A task offloading method for heterogeneous cellular networks based on SFC

Abstract

The present invention seeks protection for a heterogeneous cellular network task offloading method based on SFC, which belongs to the field of communication technology. In response to the problem that the dynamic changes in service distribution in heterogeneous cellular network edge scenarios may lead to a decrease in base station server resource utilization and deterioration in task offloading quality, a heterogeneous cellular network task offloading method based on SFC is proposed. The method establishes a maximum resource priority principle to construct an SFC mapping decision, and establishes a resource allocation decision by establishing an equal resource priority principle. Through iterative optimization of the SFC mapping and resource allocation decision, the method maximizes network resource utilization, thereby effectively improving the task offloading completion rate and reducing the resource cost per task.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a heterogeneous cellular network task offloading method based on SFC. Background Art

[0002] Mobile edge computing (MEC) effectively reduces transmission latency during task offloading by moving computing resources from cloud servers to the edge of the mobile network, providing users with a better service experience. With the emergence of various latency-sensitive and compute-intensive applications, such as facial recognition and autonomous driving, the management and scheduling of network resources has become increasingly complex. Introducing network function virtualization (NFV) technology in heterogeneous cellular networks enables more efficient and flexible scheduling and management of network resources. By abstracting underlying network resources into different virtual network functions (VNFs) and orchestrating them into service function chains (SFCs), effective network function support can be provided for a variety of emerging applications.

[0003] Existing research on task offloading in heterogeneous cellular networks mostly focuses on single-function and single-base station task offloading, but fails to consider the offloading of multifunctional tasks in heterogeneous cellular network scenarios based on SFC. Existing research on SFC deployment typically uses a fixed resource allocation approach to map each VNF, failing to consider optimizing VNF resource allocation based on physical node load and ensuring the service quality of the entire SFC service. To address the issue of dynamic changes in service distribution at the edge of heterogeneous cellular networks, which can lead to decreased base station server resource utilization and deterioration in task offloading quality, this paper proposes an SFC-based heterogeneous cellular network task offloading method. By jointly optimizing SFC mapping and resource allocation, this method effectively improves network resource utilization, thereby maximizing the task offloading completion rate and minimizing the resource cost per task. Summary of the Invention

[0004] The present invention aims to solve the above problems in the prior art. It proposes a method for offloading tasks in heterogeneous cellular networks based on SFC. The technical solution of the present invention is as follows:

[0005] A heterogeneous cellular network task offloading method based on SFC comprises the following steps:

[0006] 101. Initialize the candidate mapping node set according to the task offloading request R proposed by user u Optimal SFC deployment decision fitness ξ * =0, counting variable i=0;

[0007] 102. Based on the remaining available computing resources of each node n in the physical node set N, calculate the minimum computing delay required for user u's task R on node n. All the elements in the set N that satisfy The node joins the set N m , where T represents the task tolerance delay;

[0008] 103. If |F|≤|N m |, where F represents an ordered set of virtual network functions VNF required for task offloading, jump to step 104, otherwise jump to step 109;

[0009] 104. If i<I, let i=i+1, where I represents the m If the maximum number of service function chains SFC required for task offloading is found, the process jumps to step 105; otherwise, the process jumps to step 108;

[0010] 105. According to the set N m and network resource status, construct and update the SFC node and link mapping decision {X, Y}, if successful, jump to step 106, otherwise, jump to step 104;

[0011] 106. Construct and update the SFC resource allocation decision Z based on the node mapping decision X and the task tolerance delay T. If successful, jump to step 107; otherwise, jump to step 104.

[0012] 107. According to the SFC deployment decision {X, Y, Z}, calculate the corresponding deployment fitness ξ. If ξ * <ξ,let ξ * =ξ,{X * ,Y * ,Z *}={X,Y,Z}, jump to step 104, otherwise, jump to step 104;

[0013] 108. Output the optimal SFC deployment decision {X * ,Y * ,Z *};

[0014] 109. End;

[0015] The method for constructing and updating the SFC node and link mapping decision {X, Y} in step 105 includes the following steps:

[0016] 1) Initialize temporary set N′=N m , counting variables k=1, l=0;

[0017] 2) Determine the access node n0 of user u based on the base station to which user u belongs, and set the temporary variable a = n0;

[0018] 3) If k≤|F|, jump to step 4), otherwise, jump to step 6);

[0019] 4) According to the association logic between macro and micro base stations in the heterogeneous cellular network, for each node n in the set N′, calculate the minimum cost path P from node a to n a,n , and remove nodes that do not have the minimum cost path from N′;

[0020] 5) Calculate the mapping probability γ of each node n in N′ n , according to the mapping probability distribution {γ n | n∈N′} and the maximum resource priority principle randomly select node n from N′, add it as the kth VNF mapping node to the node mapping decision X, and add path P a,n Add the link mapping decision Y as the lth virtual link, set k = k + 1, l = l + 1, a = n, and jump to step 3);

[0021] 6) Output the node and link mapping decision {X, Y}, and the algorithm ends;

[0022] The method for constructing and updating the SFC resource allocation decision Z in step 106 includes the following steps:

[0023] 11) According to the node mapping decision X, the SFC mapping node is added to the temporary set N′. According to the link mapping decision Y, the bandwidth resource b required by the virtual link is added to the SFC resource allocation decision Z, and the total transmission delay t of the task offloading path is obtained. tra , where t tra is the cumulative transmission delay of task data through each physical link on the offloading path, and the remaining task completion delay t res =Tt tra ;

[0024] 12) Allocate the basic computing resources z0 required for VNF instantiation to each node n in the set N′ and update the remaining available computing resources according to The value of will sort the elements in N′ in ascending order;

[0025] 13) If Take the first node n from N′ and calculate the task computing resources that node n needs to allocate to the corresponding VNF instance according to the equal resource priority principle. Jump to step 14), otherwise, jump to step 16);

[0026] 14) If Jump to step 15), otherwise, let Skip to step 15);

[0027] 15) According to Update the remaining available computing resources And calculate the remaining task completion delay Skip to step 13);

[0028] 16) If t res ≥0, jump to step 17); otherwise, jump to step 18);

[0029] 17) Output resource allocation decision Z;

[0030] 18) The algorithm ends;

[0031] In step 13), computing node n needs to allocate computing resources for tasks to the corresponding VNF instance according to the principle of equal resource priority. The method is shown in formula (4):

[0032] d represents the amount of task data, and j represents the complexity of the task.

[0033] Furthermore, the minimum computational delay required for task R on node n in step 102 is The calculation method is shown in formula (1):

[0034]

[0035] in, represents the remaining available computing resources of node n, d represents the amount of task data, and j represents the complexity of the task.

[0036] Furthermore, the minimum cost path P in step 4) a,n The calculation method is as follows:

[0037] Calculate the minimum cost path P from node a to n using the minimum cost path algorithm a,n , where the link cost w of each physical link e e The calculation method is shown in formula (2):

[0038]

[0039] in, represents the remaining available bandwidth resources of physical link e, B e represents the bandwidth resource capacity of the physical link e, and b represents the transmission bandwidth requirement of the user task offloading.

[0040] Furthermore, in step 5), the mapping probability γ of each node n in N′ is calculated.n The method is shown in formula (3):

[0041]

[0042] in, Represents the remaining available computing resources of node n.

[0043] Furthermore, the remaining tasks in step 15) are completed with a delay of The calculation method is shown in formula (5):

[0044]

[0045] Furthermore, the calculation method of the fitness ξ in step 106 is shown in formula (6):

[0046]

[0047] Among them, α represents the weight factor, They represent the normalized computing resources and normalized bandwidth resources required for task offloading, The calculation method is shown in formula (7) and formula (8):

[0048]

[0049] In formula (7), Z n represents the total computing resources of node n. In formula (8), E represents the set of all nodes in the heterogeneous cellular network. Indicates the bandwidth resources allocated by physical link e to user u.

[0050] An electronic device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the SFC-based heterogeneous cellular network task offloading method as described in any one of the items is implemented.

[0051] The advantages and beneficial effects of the present invention are as follows:

[0052] The present invention discloses a method for offloading heterogeneous cellular network tasks based on SFC. Existing research on heterogeneous cellular network task offloading mostly considers the offloading of single-function and single-base station tasks, but does not consider the offloading of multifunctional tasks based on SFC in heterogeneous cellular network scenarios. Existing research on SFC deployment issues generally uses a fixed resource allocation method to map each VNF, but does not consider optimizing VNF resource allocation based on the load of physical nodes and ensuring the service quality of the entire SFC service. In heterogeneous cellular network edge scenarios, the dynamic changes in service distribution may lead to a decrease in base station server resource utilization and deterioration in task offloading quality. The present invention establishes a maximum resource priority principle to construct an SFC mapping decision based on load balancing and maximizing resource utilization. It also establishes an equal resource priority principle to construct a resource allocation decision based on maximizing resource efficiency. Through iterative optimization of SFC mapping and resource allocation decisions, network resource utilization is maximized, thereby effectively improving the task offloading completion rate and reducing the resource cost per task. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 The present invention provides a flowchart of a heterogeneous cellular network task offloading method based on SFC in a preferred embodiment. DETAILED DESCRIPTION

[0054] The following will describe the technical solutions in the embodiments of the present invention in detail with reference to the accompanying drawings. The described embodiments are only a part of the embodiments of the present invention.

[0055] The technical solution of the present invention to solve the above technical problems is:

[0056] The concepts and models involved in the content of this invention are as follows.

[0057] 1. Network Model

[0058] Assume that a heterogeneous cellular network consists of several macro base stations. Macro base stations communicate with each other via wired links. Each macro base station manages multiple micro base stations, and the macro and micro base stations communicate with each other via wired links. Communication between micro base stations belonging to different macro base stations requires forwarding between macro base stations. Users within the coverage area of ​​a micro base station access the network through that micro base station. Users outside the coverage area of ​​a micro base station but within the coverage area of ​​a macro base station access the network through that macro base station. Each macro and micro base station is equipped with an edge server with sufficient computing power, which can host any type of VNF instances. Based on user offload requirements, the corresponding VNFs can be instantiated on different edge servers, and the required SFCs can be orchestrated and constructed.

[0059] 2. Other symbols involved in the present invention are explained as follows.

[0060] u: terminal user u

[0061] N: physical node set

[0062] n: physical node n

[0063] E: Physical link set

[0064] e: physical link e

[0065] d: Task data volume

[0066] j: the number of CPU cycles required per unit data volume (computational complexity)

[0067] T: Maximum tolerable delay of the task

[0068] F: The ordered set of VNFs required for the task

[0069] Available computing resources of physical node n

[0070] Available bandwidth resources z0 of physical link e: Basic computing resources required to instantiate VNF ξ: SFC deployment fitness

[0071] Node n is the task computing resource b allocated to the user: the transmission bandwidth required for task offloading

[0072] The technical solutions of the present invention are described as follows.

[0073] 1. The minimum computational delay required for task R on node n The calculation method is shown in formula (1):

[0074]

[0075] in, represents the remaining available computing resources of node n, d represents the amount of task data, and j represents the complexity of the task.

[0076] 2. Link cost w of physical link e e , the calculation method is shown in formula (2):

[0077]

[0078] in, represents the remaining available bandwidth resources of physical link e, B e represents the bandwidth resource capacity of the physical link e, and b represents the transmission bandwidth requirement of the user task offloading;

[0079] 3. Calculate the mapping probability γ of each node n in N′ n , the calculation method is shown in formula (3):

[0080]

[0081] in, Represents the remaining available computing resources of node n;

[0082] 4. Based on the principle of equal resource priority, node n needs to allocate computing resources to the corresponding VNF The calculation method is shown in formula (4):

[0083]

[0084] 5. Delay in completing remaining tasks , the calculation method is shown in formula (5):

[0085] 6. SFC deployment fitness ξ, the calculation method is shown in formula (6):

[0086]

[0087] Among them, α represents the weight factor, They represent the normalized computing resources and normalized bandwidth resources required for task offloading respectively;

[0088] 7. Normalized computing resources required for task offloading The calculation method is shown in formula (7):

[0089]

[0090] Among them, Z n represents the total computing resources of node n;

[0091] 8. Normalized computing resources required for task offloading The calculation method is shown in formula (8):

[0092]

[0093] Where E represents the set of all physical links e in the heterogeneous cellular network, represents the bandwidth resources allocated by physical link e to user u;

[0094] 9. Sub-algorithm 1: Constructing SFC mapping decision

[0095] 1) Initialize temporary set N′=N m , counting variables k=1, l=0;

[0096] 2) Determine the access node n0 of user u based on the base station to which user u belongs, and set the temporary variable a = n0;

[0097] 3) If k≤|F|, jump to step 4), otherwise, jump to step 6);

[0098] 4) According to the association logic between macro and micro base stations in the heterogeneous cellular network, for each node n in the set N′, calculate the minimum cost path P from node a to n a,n , and remove nodes that do not have the minimum cost path from N′;

[0099] 5) Calculate the mapping probability γ of each node n in N′ n , according to the mapping probability distribution {γ n | n∈N′} and the maximum resource priority principle randomly select node n from N′, add it as the kth VNF mapping node to the node mapping decision X, and add path P a,n Add the link mapping decision Y as the lth virtual link, set k = k + 1, l = l + 1, a = n, and jump to step 3);

[0100] 6) Output the node and link mapping decision {X, Y}, and the algorithm ends.

[0101] 10. Sub-algorithm 2: Constructing resource allocation decisions

[0102] 11) According to the node mapping decision X, the SFC mapping node is added to the temporary set N′. According to the link mapping decision Y, the bandwidth resource b required by the virtual link is added to the SFC resource allocation decision Z, and the total transmission delay t of the task offloading path is obtained. tra , where t tra is the cumulative transmission delay of task data through each physical link on the offloading path, and the remaining task completion delay t res =Tt tra ;

[0103] 12) For each node n in the set N′ that does not have a corresponding VNF instance, allocate the basic computing resources z0 required for VNF instantiation and update the remaining available computing resources according to The value of will sort the elements in N′ in ascending order;

[0104] 13) If Take the first node n from N′ and calculate the task computing resources that node n needs to allocate to the corresponding VNF instance according to the equal resource priority principle. Jump to step 14), otherwise, jump to step 16);

[0105] 14) If Jump to step 15), otherwise, let Skip to step 15);

[0106] 15) According to Update the remaining available computing resources And calculate the remaining task completion delay , skip to step 13);

[0107] 16) If t res ≥0, jump to step 17); otherwise, jump to step 18);

[0108] 17) Output resource allocation decision Z;

[0109] 18) The algorithm ends.

[0110] A method for offloading tasks in a heterogeneous cellular network based on SFC, wherein the specific implementation method includes the following steps:

[0111] Step 1: Initialize the candidate mapping node set according to the task offloading request R proposed by user u Optimal SFC deployment decision fitness ξ * =0, counting variable i=0;

[0112] Step 2: Based on the remaining available computing resources of each node n in the physical node set N, calculate the minimum computing delay required for user u's task R on node n All the elements in the set N that satisfy The node joins the set N m , where T represents the task tolerance delay;

[0113] Step 3: If |F|≤|N m |, where F represents the ordered set of VNFs required for task offloading, jump to step 4, otherwise jump to step 9;

[0114] Step 4: If i<I, where I represents the m Search for the maximum number of SFCs required for task offloading, set i = i + 1, and jump to step 5, otherwise, jump to step 8;

[0115] Step 5: According to the set N m and network resource status, call Sub-Algorithm 1 to build and update the SFC node and link mapping decision {X, Y}. If successful, jump to step 6, otherwise, jump to step 4;

[0116] Step 6: Based on the node mapping decision X and the task tolerance delay T, call sub-algorithm 2 to construct and update the SFC resource allocation decision Z. If successful, jump to step 7; otherwise, jump to step 4.

[0117] Step 7: Based on the SFC deployment decision {X, Y, Z}, calculate the corresponding deployment fitness ξ. If ξ * <ξ,let ξ* =ξ,{X * ,Y * ,Z *}={X,Y,Z}, jump to step 4, otherwise, jump to step 4;

[0118] Step 8: Output the optimal SFC deployment decision {X * ,Y * ,Z *};

[0119] Step 9: The algorithm ends.

[0120] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0121] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0122] The above embodiments should be understood as merely illustrating the present invention and not as limiting the scope of protection of the present invention. After reading the contents of the present invention, technicians may make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A heterogeneous cellular network task offloading method based on SFC, characterized in that: The following steps are involved:

101. Initialize the candidate mapping node set according to the task offloading request R proposed by user u Optimal SFC deployment decision fitness ξ * =0, counting variable i=0; 102. Based on the remaining available computing resources of each node n in the physical node set N, calculate the minimum computing delay required for user u's task R on node n. All the elements in the set N that satisfy The node joins the set N m , where T represents the task tolerance delay; 103. If |F|≤|N m |, where F represents an ordered set of virtual network functions VNF required for task offloading, jump to step 104, otherwise jump to step 109; 104. If i<I, let i=i+1, where I represents the m If the maximum number of service function chains SFC required for task offloading is found, the process jumps to step 105; otherwise, the process jumps to step 108; 105. According to the set N m and network resource status, construct and update the SFC node and link mapping decision {X, Y}, if successful, jump to step 106, otherwise, jump to step 104; 106. Construct and update the SFC resource allocation decision Z based on the node mapping decision X and the task tolerance delay T. If successful, jump to step 107; otherwise, jump to step 104.

107. According to the SFC deployment decision {X, Y, Z}, calculate the corresponding deployment fitness ξ. If ξ * <ξ,let ξ * =ξ,{X * ,Y * ,Z * }={X,Y,Z}, jump to step 104, otherwise, jump to step 104; 108. Output the optimal SFC deployment decision {X * ,Y * ,Z * }; 109. End; The method for constructing and updating the SFC node and link mapping decision {X, Y} in step 105 includes the following steps: 1) Initialize temporary set N′=N m , counting variables k=1, l=0; 2) Determine the access node n0 of user u based on the base station to which user u belongs, and set the temporary variable a = n0; 3) If k≤|F|, jump to step 4), otherwise, jump to step 6); 4) According to the association logic between macro and micro base stations in the heterogeneous cellular network, for each node n in the set N′, calculate the minimum cost path P from node a to n a,n , and remove nodes that do not have the minimum cost path from N′; 5) Calculate the mapping probability γ of each node n in N′ n , according to the mapping probability distribution {γ n | n∈N′ } and the maximum resource priority principle randomly select node n from N′, add it as the kth VNF mapping node to the node mapping decision X, and add path P a,n Add the link mapping decision Y as the lth virtual link, set k = k + 1, l = l + 1, a = n, and jump to step 3); 6) Output the node and link mapping decision {X, Y}, and the algorithm ends; The method for constructing and updating the SFC resource allocation decision Z in step 106 includes the following steps: 11) According to the node mapping decision X, the SFC mapping node is added to the temporary set N′. According to the link mapping decision Y, the bandwidth resource b required by the virtual link is added to the SFC resource allocation decision Z, and the total transmission delay t of the task offloading path is obtained. tra , where t tra is the cumulative transmission delay of task data through each physical link on the offloading path, and the remaining task completion delay t res =Tt tra ; 12) Allocate the basic computing resources z0 required for VNF instantiation to each node n in the set N′ and update the remaining available computing resources according to The value of will sort the elements in N′ in ascending order; 13) If Take the first node n from N′ and calculate the task computing resources that node n needs to allocate to the corresponding VNF instance according to the equal resource priority principle. Jump to step 14), otherwise, jump to step 16); 14) If Jump to step 15), otherwise, let Skip to step 15); 15) According to Update the remaining available computing resources And calculate the remaining task completion delay Skip to step 13); 16) If t res ≥0, jump to step 17); otherwise, jump to step 18); 17) Output resource allocation decision Z; 18) The algorithm ends; In step 13), computing node n needs to allocate computing resources for tasks to the corresponding VNF instance according to the principle of equal resource priority. The method is shown in formula (4): d represents the amount of task data, and j represents the complexity of the task.

2. The SFC-based heterogeneous cellular network task offloading method according to claim 1, characterized in that: The minimum computational delay required for task R on node n in step 102 The calculation method is shown in formula (1): in, represents the remaining available computing resources of node n, d represents the amount of task data, and j represents the complexity of the task.

3. The SFC-based heterogeneous cellular network task offloading method according to claim 1, characterized in that: The minimum cost path P in step 4) a,n The calculation method is as follows: Calculate the minimum cost path P from node a to n using the minimum cost path algorithm a,n , where the link cost w of each physical link e e The calculation method is shown in formula (2): in, represents the remaining available bandwidth resources of physical link e, B e represents the bandwidth resource capacity of the physical link e, and b represents the transmission bandwidth requirement of the user task offloading.

4. The method for offloading tasks in heterogeneous cellular networks based on SFC according to claim 1, characterized in that: In step 5), the mapping probability γ of each node n in N′ is calculated. n The method is shown in formula (3): in, Represents the remaining available computing resources of node n.

5. The method for offloading tasks in heterogeneous cellular networks based on SFC according to claim 1, characterized in that: The remaining task completion delay in step 15) The calculation method is shown in formula (5):

6. The SFC-based heterogeneous cellular network task offloading method according to claim 1, characterized in that: The calculation method of the fitness ξ in step 106 is shown in formula (6): Among them, α represents the weight factor, They represent the normalized computing resources and normalized bandwidth resources required for task offloading, The calculation method is shown in formula (7) and formula (8): In formula (7), Z n represents the total computing resources of node n. In formula (8), E represents the set of all nodes in the heterogeneous cellular network. Indicates the bandwidth resources allocated by physical link e to user u.

7. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for offloading heterogeneous cellular network tasks based on SFC as claimed in any one of claims 1 to 6 is implemented.

Images