Method for optimizing multi-board-card communication path cost
By establishing a cost model for multi-board systems and combining Dijkstra and A* algorithm to tune the communication path, the problems of inefficient communication and waste of resources in the existing technology are solved, and more efficient resource utilization is achieved.
Patent Information
- Application Number
- CN202510182791.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-19
- Publication Date
- 2025-06-13
AI Technical Summary
When selecting communication paths, existing multi-board systems cannot fully consider communication load and resource limitations, resulting in inefficient communication and waste of resources.
By establishing a cost model that comprehensively considers communication latency, bandwidth utilization, energy consumption and reliability, and adopts dynamic parameters and adaptive learning mechanisms, combining Dijkstra algorithm and A* algorithm to tune the communication path between the board cards.
The communication efficiency and resource utilization of multi-board systems are improved, and resource waste in communication path selection is reduced.
Smart Images
Figure CN120146337A_ABST
Abstract
Description
Technical Field
[0001] The present invention discloses a method for optimizing the communication path cost of multiple boards, which relates to the technical field of board communication. Background Art
[0002] In a multi-board system, the selection of the communication path has an important impact on the overall performance and resource utilization rate of the system. Existing communication path selection methods often rely on simple shortest path or minimum cost principles, but these methods cannot fully consider the actual communication load and resource constraints in a multi-board environment, easily leading to low communication efficiency and resource waste. Summary of the Invention
[0003] Aiming at the problems of the prior art, the present invention provides a method for optimizing the communication path cost of multiple boards. By comprehensively considering index parameters such as communication delay, bandwidth utilization rate, energy consumption, and reliability, a cost model is established, and a dynamic parameter and an adaptive learning mechanism are used to optimize the communication path between boards, improving the communication efficiency and resource utilization rate of the multi-board system.
[0004] The specific solution proposed by the present invention is as follows:
[0005] The present invention provides a method for optimizing the communication path cost of multiple boards, including:
[0006] Step 1: Collect the bandwidth utilization rate, communication delay data, and energy consumption data of multi-board devices,
[0007] Step 2: Define the bandwidth cost, delay cost, energy consumption cost, and reliability index cost according to the collected data, and establish a cost model for the communication path of multiple boards,
[0008] Step 3: Combine the Dijkstra algorithm and the A* algorithm to select the communication path of multiple boards:
[0009] Use the Dijkstra algorithm to create a priority queue Q to store the path nodes to be processed, and initialize the cost from the starting node to itself as 0 and the cost to other path nodes as infinity. The cost from each path node to its predecessor node is empty. Use the A* algorithm to establish a heuristic function h(p) to predict the cost from the current node to the target node through the heuristic function,
[0010] Take out the path node P from the priority queue Q n , P n+1 denotes each neighbor node of the path node P n Use the cost model to calculate the cost from the starting node to P n+1 through P n If the calculated cost is lower than the currently recorded lowest cost, then update the cost of node P n+1 At the same time, update node Pn+1 The predecessor node of n is P n+1 . Then, using the cost of node P n+1 and the value of the heuristic function h(p n+1 ), their sum is used as the total cost of P n+1 . Node P
[0011] is inserted into the priority queue Q. Based on the priority queue Q, starting from the target node and backtracking through the predecessor nodes, the shortest communication cost path is obtained.
[0012] Furthermore, in step 2 of the method for optimizing the communication path cost of multiple boards, according to the collected data, the communication path delay cost is defined as the weighted sum of the delays of each link on the path.
[0013] According to the collected data, the communication path bandwidth cost is defined as the weighted sum of the bandwidth utilization rates of each link on the path.
[0014] According to the collected data, obtaining the communication reliability metric cost follows an exponentially correlated decay law.
[0015] Furthermore, in step 2 of the method for optimizing the communication path cost of multiple boards, defining the communication path delay cost as the weighted sum of the delays of each link on the path includes:
[0016] Using the formula:
[0017]
[0018] Calculate the communication path delay cost, where is the delay weight of the i-th link, d i is the delay cost of the i-th link, and n is the number of links on the path.
[0019] Furthermore, in step 2 of the method for optimizing the communication path cost of multiple boards, defining the communication path bandwidth cost as the weighted sum of the bandwidth utilization rates of each link on the path includes: Using the formula:
[0020]
[0021] Calculate the communication path bandwidth cost, where is the bandwidth utilization rate weight of the i-th link, u i is the bandwidth utilization rate of the i-th link, and n is the number of links on the path.
[0022] Further, in step 2 of the method for optimizing the communication path cost of multiple boards, the cost of obtaining the reliability index of communication follows an exponentially correlated change attenuation law, including: using the formula:
[0023] Cost r =e -λt
[0024] to illustrate the cost of the reliability index, where λ is the attenuation factor, which is a constant, indicating that as the usage time increases, the reliability of communication gradually decreases, and the cost increases due to the reliability.
[0025] Further, in step 2 of the method for optimizing the communication path cost of multiple boards, calculating the communication energy consumption cost includes:
[0026] Using the formula:
[0027] Cost e =e protocol +e topology
[0028] to calculate the communication energy consumption cost, e protocol =α*D + β*H(D), where D is the data volume, in bits, H(D) is the entropy of data transmission, representing the minimum average number of bits required to transmit the same amount of data, and α, β are weight coefficients used to adjust the importance of different factors, and e topology =γ*L + δ*C(L), L is the link length, in meters, C(L) is the link complexity function, which depends on the network topology, and γ, δ are weight coefficients used to adjust the importance of different factors.
[0029] Further, in step 2 of the method for optimizing the communication path cost of multiple boards, the cost model for establishing the communication path of multiple boards is expressed as:
[0030] Cost p =η*Cost d +ρ*Cost u +σ*Cost r +ω*Cost e
[0031] Cost d is the communication path delay cost, Cost u is the bandwidth cost of the communication path, Cost r is the reliability index cost, Cost e is the communication energy consumption cost, and η, ρ, σ, ω are weight coefficients used to adjust the importance of different factors.
[0032] Further, in step 3 of the method for optimizing the communication path cost of multiple boards, the A* algorithm is used to establish a heuristic function h(p), and the formula of the heuristic function is as follows:
[0033] h(p) = α * d(p, goal) + β * Var(L p ) + γ * E(p) + δ * R(p)
[0034] d(p, goal) is the Euclidean distance or Manhattan distance from node p to the target node, Var(L p ) is the load variance of the node, used to evaluate the load balance situation, E(p) is the estimated energy consumption value of node p, R(p) is the estimated reliability value of node p, and α, β, γ, δ are weight coefficients, used to balance the importance of different factors.
[0035] The present invention also provides an apparatus for optimizing the communication path cost of multiple boards, including an acquisition module, a model management module, and a path tuning module.
[0036] The acquisition module acquires the bandwidth utilization rate, communication delay data, and energy consumption data of the multi-board device.
[0037] The model management module defines the bandwidth cost, delay cost, energy consumption cost, and reliability index cost according to the acquired data, and establishes a cost model for the communication path of the multiple boards.
[0038] The path tuning module combines the Dijkstra algorithm and the A* algorithm to select the communication path of the multiple boards:
[0039] Use the Dijkstra algorithm to create a priority queue Q to store the path nodes to be processed, initialize the cost from the starting node to itself as 0, and the cost to other path nodes as infinity. The cost from each path node to its predecessor node is empty. Use the A* algorithm to establish a heuristic function h(p), and predict the cost from the current node to the target node through the heuristic function.
[0040] Take out the path node P from the priority queue Q n , P n+1 denotes each neighbor node of the path node P n . Use the cost model to calculate the cost from the starting node to P n+1 through P n . If the calculated cost is lower than the currently recorded lowest cost, then update the cost of node P n+1 , and at the same time update the predecessor node of node P n+1 to P n . Then, add the cost of node P n+1 and the value of the heuristic function h(p n+1 ) as the value of P n+1The total cost, insert node P n+1 into the priority queue Q. According to the priority queue Q, starting from the target node, backtrack through the predecessor nodes to obtain the shortest communication cost path.
[0041] The advantages of the present invention are:
[0042] By comprehensively considering index parameters such as communication delay, bandwidth utilization, energy consumption, and reliability, a cost model is established to evaluate the cost or price of a certain communication link. The path optimization Dijkstra algorithm is introduced into the communication path, and combined with the heuristic search A* algorithm to reduce or decrease the search space, improving the communication efficiency and resource utilization rate of the multi-board system. Brief Description of the Drawings
[0043] Figure 1 is a schematic diagram of the multi-board cluster communication framework.
[0044] Figure 2 is a schematic diagram of the multi-board communication tuning process of the present invention.
[0045] Figure 3 is a schematic diagram of the multi-board communication principle. Detailed Embodiments
[0046] The present invention will be further described below in conjunction with the drawings and specific embodiments, so that those skilled in the art can better understand the present invention and be able to implement it, but the embodiments cited are not intended to limit the present invention.
[0047] Embodiment 1
[0048] The present invention provides a method for optimizing the communication path cost of multi-boards, including:
[0049] Step 1: Collect the bandwidth utilization rate, communication delay data, and energy consumption data of multi-board devices. Among them, sensors and other sensing devices can be used to obtain the bandwidth utilization rate, communication delay situation, energy consumption situation, and reliability index parameters of a certain communication link or device, and a historical knowledge base is established based on these parameters.
[0050] Step 2: Define the bandwidth cost, delay cost, energy consumption cost, and reliability index cost according to the collected data, and establish a cost model for the communication path of the multi-board.
[0051] Among them, in Step 2, according to the collected data, the communication path delay cost is defined as the weighted sum of the delays of each link on the path, including:
[0052] Using the formula:
[0053]
[0054] Calculate the communication path delay cost, where is the delay weight of the i-th link, d i is the delay cost of the i-th link, and n is the number of links on the path. The weight is determined according to the importance and usage frequency of the link.
[0055] According to the collected data, the bandwidth cost of the communication path is defined as the weighted sum of the bandwidth utilization rates of each link on the path, including: Using the formula:
[0056]
[0057] Calculate the bandwidth cost of the communication path, where is the bandwidth utilization weight of the i-th link, u i is the bandwidth utilization rate of the i-th link, and n is the number of links on the path. The weight is determined according to the bandwidth and load conditions of the link.
[0058] According to the collected data, obtaining the cost of the communication reliability index follows an exponentially correlated change attenuation law, including: Using the formula:
[0059] Cost r = e -λt
[0060] Illustrate the cost of the reliability index, where λ is the attenuation factor, which is a constant, indicating that as the usage time increases, the reliability of the communication gradually decreases, and the cost increases due to the reliability.
[0061] According to the collected data, combined with the topological structure of the communication network and the selection of the communication protocol, calculate the communication energy consumption cost, including: Using the formula:
[0062] Cost e = e protocol + e topology
[0063] Calculate the communication energy consumption cost, e protocol = α * D + β * H(D), where D is the amount of data in bits, H(D) is the entropy of data transmission, representing the minimum average number of bits required to transmit the same amount of data, and α, β are weight coefficients used to adjust the importance of different factors, e topology = γ * L + δ * C(L), L is the link length in meters, C(L) is the link complexity function depending on the network topological structure, and γ, δ are weight coefficients used to adjust the importance of different factors.
[0064] Through the calculation of the above various parameter costs, the cost model of the communication path established in step 2 for multiple boards is expressed as:
[0065] Cost p = η * Costd + ρ * Cost u + σ * Cost r + ω * Cost e
[0066] Cost d is the communication path delay cost, Cost u is the bandwidth cost of the communication path, Cost r is the reliability index cost, Cost e is the communication energy consumption cost, η, ρ, σ, ω are weight coefficients used to adjust the importance of different factors.
[0067] Step 3: Combine the Dijkstra algorithm and the A* algorithm to select the communication path for multiple boards:
[0068] Use the Dijkstra algorithm to create a priority queue Q to store the path nodes to be processed, and initialize the cost from the starting node to itself as 0 and the cost to other path nodes as infinity. The cost from each path node to its predecessor node is empty. Use the A* algorithm to establish a heuristic function h(p), and predict the cost from the current node to the target node through the heuristic function. The formula for the heuristic function is as follows:
[0069] h(p) = α * d(p, goal) + β * Var(L p ) + γ * E(p) + δ * R(p)
[0070] d(p, goal) is the Euclidean distance or Manhattan distance from node p to the target node, Var(L p ) is the load variance of the node, used to evaluate the load balance situation, E(p) is the estimated energy consumption value of node p, R(p) is the estimated reliability value of node p, and α, β, γ, δ are weight coefficients used to balance the importance of different factors.
[0071] Take out the path node P from the priority queue Q n , P n+1 is denoted as each neighbor node of path node P n Use the cost model to calculate the cost from the starting node to P n+1 through P n . If the calculated cost is lower than the currently recorded lowest cost, then update the cost of node P n+1 , and at the same time update the predecessor node of node P n+1 as P n , and then add the cost of node P n+1 to the value of the heuristic function h(p n+1 ) as the total cost of P n+1 , and add node P n+1Insert the priority queue Q. According to the priority queue Q, starting from the target node, trace back through the predecessor nodes to obtain the shortest communication cost path.
[0072] Embodiment 2
[0073] The present invention also provides a device for optimizing the communication path cost of multiple boards, including a collection module, a model management module, and a path optimization module.
[0074] The collection module collects the bandwidth utilization rate, communication delay data, and energy consumption data of the multi-board device.
[0075] The model management module defines the bandwidth cost, delay cost, energy consumption cost, and reliability index cost according to the collected data, and establishes a cost model for the communication path of the multi-board.
[0076] The path optimization module combines the Dijkstra algorithm and the A* algorithm to select the communication path of the multi-board:
[0077] Use the Dijkstra algorithm to create a priority queue Q, store the path nodes to be processed, and initialize the cost from the starting node to itself as 0 and the cost to other path nodes as infinity. The cost from each path node to its predecessor node is empty. Use the A* algorithm to establish a heuristic function h(p), and predict the cost from the current node to the target node through the heuristic function.
[0078] Take out the path node P from the priority queue Q n , P n+1 Denote each neighbor node of the path node P n Use the cost model to calculate the cost from the starting node to P n+1 Passing through P n If the calculated cost is lower than the currently recorded lowest cost, then update the cost of node P n+1 At the same time, update the predecessor node of node P n+1 as P n , and then add the cost of node P n+1 to the value of the heuristic function h(p n+1 ) as the total cost of P n+1 . Insert the node P n+1 into the priority queue Q. According to the priority queue Q, starting from the target node, trace back through the predecessor nodes to obtain the shortest communication cost path.
[0079] Regarding the information interaction, execution process, etc. among the above-mentioned device modules, since they are based on the same concept as the method embodiment of the present invention, the specific content can be referred to the description in the method embodiment of the present invention, and will not be elaborated here.
[0080] Similarly, the device of the present invention establishes a cost model by comprehensively considering index parameters such as communication delay, bandwidth utilization, energy consumption, and reliability, which is used to evaluate the cost or price of a certain communication link. The path optimization Dijkstra algorithm is introduced into the communication path, and combined with the heuristic search A* algorithm to reduce or decrease the search space, improving the communication efficiency and resource utilization rate of the multi-board system.
[0081] It should be noted that not all steps and modules in the above-mentioned processes and device structures are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted according to needs. The system structure described in the above embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or some components in multiple independent devices can be jointly implemented.
[0082] The above-mentioned embodiments are only preferred embodiments given to fully illustrate the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or transformations made by those skilled in the art on the basis of the present invention are all within the protection scope of the present invention. The protection scope of the present invention is subject to the claims.
Claims
1. A method for optimizing the cost of multi-board communication paths, characterized in that include: Step 1: Collect bandwidth utilization, communication delay data and energy consumption data of multi-board devices. Step 2: Define bandwidth cost, delay cost, energy consumption cost, and reliability index cost based on the collected data, and establish a cost model for the communication path of multiple boards. Step 3: Combine Dijkstra algorithm and A* algorithm to select the communication path of multiple boards: Use Dijkstra algorithm to create a priority queue Q to store the path nodes to be processed, and initialize the cost from the starting node to itself to 0, the cost to other path nodes to infinity, and the cost from each path node to the predecessor node to be empty. Use A* algorithm to establish heuristic function h(p), and use the heuristic function to predict the cost from the current node to the target node. Take the path node P from the priority queue Q n , P n+1 Represented as path node P n For each neighbor node of P, the cost model is used to calculate the distance from the starting node to P n+1 After P n If the calculated cost is lower than the lowest cost currently recorded, update node P n+1 The cost of updating node P n+1 The predecessor node is P n , then take node P n+1 The cost and heuristic function h(p n+1 ) is added as P n+1 The total cost of node P n+1 Insert into the priority queue Q. According to the priority queue Q, start from the target node and backtrack through the predecessor node to obtain the path with the shortest communication cost.
2. The method for optimizing the multi-board communication path cost according to claim 1, characterized in that In step 2, based on the collected data, the communication path delay cost is defined as the weighted sum of the delays of each link on the path. Based on the collected data, the bandwidth cost of the communication path is defined as the weighted sum of the bandwidth utilization of each link on the path. According to the collected data, the cost of obtaining the reliability index of communication follows the exponentially related change attenuation law. Based on the collected data, combined with the topology of the communication network and the choice of communication protocol, the communication energy consumption cost is calculated.
3. The method for optimizing the multi-board communication path cost according to claim 2, characterized in that The communication path delay cost defined in step 2 is the weighted sum of the delays of each link on the path, including: Using the formula: Calculate the communication path delay cost, where is the delay weight of the ith link, d i is the delay cost of the i-th link, and n is the number of links on the path.
4. The method for optimizing the multi-board communication path cost according to claim 2, characterized in that The bandwidth cost of the communication path defined in step 2 is the weighted sum of the bandwidth utilization of each link on the path, including: Using the formula: Calculate the bandwidth cost of the communication path, where is the bandwidth utilization weight of the ith link, u i is the bandwidth utilization of the i-th link, and n is the number of links on the path.
5. The method for optimizing the multi-board communication path cost according to claim 2, characterized in that The reliability index cost of communication obtained in step 2 follows the exponentially related change attenuation law, including: using the formula: Cost r =e -λt It shows the reliability index cost, where λ is the attenuation factor and is a constant, indicating that as the usage time increases, the reliability of communication gradually decreases and the cost brought by reliability increases.
6. The method for optimizing the multi-board communication path cost according to claim 2, characterized in that The communication energy consumption cost is calculated in step 2, including: Using the formula: Cost e =e protocol +e topology Calculate the communication energy consumption cost, e protocol =α*D+β*H(D), where D is the amount of data in bits, H(D) is the entropy of data transmission, which indicates the minimum average number of bits required to transmit the same amount of data, and α and β are weight coefficients used to adjust the importance of different factors. topology =γ*L+δ*C(L), where L is the link length in meters, C(L) is the link complexity function, which depends on the network topology, and γ and δ are weight coefficients used to adjust the importance of different factors.
7. The method for optimizing the multi-board communication path cost according to claim 2, characterized in that In step 2, the cost model of the communication path of multiple boards is established, which is expressed as: Cost p =η*Cost d +ρ*Cost u +σ*Cost r +ω*Cost e Cost d Cost is the communication path delay cost. u Cost is the bandwidth cost of the communication path. r Cost is the reliability indicator cost. e is the communication energy consumption cost, η, ρ, σ, ω are weight coefficients used to adjust the importance of different factors.
8. The method for optimizing the multi-board communication path cost according to claim 1, characterized in that In step 3, the A* algorithm is used to establish the heuristic function h(p). The heuristic function formula is as follows: h(p)=α*d(p,goal)+β*Var(L p )+γ*E(p)+δ*R(p) d(p,goal) is the Euclidean distance or Manhattan distance from node p to the target node, Var(L p ) is the load variance of the node, which is used to evaluate the load balance. E(p) is the estimated energy consumption of node p. R(p) is the estimated reliability of node p. α, β, γ, and δ are weight coefficients used to balance the importance of different factors.
9. A device for optimizing the cost of multi-board communication paths, characterized in that Including acquisition module, model management module and path tuning module, The acquisition module collects bandwidth utilization, communication delay data and energy consumption data of multi-board devices. The model management module defines bandwidth cost, delay cost, energy consumption cost, and reliability index cost based on the collected data, and establishes a cost model for the communication path of multiple boards. The path optimization module combines the Dijkstra algorithm and the A* algorithm to select the communication path for multiple boards: Use Dijkstra algorithm to create a priority queue Q to store the path nodes to be processed, and initialize the cost from the starting node to itself to 0, the cost to other path nodes to infinity, and the cost from each path node to the predecessor node to be empty. Use A* algorithm to establish heuristic function h(p), and use the heuristic function to predict the cost from the current node to the target node. Take the path node P from the priority queue Q n , P n+1 Represented as path node P n For each neighbor node of P, the cost model is used to calculate the distance from the starting node to P n+1 After P n If the calculated cost is lower than the lowest cost currently recorded, update node P n+1 The cost of updating node P n+1 The predecessor node is P n , then take node P n+1 The cost and heuristic function h(p n+1 ) is added as P n+1 The total cost of node P n+1 Insert into the priority queue Q. According to the priority queue Q, start from the target node and backtrack through the predecessor node to obtain the path with the shortest communication cost.