between the remote monitoring station s i and the remote monitoring station s j is less than the transmission distance r, wherein l ij =1, and when the euclidean distance h ij is greater than the transmission distance r, l ij =0, obtaining y ij≤lij, wherein l ij indicates whether the remote monitoring station s i can transmit spectral data to an edge computing device disposed at the remote monitoring station s j; the sum of the proportions of transmissions from any one remote monitoring station s i to the edge computing device deployed at the other remote monitoring station s j isThe edge computing device at any one of the remote monitoring stations s j cannot receive data capacity exceeding the computing and storage capacity available at the edge computing device, thenAny one edge computing device is deployed at most at one remote monitoring station, then
Preferably, the step of solving the objective function by using the local search method according to the objective function and the constraint condition thereof includes: let f=f (z, y), let the deployment scheme z corresponding to all edge computing devices be determined, and convert the objective function intoAnd the constraint becomes: s.t.0.ltoreq.y ij≤1,yij≤lij, Wherein, s.t.0.ltoreq.y ij.ltoreq.1 represents that the arbitrary proportion y ij of the transmission spectrum data is more than or equal to 0 and less than or equal to 1,Since z and C k (k=1, 2,., N) are known, C 1 can be regarded as a known constant value; let y=Ω (z), the objective function is converted into: f=f (z, Ω (z)), where Ω represents an expression based on known z and F y; establishing edge computing device deployment variables z jk andThe one-to-one correspondence between the elements p= (j, k) e P, letObtaining a function f 1 (U) =f (z, Ω (z)), wherein z jk =1 represents that element (j, k) is selected from P, resulting in an edge computing device deployment cost, z jk =0 represents that element (j, k) is not selected from P, no edge computing device deployment cost, and function f 1 represents the effect of U on the total cost; introductionTo equivalently convert the objective function f=f (z, Ω (z)) as an identification function toAnd is also provided withWherein, (j, k) e U represents that the kth edge computing device e k is deployed at the jth remote monitoring station s j, j e M, k e N, M being the total number of remote monitoring stations, N being the total number of edge computing devices; determining that the objective function is non-negative, whereinQ is a set of P subsets; after the deployment scheme z corresponding to the edge computing equipment is obtained, the spectrum data transmission allocation proportion y among all the remote monitoring stations is obtained according to a linear programming method, and a solution of an objective function is obtained.
Preferably, according to U is notObtain f 1 (U) is not less than 0, andThe objective function is non-negative where f 1 (U) is the total cost.
The invention further provides an online test spectrum data system based on edge computing equipment, which comprises spectrum acquisition equipment, a remote monitoring station and edge computing equipment, wherein the edge computing equipment is arranged on the remote monitoring station, and the remote monitoring station is respectively in communication connection with the spectrum monitoring equipment; the remote monitoring station collects spectrum data from the spectrum acquisition device and transmits the spectrum data to the edge computing device, and the edge computing device processes the spectrum data.
Still another embodiment of the present invention provides a readable storage medium, where a deployment program of an edge computing device is stored, and when the deployment program of the edge computing device is executed by a processor, the steps of the foregoing deployment method of the edge computing device for spectrum monitoring are implemented, so as to complete a simulation calculation of a deployment process of the edge computing device.
The beneficial effects are that: the method is based on the existing three-layer architecture consisting of spectrum acquisition equipment, remote monitoring stations and a cloud computing center, the total number M and position information of the remote monitoring stations and the maximum spectrum data capacity d i correspondingly collected by each remote monitoring station are obtained, the total number N of edge computing equipment, the storage capacity c k corresponding to each edge computing equipment and the transmission distance r for receiving spectrum data are determined; the edge computing device is deployed at an appropriate remote monitoring station. And the mapping relation between the frequency spectrum acquisition equipment and the remote monitoring station is not changed when the edge computing equipment is deployed, but the edge computing equipment is selected to be deployed at a certain remote monitoring station directly by utilizing the relatively fixed mapping relation between the designed remote monitoring station and the frequency spectrum acquisition equipment in the early stage of construction. The frequency spectrum acquisition devices distributed at all corners are used for acquiring frequency spectrum data, the acquired frequency spectrum data are transmitted to the remote monitoring station, and after the remote monitoring station receives the frequency spectrum data acquired by the frequency spectrum acquisition devices, the data are transmitted to the edge computing device, and the edge computing device is used for completing data transmission, storage and computation. When the edge computing equipment is deployed, according to different deployment schemes, the deployment cost generated by deploying all the edge computing equipment and the transmission cost for transmitting all the spectrum data are calculated, and finally, the lower total cost is calculated according to the deployment cost and the transmission cost, and the optimal edge computing equipment deployment scheme is selected. In the invention, as the deployment of the edge computing equipment is closer to the data source, the time delay caused by uploading the spectrum data to the cloud computing center can be greatly reduced, so that the received spectrum data can be processed and analyzed more quickly, and the decision can be made timely and quickly aiming at the current actual electromagnetic spectrum situation change.
Drawings
FIG. 1 illustrates a prior art cloud computing center-based spectrum data transmission scenario diagram;
FIG. 2 illustrates a large-scale spectral data transmission scenario diagram based on an edge computing device;
FIG. 3 illustrates a method flow diagram of a method of deploying an edge computing device;
FIG. 4 illustrates a comparison of costs required for different edge computing device deployment methods in a small scale scenario;
FIG. 5 illustrates a comparison of costs required for different edge computing device deployment methods in a medium-scale scenario;
FIG. 6 shows a comparison of the costs required for different edge computing device deployment methods in a slightly larger scale scenario.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The terms "comprising" and "having" and any variations thereof, as used in the description, claims and drawings, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or modules is not limited to only those steps or units listed but may alternatively include other steps or units not listed or inherent to such process, method, article, or apparatus. Furthermore, the terms "first," "second," and "third," etc. are used for distinguishing between different objects and not for describing a particular sequential order.
In order to better understand the above technical solutions, the following detailed description will refer to the accompanying drawings and specific embodiments.
Referring to fig. 2, fig. 2 is a schematic diagram of a large-scale spectrum data transmission scenario based on an edge computing device 100 as a processing center, as shown in the schematic diagram, in a spectrum data transmission scenario network, the spectrum data transmission scenario network includes a remote monitoring station 10 and a spectrum acquisition device 20, the remote monitoring station 10 is communicatively connected with the spectrum acquisition device 20, the remote monitoring station 10 is closer to the spectrum acquisition device 20, at least one remote monitoring station 10 is disposed with the edge computing device 100, the edge computing device 100 is communicatively connected with the remote monitoring stations 10, and the number of the edge computing devices 100 is less than or equal to the number of the remote monitoring stations 10. The spectrum acquisition devices 20 distributed at each corner are used for acquiring spectrum data and transmitting the acquired spectrum data to the remote monitoring station 10, the remote monitoring station 10 transmits the spectrum data to the edge computing device 100 deployed at one remote monitoring station 10 after receiving the spectrum data acquired by the spectrum acquisition devices 20, and the edge computing device 100 provides communication, calculation and storage guarantee. The remote monitoring station 10 with the edge computing equipment 100 is set as a control monitoring station, the remote monitoring station 10 without the edge computing equipment 100 is set as a transmission monitoring station, the transmission monitoring station sends the received spectrum data to the edge computing equipment 100 at the control monitoring station, the edge computing equipment 100 analyzes and calculates the spectrum data, and the transmission distance of the spectrum data is equal to the distance between the two remote monitoring stations 10. In other embodiments of the present invention, the spectrum data transmission scenario network further includes a cloud computing center 30, the remote monitoring stations 10 are respectively in communication connection with the spectrum acquisition device 20 and the cloud computing center 30, the remote monitoring stations 10 are closer to the spectrum acquisition device 20, the edge computing devices 100 are disposed at the remote monitoring stations 10, and the number of the edge computing devices 100 is less than or equal to the number of the remote monitoring stations 10. The spectrum acquisition devices 20 distributed at each corner are used for acquiring spectrum data and transmitting the acquired spectrum data to the remote monitoring station 10, after the remote monitoring station 10 receives the spectrum data acquired by the spectrum acquisition devices 20, the transmission monitoring station sends the received spectrum data to the edge computing device 100 at the control monitoring station, the edge computing device 100 analyzes and computes the spectrum data, and the cloud computing center 30 is responsible for more macroscopic, complex and non-real-time data mining tasks based on the strong computing and storage capacity. Because the deployment of the edge computing device 100 is closer to the data source, the time delay caused by the uploading of the spectral data to the cloud computing center 30 can be greatly reduced, thereby more quickly processing and analyzing the received spectral data to make timely and rapid decisions for the current actual electromagnetic spectrum situation changes.
In order to strengthen the supervision of the whole spectrum resources in the area, rapidly cope with the emergency such as various interferences, and the like, reclaim idle spectrum resources and improve the utilization efficiency of the wireless spectrum, the invention monitors the spectrum big data generated in the area in real time by deploying the edge computing equipment 100 at the remote monitoring station 10, and timely and comprehensively grasps the change condition of the spectrum situation.
How the edge computing device 100 is deployed at the remote monitoring station 10, the invention has the following embodiments:
Example 1
Referring to fig. 3, fig. 3 is a flowchart illustrating a deployment method of an edge computing device according to an embodiment of the invention. In this embodiment, the deployment method includes:
Step S10, obtaining the total number M and the position information of the remote monitoring stations and the maximum spectrum data capacity d i collected by each remote monitoring station, so as to obtain the total spectrum data capacity Md i collected by all the remote monitoring stations;
Before the edge computing device is deployed, in a spectrum data transmission scene network, as shown in fig. 1, the spectrum data transmission scene network comprises a spectrum acquisition device 20, a remote monitoring station 10 and a cloud computing center 30, the remote monitoring station 10 is respectively in communication connection with the spectrum acquisition device 20 and the cloud computing center 30, the remote monitoring station 10 is closer to the spectrum acquisition device 20, the spectrum acquisition devices 20 distributed at all corners are used for acquiring spectrum data and transmitting the acquired spectrum data to the remote monitoring station 10, the remote monitoring station 10 uploads the spectrum data acquired by the spectrum acquisition device 20 to the cloud computing center 30 after receiving the spectrum data, and the cloud computing center 30 analyzes and computes the spectrum data.
Thus, a plurality of spectrum acquisition devices 20 have been built within the area to be monitored prior to deployment of the edge computing device 100. Moreover, the use condition of the spectrum acquisition device 20 in the area monitored by the remote monitoring station 10 is considered in the early stage of construction, and a relatively fixed mapping relationship is formed between the remote monitoring station 10 and the spectrum acquisition device 20. The edge computing device 100 is deployed without changing the mapping between the spectrum acquisition device 20 and the remote monitoring station 10, but rather takes full advantage of this well-established basis, choosing to deploy the edge computing device 100 directly at a remote monitoring station 10.
Thus, according to the area to be controlled actually, the total number M of the remote monitoring stations 10 in the area can be determined, and according to the total number M of the remote monitoring stations 10, any one of the remote monitoring stations can be set to be s i (i=1, 2. Location information for any one of the remote monitoring stations s i may also be determined, and the maximum spectral data capacity d i (i=1, 2,., M), the total amount Md i of the spectrum data capacity collected by all the remote monitoring stations 10 can be obtained according to the total number M of the remote monitoring stations 10 and the maximum spectrum data capacity d i corresponding to any one of the remote monitoring stations s i.
Step S20, determining the total number N of the edge computing devices and the storage capacity c k corresponding to each edge computing device according to the total spectrum data capacity Md i, where the transmission distance of the edge computing devices capable of receiving spectrum data is r;
according to the total spectrum data capacity Md i collected by all the remote monitoring stations 10 obtained in step S10, the actual spectrum data management requirement of the current managed area can be determined, and further the total number N of the edge computing devices 100 and the storage capacity c k corresponding to each edge computing device 100 are determined, wherein the transmission distance of the edge computing devices 100 for receiving the spectrum data is r, the determination of the transmission distance r needs to consider the communication performance of the edge computing devices 100 and the requirement of the real spectrum monitoring application scene on real time, and the transmission distance r is ensured to be not greater than the product of the required real time response time and the transmission speed of the spectrum data. Further, the edge computing device 100 is set to an edge computing device e k according to the total number N of edge computing devices 100, where k=1, 2.
Step S30, calculating the deployment expense generated by deploying all the edge computing devices;
Before the edge computing device 100 is deployed, the topology structure of the remote monitoring stations 10 of the control area and the position information of each remote monitoring station 10 can be obtained, and the Euclidean distance h ij between the ith remote monitoring station s i and the jth remote monitoring station s j can be calculated according to the topology structure and the position information; if the kth edge computing device e k is deployed at the jth monitoring station s j, the distance between the ith remote monitoring station s i and the kth edge computing device e k is equal to the euclidean distance h ij between the ith remote monitoring station s i and the jth remote monitoring station s j, so that the euclidean distance between the two remote monitoring stations is calculated, and the distance from the ith remote monitoring station s i to the kth edge computing device e k can be obtained; the Euclidean distance refers to the true distance between two points in m-dimensional space, or the natural length of the vector (i.e., the distance of the point from the origin). The Euclidean distance in two and three dimensions is the actual distance between two points. Wherein i=1, 2..m, j=1, 2..m, k=1, 2..m, N, M is greater than or equal to N, M is the total number of remote monitoring stations and N is the total number of edge computing devices.
Comparing the Euclidean distance h ij with the transmission distance r of the edge computing device, if h ij is less than or equal to r, which means that the distance between the ith remote monitoring station s i and the jth remote monitoring station s j is less than or equal to the transmission distance of the edge computing device for receiving the spectrum data, the ith remote monitoring station s i can transmit the spectrum data to the edge computing device deployed at the jth remote monitoring station s j, which can be expressed as that the ith remote monitoring station s i can transmit the spectrum data to the jth remote monitoring station s j; if h ij > r, which indicates that the distance between the ith remote monitoring station s i and the jth remote monitoring station s j is greater than the transmission distance of the spectral data receivable by the edge computing device, the ith remote monitoring station s i may not transmit the spectral data to the edge computing device disposed at the jth remote monitoring station s j, which may be expressed as the ith remote monitoring station s i may not transmit the spectral data to the jth remote monitoring station s j; let l ij denote whether or not spectral data can be transmitted between the i-th remote monitoring station s i and the j-th remote monitoring station s j, if spectral data can be transmitted, l ij =1, and if spectral data cannot be transmitted, l ij =0.
According to the principle that the deployment cost of an edge computing device is proportional to the specific device capacity c k and the maximum spectrum data capacity d i collected by each remote monitoring station, the deployment cost required to deploy the kth edge computing device e k at the jth monitoring station s j is determined to be w jk (j=1, 2,..m, k=1, 2,..n), while the deployment cost is generated, and it is required to determine whether the kth edge computing device e k is deployed at the jth monitoring station s j.
From the foregoing steps, it can be seen that, according to the ith remote monitoring station s i, spectrum data can be transmitted to the jth remote monitoring station s j, the euclidean distance h ij between the ith remote monitoring station s i and the jth remote monitoring station s j is less than or equal to the transmission distance r of the edge computing device, and the kth edge computing device e k is disposed at the jth monitoring station s j. Let z jk denote whether the kth edge computing device e k is deployed at the jth monitoring station s j, if the kth edge computing device e k is deployed at the jth monitoring station s j, z jk =1, if the kth edge computing device e k is not deployed at the jth monitoring station s j, z jk =0; where k=1, 2.
Thus, based on whether the kth edge computing device e k is deployed at the jth remote monitoring station s j and the deployment overhead w jk required for the kth edge computing device e k to be deployed at the jth monitoring station s j, the deployment overhead of deploying the aggregate edge computing device is determined to be
Where i=1, 2..m, j=1, 2..m, k=1, 2..m, N, M is the total number of remote monitoring stations and N is the total number of edge computing devices.
Step S40, calculating transmission cost for transmitting all the frequency spectrum data;
As can be seen from step 30, the distance between the ith remote monitoring station s i and the jth remote monitoring station s j is less than or equal to the transmission distance over which the edge computing device can receive the spectrum data, and the kth edge computing device e k is deployed at the jth monitoring station s j, so that the transmission expense of the transmission spectrum data is likely to occur between the ith remote monitoring station s i and the jth remote monitoring station s j. Based on the calculated Euclidean distance h ij between the ith remote monitoring station s i and the jth remote monitoring station s j, the cost of transmitting unit frequency spectrum data between the ith remote monitoring station s i and the jth remote monitoring station s j is determined to be t ij according to the principle that the cost of transmitting unit data is proportional to the Euclidean distance.
The ith remote monitoring station s i transmits spectral data to the jth remote monitoring station s j, corresponding to a spectral data capacity d i at s i. The ith remote monitoring station s i transmits a certain proportion of spectrum data to edge computing equipment deployed at the jth remote monitoring station s j, the proportion is y ij, y ij is more than or equal to 0 and less than or equal to 1, and the capacity of the ith remote monitoring station s i for transmitting spectrum data to the jth remote monitoring station s j is d iyij;
Thus, the transmission cost of transmitting the entire spectrum data between the i-th remote monitoring station s i and the j-th remote monitoring station s j is determined to be based on the cost of transmitting the unit spectrum data between the i-th remote monitoring station s i and the j-th remote monitoring station s j being t ij, and the data capacity of the transmission of the i-th remote monitoring station s i to the j-th remote monitoring station s j being d iyij
Where i=1, 2..m, j=1, 2..m, M and M are the total number of remote monitoring stations and N is the total number of edge computing devices.
Step S50, calculating total cost according to the deployment cost and the transmission cost; wherein i=1, 2, M, k=1, 2, N, M is greater than or equal to N, M is the total number of the remote monitoring stations, and N is the total number of the edge computing devices.
Z jk represents whether the kth edge computing device e k is deployed at the jth remote monitoring station s j, y ij represents that the ith remote monitoring station s i transmits spectrum data with the proportion of y ij to the edge computing devices deployed at the jth remote monitoring station s j, z= { z jk, 1.ltoreq.j.ltoreq.m, 1.ltoreq.k.ltoreq.n } represents a deployment scheme corresponding to all edge computing devices, y= { y ij, 1.ltoreq.i.ltoreq.m, 1.ltoreq.j.ltoreq.m } represents a spectrum data transmission allocation proportion among all remote monitoring stations, the deployment cost of the edge computing devices occupies a weight beta in the total cost, the transmission cost of transmitting all spectrum data among the remote monitoring stations occupies a weight lambda in the total cost, and beta+lambda=1;
Therefore, according to the deployment cost and the transmission cost and the weights occupied by the deployment cost and the transmission cost in the total cost respectively, the objective function is obtained as follows:
Wherein i=1, 2..m, j=1, 2..m, k=1, 2..m, N, M is greater than or equal to N, M is the total number of remote monitoring stations and N is the total number of edge computing devices.
Further, the characteristics of the actual application scene need to be considered in solving the objective function, and constraint conditions are set as follows:
s.t.0≤yij≤1
yij≤lij
zjk∈{0，1}
Specifically, since y ij indicates that the ith remote monitoring station s i has transmitted spectral data in proportion of y ij to the edge computing device disposed at the jth remote monitoring station s j, there is 0.ltoreq.y ij.ltoreq.1. Moreover, if and only if the euclidean distance h ij between the i-th remote monitoring station s i and the j-th remote monitoring station s j is within the transmission distance r, i.e., h ij r, the i-th remote monitoring station s i may transmit spectral data to the edge computing device disposed at the j-th remote monitoring station s j, at which time there corresponds to l ij =1. If h ij > r, the ith remote monitoring station s i may not transmit spectral data to the edge computing device deployed at the jth remote monitoring station s j, i.e., l ij =0, at which time the spectral data transmission ratio y ij is 0. It can be seen that y ij≤lij holds in both cases. Furthermore, for any one remote monitoring station s i, the sum of the data rates transmitted to the other remote monitoring stations s j (i.e., with edge computing devices deployed) is 1, i.e In the specific application process, the spectrum data cannot be transmitted to one remote monitoring station without limitation, and the actual situation is that the total amount of the spectrum data transmitted to one remote monitoring station s j Cannot exceed the computing and storage capacity/>, which the edge computing device atThus there isIn addition, note that the edge computing device e k is only deployed at the j-th remote monitoring station s j and is not deployed at the j-th remote monitoring station s j, and thus the variable z jk is non-0, i.e., 1. Also because an edge computing device is deployed at most at a remote monitoring station, there is
Therefore, according to the deployment cost corresponding to different edge computing equipment deployment schemes, the transmission cost of the deployed spectrum data and various practical constraint conditions which need to be met when the edge computing equipment is deployed and the spectrum data is transmitted, the local search method is utilized to solve the objective function, namely to obtain a better solution, so that the total cost of deploying the edge computing equipment is lower.
Specifically, the objective function F is solved, that is, under the condition that all constraint conditions are met, a better deployment scheme is found in deployment schemes z corresponding to all edge computing devices, and a better transmission proportion scheme is found in a spectrum data transmission allocation proportion scheme y among all remote monitoring stations, which can be expressed as f=f (z, y). When the deployment scheme z variable corresponding to the edge computing equipment is determined, the objective function of the allocation proportion scheme y variable for the frequency spectrum data transmission among all the remote monitoring stations is determined by the objective functionBecomes:
The constraints also become accordingly:
s.t.0≤yij≤1
yij≤lij
Wherein, Since z and C k (k=1, 2,..n) are known, C 1 can be regarded as a known constant value. As can be seen from the objective function F y and its corresponding constraints, the spectral data transmission ratio y ij appears in the form of a first power. Thus, when the edge computing device deployment variable z is determined, the solution of the spectral data transmission scheme variable y is converted into a linear programming solution. Thus, when the edge computing device deployment variable z is known, the existing linear programming solution method can be used to obtain the spectrum data transmission scheme variable y.
For any one edge computing device e k, it may be deployed at any one remote monitoring station s j. Thus, using a traversal method, there are M N deployment schemes to deploy N edge computing devices at M remote monitoring stations. The traversal method is adopted in a large-scale application scene, and the corresponding operation amount is huge. For example, m=100 remote monitoring stations in a certain region need to be deployed with n=10 edge computing devices, a total of 100 10＝1020 schemes are adopted by using a traversal method, and an optimal solution of an objective function is found from the schemes in 100 10＝1020, so that the operation amount is too huge. Based on the method, the real-time requirement of spectrum monitoring is considered, and the objective function is solved based on local search.
In particular, the spectral data transmission scheme variable y is denoted as y=Ω (z), where Ω denotes the edge computing device deployment based on known variables z and zSolving for the expression of y.
Therefore, the objective function F to be solved can be further expressed as: f=f (z, Ω (z)), from the expression f=f (z, Ω (z)), the objective function F can be considered to be affected only by the edge computing device deployment variable z. Based on the above expression f (z, Ω (z)), further designTo establish a one-to-one correspondence between the variable z jk of whether the ith remote monitoring station s i transmits spectral data to the jth remote monitoring station s j and the element p= (j, k) ∈p in P.
If z jk =1, it indicates that element (j, k) is selected from P, i.e., kth edge computing device e k is deployed at jth remote monitoring station s j, whereas if z jk =0, it indicates that element (j, k) is not selected from P. From the foregoing, z jk = 1 represents that the kth edge computing device e k is deployed at the jth remote monitoring station s j, which results in a corresponding deployment cost. Further, the spectral data is transmitted to the j-th remote monitoring station s j only when the edge computing device is deployed at the j-th remote monitoring station s j. If forAll have z jk =0, which means that no edge computing device is deployed at the remote monitoring station s j, so that no other remote monitoring station can send spectrum data to s j, and the corresponding deployment cost and data transmission cost are both naturally 0. Thus, from the above analysis, it can be found that the value of z jk =1, which actually affects the edge computing device deployment costs and post-deployment data transmission costs, is substantially the kth edge computing device e k is deployed at the jth remote monitoring station s j. If z jk =0 is not costly for the corresponding edge computing device deployment and spectral data transmission.
Further, a design u= { (j, k) |z jk =1, j e M, k e N }, U representing a set of elements of the kth edge computing device e k deployed at the jth remote monitoring station s j. A function f 1 is introduced to represent the effect of U on the total cost, anThis can be achieved by:
f1(U):＝f(z,Ω(z))
Further, by introducing As an identification function, the objective function F can be equivalently converted into:
Where z jk =1 for any (j, k) e U, i.e. there is a kth edge computing device e k deployed at a jth remote monitoring station s j. Since each edge computing device is deployed at most at one remote monitoring station, for There isIt can be seen that after a series of equivalent transformations, the objective function F to be solved becomes F 1, while the critical factor affecting this function is the set U of values for z jk =1.
For function f 1, ifAnd f 1 (U) > 0,It holds that f 1 is non-negative, whereRepresenting empty set,Representing a total cost of 0,/>, when no edge computing device is deployedRepresenting that the edge computing equipment is deployed, and f 1 (U) is equal to or greater than 0, representing that the total cost is greater than or equal to 0 when the edge computing equipment is deployed. f 1 (U) represents the total cost, which must be non-negative as long as the edge computing device is deployed.
Specifically, the invention deploys N edge computing devices to ensure the transmission of spectrum data corresponding to M remote monitoring stations and the related analysis and computation. If U isI.e., no edge computing device is deployed, the deployment cost is 0, and the spectral data transmission cost is 0. Thus,This is true. Correspondingly, as long as U is notA corresponding deployment cost will result and the spectral data transmission cost cannot be negative. Therefore f 1 (U) > 0,This is true.
If for anyAnd an arbitrary element p.epsilon.P andI.e., P e (p\u 2), all ：f1{U1∪{p}}-f1(U1)≥f1(U2∪{p})-f1{U2} hold, so to speak f 1 is sub-modular. Conversely, if for anyAnd any element P e p\u 2, all satisfying ：f1{U1∪{p}}-f1(U1)≤f1(U2∪{p})-f1{U2} holds, so to speak f 1 is supermodular. The set U 1、U2 includes a plurality of elements with z jk =1, but the number of elements in the set U 1 is less than or equal to the number of elements in the set U 2, and the set U 1 is included in the set U 2.
Let p= (j 1,k1),p＝(j1,k1) denote the element in P, and from the analysis and equivalent transformation process of the objective function described above, it can be seen that f 1 (U) =f (z, Ω (z)) is
Is an equivalent transformation of (a).
Based on this, a setting is madeAfter the edge computing equipment is deployed according to U 1, a frequency spectrum data transmission and distribution scheme obtained by using a linear programming method is as follows:
Wherein, Representing a set of transmission ratios y ij for transmission of spectral data between the i-th remote monitoring station s i and the j-th remote monitoring station s j when the edge computing device is deployed in accordance with U 1.
Setting upAfter the edge computing equipment is deployed according to U 1∪{j1,k1, a frequency spectrum data transmission and distribution scheme obtained by using a linear programming method is as follows:
Wherein, Representing a set of transmission proportions y ij of transmission spectrum data between an ith remote monitoring station s i and a jth remote monitoring station s j when edge computing devices are deployed according to U 1∪{j1,k1;
Setting up After the edge computing equipment is deployed according to U 2, a frequency spectrum data transmission and distribution scheme obtained by using a linear programming method is as follows:
Wherein, Representing a set of transmission proportions y ij of transmission spectrum data between an ith remote monitoring station s i and a jth remote monitoring station s j when the edge computing device is deployed according to U 2;
Setting up After the edge computing equipment is deployed according to U 2∪{j1,k1, a frequency spectrum data transmission and distribution scheme obtained by using a linear programming method is as follows:
Wherein, Representing a set of transmission proportions y ij of transmission spectrum data between an ith remote monitoring station s i and a jth remote monitoring station s j when edge computing devices are deployed according to U 2∪{j1,k1;
From the above equation, it can be calculated that:
Therefore, only comparison is needed AndIt is known whether f 1 is overmoulded or sub-moulded. That is, only the magnitude of the spectral data transmission overhead variation after the point (j 1,k1) is increased in both cases U 1 and U 2, i.e. at the remote monitoring station/>, needs to be compared with the edge computing device deployment schemeEdge computing deviceAfter that, the influence on the cost of the frequency spectrum data transmission is achieved; where (j 1,k1) represents the kth 1 edge computing deviceDeployed at the j 1 th remote monitoring stationWhere it is located.
First, the change in the spectral data transmission costs after adding point (j 1,k1) when the edge computing deployment scenario is U 1. The change in spectral data transmission costs after adding the point (j 1,k1) is mainly reflected in two aspects. In one aspect, a portion of the remote monitoring station distance (j 1,k1) is relatively close to a point (j 2,k2) in U 1. Therefore, when the point (j 1,k1) is added, the remote monitoring stations change from transmitting the spectrum monitoring data to the point (j 2,k2) to transmitting the spectrum monitoring data to the point (j 1,k1), so that the transmission cost of the data is reduced. On the other hand, (j 2,k2) in original U 1 may be fully loaded, which may have to transmit spectral data farther (j 3,k3) in U 1 by some remote monitoring stations. Now after (j 1,k1) is added, (j 2,k2) some storage capacity is left freeThe remote monitoring station near (j 2,k2) naturally changes the transmission of the spectrum data from the original transmission to the transmission of the spectrum data at (j 3,k3) to the transmission of the spectrum data at (j 2,k2), and correspondingly, the transmission cost of the spectrum data is reduced. Thus, after adding point (j 1,k1) to U 1, the total spectral data transmission overhead is reduced, i.e., f 1(U1∪{j1,k1})-f1(U1 is negative. But due toIn addition to points (j 2,k2) and (j 3,k3) mentioned in U 1 above, there may be points (j 4,k4) in U 2 that are closer to some of the remote monitoring stations. Thus, after adding point (j 1,k1), the amount of data transmission overhead reduction is relatively small compared to that in U 1, and after adding point (j 1,k1), the total spectral data transmission overhead is reduced, i.e., f 1(U1∪{j1,k1})-f1(U1) and f 1(U2∪{j1,k1})-f1(U2), and therefore:
f1(U1∪{j1,k1})-f1(U1)≤f1(U2∪{j1,k1})-f1(U2)
i.e. f 1 is supermodular.
In the invention, N is set to be more than or equal to 2, namely at least 2 edge computing devices are required to be deployed at a remote monitoring station to provide communication, computation and storage guarantee. Based on the setting that N is more than or equal to 2, the practical constraint that each edge computing device is at most deployed at one remote monitoring station is further definedU 1＝(j1,k1) represents the kth 1 edge computing deviceDeployed at the j 1 th monitoring stationWhere u 2＝(j2,k2) represents the kth 2 edge computing deviceDeployed at the j 2 th remote monitoring stationWhere Q is the set of P subsets. For (P, Q), a (P, Q) is said to be a pseudo-array if the following two conditions are satisfied: 1) IfThen U ε Q; 2) If U, V.epsilon.Q and |U|.ltoreq|V|, the presence of element v.epsilon.V causes U.mu. { V }. Epsilon.Q.
Condition 1): assume thatThe second element, which necessarily has at least 2 points in U, is identical according to the definition of Q. And because ofSo the second element in V, which must also have at least 2 points, is the same, and this conflicts with V.epsilon.Q, so the assumption does not hold, i.e. ifThen U e Q, the first condition is satisfied. Further, condition 2): it is assumed that there is no element v satisfying the condition, that is, forLet d=u ≡ v }, according to the definition of QThe second element in D, which necessarily has at least 2 points, is identical. Also, because |U| is less than or equal to |V|, the second element of at least 2 points in V is the same, which conflicts with V ε Q, so the assumption is not true. Thus, when U, V ε Q, and |U| is equal to or less than |V|, the element V ε V is present such that U { V } ∈Q. It can be seen that the second condition is also satisfied. Thus, (P, Q) is a pseudo-array.
In summary, f 1 is non-negative, overmoded, and correspondingly designed (P, Q) is a pseudo-array, so that many existing targeted sub-optimization methods or local search methods such as a linear relaxation method can be used to efficiently find a solution of the edge computing device deployment scheme z, thereby greatly reducing the computational complexity of the solution method while ensuring the edge computing device deployment quality. After z is obtained, y can be obtained by solving a linear programming method such as a simplex method or a polynomial algorithm.
Based on the existing three-layer architecture consisting of a remote monitoring station, spectrum acquisition equipment and a cloud computing center, the total number M and position information of the remote monitoring station and the maximum spectrum data capacity d i which are correspondingly collected by each remote monitoring station are obtained, the total number N of edge computing equipment, the storage capacity c k which is corresponding to each edge computing equipment and the transmission distance r for receiving spectrum data are determined; the edge computing device is deployed at an appropriate remote monitoring station. And the mapping relation between the frequency spectrum acquisition equipment and the remote monitoring station is not changed when the edge computing equipment is deployed, but the edge computing equipment is selected to be deployed at a certain remote monitoring station directly by utilizing the relatively fixed mapping relation between the designed remote monitoring station and the frequency spectrum acquisition equipment in the early stage of construction. The frequency spectrum acquisition devices distributed at all corners are used for acquiring frequency spectrum data, the acquired frequency spectrum data are transmitted to the remote monitoring stations, after the remote monitoring stations receive the frequency spectrum data acquired by the frequency spectrum acquisition devices, the frequency spectrum acquisition devices transmit the frequency spectrum data to the edge computing devices deployed at a certain remote monitoring station, and the edge computing devices are used for completing data transmission, storage and computation. In the invention, as the deployment of the edge computing equipment is closer to the data source, the time delay caused by uploading the spectrum data to the cloud computing center can be greatly reduced, so that the received spectrum data can be processed and analyzed more quickly, and the decision can be made timely and quickly aiming at the current actual electromagnetic spectrum situation change.
Further, in order to verify the great benefits of using the efficient solution method based on local search, as shown in fig. 4 to 6, the present invention designs three different scenarios for simulation.
First kind: small scale. In this scenario, m=10 remote monitoring stations are designed, the maximum capacity of the corresponding spectrum monitoring data is between [17882, 56866], n=4 edge computing devices are designed, the corresponding computing and storage capacity is between [75000, 100000], and the maximum transmission distance r of the edge computing devices is set to be 5KM in the simulation process. The location information corresponding to each remote spectrum acquisition device 20 is shown in table 1:
TABLE 1
Second kind: medium scale. In this scenario, m=30 remote monitoring stations are designed, the maximum capacity of the corresponding spectrum monitoring data is between [3594, 92969], n=8 edge computing devices are designed, the corresponding computing and storage capacity is between [110000, 145000], and the maximum transmission distance r of the edge computing devices is set to be 5KM in the simulation process. The location information corresponding to each remote spectrum acquisition device 20 is shown in table 2:
TABLE 2
Third kind: slightly larger scale. In this scenario, m=50 remote monitoring stations are designed, the maximum capacity of the corresponding spectrum monitoring data is between [3594, 92969], n=15 edge computing devices are designed, the corresponding computing and storage capacity is between [80000, 150000], and the maximum transmission distance r of the edge computing devices is set to be 5KM in the simulation process. The location information corresponding to each remote spectrum acquisition device 20 is shown in table 3:
TABLE 3 Table 3
The traversal method refers to all possible M N deployment schemes to be tried once, so that the edge computing equipment deployment scheme with the minimum cost is searched, and the corresponding edge computing equipment deployment cost, spectrum data transmission cost and total cost are calculated. The random deployment rule refers to the random deployment of N edge computing devices at some remote monitoring stations. In order to reflect the performance of the random deployment method more objectively, in the actual simulation process, each expense corresponding to the random deployment method is an average value of 100 times of simulation. It should be noted that in medium-scale and slightly large-scale scenarios, since the computer memory used for simulation cannot support running the traversal method to find the optimal deployment scheme, only the local search method and the random deployment method are compared in fig. 5 and 6. As shown in fig. 4, in a small-scale scenario, a solution method based on a local search method is compared with an existing traversal method and a random deployment method. In a small scale scenario, the total cost of the traversal method corresponding to the deployment scheme is 101250, the proposed local search method 105560, and the random deployment method 128600. In this scenario, the efficient local search method is only 4.08% more expensive than the traversal method, while it is 17.62% less expensive than the random deployment method. In a medium-scale scenario, the total cost of the efficient local search method is 246270, and the random deployment rule is 624550. In a slightly larger scale scenario, the total cost corresponding to the efficient local search method is 429502, and the random deployment method is 699530. It can be seen that the advantages of the local search method are more pronounced in medium-scale and slightly larger-scale application scenarios, compared to the random deployment method, savings in costs 60.57% and 38.61%, respectively.
In conclusion, the method for solving the problems based on local search is very suitable for large-scale frequency spectrum data real-time monitoring by combining the characteristics of practical application scenes.
Example 2:
The embodiment provides a monitoring system of edge computing equipment based on spectrum monitoring, which realizes the steps of the deployment method of the edge computing equipment for spectrum monitoring in fig. 2, and constructs the monitoring system, and comprises a remote monitoring station 10 and spectrum acquisition equipment 20, wherein the edge computing equipment 100 is arranged on the remote monitoring station 10, and the remote monitoring station 10 is in communication connection with the spectrum monitoring equipment 20; the remote monitoring station 10 receives the spectral data collected from the spectrum acquisition device 20 and transmits the spectral data to the edge computing device 100, which edge computing device 100 processes the spectral data. In other embodiments of the present invention, the monitoring system further comprises a cloud computing center 30, the cloud computing center 30 being communicatively connected to the remote monitoring station 10; the remote monitoring station 10 transmits a part of the spectrum data to the edge computing device 100, the edge computing device 100 processes the spectrum data, the remote monitoring station 10 transmits the rest of the spectrum data to the cloud computing center 30, the cloud computing center 30 processes the received spectrum data, and the rest of the spectrum data usually needs a longer processing time or is complex in data and is preferentially processed by the cloud computing center.
Example 3:
The embodiment provides a readable storage medium, on which a deployment program of an edge computing device is stored, where when the deployment program of the edge computing device is executed by a processor, an instruction as shown in fig. 3 is implemented, so as to complete a simulation calculation of a deployment process of the edge computing device. In particular, the deployment process and the beneficial effects of the edge computing device refer to the foregoing embodiments, and are not described herein.
The present application is not limited to the above-mentioned embodiments, but is intended to be limited to the following embodiments, and any modifications, equivalent changes and variations in the above-mentioned embodiments can be made by those skilled in the art without departing from the scope of the present application.
Claims (5)
1. A deployment method of edge computing equipment for spectrum monitoring, which is characterized by comprising deployment of a remote monitoring station and spectrum acquisition equipment, wherein the remote monitoring station is in communication connection with the spectrum acquisition equipment, edge computing equipment is deployed at least one remote monitoring station, the edge computing equipment is connected with the remote monitoring station, the remote monitoring station which is not deployed with edge computing equipment transmits spectrum data acquired by the spectrum acquisition equipment connected with the edge computing equipment to the remote monitoring station which is deployed with the edge computing equipment for the edge computing equipment to process the spectrum data;
the deployment method further comprises the following steps:
Acquiring the total number M and the position information of the remote monitoring stations and the maximum spectrum data capacity d i collected by each remote monitoring station correspondingly, and obtaining the total spectrum data capacity Md i collected by all the remote monitoring stations;
Determining the total number N of the edge computing devices and the storage capacity c k corresponding to each edge computing device according to the total spectrum data capacity Md i, wherein the transmission distance of the edge computing devices capable of receiving spectrum data is r;
calculating deployment costs generated by deploying all the edge computing devices;
Calculating transmission cost for transmitting all the spectrum data;
calculating total cost according to the deployment cost and the transmission cost;
Calculating the Euclidean distance h ij between the ith remote monitoring station s i and the jth remote monitoring station s j;
comparing the Euclidean distance h ij with the transmission distance r;
If h ij is less than or equal to r, the spectrum data is transmitted;
if h ij > r, the spectral data is not transmitted;
Transmitting spectrum data to edge computing equipment e k deployed at the j-th remote monitoring station s j according to the i-th remote monitoring station s i, and determining that the cost of transmitting unit spectrum data is t ij;
According to the ith remote monitoring station s i, transmitting spectrum data to edge computing equipment deployed at the jth remote monitoring station s j, setting the transmission proportion as y ij, and obtaining the data size of the ith remote monitoring station s i transmitting spectrum data as d iyij, wherein y ij is more than or equal to 0 and less than or equal to 1;
Determining the transmission cost of transmitting all the frequency spectrum data as
Setting the weight occupied by the deployment cost in the total cost as beta, and setting the weight occupied by the transmission cost in the total cost as lambda, wherein beta+lambda=1;
According to the deployment cost, the transmission cost and the weights occupied by the deployment cost and the transmission cost in the total cost respectively, an objective function is obtained as follows: Wherein z= { z jk, 1.ltoreq.j.ltoreq.M, 1.ltoreq.k.ltoreq.N } represents a deployment scheme corresponding to all the edge computing devices, y= { y ij, 1.ltoreq.i.ltoreq.M, 1.ltoreq.j.ltoreq.M } represents a spectrum data transmission allocation proportion among all the remote monitoring stations;
According to the objective function and the constraint conditions thereof, solving the objective function by utilizing a local search method;
transmitting spectral data when a euclidean distance h ij between the remote monitoring station s i and the remote monitoring station s j is smaller than the transmission distance r, corresponding to l ij =1, and if the euclidean distance h ij is greater than the transmission distance r, l ij =0, obtaining y ij≤lij, where l ij indicates whether the remote monitoring station s i transmits spectral data to an edge computing device disposed at the remote monitoring station s j;
The sum of the proportions of transmissions from any one remote monitoring station s i to the edge computing device deployed at the other remote monitoring station s j is
The data capacity received by the edge computing device at any one of the remote monitoring stations s j cannot exceed the computing and storage capacity provided at that edge computing device
Any one of the edge computing devices is deployed at most at one remote monitoring station, then
Let f=f (z, y), let the deployment scheme z corresponding to all edge computing devices be determined, and convert the objective function intoAnd the constraint is: s.t.0 is less than or equal to y ij≤1,yij≤lij,Wherein, s.t.0 is less than or equal to y ij is less than or equal to 1, and y ij is more than or equal to 0 and less than or equal to 1,/>, which represents any proportion of transmission spectrum dataSince z and C k (k=1, 2,., N) are known, C 1 is considered to be a known constant value;
let y=Ω (z), the objective function is converted into: f=f (z, Ω (z)), where Ω represents a process of solving for y based on the expression of z and F y that are known;
establishing the edge computing device deployment variables z jk and The one-to-one correspondence between the elements p= (j, k) e P, letObtaining a function f 1 (U) =f (z, Ω (z)), wherein z jk =1 represents that element (j, k) is selected from P, resulting in an edge computing device deployment cost, z jk =0 represents that element (j, k) is not selected from P, and no edge computing device deployment cost, and wherein the function f 1 represents the effect of U on the total cost;
Introduction of To equivalently convert the objective function f=f (z, Ω (z)) as an identification function toAndWherein, (j, k) e U represents that the kth edge computing device e k is deployed at the jth remote monitoring station s j, j e M, k e N, M being the total number of the remote monitoring stations, N being the total number of the edge computing devices;
determining that the objective function is non-negative, wherein,
Q is a set of P subsets;
after a deployment scheme z corresponding to the edge computing equipment is obtained, obtaining the frequency spectrum data transmission allocation proportion y among all the remote monitoring stations according to a linear programming method, and obtaining a solution of the objective function;
Wherein i=1, 2, M, k=1, 2, N, M is greater than or equal to N, M is the total number of the remote monitoring stations, and N is the total number of the edge computing devices.
2. The method of deployment of edge computing devices for spectrum monitoring of claim 1, wherein the step of computing deployment costs incurred by deploying all of the edge computing devices comprises:
Disposing a kth edge computing device e k at a jth monitoring station s j, denoted as z jk, and z jk = 1, then an ith remote monitoring station s i transmitting spectral data to an edge computing device e k disposed at the jth remote monitoring station s j, the kth edge computing device e k not disposed at the jth remote monitoring station s j if z jk = 0;
Determining a required deployment expense w jk of the kth edge computing device e k according to the deployment of the kth edge computing device e k at the jth remote monitoring station s j;
determining deployment costs to deploy all edge computing devices
Where j=1, 2,..m.
3. The deployment method of edge computing devices for spectrum monitoring of claim 2, wherein according to U is notObtain f 1 (U) is not less than 0, andThe objective function is non-negative where f 1 (U) is the total cost.
4. A monitoring system of an edge computing device based on spectrum monitoring, characterized in that the steps of implementing the deployment method of the edge computing device for spectrum monitoring according to any one of claims 1-3 are implemented, the monitoring system is constructed, and the monitoring system comprises a spectrum acquisition device, a remote monitoring station and an edge computing device, the edge computing device is arranged on the remote monitoring station, and the remote monitoring station is respectively in communication connection with the spectrum acquisition device; the remote monitoring station collects spectrum data from the spectrum acquisition device and transmits the spectrum data to the edge computing device, and the edge computing device processes the spectrum data.
5. A readable storage medium, wherein a deployment program of an edge computing device is stored on the readable storage medium, and when the deployment program of the edge computing device is executed by a processor, the steps of the deployment method of the edge computing device for spectrum monitoring according to any one of claims 1-3 are implemented, and simulation calculation of the deployment process of the edge computing device is completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210878917.6A CN115243303B (en) | 2022-07-25 | 2022-07-25 | Deployment method, system and medium for edge computing device for spectrum monitoring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210878917.6A CN115243303B (en) | 2022-07-25 | 2022-07-25 | Deployment method, system and medium for edge computing device for spectrum monitoring |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115243303A CN115243303A (en) | 2022-10-25 |
| CN115243303B true CN115243303B (en) | 2024-05-07 |
Family
ID=83675855
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210878917.6A Active CN115243303B (en) | 2022-07-25 | 2022-07-25 | Deployment method, system and medium for edge computing device for spectrum monitoring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115243303B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013185602A1 (en) * | 2012-06-13 | 2013-12-19 | 中兴通讯股份有限公司 | Capacity planning method and device for wireless broadband network |
| CN111831445A (en) * | 2020-07-13 | 2020-10-27 | 上海交通大学 | Edge server optimized deployment method |
| WO2022121097A1 (en) * | 2020-12-07 | 2022-06-16 | 南京邮电大学 | Method for offloading computing task of mobile user |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9098344B2 (en) * | 2011-12-27 | 2015-08-04 | Microsoft Technology Licensing, Llc | Cloud-edge topologies |
-
2022
- 2022-07-25 CN CN202210878917.6A patent/CN115243303B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013185602A1 (en) * | 2012-06-13 | 2013-12-19 | 中兴通讯股份有限公司 | Capacity planning method and device for wireless broadband network |
| CN111831445A (en) * | 2020-07-13 | 2020-10-27 | 上海交通大学 | Edge server optimized deployment method |
| WO2022121097A1 (en) * | 2020-12-07 | 2022-06-16 | 南京邮电大学 | Method for offloading computing task of mobile user |
Non-Patent Citations (2)
| Title |
|---|
| 无线电监测站部署效率评估与优化方法研究;徐英;周尚武;丁锋;;传感技术学报;20180915(09);全文 * |
| 移动边缘计算卸载技术综述;谢人超;廉晓飞;贾庆民;黄韬;刘韵洁;;通信学报;20181125(11);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115243303A (en) | 2022-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109413724B (en) | MEC-based task unloading and resource allocation scheme | |
| CN108541027B (en) | Communication computing resource replacement method based on edge cloud network | |
| CN109814951B (en) | Joint optimization method for task unloading and resource allocation in mobile edge computing network | |
| CN112020103B (en) | Content cache deployment method in mobile edge cloud | |
| CN111372314A (en) | Task unloading method and task unloading device based on mobile edge computing scene | |
| CN110928654A (en) | Distributed online task unloading scheduling method in edge computing system | |
| Shan et al. | A survey on computation offloading for mobile edge computing information | |
| CN111328133B (en) | V2X resource allocation method based on deep neural network | |
| CN110096362A (en) | A kind of multitask discharging method based on Edge Server cooperation | |
| CN112416603B (en) | Combined optimization system and method based on fog calculation | |
| CN116260720A (en) | Network resource deployment method and device, electronic equipment and storage medium | |
| CN113556764A (en) | Method and system for determining calculation rate based on mobile edge calculation network | |
| CN106792995B (en) | User access method for guaranteeing low-delay content transmission in 5G network | |
| CN114222371A (en) | Flow scheduling method for coexistence of eMBB (enhanced multimedia broadcast/multicast service) and uRLLC (unified radio link control) equipment | |
| El Haber et al. | Computational cost and energy efficient task offloading in hierarchical edge-clouds | |
| Wang et al. | Energy conserved computation offloading for O-RAN based IoT systems | |
| CN115243303B (en) | Deployment method, system and medium for edge computing device for spectrum monitoring | |
| Li | Optimization of task offloading problem based on simulated annealing algorithm in MEC | |
| CN115865930A (en) | MEC dynamic adjustment method based on 5G power Internet of things | |
| CN114363803A (en) | Energy-saving multi-task allocation method and system for mobile edge computing network | |
| CN117998420A (en) | DAG-based fragmented cascading distributed 6G network joint optimization method and device | |
| KR102056894B1 (en) | Dynamic resource orchestration for fog-enabled industrial internet of things networks | |
| CN112835710B (en) | Power line state analysis system based on mobile edge calculation and task processing method | |
| CN114637552B (en) | Fog computing task unloading method based on fuzzy logic strategy | |
| CN115134364A (en) | Energy-saving calculation unloading system and method based on O-RAN internet of things system |
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 |