An intelligent energy consumption control method for satellite IoT communications in remote areas

Abstract

The present invention relates to the field of satellite Internet of Things technology, and in particular to an energy consumption intelligent control method for satellite Internet of Things communications in remote areas, the method specifically comprising: determining energy consumption estimates and delay estimates for Internet of Things communications between user equipment, each edge server and each satellite in different types of remote areas according to location coordinate information and computational load information; determining a predicted moving trajectory of the user equipment, minimizing the energy consumption estimate and the delay estimate as an optimization target, combining the predicted moving trajectory of the user equipment, and using a multi-objective particle swarm optimization algorithm to dynamically select an optimal edge server and an optimal satellite for the user equipment; based on the predicted moving trajectory of the user equipment, dynamically adjusting the beam pointing parameters, transmission power parameters and beamforming weight parameters of the optimal satellite. The present invention effectively improves the energy utilization efficiency of satellite Internet of Things communications in remote areas and optimizes energy management.

Description

Technical Field

[0001] The present invention relates to the technical field of satellite Internet of Things, and in particular to an intelligent energy consumption control method for satellite Internet of Things communications in remote areas. Background Art

[0002] With the rapid advancement of Internet of Things (IoT) technology, the demand for IoT communications in remote areas has exploded. However, these areas often face severe challenges such as weak communication infrastructure and insufficient signal coverage due to their remote locations and harsh natural environments. These problems not only lead to high energy consumption in IoT communications, but also bring significant latency issues, which seriously restrict the widespread deployment and application of IoT in these areas.

[0003] In existing technologies, IoT devices mainly rely on satellites or ground base stations for communication. However, in remote areas, especially those with complex terrain and harsh climatic conditions, this communication method seems to be inadequate. Although satellite communication has a wide coverage, the signal will be interfered by many factors during the transmission process, such as the atmosphere, clouds, terrain, etc., resulting in severe signal attenuation, increased energy consumption, and increased latency. Although the ground base station communication signal is stable, the construction cost is high, and due to geographical conditions, it is difficult to cover all remote areas.

[0004] In addition, most existing communication optimization solutions only focus on optimizing a single indicator, such as reducing communication energy consumption or delay. However, in remote areas, the complexity of IoT communication is far more than that. Multiple factors such as the mobility of user devices, the dynamic changes of the communication environment, and the limited communication resources are intertwined and jointly affect the performance of communication. Therefore, it is difficult to meet the actual needs of IoT communication in remote areas by simply optimizing a single indicator.

[0005] To make matters more complicated, different remote areas have their own unique environmental parameters. For example, in ocean areas, the attenuation effect of seawater on signals is very significant, which requires communication equipment to have stronger signal penetration and anti-attenuation capabilities. In desert areas, the temperature is extremely high during the day, which may cause equipment to overheat and damage, and special heat dissipation measures need to be taken. In high-altitude areas, low air pressure makes it difficult for equipment to dissipate heat, which also requires special heat dissipation design. These unique environmental parameters not only increase the complexity of communication, but also make it more difficult to optimize communication energy consumption. Summary of the invention

[0006] The purpose of the present invention is to provide an intelligent energy consumption control method for satellite Internet of Things communications in remote areas, which effectively improves the energy utilization efficiency of satellite Internet of Things communications in remote areas, reduces the data transmission delay of satellite Internet of Things communications in remote areas, and optimizes energy management to solve at least one of the above-mentioned prior art problems.

[0007] In a first aspect, the present invention provides an intelligent energy consumption control method for satellite Internet of Things communications in remote areas, the method specifically comprising:

[0008] Determine estimated energy consumption and estimated latency for IoT communications between user devices, edge servers, and satellites in different types of remote areas based on location coordinate information and computational load information;

[0009] Determine a predicted movement trajectory of the user equipment, take minimizing the estimated energy consumption value and the estimated delay value as an optimization goal, and use a multi-objective particle swarm optimization algorithm to dynamically select an optimal edge server and an optimal satellite for the user equipment in combination with the predicted movement trajectory of the user equipment;

[0010] Based on the IoT communication relationship between the user equipment, the optimal edge server and the optimal satellite, and according to the channel environment parameters of the remote area where the user equipment is currently located, an adaptive coding modulation technology is used to establish a convex optimization model between coding energy consumption and data compression ratio through a convex optimization method, and the convex optimization model is used to determine the optimal coding parameter configuration for data coding in the IoT communication relationship;

[0011] Dynamically adjusting the optimal satellite beam pointing parameters, transmission power parameters and beamforming weight parameters based on the predicted movement trajectory of the user equipment;

[0012] The user communication behavior of the user device is predicted by a long short-term memory neural network algorithm to determine the predicted request data, and each data content in the predicted request data is cached in advance to the optimal edge server according to the priority sorting relationship by an information entropy algorithm.

[0013] In a second aspect, the present invention provides a computer device comprising: a memory and a processor and a computer program stored in the memory. When the computer program is executed on the processor, it implements an intelligent energy consumption control method for satellite Internet of Things communications in remote areas as described in any of the above methods.

[0014] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for intelligent energy consumption control for satellite Internet of Things communications in remote areas as described in any of the above methods is implemented.

[0015] Compared with the prior art, the present invention has at least one of the following technical effects:

[0016] 1. The present invention effectively improves the energy utilization efficiency of satellite Internet of Things communications in remote areas, reduces the data transmission delay of satellite Internet of Things communications in remote areas, optimizes energy management, and enhances the sustainability of the system and the user experience.

[0017] 2. Aiming at the special channel environment in remote areas, the present invention adopts adaptive coding modulation technology, and uses a convex optimization method to balance the coding energy consumption and data compression ratio, and determines the optimal coding parameter configuration, so as to optimize energy consumption while ensuring communication quality.

[0018] 3. The present invention predicts user communication behavior through a long short-term memory neural network algorithm, and caches the data that may be requested in the edge server in advance. It measures the importance of data through information entropy, realizes the priority sorting of data cache, reduces the communication delay of the satellite Internet of Things, and improves the user experience.

[0019] 4. The present invention can dynamically adjust the beam pointing parameters, transmission power parameters and beamforming weight parameters of the optimal satellite to adapt to the dynamic changes of the communication environment and ensure the stability and reliability of communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 It is a flow chart of an energy consumption intelligent control method for satellite Internet of Things communications in remote areas provided by one embodiment of the present invention;

[0022] Figure 2 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0024] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0025] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0026] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0027] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0028] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0029] In the embodiment of the present application, the execution subject of the process includes a terminal device, which includes but is not limited to: a server, a computer, a smart phone, a tablet computer, and other devices capable of executing the method disclosed in the present application. Figure 1 A flow chart of an energy consumption intelligent control method for satellite IoT communication in remote areas disclosed in one embodiment of the present invention is shown, and the details are as follows:

[0030] S101, determining an estimated energy consumption and a delay for IoT communications between a user device, each edge server, and each satellite in different types of remote areas based on location coordinate information and computational load information.

[0031] In this embodiment, it is first necessary to collect the location coordinate information of the user device, the calculation load information, and the geographical location and performance parameters of each edge server and satellite, and understand the special environmental parameters of each remote area, such as the height and slope of the mountain, the temperature and humidity of the desert, and the salinity and depth of the ocean. Based on the collected data, an energy consumption model and a delay model for IoT communication between the user device, the edge server and the satellite in different types of remote areas are established. These models will consider multiple factors, such as communication distance, signal attenuation, equipment performance, environmental parameters, etc. For example, in mountainous areas, due to the undulating terrain, the signal transmission path may become longer, resulting in increased energy consumption and increased delay. The relevant parameters in the model can be adjusted according to the height and slope of the mountain to more accurately estimate the energy consumption and delay. In desert areas, high temperatures may make it difficult for the device to dissipate heat, thereby affecting its performance. When establishing the model, it is necessary to consider the impact of the temperature in the desert area on the performance of the device, and adjust the estimated values ​​of energy consumption and delay accordingly. In marine areas, the attenuation effect of seawater on the signal is serious. In order to accurately estimate the energy consumption and delay, it is necessary to adjust the signal attenuation coefficient in the model according to parameters such as the salinity and depth of seawater. Using the established model, the energy consumption estimate and delay estimate of IoT communications between user devices, edge servers and satellites in different types of remote areas can be calculated based on the location coordinate information and computing load information of the user device.

[0032] In this embodiment, by comprehensively considering multiple factors to establish an energy consumption model and a delay model, it is possible to more accurately estimate the energy consumption and delay of user devices performing IoT communications in different remote areas, providing strong data support for subsequent optimization decisions.

[0033] In some embodiments, the above step S101 specifically includes:

[0034] According to the location coordinate information and computing load information of each user device, the K-means clustering algorithm is used to cluster each user device to obtain the clustering results of user devices in different types of remote areas;

[0035] According to the position coordinate information and calculation load information of each edge server and each satellite and the user device clustering result, the shortest distance between the current user device and each edge server, and between each edge server and each satellite is calculated by the Dijkstra algorithm;

[0036] According to the shortest distance, the communication rate between the current user device and each edge server, and between each edge server and each satellite is calculated using the Shannon information theory formula;

[0037] According to the communication rate and the computing load information of the current user equipment, each edge server, and each satellite, the estimated delay between the current user equipment and each edge server, and between each edge server and each satellite is calculated through a queuing theory model;

[0038] According to the communication rate, the communication distance between the current user equipment and each edge server, and the communication distance between each edge server and each satellite, the energy consumption estimation value between the current user equipment and each edge server, and between each edge server and each satellite is calculated using a wireless communication energy consumption model;

[0039] Obtain a first comparison result by comparing the estimated delay value and the estimated energy consumption value between the current user equipment and any edge server with their respective preset thresholds;

[0040] Obtain a second comparison result by comparing the estimated delay value and the estimated energy consumption value between each edge server and any satellite with their respective preset thresholds;

[0041] Based on the first comparison result and the second comparison result, a scheme for performing Internet of Things communication between the current user equipment, each edge server and each satellite in different types of remote areas is determined.

[0042] In this embodiment, the geographical location distribution of the user equipment is obtained according to the location coordinate information of each user equipment, and the distribution characteristics of the user equipment in different remote areas are determined by analyzing the degree of aggregation of the location coordinates. The computing resource usage of the user equipment is obtained according to the computing load information of each user equipment, and the computing power difference of different user equipment is determined by analyzing the level of computing load. The K-means clustering algorithm is adopted, and the location coordinates and computing load of the user equipment are used as clustering features to cluster each user equipment. The clustering results of the user equipment in different types of remote areas are obtained by iteratively optimizing the clustering center. According to the clustering results, the location distribution and computing load characteristics of the user equipment in each cluster cluster are obtained, and the service demand differences of the user equipment in different remote areas are determined by analyzing the characteristics of the cluster cluster. A targeted resource allocation strategy is adopted, and differentiated service plans are formulated according to the characteristics of the user equipment in different cluster clusters. By optimizing resource allocation, the service quality and user experience of the user equipment in remote areas are improved. According to the implementation effect of the service plan, the location changes and computing load changes of the user equipment are continuously obtained, and the clustering algorithm parameters are dynamically adjusted to continuously optimize the user equipment clustering results and service plans to ensure the accuracy and adaptability of the service. By real-time monitoring of the service quality indicators of user equipment, we can obtain feedback information on service improvement. Based on the feedback results, we can continuously improve the clustering algorithm and resource allocation strategy, continuously improve the service level of user equipment in remote areas, and obtain higher user satisfaction.

[0043] According to the position coordinate information and computational load information of each edge server and each satellite and the user device clustering result, the shortest distance between the current user device and each edge server is calculated by Dijkstra algorithm, and the shortest distance between each edge server and each satellite is calculated at the same time, so as to obtain the shortest distance data between the current user device and each edge server and between each edge server and each satellite. The Shannon information theory formula is adopted, and the obtained shortest distance data between each edge server and each satellite is used as input to calculate the communication rate between each edge server and each satellite, and obtain the communication rate data between each edge server and each satellite. According to the communication rate data between the current user device and each edge server and the communication rate data between each edge server and each satellite, combined with the computational load information of each edge server and each satellite, through comprehensive analysis, it is determined which edge server the current user device should communicate with and which satellite the edge server should communicate with, so as to achieve the best communication effect and computational efficiency.

[0044] For example, first, according to the location coordinate information and computational load information of each user device, the K-means clustering algorithm is used to cluster each user device, and the user devices are divided into different types of remote areas such as high altitude, mountainous areas, and islands, where the K value can be set to 5 according to actual conditions. Then, according to the location coordinate information and computational load information of each edge server and each satellite and the user device clustering result, the Dijkstra algorithm is used to calculate the shortest distance between the current user device and each edge server, and between each edge server and each satellite. Then, based on the shortest distance, the Shannon information theory formula is used to calculate the communication rate between the current user device and each edge server, and between each edge server and each satellite, where the Shannon formula is C=Blog2(1+S / N), where C is the channel capacity, B is the channel bandwidth, S is the signal power, and N is the noise power. Assuming that the calculated communication rates are 100Mbps and 10Mbps respectively, then based on the communication rate and the computing load information of the current user device, each edge server, and each satellite, the queuing theory model is used to calculate the estimated delay between the current user device and each edge server, and between each edge server and each satellite. Assuming that the arrival rate is 50Mbps and the service rate is 100Mbps, the estimated delay value is 5 seconds. At the same time, according to the communication rate, the communication distance between the current user device and each edge server, and the communication distance between each edge server and each satellite, the wireless communication energy consumption model is used to calculate the estimated energy consumption between the current user device and each edge server, and between each edge server and each satellite, wherein the energy consumption model is P=αd^2+β, α and β are energy consumption parameters, d is the communication distance, and it is assumed that the calculated estimated energy consumption values ​​are 5 watts and 10 watts respectively. Finally, by comparing the relationship between the estimated delay value and the estimated energy consumption value between the current user device and any edge server and their respective preset thresholds, a first comparison result is obtained, and then by comparing the relationship between the estimated delay value and the estimated energy consumption value between each edge server and any satellite and their respective preset thresholds, a second comparison result is obtained. Based on the first comparison result and the second comparison result, a scheme for IoT communication between the current user device, each edge server and each satellite in different types of remote areas is determined.

[0045] S102, determining a predicted moving trajectory of the user equipment, taking minimizing the energy consumption estimate and the delay estimate as an optimization goal, combining the predicted moving trajectory of the user equipment, and using a multi-objective particle swarm optimization algorithm to dynamically select an optimal edge server and an optimal satellite for the user equipment.

[0046] In this embodiment, it is necessary to use the historical mobile data of the user device, combined with machine learning or deep learning algorithms (such as long short-term memory network LSTM, recurrent neural network RNN, etc.), to predict the movement trajectory of the user device in the future. Then determine the optimization goal, which is to minimize the estimated energy consumption and delay of the user device when communicating with the edge server and satellite at different locations. These two indicators will serve as key criteria for evaluating communication quality. Next, the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm is used to dynamically select the optimal edge server and satellite for the user device. The MOPSO algorithm is an optimization algorithm based on swarm intelligence, which can balance multiple objectives and find a set of optimal solutions.

[0047] In the MOPSO algorithm, each particle represents a possible solution, that is, a combination of a user device, an edge server, and a satellite. The fitness value of the particle is determined by the estimated energy consumption and the estimated delay. The algorithm updates the position and speed of the particle through iteration, so that the particle gradually approaches the optimal solution set. During the iteration process, the predicted movement trajectory of the user device needs to be considered. Specifically, the predicted trajectory can be divided into several time periods, and an optimal edge server and satellite combination is selected for each time period, so as to ensure that the communication energy consumption and delay of the user device are optimal throughout the predicted trajectory.

[0048] After iterative optimization of the MOPSO algorithm, a set of optimal solutions will be obtained, that is, the optimal combination of user devices at different locations with edge servers and satellites. These optimal combinations will serve as the basis for dynamically selecting edge servers and satellites for user devices.

[0049] In this embodiment, by combining the predicted movement trajectory of the user device, a multi-objective particle swarm optimization algorithm is used to dynamically select the optimal edge server and satellite for the user device, which can significantly reduce the energy consumption and delay of communication, and help improve the efficiency and reliability of IoT communication. Dynamically selecting the optimal edge server and satellite can ensure the reasonable allocation and efficient use of communication resources, help avoid resource waste and congestion, and improve the performance of the entire IoT communication network.

[0050] In some embodiments, in the above step S102, determining the predicted movement trajectory of the user equipment specifically includes:

[0051] Determine the current position state parameters of the user equipment according to the position coordinate information of the user equipment, wherein the current state parameters include the current position coordinates, speed and acceleration;

[0052] Taking the current position state parameter as an input value, using the prediction equation of the Kalman filter algorithm to perform prediction, and obtaining a predicted value of the user equipment at the next moment, the predicted value including a predicted position coordinate, a predicted speed, and a predicted acceleration;

[0053] Acquire a measurement value of the user equipment at a next moment, the measurement value including a measurement position coordinate, a measurement speed, and a measurement acceleration, and obtain an optimal estimate value of the user equipment at the next moment by substituting the predicted value and the measurement value into an update equation of a Kalman filter algorithm for correction;

[0054] Using the optimal estimated value as a new input value, performing iterative calculation using a Kalman filter algorithm, and updating the optimal estimated value in real time;

[0055] Based on the optimal estimate, the prediction equation of the Kalman filter algorithm is recursively extrapolated for t time steps to obtain a predicted movement trajectory of the user device from the current moment to a time period of t in the future.

[0056] In this embodiment, the current position coordinates of the user equipment are (1212, 328), the speed is 5m / s, and the acceleration is 2m / s. 2 , these state parameters are used as the input values ​​of the Kalman filter algorithm, and the predicted position coordinates, predicted speed and predicted acceleration of the next moment are calculated through the prediction equation. Assume that the predicted position coordinates are (1213,329), the predicted speed is 6m / s, and the predicted acceleration is 22m / s 2 At the same time, the actual measurement value of the user device at the next moment is obtained. The measured position coordinates are (1214,330), the measured speed is 55m / s, and the measured acceleration is 21m / s 2 . Substitute the predicted value and the measured value into the update equation of the Kalman filter algorithm for correction, and obtain the optimal estimate at the next moment. The position coordinates are (12135,3295), the speed is 58m / s, and the acceleration is 215m / s2. The latest optimal estimate is continuously used as input, iterative calculation is performed, and the position state estimate of the user device is updated in real time. Based on the optimal estimate, the prediction equation of the Kalman filter algorithm is recursively extended for 10 time steps, each step is 1 second, to obtain the predicted movement trajectory of the user device in the next 10 seconds.

[0057] In some embodiments, in the above step S102, minimizing the estimated energy consumption value and the estimated delay value as the optimization goal, combining the predicted movement trajectory of the user equipment, and using a multi-objective particle swarm optimization algorithm to dynamically select the optimal edge server and the optimal satellite for the user equipment specifically includes:

[0058] The predicted movement trajectory of the user equipment, all the estimated energy consumption values ​​and the estimated delay values ​​are used as inputs of a multi-objective particle swarm optimization algorithm, and a fitness function is constructed with minimizing the estimated energy consumption value and the estimated delay value as the optimization target;

[0059] A multi-objective particle swarm optimization algorithm is used to solve the task offloading decision problem, and the particle position and velocity are updated through iterative search, the fitness function value of the fitness function is continuously optimized, and the Pareto optimal solution set is obtained;

[0060] Select a task offloading decision scheme whose energy consumption estimate and delay estimate are both less than their respective preset thresholds from the Pareto optimal solution set, determine the offloading scheme of the user task initiated by each user device on the edge server and the satellite, and obtain the task offloading decision result;

[0061] According to the task offloading decision results, the user tasks initiated by each user device are offloaded to the corresponding optimal edge server and optimal satellite for execution;

[0062] The movement trajectory data of the user equipment, the energy consumption data of the edge server and the satellite, and the task execution delay data are continuously collected to dynamically adjust and optimize the parameters of the multi-objective particle swarm optimization algorithm.

[0063] In this embodiment, the predicted mobile trajectory, the estimated energy consumption of the edge server and the satellite, and the estimated task execution delay are used as inputs of the multi-objective particle swarm optimization algorithm, and a fitness function with minimization of energy consumption and delay as the optimization goal is constructed. The particle position and speed are updated by iterative search, the fitness function value is optimized, and the Pareto optimal solution set is obtained. From the solution set, the task offloading decision scheme with energy consumption and delay less than the preset threshold (such as the energy consumption threshold is 1000mJ and the delay threshold is 500ms) is selected to determine the optimal offloading scheme for the user task on the edge server and the satellite. According to the offloading decision result, the user task is offloaded to the corresponding optimal edge server and satellite for execution. The user device mobile trajectory data, the energy consumption data of the edge server and the satellite, and the task execution delay data are continuously collected, and the parameters of the multi-objective particle swarm optimization algorithm are dynamically adjusted, such as the inertia weight is adjusted from 8 to 6, the acceleration factor is adjusted from 5 to 2, and the number of particles is adjusted from 50 to 80, so as to further optimize the performance of the task offloading decision.

[0064] S103, based on the IoT communication relationship between the user device, the optimal edge server and the optimal satellite, and according to the channel environment parameters of the remote area where the user device is currently located, an adaptive coding modulation technology is used to establish a convex optimization model between coding energy consumption and data compression ratio through a convex optimization method, and the convex optimization model is used to determine the optimal coding parameter configuration for data encoding in the IoT communication relationship.

[0065] In this embodiment, it is first necessary to obtain the channel environment parameters of the remote area where the user equipment is currently located. These parameters include but are not limited to signal-to-noise ratio (SNR), multipath effect, shadow fading, etc., and then select an adaptive coding and modulation technology to adapt to changes in the channel environment. The adaptive coding and modulation technology can dynamically adjust the coding and modulation methods according to the channel conditions to achieve the optimization of communication performance. Commonly used adaptive coding and modulation technologies include variable rate coding (VRC) and adaptive modulation and coding (AMC).

[0066] In order to determine the optimal coding parameter configuration, a convex optimization method is used to establish a convex optimization model between coding energy consumption and data compression ratio. The goal of this model is to minimize coding energy consumption and optimize data compression ratio while meeting certain communication quality requirements.

[0067] The general form of the convex optimization model can be expressed as:

[0068] minimize f(x)

[0069] subject to g i (x)≤0,i=1,2,...,m

[0070] h j (x)=0,j=1,2,...,n

[0071] Where f(x) is the objective function, which represents the weighted sum of encoding energy consumption and data compression ratio; g i (x) is an inequality constraint, which indicates the communication quality requirement; h j (x) is an equality constraint, which indicates some fixed conditions or restrictions on the coding parameters.

[0072] Furthermore, f(x) = αE(x) + βC(x), where α and β are weight coefficients that can be adjusted according to actual needs, E(x) is the encoding energy consumption, C(x) is the data compression ratio, and the inequality constraint g i (x) may include bit error rate (BER) requirements, signal-to-noise ratio requirements, etc. to ensure communication quality. The equality constraint h j (x) may include restrictions on fixed parameters such as coding rate and modulation method.

[0073] By solving the convex optimization model, we can obtain the optimal coding parameter configuration. This configuration is applied to the IoT communication relationship between the user device and the optimal edge server and satellite, which will achieve an optimal balance between coding energy consumption and data compression ratio, thereby meeting the communication quality requirements, reducing energy consumption and optimizing the data compression ratio.

[0074] In this embodiment, through adaptive coding modulation technology and convex optimization method, the coding parameter configuration can be dynamically adjusted according to the channel environment parameters, thereby optimizing the communication performance, helping to reduce communication energy consumption, increase data transmission rate and reduce bit error rate, thereby improving communication efficiency.

[0075] In some embodiments, in the above step S103, the step of using adaptive coding and modulation technology to establish a convex optimization model between coding energy consumption and data compression ratio by a convex optimization method according to the channel environment parameters of the remote area where the user equipment is currently located specifically includes:

[0076] Acquire an initial value of a channel environment parameter in a remote area where the user equipment is currently located according to an IoT communication relationship between the user equipment, an optimal edge server, and an optimal satellite;

[0077] An IoT communication link model is established according to the geometric position relationship between the user equipment and the optimal edge server and the optimal satellite, the initial value of the channel environment parameter is substituted into the IoT communication link model, and a channel estimation algorithm is used to calculate and obtain the channel environment parameter of the remote area where the user equipment is currently located;

[0078] According to the channel environment parameters, the coding mode and the modulation mode are determined by using the adaptive orthogonal frequency division multiplexing technology and the adaptive orthogonal amplitude modulation technology to obtain the optimal coding and modulation scheme;

[0079] By using the optimal coding modulation scheme, the relationship between coding energy consumption and data compression ratio is obtained, and a first convex optimization model with coding energy consumption and data compression ratio as optimization variables is established;

[0080] Taking minimizing encoding energy consumption and maximizing data compression ratio as optimization goals, and combining the signal environment parameters, constructing a convex optimization objective function and constraint conditions;

[0081] According to the first convex optimization model, the convex optimization objective function and the constraint condition, a target convex optimization model is formed;

[0082] The target convex optimization model is solved by the interior point method, and the value of the optimization variable is continuously updated by iterative optimization until the constraint conditions are met, thereby obtaining the optimal encoding energy consumption and data compression ratio.

[0083] In this embodiment, according to the IoT communication relationship between the user device, the optimal edge server and the optimal satellite, the initial value of the channel environment parameter of the remote area where the user device is currently located is obtained; the geometric position relationship between the user device and the optimal edge server and the optimal satellite is used to establish an IoT communication link model; the initial value of the channel environment parameter is substituted into the IoT communication link model, and the channel estimation algorithm is used for calculation to obtain the channel environment parameter of the remote area where the user device is currently located; according to the channel environment parameter of the remote area where the user device is currently located, the mean and variance of the channel environment parameter are obtained; the probability distribution model of the channel environment parameter is determined by the mean and variance; the maximum likelihood estimation method is used to estimate the channel environment parameter according to the probability distribution model to obtain the estimated channel environment parameter value; the estimated channel environment parameter value is compared with the actually measured channel environment parameter value to calculate the estimation error; it is determined whether the estimation error is less than the set threshold value, if so, the estimated channel environment parameter is used as the channel environment parameter of the remote area where the user device is currently located; otherwise, it returns to re-estimate until the estimation error meets the requirements; according to the channel environment parameter of the remote area where the user device is currently located, the key indicators such as the channel capacity and signal-to-noise ratio of the IoT communication link are obtained.

[0084] According to the channel environment parameters, such as signal-to-noise ratio, channel fading and other information, the status information of the current channel is obtained to judge the quality of the channel. Adaptive orthogonal frequency division multiplexing technology is used to select the appropriate number of subcarriers and the power allocation on each subcarrier according to the channel quality to obtain the optimal subcarrier allocation scheme. Adaptive orthogonal amplitude modulation technology is used to select the appropriate modulation order according to the channel quality, determine the modulation method on each subcarrier, and obtain the optimal modulation method. According to the selected modulation order, a suitable coding method is selected, such as convolutional code, Turbo code, etc., to obtain the optimal coding method. By combining the optimal subcarrier allocation scheme, modulation method and coding method, the optimal coding modulation scheme is obtained to meet the communication needs under different channel environments.

[0085] According to the target convex optimization model, the interior point method is used to solve, the initial encoding energy consumption and data compression ratio are obtained, and the value of the optimization variable is continuously updated by iterative optimization. According to the current encoding energy consumption and data compression ratio, the interior point method is used to calculate the objective function value and the constraint condition value to obtain whether the current solution meets the constraint conditions. If the current solution meets the constraint conditions, it is determined whether it reaches the optimal state. If it reaches the optimal state, the current solution is determined to be the optimal solution, and the optimal encoding energy consumption and data compression ratio are obtained. If the current solution does not meet the constraint conditions or does not reach the optimal state, the value of the optimization variable is updated through the iterative optimization algorithm of the interior point method to obtain a new encoding energy consumption and data compression ratio. The updated encoding energy consumption and data compression ratio are used until the constraints are met and the optimal state is reached. Finally, the optimal solution is determined to obtain the optimal encoding energy consumption and data compression ratio, and the efficient transmission of video data is completed.

[0086] S104: dynamically adjusting the optimal satellite beam pointing parameters, transmission power parameters and beamforming weight parameters based on the predicted movement trajectory of the user equipment.

[0087] In this embodiment, the beam pointing parameters of the satellite are dynamically adjusted according to the predicted movement trajectory of the user equipment to ensure that the beam can always cover the area where the user equipment is located. For example, if it is predicted that the user equipment will move to the north, the beam pointing parameters can be adjusted to deflect the beam to the north to cover the area that the user equipment is about to enter in advance.

[0088] The satellite's transmit power parameters are dynamically adjusted based on the user equipment's computational load information and the current channel's environmental parameters. When the user equipment's computational load is high or the channel environment is poor, the transmit power is appropriately increased to ensure communication quality; otherwise, the transmit power is reduced to save energy.

[0089] Beamforming is a technology that forms a beam of a specific shape by adjusting the weights of a satellite antenna array. The beamforming weight parameters are dynamically adjusted according to the predicted movement trajectory of the user device and the channel environment parameters to form a more precise beam shape, improve communication efficiency and reduce interference.

[0090] In this embodiment, by dynamically adjusting the beam pointing parameters, it is ensured that the beam always covers the area where the user equipment is located, reducing signal attenuation and transmission delay. By dynamically adjusting the transmit power parameters and beamforming weight parameters, signal transmission can be optimized according to actual communication requirements and environmental conditions to improve communication efficiency.

[0091] In some embodiments, the above step S104 specifically includes:

[0092] According to the predicted movement trajectory of the user equipment, a Newton-Raphson iteration method is used to solve and determine the optimal beam pointing angle of the optimal satellite;

[0093] According to the distance between the user equipment and the optimal satellite and the link budget result, the optimal transmission power of the optimal satellite is determined by using the gradient descent method;

[0094] Constructing an objective function and constraint conditions of a beamforming problem, and using a minimum mean square error criterion and a convex optimization algorithm to solve the objective function and the constraint conditions to obtain an optimal beamforming weight;

[0095] An optimal tracking and focusing scheme for the satellite beam of the satellite is determined according to the optimal beam pointing angle, the optimal transmission power and the optimal beamforming weight.

[0096] In this embodiment, in order to predict the movement trajectory of the user equipment, a Kalman filter algorithm can be used to predict the movement trajectory of the user equipment in the future period of time through the historical position information and speed information of the user equipment. For example, assuming that the historical position of the user equipment is (100, 200), (105, 210), (110, 220), and the speed is (5, 10), (5, 10), then it can be predicted that the position of the user equipment in the next 1 second is (115, 230). After determining the predicted movement trajectory of the user equipment, the Newton-Raphson iteration method is used to solve the optimal beam pointing angle of the satellite, and the beam pointing angle that maximizes the link signal-to-noise ratio between the satellite and the user equipment is found through iterative calculation as the optimal beam pointing angle. According to the distance between the user equipment and the optimal satellite and the link budget result, the gradient descent method is used to solve the optimal transmission power of the satellite, and the transmission power value that maximizes the link signal-to-noise ratio and meets the power constraint is found through iterative calculation. After determining the optimal beam pointing angle and optimal transmit power, the objective function and constraints of the beamforming problem are constructed, and the minimum mean square error criterion and convex optimization algorithm are used to solve the optimal beamforming weights, so that the mean square error between the beamforming pattern and the expected pattern is minimized. Finally, based on the optimal beam pointing angle, optimal transmit power and optimal beamforming weights obtained, the optimal solution for satellite beam tracking and focusing on user equipment is determined, so that the satellite can accurately track and efficiently serve user equipment.

[0097] Specifically, the predicting the moving trajectory of the user equipment and using the Newton-Raphson iteration method to solve and determine the optimal beam pointing angle of the optimal satellite specifically includes:

[0098] According to the orbital data of the optimal satellite and the predicted movement trajectory of the user equipment, an iterative equation of the satellite beam pointing angle is established by using the Newton-Raphson iteration method, and an initial value of the satellite beam pointing angle is obtained by setting an initial value of the iteration and an iteration termination condition;

[0099] According to the initial value of the satellite beam pointing angle and the predicted moving trajectory of the user equipment, the dot product formula of the beam direction vector and the position vector is used to calculate the deviation between the satellite beam center and the predicted trajectory of the user equipment to obtain the beam center deviation under the current satellite beam pointing angle;

[0100] Determine whether the beam center deviation satisfies a set iteration termination condition. If not, use the Newton-Raphson iteration formula to find the partial derivative of the beam pointing angle according to the beam center deviation to obtain a beam pointing angle correction value;

[0101] A new satellite beam pointing angle is determined according to the beam pointing angle correction amount, and the step of calculating the beam center deviation is repeated using the new satellite beam pointing angle, and the satellite beam pointing angle is continuously corrected until the beam center deviation meets the iteration termination condition, thereby obtaining the optimal satellite beam pointing angle.

[0102] In this embodiment, according to the satellite orbit parameters and the predicted position of the user equipment, the Newton-Raphson iteration method is used to establish the iterative equation of the satellite beam pointing angle, and the optimal initial value of the satellite beam pointing angle is obtained by setting the initial value of the iteration and the iteration termination condition, so as to obtain the initial satellite beam pointing angle. According to the initial value of the satellite beam pointing angle and the predicted position, the dot product formula of the beam direction vector and the position vector is used to calculate the deviation between the satellite beam center and the predicted position of the user equipment, to obtain the beam center deviation under the current satellite beam pointing angle, and to determine whether the deviation meets the set iteration termination condition. According to the beam center deviation, the Newton-Raphson iteration formula is used to obtain the partial derivative of the beam pointing angle by the deviation, to obtain the correction amount of the beam pointing angle, and to obtain the new satellite beam pointing angle. According to the new beam pointing angle, the above steps of calculating the beam center deviation are repeated, and the satellite beam pointing angle is continuously corrected by iterative calculation until the beam center deviation meets the iteration termination condition, to obtain the converged optimal satellite beam pointing angle, and to determine the optimal pointing angle for the satellite beam to align with the predicted position of the user equipment.

[0103] Assume that a satellite is operating in a circular orbit at an altitude of 1000 km above the ground, and the user equipment is located on the surface of the earth at a longitude of 116° east and a latitude of 40° north. The movement trajectory of the user equipment in the future can be predicted. For example, the device moves in the east direction at a speed of 50 km / h. First, based on the satellite orbit data and the initial position of the user equipment, the initial azimuth and elevation angle of the satellite beam pointing to the user equipment can be calculated. Assume that the initial azimuth angle is 30° and the elevation angle is 20°. Then, based on the predicted movement trajectory of the user equipment, the position of the user equipment in the future can be calculated. For example, to predict the position of the user equipment in the next 10 minutes, a simple linear extrapolation method can be used to obtain the longitude of the user equipment after 10 minutes as 11139° east and the latitude as 40° north. Next, calculate the beam center deviation at the current satellite beam pointing angle. Assuming that the 3dB beam width of the satellite beam is 2°, it can be calculated that the deviation between the beam center and the predicted position of the user equipment at the current beam pointing angle is 5°. Since the deviation of 5° is greater than the preset iteration termination condition (for example, 1°), it is necessary to continue iterating. According to the Newton-Raphson iteration formula, the correction of the beam pointing angle can be calculated. Assuming that the correction is 2°, the correction is added to the initial azimuth to obtain a new satellite beam pointing azimuth of 32°. Repeat the above steps, using the new satellite beam pointing angle to calculate the deviation between the beam center and the predicted position of the user device. If the deviation still does not meet the iteration termination condition, continue to iterate until the deviation is less than the preset iteration termination condition. Finally, the optimal satellite beam pointing angle can be obtained, so that the satellite beam can always accurately track the user device.

[0104] Specifically, the step of using a gradient descent method to solve and determine the optimal transmission power of the optimal satellite according to the distance between the user equipment and the optimal satellite and the link budget result specifically includes:

[0105] According to the distance between the user equipment and the optimal satellite, a link loss parameter is obtained by using a link budget method, and a total link loss is determined by using the link loss parameter;

[0106] Determine a bandwidth and a signal-to-noise ratio threshold of the link, and obtain a minimum receiving power required for the link using the Shannon formula based on the total loss, the bandwidth, and the signal-to-noise ratio threshold;

[0107] Determine an optimal satellite antenna gain parameter and a transmitter efficiency parameter, and obtain an optimal satellite initial transmission power by reverse calculation of a link budget method based on the minimum received power, the antenna gain parameter and the transmitter efficiency parameter;

[0108] According to the initial transmit power, the gradient descent method is used to obtain the optimal transmit power by setting the step size and the number of iterations.

[0109] In this embodiment, according to the distance between the user equipment and the satellite, the link budget method is used to obtain the free space loss, atmospheric loss and other parameters of the link, and the total loss of the link is obtained. According to the total loss of the link, the Shannon formula is used to set the bandwidth and signal-to-noise ratio threshold of the link to obtain the minimum receiving power required for the link. According to the minimum receiving power, the reverse calculation of the link budget method is used to set the parameters such as the satellite antenna gain and the transmitter efficiency to obtain the initial transmission power of the satellite. According to the initial transmission power, the gradient descent method is used to set the step size and the number of iterations to obtain the optimal transmission power that makes the link signal-to-noise ratio meet the communication quality requirements. According to the optimal transmission power, the link budget method is used to substitute various parameters to obtain the optimized link signal-to-noise ratio. According to the optimized link signal-to-noise ratio, it is judged whether it meets the communication quality requirements. If it meets the requirements, the satellite transmission power is determined to be the optimal value. If it does not meet the requirements, it returns to continue iterative optimization until the requirements are met.

[0110] Assume that the distance between the user equipment and the optimal satellite is 37,500 kilometers. First, according to the free space path loss formula, the link loss parameter can be calculated: path loss (dB) = 345 + 20lg (frequency (MHz)) + 20lg (distance (km)). Assuming the operating frequency is 2GHz, the path loss is 345 + 20lg (2000) + 20lg (37,500) = 196dB. Other link losses, such as atmospheric loss, rain attenuation loss, etc., can be calculated by looking up a table or using an empirical formula, assuming a total of 4dB. Therefore, the total loss of the link is 196dB + 4dB = 200dB.

[0111] Assume that the bandwidth of the link is 10MHz and the signal-to-noise ratio threshold is 10dB. According to Shannon's formula, the minimum receiving power required for the link can be calculated: minimum receiving power (dBW) = -174 + 10*log10 (bandwidth (Hz)) + signal-to-noise ratio threshold (dB) = -174 + 10*log10 (10*10^6) + 10 = -94dBW. Assume that the antenna gain of the optimal satellite is 40dBi and the transmitter efficiency is 80%.

[0112] According to the link budget formula: Transmit power (dBW) = Receive power (dBW) + Path loss (dB) - Transmit antenna gain (dBi) + Other losses (dB) - Transmitter efficiency (dB), the optimal initial transmit power of the satellite can be calculated: Initial transmit power (dBW) = -94dBW + 200dB - 40dBi + 4dB - (-91dB) = 731dBW.

[0113] The initial transmit power can be optimized by using the gradient descent method. Assume that the step size is 1 and the number of iterations is 100. In each iteration, the signal-to-noise ratio of the receiver at the current transmit power is calculated, and the gradient of the loss function is calculated. Then, the transmit power is updated according to the gradient direction. Through multiple iterations, the optimal transmit power that minimizes the loss function can be found. For example, after the first iteration, assuming that the calculated gradient of the loss function is -5, the updated transmit power is 731dBW-1*(-5)=736dBW. Repeat this process until the maximum number of iterations is reached, and finally the optimal transmit power is obtained.

[0114] Specifically, constructing the objective function and constraint conditions of the beamforming problem, using the minimum mean square error criterion and the convex optimization algorithm to solve the objective function and the constraint conditions to obtain the optimal beamforming weights, specifically includes:

[0115] Based on the location coordinate information of the user equipment, the satellite orbit parameters and antenna parameters of the optimal satellite, a satellite antenna beamforming mathematical model is established, wherein the satellite antenna beamforming mathematical model includes an objective function and constraint conditions of the beamforming problem;

[0116] The beamforming problem is transformed into a convex optimization problem by adopting a minimum mean square error criterion, and the objective function and the constraint condition are combined into a Lagrangian function by introducing Lagrangian multipliers, thereby obtaining a dual problem of the beamforming problem;

[0117] According to the KKT condition of the dual problem, a convex optimization algorithm is used to iteratively solve the dual problem to obtain the optimal Lagrange multiplier and dual variable;

[0118] The optimal beamforming weights of the beamforming problem are obtained by solving the problem according to the optimal Lagrange multiplier and the dual variable.

[0119] In this embodiment, according to the coverage requirements of the satellite communication system, the location information of the area where the user equipment is located is obtained, and the mathematical model of the satellite antenna beamforming is established through the satellite orbit parameters and antenna parameters to obtain the objective function and constraints of the beamforming problem. The minimum mean square error criterion is adopted to transform the beamforming problem into a convex optimization problem. By introducing the Lagrange multiplier, the objective function and the constraints are combined into the Lagrange function to obtain the dual problem of the beamforming problem. According to the KKT (Karush-Kuhn-Tucker, which refers to a set of necessary and sufficient conditions that need to be met when solving optimization problems, especially optimization problems with constraints) conditions of the dual problem, a convex optimization algorithm such as the gradient descent method is adopted to iteratively solve the dual problem to obtain the optimal Lagrange multiplier and dual variable, and then obtain the optimal solution of the original problem, that is, the beamforming weight. The obtained beamforming weight is substituted into the satellite antenna array, and the satellite beam energy is concentrated in the area where the user equipment is located by adjusting the amplitude and phase of each array element, thereby improving the coverage efficiency and service quality of the satellite communication system. The beamforming results are verified and evaluated by simulation and testing methods. By analyzing indicators such as satellite beam energy distribution and signal-to-interference-noise ratio, it is determined whether the beamforming effect meets the user's communication needs, and optimization and improvement are performed when necessary. According to the development of satellite communication services and changes in user needs, user equipment location information and satellite orbit parameters and other data are regularly updated. By repeatedly executing the above beamforming steps, dynamic optimization and adaptive adjustment of satellite beams are achieved to ensure the service quality and user experience of the satellite communication system.

[0120] Assume that the location coordinates of the user equipment are 113° longitude and 39° latitude, and the orbital parameters of the best satellite are: inclination 55°, ascending node right ascension 100°, orbit altitude 35786 km, eccentricity close to 0, antenna is 16×16 array planar antenna, and antenna unit spacing is half wavelength.

[0121] First, it is necessary to calculate the direction vector of the satellite antenna pointing to the user device based on the location coordinates of the user device and the satellite orbit parameters. The calculated direction vector is (577,577,577). Next, it is necessary to establish a mathematical model for satellite antenna beamforming, assuming that the goal is to use beamforming technology to point the main lobe of the satellite antenna beam to the user device and suppress the side lobe as much as possible. The beamforming problem can be modeled as a problem of minimizing the mean square error between the beam pattern in the direction of the user device and the ideal direction.

[0122] The objective function can be defined as: Where w represents the complex weight vector of 16×16=256 antenna elements, represents the satellite antenna array response vector, θ and Represent the elevation angle and azimuth angle of the user equipment direction, and y represents the expected received signal in the direction of the user equipment. The constraint condition is that the antenna array transmit power remains constant, that is, ||w||^2=1. In order to solve this optimization problem, the minimum mean square error (MMSE) criterion can be used to transform the problem into a convex optimization problem. The Lagrangian multiplier λ is introduced to combine the objective function and the constraint condition into a Lagrangian function: According to the KKT conditions of the dual problem, the optimal solution can be obtained to satisfy the following equation: And ||w||^2=1. This is a system of linear equations that can be solved iteratively using a convex optimization algorithm. For example, the gradient descent method can be used to iteratively update the values ​​of w and λ until the convergence condition is met. Assume that through iterative solution, the optimal Lagrange multiplier λ=01 and the corresponding dual variable w=T are obtained, where i represents the imaginary unit. According to the optimal Lagrange multiplier and dual variable, the optimal beamforming weight w for the beamforming problem can be obtained, which is T. These weights will be used to adjust the amplitude and phase of the transmitted signal of each antenna unit, thereby achieving precise control of the satellite antenna beam shape. Ultimately, through beamforming technology, the satellite antenna can concentrate the beam energy to the area where the user equipment is located, improve signal strength, while suppressing interference from other directions and improving communication quality.

[0123] S105, predicting the user communication behavior of the user device through a long short-term memory neural network algorithm, determining the predicted request data, and caching each data content in the predicted request data in advance to the optimal edge server according to the priority sorting relationship through an information entropy algorithm.

[0124] In this embodiment, historical communication data of the user device is collected, including communication time, communication content, communication frequency, etc., and the communication behavior of the user device is modeled and predicted using the LSTM algorithm. The LSTM algorithm can learn long-term dependencies, that is, the communication behavior of the user device in the past period of time may affect its future communication behavior. Through the LSTM model, the communication request data of the user device in the future period of time is predicted, including possible communication time, content type, etc.

[0125] The request data predicted by LSTM is classified according to the content type. For each type of data content, the information entropy algorithm is used to calculate its priority. Information entropy can measure the uncertainty or confusion of data. For satellite IoT communications, data with higher uncertainty and confusion tends to be more important, so its priority is also higher. The calculation formula of information entropy is: H(X) = -Σp(x)log2p(x), where p(x) represents the probability of data content x appearing. By calculating the information entropy of each type of data content, its priority ranking can be obtained.

[0126] According to the priority of the predicted request data, the high-priority data content is cached in advance to the optimal edge server. The optimal edge server is dynamically selected based on factors such as the current location of the user device, network conditions, and server load. The caching process can be based on the storage capacity and caching strategy of the edge server, such as using algorithms such as least recently used (LRU) or least frequently used (LFU) to determine which content needs to be replaced or deleted.

[0127] In this embodiment, the communication behavior of the user device is predicted by the LSTM algorithm, and the relevant data content can be prepared and cached in advance, thereby reducing communication delay and increasing data transmission speed. The information entropy algorithm can accurately measure the priority of data and ensure that the most important data content is processed and cached first. Since the data content that the user may need is cached in advance, the user can obtain the required information faster when communicating, thereby improving the user experience.

[0128] In some embodiments, the above step S105 specifically includes:

[0129] Acquire historical communication behavior data of the user device, and use a long short-term memory neural network algorithm to train the historical communication behavior data to obtain a user communication behavior prediction model;

[0130] Acquire predicted request data of the user device within a preset future time according to the user communication behavior prediction model;

[0131] Determining the probability distribution of each data content by counting the access frequency of each data content in the prediction request data;

[0132] According to the probability distribution of each data content, the information entropy algorithm is used to calculate the information entropy of each data content to obtain the information entropy value reflecting the uncertainty of the data content;

[0133] Determine the cache priority of each data content according to the information entropy value, sort the data content according to the cache priority, and obtain a priority sorting result;

[0134] According to the cache capacity of the optimal edge server and the size of each data content, a greedy algorithm is used to select the data content with the highest priority for caching through the priority sorting result, until the cache space of the optimal edge server is filled.

[0135] In this embodiment, according to the historical communication behavior data of the user, the data preprocessing technology is used to clean, convert and standardize the data to obtain a data set suitable for training the neural network model. By extracting the features of the preprocessed historical communication behavior data, key features that can reflect the user's communication behavior pattern, such as communication time, frequency, duration, etc., are obtained to determine the feature dimensions of the input layer. According to the extracted user communication behavior features, the long short-term memory neural network algorithm is used to construct a user communication behavior prediction model, set the appropriate number of hidden layers and neurons, and select appropriate activation functions and loss functions. By dividing the preprocessed historical communication behavior data into a training set and a validation set, the constructed long short-term memory neural network model is trained and optimized, and the hyperparameters of the model are continuously adjusted to obtain the user communication behavior prediction model with the best performance. According to the trained user communication behavior prediction model, the user's communication behavior data in the recent period is obtained as input, and the prediction results of the data content that the user may request in the future period are obtained through the forward propagation process. The data content type, request time and other information in the prediction results are used, combined with the user's historical preferences, to determine the specific data content that the user is most likely to request, and provide it to the user as a recommendation result. By collecting and analyzing user feedback on recommended content, we obtain the user's actual request data, compare it with the prediction results, calculate the accuracy and recall rate of the prediction model and other evaluation indicators, and judge whether the performance of the model meets the business needs. Based on the model evaluation results, we adopt the incremental learning method, regularly add new user communication behavior data to the training set, and iteratively optimize and update the user communication behavior prediction model to adapt to changes in user behavior patterns and improve the accuracy and real-time performance of the prediction.

[0136] By counting the access frequency of each data content in the predicted request data, the probability distribution of each data content is determined. According to the probability distribution of the data content, the information entropy of each data content is calculated by using the information entropy algorithm to obtain the information entropy value reflecting the uncertainty of the data content. According to the information entropy value, the cache priority of the data content is determined. The larger the information entropy value, the higher the uncertainty of the data content and the higher the priority. The priority sorting result of the data content is obtained by a priority sorting algorithm, such as heap sorting or quick sorting. According to the cache capacity of the edge server and the size of the data content, a greedy algorithm is used to select the data content with the highest priority for caching until the cache space is filled, and the final data caching solution is obtained. The selected high-priority data content is cached in the edge server in advance, and the response speed and service quality of the user request are improved by the nearest access method. The changes in the user request mode and data content are continuously monitored, and the priority and content of the data cache are dynamically adjusted according to the actual situation to ensure the timeliness and accuracy of the cache. The effect of the data cache priority sorting algorithm is evaluated through indicators such as cache hit rate and response delay, and the algorithm is optimized and the parameters are adjusted according to the feedback results to continuously improve the performance and user experience of the edge cache.

[0137] For example, first collect the user's communication records for the past three months, including call, text message, Internet access and other behavior data, sampled every 5 minutes, totaling about 250,000 records. Then, input these data into a 3-layer long short-term memory neural network with 128 neurons in each layer in chronological order for training, set the learning rate to 0.1, iterate 1,000 times, and finally obtain a user communication behavior prediction model. Using this model, predict the communication behavior that the user may generate in the next hour, and obtain a sequence of 50 data requests. Then, by counting the frequency of each data request, calculate its information entropy, and the top 10 data requests with the largest information entropy are considered to be the data that the user is most likely to request, and cache them in advance to the edge server closest to the user. Assuming that the analysis finds that the user most frequently visits news websites and video websites, then cache the latest news content and popular videos in advance. When the user actually initiates a request, the edge server can directly return the cached data, and the average response time can be shortened to 1 / 3 of the original. At the same time, the edge server will also track the user's actual request. If it finds that the predicted data is significantly different from the actual request, it will dynamically adjust the prediction model and cache strategy to ensure that the cached data matches the user's needs. This intelligent prediction and pre-caching method can not only improve the user experience, but also significantly reduce the network bandwidth pressure and the load of the central cloud server, and achieve efficient use of network resources.

[0138] Reference Figure 2The embodiment of the present invention further provides a computer device 2, comprising: a memory 202 and a processor 201 and a computer program 203 stored in the memory 202. When the computer program 203 is executed on the processor 201, an intelligent energy consumption control method for satellite Internet of Things communication in remote areas as described in any one of the above methods is implemented.

[0139] The computer device 2 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 2 may include, but is not limited to, a processor 201 and a memory 202. Those skilled in the art will appreciate that Figure 2 It is only an example of the computer device 2 and does not constitute a limitation on the computer device 2. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0140] The processor 201 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0141] In some embodiments, the memory 202 may be an internal storage unit of the computer device 2, such as a hard disk or memory of the computer device 2. In other embodiments, the memory 202 may also be an external storage device of the computer device 2, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device 2. Further, the memory 202 may also include both an internal storage unit of the computer device 2 and an external storage device. The memory 202 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory 202 may also be used to temporarily store data that has been output or is to be output.

[0142] An embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for intelligent energy consumption control for satellite Internet of Things communications in remote areas as described in any of the above methods is implemented.

[0143] In this embodiment, if the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0144] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0145] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0146] In the embodiments disclosed in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0147] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

Claims

1. An intelligent energy consumption control method for satellite Internet of Things communications in remote areas, characterized in that: The method specifically comprises: Determine estimated energy consumption and estimated latency for IoT communications between user devices, edge servers, and satellites in different types of remote areas based on location coordinate information and computational load information; The method of determining the estimated energy consumption and the estimated delay of IoT communications between the user equipment, each edge server and each satellite in different types of remote areas according to the location coordinate information and the calculation load information specifically includes: According to the location coordinate information and computing load information of each user device, the K-means clustering algorithm is used to cluster each user device to obtain the clustering results of user devices in different types of remote areas; According to the position coordinate information and calculation load information of each edge server and each satellite and the user device clustering result, the shortest distance between the current user device and each edge server, and between each edge server and each satellite is calculated by the Dijkstra algorithm; According to the shortest distance, the communication rate between the current user device and each edge server, and between each edge server and each satellite is calculated using the Shannon information theory formula; According to the communication rate and the computing load information of the current user equipment, each edge server, and each satellite, the estimated delay between the current user equipment and each edge server, and between each edge server and each satellite is calculated through a queuing theory model; According to the communication rate, the communication distance between the current user equipment and each edge server, and the communication distance between each edge server and each satellite, the energy consumption estimation value between the current user equipment and each edge server, and between each edge server and each satellite is calculated using a wireless communication energy consumption model; Determine a predicted movement trajectory of the user equipment, take minimizing the estimated energy consumption value and the estimated delay value as an optimization goal, and use a multi-objective particle swarm optimization algorithm to dynamically select an optimal edge server and an optimal satellite for the user equipment in combination with the predicted movement trajectory of the user equipment; Based on the Internet of Things communication relationship between the user equipment, the optimal edge server and the optimal satellite, according to the channel environment parameters of the remote area where the user equipment is currently located, an adaptive coding modulation technology is used to establish a convex optimization model between coding energy consumption and data compression ratio through a convex optimization method, and the convex optimization model is used to determine the optimal coding parameter configuration for data coding in the Internet of Things communication relationship, and the optimal coding parameter configuration is used to achieve an optimal balance point between coding energy consumption and data compression ratio, and the Internet of Things communication relationship refers to a scheme for Internet of Things communication in different types of remote areas; Dynamically adjusting the optimal satellite beam pointing parameters, transmission power parameters and beamforming weight parameters based on the predicted movement trajectory of the user equipment; The dynamically adjusting the optimal satellite beam pointing parameter, transmit power parameter and beamforming weight parameter based on the predicted movement trajectory of the user equipment specifically includes: According to the predicted movement trajectory of the user equipment, a Newton-Raphson iteration method is used to solve and determine the optimal beam pointing angle of the optimal satellite; According to the distance between the user equipment and the optimal satellite and the link budget result, the optimal transmission power of the optimal satellite is determined by using the gradient descent method; Constructing an objective function and constraint conditions of a beamforming problem, and using a minimum mean square error criterion and a convex optimization algorithm to solve the objective function and the constraint conditions to obtain an optimal beamforming weight; Determine an optimal satellite beam tracking and focusing scheme according to the optimal beam pointing angle, the optimal transmit power and the optimal beamforming weight; The user communication behavior of the user device is predicted by a long short-term memory neural network algorithm to determine the predicted request data, and each data content in the predicted request data is cached in advance to the optimal edge server according to the priority sorting relationship by an information entropy algorithm.

2. The method according to claim 1, characterized in that The method further includes determining an estimated energy consumption and a delay for IoT communication between a user device, each edge server, and each satellite in different types of remote areas according to the location coordinate information and the calculation load information, and then further including: Obtain a first comparison result by comparing the estimated delay value and the estimated energy consumption value between the current user equipment and any edge server with their respective preset thresholds; Obtain a second comparison result by comparing the estimated delay value and the estimated energy consumption value between each edge server and any satellite with their respective preset thresholds; Based on the first comparison result and the second comparison result, a scheme for performing Internet of Things communication between the current user equipment, each edge server and each satellite in different types of remote areas is determined.

3. The method according to claim 1, characterized in that The determining of the predicted movement trajectory of the user equipment specifically includes: Determine the current position state parameters of the user equipment according to the position coordinate information of the user equipment, wherein the current position state parameters include the current position coordinates, speed and acceleration; Taking the current position state parameter as an input value, using the prediction equation of the Kalman filter algorithm to perform prediction, and obtaining a predicted value of the user equipment at the next moment, the predicted value including a predicted position coordinate, a predicted speed, and a predicted acceleration; Acquire a measurement value of the user equipment at a next moment, the measurement value including a measurement position coordinate, a measurement speed, and a measurement acceleration, and obtain an optimal estimate value of the user equipment at the next moment by substituting the predicted value and the measurement value into an update equation of a Kalman filter algorithm for correction; Using the optimal estimated value as a new input value, performing iterative calculation using a Kalman filter algorithm, and updating the optimal estimated value in real time; Based on the optimal estimate, the prediction equation of the Kalman filter algorithm is recursively extrapolated for t time steps to obtain a predicted movement trajectory of the user device from the current moment to a time period of t in the future.

4. The method according to claim 1, characterized in that: The optimization goal is to minimize the estimated energy consumption value and the estimated delay value, and in combination with the predicted movement trajectory of the user equipment, a multi-objective particle swarm optimization algorithm is used to dynamically select the optimal edge server and the optimal satellite for the user equipment, specifically including: The predicted movement trajectory of the user equipment, all the estimated energy consumption values ​​and the estimated delay values ​​are used as inputs of a multi-objective particle swarm optimization algorithm, and a fitness function is constructed with minimizing the estimated energy consumption value and the estimated delay value as the optimization target; A multi-objective particle swarm optimization algorithm is used to solve the task offloading decision problem, and the particle position and velocity are updated through iterative search, the fitness function value of the fitness function is continuously optimized, and the Pareto optimal solution set is obtained; Select a task offloading decision scheme whose energy consumption estimate and delay estimate are both less than their respective preset thresholds from the Pareto optimal solution set, determine the offloading scheme of the user task initiated by each user device on the edge server and the satellite, and obtain the task offloading decision result; According to the task offloading decision results, the user tasks initiated by each user device are offloaded to the corresponding optimal edge server and optimal satellite for execution; The movement trajectory data of the user equipment, the energy consumption data of the edge server and the satellite, and the task execution delay data are continuously collected to dynamically adjust and optimize the parameters of the multi-objective particle swarm optimization algorithm.

5. The method according to claim 1, characterized in that The method of using adaptive coding modulation technology according to the channel environment parameters of the remote area where the user equipment is currently located, and establishing a convex optimization model between coding energy consumption and data compression ratio by a convex optimization method specifically includes: Acquire an initial value of a channel environment parameter in a remote area where the user equipment is currently located according to an IoT communication relationship between the user equipment, an optimal edge server, and an optimal satellite; An IoT communication link model is established according to the geometric position relationship between the user equipment and the optimal edge server and the optimal satellite, the initial value of the channel environment parameter is substituted into the IoT communication link model, and a channel estimation algorithm is used to calculate and obtain the channel environment parameter of the remote area where the user equipment is currently located; According to the channel environment parameters, the coding mode and the modulation mode are determined by using the adaptive orthogonal frequency division multiplexing technology and the adaptive orthogonal amplitude modulation technology to obtain the optimal coding and modulation scheme; By using the optimal coding modulation scheme, the relationship between coding energy consumption and data compression ratio is obtained, and a first convex optimization model with coding energy consumption and data compression ratio as optimization variables is established; Taking minimizing encoding energy consumption and maximizing data compression ratio as optimization goals, and combining the signal environment parameters, constructing a convex optimization objective function and constraint conditions; According to the first convex optimization model, the convex optimization objective function and the constraint condition, a target convex optimization model is formed; The target convex optimization model is solved by the interior point method, and the value of the optimization variable is continuously updated by iterative optimization until the constraint conditions are met, thereby obtaining the optimal encoding energy consumption and data compression ratio.

6. The method according to claim 1, characterized in that The method of solving and determining the optimal beam pointing angle of the optimal satellite by using the Newton-Raphson iteration method according to the predicted moving trajectory of the user equipment specifically includes: According to the orbital data of the optimal satellite and the predicted movement trajectory of the user equipment, an iterative equation of the satellite beam pointing angle is established by using the Newton-Raphson iteration method, and an initial value of the satellite beam pointing angle is obtained by setting an initial value of the iteration and an iteration termination condition; According to the initial value of the satellite beam pointing angle and the predicted moving trajectory of the user equipment, the dot product formula of the beam direction vector and the position vector is used to calculate the deviation between the satellite beam center and the predicted trajectory of the user equipment to obtain the beam center deviation under the current satellite beam pointing angle; Determine whether the beam center deviation satisfies a set iteration termination condition. If not, use the Newton-Raphson iteration formula to find the partial derivative of the beam pointing angle according to the beam center deviation to obtain a beam pointing angle correction value; A new satellite beam pointing angle is determined according to the beam pointing angle correction amount, and the step of calculating the beam center deviation is repeated using the new satellite beam pointing angle, and the satellite beam pointing angle is continuously corrected until the beam center deviation meets the iteration termination condition, thereby obtaining the optimal satellite beam pointing angle.

7. The method according to claim 1, characterized in that The step of using a gradient descent method to solve and determine the optimal transmission power of the optimal satellite according to the distance between the user equipment and the optimal satellite and the link budget result specifically includes: According to the distance between the user equipment and the optimal satellite, a link loss parameter is obtained by using a link budget method, and a total link loss is determined by using the link loss parameter; Determine a bandwidth and a signal-to-noise ratio threshold of the link, and obtain a minimum receiving power required for the link using the Shannon formula based on the total loss, the bandwidth, and the signal-to-noise ratio threshold; Determine an optimal satellite antenna gain parameter and a transmitter efficiency parameter, and obtain an optimal satellite initial transmission power by reverse calculation of a link budget method based on the minimum received power, the antenna gain parameter and the transmitter efficiency parameter; According to the initial transmit power, the gradient descent method is used to obtain the optimal transmit power by setting the step size and the number of iterations.

8. The method according to claim 1, characterized in that The objective function and constraint conditions of the beamforming problem are constructed, and the optimal beamforming weights are obtained by solving the objective function and the constraint conditions using the minimum mean square error criterion and the convex optimization algorithm, specifically including: Based on the location coordinate information of the user equipment, the satellite orbit parameters and antenna parameters of the optimal satellite, a satellite antenna beamforming mathematical model is established, wherein the satellite antenna beamforming mathematical model includes an objective function and constraint conditions of the beamforming problem; The beamforming problem is transformed into a convex optimization problem by adopting a minimum mean square error criterion, and the objective function and the constraint condition are combined into a Lagrangian function by introducing Lagrangian multipliers, thereby obtaining a dual problem of the beamforming problem; According to the KKT condition of the dual problem, a convex optimization algorithm is used to iteratively solve the dual problem to obtain the optimal Lagrange multiplier and dual variable; The optimal beamforming weights of the beamforming problem are obtained by solving the problem according to the optimal Lagrange multiplier and the dual variable.

9. The method according to claim 1, characterized in that: The method predicts the user communication behavior of the user device by using a long short-term memory neural network algorithm, determines the predicted request data, and caches each data content in the predicted request data in advance to the optimal edge server according to the priority sorting relationship by using an information entropy algorithm, specifically including: Acquire historical communication behavior data of the user device, and use a long short-term memory neural network algorithm to train the historical communication behavior data to obtain a user communication behavior prediction model; Acquire predicted request data of the user device within a preset future time according to the user communication behavior prediction model; Determining the probability distribution of each data content by counting the access frequency of each data content in the prediction request data; According to the probability distribution of each data content, the information entropy algorithm is used to calculate the information entropy of each data content to obtain the information entropy value reflecting the uncertainty of the data content; Determine the cache priority of each data content according to the information entropy value, sort the data content according to the cache priority, and obtain a priority sorting result; According to the cache capacity of the optimal edge server and the size of each data content, a greedy algorithm is used to select the data content with the highest priority for caching through the priority sorting result, until the cache space of the optimal edge server is filled.

Images