A transmission decision method for error information age
Patent Information
- Application Number
- CN202611434704.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-16
- Publication Date
- 2026-10-30
AI Technical Summary
然而，当无线信道具有时间相关性，且仅能通过发送后的确认（Acknowledgment, ACK）或否定确认（Negative Acknowledgment, NACK）反馈获知传输时隙的信道状态时，系统的状态信息同时包含无界的AoII和连续的信道信念，导致传输决策的求解较为复杂
[0009]本发明与现有技术相比，其显著优点为：利用ACK、NACK反馈和信道演化，对原始模型的可达信道信念进行精确的两参数离散表示，无需均匀划分信念区间；通过双截断获得适于数值求解的有限状态模型，并利用模型信念及信道时间相关性进行传输决策。在所测试的参数配置下，所得方法的平均AoII低于所述贪心调度方法，长期平均传输率约束较紧时优势更明显。
Smart Images
Figure CN123019929A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of wireless communication and real-time status monitoring technology, specifically a method for making transmission decisions based on the age of erroneous information under the conditions that the channel state cannot be directly observed and the long-term average transmission rate is constrained, using an accurate two-parameter discrete representation of channel belief and a double-truncation finite state model. Background Technology
[0002] In IoT monitoring, networked control, and remote state estimation systems, the receiving end needs to promptly grasp the true state of the state source. Traditional information age describes the freshness of received information but cannot directly reflect the duration of estimation errors at the receiving end; instantaneous error rate can describe whether an error is currently occurring but cannot characterize the duration of the error. Error information age (AoII), combining state inconsistency and inconsistency duration, is used to evaluate the effectiveness of real-time monitoring information.
[0003] Existing AoII scheduling research mainly focuses on scenarios where the channel state is known or easily obtainable, while some studies have addressed transmission decisions under incomplete channel information. However, when the wireless channel is time-dependent and the channel state of the transmission slot can only be obtained through post-transmission acknowledgments (ACKs) or negative acknowledgments (NACKs), the system's state information simultaneously includes unbounded AoII and continuous channel beliefs, leading to a complex solution for transmission decisions. Furthermore, greedy scheduling methods that select actions based on the current AoII and historical average transmission rate do not fully utilize the time-dependent channel state and channel belief information. Therefore, researching an AoII transmission decision-making method suitable for scenarios where the channel state is not directly observable and is constrained by long-term average transmission rates is of great significance. Summary of the Invention
[0004] This invention aims to provide a transmission decision-making method for error information age. It accurately represents the reachability belief of the original model using the most recently observed channel state and observation interval. Then, it truncates AoII and the observation interval to construct a finite-state model to derive a transmission decision strategy, which is then executed in actual operation. The technical solution to achieve this objective is as follows: Given the inconsistency process transition parameters between the source state and the receiver's estimated state, channel transition parameters, long-term average transmission rate constraints, AoII upper limit, observation interval upper limit, and the initial channel state known before the initial decision, the initial AoII and slot count are set to 0, and the following steps are performed: Step 1: Based on the most recent known channel state and observation interval, perform an accurate two-parameter discrete representation of the reachability belief in the model; then truncate AoII and the observation interval to construct the state set and transition probability of the finite state model.
[0005] Step 2: Based on the finite state model, a deterministic strategy is obtained by using transmission penalty, relative value iteration and multiplier binary search; when the deterministic strategy cannot directly satisfy the long-term average transmission rate constraint, state-by-state randomization is performed on the high transmission rate endpoint strategy and the low transmission rate endpoint strategy, and a randomization factor that satisfies the constraint is selected.
[0006] Step 3: At the beginning of each time slot, read the current truncation AoII and truncation model belief, select the send or silent action according to the strategy, and obtain ACK, NACK or no feedback at the end of the time slot.
[0007] Step 4: Obtain the actual AoII before the next time slot decision and truncate it, and update the channel state observation, observation interval and truncation model belief according to the feedback.
[0008] Step 5: Update the timeslot count; if the monitoring task has not been terminated, return to step 3 to continue execution; if the monitoring task is terminated, end the transmission decision process.
[0009] Compared with existing technologies, the significant advantages of this invention are: by utilizing ACK and NACK feedback and channel evolution, the reachability belief of the original model is accurately represented by two parameters, without the need for uniformly dividing the belief interval; a finite-state model suitable for numerical solution is obtained through double truncation, and transmission decisions are made using the model belief and channel time correlation. Under the tested parameter configuration, the average AoII of the obtained method is lower than that of the greedy scheduling method, and the advantage is more obvious when the long-term average transmission rate constraint is tight. Attached Figure Description
[0010] Figure 1 This is a schematic diagram illustrating an application scenario of the present invention; Figure 2 This is a flowchart of the method for making transmission decisions based on the age of erroneous information according to the present invention; Figure 3-4 This is a performance comparison chart of the proposed method and the greedy scheduling method under different channel transfer parameters and different inconsistent process transfer parameters, and different long-term average transmission rate constraints. Detailed Implementation
[0011] like Figure 1As shown, this embodiment includes a state source, a transmitter, and a receiver. The transmitter is equipped with a decision control module for performing transmission decisions. The transmitter observes the actual state of the state source, and the decision control module outputs a transmit or remain silent action based on the current AoII and the pre-decision channel belief.
[0012] This invention divides the time axis into equal-length time slots and sets... Represents the time slot index, where Represents the set of non-negative integers. Source real state. and receiver estimated state All values are taken from Inconsistent state indication exist Take 1 at time Set the time to 0. Send action. Setting 1 indicates sending, setting 0 indicates silence; channel state. A value of 1 indicates on, and a value of 0 indicates off. When a state update transmission fails, the inconsistency between the source state and the receiver's estimated state is a two-state Markov chain, with the following transition parameters: and The state evolution process of the state source is independent of the channel state evolution process, and the channel transition probability is... ,in Assume that the ACK and NACK feedbacks are error-free and available before the next time slot decision. Indicates the cutoff time slot The slot number where the most recent actual state matches the receiver's estimate, among which For historical time slot index, time slot The initial definition of AoII is: (1) in, , , , .
[0013] Figure 2 This is a flowchart of the method for making transmission decisions based on the age of erroneous information according to the present invention. Figure 2 As shown, a method for making transmission decisions based on the age of erroneous information is implemented according to the following steps: Step 1: Construct a discretized finite state model.
[0014] Step 1-1: Set Time Slots The channel observation history prior to the initial decision is ,in Let 1, -1, and 0 represent ACK, NACK, and no feedback, respectively. The pre-decision channel belief of the model is: , For the channel one-step prediction operator, This is based on its stable belief. The slot number corresponding to the most recently known channel state is... ,set up , ,in, Indicates the most recently known channel state. This indicates that the channel is open. This indicates that the channel is off; Indicates the time slot from the most recent known channel state. up to the current decision slot The observation interval is the number of time slots elapsed since the most recent channel state observation. It is initially set... , , express of Secondary compound, The belief that a channel is reachable can be expressed as: (2) This parameter indicates accuracy for the original model, thus eliminating the need for uniform partitioning. Furthermore, for feasible action-feedback combinations, the original model's channel beliefs are updated before the decision in the next time slot as follows: (3) in , Therefore, the next time slot channel belief corresponding to the ACK feedback is... The next time slot channel belief corresponding to the NACK feedback is .
[0015] Step 1-2: Given an integer and , As the upper limit of AoII, As the upper limit of the observation interval, the truncated AoII is defined as: (4) The truncated set of beliefs is defined as , , A finite set of states is defined as .use This indicates that the original model is in the time slot. Precise channel belief, using This indicates the truncation model in the time slot. The belief, with This indicates its current value.
[0016] Steps 1-3: Truncate the mapping by observation interval Updated channel beliefs after a silent action: (5) Steps 1-4: In the given and hour, conditional probability is used This indicates that the transition probability corresponding to the ACK feedback is: (6) The transition probability corresponding to NACK or no feedback is: (7) Steps 1-5: Settings , Let these be the current finite state and the next finite state, respectively, where the finite state transition probabilities are: (8) in, This is an indicator function that takes the value 1 if the condition is true and 0 otherwise.
[0017] Step 2: Solve for the transmission decision strategy constrained by the transmission rate.
[0018] Step 2-1: Given a long-term average transmission rate constraint and Lagrange multipliers , finite state and actions The single-slot cost is defined as follows: ,from Begin executing relative value iteration: (9) Step 2-2: Select a fixed reference state Normalize the relative value function: (10) When the relative value iteration satisfies the preset convergence condition, the minimization action obtained from the last iteration is determined as the deterministic strategy. .
[0019] Steps 2-3: Evaluate the long-term average transmission rate of strategies corresponding to different Lagrange multipliers, and update the Lagrange multipliers using binary search: (11) in, and These represent the long-term average transmission rate of the corresponding strategy being higher than and lower than the constraint value, respectively. The endpoints of the Lagrange multipliers are determined. The median of the two is taken as the new Lagrange multiplier, and the corresponding endpoints are updated according to the long-term average transmission rate of the strategy corresponding to the median multiplier.
[0020] Steps 2-4: When the deterministic strategy cannot directly satisfy the long-term average transmission rate constraint, select the high transmission rate endpoint strategy as... and low-transmission-rate endpoint strategy Then, state-by-state randomization is performed to obtain the randomization strategy. : (12) Among them, probability Select low-transmission-rate endpoint strategy With probability Select high-transmission-rate endpoint strategy To obtain the randomization strategy The long-term average transmission rate of the randomization strategy is evaluated, and the randomization factor that satisfies the long-term average transmission rate constraint and whose transmission rate is closest to the constraint value is selected. .
[0021] Step 3: In each time slot Initially, the decision control module in the transmitting end reads the current finite state. The strategy obtained in step 2 outputs the action. , The sending end sends a current status update at any time. The sending end remains silent and receives feedback at the end of the time slot. ,in and hour , and hour , hour .
[0022] Step 4: Update the state of the next time slot.
[0023] Step 4-1: In the time slot Obtain actual erroneous information about age before initial decision-making According to equation (4), the truncated updated version is obtained. .
[0024] Step 4-2: Settings , They represent time slots respectively. The initial most recent observation channel state and observation interval, when season , , , season , , .
[0025] Step 4-3: When and , At that time, keep Unchanged, settings , ;when and When, set This indicates that after the observation interval reaches its upper limit, the channel belief will be mapped to a stationary belief; when and Maintain a stable mindset until an ACK or NACK response is received, then reset as per step 4-2.
[0026] Step 5: Use the number of runtime slots as the termination condition for the monitoring task, preset as a positive integer. This is the total number of time slots for this run. After the next time slot status update is completed, the time slot number will be incremented by 1. Return to step 3. This run will end at this time. Used to limit the duration of this run.
[0027] This invention implements the method using MATLAB software. In the numerical embodiment, an upper limit for the observation interval is set. AoII upper limit Simulation time Time slot.
[0028] Figure 3-4 With other parameters remaining constant, the channel transfer parameters and the inconsistency process transfer parameters were changed sequentially. The results showed that the method proposed in this invention maintained an average AOII performance lower than that of the greedy scheduling method under different long-term average transmission rate constraints, thus verifying the effectiveness of this invention.
Claims
1. A transmission decision-making method for Age of Incorrect Information (AoII), applied to a single-source real-time monitoring system including a state source, a transmitter, a receiver, and a hidden two-state Markov channel, wherein the transmitter includes a decision control module for performing transmission decisions, the decision control module obtains the current AoII and the channel belief before the decision in each time slot, and obtains the channel state observation result of the transmission time slot through feedback of acknowledgment (ACK) or negative acknowledgment (NACK) after the transmission state update, with the goal of reducing the long-term average AoII and satisfying the long-term average transmission rate constraint, characterized in that... Given the inconsistency process transition parameters between the source state and the estimated receiver state, channel transition parameters, long-term average transmission rate constraints, AoII upper limit, observation interval upper limit, and the initial channel state known before the initial decision, and setting the initial AoII and slot count to 0, perform the following steps: Step 1: Based on the most recent known channel state and observation interval, perform an accurate two-parameter discrete representation of the reachability belief in the model; then truncate AoII and the observation interval to construct the state set and transition probabilities of the finite state model; Step 2: Based on the finite state model, a deterministic strategy is adopted using transmission penalty, relative value iteration, and multiplier binary search; When the deterministic strategy cannot directly satisfy the long-term average transmission rate constraint, state-by-state randomization is performed on the high transmission rate endpoint strategy and the low transmission rate endpoint strategy, and a randomization factor that satisfies the constraint is selected. Step 3: At the beginning of each time slot, read the current truncated AoII and truncated model belief, select the send or silent action according to the strategy, and obtain ACK, NACK or no feedback at the end of the time slot; Step 4: Obtain the actual AoII before the next time slot decision and truncate it; update the channel state observation, observation interval and truncation model belief according to the feedback. Step 5: Update the timeslot count; if the monitoring task has not been terminated, return to step 3 to continue execution; if the monitoring task is terminated, end the transmission decision process.
2. The transmission decision method for age-based erroneous information according to claim 1, characterized in that, This invention divides the time axis into equal-length time slots and sets... Represents the time slot index, where Represents the set of non-negative integers; source true state and receiver estimated state All values are taken from Inconsistent state indication exist Take 1 at time Take 0 at time; send action Setting 1 indicates sending, setting 0 indicates silence; channel state. Setting 1 indicates on, and setting 0 indicates off; when a state update is not successfully transmitted, the inconsistency between the source state and the estimated state at the receiver is a two-state Markov chain, with the following transition parameters: and The state evolution process of the state source is independent of the channel state evolution process, and the channel transition probability is... ,in Assuming ACK and NACK feedback are error-free and available before the next time slot decision; set Indicates the cutoff time slot The slot number where the most recent actual state matches the receiver's estimate, among which For historical time slot index, time slot The initial definition of AoII is: (1) in, , , , .
3. The transmission decision method for age-based erroneous information according to claim 1, characterized in that, Step 1 includes performing a two-parameter discrete representation of the reachability channel belief in the original model, and truncating the AoII and observation interval. Specific settings include: Step 1-1: Set Time Slots The channel observation history prior to the initial decision is ,in Let 1, -1, and 0 represent ACK, NACK, and no feedback, respectively; the model's pre-decision channel belief is... , For the channel one-step prediction operator, Its stable belief; the slot number corresponding to the most recently known channel state is ,set up , ,in, Indicates the most recently known channel state. This indicates that the channel is open. This indicates that the channel is off; Indicates the time slot from the most recent known channel state. up to the current decision slot The observation interval is the number of time slots elapsed since the most recent channel state observation; initially set... , , express of Secondary composite, The belief that a channel is reachable can be expressed as: (2) This parameter indicates accuracy for the original model, thus eliminating the need for uniform partitioning. Furthermore, for feasible combinations of actions and feedback, the original model's channel beliefs are updated before the decision in the next time slot as follows: (3) in , Therefore, the channel belief for the next time slot corresponding to the ACK feedback is: The next time slot channel belief corresponding to the NACK feedback is ; Step 1-2: Given an integer and , As the upper limit of AoII, As the upper limit of the observation interval, the truncated AoII is defined as: (4) The set of truncated beliefs is defined as , , A finite set of states is defined as ;use This indicates that the original model is in the time slot. Precise channel belief, using This indicates the truncation model in the time slot. The belief, with Indicates its current value; Steps 1-3: Truncate the mapping by observation interval Updated channel beliefs after a silent action: (5) Steps 1-4: In the given and hour, conditional probability is used This indicates that the transition probability corresponding to the ACK feedback is: (6) The transition probability corresponding to NACK or no feedback is: (7) Steps 1-5: Settings , Let these be the current finite state and the next finite state, respectively, where the finite state transition probabilities are: (8) in, This is an indicator function that takes the value 1 if the condition is true and 0 otherwise.
4. The transmission decision method for age-based erroneous information according to claim 1, characterized in that, In step 2, a transmission penalty is introduced for the accurate two-parameter discrete representation model. A deterministic strategy under a given Lagrange multiplier is obtained using relative value iteration, and a strategy satisfying the long-term average transmission rate constraint is obtained through multiplier binary search. When the deterministic strategy cannot directly satisfy the constraint, state-by-state randomization is performed between the high-transmission-rate endpoint strategy and the low-transmission-rate endpoint strategy to obtain a randomized strategy. The specific process is as follows: Step 2-1: Given a long-term average transmission rate constraint and Lagrange multipliers , finite state and actions The single-slot cost is defined as follows: ,from Begin executing relative value iteration: (9) Step 2-2: Select a fixed reference state Normalize the relative value function: (10) When the relative value iteration satisfies the preset convergence condition, the minimization action obtained from the last iteration is determined as the deterministic strategy. ; Steps 2-3: Evaluate the long-term average transmission rate of strategies corresponding to different Lagrange multipliers, and update the Lagrange multipliers using binary search: (11) in, and These represent the long-term average transmission rate of the corresponding strategy being higher than and lower than the constraint value, respectively. The endpoints of the Lagrange multipliers are determined; the median of the two is taken as the new Lagrange multiplier, and the corresponding endpoints are updated according to the long-term average transmission rate of the strategy corresponding to the median multiplier. Steps 2-4: When the deterministic strategy cannot directly satisfy the long-term average transmission rate constraint, select the high transmission rate endpoint strategy as... and low-transmission-rate endpoint strategy Then, state-by-state randomization is performed to obtain the randomization strategy. : (12) Among them, probability Select low-transmission-rate endpoint strategy With probability Select high-transmission-rate endpoint strategy To obtain the randomization strategy The randomization strategy is evaluated for its long-term average transmission rate, and the randomization factor that satisfies the long-term average transmission rate constraint and whose transmission rate is closest to the constraint value is selected. .
5. The transmission decision method for age-based error information according to claim 1, characterized in that, In step 3, in each time slot Initially, the decision control module in the transmitting end reads the current finite state. The strategy obtained in step 2 outputs the action. , The sending end sends a current status update at any time. The sending end remains silent and receives feedback at the end of the time slot. ,in and hour , and hour , hour .
6. The transmission decision method for age-based erroneous information according to claim 1, characterized in that, The specific process of updating the most recently observed channel state, observation interval, and model belief based on the feedback in step 4 is as follows: Step 4-1: In the time slot Obtain actual erroneous information before initial decision-making. According to equation (4), the truncated updated version is obtained. ; Step 4-2: Settings , They represent time slots respectively. The initial most recent observation channel state and observation interval, when season , , , season , , ; Step 4-3: When and , At that time, keep Unchanged, settings , ;when and When, set This indicates that after the observation interval reaches its upper limit, the channel belief will be mapped to a stationary belief; when and Maintain a stable mindset until an ACK or NACK response is received, then reset as per step 4-2.
7. The transmission decision method for age-based erroneous information according to claim 1, characterized in that, Step 5 uses the number of runtime slots as the termination condition for the monitoring task, preset as a positive integer. This is the total number of time slots for this run. After the next time slot status update is completed, the time slot number will be incremented by 1. Return to step 3. This run will end at this time. Used to limit the duration of this run.