A method and system for information age minimization sensor power allocation

Abstract

The application provides a method and system for minimizing sensor power allocation of information age, and relates to the field of network optimization design. The method for minimizing sensor power allocation of information age comprises the following steps: constructing a wireless power supply sensor network composed of a wireless energy base station, a wireless sensor and a monitoring terminal; calculating the energy collected by the wireless sensor, the energy received by the monitoring terminal according to the energy collected by the wireless sensor, and the signal-to-interference ratio of the received signal through the energy received by the monitoring terminal; calculating the residual energy of the battery; calculating the information age at the wireless sensor and the information age at the monitoring terminal; and converting the optimization problem into a Markov decision problem for solving. The problems of building a wireless industrial internet of things driven by radio frequency energy collection technology and designing a sensor transmission scheme with minimized information age in order to simultaneously solve the sensor energy supply problem and meet the demand of information timeliness are solved.

Description

Technical Field

[0001] This invention relates to the field of network optimization design technology, specifically to a method and system for minimizing sensor power allocation based on information age. Background Technology

[0002] The Industrial Internet of Things (IIoT), with its powerful sensing and transmission capabilities, is considered a key technology for realizing Industry 4.0. In the IIoT, numerous sensors, controllers, actuators, and monitors are interconnected to establish intelligent industrial control systems. These systems can monitor, collect, exchange, and analyze industrial production data to achieve intelligent industrial control. Traditionally, network latency and throughput have been the primary network performance indicators for IIoT systems. In recent years, information timeliness has become a crucial indicator in emerging status update applications (including hazard monitoring, autonomous driving, and AR services), as outdated information can have serious consequences, such as production accidents, economic losses, and even personal injury. Information age (AoI), as a new network indicator, is used to characterize information timeliness. Defined as the time interval between the moment the system last received information and the current time, it has been proposed and widely studied in various system network scenarios. Therefore, AoI-based IIoT is an important research direction for improving the information timeliness of IIoT systems.

[0003] Meanwhile, in many Industrial Internet of Things (IIoT) systems, a large number of sensor nodes (SNs) are typically required to effectively monitor environmental indicators such as temperature, humidity, and hazardous substance levels. Since powering SNs via wires or batteries usually incurs relatively high deployment and maintenance costs, wireless energy harvesting (EH), which allows SNs to extract energy from radio frequency (RF) signals, is widely considered a promising low-power wireless power supply solution for SNs due to its advantages in controllability.

[0004] Therefore, in order to simultaneously solve the sensor energy supply problem and meet the demand for information timeliness, it is necessary to build a wireless industrial Internet of Things driven by radio frequency energy harvesting technology and design a sensor transmission scheme with minimal AoI. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides a method and system for minimizing sensor power allocation based on information age. This solves the problem of designing a sensor transmission scheme that minimizes information age in order to simultaneously address sensor energy supply issues and meet the timeliness requirements of information. This requires building a wireless industrial Internet of Things driven by radio frequency energy harvesting technology.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] In a first aspect, a method for minimizing sensor power allocation to minimize information age is provided, the method comprising:

[0010] A wireless power supply sensor network is constructed, consisting of a wireless power base station, wireless sensors, and a monitoring terminal. The wireless power base station is used to transmit radio frequency signals to the wireless sensors, which collect energy from the received radio frequency signals and use the collected energy to upload environmental data collected in the industrial environment to the monitoring terminal.

[0011] The energy collected by the wireless sensor is calculated based on the transmission power of the wireless energy base station. The energy received by the monitoring terminal is calculated based on the energy collected by the wireless sensor. The signal-to-dryness ratio of the received signal is then calculated based on the energy received by the monitoring terminal.

[0012] The wireless sensor nodes are battery powered. Each wireless sensor collects energy from the wireless energy base station and stores it in the battery for wireless information transmission and to calculate the remaining energy of the battery.

[0013] Calculate the information age at the wireless sensor and the information age at the monitoring terminal, respectively;

[0014] An optimization problem is established, the objective of which is to minimize the average information age while maximizing the remaining energy of the battery.

[0015] The optimization problem is transformed into a Markov decision problem for solution, and sensor power allocation is completed.

[0016] Preferably, the energy collected by the wireless sensor is calculated based on the transmission power of the wireless energy base station, specifically including:

[0017] The i-th sensor SN i Energy collected in time slot t

[0018]

[0019] Where 1≤i≤M, P WThe transmit power of the wireless energy base station is η, which is the energy harvesting coefficient related to the sensor energy receiving circuit, and τ represents the time slot length. Indicates the transition from WPS to SN i The path loss coefficient between them, h i (t) represents the transition from WPS to SN. i Small-scale fading coefficients of the inter-channel, This represents the maximum energy that can be collected in a single time slot of SN.

[0020] Preferably, the energy received by the monitoring terminal is calculated based on the energy collected by the wireless sensor, using the following formula:

[0021]

[0022] Where, x i The i-th sensor SN i The transmitted signal g represents the path loss coefficient from SNi to the monitoring terminal. i (t) indicates that from SN i The small-scale fading coefficient of the channel between the monitoring terminal and the monitoring terminal, P i (t) represents the SN in time slot t. i The average transmission power is given by n, which is additive white Gaussian noise with zero mean and unit variance.

[0023] Preferably, the signal-to-dryness ratio of the received signal is calculated by measuring the energy received from the monitoring terminal, using the following formula:

[0024]

[0025] To enable the monitoring terminal to decode SN i For the transmitted information, the received signal-to-interference-plus-noise ratio (SIN / N) needs to be greater than the decoding threshold γ. th , i.e. γ i ≥γ th .

[0026] Preferably, the calculation of the battery's remaining energy includes:

[0027] Let SN be in time slot t i The battery capacity is B i (t), then SN i The average transmission power in time slot t should satisfy P i (t)≤min{B i (t) / τ,P max}, where τ is the time slot length;

[0028] make For SN i Energy consumed in time slot t; due to SNi A battery with a capacity of Bmax, after energy harvesting and data transmission, starts at time slot t+1 with SN. i The remaining energy of the battery is

[0029] Preferably, the information age at the wireless sensor and the information age at the monitoring terminal are calculated separately, specifically including:

[0030] Information age at the wireless sensor: In time slot t, if a new data packet arrives at SN i In SN i The information is age. It will be reset to 1, otherwise it will be increased by one unit, as shown in the formula below:

[0031]

[0032] Where r i (t) = 1 indicates that a new data packet has arrived, r i (t) = 0 means no new data packet was received;

[0033] Information age at the monitoring terminal: In time slot t, if the data from SN is successfully decoded... i The data corresponds to the age of the information. It will drop to Otherwise, it will increase by one unit, as shown in the formula:

[0034]

[0035] ack i (t) = 1 indicates that the monitoring terminal successfully decoded the signal from SN. i Data, ack i (t) = 0 indicates decoding failure;

[0036] The system uses the long-term expected sum information (ESA) age. To characterize the real-time monitoring performance of an Industrial Internet of Things (IIoT) system, in It is the initial information age of the system.

[0037] Preferably, the optimization problem is expressed as:

[0038]

[0039] stπ=(P(1),P(2),…,P(T))

[0040] P i (t)≤min{B i (t) / τ,P max}, 1≤t≤T,

[0041]

[0042] Where π represents the power adjustment strategy of the sensor, P(t)=(P0(t),P1(t),…,P M (t) represents the power distribution vector of the system at time t.

[0043] Preferably, the optimization problem is transformed into a Markov decision problem for solution, specifically including:

[0044] Markov decision problems include state space, action space, system transition probabilities, and cost function;

[0045] Specifically, the state space is defined as follows: to effectively characterize the system state, the states of all SNs are combined as the system state; let SN in time slot t... i Status is That is, the information age, battery level, and serial number at the monitoring terminal. i The channel state to the monitoring terminal, in time slot t, the system state is s(t)=(s1(t),s2(t),…,s M (t));

[0046] The action space is specific: To solve the operation problem (OP), the system's action space is set as the optimization variable of the OP, i.e., the joint average transmission power of all SNs. The joint action of the system is a(t) = (a1(t), a2(t), ..., a M (t)), where a i (t), 1≤i≤M is SN i Transmission power in time slot t;

[0047] The specific system transition probability is defined as the probability of transitioning from state s(t) to state s(t+1) after taking a joint action a(t), i.e., P(s(t+1)|s(t),a(t)).

[0048] Specifically, the cost function is C(t) representing the cost C(t) to reach state s(t+1) after taking action a(t) in state s(t). To minimize ESA, the total information age of the terminals is used as the system cost.

[0049] Preferably, the sensor power allocation is completed, specifically including:

[0050] Construct 5 artificial neural networks, including a local policy network π. θ Two central Soft Q networks {Q φ1 Q φ2} and its corresponding target network {Qφ1′ Q φ2′}, where θ,φ1,φ2,φ1',φ2' are the network parameters corresponding to the above network; the local policy network only outputs the action of a single SN, and all SNs share the same policy network, but are distinguished by their identity vectors, that is, a i (t)=π θ (s i (t),id i ), where id i This is the one-hot vector corresponding to the wireless sensor index i;

[0051] For the central Soft Q network, the following loss function is used to update the network:

[0052]

[0053] Where I represents a batch of learning experiences sampled from the experience pool. a i ~π θ Indicates the SN from the local policy network i The action is sampled, a i Indicates SN i The action is 'a', where 'a' is the joint action that samples all SNs.

[0054] For the local policy network, its loss function is:

[0055]

[0056] Where α is the entropy regularization coefficient;

[0057] Finally, update the entropy regularization coefficient; its loss function is:

[0058]

[0059] Where K is the target entropy that the strategy expects to achieve.

[0060] Secondly, a system for minimizing sensor power allocation based on information age is provided, comprising:

[0061] The building module is used to construct a wireless power supply sensor network consisting of a wireless power base station, wireless sensors, and a monitoring terminal; wherein the wireless power base station is used to transmit radio frequency signals to the wireless sensors, the wireless sensors collect energy from the received radio frequency signals, and use the collected energy to upload environmental data collected in the industrial environment to the monitoring terminal.

[0062] The first calculation module is used to calculate the energy collected by the wireless sensor based on the transmission power of the wireless energy base station, calculate the energy received by the monitoring terminal based on the energy collected by the wireless sensor, and calculate the signal-to-dryness ratio of the received signal based on the energy received by the monitoring terminal.

[0063] The second computing module is used when the wireless sensor nodes are powered by batteries, and the energy collected by each wireless sensor from the wireless energy base station is stored in the battery for wireless information transmission and to calculate the remaining energy of the battery.

[0064] The third calculation module is used to calculate the information age at the wireless sensor and the information age at the monitoring terminal, respectively.

[0065] A module is established to establish an optimization problem, the objective of which is to minimize the average information age while maximizing the remaining energy of the battery.

[0066] The solution module is used to transform the optimization problem into a Markov decision problem for solution and to complete the sensor power allocation.

[0067] (III) Beneficial Effects

[0068] This invention discloses a method and system for minimizing sensor power allocation based on information age. In this network, multiple sensor nodes (SNs) are first powered by a wireless power station (WPS), then collect environmental data from the industrial environment, and finally transmit the environmental data to a monitor using energy harvested from radio frequency signals. To reduce industrial production costs, all SNs are allowed to share the same spectrum resources and simultaneously transmit their collected status updates. While this avoids strict synchronization requirements between SNs, interference between SNs can affect data transmission and cannot be ignored, impacting the real-time performance of industrial environmental monitoring. Therefore, this patent minimizes the long-term average information age by adjusting the power allocation among sensor nodes to improve data transmission from the SNs to the monitoring end, ultimately enhancing the real-time monitoring performance of the Industrial Internet of Things (IIoT). Attached Figure Description

[0069] Figure 1 This is a flowchart of the method for minimizing sensor power allocation based on the age of the information in this invention;

[0070] Figure 2 This is a diagram of a wireless power supply sensor network in an embodiment of the present invention;

[0071] Figure 3 This is a diagram illustrating the evolution of information age in an embodiment of the present invention;

[0072] Figure 4 This is a framework diagram of the multi-agent flexible actor-critic sensor power algorithm in an embodiment of the present invention. Detailed Implementation

[0073] The technical solutions in the embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0074] Example

[0075] like Figure 1 As shown, one embodiment of the present invention provides a method for minimizing sensor power allocation based on information age, the method comprising:

[0076] A wirelessly powered sensor network is constructed, consisting of wireless power base stations, wireless sensors, and monitoring terminals. The wireless power base stations transmit radio frequency signals to the wireless sensors, which collect energy from the received signals and use this energy to upload environmental data gathered in the industrial environment to the monitoring terminal. Figure 2 The illustrated industrial IoT monitoring scenario driven by RF energy harvesting technology includes a wireless power station (WPS), M wireless sensors, and one monitoring terminal. To reduce the cost of manual battery replacement in industrial IoT scenarios, the WPS is deployed to send RF signals to charge the wireless sensors. Each wireless sensor is equipped with an energy harvesting circuit module, a battery with a maximum capacity of Bmax joules, and an energy receiving antenna and an information transmitting antenna. It can harvest energy from the received RF signals and then use the harvested energy to upload environmental data collected in the industrial environment to the monitoring terminal.

[0077] The energy collected by the wireless sensors is calculated based on the transmission power of the wireless power base station. The energy received by the monitoring terminal is calculated based on the energy collected by the wireless sensors. The signal-to-dryness ratio of the received signal is calculated based on the energy received by the monitoring terminal. In order to provide a continuous power supply to the wireless sensors, WPS continuously broadcasts radio frequency signals to them.

[0078] The wireless sensor nodes are battery powered. Each wireless sensor collects energy from the wireless power base station and stores it in the battery for wireless information transmission and to calculate the remaining energy in the battery.

[0079] The information age at the wireless sensor and the information age at the monitoring terminal are calculated separately. In industrial applications, once the system's AoI reaches the application's maximum tolerance value, it means that no new data has been received for a long time, and the monitored environmental data has become outdated. Therefore, continuing to count the AoI is not very meaningful. Therefore, using A... max This represents the maximum tolerance value for AoI. Therefore, the AoI at both the sensor and the monitoring terminal should not exceed A. max ;

[0080] An optimization problem is established, the objective of which is to minimize the average information age while maximizing the remaining energy of the battery.

[0081] The optimization problem is transformed into a Markov decision problem for solution, and sensor power allocation is completed.

[0082] Furthermore, the energy collected by the wireless sensor is calculated based on the transmission power of the wireless energy base station, specifically including:

[0083] The i-th sensor SN i Energy collected in time slot t

[0084]

[0085] Where 1≤i≤M, P W The transmit power of the wireless energy base station is η, which is the energy harvesting coefficient related to the sensor energy receiving circuit, and τ represents the time slot length. Indicates the transition from WPS to SN i The path loss coefficient between them, h i (t) represents the transition from WPS to SN. i Small-scale fading coefficients of the inter-channel, This represents the maximum energy that can be collected in a single time slot of SN.

[0086] Furthermore, the energy received by the monitoring terminal is calculated based on the energy collected by the wireless sensors, using the following formula:

[0087]

[0088] Where, x i The i-th sensor SN i The transmitted signal g represents the path loss coefficient from SNi to the monitoring terminal. i (t) indicates that from SN i The small-scale fading coefficient of the channel between the monitoring terminal and the monitoring terminal, P i (t) represents the SN in time slot t. i The average transmission power is given by n, which is additive white Gaussian noise with zero mean and unit variance.

[0089] Furthermore, the signal-to-dryness ratio of the received signal is calculated by monitoring the energy received by the terminal, using the following formula:

[0090]

[0091] To enable the monitoring terminal to decode SN i For the transmitted information, the received signal-to-interference-plus-noise ratio (SIN / N) needs to be greater than the decoding threshold γ. th , i.e. γ i ≥γ th .

[0092] Furthermore, the calculation of the battery's remaining energy includes:

[0093] Let SN be in time slot t i The battery capacity is B i (t), then SN i The average transmission power in time slot t should satisfy P i (t)≤min{B i (t) / τ,P max}, where τ is the time slot length;

[0094] make For SN i Energy consumed in time slot t; due to SN i A battery with a capacity of Bmax, after energy harvesting and data transmission, starts at time slot t+1 with SN. i The remaining energy of the battery is

[0095] An example diagram of AoI evolution is shown below. Figure 2 As shown, further, the information age at the wireless sensor and the information age at the monitoring terminal are calculated separately, specifically including:

[0096] Information age at the wireless sensor: In time slot t, if a new data packet arrives at SN i In SN i The information is age. It will be reset to 1, otherwise it will be increased by one unit, as shown in the formula below:

[0097]

[0098] Where r i (t) = 1 indicates that a new data packet has arrived, r i (t) = 0 means no new data packet was received;

[0099] Information age at the monitoring terminal: In time slot t, if the data from SN is successfully decoded... iThe data corresponds to the age of the information. It will drop to Otherwise, it will increase by one unit, as shown in the formula:

[0100]

[0101] ack i (t) = 1 indicates that the monitoring terminal successfully decoded the signal from SN. i Data, ack i (t) = 0 indicates decoding failure;

[0102] The system uses the long-term expected sum information (ESA) age. To characterize the real-time monitoring performance of an Industrial Internet of Things (IIoT) system, in It is the initial information age of the system.

[0103] like Figure 3 As shown, AoI decreases when new data is successfully transmitted. In other words, efficient data transmission can effectively reduce ESA. However, due to interference between SNs and the uncertainty of battery power, data from all SNs is difficult to be decoded by the monitoring terminal simultaneously within a single time slot. Therefore, to improve data transmission efficiency and further enhance data timeliness, an efficient power allocation strategy is needed. Therefore, this patent aims to minimize the system's ESA and constructs an optimization problem (OP) with the power strategy as the optimization variable; further, the optimization problem is expressed as:

[0104]

[0105] stπ=(P(1),P(2),…,P(T))

[0106] P i (t)≤min{B i (t) / τ,P max}, 1≤t≤T,

[0107]

[0108] Where π represents the power adjustment strategy of the sensor, P(t)=(P0(t),P1(t),…,P M (t) represents the power distribution vector of the system at time t.

[0109] However, directly solving the OP problem is very difficult because the OP is a non-convex problem, and the objective function... There is no explicit expression between the optimization variable π and the monitor. On the other hand, since the monitor and SN do not know the prior information about the channel state, they cannot know the future channel state, making it difficult to achieve the minimum ESA through power adjustment in a single time slot.

[0110] Due to the powerful ability of deep reinforcement learning to solve complex non-convex problems, this patent develops a method based on multi-agent deep reinforcement learning to efficiently solve operational problems. Before utilizing deep reinforcement learning, the original problem needs to be transformed into a Markov decision problem (MDP).

[0111] MDP modeling comprises four parts: state space, action space, system transition probabilities, and cost function, as detailed below; the optimization problem is transformed into a Markov decision problem for solution, specifically including:

[0112] Markov decision problems include state space, action space, system transition probabilities, and cost function;

[0113] Specifically, the state space is defined as follows: to effectively characterize the system state, the states of all SNs are combined as the system state; let SN in time slot t... i Status is That is, the information age, battery level, and serial number at the monitoring terminal. i The channel state to the monitoring terminal, in time slot t, the system state is s(t)=(s1(t),s2(t),…,s M (t));

[0114] The action space is specific: To solve the operation problem (OP), the system's action space is set as the optimization variable of the OP, i.e., the joint average transmission power of all SNs. The joint action of the system is a(t) = (a1(t), a2(t), ..., a M (t)), where a i (t), 1≤i≤M is SN i Transmission power in time slot t;

[0115] The specific system transition probability is defined as the probability of transitioning from state s(t) to state s(t+1) after taking a joint action a(t), i.e., P(s(t+1)|s(t),a(t)).

[0116] Specifically, the cost function is C(t) representing the cost C(t) to reach state s(t+1) after taking action a(t) in state s(t). To minimize ESA, the total information age of the terminals is used as the system cost.

[0117] The learning framework for a multi-agent flexible actor-critic sensor power allocation scheme is as follows: Figure 4 As shown, further steps include completing sensor power allocation, specifically including:

[0118] To enable SNs to rationally adjust transmission power, improve the real-time monitoring performance in industrial IoT systems, and reduce network communication overhead, this patent proposes an intelligent agent flexible actor-commentator sensor power allocation scheme. Through a framework of centralized training and distributed execution, each SN is treated as an intelligent agent, and it learns the optimal strategy to adjust power based on its local observation state without the need for large-scale network interaction.

[0119] In the learning framework for constructing a multi-agent flexible actor-critic sensor power scheme, five artificial neural networks are built, including a local policy network π. θ Two central Soft Q networks {Q φ1 Q φ2} and its corresponding target network {Q φ1′ Q φ2′}, where θ,φ1,φ2,φ1',φ2' are the network parameters corresponding to the above network; the local policy network only outputs the action of a single SN, and all SNs share the same policy network, but are distinguished by their identity vectors, i.e., a i (t)=π θ (s i (t),id i ), where id i This is the one-hot vector corresponding to the wireless sensor index i;

[0120] For the central Soft Q network, the following loss function is used to update the network:

[0121]

[0122] Where I represents a batch of learning experiences sampled from the experience pool. a i ~π θ Indicates the SN from the local policy network i The action is sampled, a i Indicates SN i The action is 'a', where 'a' is the joint action that samples all SNs.

[0123] For the local policy network, its loss function is:

[0124]

[0125] Where α is the entropy regularization coefficient;

[0126] Finally, update the entropy regularization coefficient; its loss function is:

[0127]

[0128] Where K is the target entropy that the strategy expects to achieve.

[0129] Another embodiment of the present invention provides a system for minimizing sensor power allocation based on information age, comprising:

[0130] The module is used to build a wireless power supply sensor network consisting of a wireless power base station, wireless sensors, and a monitoring terminal. The wireless power base station is used to transmit radio frequency signals to the wireless sensors. The wireless sensors collect energy from the received radio frequency signals and use the collected energy to upload environmental data collected in the industrial environment to the monitoring terminal.

[0131] The first calculation module is used to calculate the energy collected by the wireless sensor based on the transmission power of the wireless energy base station, calculate the energy received by the monitoring terminal based on the energy collected by the wireless sensor, and calculate the signal-to-dryness ratio of the received signal based on the energy received by the monitoring terminal.

[0132] The second computing module is used when the wireless sensor nodes are battery powered. Each wireless sensor collects energy from the wireless energy base station and stores it in the battery for wireless information transmission, and calculates the remaining energy of the battery.

[0133] The third calculation module is used to calculate the information age at the wireless sensor and the information age at the monitoring terminal, respectively.

[0134] A module is established to formulate an optimization problem. The objective of the optimization problem is to minimize the average information age while maximizing the remaining energy of the battery.

[0135] The solution module is used to transform the optimization problem into a Markov decision problem for solution and to complete the sensor power allocation.

[0136] Embodiments of this application may be provided as methods or computer program products. Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application may be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0137] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0138] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0139] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0140] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A method for minimizing sensor power allocation to minimize information age, characterized in that, include: Construct a wireless power supply sensor network consisting of wireless power base stations, wireless sensors, and monitoring terminals; The wireless energy base station is used to transmit radio frequency signals to wireless sensors. The wireless sensors collect energy from the received radio frequency signals and use the collected energy to upload environmental data collected in the industrial environment to the monitoring terminal. The energy collected by the wireless sensor is calculated based on the transmission power of the wireless energy base station. The energy received by the monitoring terminal is calculated based on the energy collected by the wireless sensor. The signal-to-dryness ratio of the received signal is then calculated based on the energy received by the monitoring terminal. The wireless sensor nodes are battery powered. Each wireless sensor collects energy from the wireless energy base station and stores it in the battery for wireless information transmission and to calculate the remaining energy of the battery. The information age at the wireless sensor and the information age at the monitoring terminal are calculated separately, specifically including: Information age at wireless sensors: in time slots In the middle, if a new data packet arrives at SN i In SN i The information is age. It will be reset to 1, otherwise it will be increased by one unit, as shown in the formula below: in This indicates that a new data packet has arrived. This indicates that no new data packets have been received. Information age at the monitoring terminal: in time slots In the middle, if the decoding from SN is successful i The data corresponds to the age of the information. It will drop to Otherwise, it will increase by one unit. The formula is: in This indicates that the monitoring terminal has successfully decoded the signal from SN. i Data, This indicates decoding failure; The system uses the long-term expected sum information (ESA) age. To characterize the real-time monitoring performance of an Industrial Internet of Things (IIoT) system, ,in It is the initial information age of the system; An optimization problem is established, the objective of which is to minimize the average information age while maximizing the remaining energy of the battery. The optimization problem is transformed into a Markov decision problem for solution, and sensor power allocation is completed.

2. The method for minimizing sensor power allocation based on information age according to claim 1, characterized in that: The energy collected by the wireless sensor is calculated based on the transmit power of the Wireless Power Base Station (WPS), specifically including: The i-th sensor SN i In the time slot Energy collected above in, , This refers to the transmission power of the wireless power base station. It is the energy harvesting coefficient related to the sensor's energy receiving circuit. Indicates the time slot length. Indicates from WPS to SN i Path loss coefficient between Indicates from WPS to SN i Small-scale fading coefficients of the inter-channel, This represents the maximum energy that can be collected in a single time slot of SN.

3. The method for minimizing sensor power allocation based on information age according to claim 2, characterized in that: The energy received by the monitoring terminal is calculated based on the energy collected by the wireless sensors, using the following formula: in, The i-th sensor SN i The transmitted signal This represents the path loss coefficient from SNi to the monitoring terminal. Indicates from SN i Small-scale fading coefficient of the channel between the monitoring terminal and the monitoring terminal. In order to be in SN in time slot i Average transmission power, It is additive white Gaussian noise with zero mean and unit variance.

4. The method for minimizing sensor power allocation based on information age according to claim 3, characterized in that: The signal-to-dryness ratio of the received signal is calculated based on the energy received by the monitoring terminal, using the following formula: To enable the monitoring terminal to decode SN i For the transmitted information, the received signal-to-interference-plus-noise ratio (SIN / N) needs to be greater than the decoding threshold. ,Right now .

5. The method for minimizing sensor power allocation based on information age according to claim 4, characterized in that: The calculation of the battery's remaining energy includes: Order in SN in time slot i The battery capacity is SN i exist The average transmission power in the time slot should meet the following requirements. ,in The time slot length; make For SN i In the time slot Energy consumed in the process; due to SN i Equipped with a battery of capacity Bmax, after energy harvesting and data transmission, in the time slot... SN at the beginning i The remaining energy of the battery is .

6. The method for minimizing sensor power allocation based on information age according to claim 5, characterized in that: The optimization problem is expressed as: in, This indicates the power adjustment strategy of the sensor. express The power allocation vector of the system at any given time.

7. The method for minimizing sensor power allocation based on information age according to claim 6, characterized in that: The optimization problem is transformed into a Markov decision problem for solution, specifically including: Markov decision problems include state space, action space, system transition probabilities, and cost function; Specifically, the state space is defined as follows: to effectively characterize the system state, the states of all SNs are combined as the system state; let the state in the time slot be... SN i Status is This refers to the information age, battery level, and serial number displayed on the monitoring terminal. i The channel status to the monitoring terminal, in the time slot The state of the system is ; The action space is specific: in order to solve The system's action space is set to The optimization variable, namely the joint average transmission power of all SNs, is the joint action of the system. ,in For SN i exist Transmission power over a time slot; The specific system transition probability is defined as the probability of taking joint actions. After that, from the state Transition to state The probability, i.e. ; The specific cost function: The cost function is defined in state... Take action below Later state Cost To minimize ESA, the total information age of the terminals is used as the system cost. 。 8. A system for minimizing the power allocation of a sensor based on information age, characterized in that, include: The building module is used to construct a wireless power supply sensor network consisting of wireless power base stations, wireless sensors, and monitoring terminals. The wireless energy base station is used to transmit radio frequency signals to wireless sensors. The wireless sensors collect energy from the received radio frequency signals and use the collected energy to upload environmental data collected in the industrial environment to the monitoring terminal. The first calculation module is used to calculate the energy collected by the wireless sensor based on the transmission power of the wireless energy base station, calculate the energy received by the monitoring terminal based on the energy collected by the wireless sensor, and calculate the signal-to-dryness ratio of the received signal based on the energy received by the monitoring terminal. The second computing module is used when the wireless sensor nodes are powered by batteries, and the energy collected by each wireless sensor from the wireless energy base station is stored in the battery for wireless information transmission and to calculate the remaining energy of the battery. The third calculation module is used to calculate the information age at the wireless sensor and the information age at the monitoring terminal, respectively, and specifically includes: Information age at wireless sensors: in time slots In the middle, if a new data packet arrives at SN i In SN i The information is age. It will be reset to 1, otherwise it will be increased by one unit, as shown in the formula below: in This indicates that a new data packet has arrived. This indicates that no new data packets have been received. Information age at the monitoring terminal: in time slots In the middle, if the decoding from SN is successful i The data corresponds to the age of the information. It will drop to Otherwise, it will increase by one unit. The formula is: in This indicates that the monitoring terminal has successfully decoded the signal from SN. i Data, This indicates decoding failure; The system uses the long-term expected sum information (ESA) age. To characterize the real-time monitoring performance of an Industrial Internet of Things (IIoT) system, ,in It is the initial information age of the system; A module is established to establish an optimization problem, the objective of which is to minimize the average information age while maximizing the remaining energy of the battery. The solution module is used to transform the optimization problem into a Markov decision problem for solution and to complete the sensor power allocation.

Images