Adaptive power control method for energy transmission in heterogeneous wireless Internet of Things

Abstract

The invention discloses a self-adaptive power control method for energy transmission in a heterogeneous wireless internet of things, which comprises the following steps of: S1, acquiring network parameters and equipment state information, and acquiring an equipment position, communication link quality, a network load and the like; s2, according to the information, preliminarily calculating the transmission power demand of each device, and generating a power distribution scheme; s3, optimizing energy transmission by adjusting the transmission power of each device in real time; s4, coordinating power distribution among the equipment by adopting a cooperative game algorithm, and realizing equipment cooperation under the condition of limited energy resources; s5, on the basis of a Bayesian reasoning prediction algorithm, estimating network condition changes in advance, and adjusting a power distribution scheme in real time; s6, continuously monitoring the power control effect of the equipment, and optimizing the strategy based on the monitoring data; and S7, regularly updating the power control parameters. According to the invention, through combination of adaptive power adjustment and a prediction mechanism, the efficiency and stability of energy transmission in the wireless Internet of Things are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless communications and Internet of Things technologies, and in particular to an adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things. Background Art

[0002] In modern wireless communication and IoT technologies, heterogeneous wireless IoT plays an increasingly important role in IoT applications due to its diverse devices and communication protocols. With the rapid popularization of IoT devices and the expansion of network scale, how to achieve stable and efficient energy transmission between these devices has become one of the key issues. Especially in the heterogeneous wireless IoT environment, the management and optimization of the energy transmission process becomes particularly complex due to the differences in device types, transmission protocols and communication links.

[0003] Traditional power control methods are usually optimized for a single type of network or device, and often cannot adapt to the complex environment of heterogeneous wireless IoT. In traditional wireless communications, fixed power control schemes or simple dynamic adjustment strategies can no longer meet the needs of heterogeneous networks due to the rapid increase in the number of devices, frequent environmental changes, and different energy requirements of different devices. These traditional technologies have the following defects:

[0004] The first is that the power control method lacks adaptability and flexibility. Traditional power control methods are usually based on fixed parameter settings or simple threshold adjustments, which are difficult to adapt to the dynamic changes of network environment and device status in real time. In heterogeneous wireless Internet of Things, due to the wide variety of devices, multiple factors such as communication link quality, energy demand, network load, etc. change at any time, making it difficult for power control schemes to adapt to complex and changing environments, and often resulting in energy waste or unstable transmission.

[0005] The second is the lack of coordination mechanism between devices. In the existing technology, the power control of each device is often carried out independently, and there is a lack of coordination mechanism between devices. This independent control may lead to increased mutual interference between devices in heterogeneous networks, especially in high-density device environments, where transmission interference between devices will significantly reduce the overall performance of the system. In addition, the existing power control scheme cannot reasonably coordinate the power allocation between different devices, resulting in low energy utilization efficiency and difficulty in achieving overall power optimization of the system.

[0006] Third, existing methods cannot accurately predict changes in the network environment. In a complex heterogeneous wireless IoT environment, device status and network conditions may fluctuate frequently. For example, the remaining energy of the device, link quality, or network load may change significantly. However, most existing power control schemes are based on past status information and lack the ability to predict and respond to future changes in network conditions. This lag often leads to delays in power adjustment and an inability to quickly respond to environmental changes, which affects the transmission efficiency and stability of the system.

[0007] Therefore, how to provide an adaptive power control method for energy transmission in heterogeneous wireless Internet of Things is a problem that technicians in this field need to solve urgently. Summary of the invention

[0008] One object of the present invention is to propose an adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things. The present invention makes full use of adaptive control, cooperative game algorithms and prediction algorithms based on Bayesian reasoning, and describes in detail the mechanism of dynamically adjusting device power allocation in a complex network environment to ensure collaborative cooperation and power optimization between devices. The method has the advantages of high energy utilization efficiency, strong network adaptability, good transmission stability and low interference between devices, and can significantly improve the energy transmission effect and overall system performance in a heterogeneous wireless Internet of Things environment.

[0009] An adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things according to an embodiment of the present invention includes the following steps:

[0010] S1. Collect network parameters and device status information in a heterogeneous wireless IoT environment. The network parameters include device location, communication link quality, and network load. The device status information includes the remaining energy and data transmission requirements of the device.

[0011] S2. Preliminarily calculate the transmission power requirement of each device according to the network parameters and device status information, and generate a corresponding power allocation plan;

[0012] S3, dynamically adjust the transmission power of each device according to the real-time collected network parameters and device status information;

[0013] S4. Introduce cooperative game algorithm to coordinate power allocation among multiple devices and realize normal cooperation among devices under the condition of limited energy resources;

[0014] S5. When changes in the network environment and device status are detected, a prediction algorithm based on Bayesian reasoning is used to predict the changing trend of network conditions in advance, and the power allocation plan is recalculated based on the prediction results, and then the transmission power of each device is adjusted in real time;

[0015] S6. Continuously monitor the power control effect of each device, record the power consumption, transmission efficiency and communication interference of each device during the transmission process, and optimize the power control strategy based on these monitoring data;

[0016] S7. Regularly update and store the optimized power control parameters.

[0017] Optionally, the S3 specifically includes:

[0018] S31. Define the real-time transmission power of each device as P i (t), where i represents the device number and t represents the time point;

[0019] S32. Determine the remaining energy E of each device based on the real-time status information of the device. i (t), and the quality indicator Q of the communication link i (t), and combined with the network load condition L i (t), preliminarily calculate the target transmission power requirement P of each device i,target (t);

[0020] S33, the initially calculated target transmission power requirement P i,target (t) is used as input to dynamically adjust the actual transmission power P of each device i (t), the dynamic adjustment process is achieved by minimizing the following objective function:

[0021]

[0022] Where N is the total number of devices, α, β, and γ are parameters used to balance the weights, (P i (t)-P i,target (t)) 2 represents the square of the deviation of the transmission power, Indicates the inverse ratio of link quality, L i (t) indicates the network load of the current device;

[0023] S34, by dynamically updating the transmission power adjustment strategy of each device, the reward mechanism is evaluated according to the difference between the actual transmission effect and the expected effect;

[0024] S35: Finally output the actual transmission power P of each device at time point t i (t), and is used in the current wireless energy transfer process.

[0025] Optionally, the S33 specifically includes:

[0026] S331, based on the real-time status information of each device, including the actual transmission power P i (t), link quality indicator Qi (t), network load condition L i (t), interference coefficient between devices I ij (t) and target transmission power requirement P i,target (t), construct the objective function J(t) for adaptive power adjustment:

[0027]

[0028] Among them, α, β, γ, and δ are coefficients for adjusting the actual transmission power, link quality index, network load, interference coefficient between devices, and target transmission power requirement weight. ij (t) is the interference coefficient between device i and device j, d ij is the distance between device i and device j;

[0029] S332. Apply the Q-learning algorithm in reinforcement learning to determine the optimal transmission power adjustment strategy of the device according to the value of the objective function J(t). First, define the state space S, including the current transmission power P of the device. i (t), link quality Q i (t), network negative L i (t) and the interference coefficient between devices I ij (t) state variable, then define the action space A, and finally define the reward function R(t) as the negative value of the objective function J(t);

[0030] S333. Iteratively learn the state space S and action space A through the Q-learning algorithm, and update the Q value function Q(s,a) of each state-action pair:

[0031] Q(s,a)←Q(s,a)+η[R(t)+γ·max a' Q(s',a')-Q(s,a)];

[0032] Among them, η is the learning rate, γ is the discount factor, s' represents the next state after executing action a, and max a' Q(s',a') represents the maximum Q value of taking the best action in the next state s';

[0033] S334, in the learning process of the Q-learning algorithm, dynamically select the optimal action a in each state s * , adjust the actual transmission power P of the device i (t) Close to P i,target (t), and at the same time optimize the interference effect between devices, and finally achieve the purpose of optimizing the objective function J(t).

[0034] Optionally, the S4 specifically includes:

[0035] S41. Define the utility function U of each device i (t), the utility function takes into account the energy demand of the equipment, the remaining energy E i (t), current transmission power P i (t) and interference with other devices:

[0036]

[0037] Among them, λ and μ are adjustment coefficients, Indicates the effectiveness of the equipment's energy consumption, Indicates the interference between devices;

[0038] S42, introduce a cooperative game algorithm between multiple devices, through the utility function U i (t) optimization, finding the Nash equilibrium point so that the transmission power allocation of each device reaches a normal state;

[0039] S43. Use power allocation based on cooperative game algorithm to calculate the strategy of each device in the game and optimize power allocation:

[0040]

[0041] in, represents the optimal transmission power of device i under the current game;

[0042] S44, in order to obtain the optimal transmission power of each device After that, update the actual transmission power P of each device i (t) To enable all devices to operate in a coordinated manner in the overall system and to achieve normal power allocation and energy transmission under energy-constrained conditions;

[0043] S45. Utility function U of continuous monitoring equipment i (t) and the overall system utility. When a decrease in utility is detected, the cooperative game process is restarted to adjust the power allocation strategy of the devices.

[0044] Optionally, the step S42 includes:

[0045] S421, construct a cooperative game model, in which each device i in the model acts as a participant in the game and selects a transmission power P i (t) as its strategy space A i The main decision variables of,the policy space are determined by the hardware capabilities of the,device and the current network conditions;

[0046] S422: For each device i, given the policy P of other devices j, j(t), optimize its utility function U i (t), determine the optimal strategy for device i

[0047]

[0048] in, Indicates the nonlinear relationship of energy consumption, ∈ is a small positive number to avoid the calculation problem of the logarithmic function when the denominator is zero, η is the coefficient used to adjust the dynamic impact of interference, and ν is an exponential parameter greater than 1, which is used to reflect the nonlinear impact of interference with power changes;

[0049] S424, through iterative calculation, each device adjusts the strategy P i (t) and P j (t), until the Nash equilibrium point is reached and At this time, any device i cannot change its own policy P while other devices keep their policies unchanged. i (t) to further improve its utility function U i (t);

[0050] S425. At the Nash equilibrium point Under this condition, the transmission power P of each device is fixed i (t).

[0051] Optionally, the step S5 includes:

[0052] S51, by monitoring key parameters in the network environment, including device location, link quality Q i (t), network load L i (t), equipment remaining energy E i (t) and the interference coefficient between devices I ij (t) collect this information in real time and identify changes in network environment and device status;

[0053] S52, when any of the above parameters is detected to have a significant change, triggering a prediction algorithm based on Bayesian reasoning to predict the trend of network condition changes in a future time period;

[0054] S53, based on the prediction results of the Bayesian inference algorithm, recalculate the target transmission power requirement P of each device i,target (t), in the process, the interference between devices, the nonlinear changes of network load and the dynamic adjustment of device status are considered, and the objective function J(t) is updated to reflect the predicted changes:

[0055]

[0056] Among them, log(1+(P i (t)-P i,target (t)) 2 ) introduces a nonlinear penalty for power deviation to better handle the impact of large deviations. is the link quality Q i (t) is a nonlinear processing, κ is an exponent greater than 1, which is used to adjust the sensitivity of link quality under different conditions, (L i (t)) λ Indicates network load L i (t), λ is an exponential parameter used to dynamically adjust the network load impact. This complicates the model of interference between devices. ν is an exponential parameter used to reflect the complexity of interference as power and distance change.

[0057] S54: According to the updated objective function J(t), the transmission power P of each device is i (t) Make real-time adjustments to adapt to new network conditions and device status;

[0058] S55. After completing the adjustment of the transmission power, the network environment and device status are continuously monitored. If changes in the environment and status are detected again, the above steps are repeated to maintain the adaptability of the system in a dynamic environment.

[0059] Optionally, the step S52 includes:

[0060] S521, define a hypothesis set H = {H1, H2, ..., H n}, each hypothesis H i Represents a possible change in network status, including device location, link quality Q i (t), network load L i (t) and interference coefficient I ij (t) changes;

[0061] S522, through real-time monitoring, collect evidence set E = {E1, E2,, E m}, where each piece of evidence E j Represents changes and anomalies observed in the current network environment;

[0062] S523, use the Bayesian inference model to calculate each hypothesis H i In Evidence E j The posterior probability P(H i |E j );

[0063] S524, based on the real-time updated posterior probability P(H i |Ej ), dynamically adjust the priority of each hypothesis and select the hypothesis H with the highest posterior probability best As the most likely change to the current network state;

[0064] S525, combined with the selected optimal hypothesis H best , calculate the changes in network parameters in the future time period and update the link quality Q i (t+Δt), network load L i (t+Δt), interference coefficient I ij (t+Δt) and the power demand P of the equipment i,target (t+Δt), where Δt is the prediction time interval;

[0065] S526: Use the prediction result to guide real-time power allocation adjustment.

[0066] The beneficial effects of the present invention are:

[0067] (1) The present invention effectively improves the efficiency and stability of energy transmission in heterogeneous wireless Internet of Things by introducing adaptive control mechanism, cooperative game algorithm and prediction algorithm based on Bayesian reasoning. Compared with the traditional fixed power control method, this method can dynamically adjust the power of the device in real time and adapt to the complex and changeable network environment, thereby significantly reducing energy waste and signal interference.

[0068] (2) The present invention coordinates the power allocation among multiple devices through intelligent power allocation strategy, thus improving the overall performance of the system. The cooperative game algorithm enables devices to achieve effective collaboration under limited energy resources, reduces mutual interference among devices, and ensures the reliability and effectiveness of energy transmission.

[0069] (3) The present invention combines real-time monitoring and prediction mechanisms to predict the changing trend of network conditions in advance, thereby achieving active power adjustment. This prediction capability not only enhances the adaptability of the system, but also improves the response speed in uncertain environments, ensuring that the equipment can operate in the best state. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0071] Figure 1 A flowchart of an adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things proposed by the present invention;

[0072] Figure 2The present invention provides a flow chart for dynamically adjusting transmission power in an adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things. DETAILED DESCRIPTION

[0073] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0074] refer to Figure 1-2 , an adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things, comprising the following steps:

[0075] S1. Collect network parameters and device status information in a heterogeneous wireless IoT environment. The network parameters include device location, communication link quality, and network load. The device status information includes the remaining energy and data transmission requirements of the device.

[0076] S2. Preliminarily calculate the transmission power requirement of each device according to the network parameters and device status information, and generate a corresponding power allocation plan;

[0077] S3, dynamically adjust the transmission power of each device according to the real-time collected network parameters and device status information;

[0078] S4. Introduce cooperative game algorithm to coordinate power allocation among multiple devices and realize normal cooperation among devices under the condition of limited energy resources;

[0079] S5. When changes in the network environment and device status are detected, a prediction algorithm based on Bayesian reasoning is used to predict the changing trend of network conditions in advance, and the power allocation plan is recalculated based on the prediction results, and then the transmission power of each device is adjusted in real time;

[0080] S6. Continuously monitor the power control effect of each device, record the power consumption, transmission efficiency and communication interference of each device during the transmission process, and optimize the power control strategy based on these monitoring data;

[0081] S7. Regularly update and store the optimized power control parameters.

[0082] In this implementation, S3 specifically includes:

[0083] S31. Define the real-time transmission power of each device as P i (t), where i represents the device number and t represents the time point;

[0084] S32. Determine the remaining energy E of each device based on the real-time status information of the device. i (t), and the quality indicator Q of the communication linki (t), and combined with the network load condition L i (t), preliminarily calculate the target transmission power requirement P of each device i,target (t);

[0085] S33, the initially calculated target transmission power requirement P i,target (t) is used as input to dynamically adjust the actual transmission power P of each device i (t), the dynamic adjustment process is achieved by minimizing the following objective function:

[0086]

[0087] Where N is the total number of devices, α, β, and γ are parameters used to balance the weights, (P i (t)-P i,target (t)) 2 represents the square of the deviation of the transmission power, Indicates the inverse ratio of link quality, L i (t) indicates the current network load of the device;

[0088] S34, by dynamically updating the transmission power adjustment strategy of each device, the reward mechanism is evaluated according to the difference between the actual transmission effect and the expected effect;

[0089] S35: Finally output the actual transmission power P of each device at time point t i (t), and is used in the current wireless energy transfer process.

[0090] In this implementation manner, the S33 specifically includes:

[0091] S331, based on the real-time status information of each device, including the actual transmission power P i (t), link quality indicator Q i (t), network load condition L i (t), interference coefficient between devices I ij (t) and target transmission power requirement P i,target (t), construct the objective function J(t) for adaptive power adjustment:

[0092]

[0093] Among them, α, β, γ, and δ are coefficients for adjusting the actual transmission power, link quality index, network load, interference coefficient between devices, and target transmission power requirement weight. ij (t) is the interference coefficient between device i and device j, d ij is the distance between device i and device j;

[0094] S332. Apply the Q-learning algorithm in reinforcement learning to determine the optimal transmission power adjustment strategy of the device according to the value of the objective function J(t). First, define the state space S, including the current transmission power P of the device. i (t), link quality Q i (t), network negative L i (t) and the interference coefficient between devices I ij (t) state variable, then define the action space A, and finally define the reward function R(t) as the negative value of the objective function J(t);

[0095] S333. Iteratively learn the state space S and action space A through the Q-learning algorithm, and update the Q value function Q(s,a) of each state-action pair:

[0096] Q(s,a)←Q(s,a)+η[R(t)+γ·max a' Q(s',a')-Q(s,a)];

[0097] Among them, η is the learning rate, γ is the discount factor, s' represents the next state after executing action a, and max a' Q(s',a') represents the maximum Q value of taking the best action in the next state s';

[0098] S334, in the learning process of the Q-learning algorithm, dynamically select the optimal action a in each state s * , adjust the actual transmission power P of the device i (t) Close to P i,target (t), and at the same time optimize the interference effect between devices, and finally achieve the purpose of optimizing the objective function J(t).

[0099] In this implementation manner, the S4 specifically includes:

[0100] S41. Define the utility function U of each device i (t), the utility function takes into account the energy demand of the equipment, the remaining energy E i (t), current transmission power P i (t) and interference with other devices:

[0101]

[0102] Among them, λ and μ are adjustment coefficients, Indicates the effectiveness of the equipment's energy consumption, Indicates the interference between devices;

[0103] S42, introduce a cooperative game algorithm between multiple devices, through the utility function U i (t) optimization, finding the Nash equilibrium point so that the transmission power allocation of each device reaches a normal state;

[0104] S43. Use power allocation based on cooperative game algorithm to calculate the strategy of each device in the game and optimize power allocation:

[0105]

[0106] in, represents the optimal transmission power of device i under the current game;

[0107] S44, in order to obtain the optimal transmission power of each device After that, update the actual transmission power P of each device i (t) To enable all devices to operate in a coordinated manner in the overall system and to achieve normal power allocation and energy transmission under energy-constrained conditions;

[0108] S45. Utility function U of continuous monitoring equipment i (t) and the overall system utility. When a decrease in utility is detected, the cooperative game process is restarted to adjust the power allocation strategy of the devices.

[0109] In this implementation manner, the step S42 includes:

[0110] S421, construct a cooperative game model, in which each device i in the model acts as a participant in the game and selects a transmission power P i (t) as its strategy space A i The main decision variables of,the policy space are determined by the hardware capabilities of the,device and the current network conditions;

[0111] S422: For each device i, given the policy P of other devices j, j (t), optimize its utility function U i (t), determine the optimal strategy for device i

[0112]

[0113] in, Indicates the nonlinear relationship of energy consumption, ∈ is a small positive number to avoid the calculation problem of the logarithmic function when the denominator is zero, η is the coefficient used to adjust the dynamic impact of interference, and ν is an exponential parameter greater than 1, which is used to reflect the nonlinear impact of interference with power changes;

[0114] S424, through iterative calculation, each device adjusts the strategy P i (t) and Pj (t), until the Nash equilibrium point is reached and At this time, any device i cannot change its own policy P while other devices keep their policies unchanged. i (t) to further improve its utility function U i (t);

[0115] S425. At the Nash equilibrium point Under this condition, the transmission power P of each device is fixed i (t).

[0116] In this implementation, the step S5 includes:

[0117] S51, by monitoring key parameters in the network environment, including device location, link quality Q i (t), network load L i (t), equipment remaining energy E i (t) and the interference coefficient between devices I ij (t) collect this information in real time and identify changes in network environment and device status;

[0118] S52, when any of the above parameters is detected to have a significant change, triggering a prediction algorithm based on Bayesian reasoning to predict the trend of network condition changes in a future time period;

[0119] S53, based on the prediction results of the Bayesian inference algorithm, recalculate the target transmission power requirement P of each device i,target (t), in the process, the interference between devices, the nonlinear changes of network load and the dynamic adjustment of device status are considered, and the objective function J(t) is updated to reflect the predicted changes:

[0120]

[0121] Among them, log(1+(P i (t)-P i,target (t)) 2 ) introduces a nonlinear penalty for power deviation to better handle the impact of large deviations. is the link quality Q i (t) is a nonlinear processing, κ is an exponent greater than 1, which is used to adjust the sensitivity of link quality under different conditions, (L i (t)) λ Indicates network load L i (t), λ is an exponential parameter used to dynamically adjust the network load impact. This complicates the model of interference between devices. ν is an exponential parameter used to reflect the complexity of interference as power and distance change.

[0122] S54: According to the updated objective function J(t), the transmission power P of each device is i (t) Make real-time adjustments to adapt to new network conditions and device status;

[0123] S55. After completing the adjustment of the transmission power, the network environment and device status are continuously monitored. If changes in the environment and status are detected again, the above steps are repeated to maintain the adaptability of the system in a dynamic environment.

[0124] In this implementation manner, the step S52 includes:

[0125] S521, define a hypothesis set H = {H1, H2, ..., H n}, each hypothesis H i Represents a possible change in network status, including device location, link quality Q i (t), network load L i )t) and interference coefficient I ij (t) changes;

[0126] S522, through real-time monitoring, collect evidence set E = {E1, E2,, E m}, where each piece of evidence E j Represents changes and anomalies observed in the current network environment;

[0127] S523, use the Bayesian inference model to calculate each hypothesis H i In Evidence E j The posterior probability P(H i |E j );

[0128] S524, based on the real-time updated posterior probability P(H i |E j ), dynamically adjust the priority of each hypothesis and select the hypothesis H with the highest posterior probability best As the most likely change to the current network state;

[0129] S525, combined with the selected optimal hypothesis H best , calculate the changes in network parameters in the future time period and update the link quality Q i (t+Δt), network load L i (t+Δt), interference coefficient I ij (t+Δt) and the power demand P of the equipment i,target (t+Δt), where Δt is the prediction time interval;

[0130] S526: Use the prediction result to guide real-time power allocation adjustment.

[0131] Embodiment 1:

[0132] In Company B, a company located in China that focuses on the research and development of IoT devices, the adaptive power control method for energy transmission in heterogeneous wireless IoT of the present invention is deployed. Company B has multiple types of IoT devices, including smart sensors, data acquisition terminals and wireless communication modules, with an annual output of millions of units. These devices are widely used in smart cities, agricultural monitoring and industrial automation.

[0133] In this company, each IoT device requires a stable energy supply when working. Traditional power control methods are mainly based on fixed power settings, which are difficult to adapt to environmental changes in real time, resulting in serious energy waste of devices. Especially in a high-density device environment, mutual interference between devices will also significantly affect the stability of energy transmission. In order to solve this problem, Company B introduced the adaptive power control method described in this invention.

[0134] During the six-month implementation, Company B tested adaptive power control of energy transmission on approximately 100,000 IoT devices under different environmental conditions. These devices include applications in urban monitoring, agricultural irrigation, and industrial production lines, with complex and changing working environments. During the test, the average power demand of the devices ranged from 5W to 20W, and the communication link quality varied from 0.5 to 2.5 meters.

[0135] Table 1 Energy transmission effect comparison data table

[0136]

[0137] As shown in Table 1, before implementation, the average transmission power of each device was 15W, and after using the method of the present invention, the average transmission power was reduced to 10W, and the energy utilization efficiency was increased from 70% to 90%, with an increase of 28.57%. During the six-month implementation process, the total energy consumption was reduced from 1,500,000kWh to 900,000kWh, saving 600,000kWh of energy. At the same time, the interference rate between devices was reduced from 5% to 1%, significantly improving the stability of the system.

[0138] By implementing the adaptive power control method of the present invention, Company B significantly improved the energy utilization efficiency of the equipment, reduced energy waste, reduced interference between devices, and improved the stability of the entire system. In addition, the method also reduced the failure rate of the equipment from 3% to 0.5%, improving the reliability of the equipment and customer satisfaction.

[0139] In summary, the present invention combines adaptive power adjustment and prediction mechanisms, and can effectively solve the stability and efficiency issues of energy transmission in heterogeneous wireless Internet of Things through real-time monitoring and dynamic adjustment. The implementation of this method has laid a solid foundation for Company B to further expand its market in the field of intelligent Internet of Things in the future, and has strong promotion value.

[0140] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things, characterized in that: The steps include: S1. Collect network parameters and device status information in a heterogeneous wireless IoT environment. The network parameters include device location, communication link quality, and network load. The device status information includes the remaining energy and data transmission requirements of the device. S2. Preliminarily calculate the transmission power requirement of each device according to the network parameters and device status information, and generate a corresponding power allocation plan; S3, dynamically adjust the transmission power of each device according to the real-time collected network parameters and device status information; S4. Introduce cooperative game algorithm to coordinate power allocation among multiple devices and realize normal cooperation among devices under the condition of limited energy resources; S5. When changes in the network environment and device status are detected, a prediction algorithm based on Bayesian reasoning is used to predict the changing trend of network conditions in advance, and the power allocation plan is recalculated based on the prediction results, and then the transmission power of each device is adjusted in real time; S6. Continuously monitor the power control effect of each device, record the power consumption, transmission efficiency and communication interference of each device during the transmission process, and optimize the power control strategy based on these monitoring data; S7. Regularly update and store the optimized power control parameters.

2. The adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things according to claim 1, characterized in that: The S3 specifically includes: S31. Define the real-time transmission power of each device as P i (t), where i represents the device number and t represents the time point; S32. Determine the remaining energy E of each device based on the real-time status information of the device. i (t), and the quality indicator Q of the communication link i (t), and combined with the network load condition L i (t), preliminarily calculate the target transmission power requirement P of each device i，target (t); S33, the initially calculated target transmission power requirement P i，target (t) is used as input to dynamically adjust the actual transmission power P of each device i (t), the dynamic adjustment process is achieved by minimizing the following objective function: Where N is the total number of devices, α, β, and Y are parameters used to balance the weights, (P i (t)-P i，target (t)) 2 represents the square of the deviation of the transmission power, Indicates the inverse ratio of link quality, L i (t) indicates the current network load of the device; S34, by dynamically updating the transmission power adjustment strategy of each device, the reward mechanism is evaluated according to the difference between the actual transmission effect and the expected effect; S35: Finally output the actual transmission power P of each device at time point t i (t), and is used in the current wireless energy transfer process.

3. The adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things according to claim 2, characterized in that: The S33 specifically includes: S331, based on the real-time status information of each device, including the actual transmission power P i (t), link quality indicator Qi(t), network load condition L i (t), interference coefficient between devices I ij (t) and target transmission power requirement P i,target (t), construct the objective function J(t) for adaptive power adjustment: Among them, α, β, y, and 6 are coefficients for adjusting the actual transmission power, link quality index, network load, interference coefficient between devices, and target transmission power requirement weight. ij (t) is the interference coefficient between device i and device j, d ij is the distance between device i and device j; S332. Apply the Q-learning algorithm in reinforcement learning to determine the optimal transmission power adjustment strategy of the device according to the value of the objective function J(t). First, define the state space S, including the current transmission power P of the device. i (t), link quality Q i (t), network negative L i (t) and the interference coefficient between devices I ij (t) state variable, then define the action space A, and finally define the reward function R(t) as the negative value of the objective function J(t); S333. Perform iterative learning on the state space s and the action space A through the Q-learning algorithm, and update the Q value function Q(s, a) of each state-action pair: Q(s,a)←Q(s,a)+η[R(t)+γ·max a′ Q(s′,a′)-Q(s,a)]; Among them, η is the learning rate, γ is the discount factor, s′ represents the next state after executing action a, and max a′ Q(s′, a′) represents the maximum Q value of taking the best action in the next state s′; S334, in the learning process of the Q-learning algorithm, dynamically select the optimal action a in each state s * , adjust the actual transmission power P of the device i (t) Close to P i，target (t), and at the same time optimize the interference effect between devices, and finally achieve the purpose of optimizing the objective function J(t).

4. The adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things according to claim 1, characterized in that: The S4 specifically includes: S41. Define the utility function U of each device i (t), the utility function takes into account the energy demand of the equipment, the remaining energy E i (t), current transmission power P i (t) and interference with other devices: Among them, λ and μμ are adjustment coefficients, Indicates the effectiveness of the equipment's energy consumption, Indicates the interference between devices; S42, introduce a cooperative game algorithm between multiple devices, through the utility function U i (t) optimization, finding the Nash equilibrium point so that the transmission power allocation of each device reaches a normal state; S43. Use power allocation based on cooperative game algorithm to calculate the strategy of each device in the game and optimize power allocation: in, represents the optimal transmission power of device i under the current game; S44, in order to obtain the optimal transmission power of each device After that, update the actual transmission power P of each device i (t) To enable all devices to operate in a coordinated manner in the overall system and to achieve normal power allocation and energy transmission under energy-constrained conditions; S45. Utility function U of continuous monitoring equipment i (t) and the overall system utility. When a decrease in utility is detected, the cooperative game process is restarted to adjust the power allocation strategy of the devices.

5. The adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things according to claim 4, characterized in that: The step S42 includes: S421, construct a cooperative game model, in which each device i in the model acts as a participant in the game and selects a transmission power P i (t) is the main decision variable in its policy space Ai, the scope of which is determined by the hardware capabilities of the device and the current network conditions; S422: For each device i, given the policy P of other devices j, j (t), optimize its utility function U i (t), determine the optimal strategy for device i in, Indicates the nonlinear relationship of energy consumption, ∈ is a small positive number to avoid the calculation problem of the logarithmic function when the denominator is zero, η is the coefficient used to adjust the dynamic impact of interference, and ν is an exponential parameter greater than 1, which is used to reflect the nonlinear impact of interference with power changes; S424, through iterative calculation, each device adjusts the strategy P i (t) and P j (t), until the Nash equilibrium point is reached and At this time, any device i cannot change its own policy P while other devices keep their policies unchanged. i (t) to further improve its utility function U i (t); S425. At the Nash equilibrium point Under this condition, the transmission power P of each device is fixed i (t).

6. The adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things according to claim 1, characterized in that: The S5 step includes: S51, by monitoring key parameters in the network environment, including device location, link quality Qi(t), network load L i (t), equipment remaining energy E i (t) and the interference coefficient between devices I ij (t) collect this information in real time and identify changes in network environment and device status; S52, when any of the above parameters is detected to have a significant change, triggering a prediction algorithm based on Bayesian reasoning to predict the trend of network condition changes in a future time period; S53, based on the prediction results of the Bayesian inference algorithm, recalculate the target transmission power requirement P of each device i，target (t), in the process, the interference between devices, the nonlinear changes of network load and the dynamic adjustment of device status are considered, and the objective function J(t) is updated to reflect the predicted changes: Among them, log(1+(P i (t)-P i,target (t)) 2 ) introduces a nonlinear penalty for power deviation to better handle the impact of large deviations. is the link quality Q i (t) is a nonlinear processing, κ is an exponent greater than 1, which is used to adjust the sensitivity of link quality under different conditions, (L i (t)) λ Indicates network load L i (t), λ is an exponential parameter used to dynamically adjust the network load impact. This complicates the model of interference between devices. ν is an exponential parameter used to reflect the complexity of interference as power and distance change. S54: According to the updated objective function J(t), the transmission power P of each device is i (t) Make real-time adjustments to adapt to new network conditions and device status; S55. After completing the adjustment of the transmission power, the network environment and device status are continuously monitored. If changes in the environment and status are detected again, the above steps are repeated to maintain the adaptability of the system in a dynamic environment.

7. The adaptive power control method for energy transmission in a heterogeneous wireless Internet of Things according to claim 6, characterized in that: The step S52 includes: S521, define a hypothesis set H of network condition changes = {H1, H2, ..., H n }, each hypothesis H i Represents a possible change in network status, including device location, link quality Q i (t), network load L i (t) and interference coefficient I ij (t) changes; S522, through real-time monitoring, collect evidence set E = {E1, E2, E m }, where each piece of evidence E j Represents changes and anomalies observed in the current network environment; S523, use the Bayesian inference model to calculate each hypothesis H i In Evidence E j The posterior probability P(H i |E j ); S524, based on the real-time updated posterior probability P(H i |E j ), dynamically adjust the priority of each hypothesis and select the hypothesis H with the highest posterior probability best As the most likely change to the current network state; S525, combined with the selected optimal hypothesis H best , calculate the changes in network parameters in the future time period and update the link quality Q i (t+Δt), network load L i (t+Δt), interference coefficient I ij (t+Δt) and the power demand P of the equipment i,target (t+Δt), where Δt is the prediction time interval; S526: Use the prediction result to guide real-time power allocation adjustment.

Images