Parallel method and system for latency optimization task offloading and resource allocation in vehicle-mounted edge computing environment

Abstract

The application discloses a time delay optimization task unloading and resource allocation parallel method and system for a vehicle-mounted edge computing environment, and the method comprises the following steps: S1, a model for cooperative computing of a target vehicle and idle vehicles under RSU coverage is established; S2, a motion model of the target vehicle and the idle vehicles, a computing and unloading option of the target vehicle, a task allocation ratio and a communication range are configured; S3, a transmission rate formula is given, and a task transmission time delay is obtained; S4, a task time delay is calculated; S5, an optimization model for joint task allocation and time delay minimization is established by taking the minimization of the parallel system time delay as a target, according to the task allocation ratio constraint and the time delay constraint of vehicle task computing and unloading; and S6, the task allocation ratio is solved by making the time delay of local processing of the target vehicle, the time delay of mobile edge computing server processing equipped by a roadside unit and the time delay of processing of the idle vehicles equal. The minimum time delay of the target vehicle for solving the task is obtained.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of wireless communication, and particularly relates to a time delay optimization task offloading and resource allocation parallel method and system for a vehicle-mounted edge computing environment. BACKGROUND

[0002] With the rapid development of wireless communication technology, high-demand services and applications will further test the cellular and wireless networks. Mobile edge computing, as a promising technology for providing computing services in mobile edge networks, is proposed to solve the above problems. Mobile edge computing deploys computing and storage resources at the edge of the mobile network to provide an IT service environment and cloud computing capabilities for the mobile network, thereby providing a network service solution with ultra-low latency and high bandwidth for users, and ultimately minimizing system latency or energy consumption.

[0003] Research on mobile edge computing offloading resource allocation algorithms has made great progress, however, these studies mostly consider tasks as a whole and do not split them, only selecting a single device for offloading computing. In fact, the computing capacity of each computing device is limited, and assigning tasks to a single device for computing will result in longer latency. After considering task segmentation, scheduling and transmission, etc., tasks can be reasonably allocated to multiple computing devices for parallel processing, which can achieve higher efficiency. In addition, the edge computing nodes in these studies are usually fixed, without considering the dynamic nature of edge computing nodes in real life. Based on this, the application proposes a time delay optimization task offloading and resource allocation parallel method and system for a vehicle-mounted edge computing environment. SUMMARY

[0004] In view of the above defects in the prior art, based on a mobile vehicle edge computing network, the application discloses a time delay optimization task offloading and resource allocation parallel method and system for a vehicle-mounted edge computing environment. The application targets the application scenario of mobile vehicles, aims to minimize the parallel system latency, establishes an optimization model of joint task allocation and latency minimization according to the task allocation proportion constraint and the latency constraint of vehicle task computing offloading, and uses a task allocation strategy that makes the task processing time on the local and offloaded devices equal to achieve optimal allocation of task proportions.

[0005] In order to achieve the purpose of the application, the application adopts the following technical solutions:

[0006] A time delay optimization task offloading and resource allocation parallel method for a vehicle-mounted edge computing environment, comprising the following steps:

[0007] S1, model establishment: a model for cooperative computing of target vehicles and idle vehicles under the coverage of a road side unit (RSU) is established;

[0008] S2, initial system configuration: motion model of target vehicle and idle vehicle, computing and offloading options of target vehicle, task allocation ratio and communication range;

[0009] S3, analyze communication model: give the transmission rate formula, get the task transmission delay;

[0010] S4, analyze the computing model: calculate the task delay;

[0011] S5, establish optimization model: minimize parallel system delay as the goal, according to the task allocation ratio constraint and the delay constraint of vehicle task computing and offloading, establish the optimization model of joint task allocation and delay minimization;

[0012] S6, solve the optimization model: by making the delay of target vehicle local processing, the delay of mobile edge computing server (MEC server) processing equipped by roadside unit, and the delay of idle vehicle processing equal, the task allocation ratio is solved.

[0013] Further, in step S1, the model established is specifically:

[0014] Consider a one-way straight road, along one side of the road a roadside unit is deployed, in addition to the target vehicle, there is also a friendly assisting idle vehicle in the system, they all travel in the same direction under the RSU coverage.

[0015] Further, in step S2, the initial system configuration is:

[0016] The vehicle speed is constant, the communication coverage of RSU is considered as a cell, and the RSU is equipped with a fixed mobile edge computing server as an edge computing node. Consider that the target vehicle starts driving from the left end of the RSU coverage area until the rightmost end of the coverage area. The radius of the RSU is R, and the angle between the road and the RSU coverage range is θ, so the driving distance of the vehicle under the RSU is represented as:

[0017] K=2Rcosθ

[0018] Use <D, F> to represent the task generated on the target vehicle, where D represents the input data size and F represents the CPU cycles required to complete the task. The target vehicle can choose to execute its task locally or offload it to the MEC server of the RSU and execute it in parallel. The ratio of task allocation to each device is denoted as A={k0, k1, k2}. k0, k1, k2 represent the proportion of tasks in local computing and offloading to the RSU (MEC server) and idle vehicle, respectively, and the proportion of the three parts is weighted to 1, which can be represented as:

[0019] k0+k1+k2=1

[0020] According to the distribution ratio, the task size allocated to the local calculation and the number of CPU cycles required to complete this part of the task are respectively represented as:

[0021]

[0022] The task size allocated to the MEC server on the RSU and the number of CPU cycles required to complete this part of the task are respectively represented as:

[0023]

[0024] The task size allocated to the idle vehicle and the number of CPU cycles required to complete this part of the task are respectively represented as:

[0025]

[0026] Further, in step S3, the communication model is specifically:

[0027] The orthogonal frequency division multiplexing technology is adopted. Therefore, the data transmission rate of the target vehicle to the road side unit (RSU equipped with MEC server) can be represented as:

[0028]

[0029] The data transmission rate of the target vehicle and the idle vehicle can be represented as:

[0030]

[0031] Wherein, B M , B V2V respectively represent the channel bandwidth of the uplink between the target vehicle and the RSU (MEC server) and the idle vehicle, P represents the transmission power of the target vehicle device data, d TM , d V2V respectively represent the distance between the target vehicle and the RSU (MEC server) and the target vehicle and the idle vehicle, ζ is the path loss index, h M and h V2V respectively represent the channel gain of the target vehicle and the RSU (MEC server) and the idle vehicle, N0 is the additive white noise power. In order to simplify the calculation, R is used to represent d TM , which is represented as:

[0032] d TM = R

[0033] The transmission delay of the part of the task generated by the target vehicle uploaded to the RSU (MEC server) is represented as:

[0034]

[0035] The transmission delay of the part of the task generated by the target vehicle uploaded to the idle vehicle is expressed as:

[0036]

[0037] Further, in step S4, the calculation model is specifically:

[0038] The calculation task <D, F> generated on the target vehicle can be assigned to the local vehicle, the MEC server on the RSU and the idle vehicle SV for parallel processing. The calculation delay corresponding to the local calculation is:

[0039]

[0040] The calculation delay of the part of the task uploaded to the RSU (MEC server) is:

[0041]

[0042] The calculation delay of the part of the task uploaded to the idle vehicle is:

[0043]

[0044] Wherein, f L , f M , f S respectively represent the calculation capability of the target vehicle locally, the MEC server equipped on the RSU and the idle vehicle.

[0045] Further, in step S5, the optimization problem is:

[0046] With the objective of minimizing the delay, the optimization variable is the offloading ratio of each device. In order to make the optimal offloading decision in terms of delay, first, the total delay function is formulated, considering that the local, MEC server and idle vehicle process tasks in parallel, the total delay function is:

[0047]

[0048] Therefore, the optimization problem expression is:

[0049]

[0050] s.t C1:k0+k1+k2=1

[0051] C2:

[0052] C3:

[0053] C4:

[0054] Wherein, v represents the driving speed of the vehicle; constraint condition C1 represents that the distribution proportion weighted sum is 1; constraint condition C2 represents that the local calculation amount cannot exceed the maximum calculation amount; constraint conditions C3 and C4 represent that the uploaded data amount cannot exceed the maximum data amount.

[0055] Further, step S6, solving the optimization model:

[0056] A time delay optimization task offloading and resource allocation parallel algorithm for a vehicle-mounted edge computing environment is used to solve the optimization problem. The shortest time delay load balancing algorithm aims to uniformly distribute tasks to different execution units to achieve simultaneous completion of tasks and reduce the overall execution time. In this case, the tasks are divided into three parts, which are processed by the target vehicle locally, the MEC server and the idle vehicle SV respectively. At this time, when the time delays of the three parts are equal, k0, k1 and k2 are obtained respectively:

[0057]

[0058] Wherein,

[0059] The application also discloses a time delay optimization task offloading and resource allocation parallel system for a vehicle-mounted edge computing environment, which is used for executing the above method and comprises the following modules:

[0060] The model establishing module establishes a model for cooperative calculation of the target vehicle and the idle vehicle under the coverage of a road side unit (RSU).

[0061] The initial system configuration module configures the motion model of the target vehicle and the idle vehicle, the calculation and offloading options of the target vehicle, the task distribution proportion and the communication range.

[0062] The communication model analysis module gives a transmission rate formula to obtain the task transmission time delay.

[0063] The calculation model analysis module calculates the task time delay.

[0064] The optimization model establishing module establishes an optimization model of joint task distribution and time delay minimization by taking the minimization of the parallel system time delay as the target, according to the task distribution proportion constraint and the time delay constraint of vehicle task calculation and offloading.

[0065] The optimization model solving module solves the task distribution proportion by making the time delays of the target vehicle local processing, the mobile edge computing server processing of the road side unit and the idle vehicle processing equal.

[0066] The application has the following advantages:

[0067] (1) The present invention provides a parallel method and system for delay-optimized task offloading and resource allocation in an on-vehicle edge computing environment. By optimizing the task allocation ratio and making full use of various computing resources, the total delay of the target vehicle in completing the task is significantly reduced.

[0068] (2) In intelligent transportation and IoT applications, real-time performance is crucial. This invention uses mobile vehicles as edge computing nodes. By considering the dynamic mobility of vehicles, tasks can be better scheduled to ensure timely unloading and completion of tasks at the target vehicle, reducing latency. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a flowchart of a parallel method for delay optimization task offloading and resource allocation for an on-vehicle edge computing environment in a preferred embodiment of the present invention.

[0070] Figure 2 A diagram showing the relationship between task size and latency under different resource allocation schemes.

[0071] Figure 3 This is a diagram showing the relationship between the local computing power and latency of the target vehicle under different resource allocation schemes.

[0072] Figure 4 This is a diagram showing the relationship between MEC server computing power and latency under different resource allocation schemes.

[0073] Figure 5 The relationship between the computing power and latency of idle vehicle SV under different resource allocation schemes.

[0074] Figure 6 This is a diagram showing the relationship between the target vehicle's transmission power and delay under different resource allocation schemes.

[0075] Figure 7 This is a block diagram of a parallel system for latency-optimized task offloading and resource allocation in a vehicle-mounted edge computing environment, according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0076] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0077] like Figure 1 As shown, this embodiment provides a parallel method for delay optimization task offloading and resource allocation for an in-vehicle edge computing environment, including the following steps:

[0078] S1, model establishment: a model for target vehicle and idle vehicle to perform cooperative calculation under the coverage of a road side unit (RSU) is established. The model is as follows: this embodiment considers a one-way straight road, a road side unit is deployed along one side of the road, and in addition to the target vehicle, there is also an idle friendly assisting vehicle in the system, which all travel in the same direction under the RSU.

[0079] S2, initial system configuration: the communication coverage of the RSU is regarded as a cell, and the RSU is equipped with a fixed mobile edge computing server, which can help the target vehicle to perform its calculation task. It is assumed that the vehicle speed is constant. It is considered that the target vehicle starts to travel from the left end of the RSU coverage area and travels to the rightmost end of the coverage area. The radius of the RSU is R, and at the junction of the road and the coverage range of the RSU, the angle between the road and the radius of the RSU is θ, therefore, the distance traveled by the vehicle under the RSU is represented as:

[0080] K = 2Rcosθ (1)

[0081] In order to make the task calculated locally by the target vehicle completed before leaving the RSU, the constraint can be represented as:

[0082]

[0083] In order to make the transmission process of the target vehicle unloading to the MEC server of the RSU completed before leaving the RSU, the constraint can be represented as:

[0084]

[0085] In order to make the transmission process of the target vehicle unloading to the idle vehicle SV completed before leaving the RSU, the constraint can be represented as:

[0086]

[0087] The task generated on the target vehicle is represented as <D, F>, wherein D represents the input data size, and F represents the CPU cycle required to complete the task. The target vehicle can choose to execute its task locally or unload it to the MEC server of the RSU and execute it in parallel. The proportion of the task allocated to each device is denoted as A = {k0, k1, k2}. k0, k1, and k2 represent the proportion of the task calculated locally and unloaded to the RSU (MEC server) and unloaded to the idle vehicle respectively, and the proportion of the three parts is weighted to 1, which can be represented as:

[0088] k0+k1+k2=1 (5)

[0089] According to the allocation proportion, the task size allocated to local calculation and the CPU cycle number required to complete this part of the task are represented as:

[0090]

[0091] The task size allocated to the MEC server on the RSU and the number of CPU cycles required to complete this part of the task are denoted as:

[0092]

[0093] The task size allocated to the idle vehicle and the number of CPU cycles required to complete this part of the task are denoted as:

[0094]

[0095] S3, Analyzing the communication model: At the physical layer, adaptive modulation and coding are assumed and adopted to support multiple transmission modes. In this system, the uplink and downlink channels between the target vehicle and the road facility unit and the idle vehicle are assumed to be frequency-flat block-fading Rayleigh channels. To transmit data, this embodiment adopts orthogonal frequency division multiplexing technology to ensure the orthogonality between different communication links and effectively reduce the influence of interference. Therefore, the data transmission rate of the target vehicle to the road side unit RSU can be represented as:

[0096]

[0097] The data transmission rate of the target vehicle and the idle vehicle SV can be represented as:

[0098]

[0099] where B M , B V2V represent the channel bandwidth of the uplink between the target vehicle and the RSU (MEC server) and the idle vehicle SV, respectively, P represents the transmit power of the target vehicle device data, d TM , d V2V represent the distance between the target vehicle and the RSU (MEC server) and the target vehicle and the idle vehicle, respectively, ζ is the path loss exponent, h M and h V2V represent the channel gain of the target vehicle and the RSU (MEC server) and the idle vehicle SV, respectively, and N0 is the additive white noise power. For convenience of calculation, R is used to represent d TM , which can be represented as:

[0100] d TM = R (11)

[0101] Therefore, the transmission delay of the task generated by the target vehicle uploaded to the RSU (MEC server) is:

[0102]

[0103] The transmission delay of the task uploaded to the idle vehicle SV is:

[0104]

[0105] S4, analyzing and calculating the model, specifically:

[0106] When the target vehicle has a computing task, the computing task can be split to make the delay of completing the task shortest, therefore, the computing task <D, F> generated on the target vehicle can be simultaneously allocated to the local vehicle, the MEC server on the RSU and the idle vehicle for parallel processing. The computing delay corresponding to the local computing is:

[0107]

[0108] The computing delay of the task transmitted to the RSU (MEC server) part is:

[0109]

[0110] The computing delay of the task transmitted to the idle vehicle part is:

[0111]

[0112] Wherein, f L , f M , f S respectively represent the computing capacity of the local, the MEC server equipped on the RSU and the idle vehicle.

[0113] S5, establishing an optimization model: by making the delay of the target vehicle local processing, the delay of the mobile edge computing server equipped on the roadside unit processing and the delay of the idle vehicle processing equal, the task allocation proportion is solved.

[0114] In order to make the optimal delay unloading decision, first of all, the total delay function is formulated, considering that the local, MEC server and idle vehicle processing task is parallel, the total delay function is:

[0115]

[0116] Therefore, the optimization problem expression is:

[0117]

[0118] Wherein, v represents the driving speed of the vehicle. Constraint condition C1 represents that the weighted sum of the allocation proportion is 1; constraint condition C2 represents that the local computing amount cannot exceed the maximum computing amount; constraint conditions C3-4 represent that the uploaded data amount cannot exceed the maximum data amount.

[0119] S6, solving the optimization model, specifically:

[0120] A latency optimization task offloading and resource allocation parallel algorithm for vehicle-mounted edge computing environment is used to solve the optimization problem. The shortest latency load balancing algorithm aims to evenly distribute tasks to different execution units to achieve simultaneous completion of tasks and reduce overall execution time. In this case, the task is divided into three parts, which are processed by the target vehicle locally, the MEC server, and the idle vehicle SV, respectively. At this time, when the three parts of the latency are equal, k0, k1, and k2 can be obtained as follows:

[0121]

[0122] wherein,

[0123] Lemma: When , the optimization target total latency T can take the minimum value.

[0124] Proof: Assume that at a certain point , there is Consider the first-order derivatives of T with respect to k0, k1, and k2 and

[0125] According to the definition, there is:

[0126]

[0127] According to the definition of , the values of these derivatives can be calculated as follows:

[0128]

[0129] Therefore, at , these derivatives are all positive, that is, the three short times all increase with the increase of k value. However, k0+k1+k2=1, which means that as long as one of the three proportions decreases, the other will increase, and then T will increase.

[0130] In summary, when and k0+k1+k2=1, the total latency T is locally minimal.

[0131] The parallel method for task offloading and resource allocation of the time delay optimization in the vehicle-mounted edge computing environment according to the present application obtains the optimal local, RSU, V2V (LRV) cooperative strategy. In the simulation analysis, in order to analyze the performance of the LRV allocation strategy, it is compared with four schemes. The four benchmark algorithms are as follows: scheme 1: a strategy of only MEC server and idle vehicle SV auxiliary computing; scheme 2: a strategy of only target vehicle local and MEC server computing; scheme 3: a strategy of only target vehicle local and idle vehicle SV computing; and scheme 4: a strategy of only target vehicle local computing.

[0132] Figure 2 The relationship between the task size and the time delay under the five schemes is plotted. It can be noted that, as the task size increases, the time delay of the V2V, RSU cooperative strategy is gradually greater than that of the local, RSU cooperative strategy, because when the task size is small, the transmission time delay and the computing time delay of the offloading to other device computing are greater than the time delay of the direct local computing, but as the task size becomes large, the offloading to the MEC server for computing has stronger computing capacity, so that the total time delay is smaller. As can be seen from the figure, compared with the other four schemes, the method designed in the present application has lower task processing time delay.

[0133] Figure 3 、 4 , 5 the influence of different local computing capacity, different MEC computing capacity, different idle vehicle computing capacity and different target vehicle transmission power on the total time delay is plotted respectively. As can be seen from Figure 3 、 4 , 5 when the local computing capacity, the MEC computing capacity and the idle vehicle SV computing capacity are enhanced, the total time delay of the cooperative strategy containing the corresponding computing capacity is correspondingly smaller, and compared with the other four schemes, the method designed in the present application has lower task processing time delay at the same period.

[0134] Figure 6 The relationship between the target vehicle transmission power and the time delay under the five schemes is shown. It can be seen that the increase of the transmission power has little influence on the total time delay, and the total time delay only slightly decreases, because the total time delay is mainly determined by the processing time of the task, not only the transmission delay. Changing the transmission power does not directly affect the processing time of the task on the local, idle vehicle SV or MEC server. Therefore, the change of the transmission power may have limited influence on the total time delay. Under the limited influence, the method designed in the present application still has the lowest time delay.

[0135] As shown in Figure 7 , the embodiment discloses a time delay optimization task offloading and resource allocation parallel system for a vehicle-mounted edge computing environment, for executing the above-mentioned method embodiment, comprising the following modules:

[0136] A model establishing module: a model of target vehicles and idle vehicles performing cooperative calculation under the coverage of a road side unit (RSU) is established;

[0137] An initial system configuration module: a motion model of the target vehicles and the idle vehicles, a calculation and unloading option of the target vehicles, a task allocation ratio, and a communication range are configured;

[0138] A communication model analysis module: a transmission rate formula is given, and a task transmission delay is obtained;

[0139] A calculation model analysis module: a task delay is calculated;

[0140] An optimization model establishing module: an optimization model of joint task allocation and minimum delay is established, with the minimum parallel system delay as a target, according to a task allocation ratio constraint and a delay constraint of vehicle task calculation and unloading;

[0141] An optimization model solving module: the task allocation ratio is solved by making the delay of local processing of the target vehicles, the delay of mobile edge computing server processing equipped in the RSU, and the delay of idle vehicle processing equal.

[0142] Other contents of the embodiment can refer to the above method embodiments.

[0143] In summary, the resource allocation method designed by the application can significantly improve the device resource utilization rate and make up for the deficiency of the single unloading device of the existing edge computing resource allocation algorithm.

[0144] The preferred embodiments and principles of the application are described in detail above, and for those skilled in the art, the specific implementation manner can be changed according to the idea provided by the application, and these changes should be considered as the protection scope of the application.

Claims

1. A latency-optimized task offloading and resource allocation parallel method for a vehicle-mounted edge computing environment, characterized in that: The method comprises the following steps: S1, model establishment: a model for target vehicle and idle vehicle to perform cooperative calculation under a road side unit (RSU) coverage is established; S2, initial system configuration: a motion model of the target vehicle and the idle vehicle, a calculation and unloading option of the target vehicle, a task allocation ratio and a communication range are configured; S3, analysis of the communication model: a transmission rate formula is given, and a task transmission delay is obtained; S4, analysis of the calculation model: a task calculation delay is obtained; S5, establishment of an optimization model: an optimization model for joint task allocation and delay minimization is established, with the objective of minimizing the parallel system delay, according to the task allocation ratio constraint and the delay constraint of vehicle task calculation and unloading; S6, solving of the optimization model: the task allocation ratio is solved by equating the delay of local processing of the target vehicle, the delay of processing by the mobile edge computing server of the RSU and the delay of processing by the idle vehicle; In step S1, the model for cooperative calculation is as follows: A one-way straight road is provided, and a road side unit is arranged along one side of the road; in addition to the target vehicle, there is also an idle vehicle for assistance, and they all travel in the same direction under the RSU coverage; In step S2, the initial system configuration is as follows: The vehicle speed is constant, the communication coverage of the RSU is regarded as a cell, and the RSU is provided with a fixed mobile edge computing server as an edge computing node; it is assumed that the target vehicle starts to travel from the left end of the RSU coverage area and travels to the right end of the coverage area; the radius of the RSU is R, and the angle between the road and the radius of the RSU at the intersection of the road and the coverage range of the RSU is θ, so the distance of the vehicle traveling under the RSU is represented as: K=2Rcosθ The task generated on the target vehicle is represented as <D, F>, wherein D represents the input data size, and F represents the CPU cycle required to complete the task; the target vehicle selects to execute the task locally or to unload the task to the MEC server of the RSU for parallel execution; the allocation ratio of the task to each device is represented as A={k0, k1, k2}, wherein k0, k1 and k2 respectively represent the proportions of the task in local calculation, unloading to the RSU and unloading to the idle vehicle, and the proportions of the three parts are weighted as 1, represented as: k0+k1+k2=1 According to the allocation ratio, the task size allocated to local calculation and the number of CPU cycles required to complete the task are respectively represented as: The task size allocated to the MEC server of the RSU and the number of CPU cycles required to complete the task are respectively represented as: The task size allocated to the idle vehicle and the number of CPU cycles required to complete the task are respectively represented as: In step S3, the communication model is as follows: The orthogonal frequency division multiplexing technology is adopted; the data transmission rate of the target vehicle to the road side unit is represented as: The data transmission rate of the target vehicle and the idle vehicle is represented as: wherein B M , B V2V respectively represent the channel bandwidth of uplink between the target vehicle, RSU and idle vehicle, P represents the transmission power of the target vehicle device data, d TM , d V2V respectively represent the distance between the target vehicle and RSU and the target vehicle and idle vehicle, ζ is the path loss index, h M , h V2V respectively represent the channel gain of the target vehicle, RSU and idle vehicle, N0 is the additive white noise power; d TM is represented by R as follows: d TM = R The transmission delay of the part of the task generated by the target vehicle uploaded to the RSU is represented as: The transmission delay of the part of the task generated by the target vehicle uploaded to the idle vehicle is represented as: In step S4, the generated computing task <D, F> on the target vehicle is assigned to the local vehicle, the MEC server on the RSU and the idle vehicle for parallel processing; the computing time delay corresponding to the local computing is: The computing time delay of the task transmitted to the RSU is: The computing time delay of the task transmitted to the idle vehicle is: where f L , f M , f S represent the computing power of the target vehicle locally, the RSU equipped MEC server, and the idle vehicles, respectively; Step S5 is specifically: Taking the minimum delay as the target, the optimization variable is the unloading proportion of each device; the total delay function is formulated, considering that the local, MEC server and idle vehicle processing tasks are parallel, the total delay function is: Therefore, the optimization problem expression is: Wherein, v represents the driving speed of the vehicle; constraint condition C1 represents that the weighted sum of the allocation proportion is 1; constraint condition C2 represents that the local computing amount cannot exceed the maximum computing amount; constraint conditions C3 and C4 represent that the uploaded data amount cannot exceed the maximum data amount; In step S6, the optimization model solving process is as follows: The task is divided into three parts, which are processed by the target vehicle locally, the MEC server and the idle vehicle; when the time delays of the three parts are equal, k0, k1 and k2 are obtained as: wherein, 2. A latency-optimized task offloading and resource allocation parallel system for vehicular edge computing environment, for performing the method of claim 1, characterized in that: Comprise the following modules: The model establishing module establishes the model of the target vehicle and the idle vehicle for cooperative computing under the coverage of a road side unit (RSU); The initial system configuration module configures the motion model of the target vehicle and the idle vehicle, the computing and unloading options of the target vehicle, the task allocation proportion and the communication range; The communication model analysis module gives the transmission rate formula to obtain the task transmission time delay; The computing model analysis module calculates the task time delay; The optimization model establishing module takes the minimum parallel system delay as the target, establishes the optimization model of joint task allocation and minimum delay according to the task allocation proportion constraint and the time delay constraint of vehicle task computing and unloading; The optimization model solving module solves the task allocation proportion by making the time delays of the target vehicle local processing, the mobile edge computing server processing of the road side unit and the idle vehicle processing equal.

Images