Internet of vehicles computing unloading method and system

Abstract

The invention relates to the field of Internet of Vehicles edge computing and network security, in particular to an Internet of Vehicles computing unloading method and system, and the method comprises the steps: 1, constructing an Internet of Vehicles system architecture comprising a vehicle, a roadside unit, an edge server and a block chain network; 2, defining and calculating an unloading decision variable, and using # imgabs0 # to represent whether the task of the vehicle # imgabs1 # is unloaded to a local or edge server; 3, calculating the fitness value of each vehicle node task unloading scheme; 4, performing iterative optimization by using an aurora optimization algorithm; 5, judging whether the aurora optimization algorithm reaches a preset maximum number of iterations or not; an unloading decision is optimized by deeply fusing an Internet of Vehicles architecture and a block chain technology, taking an aurora optimization algorithm as a core and comprehensively considering multiple factors. Through the unique operation of simulating the aurora motion, the approximate optimal strategy is quickly searched in the complex Internet of Vehicles environment, and the tasks are reasonably allocated to the vehicle, the local or the edge, so that the system cost is reduced, and the resource utilization efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the fields of Internet of Vehicles (IoV) edge computing and network security, and particularly to an IoV computing offloading method and its system. Background Art

[0002] With the rapid deployment and development of IoV, various interconnected services and applications have emerged continuously, and users' requirements for aspects such as network service quality and latency have also increased. At the same time, due to the increase in data volume, vehicles usually cannot process computationally intensive or latency-sensitive computing tasks in a timely manner, and existing IoV computing offloading technologies also have deficiencies in decision optimization, data security, and trust. There is an urgent need for an IoV computing offloading method based on the Aurora optimization algorithm and blockchain technology, which can integrate existing optimization algorithms and blockchain technology into the IoV computing offloading scenario, comprehensively solve these problems, and achieve efficient, secure, and reliable IoV computing offloading. Summary of the Invention

[0003] In view of the above problems, the purpose of the present invention is to provide an IoV computing offloading method and its system, which can perform IoV computing offloading safely and efficiently in a complex environment, and effectively reduce the energy consumption, latency, and system overhead of offloading.

[0004] An IoV computing offloading method provided by the present invention specifically includes the following steps: Step 1: Construct an IoV system architecture including vehicles, roadside units (RSUs), multi-access edge computing (MEC) servers, and a blockchain network; Step 2: Define a computing offloading decision variable, and use to represent whether the task of vehicle is offloaded to the local or the edge server; Step 3: According to the offloading decision variable in Step 2, calculate the fitness value of the task offloading scheme of each vehicle node. With the goal of minimizing the total system overhead, considering factors such as task latency, energy consumption, and blockchain data processing overhead, encrypt the key intermediate data in the fitness value calculation process and record it in the blockchain network, and finally verify and store it through the consensus mechanism in the blockchain network; Step 4: Use the Aurora optimization algorithm for iterative optimization. In each iteration, update the offloading decision variable according to the optimal solution and individual solution in the current population through the motion behavior simulation mechanism of the Aurora optimization algorithm, and perform a mutation operation at the same time. Finally, broadcast the updated information to the blockchain network; Step 5: Determine whether the Aurora optimization algorithm reaches the preset maximum number of iterations; when the convergence condition is reached, output the globally optimal computing offloading scheme, and encrypt and record the final offloading scheme in the blockchain network; when the convergence condition is not reached, return to Step 4 to continue iterative optimization.

[0005] Preferably in the present invention, the calculation methods of energy consumption and time delay in the third step are as follows: Vehicle The energy consumption generated by the task offloading is ; The generated time delay is ; In the formula, is the offloading decision variable of the task. The offloading decision variable refers to offloading to the local or edge server. When = 0, it means that the task is calculated locally, and the energy consumption is , and the time delay is ; when = 1, it means that the task is calculated on the edge server, and the energy consumption is , and the time delay is , where represents the energy consumption consumed by the task offloading to the local for calculation, represents the energy consumption consumed by the task offloading to the MEC server for calculation, represents the energy consumption consumed by executing the consensus mechanism in the blockchain network, represents the time delay generated by the task offloading to the local for calculation, represents the time delay generated by the task offloading to the MEC server for calculation, represents the time delay generated by executing the consensus mechanism in the blockchain network.

[0006] Preferably in the present invention, the calculation method of the overhead in the third step is as follows: The total overhead of the system is ; In the formula, and are the coefficients of energy consumption and time delay, where , and .

[0007] Preferably in the present invention, the update and mutation operations of the aurora optimization algorithm in the fourth step are specifically as follows: Step Four One, update operation: The update operation includes rotational motion and aurora elliptical walk. Among them, the rotational motion uses the optimal solution in the current population as the reference point to guide other individual solutions to perform rotational position adjustment around it. The aurora elliptical walk draws on the characteristics of the aurora moving on an elliptical orbit to let the individual solutions move in a walking manner along a path similar to an ellipse; Step Four Two, mutation operation: The mutation operation includes particle collision, specifically randomly selecting two individual solutions and simulating the effect of particle collision between them.

[0008] Preferably, in the present invention, the threshold value in the convergence condition in step five is adjusted according to the accuracy requirements, stability requirements of the vehicle networking system, and the performance of the blockchain network. The specific adjustment method is as follows: when the vehicle networking system has high requirements for computing accuracy (such as real-time decision-making scenarios), the threshold value is set lower to ensure that the Aurora optimization algorithm can fully iterate to find a more accurate global optimal offloading solution; when the system has high requirements for stability (such as in scenarios of high-speed movement or large network fluctuations), the threshold value is set higher to accelerate the convergence speed of the Aurora optimization algorithm and avoid system response delays or policy instability caused by excessive iteration.

[0009] Preferably, in the present invention, the consensus mechanism in the blockchain network adopts the Delegated Byzantine Fault Tolerance algorithm (DBFT) to ensure the consistency and effectiveness of the update of decision variables.

[0010] Another object of the present invention is to provide a vehicle networking computing offloading system, including: an environment and task management module, a decision variable initialization and blockchain storage module, a fitness calculation and blockchain recording module, an optimization iteration and blockchain consensus module, and a convergence judgment and result output module; The environment and task management module is used to construct the vehicle networking system architecture, initialize the parameters of vehicle nodes, edge computing nodes, and blockchain nodes, collect vehicle node computing task information, and interact with the blockchain network to record task data; The decision variable initialization and blockchain storage module is responsible for defining and initializing the offloading decision variable matrix and algorithm parameters, and encrypting and storing them in the blockchain network; The fitness calculation and blockchain recording module is used to calculate the fitness value based on the offloading decision variables, consider various cost factors, and record key data in the evaluation process in the blockchain. Among them, the various cost factors include: computing energy consumption costs (such as local computing energy consumption , MEC server computing energy consumption and blockchain consensus energy consumption ), computing delay costs (such as local computing delay , MEC server computing delay and blockchain consensus delay ); The optimization iteration and blockchain consensus module is used to implement the iteration of the Aurora optimization algorithm, including decision variable update, new solution evaluation and selection, and at the same time use the consensus mechanism in the blockchain network to make the data consistent; The convergence judgment and result output module is used to judge the convergence of the aurora optimization algorithm, output the optimal offloading scheme and record it in the blockchain network. If it does not converge, it will feedback to the optimization iteration module for continuous optimization. Among them, the optimization iteration module evaluates the fitness of the newly generated solution by dynamically adjusting the offloading decision variable matrix (including resource allocation scheme and blockchain node selection strategy), considering factors such as energy consumption, delay, and communication overhead, and selects the optimal solution as the optimal scheme for the current iteration.

[0011] The beneficial effects of the present invention are as follows: 1. The method of the present invention deeply integrates the vehicle networking architecture and blockchain technology, takes the aurora optimization algorithm as the core, and comprehensively considers multiple factors to optimize the offloading decision. Through the unique operation of simulating the movement of aurora, it quickly explores near-optimal strategies in a complex vehicle networking environment, reasonably allocates tasks to vehicles, local or edge, reduces system costs, and improves resource utilization efficiency.

[0012] 2. The method of the present invention uses the consensus mechanism to ensure the consistency of node data and the effectiveness of decision update, establishes a reliable trust relationship between nodes, and enhances the stability and security of the system.

[0013] 3. The method of the present invention combines the mutation and update operations in the aurora optimization algorithm to increase the diversity of solution space exploration, avoid falling into local optima, and improve the probability of finding the global optimal strategy.

[0014] 4. In the method of the present invention, the blockchain and the computing offloading process cooperate closely, and the blockchain is deeply involved in each stage of the task. This collaborative optimization process improves the reliability and scalability of the system, promotes node cooperation, and enhances the overall efficiency and intelligence level. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Through the following description with reference to the accompanying drawings, and with a more comprehensive understanding of the present invention, other objects and results of the present invention will become more obvious and easier to understand. In the drawings: Figure 1 is a flowchart of a vehicle networking computing offloading method based on the aurora optimization algorithm and blockchain technology of the present invention; Figure 2 is a schematic diagram of the network model of a vehicle networking computing offloading method based on the aurora optimization algorithm and blockchain technology of the present invention; Figure 3 is a schematic diagram of blockchain consensus simulation of a vehicle networking computing offloading method based on the aurora optimization algorithm and blockchain technology of the present invention; Figure 4 is a schematic diagram of aurora optimization algorithm simulation of a vehicle networking computing offloading method based on the aurora optimization algorithm and blockchain technology of the present invention; Figure 5It is one of the schematic diagrams of the task volume of the simulation result of the specific implementation mode of a vehicle networking computing offloading method based on the aurora optimization algorithm and blockchain technology according to the present invention; Figure 6 It is the second schematic diagram of the task volume of the simulation result of the specific implementation mode of a vehicle networking computing offloading method based on the aurora optimization algorithm and blockchain technology according to the present invention; Figure 7 It is the third schematic diagram of the task volume of the simulation result of the specific implementation mode of a vehicle networking computing offloading method based on the aurora optimization algorithm and blockchain technology according to the present invention. Specific implementation mode

[0016] Embodiment 1 Refer to Figure 1-7 The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0017] As Figure 1 shown, the embodiment of the present invention provides a vehicle networking computing offloading method, including the following steps: Step 1: Construct a vehicle networking system architecture including vehicles, roadside units (RSUs), mobile edge computing (MEC) servers, and a blockchain network; Step 2: Define the computing offloading decision variable, and use to represent whether the task of vehicle is offloaded to the local or the edge server; Step 3: According to the offloading decision variable in Step 2, calculate the fitness value of the task offloading scheme of each vehicle node. With the goal of minimizing the total overhead of the system, considering factors such as task latency, energy consumption, and blockchain data processing overhead, encrypt the key intermediate data in the fitness value calculation process and record it in the blockchain network, and finally verify and store it through the consensus mechanism in the blockchain network; Step 4: Use the aurora optimization algorithm for iterative optimization. In each iteration, according to the optimal solution and individual solution in the current population, update the offloading decision variable through the motion behavior simulation mechanism of the aurora optimization algorithm, and at the same time perform a mutation operation, and finally broadcast the updated information to the blockchain network; Step 5: Determine whether the aurora optimization algorithm has reached the preset maximum number of iterations; when the convergence condition is reached, output the globally optimal computing offloading scheme, and encrypt and record the final offloading scheme in the blockchain network; when the convergence condition is not reached, return to Step 4 to continue iterative optimization.

[0018] Furthermore, Step 1 in this embodiment further includes the following specific steps: The network model constructed in this embodiment is as Figure 2As shown in the figure, the network model consists of vehicles, roadside units, edge servers, and a blockchain network. Each RSU acts as a relay node to realize information interaction between vehicles and edge servers within its communication coverage. The vehicle travels at a speed at a constant speed. RSUs are equidistantly installed along the roadside, and an MEC server is deployed in each RSU. The two are connected by a fiber-optic wired link. The set of vehicles is defined as , and the set of RSUs is defined as . In the blockchain system, all RSUs form blockchain nodes, which can be divided into two categories. One category is ordinary nodes that only undertake transmission and reception work and do not participate in the consensus process; the other category is consensus nodes selected by voting based on trust values, denoted as . Such nodes are responsible for generating new blocks and implementing the consensus process.

[0019] Furthermore, Steps 2 and 3 in this embodiment further include the following specific steps: The tasks generated by vehicles follow the Poisson distribution law, that is, each vehicle has one and only one computing task , where is the size of the input data required for the task, is the number of CPU cycles required to complete the task, is the maximum latency that the task completion can tolerate, is the maximum time that the blockchain consensus process can tolerate. The task uses the offloading decision variable to determine whether to process locally or offload to the MEC server for processing. When = 0, it means the task is calculated locally. When = 1, it means the task is offloaded to the MEC server for calculation; The vehicle offloads the task to the MEC server through the RSU. After the server completes the calculation, it then feeds back the result to the vehicle. According to Shannon's formula, the data transmission rate between the vehicle and the RSU is defined as: (Formula 1); In the formula, represents the channel bandwidth between the vehicle and the MEC server, represents the channel gain between the vehicle and the RSU, represents the transmission power of the vehicle, represents the power of Gaussian white noise; (1) Local calculation; When the vehicle executes the computing task locally, the calculation latency and energy consumption depend on the vehicle's own computing power. Therefore, the energy consumption and latency of the task calculated locally are respectively: (Formula 2); (Formula 3); In the formula, represents the computing power of the vehicle ; represents the power consumed by the core operation of each vehicle, represents the effective open capacitance coefficient of the chip structure corresponding to the vehicle; (2) Edge server computing; When the vehicle executes a computationally intensive and latency-sensitive task and uses an edge server for processing, since the size of the output data after computing on the edge server is much smaller than the size of the input data, the latency of the result feedback to the vehicle can be ignored; therefore, the energy consumption and latency of the task on the edge server are respectively: (Formula 4); (Formula 5); (Formula 6); (Formula 7); In the formula, represents the wireless transmission rate between the task vehicle and the edge server, represents the transmission power of the task vehicle, represents the transmission latency between the task vehicle and the edge server, represents the processing latency of the computing task in the edge server, represents the computing power of the MEC server.

[0020] (3) Blockchain consensus computing; The latency in the consensus computing process is defined as , and the energy consumption is defined as .

[0021] Mobility of the vehicle: The vehicle travels at a constant speed of . To prevent the task transmission from being interrupted due to RSU handover, the task should be completed within the range of the current RSU. Assuming that the vehicle is within the RSU coverage range, the estimated residence time of the vehicle within this range is: (Formula 8); In the formula, is the radius of the RSU coverage range, is the horizontal distance between the vehicle and the RSU, which is expressed as: (Formula 9); Wherein, represents the time required for task processing; In the blockchain network designed in this embodiment, the process of executing the consensus mechanism is as Figure 3 shown. The process of executing the consensus mechanism is divided into four stages: prepare request, prepare reply, commit, and generate and broadcast a block; each consensus stage will generate energy consumption and latency. During the consensus process, assume the number of consensus nodes is ( is the maximum tolerable number of malicious nodes), in the prepare request stage, the host (primary node) needs to generate 1 signature and information verification codes (MACs). Generating and verifying a signature and a MAC respectively require and central processing unit (CPU) cycles. Among them, the energy consumption and latency generated in the prepare request stage are respectively: (Formula 10); (Formula 11); Wherein, represents the CPU cycles required by the host, represents the average computing power of the nodes, represents the average computing power of the nodes; The energy consumption and latency generated in the prepare reply stage are respectively: (Formula 12); (Formula 13); Wherein, represents the CPU cycles required by all participants, represents the average size of the block, represents the average size of the transaction; The energy consumption and latency generated in the commit stage are respectively: (Formula 14); (Formula 15); Wherein, represents the CPU cycles required by all participating consensus nodes, represents the maximum tolerable number of malicious nodes; Energy consumption and latency generated during the block generation and broadcasting phase are respectively as follows (Equation 16); (Equation 17); In the formula represents the CPU cycles required by all participants In summary, the total energy consumption and total latency generated during the consensus process are respectively (Equation 18); (Equation 19).

[0022] Furthermore, Step 4 in this embodiment further includes the following specific steps (1) Initialize the population using the random initialization method (Equation 20); In the formula and respectively represent the upper and lower boundaries of the solution space represents a random value within the range of [0, 1] represents row column array population (2) Update the population position First, calculate two adaptive weights and : (Equation 21); (Equation 22); In the formula, the initial fitness evaluation count , represents the maximum evaluation count Then perform the rotational motion : (Equation 23); In the formula is a constant 、 represent the charged particle charge and mass is the Earth's magnetic field strength is the damping factor Then perform the elliptical walk motion : (Formula 24); In the formula, is the global search step size, is the centroid position of the high-energy particle swarm, represents the position where the current particle is located (the th row and the th column), , is the interference generated by factors such as the environment on the particle, and the value range is [0, 1]; Finally, according to the rotational motion and elliptical walk motion, calculate the final population update position : (Formula 25); (3) Population mutation, To avoid the situation of local optimum, the high-energy particles need to perform mutation operations: (Formula 26); In the formula, 、 are all random values in [0, 1], is the mutation probability, is any particle in the particle cluster; Refer to Figure 5-7 , which compares the energy consumption, latency and overhead of the local offloading scheme, average offloading scheme and Aurora optimization scheme under different sizes of offloadable tasks. It can be clearly found that compared with other offloading schemes, the Aurora optimization scheme shows extremely significant advantages. It has a significant reduction in energy consumption, can greatly reduce energy consumption, and effectively improve energy utilization efficiency; it also performs better in terms of latency and can significantly shorten the time consumed for data transmission and processing; at the same time, this scheme has also achieved excellent results in cost control, successfully reducing various cost expenditures in the implementation process and having higher cost performance.

[0023] Example 2 This embodiment provides a vehicle networking computing offloading system, including: an environment and task management module, a decision variable initialization and blockchain storage module, a fitness calculation and blockchain recording module, an optimization iteration and blockchain consensus module, and a convergence judgment and result output module; The environment and task management module is used to construct the vehicle networking system architecture, initialize the parameters of vehicle nodes, edge computing nodes, and blockchain nodes, collect vehicle node computing task information, and interact with the blockchain network to record task data; The decision variable initialization and blockchain storage module is responsible for defining and initializing the offloading decision variable matrix and algorithm parameters, and encrypting and storing them in the blockchain network; The fitness calculation and blockchain recording module is used to calculate the fitness value according to the offloading decision variables, consider various cost factors and record key data during the evaluation process in the blockchain. Among them, the various cost factors include: computing energy consumption cost (such as local computing energy consumption , MEC server computing energy consumption and blockchain consensus energy consumption ), computing delay cost (such as local computing delay , MEC server computing delay and blockchain consensus delay ); The optimization iteration and blockchain consensus module is used to implement the iteration of the Aurora optimization algorithm, including decision variable update, new solution evaluation and selection, and at the same time use the consensus mechanism in the blockchain network to ensure data consistency; The convergence judgment and result output module is used to judge the convergence of the Aurora optimization algorithm, output the optimal offloading scheme and record it in the blockchain network. If it does not converge, it will feedback to the optimization iteration module for continuous optimization. Among them, the optimization iteration module evaluates the fitness of the newly generated solution by dynamically adjusting the offloading decision variable matrix (including resource allocation scheme, blockchain node selection strategy), considers energy consumption, delay, and communication overhead factors, and selects the optimal solution as the optimal scheme for the current iteration.

[0024] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for offloading computing in an Internet of Vehicles, characterized in that: The following steps are involved: Step 1: Build the Internet of Vehicles system architecture including vehicles, roadside units, edge servers and blockchain networks; Step 2: Define the calculation unloading decision variables, using Indicates vehicle Whether the task is offloaded to the local or edge server; Step 3: Based on the offloading decision variables in step 2, calculate the fitness value of each vehicle node task offloading scheme, with the goal of minimizing the total overhead of the system, taking into account the task delay, energy consumption and blockchain data processing overhead factors, and encrypt the key intermediate data in the fitness value calculation process and record it in the blockchain network, and finally verify and store it through the consensus mechanism in the blockchain network; Step 4: Use the Aurora optimization algorithm to iterate. In each iteration, according to the optimal solution and individual solution in the current population, the unloading decision variables are updated through the motion behavior simulation mechanism of the Aurora optimization algorithm, and mutation operations are performed at the same time. Finally, the update information is broadcast to the blockchain network; Step 5: Determine whether the Aurora optimization algorithm has reached the preset maximum number of iterations; When the convergence condition is reached, the globally optimal computation offloading solution is output, and the final offloading solution is encrypted and recorded in the blockchain network; If the convergence condition is not met, return to step 4 and continue iterating.

2. The method for offloading computing in an Internet of Vehicles according to claim 1, characterized in that: The energy consumption and delay in step 3 are calculated as follows: vehicle The energy consumption of task offloading is ; The resulting delay is ; In the formula, is the task offloading decision variable, which refers to offloading to the local or edge server. =0, indicating that the task is calculated locally, and the energy consumption is , the delay is ;when =1, indicating that the task is calculated on the edge server, and the energy consumption is , the delay is ,in, It represents the energy consumption of offloading tasks to local computing. Indicates the energy consumption of offloading tasks to MEC servers for computing. Represents the energy consumed by executing the consensus mechanism in the blockchain network, Indicates the delay caused by offloading the task to the local computer for calculation. Indicates the delay caused by offloading the task to the MEC server for calculation. Represents the delay in executing the consensus mechanism in the blockchain network.

3. The method for offloading computing in an Internet of Vehicles according to claim 1, characterized in that: The calculation method of the cost in step 3 is as follows: The total cost of the system is ; In the formula, and is the coefficient of energy consumption and delay, where , and .

4. The method for offloading computing in an Internet of Vehicles according to claim 1, characterized in that: The update and mutation operations of the Aurora optimization algorithm in step 4 are specifically as follows: Step 41. Update operation: The update operation includes rotation movement and auroral ellipse walk, in which the rotation movement takes the optimal solution in the current population as a reference point and guides the individual solution to perform rotational position adjustment around it; Step 42: Mutation operation: The mutation operation includes particle collision, specifically, randomly selecting two individual solutions and simulating the effect of particle collision between the two individual solutions.

5. The method for offloading computing in an Internet of Vehicles according to claim 1, characterized in that: The threshold in the convergence condition in step 5 is adjusted according to the accuracy requirements, stability requirements and blockchain network performance of the Internet of Vehicles system; the specific adjustment method is as follows: when the Internet of Vehicles system has high requirements for calculation accuracy, the threshold is set to low, and when the system has high requirements for stability, the threshold is set to high.

6. The method for offloading computing in an Internet of Vehicles according to claim 1, characterized in that: The consensus mechanism in the blockchain network adopts the authorized Byzantine fault-tolerant algorithm.

7. A vehicle networking computing offloading system for implementing the method of claim 1, characterized in that: include: Environment and task management module, decision variable initialization and blockchain storage module, fitness calculation and blockchain recording module, optimization iteration and blockchain consensus module, convergence judgment and result output module; The environment and task management module is used to build the Internet of Vehicles system architecture, initialize the parameters of vehicle nodes, edge computing nodes, and blockchain nodes, collect vehicle node computing task information, interact with the blockchain network, and record task data; The decision variable initialization and blockchain storage module is responsible for defining and initializing the unloading decision variable matrix and algorithm parameters, and encrypting and storing them in the blockchain network; The fitness calculation and blockchain recording module is used to calculate the fitness value according to the unloading decision variable, consider multiple cost factors and record the key data of the evaluation process in the blockchain, wherein the multiple cost factors include: calculation energy consumption cost, calculation delay cost; The optimization iteration and blockchain consensus module is used for the iteration of the Aurora optimization algorithm, including decision variable update, new solution evaluation and selection, while using the consensus mechanism in the blockchain network to ensure data consistency; The convergence judgment and result output module is used to judge the convergence of the Aurora optimization algorithm, output the optimal unloading solution and record it in the blockchain network. If it does not converge, it will be fed back to the optimization iteration module for further optimization. The optimization iteration module dynamically adjusts the unloading decision variable matrix, evaluates the fitness of the newly generated solution, considers energy consumption, delay, and communication overhead factors, and selects the optimal solution as the optimal solution for the current iteration.

Images