A NOMA-VLC power allocation method based on sparrow search algorithm

Abstract

A NOMA-VLC power allocation method based on a sparrow search algorithm is proposed. The method determines the relevant parameters of the transmitter and user's receiving power distribution (PD). A line-of-sight link is used as the simulated link type. A mathematical model is established to maximize system and rate based on user fairness, and the relevant parameters in the model are determined. An objective function and constraints are determined, and the logarithmic utility function is applied to the user's rate and summed to obtain the objective function. The objective function is used as the fitness function of the sparrow search algorithm, and the parameters of the sparrow search algorithm are determined. The optimal sparrow's position and fitness value are found through the sparrow search algorithm, thereby obtaining the optimal power allocation result and the maximum system and rate. The sparrow search algorithm proposed in this invention exhibits good performance in terms of search accuracy, convergence speed, and stability. It supports arbitrary scenario parameters, arbitrary transmitter / receiver parameters, and arbitrary number of users. The simulation results and analysis provide guidance for multi-user power allocation.

Description

Technical Field

[0001] This invention belongs to the field of communication technology and mainly relates to power allocation in non-orthogonal multiple access visible light communication systems, specifically a NOMA-VLC power allocation method based on the sparrow search algorithm. Background Technology

[0002] Performance evaluation of NOMA systems primarily focuses on three metrics: total system throughput, energy efficiency, and user fairness. Among these, the power allocation algorithm significantly impacts the system's performance in these three metrics. Most existing NOMA downlink power allocation algorithms aim to maximize system throughput or energy efficiency while ensuring a certain user data rate. Although these methods make some efforts to guarantee user fairness, the results in the literature are limited to comparisons of algorithms in terms of total system throughput or energy efficiency, lacking specific quantitative metrics to measure user fairness. Therefore, it is impossible to intuitively assess the balance between total system throughput and user fairness achieved by the algorithm.

[0003] While some progress has been made in research on multi-user power allocation in NOMA-VLC systems, some problems still remain, such as:

[0004] "Research and Optimization of User Packet and Power Allocation Algorithm in NOMA Downlink," Beijing University of Posts and Telecommunications, Xie Xuexiao. This scheme does not use the logarithmic utility function of the sum and rate as the objective function, failing to achieve a balance between system sum and rate and user fairness. Furthermore, its constraints are insufficient, failing to consider the luminous capability of LED devices and the need to protect human eyes, and also failing to consider that the total allocated power is limited by the maximum transmit power.

[0005] The paper, "Optimization of Non-Orthogonal Multiple Access Based Visible Light Communication Systems," published in IEEE Communication Letters by Zaib Tahira et al., focuses on the sum and rate of the NOMA system performance, neglecting user fairness. Furthermore, it only discusses power allocation for two users, raising questions about its applicability to multi-user power allocation.

[0006] Therefore, existing multi-user power allocation methods in NOMA-VLC systems have the following drawbacks: (1) Existing technologies cannot achieve a balance between the total system throughput and user fairness. (2) Existing technologies do not comprehensively consider the setting of constraints. (3) There is no verification that the algorithm is suitable for multi-user power allocation. These drawbacks reduce system performance and user fairness. Therefore, it is essential to propose a power allocation method that can simultaneously guarantee user fairness and system performance. Summary of the Invention

[0007] In order to overcome the shortcomings of the prior art, the present invention aims to provide a NOMA-VLC power allocation method based on the sparrow search algorithm, in order to solve the problem that the existing power allocation algorithm cannot simultaneously guarantee user fairness and system and rate, and to support arbitrary scenario-related parameters, arbitrary transmitter / receiver-related parameters, and arbitrary number of users.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] Firstly, a NOMA-VLC power allocation method based on the sparrow search algorithm is provided, comprising the following steps:

[0010] Step 1: Determine the relevant parameters of the transmitter and the user's receiving PD;

[0011] Step 2: Use line-of-sight links as the link type for simulation;

[0012] Step 3: Establish a mathematical model based on maximizing user fairness in the system and rate, and determine the relevant parameters in the model;

[0013] Step 4: Determine the objective function and constraints. Apply the logarithmic utility function to the user's rate and sum the results as the objective function.

[0014] Step 5: Using the objective function as the fitness function of the sparrow search algorithm, determine the parameters of the sparrow search algorithm, including: maximum number of iterations, total number of sparrows, ratio of discoverers, followers and vigilants, current iteration number and the position of all sparrows, where the position of each sparrow represents the allocated power of each user in the NOMA system;

[0015] Step 6: Find the optimal sparrow's position and fitness value using the sparrow search algorithm, thereby obtaining the optimal power allocation result and the maximum system and rate.

[0016] In a second aspect, a computer-readable storage medium is provided, characterized in that it is used to store a computer program that causes a computer to execute the NOMA-VLC power allocation method based on the sparrow search algorithm.

[0017] Thirdly, a computer program product containing instructions is provided, characterized in that, when the instructions are executed on a computer, the NOMA-VLC power allocation method based on the sparrow search algorithm is made to work.

[0018] Most existing NOMA downlink power allocation algorithms aim to maximize system throughput or energy efficiency while ensuring a certain user data rate. Although these methods have made some efforts to ensure user fairness, the results in the literature are limited to comparing the algorithms' total system throughput or energy efficiency, without specific quantitative indicators to measure user fairness. They fail to achieve a balance between total system throughput and user fairness, and the constraints are not comprehensive, neglecting the luminous capability of LED devices, the need to protect human eyes, and the limitation of total allocated power by the maximum transmit power. Therefore, current NOMA-VLC power allocation methods cannot simultaneously guarantee user fairness and system data rate, which interferes with research on NOMA-VLC multi-user power allocation.

[0019] This invention solves the problem that existing power allocation algorithms cannot simultaneously guarantee user fairness and system and speed, and has the following advantages compared with existing technologies:

[0020] (1) Achieving a balance between the total throughput of the system and user fairness: Based on a NOMA-VLC downlink system with a single transmitter and multiple users, this invention establishes a mathematical model based on maximizing system and rate by using a logarithmic utility function for the user rate and summing it as the objective function in order to ensure user fairness.

[0021] (2) Obtaining the optimal power allocation of NOMA in VLC systems more accurately: This invention proposes a NOMA-VLC power allocation strategy based on the sparrow search algorithm. The sparrow search algorithm has good performance in terms of search accuracy, convergence speed and stability, and can effectively obtain the optimal power allocation of NOMA in VLC systems.

[0022] (3) Engineering application value: This invention fully considers the objective function and constraints of the power allocation optimization problem, and supports arbitrary scenario-related parameters, arbitrary transmitter / receiver-related parameters, and arbitrary number of users. The simulation results and analysis have guiding significance for multi-user power allocation. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0024] Figure 1 This is a flowchart of the present invention.

[0025] Figure 2 This is a system model diagram of the present invention.

[0026] Figure 3 This is the schematic diagram of SiC.

[0027] Figure 4 This is a flowchart of the sparrow search algorithm iteration process.

[0028] Figure 5 This is a graph showing the relationship between the number of users and the data rate.

[0029] Figure 6 This is a graph showing the relationship between the number of users and the Jain Fairness Index.

[0030] Figure 7 The relationship between transmit power and transmission rate is shown for different numbers of users. (a) to (f) represent the number of users as 2, 3, 4, 5, 6, and 7, respectively. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the present invention clearer, the following detailed description is provided in conjunction with the accompanying drawings and specific embodiments.

[0032] Example 1

[0033] Performance evaluation of NOMA systems primarily focuses on three metrics: total system throughput, energy efficiency, and user fairness. Among these, the power allocation algorithm significantly impacts the system's performance in these three metrics. Most existing NOMA downlink power allocation algorithms aim to maximize system throughput or energy efficiency while ensuring a certain user data rate. Although these methods make some efforts to guarantee user fairness, the results in current technologies are limited to comparing the algorithms' performance in terms of total system throughput or energy efficiency, lacking specific quantitative metrics to measure user fairness. Therefore, it is impossible to intuitively assess the balance between total system throughput and user fairness achieved by the algorithm.

[0034] To address the issue that existing technologies cannot simultaneously guarantee user fairness and system performance / rate, a mathematical model based on maximizing system performance / rate while ensuring user fairness is established. Considering the non-negativity of VLC signals and eye safety, optical power constraints are set, and a NOMA-VLC power allocation method based on the sparrow search algorithm is proposed. (See [link to relevant documentation]). Figure 1 , Figure 1 The flowchart of this invention includes the following steps:

[0035] Step 1: Determine the relevant parameters of the transmitter (generally an LED transmitter in the NOMA system of this invention) and the user's receiving PD: Establish a Cartesian coordinate system with the lower left corner of the scene as the origin, the width direction as the X-axis, the length direction as the Y-axis, and the height direction as the Z-axis. The system model is as follows: Figure 2 As shown. Specifically, the parameters mainly include the transmitter's coordinates, the vertical height L of the transmitter above the ground, and the Euclidean distance r between the location of the i-th user and the transmitter's projected position on the receiving plane. i The line-of-sight link distance d between the transmitter and the i-th user i And the radius of the illuminated area.

[0036] Since the users are evenly distributed near the transmitter, r i It can be obtained through Monte Carlo simulation, r i The probability density function can be given as:

[0037]

[0038] Where r represents the distance from the user to the center of the circle.

[0039] Line-of-sight link distance between the transmitter and the i-th user

[0040] Step 2: Determine the link type for simulation. Visible light communication channels include two types of communication links: line-of-sight (LOS) links and NLOS links. Specifically, an LOS link refers to a transmitted light signal that propagates directly to the receiver via a straight path without being blocked or reflected by anything; it constitutes the majority of visible light communication systems. An NLOS link refers to a transmitted light signal that does not propagate directly to the receiver in a straight line, but rather reaches the receiver after being reflected by some object (such as a wall). A first-order NLOS link means the signal reaches the receiver after only one reflection, a second-order link means the signal reaches the receiver after two reflections, and so on.

[0041] The power of a first-order reflection is much less than the power of a LOS link, and the power obtained from higher-order reflections is even less than the power of a first-order reflection. This invention only considers LOS links.

[0042] Step 3: Establish a mathematical model based on maximizing user fairness and system speed, and determine the relevant parameters in the model, mainly including: DC bias I. DC Peak light intensity B, maximum transmission power p max The effective area A of the receiver PD The receiver's field of view φ 1 / 2 and the optical filter gain T of the receiver s (φ).

[0043] In this invention, all users employ M-ary Pulse Position Modulation (PPM), and the modulation signal s of the i-th user is... i The average value is zero, so we can get s. i ∈[-δ,δ], i=1,2,3,…,N, where N is the number of users. Here, δ>0 is a coefficient determined by the modulation order M, p i It is the power allocated to different users, and p i >0.

[0044] To ensure the non-negativity of the transmitted signal, the transmitted signal S can be represented as:

[0045]

[0046] The data is modulated using an 8-ary pulse position modulation scheme, because p i >0, according to the above formula, we can obtain:

[0047]

[0048] In NOMA-VLC systems using IM / DD, optical power is proportional to electrical amplitude. Based on the luminous capability of LED devices and the need to protect human eyes, the transmitted light intensity should be limited by the permissible peak light intensity B, i.e.:

[0049]

[0050] In addition, the total power of all users should be less than the maximum transmission power, expressed as:

[0051]

[0052] Receiver field of view φ 1 / 2 This indicates the maximum angle at which the receiver can receive incident light, when the angle between the incident light and the receiver's normal vector is greater than the receiver's field of view φ. 1 / 2 When the light signal is at a certain time, no light signal will be received; conversely, when it is at a certain time, a light signal will be received.

[0053] The optical filter gain T of the receiver s (φ) represents the gain of the optical filter for different wavelengths of light signals. An optical filter is an optical element that can selectively allow or block light of a specific wavelength. In a receiver, optical filters are typically used to select a specific wavelength range of the received light. Generally, T s (φ) = 1, meaning that light of all wavelengths can pass through the filter.

[0054] The superimposed signals are received after passing through the VLC channel, and independent zero-mean additive white Gaussian noise n is introduced. i Furthermore, the constant DC offset is removed. The signal received at the i-th user can be expressed as:

[0055]

[0056] in, This represents the interference of the first i-1 users on the i-th user. This represents the useful information of the i-th user. This represents the interference from users i+1 to N on user i, where n i The channel noise between the i-th user and the transmitter is additive Gaussian noise with a mean of 0 and a variance of σ. 2 The user's channel gain can be expressed as:

[0057]

[0058] Among them, A PD Let d1 represent the effective area of ​​the receiver, d2 represent the distance between the transmitter and receiver, FOV represent the receiver's field of view, θ represent the angle between the emitted ray and the transmitter's normal vector, φ represent the angle between the incident ray and the receiver's normal vector, and m represent the Lambertian emission coefficient, which can be expressed by the formula m = -ln2 / (lncos(φ)). 1 / 2 ))Calculation yielded φ 1 / 2 Take the half-angle of the half-power point, T s (φ) represents the optical filter gain, and G(φ) represents the condenser gain, specifically in the following form:

[0059]

[0060] Where η represents the refractive index.

[0061] Step 4: Determine the objective function and constraints: When a user is within the coverage area of ​​the LED beam, the channel gain will vary depending on the distance. Without loss of generality, sort the channel gains in ascending order as |h1|. 2 ≤|h2| 2 ≤|h3| 2 …|h N | 2 Using SIC technology at the receiving end can decode the signal of user i with weak channel conditions, and eliminate inter-user interference for the remaining users. Figure 3This is a schematic diagram of the SIC principle. When performing multi-user detection, the SIC receiver detects users one by one in a specific order. Generally, the higher the signal power, the higher the priority of the detection order; that is, the first user signal detected has the highest power, followed by users with lower power. For a series of superimposed signals, the first user signal detected is directly decoded from the original received signal, and then this signal is removed from the initial superimposed signal to decode the next highest power signal. This process continues until all user signals are demodulated.

[0062] Based on the SIC principle described above, user signals are decoded, and the total bandwidth is normalized to a unit. The i-th user employs serial interference cancellation technology, which can perfectly decode user signals with weak channels and partially eliminate inter-user interference. Therefore, the lower bound of the rate for the i-th user is given as follows:

[0063]

[0064] The sum rate of the system can be expressed as:

[0065]

[0066] Furthermore, the logarithmic utility function has been proven to achieve good user fairness in many scenarios of multi-user wireless communication. Therefore, to ensure user fairness, this invention uses the logarithmic utility function on the user rate and sums it as the objective function, resulting in:

[0067]

[0068] Among them, h i For the user's channel gain, p i The power allocated to different users, and p i >0, N is the number of users, σ 2 Let be the variance of the additive Gaussian noise in the channel between the i-th user and the transmitter.

[0069] This invention proposes a problem for maximizing the sum and rate of a NOMA-VLC system based on fairness. To ensure user fairness, a logarithmic utility function is introduced to address the fairness problem, using the logarithmic utility function that maximizes the sum and rate as the objective function. Furthermore, an iterative algorithm is proposed, which can effectively obtain the optimal power allocation for NOMA in a VLC system. In summary, the objective function and constraints of this invention can be expressed as:

[0070]

[0071]

[0072]

[0073] p i ≥0

[0074] Where C = min{I DC BI DC}

[0075] Step 5: Determine two power allocation strategies to compare with the sparrow search algorithm:

[0076] Fixed Power Allocation (FPA) allocates power to users based on the total number of users and their SIC (Self-Induced Channel Gain) order, and is considered a static power allocation scheme. For an N-user NOMA system, users are sorted in ascending order of channel gain, and the ratio of transmit power allocated to any two adjacent users is a fixed value. Specifically, the power allocated to user i is:

[0077] P i =αP i-1 i = 2, 3, ..., N

[0078] Where α (0 < α < 1) represents the power allocation factor. As can be seen from the formula, power is allocated to each user in descending order, and users with lower channel gain can receive higher allocated power. Since the value of the power allocation factor α is fixed, the advantage of this algorithm is its low computational complexity, but its disadvantage is that it does not consider the channel quality of users when allocating power, resulting in poor performance.

[0079] Gain Ratio Power Allocation (GRPA) is a dynamic power allocation algorithm that considers the impact of channel conditions on power allocation and distributes power proportionally to user path loss. It introduces a power allocation factor and adjusts it based on differences in channel gain among users, ensuring that users with poorer channel conditions receive more power, thus achieving fairness in power allocation. In a system with N users, the channel gains are sorted in ascending order, and the power allocated to the i-th user is:

[0080]

[0081] Step 6: Determine the parameters of the sparrow search algorithm, including: maximum number of iterations (iter). max Total number of sparrows, the ratio of discoverers, followers, and vigilants, initial iteration number, and the location of all sparrows. The position of each sparrow represents the allocated power of each user in the NOMA system. The rate is calculated, and the logarithmic utility function is applied to the user's rate and the sum is used as the fitness function.

[0082] Step 7: Calculate the power allocation for users in the sparrow search algorithm: By iteratively obtaining the optimal sparrow's position and fitness value through the sparrow search algorithm, the optimal power allocation result and the maximum system and rate can be obtained, that is, the optimal value of the objective function in Step 4 can be found. The iterative flowchart of the sparrow search algorithm is shown below. Figure 4 As shown, it can be summarized into the following steps:

[0083] Step 1: Initialize the algorithm parameters, mainly including the maximum number of iterations (iter). max The total number of sparrows, the ratio of discoverers, followers, and vigilants, and the initial number of iterations.

[0084] Step 2: Randomly initialize the positions of all sparrows.

[0085] Step 3: Based on the fitness function Substitute the position of each sparrow, calculate the initial fitness value of all sparrows and sort them, and record the best and worst positions of the sparrow population as the initial positions.

[0086] Step 4: Utilize Update the discoverer's location.

[0087] Step 5: Utilize Update follower positions.

[0088] Step Six: Utilize Update the location of the vigilant.

[0089] Step 7: Based on the latest positions of each sparrow after the update, substitute them into... Recalculate the fitness values ​​of all sparrows, compare the obtained fitness values ​​with the original optimal values, update the current optimal and worst values ​​of the sparrow population, and update the global optimal and worst values.

[0090] Step 8: Determine if the current iteration count is less than the set maximum iteration count. If it is less, increment the iteration count by one and continue repeating steps 2 to 7. Otherwise, stop the optimization process, output the optimal sparrow position and fitness value, and thus obtain the optimal power allocation result and the maximum system and speed.

[0091] Step 8: Determine the power allocation factors α1 and α2 for the fixed power allocation algorithm and the gain ratio power allocation algorithm. Based on Step 5, the power allocation factor α1 for the fixed power allocation algorithm is a fixed value set between 0 and 1. The power allocation factor α2 for the gain ratio power allocation algorithm...

[0092] Step 9: Calculate the system sum, rate, and user fairness: Based on steps 7 and 8, the user allocation power under the three algorithms can be obtained, and then the system sum, rate, and user fairness index can be calculated. The system sum and rate can be expressed as:

[0093]

[0094] To evaluate the fairness of the system obtained by the algorithm used, the Jain fairness index is introduced. The fairness index is generally used to represent the fairness of the system, taking a continuous value between 0 and 1, with the closer to 1 indicating a fairer allocation of resources.

[0095]

[0096] Step 10: Calculate the system sum rate and user fairness index under different user numbers and different maximum transmit power conditions: Change the number of users and compare the relationship between the sum rate and the number of users for different algorithms, such as... Figure 5 As shown. By changing the number of users, we can further compare the relationship between the Jain Fairness Index and the number of users for different algorithms, as shown below. Figure 6 As shown. Determine the number of users, change the transmit power, obtain the power allocation factors for the three algorithms according to steps 7 and S8, calculate the system and rate, and compare the relationship between the transmit power and rate of different algorithms, as shown. Figure 7 As shown.

[0097] This invention, based on a NOMA-VLC downlink system with a single transmitter and multiple users, establishes a mathematical model that maximizes system and data rate based on user fairness, thereby improving both system and data rate performance and user fairness. Considering the non-negativity of VLC signals and eye safety, this invention sets optical power constraints and proposes a NOMA power allocation strategy based on a sparrow search algorithm, improving search accuracy, convergence speed, and stability. The proposed sparrow search algorithm supports arbitrary scene-related parameters, arbitrary transmitter / receiver-related parameters, and arbitrary number of users. Simulation results and analysis provide guidance for multi-user power allocation.

[0098] Example 2

[0099] A NOMA-VLC power allocation method based on the sparrow search algorithm is the same as in Embodiment 1. In this embodiment, step 7, calculating the allocated power of users in the sparrow search algorithm, includes the following steps:

[0100] Step 7.1, the sparrow search algorithm simulates the competition, cooperation, and surveillance behaviors of sparrows in foraging and predation. The discoverer typically possesses higher energy reserves and is responsible for searching food-rich areas, providing foraging areas and directions for all followers. Followers, guided by the discoverer, gather after the explorer finds the target to increase their own predation rate. The watchdog is responsible for protecting the entire sparrow population, guarding against predators. It exhibits good performance in terms of search accuracy, convergence speed, and stability. The sparrow search algorithm has been applied in various fields. In this invention, X represents the sparrow position matrix, where each element represents a possible solution to the target problem, resulting in:

[0101]

[0102] Where n is the number of sparrows, d is the dimension of the parameter to be optimized, and N S It represents the population size, and the elements in the matrix represent the values ​​of each parameter to be optimized, x. n,d Let d represent the parameter to be optimized for the nth sparrow, and the position of each sparrow represents the allocated power for each user in the NOMA system.

[0103] Step 7.2, therefore, the fitness value of each sparrow can be expressed as:

[0104]

[0105] In each row f([x i,1 x i,2 … … x i,d The value of ]) represents the fitness value of each individual, which is also the obtained system and velocity. Discoverers with higher fitness values ​​have priority in obtaining food during the search process. When updating the location of a discoverer, it is necessary to determine the magnitude of the alarm value and the safety threshold. R2 (R2∈[0,1]) and S are used respectively. t (S t ∈[0.5,1.0]) represents the alarm value and the safety threshold. When R2 t When R² ≥ S, it indicates that there are no predators nearby, and the discoverer can conduct a wide search. t When this occurs, it indicates that some sparrows have spotted a predator, triggering a warning that all sparrows must quickly fly to a safe area. Therefore, the location update method for the discoverers can be represented as:

[0106]

[0107] Where t represents the current iteration round, and j = 1, 2, ..., d. Iter represents the j-th dimension value of the i-th sparrow in the t-th iteration. max ​α is the maximum number of iterations, and α∈(0,1) is a random number. Q is a random number that follows a normal distribution. L represents a 1×d matrix where each element is 1.

[0108] In step 7.3, followers will frequently monitor the discoverer. When followers notice that the discoverer has found good food, they will leave their current location and go to the discoverer's location to compete for the food. After successfully stealing food, the followers will move to a new location. The formula for updating the follower's location is described below:

[0109]

[0110] X P It is the optimal position occupied by the discoverer. X worst Let A represent the current global position farthest from food, where A is a 1×d matrix and each element is randomly assigned either 1 or -1. + =A T (AA T ) -1 When i > n / 2, it indicates that the i-th follower with the poorest fitness value is most likely to be in a state of starvation. This can be further simplified to:

[0111]

[0112] Step 7.4: The initial positions of alert sparrows are randomly generated within the population. The mathematical model for the initial positions of alert sparrows can be expressed as:

[0113]

[0114] Where X best It represents the global optimal position, β represents the step size control parameter, which follows a random normal distribution with a mean of 0 and a variance of 1, and K∈[-1,1] is a random number, f i This is the current suitability value for sparrows, f g and f w These are the current best and worst fitness values, respectively, with ε being a minimum constant to avoid division by zero error.

[0115] For simplicity, when f i >f g This indicates that the sparrow is located on the edge of the group. X best It represents the central location; the area around it is safe. i =f g This indicates that the sparrow in the middle of the population is aware of the danger and needs to move closer to the other sparrows. K represents the direction of the sparrow's movement and is also the step size control coefficient.

[0116] Step 7.5: Substitute the latest positions of each sparrow into the following formula:

[0117]

[0118] Recalculate the fitness values ​​of all sparrows, compare the obtained fitness values ​​with the original optimal values, update the optimal and worst values ​​of the current sparrow population, and update the global optimal and worst values.

[0119] Step 7.6: Determine if the current iteration count is less than the set maximum iteration count. If it is less, increment the iteration count by one and continue repeating steps 7.2 to 7.5. Otherwise, stop the optimization process, output the optimal sparrow position and fitness value, and thus obtain the optimal power allocation result and the maximum system and speed.

[0120] Example 3

[0121] A NOMA-VLC power allocation method based on the sparrow search algorithm is the same as in Examples 1-2. In this example, step 9 calculates the system and rate, as well as the user fairness index, to evaluate the allocation results. This includes the following steps:

[0122] Step 9.1, calculate the user's channel gain h. i Based on steps 7 and 8, the user's allocated power under the three algorithms can be obtained.

[0123] Step 9.2, calculate the system sum rate and user fairness index: Based on the user power allocation under the three algorithms, the system sum rate can be expressed as:

[0124]

[0125] Step 9.3 introduces the Jain fairness index J to evaluate the fairness of the system obtained by the algorithm. The fairness index is generally used to represent the fairness of the system, taking a continuous value between 0 and 1. The closer it is to 1, the fairer the resource allocation.

[0126]

[0127] Example 4

[0128] A NOMA-VLC power allocation method based on the sparrow search algorithm is the same as in Examples 1-3. In this example, step 10, which calculates the system, rate, and user fairness index under different conditions, includes the following steps:

[0129] Step 10.1 calculates the system sum rate under different numbers of users: Based on different maximum transmit power and user numbers, Monte Carlo simulations are used to evaluate the sum rate and fairness performance of the optimized system, and the results are compared with those of the Fixed Power Allocation (FPA) algorithm and the Gain Ratio Allocation (GRPA) algorithm. Specifically, by changing the number of users, the relationship between the sum rate of different algorithms and the number of users is compared, such as... Figure 5 As shown.

[0130] Step 10.2, calculate the user fairness index under different user numbers: change the number of users and compare the relationship between the user fairness index of different algorithms and the number of users, such as... Figure 6 As shown.

[0131] Step 10.3: Calculate the system and rate under different maximum transmit powers: Determine the number of users, change the transmit power, obtain the power allocation factors for the three algorithms based on steps 7 and 8, calculate the system and rate, and compare the relationship between the transmit power and rate of different algorithms, such as... Figure 7 As shown.

[0132] Based on the calculated system and rate, and user fairness index under different conditions, the sparrow search algorithm proposed in this invention supports arbitrary scenario-related parameters, arbitrary transmitter / receiver-related parameters, and arbitrary number of users. Simulation results and analysis provide guidance for multi-user power allocation.

[0133] The invention is further illustrated below with a more complete and specific example.

[0134] Example 5

[0135] A NOMA-VLC power allocation method based on the sparrow search algorithm, similar to Examples 1-4, includes the following steps:

[0136] Step 1: Determine the relevant parameters of the transmitter and the user's receiving PD, namely the coordinates of the LED (2.5, 2.5, 3), the vertical height of the transmitter above the ground L = 3m, the users are evenly distributed near the transmitter, and the Euclidean distance r between the location of user i and the projection position of the LED on the receiving plane. i It can be obtained through Monte Carlo simulation, r i The probability density function can be given as: Line-of-sight link distance between the transmitter and the i-th user The radius of the illuminated area is D = 2.5m.

[0137] Step 2: Determine the link type to be simulated, namely line-of-sight (LOS) link.

[0138] Step 3: Determine the relevant parameters in the system model, namely the DC bias. Peak light intensity Maximum transmission power p max =12mV, receiving surface area A=1cm 2 The receiver's field of view φ 1 / 2 =60deg, receiver optical filter gain T s (φ)=1.

[0139] Step 4: Determine the objective function and constraints. Apply the logarithmic utility function to the user's rate and sum the results to obtain the objective function. The objective function and constraints can be expressed as:

[0140]

[0141]

[0142]

[0143] p i ≥0

[0144] Where C = min{I DC BI DC}

[0145] Step 5: Determine two power allocation strategies to compare with the sparrow search algorithm: fixed power allocation algorithm and gain ratio power allocation algorithm.

[0146] Step 6: Determine the parameters of the sparrow search algorithm, setting the maximum number of iterations to iter max =300, total number of sparrows n=100, ratio of discoverers, followers and vigilants is set to 0.7:0.2:0.1, initial number of iterations is set to t=0.

[0147] Step 7: Calculate the user allocation power in the sparrow search algorithm: Find the optimal value of the objective function in step 4 using the sparrow search algorithm.

[0148] Step 8: Determine the power allocation factors α1 and α2 for the fixed power allocation algorithm and the gain ratio power allocation algorithm, and the power allocation factor for the fixed power allocation algorithm. Set between 0 and 1, the gain is the power allocation factor of the power allocation algorithm. It is related to the user channel gain.

[0149] After calculating the user allocation power in the NOMA-VLC system, in order to study the effectiveness of the sparrow search algorithm, it is also necessary to calculate the system and rate of the three algorithms and the user fairness index.

[0150] The following is an example of an engineering application: using the allocated power of NOMA-VLC users calculated by this invention, the system, rate, and user fairness index are calculated and analyzed under different user numbers and different maximum transmit power conditions.

[0151] Example 6

[0152] A NOMA-VLC power allocation method based on the sparrow search algorithm, similar to Examples 1-5, includes the following steps:

[0153] Set simulation parameters: The system simulation parameters are the same as in Example 5.

[0154] Simulation content: Calculate the system sum rate and user fairness index under different user numbers and maximum transmit power conditions. By changing the user number, compare the sum rate of different algorithms with the user number. Further, by changing the user number, compare the Jain fairness index of different algorithms with the user number. With the user number fixed, change the transmit power to obtain the power allocation factors for the three algorithms, calculate the system sum rate, and compare the transmit power with the sum rate of different algorithms.

[0155] Simulation Result Analysis: See simulation results. Figures 5-7 Specifically, the Sparrow Search algorithm calculates the following: When there are 2 users, the calculated sum rate is 2.0590 bps / Hz, and the user fairness index is 0.9495. When there are 3 users, the calculated sum rate is 2.0838 bps / Hz, and the user fairness index is 0.93759. When there are 4 users, the calculated sum rate is 2.0964 bps / Hz, and the user fairness index is 0.93425. When there are 5 users, the calculated sum rate is 2.0991 bps / Hz, and the user fairness index is 0.93179. When there are 6 users, the calculated sum rate is 2.1095 bps / Hz, and the user fairness index is 0.93365. When there are 7 users, the calculated sum rate is 2.1133 bps / Hz, and the user fairness index is 0.93305.

[0156] Fixed power allocation algorithm: Specifically, when the number of users is 2, the calculated sum rate is 1.9906 bps / Hz, and the user fairness index is 0.9756. When the number of users is 3, the calculated sum rate is 1.8851 bps / Hz, and the user fairness index is 0.91817. When the number of users is 4, the calculated sum rate is 1.7604 bps / Hz, and the user fairness index is 0.79689. When the number of users is 5, the calculated sum rate is 1.6275 bps / Hz, and the user fairness index is 0.67392. When the number of users is 6, the calculated sum rate is 1.5327 bps / Hz, and the user fairness index is 0.56694. When the number of users is 7, the calculated sum rate is 1.4332 bps / Hz, and the user fairness index is 0.48609. Gain-to-power allocation algorithm: Specifically, when the number of users is 2, the calculated sum rate is 1.9817 bps / Hz, and the user fairness index is 0.9472. When the number of users is 3, the calculated sum rate is 1.8487 bps / Hz, and the user fairness index is 0.88577. When the number of users is 4, the calculated sum rate is 1.7032 bps / Hz, and the user fairness index is 0.75328. When the number of users is 5, the calculated sum rate is 1.5582 bps / Hz, and the user fairness index is 0.61841. When the number of users is 6, the calculated sum rate is 1.4711 bps / Hz, and the user fairness index is 0.52221. When the number of users is 7, the calculated sum rate is 1.3773 bps / Hz, and the user fairness index is 0.4449. With 7 users and a maximum transmit power of 12mV, the sum rate obtained using the Sparrow Algorithm is 40.45% and 53.44% higher than that obtained using the FPA and GRPA algorithms, respectively.

[0157] In summary, this invention is a NOMA-VLC power allocation method based on the Sparrow Search algorithm. It can be used to solve the problem that existing power allocation algorithms cannot simultaneously guarantee user fairness and system rate. The method includes: determining the receiver PD-related parameters of the transmitter and users; determining the link type for simulation; determining the relevant parameters in the system model; determining the objective function and constraints; determining two power allocation strategies for comparison with the Sparrow Search algorithm; determining the parameters of the Sparrow Search algorithm; calculating the allocated power to users in the Sparrow Search algorithm; determining the power allocation factors α1 and α2 for the fixed power allocation algorithm and the gain ratio power allocation algorithm; calculating the system rate and user fairness; calculating the relationship between the rate and the number of users for different algorithms; calculating the Jain fairness index of the algorithm and the number of users; and calculating the relationship between the transmit power and the rate of the algorithm.

[0158] This invention, based on a NOMA-VLC downlink system with a single transmitter and multiple users, establishes a mathematical model that maximizes system and data rate based on user fairness, thereby improving both system and data rate performance and user fairness. Considering the non-negativity of VLC signals and eye safety, this invention sets optical power constraints and proposes a NOMA power allocation strategy based on a sparrow search algorithm, improving search accuracy, convergence speed, and stability. The proposed sparrow search algorithm supports arbitrary scene-related parameters, arbitrary transmitter / receiver-related parameters, and arbitrary number of users. Simulation results and analysis provide guidance for multi-user power allocation.

Claims

1. A NOMA-VLC power allocation method based on sparrow search algorithm, characterized in that, Includes the following steps: Step 1, determining the PD related parameters of the transmitter and the user, including: the coordinates of the transmitter, the vertical height of the transmitter and the ground , the Euclidean distance between the position of the first user and the position of the transmitter in the projection position of the receiving plane , the distance of the line-of-sight link between the transmitter and the first user , and the radius of the illumination area ; Users are evenly distributed around the transmitter. The probability density function obtained through Monte Carlo simulation is expressed as follows: Represented as: in r This represents the distance from the user to the center of the circle; Step 2: Use line-of-sight links as the link type for simulation; Step 3: Establish a mathematical model based on maximizing user fairness and system speed, and determine relevant parameters in the model, including: DC bias. Peak light intensity Maximum transmission power Effective area of ​​the receiver Receiver field of view and the optical filter gain of the receiver The established mathematical model includes: No. Each user employs serial interference cancellation technology, with a lower bound on its rate. The sum rate of the system is expressed as: in, For the user's channel gain, Power allocated to different users, and , N For the number of users, For the first The variance of the additive Gaussian noise in the channel between each user and the transmitter; Step 4: Determine the objective function and constraints. Apply the logarithmic utility function to the user's rate and sum the results to obtain the objective function. The objective function and constraints are expressed as follows: in ; Step 5: Using the objective function as the fitness function of the sparrow search algorithm, determine the parameters of the sparrow search algorithm, including: maximum number of iterations, total number of sparrows, ratio of discoverers, followers and vigilants, current iteration number and the position of all sparrows, where the position of each sparrow represents the allocated power of each user in the NOMA system; Step 6: Find the optimal sparrow's position and fitness value using the sparrow search algorithm, thereby obtaining the optimal power allocation result and the maximum system and rate.

2. The NOMA-VLC power allocation method based on the sparrow search algorithm according to claim 1, characterized in that, The user's channel gain is: in, This indicates the distance between the transmitter and the receiver. FOV Indicates the receiver's field of view. This represents the angle between the emitted ray and the transmitter's normal vector. This represents the angle between the incident ray and the receiver's normal vector. The Lambert emission coefficient is expressed by the formula. Calculations show that Take half-power point half angle, The gain of the condenser is expressed in the following form: in, It represents the refractive index.

3. The NOMA-VLC power allocation method based on the sparrow search algorithm according to claim 1, characterized in that, Step 6 involves calculating the user's allocated power, as follows: Step 6.1, using Let the matrix represent the positions of the sparrows, where each element represents a possible solution to the objective problem. We then obtain: in, This represents the dimension of the parameter to be optimized, and the elements in the matrix represent the values ​​of each parameter to be optimized. Indicates the first n The first sparrow d One parameter to be optimized; Step 6.2, the fitness value of each sparrow is represented as: Each row The value represents the fitness value of each individual, that is, the obtained system and velocity, respectively represented by... and The alarm value and safety threshold are represented, and the location update method for the discoverer is expressed as follows: in, , Indicates the first Only sparrows in the first The first iteration Dimensional value, It is the maximum number of iterations. It is a random number. It is a random number that follows a normal distribution. Represent a A matrix where each element is 1; Step 6.3, the follower position update formula is as follows: It is the optimal position occupied by the discoverer. This indicates the current global position furthest from the food. A Represent a A matrix, wherein each element is randomly assigned either 1 or -1. ; Step 6.4, the initial position of the alert person is represented as: in It is the globally optimal position. This represents the step size control parameter, which follows a random normal distribution with a mean of 0 and a variance of 1. It is a random number. This is the current suitability value for sparrows. and These are the current best and worst fitness values, respectively. It is the smallest constant to avoid division by zero error; Step 6.5, based on the latest positions of each sparrow after the update, substitute into the following formula: Recalculate the fitness values ​​of all sparrows, compare the obtained fitness values ​​with the original optimal values, update the optimal and worst values ​​of the current sparrow population, and update the global optimal and worst values. Step 6.6: Determine if the current iteration count is less than the set maximum iteration count. If it is less, increment the iteration count by one and continue repeating steps 6.2 to 6.

5. Otherwise, stop the optimization process, output the optimal sparrow position and fitness value, and thus obtain the optimal power allocation result and the maximum system and speed.

4. The NOMA-VLC power allocation method based on the sparrow search algorithm according to claim 3, characterized in that, In step 6.3, when When, it indicates that the fitness value is poor. If a follower is most likely to be in a state of hunger, then: 。 5. The NOMA-VLC power allocation method based on the sparrow search algorithm according to claim 1, characterized in that, The system and speed, as well as the user fairness index, are calculated to evaluate the allocation results. The calculation method is as follows: Step 7.1, calculate the user's channel gain. ; Step 7.2, calculate the system and rate. ; Step 7.3, introduce the Jain Fairness Index To represent the fairness of the system, a continuous value between 0 and 1 is used. The closer the value is to 1, the fairer the resource allocation. The calculation is as follows: 。 6. A computer-readable storage medium, characterized in that, Used to store a computer program that causes a computer to perform the method as described in any one of claims 1-5.

7. A computer program product containing instructions, characterized in that, When the instructions are executed on a computer, the computer causes the computer to perform the method of any one of claims 1-5.

Images