A short packet-MEC network transmission optimization method based on information age and delay

Abstract

The present invention relates to the field of wireless communications technology and discloses a short packet MEC network transmission optimization method based on information age and delay. The method comprises: initializing relevant parameters of the short packet MEC network, the short packet MEC network comprising a local device and a mobile edge server; generating state update information P based on the state of the mobile edge server; determining the first-order moment of the state update information delay #imgabs0# and the first-order moment of the state update information calculation completion interval #imgabs1# based on the state update information P; determining the average delay and average peak information age of the state update information P; establishing a multi-objective optimization function based on the average delay and average peak information age; iterating the multi-objective optimization function to determine an optimal solution of the multi-objective optimization function, and determining optimal parameters of the short packet MEC network based on the optimal solution of the multi-objective optimization function. The present invention establishes a multi-objective optimization function taking into account delay and information age to obtain an optimal solution, thereby improving network performance.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technologies, and in particular to a short packet-MEC network transmission optimization method based on information age and delay. Background Art

[0002] With the rapid development of IoT technology, a wide variety of smart devices and applications are emerging, leading to an explosive growth in computing demand. Traditional cloud computing architectures, due to their centralized computing model, struggle to meet requirements such as low latency, high bandwidth, and high reliability. Therefore, bringing computing power to the network edge, namely mobile edge computing (MEC), has become an important approach to addressing this growing computing demand. By deploying computing, storage, and network resources at the edge of the network, MEC brings them closer to users, reducing latency, improving network efficiency, and enhancing user experience.

[0003] As a low-latency information transmission method, short packet communication plays a key role in real-time applications. However, from a perception perspective, simply ensuring latency is insufficient to characterize the freshness of information. Therefore, the Age of Information (AoI) has been proposed as a timeliness metric that better reflects information freshness and is receiving increasing attention.

[0004] Currently, studying the latency and AoI performance of short packet MEC (SP-MEC) networks is an active research direction, but existing studies have not fully explored their intrinsic correlation. Summary of the Invention

[0005] Based on this, it is necessary to propose a short packet-MEC network transmission optimization method based on information age and delay to address the above problems.

[0006] A short packet-MEC network transmission optimization method based on information age and delay, the method comprising the following steps:

[0007] Initialize relevant parameters of the short packet-MEC network, which includes local devices and mobile edge servers;

[0008] Generate state update information P according to the state of the mobile edge server;

[0009] Determine the status update information delay according to the status update information P The first-order moment and status update information calculation completion interval The first moment of

[0010] Determining an average delay and an average peak information age of the status update information P;

[0011] Establishing a multi-objective optimization function based on the average time delay and the average peak information age;

[0012] Iterating the multi-objective optimization function to determine an optimal solution of the multi-objective optimization function;

[0013] The optimal parameters of the short packet-MEC network are determined according to the optimal solution of the multi-objective optimization function.

[0014] In the above solution, when the mobile edge server is idle, the local device generates new status update information and performs computational processing on the first part of the information to obtain the processed information;

[0015] The local device offloads the processed information and information other than the first part to the mobile edge server via a wireless communication channel;

[0016] The overall decoding error probability of the transmission process is characterized using finite code length field theory;

[0017] If the status update information fails to be unloaded, the local device retransmits the processed information and information other than the first part, and presets the maximum number of retransmissions allowed. .

[0018] In the above scheme, the parameters related to the initialization short packet-MEC network specifically include: status update information P, status update information size , update information uninstall ratio , static channel gain , block length , block error probability under finite block length system , status update information delay , status update information calculation completion interval , local device calculation time , computing time of mobile edge server , the number of CPU cycles required to process 1NAT task , the local device calculates the frequency , mobile edge server calculation frequency , the maximum number of retransmissions allowed , iterate index .

[0019] In the above solution, the state update information delay is determined according to the state update information P. The first-order moment and status update information calculation completion interval The first-order moments of include:

[0020] Determine the status update information delay according to the following expression The first moment of :

[0021]

[0022] in, is the block transfer time, is the maximum number of transmissions, Status update information delay The first moment of Calculate time for the local device, Calculate time for mobile edge servers, is the block error probability under the finite block length system, Transmitting status information The probability of failure every time;

[0023] Determine the status update information calculation completion interval according to the following expression The first moment of :

[0024]

[0025] in, Calculates time for the local device, Calculate time for mobile edge servers, is the block transfer time, is the block error probability under the finite block length system, Calculate completion interval for status update information The first moment of Transmitting status information The probability of failure.

[0026] In the above solution, determining the average delay and average peak information age of the status update information P specifically includes:

[0027] The average delay of the status update information P is determined according to the following expression:

[0028]

[0029] in, Calculate time for the local device, Calculate time for mobile edge servers, is the block transfer time, is the block error probability under the finite block length system, is the maximum number of transmissions, is the average delay of the state update information P, Status update information delay The first moment of Transmitting status information The probability of failure each time;

[0030] The average peak information age of the status update information P is determined according to the following expression:

[0031]

[0032] in, Status update information delay The first moment of Calculate completion interval for status update information The first moment of Calculate time for the local device, Calculate time for mobile edge servers, is the block transfer time, is the maximum number of transmissions, is the block error probability under the finite block length system, is the average peak information age of the state update information P, Transmitting status information The probability of failure.

[0033] In the above solution, the multi-objective optimization function is established according to the average delay and the average peak information age, specifically including:

[0034] The multi-objective optimization function is based on the block length , the maximum number of transmission times and the uninstall ratio Minimize the weighted sum of the average peak information age and the average delay:

[0035]

[0036] in, is the weight factor, and are the lower and upper limits of the packet length respectively, is the maximum number of retransmissions allowed, is the average delay of the state update information P, is the average peak information age of the state update information P, is a multi-objective optimization function, st represents the following expression as a constraint, C1 is the first constraint, C2 is the second constraint, and C3 is the third constraint. To update the information uninstall ratio, is the maximum number of transmissions, is the block length, N + is a set of positive integers.

[0037] In the above solution, iterating the multi-objective optimization function to determine the optimal solution of the multi-objective optimization function specifically includes:

[0038] Convert the constraint variables into positive real variables;

[0039] Dividing the multi-objective optimization function into several sub-problem functions;

[0040] Based on the maximum number of transmissions Iterating the plurality of sub-problem functions to obtain an optimal parameter combination;

[0041] The parameter combination that minimizes the multi-objective optimization function is determined among the optimal parameter combinations as the optimal solution of the multi-objective optimization function.

[0042] In the above solution, the multi-objective optimization function is divided into several sub-problems, specifically including:

[0043] Based on the maximum number of transmissions and the update information uninstall ratio , and get the first subproblem function:

[0044]

[0045]

[0046]

[0047]

[0048] in, is the first sub-problem function, st C1 indicates that the constraint is the first constraint, For the time When the first concave function of the block length is For the time When the second convex function of the block length is For the time When , the third convex function of the block length is, is the size of the status update information, To update the information uninstall ratio, is the block transfer time, is the maximum number of transmissions, is the block error probability under the finite block length system, Calculate time for the local device, Calculate time for mobile edge servers, is the weight factor, The probability that the state information transmission fails G times;

[0049] Based on the maximum number of transmissions and the block length , and get the second sub-problem function:

[0050]

[0051]

[0052]

[0053] in, is the second sub-problem function, st C2 indicates that the constraint is the second constraint, For the time When updating information about uninstall rate The first concave function of For the time When , the second convex function of the update information offloading ratio is, is the weight factor, is the block transfer time, is the block error probability under the finite block length system, is the maximum number of transmissions, is the size of the status update information, To update the information uninstall ratio, The number of CPU cycles required to process 1NAT tasks, Calculate frequency for local devices, Calculate frequency for mobile edge servers, Transmitting status information The probability of failure.

[0054] The embodiments of the present invention have the following beneficial effects: by initializing network parameters, generating update information based on server status, determining relevant first-order moments, obtaining average latency and peak information age, constructing and iterating a multi-objective optimization function to obtain the optimal solution, and ultimately determining the optimal parameters of the short packet-MEC network, network transmission is comprehensively optimized in terms of information freshness (measured by information age) and transmission time (i.e., latency), effectively improving network performance, and making data transmission between local devices and mobile edge servers more efficient and information more timely. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0056] in:

[0057] Figure 1 1. A flowchart of a short packet-MEC network transmission optimization method based on information age and latency in one embodiment;

[0058] Figure 2 This is a system model diagram of the short packet-MEC network in the example of the present invention;

[0059] Figure 3 is a schematic diagram of information age changing over time;

[0060] Figure 4 Schematic diagram of the functional relationship between average AoI / PAoI and average delay and the maximum allowed number of retransmissions;

[0061] Figure 5 Schematic diagram of the relationship between average PAoI and average delay;

[0062] Figure 6 Schematic diagram of the relationship between the local calculation frequency and the weighted sum of the average delay and average PAoI;

[0063] Figure 7 Schematic diagram of the relationship between edge computing frequency and the weighted sum of average latency and average PAoI. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0065] In the following description, numerous specific details are given to provide a more thorough understanding of the present invention; however, it will be apparent to those skilled in the art that the present invention may be practiced without one or more of these details; in other examples, some technical features known in the art are not described to avoid confusion with the present invention; it should be understood that the present invention may be practiced in different forms and should not be construed as limited to the embodiments set forth herein.

[0066] The short packet-MEC network mentioned in this application refers to a network architecture that combines short packet transmission technology and Mobile Edge Computing (MEC). The following is an explanation of these two concepts:

[0067] Short packet transmission is a data transmission method that typically involves transmitting small data packets across a network. Short packet transmission is crucial in real-time communications, the Internet of Things (IoT), and mobile communications, as these applications often require fast, low-latency data transmission. Short data packets can reduce transmission latency and improve the real-time nature of communications.

[0068] Mobile Edge Computing (MEC): This is a network architecture that moves computing resources from cloud centers to the edge of the network, closer to the data source. MEC aims to reduce the distance data must travel, lower latency, increase processing speed, and provide a better service experience for users. In a MEC environment, data processing and analysis occurs at the edge of the network, rather than in remote data centers.

[0069] This application combines the two to create a short packet MEC network that can provide extremely low latency and higher bandwidth, thereby supporting high-speed transmission of a large number of short data packets.

[0070] In order to thoroughly understand the present invention, a detailed structure will be presented in the following description to illustrate the technical solution proposed by the present invention; optional embodiments of the present invention are described in detail below, but in addition to these detailed descriptions, the present invention may also have other implementation methods.

[0071] like Figure 1 As shown, in one embodiment, a short packet-MEC network transmission optimization method based on information age and delay is provided. The short packet-MEC network transmission optimization method based on information age and delay includes steps S101 to S107, which are detailed as follows:

[0072] S101. Initialize relevant parameters of the short packet-MEC network, which includes local devices and mobile edge servers;

[0073] This step sets the basic configuration of the network, ensuring that local devices and mobile edge servers can start working under preset conditions, providing a foundation for subsequent optimization.

[0074] In some embodiments, when the mobile edge server is idle, the local device generates new status update information and performs computational processing on the first portion of information to obtain processed information;

[0075] The local device offloads the processed information and information other than the first part to the mobile edge server via a wireless communication channel;

[0076] The overall decoding error probability of the transmission process is characterized using finite code length field theory;

[0077] If the status update information fails to be unloaded, the local device retransmits the processed information and the information other than the first part, and presets the maximum number of retransmissions allowed .

[0078] The maximum number of retransmissions allowed is It is a restriction on the maximum number of transmissions G, that is, the upper limit of the maximum number of transmissions G.

[0079] This solution allows local devices to generate new status updates when the mobile edge server is idle, process some of the information, and offload it to the server along with the remaining information. Finite code length field theory is used to characterize the overall decoding error probability of the transmission. If the offload fails, the local device will retransmit the message a preset maximum number of times, ensuring the reliability and integrity of the information transmission and improving the success rate and efficiency of short-packet MEC network status update transmission.

[0080] Specifically, such as Figure 2 The figure shows the system model diagram of the short packet-MEC network in the embodiment of the present invention, including: a local device and a mobile edge server; when the mobile edge server is idle, the local device generates a packet of size The data packet of nats is divided into two parts, the size of The data is processed on the local device, and the remaining size is The data is encapsulated into a short data packet and unloaded to the mobile edge server via the wireless communication channel for further processing; the finite code length field theory is used to characterize the overall decoding error probability of the transmission process. When an error occurs in the decoding of a data packet, the data packet cannot be unloaded to the mobile edge server, and the local device will retransmit the data packet. The maximum allowed number of retransmissions is .

[0081] In some embodiments, the parameters related to initializing the short packet-MEC network specifically include: status update information P, status update information size , update information uninstall ratio , static channel gain , block length , block error probability under finite block length system , status update information delay , status update information calculation completion interval , local device calculation time , computing time of mobile edge server , the number of CPU cycles required to process 1NAT task , the local device calculates the frequency , mobile edge server calculation frequency , the maximum number of retransmissions allowed , iterate index .

[0082] Channel gain plays a key role in the information transmission process from local devices to mobile edge devices. The expression is: ,in, is the path loss constant, is the distance between the local device and the mobile edge server, is the path loss exponent, is the Rayleigh block fading coefficient with zero mean and unit variance, is the channel gain.

[0083] The approximate characterization of the overall decoding error probability of the transmission process using finite code length field theory is:

[0084]

[0085] in, is the first coefficient, is the transmission power of the mobile edge server, is the noise power, is a constant, is an approximate representation of the overall decoding error probability, is the block length.

[0086] Specifically, , , , The packet size calculated to be encapsulated into short packets and offloaded to the mobile edge server, is the size of the status update information, To update the information uninstall ratio, is the transmission power of the mobile edge server, is the noise power, is the block length, is the distance between the local device and the mobile edge server, is the path loss exponent, is the path loss constant.

[0087] When a decoding error occurs and the data packet cannot be offloaded to the mobile edge server, the local device will resend the data packet until the maximum allowed retransmission times are reached. ,If the unloading is not successful at this time, the data packet is discarded, and when the mobile edge server is idle, the local device generates a new data packet.

[0088] like Figure 3The figure shows the information age changing over time. The time for the mobile edge server to complete the process is and . Its calculation is completed at interval for: .

[0089] S102: Generate status update information P according to the status of the mobile edge server;

[0090] By capturing the status changes of mobile edge servers in real time and generating corresponding status update information, data support can be provided for network adjustments.

[0091] S103: Determine the status update information delay according to the status update information P. The first-order moment and status update information calculation completion interval The first moment of

[0092] By calculating the first-order moment of the latency and the interval between completions, the transmission and computational efficiency of the status update information can be quantified, providing indicators for evaluating network performance.

[0093] In some embodiments, the status update information delay is determined based on the status update information P. The first-order moment and status update information calculation completion interval The first-order moments of include:

[0094] Determine the status update information delay according to the following expression The first moment of :

[0095]

[0096] in, is the block transfer time, is the maximum number of transmissions, Status update information delay The first moment of Calculate time for the local device, Calculate time for mobile edge servers, is the block error probability under the finite block length system, The probability that the state information transmission fails G times;

[0097] Determine the status update information calculation completion interval according to the following expression The first moment of :

[0098]

[0099] in, Calculate time for the local device, Calculate time for mobile edge servers, is the block transfer time, is the block error probability under the finite block length system, Calculate completion interval for status update information The first moment of is the probability that the state information transmission fails G times.

[0100] Specifically, using expressions for block transmission time, maximum number of transmissions, local device and mobile edge server computation time, and block error probability, the first-order moment of the status update information delay can be accurately calculated. This helps clearly understand the time delay characteristics of the status update information during transmission. Simultaneously, using another expression that includes factors such as local device and mobile edge server computation time, block transmission time, and block error probability, the first-order moment of the status update information calculation completion interval can be determined, providing a quantitative understanding of the time characteristics of the status update information calculation completion interval. The determination of these two first-order moments provides important data support for in-depth analysis of network performance and optimization of transmission mechanisms.

[0101] S104, determining the average delay and average peak information age of the status update information P;

[0102] By calculating the average latency and average peak information age, we can further evaluate the timeliness of network transmission and the freshness of information, providing a quantitative basis for optimization goals.

[0103] In some embodiments, determining the average latency and average peak information age of the status update information P specifically includes:

[0104] The average delay of the status update information P is determined according to the following expression:

[0105]

[0106] in, Calculate time for the local device, Calculate time for mobile edge servers, is the block transfer time, is the block error probability under the finite block length system, is the maximum number of transmissions, is the average delay of the state update information P, Status update information delay The first moment of Transmitting status information The probability of failure each time;

[0107] The average peak information age of the state update information P is determined according to the following expression:

[0108]

[0109] in, Status update information delay The first moment of Calculate completion interval for status update information The first moment of Calculate time for the local device, Calculate time for mobile edge servers, is the block transfer time, is the maximum number of transmissions, is the block error probability under the finite block length system, is the average peak information age of the state update information P, Transmitting status information The probability of failure.

[0110] Specifically, the given expressions enable precise calculation of the average latency and average peak information age of state update information. By combining factors such as local device computation time, mobile edge server computation time, block transmission time, block error probability under a finite block length system, and the maximum number of transmissions, the average latency of state update information can be accurately determined. This provides a comprehensive understanding of the average latency of state update information during transmission. The calculation of the average peak information age not only considers these factors but also incorporates the first-order moment of the state update information latency and the first-order moment of the state update information calculation completion interval. This allows us to measure the freshness of information and understand the average aging of information at its peak throughout the entire process. The determination of these two metrics provides a key basis for evaluating and optimizing the transmission performance of short-packet MEC networks, enabling more effective measures to improve network real-time performance and information validity in practical applications.

[0111] S105, establishing a multi-objective optimization function based on the average delay and the average peak information age;

[0112] A multi-objective optimization function including average delay and average peak information age is constructed to quantify network performance and provide a mathematical model for finding the optimal solution.

[0113] In some embodiments, a multi-objective optimization function is established based on the average latency and the average peak information age, specifically including:

[0114] The multi-objective optimization function is based on the block length , maximum number of transmissions and update information uninstall ratio Minimize the weighted sum of average peak information age and average delay:

[0115]

[0116] in, is the weight factor, and are the lower and upper limits of the packet length respectively, is the maximum number of retransmissions allowed, is the average delay of the state update information P, is the average peak information age of the state update information P, is a multi-objective optimization function, st represents the following expression as a constraint, C1 is the first constraint, C2 is the second constraint, and C3 is the third constraint. To update the information uninstall ratio, is the maximum number of transmissions, is the block length, N + is a set of positive integers.

[0117] in, .

[0118] The multi-objective optimization function constructed in this application is based on block length, maximum number of transmissions, and update information offload ratio, aiming to minimize the weighted sum of average peak information age and average latency. By introducing weighting factors, this function allows for flexible adjustment of the relative importance of average peak information age and average latency in the optimization objectives. It also considers constraints such as upper and lower limits on packet length and maximum number of transmissions, making the optimization more tailored to actual network scenarios. The average latency and average peak information age of status update information, as key components of the function, reflect the network's performance in terms of information transmission time and information freshness. Solving and analyzing this multi-objective optimization function can provide guidance for parameter selection (such as block length, maximum number of transmissions, and update information offload ratio) in short-packet MEC networks. This allows for optimized network performance while comprehensively considering information age and latency, helping to meet the needs of application scenarios with high real-time and information timeliness requirements.

[0119] S106, iterating the multi-objective optimization function to determine the optimal solution of the multi-objective optimization function;

[0120] S107. Determine the optimal parameters of the short packet-MEC network according to the optimal solution of the multi-objective optimization function.

[0121] By using an iterative algorithm to search step by step, a set of solutions can be found that can achieve relatively optimal levels of average delay and average peak information age. This provides theoretical guidance for optimizing network performance and applies the theoretical optimal solution to the actual network, adjusting the short packet-MEC network parameters to achieve network transmission optimization in terms of information age and delay.

[0122] In some embodiments, iterating the multi-objective optimization function to determine the optimal solution of the multi-objective optimization function specifically includes:

[0123] Convert the constraint variables into positive real variables;

[0124] Divide the multi-objective optimization function into several sub-problem functions;

[0125] Based on the maximum number of transmissions Iterate several sub-problem functions to obtain the optimal parameter combination;

[0126] The parameter combination that minimizes the multi-objective optimization function is determined among the optimal parameter combinations as the optimal solution of the multi-objective optimization function.

[0127] Converting the constraint variables into positive real variables simplifies their properties, facilitating subsequent calculations and analysis, and allowing the optimization process to proceed in a more concise mathematical environment. Next, the multi-objective optimization function is split into several sub-problem functions. This approach decomposes the complex multi-objective problem into relatively simple sub-problems, reducing the difficulty of solving the problem. These sub-problem functions are then iterated based on the maximum number of transmissions. During the iteration process, the parameters are continuously adjusted to gradually approach the optimal parameter combination. This process fully utilizes the maximum number of transmissions as a key factor to explore the optimal solution space of the function. Finally, the parameter combination that minimizes the value of the multi-objective optimization function is selected from the obtained optimal parameter combinations and determined as the optimal solution to the multi-objective optimization function. This optimal solution comprehensively considers the weighted sum minimization of the average peak information age and average latency, providing a practical parameter configuration scheme for optimizing short-packet MEC networks, effectively improving the overall performance of the network in terms of information age and latency.

[0128] In some embodiments, the multi-objective optimization function is divided into several sub-problems, specifically including:

[0129] Based on the maximum number of transmissions and update information uninstall ratio , and get the first subproblem function:

[0130]

[0131]

[0132]

[0133]

[0134] in, is the first sub-problem function, st C1 indicates that the constraint is the first constraint, For the time When the first concave function of the block length is For the time When the second convex function of the block length is For the time When , the third convex function of the block length is, is the size of the status update information, To update the information uninstall ratio, is the block transfer time, is the maximum number of transmissions, is the block error probability under the finite block length system, Calculate time for the local device, Calculate time for mobile edge servers, is the weight factor, is the probability that the state information transmission fails G times.

[0135] Among them, when hour, is the first concave function of the block length, is the second convex function of the block length, is the third convex function of the block length. Therefore, the first sub-problem can be solved as follows:

[0136] when When , the subproblem can be converted into a convex difference programming problem, which can be solved using the convex-concave process (CCP). The core idea of CCP is to iteratively solve a series of convex proxy problems, each of which is constructed by linearizing the concave term. When , the optimal block length of this subproblem can be found through linear search.

[0137] Based on the maximum number of transmissions and block length , and get the second sub-problem function:

[0138]

[0139]

[0140]

[0141] in, is the second sub-problem function, st C2 indicates that the constraint is the second constraint, For the time When updating information about uninstall rate The first concave function of For the time When , the second convex function of the update information offloading ratio is, is the weight factor, is the block transfer time, is the block error probability under the finite block length system, is the maximum number of transmissions, is the size of the status update information, To update the information uninstall ratio, The number of CPU cycles required to process 1NAT tasks, Calculate frequency for local devices, Calculate frequency for mobile edge servers, Transmitting status information The probability of failure.

[0142] Among them, when hour, is the first concave function of the update information offloading ratio, is the second convex function of the update information offloading ratio. As in the solution of the first sub-problem, when When , the second subproblem is a convex difference programming problem, which is solved using the concave-convex process (CCP). ,The optimal update information offloading ratio for the second sub-problem can be found by linear search.

[0143] By decomposing the multi-objective optimization function into these two sub-problem functions, the originally complex multi-factor optimization problem is rationally split, making the study of each factor more detailed and in-depth, providing a clearer idea and more effective method for the subsequent optimal solution of the multi-objective optimization function, and helping to achieve more accurate performance optimization and meet the strict requirements of different application scenarios on information age and latency.

[0144] In some embodiments, by setting the maximum number of transmissions Performing a one-dimensional search can iteratively solve the optimization problem, specifically:

[0145] First, the maximum number of transmissions Perform a one-dimensional search, initialize the iteration index r = 0, and the iteration range is from 1 to the maximum allowed number of retransmissions , randomly generate the initial point ,in and They represent the block length and update information offloading ratio respectively;

[0146] if , then the first sub-problem function and the second sub-problem function are updated using the linear search method and ,in, is the size of the status update information, is the product of the rth update information offloading ratio and the state update information size, is pi, is the length of the r+1th block, The offloading ratio for the r+1th update information;

[0147] if , use linear search to update the first subproblem , use the bump process to update the second subproblem ,in, For status update information, is the product of the rth update information offloading ratio and the state update information size, is pi, is the length of the r+1th block, The offloading ratio for the r+1th update information;

[0148] if , then the first and second sub-problems are updated using the concave-convex process respectively. and ,in, is the size of the status update information, is the product of the rth update information offloading ratio and the state update information size, is pi, is the length of the r+1th block, The offloading ratio of the r+1th update information.

[0149] After each iteration is completed, the iteration index Updated to When the convergence criteria are met, stop the iteration process under the current maximum number of transmissions and record the current maximum number of transmissions and the current block length. And the current update information uninstall ratio , using the current maximum number of transmissions and the current block length And the current update information uninstall ratio Calculate the objective function The value of ; where is the weight factor, is the average peak information age of the state update information P, is the average delay of the state update information P, is the objective function.

[0150] When all the maximum number of transmissions are completed After the value is set, each maximum number of transmissions has been The objective function is calculated and the first parameter combination that minimizes the objective function is selected within this range. , as the preliminary optimal parameter combination.

[0151] After getting the first parameter combination After that, further in block length Within the preset range, find the second parameter combination that minimizes the objective function to obtain the final optimal solution.

[0152] like Figure 4 The relationship between average AoI / PAoI and average latency as a function of the maximum allowed number of retransmissions is described. The results show that average AoI / PAoI can be reduced by increasing the maximum allowed number of retransmissions. However, the benefit of information age is offset by the loss of average latency, thus demonstrating the necessity of achieving a latency tradeoff. The average AoI (Average Age of Information) refers to the average time from information generation to user reception. Specifically, it is the expected value of the time difference between the time information arrives and the time it is generated. AoI can be used to measure the freshness of information; the smaller the AoI, the fresher the information. The average peak age of information (PAoI) refers to the average of the maximum AoI reached before the arrival of an information update. It measures the worst-case delay between information updates.

[0153] Figure 5 The figure depicts the trade-off between average PAoI and average latency. The figure shows that the performance trade-off of our proposed algorithm is consistent with the Pareto frontier, achieving Pareto optimality.

[0154] Figure 6 The effect of local computation frequency on the optimized weighted sum of average delay and average PAoI is studied. The proposed algorithm is used to benchmark the optimization results of exhaustive search, no offloading scheme, and offloading scheme with ARQ (Automatic Repeat Request) retransmission. It can be seen that the proposed algorithm achieves optimality close to that of exhaustive search. Figure 6 We can see that when local computing power is insufficient, local devices tend to offload all update information to the MES (Mobile Edge Computing Server). In this case, the advantages of this application are very prominent compared with no offloading and ARQ offloading. As the local computing frequency increases, the performance of the no offloading scheme and the offloading scheme with ARQ is still poor, but this difference gradually disappears due to the shrinking offloading demand. As the edge computing frequency increases, the weighted sum of all schemes except the no offloading scheme can be reduced, such as Figure 7The proposed scheme outperforms other comparison baselines, thus illustrating the benefits of jointly considering block length, offload ratio, and maximum allowed retransmission times in SP-MEC networks.

[0155] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory.

[0156] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0157] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person skilled in the art would be able to make numerous modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. The above disclosures are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A short packet-MEC network transmission optimization method based on information age and delay, characterized in that: The method comprises: Initialize relevant parameters of the short packet-MEC network, which includes local devices and mobile edge servers; Generate state update information P according to the state of the mobile edge server; Determine the status update information delay according to the status update information P The first-order moment and status update information calculation completion interval The first moment of Determining an average delay and an average peak information age of the status update information P; Establishing a multi-objective optimization function based on the average time delay and the average peak information age; Iterating the multi-objective optimization function to determine an optimal solution of the multi-objective optimization function; Determining optimal parameters of the short packet-MEC network according to the optimal solution of the multi-objective optimization function; Determining the status update information delay according to the status update information P The first-order moment and status update information calculation completion interval The first-order moments of include: Determine the status update information delay according to the following expression The first moment of : in, is the block transfer time, is the maximum number of transmissions, Status update information delay The first moment of Calculate time for the local device, Calculate time for mobile edge servers, is the block error probability under the finite block length system, Transmitting status information The probability of failure every time; Determine the status update information calculation completion interval according to the following expression The first moment of : in, Calculate time for the local device, Calculate time for mobile edge servers, is the block transfer time, is the block error probability under the finite block length system, Calculate completion interval for status update information The first moment of Transmitting status information The probability of failure every time; Determining the average delay and average peak information age of the status update information P specifically includes: The average delay of the status update information P is determined according to the following expression: in, Calculate time for the local device, Calculate time for mobile edge servers, is the block transfer time, is the block error probability under the finite block length system, is the maximum number of transmissions, is the average delay of the state update information P, Status update information delay The first moment of The probability that the state information transmission fails G times; The average peak information age of the status update information P is determined according to the following expression: in, Status update information delay The first moment of Calculate completion interval for status update information The first moment of Calculate time for the local device, Calculate time for mobile edge servers, is the block transfer time, is the maximum number of transmissions, is the block error probability under the finite block length system, is the average peak information age of the state update information P, is the probability that the state information transmission fails G times.

2. The short packet-MEC network transmission optimization method based on information age and delay according to claim 1 is characterized in that: When the mobile edge server is idle, the local device generates new status update information and performs calculation processing on the first part of information to obtain processed information; The local device offloads the processed information and information other than the first part to the mobile edge server via a wireless communication channel; The overall decoding error probability of the transmission process is characterized using finite code length field theory; If the status update information fails to be unloaded, the local device retransmits the processed information and information other than the first part, and presets the maximum number of retransmissions allowed. .

3. The short packet-MEC network transmission optimization method based on information age and delay according to claim 1 is characterized in that: The parameters related to the initialization short packet-MEC network specifically include: status update information P, status update information size , update information uninstall ratio , static channel gain , block length , block error probability under finite block length system , status update information delay , status update information calculation completion interval , local device calculation time , computing time of mobile edge server , the number of CPU cycles required to process 1NAT task , the local device calculates the frequency , mobile edge server calculation frequency , the maximum number of retransmissions allowed , iterate index .

4. The short packet-MEC network transmission optimization method based on information age and delay according to claim 1 is characterized in that: The establishing of a multi-objective optimization function according to the average delay and the average peak information age specifically includes: The multi-objective optimization function is based on the block length , maximum number of transmissions and update information uninstall ratio Minimize the weighted sum of the average peak information age and the average delay: in, is the weight factor, and are the lower and upper limits of the packet length respectively, is the maximum number of retransmissions allowed, is the maximum number of transmissions, is the average delay of the state update information P, is the average peak information age of the state update information P, is a multi-objective optimization function, st represents the following expression as a constraint, C1 is the first constraint, C2 is the second constraint, and C3 is the third constraint. To update the information uninstall ratio, is the block length, N + is a set of positive integers.

5. The short packet-MEC network transmission optimization method based on information age and delay according to claim 4 is characterized in that: The iterating the multi-objective optimization function to determine the optimal solution of the multi-objective optimization function specifically includes: Convert the constraint variables into positive real variables; Dividing the multi-objective optimization function into several sub-problem functions; Based on the maximum number of transmissions Iterating the plurality of sub-problem functions to obtain an optimal parameter combination; The parameter combination that minimizes the multi-objective optimization function is determined among the optimal parameter combinations as the optimal solution of the multi-objective optimization function.

6. The short packet-MEC network transmission optimization method based on information age and delay according to claim 5 is characterized in that: The multi-objective optimization function is divided into several sub-problems, specifically including: Based on the maximum number of transmissions and the update information uninstall ratio , and get the first subproblem function: in, is the first sub-problem function, st C1 indicates that the constraint is the first constraint, For the time When the first concave function of the block length is For the time When the second convex function of the block length is For the time When , the third convex function of the block length is, is the size of the status update information, To update the information uninstall ratio, is the block transfer time, is the maximum number of transmissions, is the block error probability under the finite block length system, Calculate time for the local device, Calculate time for mobile edge servers, is the weight factor, The probability that the state information transmission fails G times; Based on the maximum number of transmissions and the block length , and get the second sub-problem function: in, is the second sub-problem function, st C2 indicates that the constraint is the second constraint, For the time When updating information about uninstall rate The first concave function of For the time When , the second convex function of the update information offloading ratio is, is the weight factor, is the block transfer time, is the block error probability under the finite block length system, is the maximum number of transmissions, is the size of the status update information, To update the information uninstall ratio, The number of CPU cycles required to process 1NAT tasks, Calculate frequency for local devices, Calculate frequency for mobile edge servers, Transmitting status information The probability of failure.

Images