CN116582949A - Resource scheduling method integrating security decision and calculation acceleration - Google Patents

Resource scheduling method integrating security decision and calculation acceleration Download PDF

Info

Publication number
CN116582949A
CN116582949A CN202310617535.2A CN202310617535A CN116582949A CN 116582949 A CN116582949 A CN 116582949A CN 202310617535 A CN202310617535 A CN 202310617535A CN 116582949 A CN116582949 A CN 116582949A
Authority
CN
China
Prior art keywords
representing
data
task
edge
power
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310617535.2A
Other languages
Chinese (zh)
Inventor
杨鹏
方程
易梦
杜淼
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southeast University
Original Assignee
Southeast University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Southeast University filed Critical Southeast University
Priority to CN202310617535.2A priority Critical patent/CN116582949A/en
Publication of CN116582949A publication Critical patent/CN116582949A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/535—Allocation or scheduling criteria for wireless resources based on resource usage policies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/03—Protecting confidentiality, e.g. by encryption
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/54—Allocation or scheduling criteria for wireless resources based on quality criteria
    • 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
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Quality & Reliability (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The application discloses a resource scheduling method integrating security decision and calculation acceleration, which comprises the following steps: step 1: calculating the processing time and energy consumption of a local task; step 2: calculating delay and energy consumption of a communication process; step 3: calculating QoE demand parameters of a user; step 4: a security decision for the user is determined. Step 5: and optimizing and solving a resource scheduling strategy. According to the method, the resource allocation, the task unloading, the QoE requirements of the user and the data security are comprehensively considered, the data security of the task unloading process is enhanced, the calculation acceleration is used as an index for quantifying the QoE requirements, the method is further enabled to be more fit with the real experience process of the user, the increasingly sensitive experience quality requirements of the user are met, and the trade-off among the four time delay, the energy consumption, the QoE and the data security is further sought.

Description

Resource scheduling method integrating security decision and calculation acceleration
Technical Field
The application relates to a resource scheduling method integrating security decision and calculation acceleration, belonging to the field of Internet and edge calculation.
Background
In recent years, with mobile internet, especially 5G technology support, internet of things (IoT) has developed rapidly, and has been widely used in various aspects of production and living, such as virtual reality/augmented reality, mobile medical treatment, etc. These traffic scenarios not only require very low, deterministic network latency, but also require massive, heterogeneous, and diverse data access. The traditional centralized computing processing mode of cloud computing faces huge computing and network pressure, and cannot meet the application requirements of everything interconnection. In this context, the edge calculation mode becomes an effective solution. In the edge computing mode, in order to better support high-density, large-bandwidth and low-delay service scenes, a more effective mode is to construct a service platform (edge data center) at the network edge close to a user, provide storage, computing, network and other resources, sink part of key service applications to the edge of an access network, and reduce bandwidth and delay loss caused by network transmission and multistage forwarding.
In edge computing, there are not only static terminal devices (e.g., sensors in smarthouses, cameras in public places) but also dynamic terminal devices (e.g., drones and vehicles, etc.), which makes resource management more challenging. And proper scheduling of resources may alleviate this situation. If the resource scheduling cannot effectively utilize the scattered resources, the utilization rate of the practically available resources is too low, and thus problems of high delay, high energy consumption and the like are caused. These problems are unacceptable for new applications that are computationally intensive and delay sensitive. If an efficient resource scheduling strategy is applied, the resources can be effectively combined together to establish an available and economical and efficient computing resource pool, thereby solving the resource management problem.
Thus, considering the importance of resource scheduling, some work begins to optimize the weighted sum of latency and energy consumption of edge computing systems and study the trade-off relationship between the two. However, in a practical scenario, besides latency and energy consumption, the resource scheduling process also needs focused factors such as quality of experience (QoE) and data security. QoE requirements are indicators that reflect the user's real experience. More and more users currently begin to pay attention to their subjective feelings, their QoE requirements are sensitive, and the low latency requirements do not fully cover the QoE requirements of the users, thus encouraging attention to this index during resource scheduling.
In addition, the QoE requirement is closely related to the data security, so that an attacker can easily insert interference factors into unencrypted task data to achieve the purpose of destroying the good experience of users, and wireless interference between MUs (user equipment) can also unintentionally reduce the QoE index of the users due to sharing wireless channels, so that the integrity of the data can be ensured by encrypting the task data. Therefore, it is also necessary to consider the element of data security while taking QoE into consideration. Currently, while there have been some efforts focused on four factors of latency, energy consumption, qoE and data security, most studies are joint analysis of two to three of them and they do not sufficiently focus on the tight coupling relationship between QoE and data security, and related studies do not combine these four factors well for analysis. The trade-off between four factors of latency, energy consumption, qoE requirements and data security is in need of solution.
Disclosure of Invention
In order to solve the problems and the shortcomings in the prior art, the application provides a resource scheduling method integrating security decision and calculation acceleration, which comprehensively considers the factors such as resource allocation, calculation unloading, data security, qoE and the like, aims at minimizing the cost of the whole system, and seeks the trade-off among time delay, energy consumption, qoE, data security and the like.
In order to achieve the above object, the technical scheme of the present application is as follows: a resource scheduling method integrating security decision and calculation acceleration comprises the following steps:
step 1: and calculating the processing time and the energy consumption of the local task. Obtaining the CPU core number n of different mobile users MUi i And per-core processing power f i l Re-calculating the power of each coreWherein the calculation time of the serialized part is then obtained based on Amdahl's law +.>And calculation time of parallelization part +.>Thereby the total local calculation time +.>And total local calculation energy consumption, i.e. +.>Wherein T is i Representing each MUi generated task, c i Representing T i The treatment density of D i Representing the size, alpha, of task data generated by MUi i Representing T i Parallel score, lambda i Unloading decision variable k representing task Ti generated by MUi 1 Is a coefficient reflecting the relationship between the processing power and the power consumption of the user mobile equipment.
Step 2: and calculating the delay and the energy consumption of the communication process. Wireless transmission model based on shannon formula, and representing data transmission rate r according to channel power gain and channel noise power density i . Wherein let P i Is the transmission power, h, of the mobile user equipment MUi i Is the channel power gain, ω 0 Representing the channel noise power density and B represents the radio bandwidth. Then calculate the transmission delayAnd transmission energy consumption->Then calculate edge processing time +.>And edge processing energy consumption->Wherein m is i Represented as edge server notThe number of cores allocated to the task processing generated by MUi, f i e Representing the processing power (CPU frequency in cycles/second) per core of the edge server assigned to MUi, k 2 Is a coefficient reflecting the relationship between the edge-side processing power and power consumption. Finally giving the transmission delay of the result return +.>And corresponding energy consumption->Wherein gamma is i Data size and T representing return result i The ratio of the initial task data size of P 0 Is the transmission power of the edge server.
Step 3: and calculating the QoE demand parameters of the user. Firstly, a mobile user sends a reconnaissance packet with small data volume to a base station, then the base station returns result information, after the two operations, the CPU core number of the mobile user equipment and the CPU core number of an edge server are obtained, and then the local calculation acceleration is calculated according to Amdahl lawEdge computing acceleration value ++>Wherein alpha is i Representing T i Can be parallel fractional, m i Representing the number of cores allocated for the edge server to process tasks generated by different MUs, n i Representing the number of CPU cores of the different mobile users MU i.
Step 4: a security decision for the user is determined. First S is conducted i E {0,1} represents each mobile user MU i Is made by each MU personalized according to the privacy requirements of the application data. Wherein S is i =0 denotes a mobile terminal device MU i The computing task is unloaded in an unencrypted manner; and S is i =1 indicates the mobile terminal device MU i Will be atThe ChaCha20 encryption technique is used to encrypt the computing tasks and their data prior to transmission to the edge server. After receiving the task and the data, the edge server further decrypts the data, then performs the calculation task and sends the processed result back to the MU i 。
Step 5: and optimizing and solving a resource scheduling strategy. Specifically, according to the first four steps, the total cost of the MU for locally executing all tasks is solvedAnd the total overhead of executing all tasks on edge servers +.>Then setting the weighting parameter w of the execution time and the energy consumption 1 And w is equal to 2 Wherein w is 1 ,w 2 ∈[0,1]. And let w 1 +w 2 =1. When the targets are different, the set weight values are changed correspondingly. After which a question definition is established, with the aim of minimizing the cost of the system +.>Thereafter, the optimal solution of the above objective equation is solved using a branch-and-bound method, the optimal solution being used to offload the decision set O (λ I )＝{λ 1 ,...,λ N Form representation. And finally, taking average value as a final return result after addition through multiple rounds of tests.
Compared with the prior art, the application has the following advantages:
1) Aiming at the multi-user resource scheduling problem of the edge computing system, a mixed integer linear programming method is adopted to comprehensively consider the factors such as resource allocation, calculation unloading, safety, qoE and the like, and the system realization cost can be minimized by integrating the safety decision and the calculation acceleration resource scheduling method to solve.
2) The ChaCha20 encryption technology is introduced as a security decision, so that task data can be prevented from being maliciously stolen by an attacker in the unloading process.
3) The calculation acceleration is used as an index for quantifying QoE requirements, so that the service experience requirements of users can be reflected more truly. Meanwhile, the quantitative index of calculation acceleration can be well coupled with the security decision, so that the resource scheduling method integrating the calculation acceleration and the security decision can well balance time delay, energy consumption, qoE requirements and data security.
Drawings
FIG. 1 is an overall flow chart of an embodiment of the present application.
Fig. 2 is a schematic diagram of an embodiment of the present application.
FIG. 3 is a computational graph of the present application implementing the optimization objective Obj.
Detailed Description
The application is further illustrated below in conjunction with specific embodiments in order to enhance the understanding and appreciation of the application.
Example 1: a resource scheduling method integrating security decision and calculation acceleration includes the steps that firstly, whether a task is to be processed locally or not; then, referring to fig. 2, the proposed three-layer model is adopted as an overall architecture, then the multi-user resource scheduling problem is abstracted into a mixed integer linear programming problem, the defined optimization problem Obj is solved according to a branch-and-bound method, the process of calculating Obj is referred to fig. 3, and finally, the set constraint conditions are combined to obtain the optimal solution of the optimization problem. The detailed implementation steps are as follows:
step 1: calculating the processing time and energy consumption of the local task to obtain the CPU core number n of the MU i of different mobile users i And per-core processing power f i l Re-calculating the power of each coreWherein the calculation time of the serialized part is then obtained based on Amdahl's law +.>And computation time of parallelized partTo derive the total local computationMeta->And total local calculation energy consumption, i.e. +.>Wherein T is i Representing each MUi generated task, c i Representing T i The treatment density of D i Representing the size, alpha, of task data generated by MUi i Representing T i Parallel score, lambda i Unloading decision variable k representing task Ti generated by MUi 1 Is a coefficient reflecting the relationship between the processing power and the power consumption of the user mobile equipment.
Step 2: calculating delay and energy consumption of communication process, wireless transmission model based on shannon formula, and representing data transmission rate r according to channel power gain and channel noise power density i The method comprises the steps of carrying out a first treatment on the surface of the The transmission delay and transmission energy consumption are then calculated. Calculating edge processing time and edge processing energy consumption; and finally, giving the transmission delay and the corresponding energy consumption of the result return. The method comprises the following steps: the implementation is divided into the following sub-steps:
substep 2-1: the data transmission process is generally constructed based on shannon's formula, the data transmission rate needs to have wireless bandwidth parameter B, and the transmission power P of the mobile user equipment MU i i Channel power gain h i And channel noise power density omega 0 The data transmission rate can be expressed by the above parameters as:
equation (1) represents the transmission rate of data as it is offloaded from the mobile user equipment to the edge over the designated wireless bandwidth B.
Substep 2-2: task T generated by MUi i Is to be used for unloading transmission delay: based on the analysis, lambda was calculated i D i The transfer delay of bit data offloading to the edge can be throughObtained in the following manner:
wherein lambda is i An unload decision variable, D, representing task Ti generated by MUi i The task data size generated by MU i is represented.
Transmission energy consumption consumed by the task offloading process: lambda of transmission unloading i D i The energy consumption of the terminal device of the bit data is expressed as:
wherein lambda is i Task T representing MUi generation i Is a decision variable of unloading of D i Representing the size, P, of task data generated by MUi i Representing the transmission power of the mobile user equipment MU i.
Substep 2-3: computing time required by the edge server to process tasks: lambda (lambda) i D i After the bit data is offloaded to the edge, the edge will process the data, let m be i Representing the number of cores allocated for the edge server to process tasks generated by different MUs i, f i e Representing the processing power (CPU frequency in cycles/second) per core allocated to the edge server for MU i,
f i e ＞＞f i l (16)
the power consumption of each core of edge processing data can be expressed as:
wherein k is 2 Is a coefficient reflecting the relationship between edge side processing power and power consumption, unloaded lambda i D i The computation time of the bit data includes the computation time of the serialization portion and the computation time of the parallelizable portion, the computation time of the serialization portion being expressed as:
c i representing T i The treatment density of D i Representing the size, alpha, of task data generated by MUi i Representing T i Parallel score, lambda i Task T representing MUi generation i Is a decision variable for offloading.
The computation time of the parallel portion can be expressed as:
m i representing the number of cores allocated for the edge server to process tasks generated by different MUs i, f i e Representing the processing power (CPU frequency in cycles/second) per core allocated to the edge server for MU i. The calculation time required for processing tasks by the edge server can be expressed as:
calculation time sum representing the serialization part +.>Representing the computation time of the parallelizable portion.
Substep 2-4: energy consumption consumed by the edge server to process tasks: calculation data lambda i D i The edge energy consumption formula of the bits is as follows:
wherein k is 2 Is a coefficient reflecting the relationship between the edge side processing power and power consumption, c i Representing T i Alpha, alpha i Representing T i Parallel score, lambda i Task T representing MUi generation i Is a decision variable for offloading. f (f) i e Representing the processing power of each core assigned to the edge server for MU i.
Substep 2-5: and (5) returning a result: task T i After the processing is finished, the result is returned to the mobile terminal equipment, r is set i ' indicates the data transmission rate in the result return process, P 0 Is the transmission power of the edge server, the data transmission rate needs to have a wireless bandwidth parameter B, and the channel power gain h i And channel noise power density omega 0 . R is similar to the offloaded data transfer rate i ' can also be formulated as:
transmission delay of the result return: based on the above analysis, γ i D i The transmission delay of the bit result return can be expressed as:
wherein gamma is i Data size and T representing return result i Is a ratio of the initial task data size of (a) to (b).
Thus, gamma of the result of the processing i D i The edge server power consumption of the bit transfer to the mobile terminal device is expressed as:
many research efforts ignore the return process because of the small values of the processing results. Different from the work, the patent pays more attention to the returning process of the result, pays more attention to the information carried by the returning value of the result, and can reflect the real wireless communication process.
Step 3: calculating QoE demand parameters of users, firstly, transmitting a reconnaissance packet with small data volume to a base station by a mobile user, then, returning result information by the base station, obtaining the CPU core number of mobile user equipment and the CPU core number of an edge server after the two operations, and then, respectively calculating local calculation acceleration according to Amdahl lawEdge computing acceleration value ++>Wherein alpha is i Representing T i Can be parallel fractional, m i Representing the number of cores allocated for the edge server to process tasks generated by different MUs, n i Representing the number of CPU cores of the different mobile users MU i.
Step 4: determining the security decision of the user, firstly, S i E {0,1} represents a binary security decision for each mobile user MUi, which is made by each MU personalized according to the privacy requirements of the application data. Wherein S is i =0 means that the mobile terminal device MU i will offload computing tasks in an unencrypted manner; and S is i =1 means that the mobile terminal device MU i will encrypt the computing task and its data using the ChaCha20 encryption technique before transmission to the edge server. The edge server further decrypts the data after receiving the task and the data, then performs the calculation task and sends the processed result back to the MU i.
Step 5: the resource scheduling strategy is optimized and solved, in the embodiment, a PC provided with an AMD Ryzen 5 4600H CPU, a 3.0GHz frequency and a 16GB RAM capacity is adopted, an edge computing system is provided with 60 MUs, and the edge computing system is operated to perform multi-round test and weighted average. More specifically, from the first four steps, the MU's total overhead of performing all tasks locally is solvedAnd the total overhead of executing all tasks on edge servers +.>Then setting the weighting parameter w of the execution time and the energy consumption 1 And w is equal to 2 Wherein w is 1 ,w 2 ∈[0,1]. And let w 1 +w 2 =1. When the targets are different, the set weight values are changed correspondingly. After which a question definition is established, with the aim of minimizing the cost of the system +.>Thereafter, the optimal solution of the above objective equation is solved using a branch-and-bound method, the optimal solution being used to offload the decision set O (λ I )＝{λ 1 ,...,λ N Form representation. And finally, taking average value as a final return result after addition through multiple rounds of tests.
Based on the same inventive concept, the resource scheduling method integrating security decision and calculation acceleration comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the resource scheduling method integrating security decision and calculation acceleration is realized when the computer program is loaded to the processor.
It will be appreciated by those skilled in the art that the embodiments described herein are for the purpose of aiding the reader in understanding the principles of the application, and it is to be understood that the embodiments are merely illustrative of the application and not limiting the scope of the application, as various equivalent modifications to the application will fall within the scope of the application as defined in the claims after reading the application.

Claims (7)

1. The resource scheduling method integrating the security decision and the calculation acceleration is characterized by comprising the following steps:
step 1: calculating the processing time and energy consumption of the local task,
step 2: the delay and energy consumption of the communication process construction are calculated,
step 3: the user QoE demand parameter is calculated,
step 4: a security decision of the user is determined,
step 5: and optimizing and solving a resource scheduling strategy.
2. The method for resource scheduling with fused security decision and computational acceleration according to claim 1, wherein step 1: calculating the processing time and energy consumption of the local task to obtain the CPU core number n of different mobile users MUi i And per-core processing power f i l Re-calculating the power of each coreWherein the calculation time of the serialized part is then obtained based on Amdahl's law +.>And calculation time of parallelization part +.>Thereby the total local calculation time +.>And total local computing energy consumption, i.eWherein T is i Representing each MUi generated task, c i Representing T i The treatment density of D i Representing MUi generated task data size, alpha i Representing T i Parallel score, lambda i An offload decision variable, k, representing MUi generated task Ti 1 Is a coefficient reflecting the relationship between the processing power and the power consumption of the user mobile equipment.
3. The fused security decision and computation-accelerated resource of claim 1The scheduling method is characterized by comprising the following steps of: calculating delay and energy consumption constructed in the communication process, a wireless transmission model based on a shannon formula, and representing the transmission rate r of data according to the channel power gain and the channel noise power density i Wherein let P i Is the transmission power of the mobile user equipment MUi, h i Is the channel power gain, ω 0 Representing channel noise power density, B representing radio bandwidth, and then calculating transmission delayAnd transmission energy consumption->Then calculate edge processing time +.>And edge processing energy consumption->Wherein m is i Representing the number of cores allocated for the edge server to process tasks generated by different MUs i, f i e Representing the processing power (CPU frequency in cycles/second) per core of the edge server assigned to MUi, k 2 Is a coefficient reflecting the relation between the edge side processing power and the power consumption, and finally gives the transmission delay of the result return +.>And corresponding energy consumption->Wherein gamma is i Data size and T representing return result i The ratio of the initial task data size of P 0 Is the transmission power of the edge server.
4. The method for fusing security decisions and computational acceleration of resource scheduling of claim 1, wherein step 2: the delay and energy consumption of the communication process construction are calculated, and the method is concretely as follows: the implementation is divided into the following sub-steps:
substep 2-1: the data transmission process is generally constructed based on shannon's formula, the data transmission rate needs to have wireless bandwidth parameter B, and the transmission power P of the mobile user equipment MU i i Channel power gain h i And channel noise power density omega 0 The data transmission rate can be expressed by the above parameters as:
equation (1) represents the transmission rate representation of data as it is offloaded from the mobile user equipment to the edge over the designated wireless bandwidth B,
substep 2-2: task T generated by MUi i Is to be used for unloading transmission delay: based on the analysis, lambda was calculated i D i The transfer delay of bit data offloading to the edge can be obtained by:
wherein lambda is i An unload decision variable, D, representing task Ti generated by MUi i Representing the size of the task data generated by the MU i,
transmission energy consumption consumed by the task offloading process: lambda of transmission unloading i D i The energy consumption of the terminal device of the bit data is expressed as:
wherein lambda is i Task T representing MUi generation i Is a decision variable of unloading of D i Representing the size, P, of task data generated by MUi i Representing the transmission power of the mobile user equipment MU i,
substep 2-3: computing time required by the edge server to process tasks: lambda (lambda) i D i After the bit data is offloaded to the edge, the edge will process the data, let m be i Representing the number of cores allocated for the edge server to process tasks generated by different MUs i, f i e Representing the processing power (CPU frequency in cycles/second) per core allocated to the edge server for MU i,
f i e ＞＞f i l (4)
the power consumption of each core of edge processing data can be expressed as:
wherein k is 2 Is a coefficient reflecting the relationship between edge side processing power and power consumption, unloaded lambda i D i The computation time of the bit data includes the computation time of the serialization portion and the computation time of the parallelizable portion, the computation time of the serialization portion being expressed as:
c i representing T i The treatment density of D i Representing the size, alpha, of task data generated by MUi i Representing T i Parallel score, lambda i Task T representing MUi generation i Is used to unload the decision variables,
the computation time of the parallel portion can be expressed as:
m i representing the number of cores allocated for the edge server to process tasks generated by different MUs i, f i e Representing the processing power (CPU frequency in cycles/second) of each core of the edge server assigned to MU i, the computation time required to compute the edge server processing task from the above can be expressed as:
calculation time sum representing the serialization part +.>Representing the computation time of the parallelizable sections,
substep 2-4: energy consumption consumed by the edge server to process tasks: calculation data lambda i D i The edge energy consumption formula of the bits is as follows:
wherein k is 2 Is a coefficient reflecting the relationship between the edge side processing power and power consumption, c i Representing T i Alpha, alpha i Representing T i Parallel score, lambda i Task T representing MUi generation i Is a load-off decision variable f i e Representing the processing power of each core allocated to the edge server for MU i,
substep 2-5: and (5) returning a result: task T i After the processing is finished, the result is returned to the mobile terminal equipment, r is set i ' indicates the data transmission rate in the result return process, P 0 Is the transmission power of the edge server, the data transmission rate needs to have a wireless bandwidth parameter B, and the channel power gain h i And channel noise power density omega 0 R is similar to the offloaded data transfer rate i ' can also be formulated as：
Transmission delay of the result return: based on the above analysis, γ i D i The transmission delay of the bit result return can be expressed as:
wherein gamma is i Data size and T representing return result i The ratio of the initial task data sizes of (2) and thus, the gamma of the processing result i D i The edge server power consumption of the bit transfer to the mobile terminal device is expressed as:
wherein P is 0 Is the transmission power of the edge server.
5. The method for fusing security decisions and computational acceleration of resource scheduling of claim 1, wherein step 3: calculating QoE demand parameters of users, firstly, transmitting a reconnaissance packet with small data volume to a base station by a mobile user, then, returning result information by the base station, obtaining the CPU core number of mobile user equipment and the CPU core number of an edge server after the two operations, and then, respectively calculating local calculation acceleration according to Amdahl lawEdge computing acceleration value ++>Wherein alpha is i Representing T i Can be parallel fractional, m i Representing the number of cores allocated for the edge server to process tasks generated by different MUs, n i Representing the number of CPU cores of the different mobile users MU i.
6. The method for fusing security decisions and computational acceleration of resource scheduling of claim 1, wherein step 4: determining the security decision of the user, firstly, S i E {0,1} is represented as a binary security decision for each mobile user MUi, which is made by each MU personalized according to the privacy requirements of the application data, where S i =0 means that the mobile terminal device MU i will offload computing tasks in an unencrypted manner; and S is i =1 means that the mobile terminal device MU i will encrypt the computing task and its data using ChaCha20 encryption technique before transmission to the edge server, which receives the task and data, decrypts the data further, then performs the computing task and sends the processed result back to MU i.
7. The method for fusing security decisions and computational acceleration of resource scheduling of claim 1, wherein step 5: optimizing and solving resource scheduling strategy, specifically, according to the first four steps, solving the total cost of MU executing all tasks locallyAnd the total overhead of executing all tasks on edge servers +.>Then setting the weighting parameter w of the execution time and the energy consumption 1 And w is equal to 2 Wherein w is 1 ,w 2 ∈[0,1]And let w 1 +w 2 When the targets are different, the set weight values are changed accordingly, after which a problem definition is established, the target is to minimize the cost of the system +.>After that, the process is carried out,using branch-and-bound method to solve the optimal solution of the above objective equation to offload the decision set O (λ I )＝{λ 1 ,...,λ N The form of } represents that the final result is averaged over the sum as a final return result by multiple rounds of testing.
CN202310617535.2A 2023-05-29 2023-05-29 Resource scheduling method integrating security decision and calculation acceleration Pending CN116582949A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310617535.2A CN116582949A (en) 2023-05-29 2023-05-29 Resource scheduling method integrating security decision and calculation acceleration

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310617535.2A CN116582949A (en) 2023-05-29 2023-05-29 Resource scheduling method integrating security decision and calculation acceleration

Publications (1)

Publication Number Publication Date
CN116582949A true CN116582949A (en) 2023-08-11

Family

ID=87545160

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310617535.2A Pending CN116582949A (en) 2023-05-29 2023-05-29 Resource scheduling method integrating security decision and calculation acceleration

Country Status (1)

Country Link
CN (1) CN116582949A (en)

Similar Documents

Publication Publication Date Title
Lu et al. Low-latency federated learning and blockchain for edge association in digital twin empowered 6G networks
Guo et al. Energy-aware computation offloading and transmit power allocation in ultradense IoT networks
Hao et al. Energy efficient task caching and offloading for mobile edge computing
CN113242568A (en) Task unloading and resource allocation method in uncertain network environment
Liu et al. Multiobjective optimization for computation offloading in fog computing
CN111010434B (en) Optimized task unloading method based on network delay and resource management
Xu et al. Multiobjective computation offloading for workflow management in cloudlet‐based mobile cloud using NSGA‐II
Zhou et al. Stackelberg-game-based computation offloading method in cloud–edge computing networks
CN111163519A (en) Wireless body area network resource allocation and task unloading algorithm with maximized system benefits
CN114189892A (en) Cloud-edge collaborative Internet of things system resource allocation method based on block chain and collective reinforcement learning
CN111711962B (en) Cooperative scheduling method for subtasks of mobile edge computing system
CN113590307B (en) Edge computing node optimal configuration method and device and cloud computing center
Li et al. Security and energy-aware collaborative task offloading in D2D communication
CN115473894A (en) Calculation force security sensing and cooperation method based on block chain in calculation network
CN111526526B (en) Task unloading method in mobile edge calculation based on service mashup
WO2023116460A1 (en) Multi-user multi-task computing offloading method and system in mobile edge computing environment
Zhang et al. Enhanced adaptive cloudlet placement approach for mobile application on spark
CN117591886B (en) Method and device for quickly aggregating heterogeneous data based on cluster federation learning
Li et al. Multi-task offloading scheme for UAV-enabled fog computing networks
Ye et al. Processing capability and QoE driven optimized computation offloading scheme in vehicular fog based F-RAN
Karatalay et al. Energy-efficient resource allocation for D2D-assisted fog computing
Lyu et al. Optimal resource allocation for u-shaped parallel split learning
CN113504949A (en) Task unloading and parameter optimization method and system for MAR client in edge computing
CN116582949A (en) Resource scheduling method integrating security decision and calculation acceleration
Li Optimization of task offloading problem based on simulated annealing algorithm in MEC

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