Resource allocation optimization method, device and storage medium based on mobile edge computing

Abstract

The present invention discloses a resource allocation optimization method, device and storage medium based on mobile edge computing, the method comprising: constructing a system model based on mobile edge computing, the system model comprising at least one edge server and at least one base station; based on the system model, calculating the transmission delay of the task on the wireless side and the local delay of the task being executed locally; calculating the unloading delay of the task to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal; establishing the optimization goal and constraint conditions of minimizing the total energy consumption of the system model according to the transmission delay, local delay and unloading delay; calculating the optimal solution of the optimization goal according to the constraint conditions, and obtaining the optimal resource allocation strategy of the system model. The present invention jointly considers user mobility, unloading strategy issues and computing resource allocation, takes the minimum total energy consumption of the system processing business as the optimization goal, and can quickly obtain the resource allocation strategy.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technologies, and in particular to a resource allocation optimization method, device and storage medium based on mobile edge computing. Background Art

[0002] Due to the limited computing power and battery capacity of mobile devices, mobile cloud computing has emerged to solve the problem of insufficient local terminal computing power caused by computing-intensive applications brought about by the 5G era. Mobile cloud computing refers to offloading some or all tasks of mobile terminals to cloud servers to solve the problem of insufficient computing power. However, this method has limitations on backhaul line resources, and the offloaded tasks will have a high latency, which does not meet the requirements of uRLLc tasks. Subsequently, some scholars proposed mobile edge computing, which solves the problems of user latency and limited computing resources by offloading applications to edge servers close to the user side.

[0003] However, since traditional offloading management is separated from user mobility management, that is, task offloading does not take user mobility into consideration, the final decision result often deviates from the optimal value of the system. Therefore, it is urgent to find a new mechanism to optimize the offloading strategy of user mobility, so as to reduce the network switching cost, increase network capacity, and form an optimal configuration of network resources. Summary of the invention

[0004] The technical problem to be solved by the embodiments of the present invention is to provide a resource allocation optimization method, device and storage medium based on mobile edge computing, which jointly considers user mobility, offloading strategy issues and computing resource allocation, takes the minimum total energy consumption of system processing business as the optimization goal, and can quickly obtain a resource allocation strategy.

[0005] In order to achieve the above object, an embodiment of the present invention provides a resource allocation optimization method based on mobile edge computing, including:

[0006] Constructing a system model based on mobile edge computing, the system model includes at least one edge server and at least one base station, the edge server and the base station providing computing and communication services for all user terminals within their coverage areas;

[0007] Based on the system model, calculating the transmission delay of the task on the wireless side and the local delay of the task being executed locally;

[0008] Calculate the unloading delay of the task to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal;

[0009] According to the transmission delay, the local delay and the unloading delay, an optimization goal and constraint conditions for minimizing the total energy consumption of the system model are established;

[0010] The optimal solution of the optimization target is calculated according to the constraint conditions to obtain the optimal resource allocation strategy of the system model.

[0011] As an improvement to the above solution, each task generated by the user terminal is represented by a triple Indicates that, D i Indicates the task size, C i Indicates the computing resources required for the task, tmax i Indicates the maximum tolerable delay of the task;

[0012] Using 0-1 variable a ij Indicates the way the task is executed, namely:

[0013]

[0014] in, It means that task i can only choose the candidate edge server set Br in the system to unload or choose local unloading; the set V = {1, 2, ..., v} represents all user terminals within the coverage area of ​​the edge server and the base station; the set A = {1, 2, ..., m} represents all edge servers in the entire system.

[0015] As an improvement of the above solution, the calculation of the transmission delay of the task on the wireless side and the local delay of the task being executed locally based on the system model specifically includes:

[0016] Obtain the bandwidth B of the upload link channel, the Gaussian white noise power N0, the transmit power G of the user terminal, the channel fading factor h of the upload link, and the average distance d_ from the start to the end of the user terminal transmission task in the system model;

[0017] According to the channel fading factor h, the average distance And the formula Calculate the channel gain parameter H i ;

[0018] According to the bandwidth B, the Gaussian white noise power N0, the transmission power G, the channel gain parameter H i And the formula Calculate the average data transfer rate r of the task i ;

[0019] According to the task size D i , the average data transmission rate r i And the formula Calculate the transmission delay of the task on the wireless side

[0020] According to the computing resources C required by the task i , the processing capability of the user terminal i And the formula Calculate the local latency of the task when it is executed locally

[0021] As an improvement of the above solution, the calculation of the unloading delay of the task to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal specifically includes:

[0022] The normal distribution probability density function f(x) is used to represent the preference of the user terminal for the time it stays in a certain area, that is, Among them, μ c represents the average value, δ c represents variance;

[0023] The probability of the user terminal staying in a certain area at a certain time is obtained according to the normal distribution probability density function f(x). Among them, x1 is the previous moment of x2;

[0024] According to the current moving speed v of the user terminal i , the distance l from the location of the user terminal in the coverage area of ​​a certain edge server to the location outside the coverage area, and the formula Calculate the current stay time of the user terminal at the location

[0025] According to the stay probability and the current stay time Calculating and obtaining a weighting factor for allocating computing resources to the user terminal;

[0026] According to the transmission delay, the weighting factor and the formula Calculate the offloading delay of the task to the edge server in, represents the transmission delay of the task on the wireless side, C i represents the computing resources required for the task, ε ir represents the weighting factor, Br represents the set of candidate edge servers in the system, and f MEC Represents the computing resources of the edge server.

[0027] As an improvement of the above solution, the calculation formula of the weighting factor is:

[0028]

[0029] Among them, tmax i represents the maximum tolerable delay of the task, and p(t) represents the probability of the user terminal staying in a certain area at time t.

[0030] As an improvement of the above solution, the optimization goal and constraint conditions for minimizing the total energy consumption of the system model are established according to the transmission delay, the local delay and the unloading delay, specifically including:

[0031] According to the local delay and formula Calculate the first energy consumption of the task processed locally in, represents the local delay, P i Indicates the power of the user terminal processing tasks;

[0032] According to the transmission delay, the unloading delay and the formula Calculate the second energy consumption of the task processed by the edge server in, represents the offloading delay of the task to the edge server, represents the transmission delay of the task on the wireless side, P MEC represents the power of edge server processing tasks, P t represents the power when the user terminal sends data to the edge server, and P0 represents the power in the idle state when waiting for the edge server to process the task;

[0033] According to the first energy consumption And the second energy consumption The optimization objective and constraints of minimizing the total energy consumption of the system model are established.

[0034] As an improvement of the above scheme, the optimization goal is

[0035] The constraints are

[0036] in, Indicates the current stay time. Indicates the transmission delay of the task on the wireless side.

[0037] The embodiment of the present invention also provides a resource allocation optimization device based on mobile edge computing, including:

[0038] A model building module, used to build a system model based on mobile edge computing, wherein the system model includes at least one edge server and at least one base station, and the edge server and the base station provide computing and communication services for all user terminals within their coverage areas;

[0039] A first calculation module, used to calculate the transmission delay of the task on the wireless side and the local delay of the task being executed locally based on the system model;

[0040] A second calculation module, configured to calculate an unloading delay of the task to be unloaded to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal;

[0041] An optimization module, used to establish an optimization goal and constraint conditions for minimizing the total energy consumption of the system model according to the transmission delay, the local delay and the unloading delay;

[0042] The third calculation module is used to calculate the optimal solution of the optimization target according to the constraint conditions to obtain the optimal resource allocation strategy of the system model.

[0043] An embodiment of the present invention also provides a resource allocation optimization device based on mobile edge computing, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements any of the resource allocation optimization methods based on mobile edge computing described above.

[0044] An embodiment of the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any of the above-mentioned resource allocation optimization methods based on mobile edge computing.

[0045] Compared with the prior art, the beneficial effect of a resource allocation optimization method, device and storage medium based on mobile edge computing provided by the embodiment of the present invention is that: user mobility, unloading strategy issues and computing resource allocation are jointly considered, different computing resources are allocated according to the length of the user terminal's stay time, and the shorter the stay time, the more computing resources are allocated to the user terminal, thereby meeting the user terminal's requirements for task computing delay during movement. In addition, the embodiment of the present invention also takes into account the uncertainty of user terminal movement. Simply using the current speed of the user terminal to calculate the user terminal's stay time has great uncertainty. Therefore, the user terminal stay time is obtained through big data to construct a probability density function, thereby reflecting the user terminal's stay probability at each location. By integrating the user terminal's stay probability and the user terminal's current stay time as weighted factors for edge server cluster resource allocation, and applying the weighted factors to the edge server cluster processing service delay. Based on the server computing delay obtained by the weighted factor, the unloading strategy and the resource allocation strategy are combined, and the total energy consumption of the system processing service is minimized as the optimization goal, and the resource allocation strategy can be quickly obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of a preferred embodiment of a resource allocation optimization method based on mobile edge computing provided by the present invention;

[0047] Figure 2 It is a structural schematic diagram of a preferred embodiment of a resource allocation optimization device based on mobile edge computing provided by the present invention;

[0048] Figure 3 It is a structural schematic diagram of another preferred embodiment of a resource allocation optimization device based on mobile edge computing provided by the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] See also Figure 1 , Figure 1 The present invention provides a method for optimizing resource allocation based on mobile edge computing, and the method comprises:

[0051] S1, constructing a system model based on mobile edge computing, wherein the system model includes at least one edge server and at least one base station, wherein the edge server and the base station provide computing and communication services for all user terminals within their coverage areas;

[0052] S2, based on the system model, calculating the transmission delay of the task on the wireless side and the local delay of the task being executed locally;

[0053] S3, calculating the unloading delay of the task to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal;

[0054] S4, establishing an optimization goal and constraint conditions for minimizing total energy consumption of the system model according to the transmission delay, the local delay and the unloading delay;

[0055] S5, calculating the optimal solution of the optimization target according to the constraint conditions to obtain the optimal resource allocation strategy of the system model.

[0056] Specifically, the embodiment of the present invention first constructs a system model based on mobile edge computing, which includes at least one edge server and at least one base station. The edge server and the base station provide computing and communication services for all user terminals within their coverage. Then, based on the system model, the transmission delay of the task on the wireless side and the local delay of the task executed locally are calculated. According to the calculated transmission delay, the stay probability of the user terminal, and the current stay time, the unloading delay of the task to the edge server is calculated, wherein the stay probability of the user terminal is calculated by constructing a probability density function by obtaining the stay time of the user terminal through big data. Secondly, according to the transmission delay of the task on the wireless side, the local delay of the task executed locally, and the unloading delay of the task to the edge server, the optimization goal and constraint conditions of minimizing the total energy consumption of the system model are established. Finally, the optimal solution of the optimization goal is calculated according to the constraint conditions, and then the optimal resource allocation strategy of the system model is obtained.

[0057] The embodiment of the present invention takes into account the impact of user mobility on task computing delay, jointly considers user mobility, offloading strategy issues and computing resource allocation, and constructs a probability density function by obtaining the user terminal residence time through big data, thereby reflecting the user terminal's residence probability at each location. By combining the user terminal's residence probability and the user terminal's current residence time as weighting factors for edge server cluster resource allocation, and applying the weighting factors to the edge server cluster's service processing delay, the user terminal's requirements for task computing delay during mobility are met. With the lowest total energy consumption of the system processing services as the optimization goal, a resource allocation strategy can be quickly obtained.

[0058] In another preferred embodiment, each task generated by the user terminal is represented by a triple Indicates that, D i Indicates the task size, C i Indicates the computing resources required for the task, tmax i Indicates the maximum tolerable delay of the task;

[0059] Using 0-1 variable a ij Indicates the way the task is executed, namely:

[0060]

[0061] in, It means that task i can only choose the candidate edge server set Br in the system to unload or choose local unloading; the set V = {1, 2, ..., v} represents all user terminals within the coverage area of ​​the edge server and the base station; the set A = {1, 2, ..., m} represents all edge servers in the entire system.

[0062] Specifically, the embodiment of the present invention uses a set A = {1, 2, ..., m} to represent all edge servers of the entire system, and the computing resources of each edge server are f MEC , the set V = {1,2,…,v} represents all user terminals within the coverage of the edge server and the base station, and the computing power of each user terminal is f i The task generated by each user terminal is represented by a triple Indicates that, D i Indicates the task size, C i Indicates the computing resources required for the task, tmax i Indicates the maximum tolerable delay of the task.

[0063] Using 0-1 variable a ij Indicates the way the task is executed, namely:

[0064]

[0065] in, It means that task i can only choose to uninstall from the candidate edge server set Br in the system or choose to uninstall locally.

[0066] It should be noted that the candidate edge server refers to an edge server that is available within the service range of the user terminal. In mobile communications, the range where a user terminal is located is generally covered by 6-14 base stations, but when the user terminal uses the service, there is only one main base station for service. Therefore, when the edge server of the main base station where the user terminal is located cannot accept new tasks, other base stations that can be covered (edge ​​servers are hung on the base station side) will receive the tasks of the user terminal wirelessly and unload the tasks on the server on the base station side. The embodiment of the present invention also proposes a concept of edge servers collaboratively processing tasks unloaded by user terminals. The unloading of user terminal tasks to the server is not a single server processing, but a server cluster processing concept. The tasks of the user terminal can be divided into different candidate edge servers for processing. Br is a set of candidate edge servers, and Br in different regions is different. Therefore, the task processing capacity of different regions is related to the number of candidate edge servers in the region.

[0067] In another preferred embodiment, the S2, based on the system model, calculates the transmission delay of the task on the wireless side and the local delay of the task being executed locally, specifically including:

[0068] S201, obtaining the bandwidth B of the upload link channel in the system model, the Gaussian white noise power N0, the transmission power G of the user terminal, the channel fading factor h of the upload link, and the average distance from the start to the end of the user terminal transmission task

[0069] S202, according to the channel fading factor h, the average distance And the formula Calculate the channel gain parameter H i ;

[0070] S203, based on the bandwidth B, the Gaussian white noise power N0, the transmit power G, and the channel gain parameter H i And the formula Calculate the average data transfer rate r of the task i ;

[0071] S204, according to the task size D i , the average data transmission rate r i And the formula Calculate the transmission delay of the task on the wireless side

[0072] S205, based on the computing resources C required for the task i , the processing capability of the user terminal i And the formula Calculate the local latency of the task when it is executed locally

[0073] Specifically, obtain the bandwidth B of the upload link channel, the Gaussian white noise power N0, the transmission power G of the user terminal, the channel fading factor h of the upload link, and the average distance from the start to the end of the user terminal transmission task in the system model Since the average data transmission rate of the task is related to the channel gain parameter, we first calculate the channel fading factor h and the average distance And the formula Calculate the channel gain parameter H i Then, according to the bandwidth B of the upload link channel, the Gaussian white noise power N0, the transmit power G of the user terminal, and the channel gain parameter H i And the formula Calculate the average data transfer rate r of the task i Then according to the task size D i , the average data transfer rate of the task r i And the formula Calculate the transmission delay of the task on the wireless side According to the computing resources C required by the task i , the processing capability of the user terminal i And the formula Calculate the local latency of the task when it is executed locally

[0074] In another preferred embodiment, the S3, calculating the unloading delay of the task to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal, specifically includes:

[0075] S301, using a normal distribution probability density function f(x) to represent the preference of the user terminal for the time it stays in a certain area, that is, Among them, μ c represents the average value, δ c represents variance;

[0076] S302, obtaining the probability of the user terminal staying in a certain area at a certain time according to the normal distribution probability density function f(x) Among them, x1 is the previous moment of x2;

[0077] S303, based on the current moving speed v of the user terminal i , the distance l from the location of the user terminal in the coverage area of ​​a certain edge server to the location outside the coverage area, and the formula Calculate the current stay time of the user terminal at the location

[0078] S304: Based on the stay probability and the current stay time Calculating and obtaining a weighting factor for allocating computing resources to the user terminal;

[0079] S305, according to the transmission delay, the weighting factor and the formula Calculate the offloading delay of the task to the edge server in, represents the transmission delay of the task on the wireless side, C i represents the computing resources required for the task, ε ir represents the weighting factor, Br represents the set of candidate edge servers in the system, and f MEC Represents the computing resources of the edge server.

[0080] Specifically, since the user terminal is mobile, considering the issue of the length of time the user terminal moves, the embodiment of the present invention allocates different computing resources according to the length of the residence time, and the shorter the residence time, the more computing resources are allocated to the user terminal. The length of the user terminal's residence time needs to take into account two factors: the probability of residence and the residence time of the user terminal in the area predicted by the speed of the current user terminal. The weighting factor used by the edge server is calculated based on the residence probability and the residence time. Assuming that the user terminal has a preference for the residence time in each area, the normal distribution probability density function is used to represent the preference of the user terminal's residence time in a certain area, that is:

[0081]

[0082] Among them, μ c represents the average value, δ c Represents variance.

[0083] In this embodiment, the parameters can be 0≤x≤24, μ c =10,δ c =0.741, indicating that the user terminal stays in a certain area the longest at 10 o'clock. This is just the stay preference of a certain area. The stay preferences of different areas are different, that is, the average value and variance of the formula are different. Simulation is performed based on historical statistical data, and the average value and variance can be calculated through statistical data.

[0084] Based on the normal distribution probability density function f(x), the probability of a user terminal staying in a certain area at a certain time is obtained Where x1 is the previous moment of x2. According to the current moving speed v of the user terminal i , the distance l from the location where the user terminal is located in the coverage area of ​​a certain edge server to the location outside the coverage area, and the formula Calculate the current residence time of the user terminal at this location Combined with the user terminal's stay probability and current stay time The weighted factor of the computing resources allocated to the user terminal is calculated. According to the transmission delay of the task on the wireless side, the weighted factor of the computing resources allocated to the user terminal and the formula Calculate the offloading delay of the task to the edge server in, represents the transmission delay of the task on the wireless side, C i represents the computing resources required for the task, ε ir represents the weighting factor, Br represents the set of candidate edge servers in the system, and f MEC Indicates the computing resources of the edge server. In the embodiment of the present invention, a task can be divided into different subtasks and offloaded to different edge servers for processing. MEC Indicates the total processing capacity of candidate servers in this area.

[0085] As a preferred solution, the calculation formula of the weighting factor is:

[0086]

[0087] Among them, tmax i represents the maximum tolerable delay of the task, and p(t) represents the probability of the user terminal staying in a certain area at time t.

[0088] Specifically, the user terminal's stay probability and current stay time are combined Calculate the weighted factor for allocating computing resources to the user terminal. The calculation formula is:

[0089]

[0090] Among them, tmax i represents the maximum tolerable delay of the task, and p(t) represents the probability of the user terminal staying in a certain area at time t. The smaller the stay probability, the more computing resources are allocated, and vice versa.

[0091] It should be noted that if the user terminal stays in a certain area for a very short time, that is, less than the data transmission time, the system model will not allocate computing resources to it.

[0092] In another preferred embodiment, the S4, based on the transmission delay, the local delay and the unloading delay, establishes the optimization goal and constraint conditions for minimizing the total energy consumption of the system model, specifically including:

[0093] S401, according to the local delay and formula Calculate the first energy consumption of the task processed locally in, represents the local delay, P i Indicates the power of the user terminal processing tasks;

[0094] S402, according to the transmission delay, the unloading delay and the formula Calculate the second energy consumption of the task processed by the edge server in, represents the offloading delay of the task to the edge server, represents the transmission delay of the task on the wireless side, P MEC represents the power of edge server processing tasks, P t represents the power when the user terminal sends data to the edge server, and P0 represents the power in the idle state when waiting for the edge server to process the task;

[0095] S403, according to the first energy consumption And the second energy consumption The optimization objective and constraints of minimizing the total energy consumption of the system model are established.

[0096] Specifically, when calculating energy consumption, we consider the energy consumption of multiple resource selections when processing task i. First, based on the local latency of the task execution and the formula Calculate the first energy consumption of the task processed locally in, represents the local delay, P i represents the power of the user terminal processing the task. Then, according to the transmission delay of the task on the wireless side, the unloading delay of the task to the edge server, and the formula Calculate the second energy consumption of the task processed by the edge server in, represents the offloading delay of the task to the edge server, represents the transmission delay of the task on the wireless side, P MEC represents the power of edge server processing tasks, P t represents the power when the user terminal sends data to the edge server, and P0 represents the power in the idle state when waiting for the edge server to process the task. And the second energy consumption Establish the optimization objective and constraints of minimizing the total energy consumption of the system model.

[0097] It should be noted that the formula The first part represents the energy consumption of edge service processing tasks, the second part represents the energy consumption of data transmission to the wireless base station, and the third part represents the energy consumption of the user terminal waiting for task processing.

[0098] As a preferred solution, the optimization target is

[0099] The constraints are

[0100] in, Indicates the current stay time. Indicates the transmission delay of the task on the wireless side.

[0101] Specifically, according to the first energy consumption And the second energy consumption The optimization goal of minimizing the total energy consumption of the system model is

[0102] The constraints are

[0103] The first constraint is that the processing delay of each task is less than the maximum delay tolerance value; the second constraint is that the user terminal residence time must be greater than the data wireless transmission time, because if the user terminal stays in a certain area for less than the data transmission time, the system model will not allocate computing resources to it; the third constraint is that task i can only choose to unload from the candidate edge server set Br in the system or choose to unload locally.

[0104] Correspondingly, the present invention also provides a resource allocation optimization device based on mobile edge computing, which can implement all processes of the resource allocation optimization method based on mobile edge computing in the above-mentioned embodiment.

[0105] See also Figure 2 , Figure 2 1 is a schematic diagram of a preferred embodiment of a resource allocation optimization device based on mobile edge computing provided by the present invention. The resource allocation optimization device based on mobile edge computing includes:

[0106] A model building module 201 is used to build a system model based on mobile edge computing, wherein the system model includes at least one edge server and at least one base station, and the edge server and the base station provide computing and communication services for all user terminals within their coverage areas;

[0107] A first calculation module 202, configured to calculate the transmission delay of the task on the wireless side and the local delay of the task being executed locally based on the system model;

[0108] A second calculation module 203 is used to calculate the unloading delay of the task to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal;

[0109] An optimization module 204, configured to establish an optimization target and constraint conditions for minimizing the total energy consumption of the system model according to the transmission delay, the local delay, and the unloading delay;

[0110] The third calculation module 205 is used to calculate the optimal solution of the optimization target according to the constraint conditions to obtain the optimal resource allocation strategy of the system model.

[0111] Preferably, each task generated by the user terminal is represented by a triple Indicates that, D i Indicates the task size, C i Indicates the computing resources required for the task, tmax i Indicates the maximum tolerable delay of the task;

[0112] Using 0-1 variable a ij Indicates the way the task is executed, namely:

[0113]

[0114] in, It means that task i can only choose the candidate edge server set Br in the system to unload or choose local unloading; the set V = {1, 2, ..., v} represents all user terminals within the coverage area of ​​the edge server and the base station; the set A = {1, 2, ..., m} represents all edge servers in the entire system.

[0115] Preferably, the first calculation module 202 specifically includes:

[0116] The acquisition unit 212 is used to obtain the bandwidth B of the upload link channel in the system model, the Gaussian white noise power N0, the transmission power G of the user terminal, the channel fading factor h of the upload link, and the average distance from the start to the end of the user terminal transmission task

[0117] The gain calculation unit 222 is used to calculate the channel fading factor h and the average distance And the formula Calculate the channel gain parameter H i ;

[0118] The transmission rate calculation unit 232 is used to calculate the transmission rate according to the bandwidth B, the Gaussian white noise power N0, the transmission power G, and the channel gain parameter H. i And the formula Calculate the average data transfer rate r of the task i ;

[0119] The transmission delay calculation unit 242 is used to calculate the transmission delay according to the task size D i , the average data transmission rate r i And the formula Calculate the transmission delay of the task on the wireless side

[0120] The local delay calculation unit 252 is used to calculate the required computing resources C according to the task. i , the processing capability of the user terminal i And the formula Calculate the local latency of the task when it is executed locally

[0121] Preferably, the second calculation module 203 specifically includes:

[0122] The probability function construction unit 213 is used to use the normal distribution probability density function f(x) to express the preference of the user terminal for the residence time in a certain area, that is, Among them, μ c represents the average value, δ c represents variance;

[0123] The stay probability calculation unit 223 is used to obtain the stay probability of the user terminal in a certain area at a certain time according to the normal distribution probability density function f(x). Among them, x1 is the previous moment of x2;

[0124] The dwell time calculation unit 233 is configured to calculate the dwell time according to the current moving speed v of the user terminal. i , the distance l from the location of the user terminal in the coverage area of ​​a certain edge server to the location outside the coverage area, and the formula Calculate the current stay time of the user terminal at the location

[0125] The weighted factor calculation unit 243 is used to calculate the weighted factor according to the stay probability and the current stay time. Calculating and obtaining a weighting factor for allocating computing resources to the user terminal;

[0126] The unloading delay calculation unit 253 is used to calculate the unloading delay according to the transmission delay, the weighting factor and the formula Calculate the offloading delay of the task to the edge server in, represents the transmission delay of the task on the wireless side, C i represents the computing resources required for the task, ε ir represents the weighting factor, Br represents the set of candidate edge servers in the system, and f MEC Represents the computing resources of the edge server.

[0127] Preferably, the calculation formula of the weighting factor is:

[0128]

[0129] Among them, tmax i represents the maximum tolerable delay of the task, and p(t) represents the probability of the user terminal staying in a certain area at time t.

[0130] Preferably, the optimization module 204 specifically includes:

[0131] The first energy consumption calculation unit 214 is used to calculate the energy consumption according to the local delay and the formula Calculate the first energy consumption of the task processed locally in, represents the local delay, P i Indicates the power of the user terminal processing tasks;

[0132] The second energy consumption calculation unit 224 is used to calculate the energy consumption according to the transmission delay, the unloading delay and the formula Calculate the second energy consumption of the task processed by the edge server in, represents the offloading delay of the task to the edge server, represents the transmission delay of the task on the wireless side, P MEC represents the power of edge server processing tasks, P t represents the power when the user terminal sends data to the edge server, and P0 represents the power in the idle state when waiting for the edge server to process the task;

[0133] The optimization unit 234 is configured to optimize the energy consumption according to the first And the second energy consumption The optimization objective and constraints of minimizing the total energy consumption of the system model are established.

[0134] Preferably, the optimization target is

[0135] The constraints are

[0136] in, Indicates the current stay time. Indicates the transmission delay of the task on the wireless side.

[0137] In the specific implementation, the working principle, control process and technical effect of the resource allocation optimization device based on mobile edge computing provided in the embodiment of the present invention are the same as those of the resource allocation optimization method based on mobile edge computing in the above-mentioned embodiment, and will not be repeated here.

[0138] See also Figure 3 , Figure 3 It is a structural diagram of another preferred embodiment of a resource allocation optimization device based on mobile edge computing provided by the present invention. The resource allocation optimization device based on mobile edge computing includes a processor 301, a memory 302, and a computer program stored in the memory 302 and configured to be executed by the processor 301. When the processor 301 executes the computer program, the resource allocation optimization method based on mobile edge computing described in any of the above embodiments is implemented.

[0139] Preferably, the computer program can be divided into one or more modules / units (such as computer program 1, computer program 2, ...), and the one or more modules / units are stored in the memory 302 and executed by the processor 301 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the resource allocation optimization device based on mobile edge computing.

[0140] The processor 301 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor 301 can also be any conventional processor. The processor 301 is the control center of the resource allocation optimization device based on mobile edge computing, and various interfaces and lines are used to connect the various parts of the resource allocation optimization device based on mobile edge computing.

[0141] The memory 302 mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory 302 can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, and a flash card (Flash Card), etc., or the memory 302 can also be other volatile solid-state storage devices.

[0142] It should be noted that the above-mentioned resource allocation optimization device based on mobile edge computing may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that Figure 3 The structural diagram is only an example of the above-mentioned resource allocation optimization device based on mobile edge computing, and does not constitute a limitation of the above-mentioned resource allocation optimization device based on mobile edge computing. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components.

[0143] An embodiment of the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the resource allocation optimization method based on mobile edge computing described in any of the above embodiments.

[0144] The embodiment of the present invention provides a resource allocation optimization method, device and storage medium based on mobile edge computing, which jointly considers user mobility, unloading strategy issues and computing resource allocation, allocates different computing resources according to the length of the user terminal's stay time, and the shorter the stay time, the more computing resources are allocated to the user terminal, so as to meet the user terminal's requirements for task computing delay during the mobile process. In addition, the embodiment of the present invention also takes into account the uncertainty of user terminal movement. Simply using the current speed of the user terminal to calculate the user terminal's stay time has great uncertainty. Therefore, the user terminal stay time is obtained through big data to construct a probability density function, so as to reflect the user terminal's stay probability at each location. The user terminal's stay probability and the user terminal's current stay time are used as weighting factors for edge server cluster resource allocation, and the weighting factors are applied to the edge server cluster processing service delay. Based on the server computing delay obtained by the weighting factor, the unloading strategy and the resource allocation strategy are combined, and the total energy consumption of the system processing service is minimized as the optimization goal, so that the resource allocation strategy can be quickly obtained.

[0145] It should be noted that the system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the system embodiment provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0146] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A resource allocation optimization method based on mobile edge computing, characterized in that: include: Constructing a system model based on mobile edge computing, the system model includes at least one edge server and at least one base station, the edge server and the base station providing computing and communication services for all user terminals within their coverage areas; Based on the system model, calculating the transmission delay of the task on the wireless side and the local delay of the task being executed locally; Calculate the unloading delay of the task to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal; According to the transmission delay, the local delay and the unloading delay, an optimization goal and constraint conditions for minimizing the total energy consumption of the system model are established; Calculating the optimal solution of the optimization objective according to the constraint conditions to obtain the optimal resource allocation strategy of the system model; The calculating, based on the system model, the transmission delay of the task on the wireless side and the local delay of the task being executed locally specifically includes: Obtain the bandwidth B of the upload link channel in the system model, the Gaussian white noise power N0, the transmission power G of the user terminal, the channel fading factor h of the upload link, and the average distance from the start to the end of the user terminal transmission task According to the channel fading factor h, the average distance And the formula Calculate the channel gain parameter H i ; According to the bandwidth B, the Gaussian white noise power N0, the transmission power G, the channel gain parameter H i And the formula Calculate the average data transfer rate r of the task i ; According to the task size D i , the average data transmission rate r i And the formula Calculate the transmission delay of the task on the wireless side According to the computing resources C required by the task i , the processing capability of the user terminal i And the formula Calculate the local latency of the task when it is executed locally The step of calculating the unloading delay of the task to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal specifically includes: The normal distribution probability density function f(x) is used to represent the preference of the user terminal for the time it stays in a certain area, that is, Among them, μ c represents the average value, δ c represents variance; The probability of the user terminal staying in a certain area at a certain time is obtained according to the normal distribution probability density function f(x). Among them, x1 is the previous moment of x2; According to the current moving speed v of the user terminal i , the distance l from the location of the user terminal in the coverage area of ​​a certain edge server to the location outside the coverage area, and the formula Calculate the current stay time of the user terminal at the location According to the stay probability and the current stay time Calculating and obtaining a weighting factor for allocating computing resources to the user terminal; According to the transmission delay, the weighting factor and the formula Calculate the offloading delay of the task to the edge server in, represents the transmission delay of the task on the wireless side, C i represents the computing resources required for the task, ε ir represents the weighting factor, Br represents the set of candidate edge servers in the system, and f MEC represents the computing resources of the edge server; The optimization objective and constraint conditions for minimizing the total energy consumption of the system model are established according to the transmission delay, the local delay and the unloading delay, and specifically include: According to the local delay and formula Calculate the first energy consumption of the task processed locally in, represents the local delay, P i Indicates the power of the user terminal processing tasks; According to the transmission delay, the unloading delay and the formula Calculate the second energy consumption of the task processed by the edge server in, represents the offloading delay of the task to the edge server, represents the transmission delay of the task on the wireless side, P MEC represents the power of edge server processing tasks, P t represents the power when the user terminal sends data to the edge server, and P0 represents the power in the idle state when waiting for the edge server to process the task; According to the first energy consumption And the second energy consumption The optimization objective and constraints of minimizing the total energy consumption of the system model are established.

2. The resource allocation optimization method based on mobile edge computing according to claim 1, characterized in that: Each task generated by the user terminal is represented by a triple Indicates that, D i Indicates the task size, C i Indicates the computing resources required for the task, tmax i Indicates the maximum tolerable delay of the task; Using 0-1 variable a ij Indicates the way the task is executed, namely: in, It means that task i can only choose the candidate edge server set Br in the system to unload or choose local unloading; the set V = {1, 2, ..., v} represents all user terminals within the coverage of the edge server and the base station, and v represents the number of user terminals within the coverage of the edge server and the base station; the set A = {1, 2, ..., m} represents all edge servers in the entire system, and m represents the number of edge servers in the entire system.

3. The resource allocation optimization method based on mobile edge computing according to claim 1, characterized in that: The calculation formula of the weighting factor is: Among them, tmax i represents the maximum tolerable delay of the task, and p(t) represents the probability of the user terminal staying in a certain area at time t.

4. The resource allocation optimization method based on mobile edge computing according to claim 2, characterized in that: The optimization goal is The constraints are in, Indicates the current stay time. Indicates the transmission delay of the task on the wireless side.

5. A resource allocation optimization device based on mobile edge computing, characterized in that: include: A model building module, used to build a system model based on mobile edge computing, wherein the system model includes at least one edge server and at least one base station, and the edge server and the base station provide computing and communication services for all user terminals within their coverage areas; A first calculation module, used to calculate the transmission delay of the task on the wireless side and the local delay of the task being executed locally based on the system model; A second calculation module, configured to calculate an unloading delay of the task to be unloaded to the edge server according to the transmission delay and the stay probability and current stay time of the user terminal; An optimization module, used to establish an optimization goal and constraint conditions for minimizing the total energy consumption of the system model according to the transmission delay, the local delay and the unloading delay; A third calculation module is used to calculate the optimal solution of the optimization target according to the constraint conditions to obtain the optimal resource allocation strategy of the system model; The first calculation module specifically includes: An acquisition unit is used to acquire the bandwidth B of the upload link channel in the system model, the Gaussian white noise power N0, the transmission power G of the user terminal, the channel fading factor h of the upload link, and the average distance from the start to the end of the user terminal transmission task A gain calculation unit is used to calculate the channel fading factor h and the average distance And the formula Calculate the channel gain parameter H i ; The transmission rate calculation unit is used to calculate the transmission rate according to the bandwidth B, the Gaussian white noise power N0, the transmission power G, and the channel gain parameter H. i And the formula Calculate the average data transfer rate r of the task i ; The transmission delay calculation unit is used to calculate the transmission delay according to the task size D i , the average data transmission rate r i And the formula Calculate the transmission delay of the task on the wireless side The local delay calculation unit is used to calculate the required computing resources C according to the task i , the processing capability of the user terminal i And the formula Calculate the local latency of the task when it is executed locally The second calculation module specifically includes: The probability function construction unit is used to use the normal distribution probability density function f(x) to express the preference of the user terminal for the residence time in a certain area, that is, Among them, μ c represents the average value, δ c represents variance; A stay probability calculation unit, used to obtain the stay probability of the user terminal in a certain area at a certain time according to the normal distribution probability density function f(x) Among them, x1 is the previous moment of x2; The dwell time calculation unit is used to calculate the dwell time according to the current moving speed v of the user terminal. i , the distance l from the location of the user terminal in the coverage area of ​​a certain edge server to the location outside the coverage area, and the formula T i s =l / v i , calculate the current stay time T of the user terminal at the location i s ; A weighting factor calculation unit is used to calculate the weighting factor according to the stay probability and the current stay time T i s , calculate and obtain a weighting factor for allocating computing resources to the user terminal; An unloading delay calculation unit is used to calculate the transmission delay, the weighting factor and the formula Calculate the offloading delay of the task to the edge server Among them, t i w represents the transmission delay of the task on the wireless side, C i represents the computing resources required for the task, ε ir represents the weighting factor, Br represents the set of candidate edge servers in the system, and f MEC represents the computing resources of the edge server; The optimization module specifically includes: The first energy consumption calculation unit is used to calculate the energy consumption according to the local delay and the formula Calculate the first energy consumption of the task processed locally in, represents the local delay, P i Indicates the power of the user terminal processing tasks; The second energy consumption calculation unit is used to calculate the energy consumption according to the transmission delay, the unloading delay and the formula Calculate the second energy consumption of the task processed by the edge server in, represents the offloading delay of the task to the edge server, represents the transmission delay of the task on the wireless side, P MEC represents the power of edge server processing tasks, P t represents the power when the user terminal sends data to the edge server, and P0 represents the power in the idle state when waiting for the edge server to process the task; An optimization unit is configured to optimize the energy consumption according to the first And the second energy consumption The optimization objective and constraints of minimizing the total energy consumption of the system model are established.

6. A resource allocation optimization device based on mobile edge computing, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the resource allocation optimization method based on mobile edge computing as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the resource allocation optimization method based on mobile edge computing as described in any one of claims 1 to 4.

Images