A Cellular Resource Scheduling Method Based on Information Freshness

Abstract

This invention belongs to the field of wireless communication technology, and particularly relates to a cellular resource scheduling method based on information freshness. It utilizes Shannon's formula and the graphical properties of information age to derive the expression for the average information age at the base station; compares the advantages and disadvantages of amplified forwarding and decoded forwarding methods, as well as their trends with transmission power, spectral efficiency constraints, and distance, to select an appropriate forwarding method; while ensuring the success rate of D2D user information transmission, it uses an improved tundra swarm algorithm to allocate the transmission power of D2D users, minimizing the average information age of the information obtained at the base station.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, and in particular relates to a cellular resource scheduling method based on information freshness. Background Technology

[0002] With the development and deployment of 5G technology, more and more emerging IoT applications are appearing in people's daily lives. Future IoT applications will increasingly rely on the exchange of latency-sensitive information for environmental and status monitoring and control. Systems such as smart homes, smart transportation, and smart healthcare largely depend on real-time status information. For this type of time-sensitive information, if the destination receives outdated information, it will lead to ineffective decision-making and erroneous control, and even systemic disasters. Previous research has typically used latency or throughput to characterize the timeliness of information in a system. However, in the aforementioned real-time service applications, these two indicators cannot fully measure the timeliness of information. Therefore, information age, as a new indicator for quantifying the freshness of information, has received widespread research attention.

[0003] To date, Age of Information (AoI) has been studied in various systems as an emerging concept, performance metric, and tool. Currently, AoI research is still in its early stages, mostly considering data transmission under ideal conditions. However, in practical applications, the impact of co-channel interference on AoI cannot be ignored. Taking cellular networks as an example, when D2D users and cellular users share a channel, interference between them can impair the timely arrival of information. Therefore, the issue of AoI guarantee under interference environments cannot be overlooked. Summary of the Invention

[0004] To overcome the above problems, this invention provides a cellular resource scheduling method based on information freshness, which can allocate the transmission power of D2D users, thereby minimizing the information age of the information obtained at the base station.

[0005] A cell resource scheduling method based on information freshness includes the following:

[0006] Step 1: Set up the cellular system model:

[0007] In a cellular cell centered on a base station, there are C cellular users and D D2D users simultaneously, and the cellular users and D2D users share channel resources, resulting in co-channel interference.

[0008] Cellular users send status information to repeaters. If the information sent by the cellular user meets the spectrum efficiency constraint, the information is successfully transmitted; otherwise, it will be retransmitted. After receiving the information, the repeater forwards the information to the base station. If the forwarded information meets the spectrum efficiency constraint, the information is successfully transmitted; otherwise, it will be retransmitted. After the information is successfully sent to the base station, the cellular user proceeds to the next information transmission.

[0009] The repeater operates under interference-limited conditions and is subject to interference from q multiplexed channels' D2D users. The base station noise power is denoted as σ. 2 ; stipulate | h C,M | 2 ,|h M,B | 2 ,|g l | 2 Let represent the channel gain between the cellular user and the repeater, the channel gain between the repeater node and the base station, and the interference channel gain caused by D2D users to the repeater node, respectively. All channels satisfy Rayleigh block fading and follow the parameters λ. C,M , λ M,B τ is an exponential distribution, l∈{1,2...q};

[0010] Step two, to determine the repeater's forwarding method, calculate the probability of a cellular user or repeater successfully sending information under different modes:

[0011] The probability of different users successfully sending information under both decode-forward and amplify-forward modes is calculated using Shannon's theorem and probability theory. The calculation method is as follows:

[0012] 1. Decoding and forwarding mode:

[0013] The probability π of cellular users and repeaters successfully transmitting information in decode-forward mode C1 and π M1 Calculate according to the following formulas:

[0014]

[0015]

[0016] Where: P{} represents the probability of taking the formula within {}, and the cellular user transmit power is P. C The repeater's transmit power is P M The D2D user transmit power is P D r refers to the minimum amount of information required for each transmitting channel to successfully send information;

[0017] Second, amplified forwarding mode:

[0018] The probability π of cellular users and repeaters successfully transmitting information in amplified forwarding mode C2 and π M2 Calculate according to the following formulas:

[0019]

[0020]

[0021] Where: β is the amplification factor of the repeater, and the calculation formula is: μ2=σ 2 P C , x is the integration variable, representing the integral over time;

[0022] Step 3: Calculating the average information age under different models:

[0023] The average information age is calculated using the image properties of information age and Taylor series when K transmissions are successfully completed in the entire transmission process from cellular user to repeater to base station. The average information age is then used to calculate this value. Calculate using the following formula:

[0024]

[0025] Where: X is the time interval for the base station to update data packets, and E(X) represents the expected value of the time interval X, calculated by the following formula:

[0026]

[0027] E(X 2 The expected value of the square of the time interval X is calculated using the following formula:

[0028]

[0029] Where m is the number of cellular user transmissions, n is the number of repeater forwardings, and π C and π M These represent the probabilities of cellular users and repeaters successfully sending information in the corresponding modes;

[0030] Step 4: Based on the calculation results of Step 2 and Step 3, compare the average information age when the entire transmission process from cellular user to repeater to base station is successfully completed K times under different forwarding modes, and take the decoding and forwarding mode with the smaller average information age as the forwarding mode determined by this method.

[0031] Step 5, Cellular Resource Scheduling Strategy: An enhanced stochastic tunic swarm algorithm is proposed to allocate transmit power for D2D users in a cell, minimizing the average information age at the base station. The specific strategy is as follows:

[0032] Step 5.1 stipulates that D D2D users reuse cellular channels using a random allocation method, while ensuring that a single cellular user is reused by only one D2D user, and a D2D user reuses only one cellular channel. Furthermore, the information transmitted by both D2D users and cellular users must satisfy the following spectral efficiency constraints:

[0033] log(1+γ)≥r

[0034] Where: γ is the signal-to-interference-plus-noise ratio;

[0035] Step 5.2, Initialize the location of the salps population: that is, randomly set the location of each salps in the population, where the location of the salps represents the possible transmission power of the D2D user that minimizes the average information age. The location of the salps includes three categories: food source location, salps leader location, and salps follower location.

[0036] Step 5.3: Calculate the fitness f of the locations of all tunicates in the initial population. The location of the tunicate with the smallest f value corresponds to the food source location. In other words, the D2D user's transmit power P is at the smallest f value. D This minimizes the final average information age, and the minimum value of f guarantees the objective: minimizing the average information age while satisfying the spectral efficiency constraint; the fitness f at the location of the tunic is calculated by the following formula:

[0037]

[0038]

[0039] In the formula, μ is the penalty factor, and μ > 0. The fitness f includes two parts: the objective function and the constraints. The objective function and constraints are as follows:

[0040]

[0041] log2(1+γ M1 )≥r,

[0042] log2(1+γ B1 )≥r,

[0043] log2(1+γ d )≥r,d∈D

[0044]

[0045]

[0046] C > D

[0047] 0 < PD <P Dmax

[0048] Where: when reuse parameters When, it indicates that D2D user d is multiplexing the channel of cellular user c; otherwise... γ M1 Indicates the signal-to-interference-plus-noise ratio (SIR) of the repeater in decode-and-forward mode; γ B1 P represents the signal-to-interference-plus-noise ratio (SIR) of the base station in decode-and-forward mode. Dmax This represents the maximum transmit power of D2D users, where C is the number of cellular users, D is the number of D2D users, and γ is the maximum transmit power of D2D users. d Indicates the signal-to-interference-plus-noise ratio (SIR) for D2D users;

[0049] Step 5.4, update the location of the leader of the tunicate sea lizard according to the following formula:

[0050]

[0051] Where: the j-th dimension represents the j-th D2D user in the cellular system model. The updated position of the tunicate leader, i.e., the position of the tunicate leader in the j-th dimension, is the transmit power of the D2D user among the j D2D users that may minimize the average information age, F. j Let ub be the location of the food source in the j-th dimension, that is, the location of the tunicate corresponding to the minimum f value in step 5.3. j Let lb be the upper bound of the j-th dimension, i.e., the maximum possible transmit power of a D2D user. j Let c2 and c3 be the lower bound of the j-th dimension, i.e., the possible minimum transmit power of a D2D user, where c2 and c3 are random numbers, Cauchy is the Cauchy variation factor, and a is the lower bound of the j-th dimension. max and a min Let a and b be the maximum and minimum values ​​of the convergence factor a, respectively, where a(t) = (a max -a min )·rand+σ·randn, where rand is a random number in the interval [0,1], randn is a normally distributed random number, σ is used to measure the degree of deviation between the convergence factor a and its mathematical expectation, and t is time;

[0052] Then update the position of the salver followers according to the following formula:

[0053]

[0054] Where: i≥2, This represents the position of the i-th follower of the salver besides the leader of the salver, in the j-dimensional space. That is, the position of the follower of the salver is the transmission power of the j-th D2D user other than the transmission power of the D2D user corresponding to the position of the salver leader.

[0055] Step 6: Set the maximum number of iterations, and then perform iterative calculations on steps 5.3-5.4 in step 5 until the maximum number of iterations is reached. Output the food source location obtained from the last iteration as the optimal location, thus obtaining the transmit power value of the D2D user that minimizes the average information age at the base station.

[0056] In step 3 of step four, under the decoding and forwarding mode, the signal-to-interference-plus-noise ratio (SIR) γ of the repeater and the base station... M1 γ B1 Calculate according to the following formulas:

[0057]

[0058]

[0059] In step 5.3 of step five, γ d The signal-to-interference-plus-noise ratio (SIR) of a D2D user is calculated using the following formula:

[0060]

[0061] Where: |h d | 2 Represents the channel gain between D2D users; |g d | 2 This indicates the link gain between the repeater and the D2D user.

[0062] The beneficial effects of this invention are:

[0063] This invention uses information age as a metric to evaluate the timeliness of state information updates. It introduces the Shannon formula and AoI (Aspect-Oriented Information) image properties. In the presence of co-channel interference, the repeater forwards information from cellular users to the base station. The invention discusses the repeater's forwarding modes, comparing the advantages and disadvantages of amplified forwarding and decoded forwarding methods, and analyzing their trends with transmit power, spectral efficiency constraints, and distance to select an appropriate forwarding method. While ensuring the probability of successful D2D user information transmission, an improved tundra swarm algorithm is used to allocate the transmit power of D2D users, minimizing the average AoI of the information received at the base station, thus achieving timely information delivery.

[0064] This invention utilizes Shannon's formula and AoI image properties, combined with the proposed improved tunicate swarm algorithm for resource allocation, which can significantly reduce the average information age of the system. Attached Figure Description

[0065] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the content of the embodiments of the present invention and these drawings without creative effort.

[0066] Figure 1 This is a model diagram of the interference cellular cell of the present invention;

[0067] Figure 2 This describes the information age change trend of the discrete system of the present invention;

[0068] Figure 3 This is a flowchart of the enhanced random tunicate swarm algorithm in step five of this invention;

[0069] Figure 4 This is the trend of the average AoI as a function of cellular user transmit power in this invention;

[0070] Figure 5 This shows the trend of the average AoI as a function of spectral efficiency constraints in this invention.

[0071] Figure 6 This invention describes the trend of average AoI (Average Area of ​​Indicator) as the distance between the cellular user and the base station changes with the distance between the cellular user and the repeater when the distance between the cellular user and the base station is fixed.

[0072] Figure 7 This invention provides the average AoI convergence curve when the number of D2D users D=15 and the number of cellular users C=50.

[0073] Figure 8 This is the average AoI convergence curve when the number of D2D users D=25 and the number of cellular users C=50 in this invention;

[0074] Figure 9 The present invention provides the average AoI convergence curve when the number of D2D users D=15 and the number of cellular users C=75. Detailed Implementation

[0075] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0076] Example 1

[0077] A cell resource scheduling method based on information freshness includes the following:

[0078] Step 1: Set up the cellular system model (referred to as the cellular system):

[0079] In a cellular cell centered on a base station, there are C cellular users and D D2D users simultaneously, and the cellular users and D2D users share channel resources, resulting in co-channel interference.

[0080] Cellular users send status information to repeaters. If the information sent by the cellular user meets the spectrum efficiency constraint, the information is successfully transmitted; otherwise, it will be retransmitted. After receiving the information, the repeater forwards the information to the base station. If the forwarded information meets the spectrum efficiency constraint, the information is successfully transmitted; otherwise, it will be retransmitted. After the information is successfully sent to the base station, the cellular user proceeds to the next information transmission.

[0081] When the repeater operates under interference-limited conditions, it will be subject to interference from q multiplexed channels of D2D users. The base station noise power is denoted as σ. 2 ; stipulate | h C,M | 2 ,|h M,B | 2 ,|g l | 2 Let represent the channel gain between the cellular user and the repeater, the channel gain between the repeater node and the base station, and the interference channel gain caused by D2D users to the repeater node, respectively. All channels satisfy Rayleigh block fading and follow the parameters λ. C,M , λ M,B τ is an exponential distribution, l∈{1,2...q};

[0082] Step two, to determine the repeater's forwarding method, calculate the probability of a cellular user or repeater successfully sending information under different modes:

[0083] The probability of different users successfully sending information under both decode-forward and amplify-forward modes is calculated using Shannon's theorem and probability theory. The calculation method is as follows:

[0084] 1. Decoding and forwarding mode:

[0085] The probability π of cellular users and repeaters successfully transmitting information in decode-forward mode C1 and π M1 Calculate according to the following formulas:

[0086]

[0087]

[0088] Where: P{} represents the probability of taking the formula within {}, and the cellular user transmit power is P. C The repeater's transmit power is P M The D2D user transmit power is P D r refers to the minimum amount of information required for each transmitting channel to successfully send information;

[0089] Second, amplified forwarding mode:

[0090] The probability π of cellular users and repeaters successfully transmitting information in amplified forwarding mode C2 and π M2 Calculate according to the following formulas:

[0091]

[0092]

[0093] Where: β is the amplification factor of the repeater, and the calculation formula is: μ2=σ 2 P C , x is the integration variable, representing the integral over time;

[0094] Step 3: Calculating the average information age under different models:

[0095] The average information age is calculated using the image properties of information age and Taylor series when K transmissions are successfully completed in the entire transmission process from cellular user to repeater to base station. The average information age is then used to calculate this value. Calculate using the following formula:

[0096]

[0097] Where: X is the time interval for the base station to update data packets, and E(X) represents the expected value of the time interval X, calculated by the following formula:

[0098]

[0099] E(X 2 The expected value of the square of the time interval X is calculated using the following formula:

[0100]

[0101] Where m is the number of cellular user transmissions, n is the number of repeater forwardings, and π C and π M These represent the probabilities of cellular users and repeaters successfully sending information in the corresponding modes;

[0102] Step 4: Based on the calculation results of Step 2 and Step 3, compare the average information age when the entire transmission process from cellular user to repeater to base station is successfully completed K times under different forwarding modes, and take the decoding and forwarding mode with the smaller average information age as the forwarding mode determined by this method.

[0103] Step 5, Cellular Resource Scheduling Strategy: An enhanced stochastic tunic swarm algorithm is proposed to allocate transmit power for D2D users in a cell, minimizing the average information age at the base station. The specific strategy is as follows:

[0104] Step 5.1 stipulates that D D2D users reuse cellular channels using a random allocation method, while ensuring that a single cellular user is reused by only one D2D user, and a D2D user reuses only one cellular channel. Furthermore, the information transmitted by both D2D users and cellular users must satisfy the following spectral efficiency constraints:

[0105] log(1+γ)≥r

[0106] Where: γ is the signal-to-interference-plus-noise ratio;

[0107] Step 5.2, Initialize the location of the salps population: This involves randomly setting the location of each salps in the population. The location of the salps represents the possible transmission power of the D2D user that minimizes the average information age. The location of the salps includes three categories: food source location, salps leader location, and salps follower location. During the iteration process, the salps population will move closer to the food source location, and the salps leader will lead the entire group to move closer first. That is, the leader's location update is affected by the food source location.

[0108] Step 5.3: Calculate the fitness f of the locations of all tunicates in the initial population. The location of the tunicate with the smallest f value corresponds to the food source location. In other words, the D2D user's transmit power P is at the smallest f value. D This minimizes the final average information age, and the minimum value of f guarantees the objective: minimizing the average information age while satisfying the spectral efficiency constraint; the fitness f at the location of the tunic is calculated by the following formula:

[0109]

[0110]

[0111] In the formula, μ is the penalty factor, and μ > 0. The fitness f includes two parts: the objective function and the constraints. The objective function and constraints are as follows:

[0112]

[0113] log2(1+γM1 )≥r,

[0114] log2(1+γ B1 )≥r,

[0115] log2(1+γ d )≥r,d∈D

[0116]

[0117]

[0118] C > D

[0119] 0 < P D <P Dmax

[0120] Where: when reuse parameters When, it indicates that D2D user d is multiplexing the channel of cellular user c; otherwise... γ M1 Indicates the signal-to-interference-plus-noise ratio (SIR) of the repeater in decode-and-forward mode; γ B1 P represents the signal-to-interference-plus-noise ratio (SIR) of the base station in decode-and-forward mode. Dmax This represents the maximum transmit power of D2D users, where C is the number of cellular users, D is the number of D2D users, and γ is the maximum transmit power of D2D users. d Indicates the signal-to-interference-plus-noise ratio (SIR) for D2D users;

[0121] Step 5.4: Update the location of the salver leader according to the following formula, where the location of the salver leader represents the possible transmission power of the D2D user that minimizes the average information age, and continuously moves closer to the food source location due to the influence of the food source location:

[0122]

[0123] Where: the j-th dimension represents the j-th D2D user in the cellular system model. The updated location of the tunicate leader, i.e., the position of the tunicate leader in the j-th dimension, is the transmit power of the j-th D2D user that may minimize the average information age. F j Let ub be the location of the food source in the j-th dimension, that is, the location of the tunicate corresponding to the minimum f value in step 5.3. j Let lb be the upper bound of the j-th dimension, i.e., the maximum possible transmit power of a D2D user. j Let c2 and c3 be the lower bound of the j-th dimension, i.e., the possible minimum transmit power of a D2D user, where c2 and c3 are random numbers, Cauchy is the Cauchy variation factor, and a is the lower bound of the j-th dimension. max and a minLet a and b be the maximum and minimum values ​​of the convergence factor a, respectively, where a(t) = (a max -a min )·rand+σ·randn, where rand is a random number in the interval [0,1], randn is a normally distributed random number, σ (variance) is used to measure the degree of deviation between the convergence factor a and its mathematical expectation, and t is time;

[0124] Then, update the position of the salver followers according to the following formula. The salver followers are influenced by the salver leader and continuously move closer to the salver leader:

[0125]

[0126] Where: i≥2, This represents the position of the i-th tunicate follower other than the tunicate leader in the j-dimensional space. It is the possible solution set in the process of finding the optimal solution, that is, the position of the tunicate follower, which is the transmission power of the j-th D2D user other than the transmission power of the D2D user corresponding to the position of the tunicate leader, which may minimize the average information age.

[0127] Step 6: Set the maximum number of iterations, and then perform iterative calculations on steps 5.3-5.4 in step 5 until the maximum number of iterations is reached. Output the food source location obtained from the last iteration as the optimal location, thus obtaining the transmit power value of the D2D user that minimizes the average information age at the base station.

[0128] In step 3 of step four, under the decoding and forwarding mode, the signal-to-interference-plus-noise ratio (SIR) γ of the repeater and the base station... M1 γ B1 Calculate according to the following formulas:

[0129]

[0130]

[0131] In step 5.3 of step five, γ d The signal-to-interference-plus-noise ratio (SIR) of a D2D user is calculated using the following formula:

[0132]

[0133] Where: |h d | 2 Represents the channel gain between D2D users; |g d | 2 This indicates the link gain between the repeater and the D2D user.

[0134] Example 2

[0135] The cellular system model simultaneously includes D2D users who share the channel with cellular users, resulting in co-channel interference. Under interference conditions, the average information age expression at the base station is derived using Shannon's formula and the graphical properties of information age. The relay forwarding modes are analyzed, comparing the advantages and disadvantages of amplification-forwarding and decoding-forwarding methods, and considering their variations with transmit power, spectral efficiency constraints, and distance, to select the appropriate forwarding method. While ensuring the success rate of D2D user information transmission, an improved tunic swarm algorithm is used to allocate the transmit power of D2D users, minimizing the average information age of the information received at the base station. This algorithm, based on the tunic swarm algorithm, adds random parameter selection and a Cauchy mutation operator, ensuring population diversity and achieving a globally optimal search.

[0136] In the cellular system model, only the decode-forward mode is used because, in the simulations described later, i.e. Figure 4 As can be seen from 5 and 6, decoding and forwarding are more effective than amplification and forwarding. Therefore, the decoding and forwarding mode was adopted in the cellular system model.

[0137] The cellular system model described above needs to simultaneously ensure the success rate of D2D user information transmission and minimize the average age of cellular user information received by the base station.

[0138] The improved tunicate swarm algorithm utilizes random parameter selection and the Cauchy mutation operator to ensure population diversity and achieve global optimization.

[0139] like Figure 1 As shown, in a cellular cell centered on a base station, there are C cellular users and D D2D users simultaneously, and the cellular users and D2D users share channel resources, resulting in co-channel interference.

[0140] Assume that each data transfer between users consumes one unit of time; every data transfer by any user consumes time, and one unit of time is... Figure 2 Each small grid on the horizontal axis is collectively referred to as a unit of time.

[0141] Cellular users send status information to repeaters. If the information sent by the cellular user meets the spectrum efficiency constraint, the information is successfully sent; otherwise, it will be retransmitted. After receiving the information, the repeater forwards the information to the base station. If the forwarded information meets the spectrum efficiency constraint, the information is successfully sent; otherwise, it will be retransmitted. After the information is successfully sent to the base station, the cellular user proceeds to the next information transmission. Therefore, there is no queuing phenomenon.

[0142] When a repeater operates under interference-limited conditions, it will be subject to interference from q multiplexed channels of D2D users, using Q... D ={I1,I2,...,I q} represents, where QD I represents the set of all D2D users on multiplexed channels. q Let q represent the D2D user in the q-th multiplexed channel. Since the repeater is mainly affected by interfering users, the noise experienced by the repeater can be ignored. However, the base station is affected by additive white Gaussian noise, and the base station noise power is denoted as σ. 2 ; stipulate | h C,M | 2 ,|h M,B | 2 ,|g l | 2 Let represent the channel gain between the cellular user and the repeater node, the channel gain between the repeater node and the base station, and the interference channel gain caused by the D2D user transmitter to the repeater node, respectively. All channels satisfy Rayleigh block fading and follow the parameters λ. C,M , λ M,B τ follows an exponential distribution, l∈{1,2...q}.

[0143] Calculation of the probability of a cellular user or repeater successfully transmitting information under different modes:

[0144] The probability of a cellular user or repeater successfully transmitting information under both decode-forward and amplify-forward modes is calculated using Shannon's theorem and probability theory. Cellular users or repeaters are uniformly referred to as "users" below, and the calculation method is as follows:

[0145] Decoding and forwarding:

[0146] To ensure successful transmission, the information sent by the user must meet the spectral efficiency constraint, namely:

[0147] log(1+γ)≥r (1)

[0148] Where: γ is the signal-to-interference-plus-noise ratio (SINR), and r refers to the minimum amount of information required for each channel to successfully transmit information. In the decode-and-forward mode, the forwarded signal (i.e., the signal sent by the repeater to the base station, and also the signal received by the repeater from the cellular user) is obtained by the user demodulating, decoding, re-encoding, and modulating the received signal. Assuming... Figure 1 The cell user transmit power in the D2D multiplexing cellular model system shown in the interference environment is P. C The repeater's transmit power is P M The D2D user transmit power is P D Therefore, the signal-to-interference-plus-noise ratios (SIRs) of the repeater and the base station are respectively:

[0149]

[0150]

[0151] Where C represents cellular users, M represents repeaters, and D represents D2D users; substituting Equations 2 and 3 into the spectral efficiency constraint (i.e., Equation 1), the probabilities of cellular users and repeaters successfully transmitting information are obtained as follows:

[0152]

[0153]

[0154] Where: λ C,M τ refers to the parameter that the channel gain between cellular users and repeaters follows an exponential distribution, τ is the parameter that the interference channel gain causing interference from the D2D user transmitter to the repeater node follows an exponential distribution, and λ is the parameter that λ follows an exponential distribution. M,B Let σ be the parameter of the exponential distribution that the channel gain between the repeater node and the base station follows. 2 This refers to the base station noise power.

[0155] Amplify and forward:

[0156] In the amplification-retransmission mode, the repeater directly amplifies the received signal using a fixed gain method, meaning the power amplification factor is a fixed value. This method is also known as "semi-blind." The instantaneous forwarding power in the fixed-gain mode varies with the received power, but the repeater does not need to obtain precise first-hop channel state information. In this case, the signal-to-interference-plus-noise ratio (SNR) at the repeater and the base station are respectively:

[0157]

[0158]

[0159] Where β is the amplification factor of the repeater, expressed as:

[0160]

[0161] Similarly, after substituting the spectral efficiency constraint, the probabilities of cellular users and repeaters successfully transmitting information are obtained as follows:

[0162]

[0163]

[0164] in μ2=σ 2 P C , x is the integration variable, representing the integral over time;

[0165] Calculation of average information age:

[0166] The average information age of the system is calculated using the properties of information age images and Taylor series; the calculation method is as follows:

[0167] like Figure 2 The figure shows the information age change trend of a discrete system. According to the information age change trend of a discrete system, in time slot n, the information age Δ(n) is defined as the difference between the current time and the generation time U(n) of the latest data packet received by the destination, i.e., Δ(n) = nU(n); whenever the destination user receives a new data packet, the information age will be reset to 1. Figure 2 n k X represents the time when the information source (in the communication model, nodes that send updates are collectively referred to as information sources; in this model, cellular users and D2D users can both be collectively referred to as information sources) begins its k-th update. k This represents the update time interval between the k-th and k+1-th base station update data packets;

[0168] When K transmissions are successfully completed in N time slots (referring to the entire transmission process from the cellular user to the repeater and then to the base station), the average information age is expressed as:

[0169]

[0170] where Q k This represents the area under the curve representing the information age Δ(n) at the k-th update. As N→∞, Figure 1 The average information age of the system shown is expressed as:

[0171]

[0172] in To update the generated steady-state rate, E(Q) is the expectation of Q, i.e. the mean, where Q represents the area under the entire information age curve;

[0173] Q k Split into X k The sum of the lengths of the rectangles, where 1 ≤ j ≤ X. k The width is 1, and the number of rectangles and the time interval are numerically equal, therefore Q k Represented as:

[0174]

[0175] After introducing the expectation, the mean area under the information age curve is expressed as:

[0176]

[0177] If a user transmits a data packet k times, it means the first k-1 transmissions failed and the kth transmission succeeded. Therefore, using Taylor series simplification, the average interval time can be expressed as:

[0178]

[0179] Similarly, we have:

[0180]

[0181] Where: E(X) represents the expectation of the time interval X, E(X) 2 ) is the expected value of the square of the time interval X, where m is the number of transmissions by the cellular user, n is the number of repeater forwardings, and π is the expected value. C and π M These represent the probabilities of cellular users and repeaters successfully sending information in the corresponding modes, respectively. Substituting (15) and (16) into the system's average information age expression (12), we obtain the final expression for the average information age:

[0182]

[0183] Cellular resource scheduling strategy:

[0184] The algorithm for tunic swarm optimization is improved by using random parameter selection and Cauchy mutation to schedule channel resources and minimize the average information age.

[0185] This invention proposes an enhanced randomized tunic swarm algorithm to allocate transmit power for D2D users in a cellular cell, minimizing the average information age at the base station; the specific strategy is as follows:

[0186] The salp swarm algorithm (SSA) simulates the aggregation behavior of salps, which form chains and then engage in predation and movement. A salp chain consists of two types of salps: leaders and followers. The leader is the salps at the front of the food chain, while the rest are considered followers. It is assumed that there is a food source, F, in the search space as the target of the population.

[0187] To update the leader's position, the following formula exists:

[0188]

[0189] Let F be the position of the first tunicate (leader) in the j-th dimension. j Let ub be the position of the food source in the j-th dimension. j Let lb be the upper bound of the j-th dimension. j c1, c2, c3 are random numbers and represent the lower bound of the j-th dimension.

[0190] The coefficient c1 is the convergence factor of the SSA algorithm. During the iteration process, it balances the algorithm's global search and local exploitation. Its definition is as follows:

[0191]

[0192] In the formula, l is the current iteration number, L is the maximum iteration number, and e is the base of the natural logarithm function.

[0193] Parameters c2 and c3 are random numbers uniformly generated within the interval [0,1]. To update the position of the follower, Newton's laws of motion are used:

[0194]

[0195] In the formula, i≥2, Let represent the position of the i-th tunicate (follower) in the j-th dimension, t represent time, v0 represent the initial velocity, and a0 represent the acceleration. Since time is represented by the number of iterations during algorithm optimization, and the interval between iterations is 1, t = 1. At the start of the iteration, the tunicate's initial velocity is 0, hence v0 = 0. Considering a0 = (v... end -v0) / t, Therefore, it can also be expressed as

[0196]

[0197] The above formulas establish the mathematical model of the SSA algorithm. When solving the optimization problem, the positions of salps are first randomly initialized according to the upper and lower limits of the population position. Then, the fitness value of each salps is calculated. After finding the optimal salps position and assigning it to the food source, the salps leader and followers update their positions sequentially according to the formula.

[0198] This invention improves upon the shortcomings of the tunic group algorithm by proposing an enhanced random tunic group algorithm.

[0199] First, since parameter c1 decreases linearly with the number of iterations, this linear decrease is likely to cause the algorithm to get stuck in local optima and become inefficient. However, the strategy of randomly adjusting control parameters (see Liu Zhusong and Li Sheng, "Sine-Co-chaotic Bisine Whale Optimization Algorithm," Computer Engineering and Applications, 2018, vol.54, no.3, pp.159-163+212) allows the convergence factor a to be randomly selected without being constrained by the number of iterations, thus enabling the algorithm to escape local optima. This invention changes parameter c1 to the convergence factor a, as shown in the following mathematical formula:

[0200] a(t)=(a max -a min)·rand+σ·randn (22)

[0201] Among them, a max and a min These are the maximum and minimum values ​​of the convergence factor a, respectively. rand is a random number in the interval [0,1], randn is a normally distributed random number, and σ (variance) is used to measure the degree of deviation between the convergence factor a and its mathematical expectation, and can control the error of the control parameter in the value selection.

[0202] Meanwhile, swarm intelligence optimization algorithms share a common drawback in the later stages of iteration: as the number of iterations increases, the population tends to converge towards a single optimal individual region, thus losing the opportunity to explore other locations in space. In other words, the loss of population diversity makes the algorithm prone to getting trapped in local optima. Therefore, this method improves upon this by perturbing the population after each iteration to maintain its diversity and thus search for the global optimum as much as possible. The Cauchy distribution has been proven to produce a larger variation length than Gaussian mutation, exhibiting better perturbation and search capabilities. This method combines the position update of the tunicate with the Cauchy mutation factor, whose mathematical expression is defined as follows:

[0203]

[0204] in, This represents the optimal search agent in the j-th dimension of the current search space, i.e., the optimal search position, where Cauchy is the Cauchy mutation factor.

[0205] Based on the above optimizations, the enhanced random tunic position update formula of the present invention can be expressed as:

[0206]

[0207] This method first enables D D2D users to reuse cellular channels using a random allocation method. Simultaneously, it must satisfy the following conditions: a single cellular user is reused by only one D2D user, and a D2D user reuses only one cellular channel. Both D2D users and cellular users meet spectral efficiency constraints. Then, an enhanced randomized tunic algorithm is used to allocate the transmit power of the D2D users, achieving the minimum information age (AoI) at the base station. The objective function and constraints are as follows:

[0208]

[0209] Where the reuse parameter When, it indicates that D2D user d is multiplexing the channel of cellular user c; otherwise...

[0210] γ M1 Indicates the signal-to-interference-plus-noise ratio (SIR) of the repeater in decode-and-forward mode; γB1 P represents the signal-to-interference-plus-noise ratio (SIR) of the base station in decode-and-forward mode. Dmax This represents the maximum transmit power for D2D users, where C is the number of cellular users, and r refers to the minimum amount of information used per channel (r bits); γ d The signal-to-interference-plus-noise ratio (SIR) of a D2D user is calculated using the following formula:

[0211]

[0212] Where: h d 2 Indicates the channel gain between D2D users; g d 2 This indicates the link gain between the repeater and the D2D user;

[0213] Constraint 1 is the spectral efficiency constraint satisfied by cellular users; Constraint 2 is the spectral efficiency constraint satisfied by repeaters; Constraint 3 is the spectral efficiency constraint satisfied by D2D users.

[0214] Based on the above objective function and constraints, the fitness function of this invention is:

[0215]

[0216]

[0217] In the formula, μ is the penalty factor, μ > 0; the value of f is continuously calculated during the iteration process, and the transmit power P of the D2D user is minimized when the value of f reaches its minimum. D This approach minimizes the final average information age, thus ensuring the objective of minimizing the average information age while satisfying the spectral efficiency constraint.

[0218] The optimal solution is the power of D2D users that minimizes the final average information age; f is defined by the objective function, which includes both minimizing the information age at the base station (the part before the plus sign) and the spectral efficiency constraint of D2D users (the part after the plus sign). The algorithm continuously calculates the value of f during the iteration process. When the value of f reaches its minimum, the objective of this invention is guaranteed: minimizing the information age while satisfying the spectral efficiency constraint. f serves as the standard for the algorithm to find the optimal solution for the objective value.

[0219] The effects of this invention:

[0220] First, to explore the advantages and disadvantages of different forwarding modes of repeaters under interference conditions, the average AoI change trend of base stations under decoding forwarding and amplification forwarding modes was simulated. Cellular user transmit power, spectral efficiency constraints, and distance between users were used as variables to compare the information transmission effects of different forwarding modes.

[0221] Figure 4 The simulation results show the trend of average AoI with cellular user transmit power. As can be seen from the simulation graph, the average AoI in the amplification-forward scenario is greater than that in the decoding-forward scenario. This is because in the amplification-forward scenario, the repeater amplifies and forwards both the information transmitted by the cellular user and the co-channel interference from D2D users simultaneously, leading to a decrease in the probability of successful information transmission and an increase in average AoI. Secondly, in both scenarios, the average AoI decreases as the cellular user transmit power increases. This is because increasing power allows the cellular user to transmit more information, thus reducing the average AoI. Finally, by comparing the number of interferences under the same conditions, it can also be seen that increased interference also increases the AoI.

[0222] Figure 5 The figure shows the trend of average AoI as a function of spectral efficiency constraints. It is still evident from the figure that the average AoI is higher in the amplified forwarding scenario, and that increased interference negatively impacts the destination AoI. Meanwhile, the average AoI decreases in both the decoded forwarding and amplified forwarding scenarios as the spectral efficiency constraint decreases. This is because increasing the spectral efficiency constraint raises the threshold for successful forwarding, increasing the probability of forwarding failure and reducing the base station's information update speed, thus increasing the average AoI.

[0223] Figure 6 The figure shows the trend of average AoI (Average Address Value) as the distance between the cellular user and the base station changes with the distance between the cellular user and the repeater when the distance between the cellular user and the base station is fixed. As can be seen from the figure, regardless of whether amplification-forwarding or decoding-forwarding is used, the AoI is not necessarily lower the closer the repeater is to the cellular user. In each case, a point can be found that minimizes the average AoI of the base station. Furthermore, the optimal point for amplification-forwarding is closer to the cellular user than for decoding-forwarding. This is because for amplification-forwarding, the micro base station needs to amplify the cellular user information before forwarding, which can be considered an analog signal processing method; while for decoding-forwarding, the repeater needs to detect and estimate the received signal before forwarding, which can be considered a digital signal processing method. The advantages of digital signal processing over analog signal processing are mainly accurate transmission, strong anti-interference capability, and fast long-distance transmission speed without distortion. Therefore, under the condition of fixed cellular user and base station distances, amplification-forwarding requires the repeater to be closer to the signal source.

[0224] In the cellular model, that is Figure 4 As can be seen from 5 and 6, decoding and forwarding are more effective than amplification and forwarding, so the decoding and forwarding mode was adopted in the cellular model.

[0225] Meanwhile, to evaluate the technical effectiveness of the cell resource scheduling method based on information freshness, this method was compared with existing Particle Swarm Optimization (PSO) and Salp swarm algorithm (SSA). The number of cellular users and D2D users were also used as variables to compare the superiority of the algorithm under different numbers of users.

[0226] Figure 7 The graph compares three resource scheduling strategies for a D2D user count of 15 and a cellular user count of 50. The simulation results show that PSO reaches its optimal value at 137 iterations, with an optimal value of 8.678 * 10^6. 5 SSA reaches its optimal value at 415 runs, with an optimal value of 1.389 * 10. 5 ASSSA reaches its optimal value at 122 runs, with an optimal value of 2.531 * 10. 4 Therefore, it can be seen that the tunic swarm algorithm can obtain a smaller average AoI compared with the particle swarm algorithm, but the convergence speed is slower. The proposed enhanced stochastic tunic algorithm is significantly better than the tunic algorithm and the particle swarm algorithm in terms of convergence speed. At the same time, the average AoI obtained at the base station also has a significant advantage over the other two algorithms.

[0227] Figure 8 The chart compares three resource scheduling strategies for a D2D user count of 25 and a cellular user count of 50. The results show that PSO reaches its optimal value in 143 runs, with an optimal value of 3.459 * 10^13. 7 SSA reaches its optimal value at 387 runs, with an optimal value of 6.733 * 10. 6 ASSSA reaches its optimal value at 304 runs, with an optimal value of 4.915 * 10. 6 It can be seen that the proposed algorithm still outperforms the other two algorithms in both convergence speed and minimizing the average AoI. Furthermore, based on the number of D2D users, it can be observed that the average AoI at the base station increases with the increase in the number of D2D users. This is because the increase in D2D users leads to more reused cellular users, i.e., increased interference, thus increasing the AoI at the base station.

[0228] Figure 9 The chart compares three resource scheduling strategies for a D2D user count of 15 and a cellular user count of 75. The results show that PSO reaches its optimal value at 254 iterations, with an optimal value of 1.461 * 10^- ... 6 SSA reaches its optimal value at 268 runs, with an optimal value of 1.851 * 10^6. 5 ASSSA reaches its optimal value at 251 runs, with an optimal value of 6.051 * 10^25. 4By comparison Figure 5 The number of cellular users varies, indicating that changes in the number of cellular users do not significantly affect the AoI at the base station. However, compared to the scenario with an increase in D2D users, the proposed improved tunic algorithm has a more obvious advantage in the scenario with an increase in the number of cellular users.

[0229] In summary, simulation results show that, compared with other existing algorithms, the proposed cell resource scheduling method based on information freshness significantly reduces the average information age.

[0230] This method not only reduces the information age of the system, but also improves the convergence speed of the algorithm.

[0231] This method calculates the probability of a user successfully sending a message by using Shannon's theorem and probability theory, and calculates the closed expression of the system's average message age based on the properties of the message age image and Taylor series.

[0232] This method compares the relay forwarding methods under interference conditions, solves the information age expression in two modes, and compares the advantages and disadvantages of the two forwarding methods through simulation. It also shows that a point can be found in both modes that minimizes the average AoI of the base station.

[0233] This method addresses cellular systems under interference. Considering D2D multiplexing of cellular channels for information transmission, it proposes a new resource scheduling strategy based on information freshness metrics. Compared to existing strategies, this strategy can further reduce the average information age of the system and improve the freshness of the information.

[0234] This method is not limited to the above embodiments, and the number of cellular users and D2D users can be changed according to the actual situation.

[0235] The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the scope of protection of the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, any person skilled in the art can make equivalent substitutions or changes based on the technical solution and inventive concept of the present invention within the scope of the technology disclosed in the present invention. These simple modifications are all within the scope of protection of the present invention.

[0236] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not describe the various possible combinations separately.

[0237] Furthermore, various different embodiments of the present invention can be combined in any way, as long as they do not violate the spirit of the present invention, they should also be regarded as the content disclosed by the present invention.

Claims

1. A cellular resource scheduling method based on information freshness, characterized in that... Includes the following: Step 1: Set up the cellular system model: In a cellular cell centered on a base station, there are C cellular users and D D2D users simultaneously, and the cellular users and D2D users share channel resources, resulting in co-channel interference. Cellular users send status information to repeaters. If the information sent by the cellular user meets the spectrum efficiency constraint, the information is successfully transmitted; otherwise, it will be retransmitted. After receiving the information, the repeater forwards the information to the base station. If the forwarded information meets the spectrum efficiency constraint, the information is successfully transmitted; otherwise, it will be retransmitted. After the information is successfully sent to the base station, the cellular user proceeds to the next information transmission. When the repeater operates under interference-limited conditions, it will be subject to interference from q multiplexed channels of D2D users. The base station noise power is denoted as σ. 2 ; stipulate | h C,M | 2 ,|h M,B | 2 ,|g l | 2 Let represent the channel gain between the cellular user and the repeater, the channel gain between the repeater node and the base station, and the interference channel gain caused by D2D users to the repeater node, respectively. All channels satisfy Rayleigh block fading and follow the parameters λ. C,M , λ M,B τ is an exponential distribution, l∈{1,2...q}; Step two, to determine the repeater's forwarding method, calculate the probability of a cellular user or repeater successfully sending information under different modes: The probability of different users successfully sending information under both decode-forward and amplify-forward modes is calculated using Shannon's theorem and probability theory. The calculation method is as follows:

1. Decoding and forwarding mode: The probability π of cellular users and repeaters successfully transmitting information in decode-forward mode C1 and π M1 Calculate according to the following formulas: Where: P{} represents the probability of taking the formula within {}, and the cellular user transmit power is P. C The repeater's transmit power is P M The D2D user transmit power is P D r refers to the minimum amount of information required for each transmitting channel to successfully send information; Second, amplified forwarding mode: The probability π of cellular users and repeaters successfully transmitting information in amplified forwarding mode C2 and π M2 Calculate according to the following formulas: Where: β is the amplification factor of the repeater, and the calculation formula is: x is the integration variable, representing the integral over time; Step 3: Calculating the average information age under different models: The average information age is calculated using the image properties of information age and Taylor series when K transmissions are successfully completed in the entire transmission process from cellular user to repeater to base station. The average information age is then used to calculate this value. Calculate using the following formula: Where: X is the time interval for the base station to update data packets, and E(X) represents the expected value of the time interval X, calculated by the following formula: E(X 2 The expected value of the square of the time interval X is calculated using the following formula: Where m is the number of cellular user transmissions, n is the number of repeater forwardings, and π C and π M These represent the probabilities of cellular users and repeaters successfully sending information in the corresponding modes; Step 4: Based on the calculation results of Step 2 and Step 3, compare the average information age when the entire transmission process from cellular user to repeater to base station is successfully completed K times under different forwarding modes, and take the decoding and forwarding mode with the smaller average information age as the forwarding mode determined by this method. Step 5, Cellular Resource Scheduling Strategy: An enhanced stochastic tunic swarm algorithm is proposed to allocate transmit power for D2D users in a cell, minimizing the average information age at the base station. The specific strategy is as follows: Step 5.1 stipulates that D D2D users reuse cellular channels using a random allocation method, while ensuring that a single cellular user is reused by only one D2D user, and a D2D user reuses only one cellular channel. Furthermore, the information transmitted by both D2D users and cellular users must satisfy the following spectral efficiency constraints: log(1+γ)≥r Where: γ is the signal-to-interference-plus-noise ratio; Step 5.2, Initialize the location of the salps population: that is, randomly set the location of each salps in the population, where the location of the salps represents the possible transmission power of the D2D user that minimizes the average information age. The location of the salps includes three categories: food source location, salps leader location, and salps follower location. Step 5.3: Calculate the fitness f of the locations of all tunicates in the initial population. The location of the tunicate with the smallest f value corresponds to the food source location. In other words, the D2D user's transmit power P is at the smallest f value. D This minimizes the final average information age, and the minimum value of f guarantees the objective: minimizing the average information age while satisfying the spectral efficiency constraint; the fitness f at the location of the tunic is calculated by the following formula: In the formula, μ is the penalty factor, and μ > 0. The fitness f includes two parts: the objective function and the constraints. The objective function and constraints are as follows: C > D 0<P D <P Dmax Where: when reuse parameters When, it indicates that D2D user d is multiplexing the channel of cellular user c; otherwise... γ M1 Indicates the signal-to-interference-plus-noise ratio (SIR) of the repeater in decode-and-forward mode; γ B1 P represents the signal-to-interference-plus-noise ratio (SIR) of the base station in decode-and-forward mode. Dmax This represents the maximum transmit power of D2D users, where C is the number of cellular users, D is the number of D2D users, and γ is the maximum transmit power of D2D users. d Indicates the signal-to-interference-plus-noise ratio (SIR) for D2D users; Step 5.4, update the location of the leader of the tunicate sea lizard according to the following formula: Where: the j-th dimension represents the j-th D2D user in the cellular system model. The updated position of the tunicate leader, i.e., the position of the tunicate leader in the j-th dimension, is the transmit power of the D2D user among the j D2D users that may minimize the average information age, F. j Let ub be the location of the food source in the j-th dimension, that is, the location of the tunicate corresponding to the minimum f value in step 5.

3. j Let lb be the upper bound of the j-th dimension, i.e., the maximum possible transmit power of a D2D user. j Let c2 and c3 be the lower bound of the j-th dimension, i.e., the possible minimum transmit power of a D2D user, where c2 and c3 are random numbers, Cauchy is the Cauchy variation factor, and a is the lower bound of the j-th dimension. max and a min Let a and b be the maximum and minimum values ​​of the convergence factor a, respectively, where a(t) = (a max -a min )·rand+σ·randn, where rand is a random number in the interval [0,1], randn is a normally distributed random number, σ is used to measure the degree of deviation between the convergence factor a and its mathematical expectation, and t is time; Then update the position of the salver followers according to the following formula: Where: i≥2, This represents the position of the i-th follower of the salver besides the leader of the salver, in the j-dimensional space. That is, the position of the follower of the salver is the transmission power of the j-th D2D user other than the transmission power of the D2D user corresponding to the position of the salver leader. Step 6: Set the maximum number of iterations, and then perform iterative calculations on steps 5.3-5.4 in step 5 until the maximum number of iterations is reached. Output the food source location obtained from the last iteration as the optimal location, thus obtaining the transmit power value of the D2D user that minimizes the average information age at the base station.

2. The cellular resource scheduling method based on information freshness according to claim 1, characterized in that... In step 5.3 of step five, under the decoding and forwarding mode, the signal-to-interference-plus-noise ratio γ of the repeater and the base station M1 γ B1 Calculate according to the following formulas:

3. The cellular resource scheduling method based on information freshness according to claim 1, characterized in that... In step 5.3 of step five, γ d The signal-to-interference-plus-noise ratio (SIR) of a D2D user is calculated using the following formula: Where: |h d | 2 Represents the channel gain between D2D users; |g d | 2 This indicates the link gain between the repeater and the D2D user.

Images