A low-power data communication method and system for a smart meter

Abstract

The application relates to the technical field of digital information transmission, and discloses a low-power data communication method and system for a smart electric energy meter, wherein the method comprises the following steps: receiving a wake-up confirmation frame carrying a predicted number of to-be-responded electric energy meters; obtaining channel evaluation information by using Zadoff-Chu sequence sliding correlation detection; calculating an external interference occupancy rate and a concurrent load; determining a final contention window by combining priority compensation and window limiting; performing truncated negative exponential sampling according to a historical collision probability, initializing a backoff counter, and generating a wake-up request frame; determining collision intensity according to a base station response, updating the collision probability, and calculating a retransmission window. Compared with fixed backoff, the application connects same-frequency activity identification, external interference quantification, service priority and collision feedback in series, so that the contention window and the retransmission parameter are adjusted according to the load and the collision degree, interference misjudgment, invalid backoff and repeated wake-up power consumption are reduced, and the access timeliness of high-priority data is considered.

Description

Technical Field

[0001] This invention relates to the field of digital information transmission technology, and in particular to a low-power data communication method and system for smart meters. Background Technology

[0002] Currently, in scenarios such as densely populated residential areas, smart meters widely employ low-power wireless ad hoc networks for remote automatic meter reading. To extend battery life, these meters typically remain in a dormant state for extended periods, relying solely on intermittent wake-up frames or wake-up trigger commands periodically broadcast by the base station to initiate communication. Existing technologies primarily utilize contention-based random backoff mechanisms for wake-up frame collision handling, such as Carrier Sense Multiple Access / Collision Avoidance Protocols (CSMP). The process involves the base station broadcasting a wake-up confirmation frame, after which the responding meter estimates the response size based on the received signal. Then, within a fixed or binary exponential backoff window, it randomly selects a backoff time slot. The device that counts down to zero first sends a wake-up request frame to participate in channel contention. This approach attempts to mitigate the collision risk of dense responses through simple randomization and uses energy detection or received signal strength indication as the basis for channel busy / idle judgment to reduce mutual interference between wake-up frames.

[0003] For example, in a typical high-rise residential community, a single concentrator may need to manage hundreds of electricity meters. To meet the requirement of second-level data collection delay, the base station often triggers a large number of electricity meters to simultaneously transition from dormant to active state and report data within a beacon cycle, forming a highly synchronized burst load. At this time, traditional fixed contention windows or simple exponential backoff windows lack the ability to perceive real-time load. When strong broadband interference such as Wi-Fi in the same frequency band is present, simple energy detection is very likely to misjudge external interference as signals from the same network, resulting in a large number of invalid backoffs and wasted channel resources. At the same time, the wake-up requests of a large number of electricity meters under dense burst loads are very likely to cause synchronization collisions, and uniform random backoff cannot distinguish service priorities, making emergency alarm messages and regular meter reading data face the same collision risk. If the retransmission after the collision adopts a fixed window doubling or a preset retransmission limit, the wake-up overhead will increase approximately exponentially with the number of retransmissions. In dense deployment and high interference environments, power consumption will rise sharply, seriously deviating from the system's low power consumption design target and significantly shortening battery life.

[0004] Therefore, there is an urgent need for a method that can accurately detect collisions and adaptive backoff mechanisms using low-power wake-up frames in scenarios with dense multi-meter deployment and external wireless interference. This method dynamically adjusts the contention window and backoff distribution based on real-time channel occupancy, collision intensity, and service priority. In complex environments where collision retransmissions occur frequently, the wake-up overhead can still be reduced to less than 5% of the total communication energy consumption, while ensuring second-level meter reading delay. This will comprehensively improve the anti-interference capability and long-term operational reliability of smart meter wireless communication. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the purpose of this invention is to propose a low-power data communication method for smart meters. This method aims to solve the problem that traditional fixed random backoff technology cannot adapt to sudden loads and interference from co-frequency WIFI, resulting in frequent wake-up collisions and a surge in retransmission power consumption. In particular, under conditions of multiple meters deployed in dense residential areas with co-frequency WIFI interference, it is impossible to reduce the wake-up overhead to less than 5% of the total communication power consumption while ensuring second-level meter reading delay.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a low-power data communication method for smart energy meters.

[0007] The low-power data communication method for smart meters includes: Step S10: Obtain the wake-up confirmation frame broadcast by the base station carrying the estimated number of energy meters to be responded to, and perform channel evaluation and collision source priority identification tasks based on the wake-up confirmation frame using Zadoff-Chu sequence sliding correlation detection, and output channel evaluation information; Step S20: Based on the channel evaluation information, perform the environmental parameter determination task by calculating the external interference occupancy rate and the concurrent load classification, and output the environmental parameter information; Step S30: Based on the environmental parameter information, perform the competition window determination task by using priority compensation and competition window limiting calculation, and output the final competition window; Step S40: Based on the final contention window, the backoff counter initialization and wake-up request frame generation tasks are performed using a truncated negative exponential distribution sampling with historical collision probability feedback, and the wake-up request frame to be sent is output. Step S50: Send the wake-up request frame to be sent, and perform collision intensity determination, historical collision probability update and retransmission contention window determination tasks according to the base station's response to the wake-up request frame to be sent, and output retransmission control parameters.

[0008] Preferably, step S10, which involves acquiring a wake-up confirmation frame broadcast by the base station carrying an estimated number of energy meters to be responded to, performing channel evaluation and collision source priority identification tasks based on the wake-up confirmation frame using Zadoff-Chu sequence sliding correlation detection, and outputting channel evaluation information, specifically includes: Step S101: Receive the wake-up confirmation frame broadcast by the base station, and parse the estimated number of energy meters to be responded to from the wake-up confirmation frame; wherein, the base station pre-stores response records of multiple historical communication cycles, and counts the response device identifiers in the response records to obtain the estimated number of energy meters to be responded to. Step S102: During the wake-up access period, the received signal is collected according to the preset sampling period. The pilot sequence in the received signal is subjected to sliding correlation operation with the locally stored Zadoff-Chu base sequence to obtain the correlation peak value, correlation peak delay and correlation peak phase corresponding to each sampling time, and the total number of sampling times is recorded. Step S103: Filter out valid correlation peaks whose correlation peak values ​​are greater than a preset correlation peak threshold, and calculate the correlation peak delay dispersion corresponding to the valid correlation peak; when a valid correlation peak is filtered out and the correlation peak delay dispersion is less than a preset delay dispersion threshold, it is determined that a co-frequency wake-up activity has been detected, the number of co-frequency correlation peaks is counted, and the correlation peak phase rotation mode corresponding to the valid correlation peak is matched with a preset priority phase rotation codebook to obtain the collision source priority; when a valid correlation peak is filtered out and the correlation peak delay dispersion is not less than a preset delay dispersion threshold, it is determined that external narrowband interference has been detected, the number of external interference samplings is counted, and the average energy of external interference corresponding to the number of external interference samplings is calculated; when no valid correlation peak is filtered out, the number of co-frequency correlation peaks, the number of external interference samplings, and the average energy of external interference are set to zero respectively; the estimated number of energy meters to be responded to, the total number of samplings, the number of co-frequency correlation peaks, the number of external interference samplings, the average energy of external interference, and the collision source priority are encapsulated into the channel evaluation information.

[0009] Preferably, step S20, which involves performing the environmental parameter determination task based on the channel evaluation information using external interference occupancy rate calculation and concurrent load classification calculation, and outputting environmental parameter information, specifically includes: Step S201: Extract the number of external interference samples and the total number of samples from the channel evaluation information, and determine the ratio of the number of external interference samples to the total number of samples as the external interference channel occupancy rate; Step S202: Extract the average energy of external interference from the channel evaluation information and obtain a predetermined reference energy threshold; add one to the ratio of the average energy of external interference to the reference energy threshold and take the natural logarithm; multiply the obtained natural logarithm value, the external interference channel occupancy rate and the preset interference intensity adjustment coefficient to obtain the external interference intensity quantification index. Step S203: Read the number of co-frequency correlation peaks corresponding to multiple consecutive communication cycles from the local historical channel evaluation record, and calculate the average number of concurrent nodes corresponding to the multiple consecutive communication cycles; when the estimated number of energy meters to be responded to is greater than the preset dense scene judgment threshold, determine the ratio of the estimated number of energy meters to be responded to to the average number of concurrent nodes as the load factor; when the estimated number of energy meters to be responded to is not greater than the preset dense scene judgment threshold, set the load factor to one; encapsulate the external interference intensity quantification index, load factor, estimated number of energy meters to be responded to, and collision source priority in the channel evaluation information into the environmental parameter information.

[0010] Preferably, in step S202, the step of determining the reference energy threshold includes: during the deployment and initialization phase of the smart energy meter, selecting a silent period that does not contain wake-up signals and data transmission signals; collecting multiple air interface noise energy samples within the silent period according to a preset noise sampling period; comparing each of the air interface noise energy samples with a preset abnormal noise threshold, and removing air interface noise energy samples with energy values ​​greater than the preset abnormal noise threshold; calculating the arithmetic mean of the air interface noise energy samples retained after the removal process, and determining the arithmetic mean as the reference energy threshold.

[0011] Preferably, step S30, which involves performing a contention window determination task based on the environmental parameter information using priority compensation and contention window limiting calculation, and outputting the final contention window, specifically includes: Step S301: Extract the external interference intensity quantification index from the environmental parameter information. Loading factor and the estimated number of electricity meters to be responded to And read the historical collision probabilities stored locally. The basic competition window is calculated according to the following formula. :

[0012] in, To preset the minimum competition window, To preset the maximum competition window, The preset load window gain is expressed in terms of the number of wake-up access slots. To preset the interference adjustment coefficient, This indicates rounding to the nearest integer. Indicates the variable Limited to the lower limit and upper limit Within the closed interval formed; Step S302: Obtain the local service priority, the urgency of the data to be transmitted, and the remaining local battery power. Input the urgency of the data to be transmitted and the remaining local battery power into a preset two-dimensional compensation mapping table to obtain the priority window compensation amount. ; Step S303: Extract the collision source priority from the environmental parameter information; when the local service priority is high and the collision source priority is high, calculate the final contention window according to the following formula. :

[0013] When the local service priority is not high priority, or the collision source priority is not high priority, the final contention window is calculated according to the following formula. :

[0014] in, To pre-set the window reduction amount, The increment of the preset yield window is used as the basis for calculating the basic contention window, the final contention window, the priority window compensation amount, the preset preemption window reduction amount, and the preset yield window increment amount. All of these are expressed in terms of the number of wake-up access slots and are non-negative integers.

[0015] Preferably, step S40, which involves initializing the backoff counter and generating the wake-up request frame based on the truncated negative exponential distribution sampling using historical collision probability feedback within the final contention window, and outputting the wake-up request frame to be sent, specifically includes: Step S401: Obtain the preset maximum retreat scale parameter Calculate the retreat scale parameters according to the following formula. :

[0016] The final competition window As the upper limit of the interval truncated by the negative exponential distribution, the initial value of the backoff counter is generated according to the following formula. :

[0017] Wherein, the historical collision probability The preset maximum retreat scale parameter is greater than or equal to zero and less than 1. The backoff metric parameter is represented by the number of wake-up access slots and is greater than zero. Expressed in terms of the number of wake-up access slots, A random number generated uniformly within an interval greater than or equal to zero and less than 1. Represents the natural logarithm function. Represents an exponential function. This indicates rounding down; the initial value of the backoff counter. greater than or equal to zero and less than the final competition window Integers; Step S402: Decrement the initial value of the backoff counter sequentially according to the wake-up access time slot; when the decremented backoff counter returns to zero, obtain the local service priority, select the perturbation phase sequence corresponding to the local service priority from the locally pre-stored priority phase rotation codebook, and use the perturbation phase sequence to perform phase modulation on the locally pre-stored Zadoff-Chu base sequence to obtain the composite pilot sequence. Step S403: Obtain the short identifier of the local device, and write the short identifier of the local device and the composite pilot sequence into the wake-up request frame to obtain the wake-up request frame to be sent.

[0018] Preferably, step S50, which involves sending the wake-up request frame to be sent, performing collision intensity determination, historical collision probability update, and retransmission contention window determination tasks based on the base station's response to the wake-up request frame, and outputting retransmission control parameters, specifically includes: Step S501: Send the wake-up request frame to be sent, and detect the response frame returned by the base station within a preset response timeout period; when no response frame is detected within the preset response timeout period, it is determined that a wake-up access collision has occurred, pilot correlation peaks are extracted from the locally received signal, and the number of pilot correlation peaks with different delays is determined as the collision intensity index; when the response frame is detected within the preset response timeout period, the collision intensity index is set to zero. Step S502: Read the historical collision probability and preset forgetting factor stored locally before the update, divide the conflict intensity index by the sum of the conflict intensity index and 1 to obtain the collision characterization value of the current communication cycle; use the preset forgetting factor as the weight of the historical collision probability before the update, and use the difference between 1 and the preset forgetting factor as the weight of the collision characterization value, and perform a weighted summation of the historical collision probability before the update and the collision characterization value to obtain the updated historical collision probability; Step S503: When the conflict intensity index is greater than zero, multiply the conflict intensity index, the preset window enhancement coefficient, and the final competition window to obtain the candidate window increase; calculate the difference between the preset maximum competition window and the final competition window to obtain the increaseable window amount; determine the smaller value between the candidate window increase and the increaseable window amount as the target window increase; add the target window increase to the final competition window to obtain the retransmission competition window; when the conflict intensity index is equal to zero, determine the final competition window as the retransmission competition window; encapsulate the updated historical collision probability and the retransmission competition window into the retransmission control parameters.

[0019] The present invention also provides a low-power data communication system for smart meters, the low-power data communication system for smart meters comprising: The channel evaluation module is used to acquire the wake-up confirmation frame broadcast by the base station carrying the estimated number of energy meters to be responded to, and to perform channel evaluation and collision source priority identification tasks based on the wake-up confirmation frame using Zadoff-Chu sequence sliding correlation detection, and output channel evaluation information. The environmental parameter determination module is used to perform the environmental parameter determination task based on the channel evaluation information by calculating the external interference occupancy rate and the concurrent load classification, and output the environmental parameter information. The contention window determination module is used to perform the contention window determination task based on the environmental parameter information by using priority compensation and contention window limiting calculation, and output the final contention window. The wake-up request frame generation module is used to perform backoff counter initialization and wake-up request frame generation tasks based on the final contention window and the truncated negative exponential distribution sampling with historical collision probability feedback, and output the wake-up request frame to be sent. The retransmission control module is used to send the wake-up request frame to be sent, perform collision intensity determination, historical collision probability update and retransmission contention window determination tasks according to the base station's response to the wake-up request frame to be sent, and output retransmission control parameters.

[0020] The present invention also provides a low-power data communication device for a smart energy meter, the low-power data communication device for a smart energy meter comprising: a memory, a processor, and a low-power data communication program for a smart energy meter stored in the memory and executable on the processor, wherein the low-power data communication program for a smart energy meter implements the above-described method when executed by the processor.

[0021] The present invention also provides a computer program product, the computer program product including a low-power data communication program for a smart energy meter, wherein the low-power data communication program for a smart energy meter implements the above-described method when executed by a processor.

[0022] The beneficial effects of this invention are as follows: 1. This invention utilizes Zadoff-Chu sequence sliding correlation detection, applying correlation peak value, correlation peak delay dispersion, and correlation peak phase to effective peak selection, classify co-frequency wake-up activities and external narrowband interference, and identify collision source priorities. It also uses the estimated number of energy meters to be responded to, the total number of samples, the number of co-frequency correlation peaks, the number of external interference samples, and the average energy to form channel evaluation information. Subsequently, it calculates the external interference channel occupancy rate and external interference intensity quantification indicators, and determines the load factor based on the estimated number of energy meters to be responded to and the average number of concurrent nodes. This avoids the misjudgment of external interference as network contention by simple energy detection, allowing subsequent window calculations to simultaneously obtain the interference duration, relative energy, concurrent load, and collision source priority.

[0023] 2. This invention determines and limits the basic contention window based on external interference intensity, load factor, estimated number of energy meters awaiting response, and historical collision probability. It then combines this with the urgency of the data to be transmitted, remaining local battery power, and collision source priority to perform priority compensation or preemption correction, resulting in the final contention window. Subsequently, the final contention window is used as the upper limit of the interval for truncating the negative exponential distribution, and the backoff scale is adjusted using historical collision probability. An initial value for the backoff counter is generated within this window, and the perturbation phase sequence corresponding to the service priority is modulated into the Zadoff-Chu base sequence to form a composite pilot sequence. This allows the backoff range and distribution pattern to change with real-time load and historical collision status, reducing concentrated collisions and invalid waiting caused by a fixed window while reserving faster access conditions for high-priority data.

[0024] 3. This invention distinguishes between successful access and wake-up access collisions based on the base station's response after sending a wake-up request frame. If no response is received, the collision intensity is determined by the number of pilot correlation peaks with different delays. The collision intensity is then used to form the current collision characterization value, which is used to weight and update the historical collision probability. The smaller value between the increase in the candidate window and the remaining space of the maximum contention window is taken to obtain a retransmission contention window constrained by an upper limit. This allows for the differentiation between occasional collisions involving a small number of nodes and synchronous contention involving a large number of nodes. The historical collision probability is responsible for smoothing long-term states, while the retransmission window is responsible for matching the scale of the current collision. This avoids excessive window expansion during mild collisions or insufficient expansion during severe collisions due to binary exponential backoff, thereby reducing wake-up power consumption caused by ineffective backoff and continuous retransmissions. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating the first embodiment of a low-power data communication method for smart meters according to the present invention. Detailed Implementation

[0026] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0027] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Example 1: As Figure 1 The diagram shown is a flowchart of a first embodiment of a low-power data communication method for smart meters according to the present invention.

[0029] In the first embodiment, the low-power data communication method for smart meters includes: Step S10: Obtain the wake-up confirmation frame broadcast by the base station carrying the estimated number of energy meters to be responded to, and perform channel evaluation and collision source priority identification tasks based on the wake-up confirmation frame using Zadoff-Chu sequence sliding correlation detection, and output channel evaluation information; Channel assessment information is a data set that correlates and describes the scale of current wake-up access, related characteristics of received signals, external interference status, and collision source priority. Before broadcasting the wake-up confirmation frame, the base station reads response records from multiple historical communication cycles, counts the responding device identifiers appearing in these records, obtains an estimated number of energy meters to be responded to, and writes this number into the wake-up confirmation frame. After receiving this frame, the energy meter first parses the estimated number, so that subsequent load calculations have an overall scale basis from the base station side, rather than relying solely on the results of a single local listening.

[0030] After entering the wake-up access period, the energy meter collects the received signal according to a preset sampling period and performs a sliding correlation between the pilot sequence in the received signal and the locally stored Zadoff-Chu base sequence at each sampling position. Each calculation records the correlation peak value, correlation peak delay, and correlation peak phase, while accumulating the total number of samples. The correlation peak value reflects the degree of matching between the received pilot and the local base sequence; the correlation peak delay is used to determine the concentration of the peak value on the time axis; and the correlation peak phase is used to identify the priority phase rotation mode adopted by the transmitting node. These three types of features serve activity detection, interference classification, and priority identification respectively, and cannot be substituted for each other.

[0031] The controller first filters out valid correlation peaks whose peak values ​​are greater than a preset correlation peak threshold, and then calculates the correlation peak delay dispersion of these valid correlation peaks. When a valid correlation peak exists and the delay dispersion is less than the preset threshold, it is determined that a co-frequency wake-up activity has been detected. The number of co-frequency correlation peaks is counted, and the phase rotation mode of the valid correlation peaks is matched with a preset priority phase rotation codebook to obtain the collision source priority. When a valid correlation peak exists but the delay dispersion is not less than the threshold, it is determined to be external narrowband interference. The number of external interference samples is counted, and the corresponding average energy is calculated. When no valid correlation peak is found, the above quantity and energy are set to zero. Finally, the estimated number of energy meters to be responded to, the total number of samples, the number of co-frequency correlation peaks, the number of external interference samples, the average energy of external interference, and the collision source priority are encapsulated into channel evaluation information.

[0032] The aforementioned branches ensure that each output has a clear source. The total number of samples provides the denominator for S20's calculation of the external interference channel occupancy rate. The number of external interference samples and the average energy together characterize the duration and energy level of the interference. The number of co-frequency correlation peaks is used to form the statistics of concurrent nodes in continuous communication cycles, and the estimated number of energy meters awaiting response is used to determine whether a dense response scenario is in place. In addition to being transmitted to S20 along with the channel evaluation information, the collision source priority continues to enter the contention window branch selection in S30. Therefore, the signal detection results can be continuously used along the environmental parameters and window calculation chain.

[0033] Traditional energy sensing typically only provides channel busy / idle results, failing to distinguish whether received energy originates from a network wake-up signal or external narrowband interference, nor can it identify service priorities from conflict signals. In densely populated residential areas with nearby wireless devices, simply using received energy exceeding a threshold as the condition for freeze-back can easily misinterpret external interference as network contention, causing the energy meter to continue waiting within its actual access time slot. This step utilizes correlation peak amplitude, delay dispersion, and phase rotation mode to handle matching, classification, and priority identification tasks respectively, ensuring that the count and energy data received by S20 are no longer mixed from different physical sources.

[0034] Taking a wake-up access period containing 100 sampling moments as an example, the base station estimates the number of energy meters awaiting response to be 300 based on the response device identifiers from the most recent multiple communication cycles. A certain energy meter detects 8 sets of valid correlation peaks exceeding the correlation peak threshold. Among them, 5 sets have a delay dispersion of 2 sampling points, less than the threshold of 3 sampling points, and are counted as co-frequency correlation peaks; their phase rotation mode is consistent with the high-priority codeword, and the collision source priority is recorded as high priority. Another 3 sets of valid correlation peaks have a more dispersed delay distribution, with a dispersion not less than the threshold, and are counted as external narrowband interference samples used to calculate the average energy. The resulting channel evaluation information simultaneously retains six types of information: 300, 100, 5, 3, external interference average energy, and high priority.

[0035] Step S20: Based on the channel evaluation information, perform the environmental parameter determination task by calculating the external interference occupancy rate and the concurrent load classification, and output the environmental parameter information; Environmental parameter information is used to convert the discrete detection results output by S10 into external interference intensity, concurrent load, and priority conditions that can be used for window calculation. The controller extracts the number of external interference samples and the total number of samples from the channel evaluation information, and divides the former by the latter to obtain the external interference channel occupancy rate. This occupancy rate describes the proportion of time that external narrowband interference occurs during the current sampling period; when the total number of samples is accumulated according to a preset sampling period, the occupancy rates obtained in different communication periods have the same statistical caliber and can be used to compare the degree of interference persistence.

[0036] The baseline energy threshold is determined during the smart meter deployment initialization phase. The controller selects a quiet period that does not contain wake-up signals or data transmission signals, and collects multiple air interface noise energy samples according to a preset noise sampling period. Each sample is compared with a preset abnormal noise threshold, and samples with energy values ​​exceeding the threshold are discarded. The remaining samples are then calculated as an arithmetic mean. The baseline energy threshold obtained in this way represents the normal air interface noise level at the installation location; occasional strong pulses will not raise the baseline value. If all samples are discarded, the current quiet period cannot form a valid baseline, and a new quiet period that meets the quiet conditions should be selected to complete the same determination process.

[0037] The external interference intensity quantification index uses both the interference duration ratio and relative energy. The controller first divides the average external interference energy by a baseline energy threshold, adds one to the resulting ratio, and takes the natural logarithm. This natural logarithm value is then multiplied by the external interference channel occupancy rate and a preset interference intensity adjustment coefficient. The addition ensures that logarithmic calculations can still be performed when the average energy is close to zero. The natural logarithm helps to compress the range of energy ratio variation, while the occupancy rate prevents a short-duration strong interference from being equated to continuous interference. With these three factors working together, the output value increases with the duration and relative energy of the external interference.

[0038] Concurrent load is determined by a combination of global estimated scale and local historical detection. The energy meter reads the number of co-frequency correlation peaks from local channel evaluation records across multiple consecutive communication cycles and averages them to obtain the average number of concurrent nodes. When the estimated number of energy meters awaiting response exceeds a preset dense scenario judgment threshold, the load factor is obtained by dividing the estimated number by the average number of concurrent nodes; when the estimated number is not greater than the threshold, the load factor is set to one. The external interference intensity quantification index, load factor, estimated number of energy meters awaiting response, and collision source priority are then encapsulated into environmental parameter information for S30 to calculate the basic contention window and select priority branches.

[0039] Expanding the window solely based on whether there was a collision in the previous round would confuse unresponsiveness caused by external interference with collisions with nodes on the same network; using only the estimated number broadcast by the base station would fail to reflect the actual detectable concurrency level at the local installation location. This step quantifies the interference occupancy rate, relative energy, global estimated scale, and local co-frequency correlation peak history separately, and then assembles environmental parameters according to the conditions defined in the claims. In this way, S30 can both increase the corresponding window items under continuous external interference and expand the basic window in advance based on the load factor when the estimated response scale exceeds the density threshold, reducing the lag caused by relying solely on post-collision feedback for window adjustment.

[0040] Using the aforementioned sampling scenario, when external interference occurs 15 times out of 100 samples, the external interference channel occupancy rate is 0.15. If the baseline energy threshold obtained after removing abnormal samples during the silent period is 0.01 microwatts, the average energy of external interference in this round is 0.04 microwatts, and the interference intensity adjustment coefficient is 0.8, then the energy ratio of 4 is first obtained, one is added to it, the natural logarithm is taken, and then multiplied by 0.15 and 0.8, the external interference intensity quantification index is approximately 0.193. If the number of co-frequency correlation peaks in the most recent 5 communication cycles are 4, 5, 5, 6, and 5 respectively, the average number of concurrent nodes is 5, and the estimated number of electricity meters to be responded to is 300, which is higher than the threshold of 200 in a dense scenario, then the load factor is 60; the above indicators, along with the estimated number and high-priority collision source information, are entered into S30.

[0041] Step S30: Based on the environmental parameter information, perform the competition window determination task by using priority compensation and competition window limiting calculation, and output the final competition window; The final contention window is the upper limit of the time slot range used by the energy meter when generating the backoff counter in this round. It is determined by the basic contention window, the priority window compensation amount, and the priority contention branch. The controller extracts the external interference intensity quantification index, load factor, and estimated number of energy meters to be responded to from the environmental parameter information, and reads the locally stored historical collision probabilities. The calculation of the basic contention window starts from the preset minimum contention window, and then adds the load term and external interference term in sequence, followed by rounding and upper and lower limit constraints.

[0042] The load term is obtained by multiplying the preset load window gain, historical collision probability, estimated number of energy meters to be responded to, and load factor, reflecting the combined impact of the current potential response scale and past collision levels. The external interference term is obtained by multiplying the preset interference adjustment coefficient, external interference intensity quantification index, and preset maximum contention window, reflecting the impact of external narrowband interference on available access time slots. The controller adds the minimum contention window to the two increments and rounds them off, then limits the result to a closed interval formed by the preset minimum contention window and the preset maximum contention window to obtain the basic contention window, preventing extreme loads or interference from causing the window to exceed the device's allowable range.

[0043] The priority window compensation amount is derived from a preset two-dimensional compensation mapping table. The energy meter reads the local service priority, the urgency of the data to be transmitted, and the remaining local battery power. Using the urgency level and remaining power as a two-dimensional index, it looks up the table to obtain a non-negative compensation amount expressed as the number of wake-up access slots. This mapping allows regular meter readings, low battery alerts, and fault alarms to generate different window corrections under the same environmental parameters without requiring additional negotiation messages during access. The compensation amount is retained as an independent object and is not recalculated with the collision source priority or external interference intensity.

[0044] The controller continues to extract collision source priorities from environmental parameter information and determines the final contention window based on two conditions. When both the local service priority and the collision source priority are high priorities, a preset preemption window reduction is subtracted from the basic contention window, and the result is then limited to between the minimum and maximum contention windows. If this condition is not met, a priority window compensation and a preset backoff window increase are added to the basic contention window, and the same limiting is applied. Both branches output non-negative integer time slot numbers, enabling S40 to directly use the final contention window as the upper limit of the interval for truncating the negative exponential distribution.

[0045] Fixed contention windows cannot adapt to the scale of dense responses, and binary exponential backoff can only expand the window by a predetermined multiple after a collision, making it difficult to simultaneously handle external interference, service urgency, and contention with other high-priority nodes. This step first uses environmental parameters to determine a base window that matches the network state, then uses a two-dimensional mapping table and priority branches to complete service-side corrections, while maintaining window limiting at each step. This avoids channel idleness caused by excessively large windows under light loads, and also allows for the expansion of selectable time slots in advance when both dense loads and external interference coexist, while reserving clear preemption conditions for high-priority nodes.

[0046] For example, the preset minimum contention window is 16 time slots, the maximum contention window is 128 time slots, the historical collision probability is 0.25, the estimated number of energy meters awaiting response is 300, the load factor is 60, and the external interference intensity quantification index is 0.193. After calculating two increments based on the configured load window gain and interference adjustment coefficient, rounding and limiting, a basic contention window of 73 time slots is obtained. When an energy meter sends a low-battery alert, the two-dimensional compensation mapping table gives a compensation amount of 12 time slots; since the local unit is not a high-priority unit, it does not enter the preemption branch, and an additional preset yield window increment of 8 time slots is added, resulting in a final contention window of 93 time slots. If both the local unit and the collision source are high-priority units and a preemption reduction of 20 time slots is used, the same basic window will form a final contention window of 53 time slots.

[0047] Step S40: Based on the final contention window, the backoff counter initialization and wake-up request frame generation tasks are performed using a truncated negative exponential distribution sampling with historical collision probability feedback, and the wake-up request frame to be sent is output. This step converts the final contention window given by S30 into the initial value of the backoff counter, and generates a wake-up request frame with device identifier and priority pilot characteristics after the count reaches zero. The controller first obtains the preset maximum backoff scale parameter expressed in terms of the number of wake-up access slots, and then multiplies the difference obtained by subtracting the historical collision probability by the maximum scale parameter to obtain the current backoff scale parameter. Since the historical collision probability is limited to greater than or equal to zero and less than 1, the backoff scale parameter remains positive and can be used for subsequent truncated negative exponential distribution sampling.

[0048] During sampling, the final contention window serves as the upper limit of the interval for truncating the negative exponential distribution. The controller uniformly generates random numbers within an interval greater than or equal to zero and less than 1. First, the final contention window is divided by the backoff scale parameter and the result is negative. Then, the corresponding exponential value is calculated. This exponential value is subtracted from 1 to form a truncation interval correction. This correction is multiplied by the random number and then subtracted from 1. The natural logarithm of the result is taken, multiplied by the negative value of the backoff scale parameter, and finally rounded down. The resulting initial backoff counter value is an integer greater than or equal to zero and less than the final contention window, eliminating the need for the modulo operation used in the old scheme.

[0049] After the counter is loaded, the energy meter decrements sequentially according to the wake-up access time slot. When the decrement result reaches zero, the controller reads the local service priority, selects the corresponding perturbation phase sequence from the locally pre-stored priority phase rotation codebook, and uses this sequence to perform phase modulation on the Zadoff-Chu base sequence to obtain a composite pilot sequence. Subsequently, it reads the local device short identifier, writes the short identifier and the composite pilot sequence into the wake-up request frame, forming a wake-up request frame to be sent. The short identifier is used by the base station to identify the response device, while the composite pilot sequence retains the relevant detection and priority phase patterns for subsequent response and collision analysis.

[0050] The final competition window controls the range of selectable time slots. The historical collision probability is controlled by the backoff scale parameter, which controls the distribution of samples in each time slot within the same range. Random numbers ensure that different energy meters do not necessarily obtain the same count value even if their parameters are the same. Thus, S30 is responsible for determining the overall dispersion space, and S40 is responsible for forming a non-uniform random backoff within this space, influenced by the historical collision state. The composite pilot sequence in the output frame corresponds to the local base sequence and priority codebook used in S10, enabling the transmission results of this round to be identified by the base station or other nodes according to the same relevant features.

[0051] Uniform random backoff ensures that each time slot within the final competition window has the same selection probability. However, when hundreds of electricity meters are simultaneously woken up, multiple nodes may still concentrate on selecting similar time slots. Expanding the window only after a collision would increase the average waiting time along with the overall window size. This step uses a truncated negative exponential distribution of historical collision probability feedback within a finite window. It does not change the window boundary determined by S30, but adjusts the sampling scale within the boundary. At the same time, priority phase information is written into the composite pilot to avoid backoff and collision identification becoming two unrelated processes.

[0052] Taking the final competition window of 93 slots as an example, assuming a historical collision probability of 0.25 and a preset maximum backoff scale parameter of 80 slots, the current backoff scale parameter is 60 slots. The energy meter generates a random number 0.4. Following the literal arithmetic order of the truncated negative exponential distribution, the truncation interval correction corresponding to the window and scale is first calculated, and then the natural logarithm and rounding down are performed to obtain the initial value of the backoff counter, 22. After 22 wake-up access slots, the counter is reset to zero. The node selects the perturbation phase sequence modulation Zadoff-Chu base sequence corresponding to its local service priority, and writes the resulting composite pilot sequence and the device short identifier into the wake-up request frame, which is then sent and the response is detected by the S50.

[0053] Step S50: Send the wake-up request frame to be sent, and perform collision intensity determination, historical collision probability update and retransmission contention window determination tasks according to the base station's response to the wake-up request frame to be sent, and output retransmission control parameters.

[0054] The retransmission control parameters consist of the updated historical collision probability and the retransmission contention window, used to describe the collision state and window upper limit required for the next backoff. After the energy meter sends the wake-up request frame generated by S40, it continuously monitors the response frames returned by the base station within a preset response timeout period. If no response frame is detected within the timeout period, a wake-up access collision is determined to have occurred. Pilot correlation peaks are extracted from the locally received signal, and the number of pilot correlation peaks with different delays is determined as the collision intensity index. When a response frame is detected, the collision intensity index is set to zero.

[0055] Historical collision probabilities are updated using collision characterization values ​​from the current communication cycle. The controller first divides the collision intensity index by the sum of that index and 1 to obtain a collision characterization value that is not less than zero and less than 1; as the number of collision peaks increases, this characterization value gradually approaches one. Subsequently, a preset forgetting factor is used as the weight of the historical collision probability before the update, and the difference between 1 and the forgetting factor is used as the weight of the current collision characterization value. The two are then weighted and summed to obtain the updated historical collision probability. This process preserves the trend of previous communication cycles and allows the current round of collision intensity to participate in the calculation of the next backoff scale.

[0056] The retransmission contention window is determined based on the collision intensity and the equipment's allowable range. When the collision intensity is greater than zero, the collision intensity, the preset window enhancement coefficient, and the final contention window of this round are multiplied to obtain the candidate window increase amount; then, the preset maximum contention window is subtracted from the final contention window of this round to obtain the current available window increase amount; the smaller value between the candidate window increase amount and the available window increase amount is selected as the target window increase amount, and this increase amount is added to the final contention window to obtain the retransmission contention window. When the collision intensity is equal to zero, no window is added, and the final contention window of this round is directly determined as the retransmission contention window.

[0057] The two outputs described above serve different feedback functions. The updated historical collision probability redetermines the backoff scale parameter in the next S40 iteration, ensuring the distribution pattern follows long-term collision state changes. The retransmission contention window directly limits the range of time slots available for the next retransmission, ensuring the window expansion corresponds to the intensity of the current collision. The increase in candidates is also limited by the remaining space of the maximum contention window, so even if a large number of correlation peaks are detected, the retransmission window will not exceed the device's preset upper limit. If a response is received from the base station in this round, the collision intensity is zero, the window remains unchanged, and the historical collision probability decays according to the forgetting factor, preventing the continued use of a high-collision state after successful communication.

[0058] Binary exponential backoff typically treats every non-response as the same collision, directly doubling the window size. This fails to distinguish between sporadic collisions among a few nodes and synchronous competition among a large number of nodes. Doubling the window for minor collisions introduces additional waiting time, while severe collisions may require multiple rounds of doubling to create sufficient dispersion space. This step quantifies the collision intensity by counting pilot correlation peaks with different time delays, and then updates the probability and window size separately. The probability feedback is used to smooth the long-term state, while the window feedback is used to handle the current retransmission scale, thereby reducing invalid waiting time and continuous retransmissions caused by excessive or insufficient window expansion.

[0059] For example, after a power meter sends a wake-up request frame, it does not receive a response from the base station within a 50 ms response timeout period and detects four pilot correlation peaks with different delays in the received signal, resulting in a collision intensity index of 4. Before the update, the historical collision probability was 0.35, and the forgetting factor was 0.7. The current collision characteristic value is 4 divided by 5, which is 0.8. After weighting by the two factors, the updated historical collision probability is 0.485. With a final contention window of 93 time slots, a maximum contention window of 128 time slots, and a window enhancement coefficient of 0.2, the increase in the candidate window exceeds the 35 time slots that can be added. Therefore, 35 time slots are taken as the target increase, and the retransmission contention window is 128 time slots. The updated 0.485 and 128 time slots together constitute the retransmission control parameters.

[0060] Example 2: Furthermore, the present invention provides a low-power data communication system for smart meters, employing a low-power data communication method for smart meters as described in the above embodiments, which can solve the technical problem of low-power data communication for smart meters. The beneficial effects of the low-power data communication system for smart meters provided by the present invention are the same as those of the low-power data communication method for smart meters provided in the above embodiments, and other technical features of the low-power data communication system for smart meters are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0061] Example 3: This invention provides a low-power data communication device for smart meters. The device includes at least one processor and a memory communicatively connected to the processor. The memory stores instructions executable by the processor, which, when executed, enable the processor to perform the low-power data communication method for smart meters described in Example 1. This low-power data communication device for smart meters can include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. This low-power data communication device for smart meters is merely an example and should not limit the functionality or scope of the invention. A low-power data communication device for a smart meter may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory or a program loaded from a storage device into a random access memory. The random access memory also stores various programs and data required for the operation of the low-power data communication device for the smart meter. The processing unit, the read-only memory, and the random access memory are interconnected via a bus. An I / O interface is also connected to the bus. Typically, the following systems can be connected to the I / O interface: input devices including touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices including liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices including magnetic tapes, hard disks, etc.; and communication devices. The communication device allows the low-power data communication device for a smart meter to communicate wirelessly or wiredly with other devices to exchange data. While a low-power data communication device for a smart meter with various systems has been described, it should be understood that implementation or possession of all the described systems is not required. It can be implemented alternatively or with more or fewer systems.

[0062] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the low-power data communication method for smart meters described above. The computer program product provided by this invention can solve the technical problem of low-power data communication for smart meters. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as those of the low-power data communication method for smart meters provided in the above embodiments, and will not be repeated here.

[0063] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a read-only memory. When the computer program is executed by a processing device, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0064] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0065] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the present invention and its equivalents, the present invention also intends to include these modifications and variations.

Claims

1. A low-power data communication method for smart meters, characterized in that, The method includes: Step S10: Obtain the wake-up confirmation frame broadcast by the base station carrying the estimated number of energy meters to be responded to, and perform channel evaluation and collision source priority identification tasks based on the wake-up confirmation frame using Zadoff-Chu sequence sliding correlation detection, and output channel evaluation information; Step S20: Based on the channel evaluation information, perform the environmental parameter determination task by calculating the external interference occupancy rate and the concurrent load classification, and output the environmental parameter information; Step S30: Based on the environmental parameter information, perform the competition window determination task by using priority compensation and competition window limiting calculation, and output the final competition window; Step S40: Based on the final contention window, the backoff counter initialization and wake-up request frame generation tasks are performed using a truncated negative exponential distribution sampling with historical collision probability feedback, and the wake-up request frame to be sent is output. Step S50: Send the wake-up request frame to be sent, and perform collision intensity determination, historical collision probability update and retransmission contention window determination tasks according to the base station's response to the wake-up request frame to be sent, and output retransmission control parameters.

2. The low-power data communication method for smart meters as described in claim 1, characterized in that, Step S10 involves obtaining a wake-up confirmation frame broadcast by the base station carrying an estimated number of energy meters to be responded to, performing channel evaluation and collision source priority identification tasks based on the wake-up confirmation frame using Zadoff-Chu sequence sliding correlation detection, and outputting channel evaluation information. Specifically, this includes: Step S101: Receive the wake-up confirmation frame broadcast by the base station, and parse the estimated number of energy meters to be responded to from the wake-up confirmation frame; wherein, the base station pre-stores response records of multiple historical communication cycles, and counts the response device identifiers in the response records to obtain the estimated number of energy meters to be responded to. Step S102: During the wake-up access period, the received signal is collected according to the preset sampling period. The pilot sequence in the received signal is subjected to sliding correlation operation with the locally stored Zadoff-Chu base sequence to obtain the correlation peak value, correlation peak delay and correlation peak phase corresponding to each sampling time, and the total number of sampling times is recorded. Step S103: Filter out valid correlation peaks whose correlation peak values ​​are greater than a preset correlation peak threshold, and calculate the correlation peak delay dispersion corresponding to the valid correlation peak; when a valid correlation peak is filtered out and the correlation peak delay dispersion is less than a preset delay dispersion threshold, it is determined that a co-frequency wake-up activity has been detected, the number of co-frequency correlation peaks is counted, and the correlation peak phase rotation mode corresponding to the valid correlation peak is matched with a preset priority phase rotation codebook to obtain the collision source priority; when a valid correlation peak is filtered out and the correlation peak delay dispersion is not less than a preset delay dispersion threshold, it is determined that external narrowband interference has been detected, the number of external interference samplings is counted, and the average energy of external interference corresponding to the number of external interference samplings is calculated; when no valid correlation peak is filtered out, the number of co-frequency correlation peaks, the number of external interference samplings, and the average energy of external interference are set to zero respectively; the estimated number of energy meters to be responded to, the total number of samplings, the number of co-frequency correlation peaks, the number of external interference samplings, the average energy of external interference, and the collision source priority are encapsulated into the channel evaluation information.

3. The low-power data communication method for smart meters as described in claim 1, characterized in that, Step S20, which involves determining environmental parameters based on the channel evaluation information using external interference occupancy rate calculation and concurrent load classification calculation, and outputting environmental parameter information, specifically includes: Step S201: Extract the number of external interference samples and the total number of samples from the channel evaluation information, and determine the ratio of the number of external interference samples to the total number of samples as the external interference channel occupancy rate; Step S202: Extract the average energy of external interference from the channel evaluation information and obtain a predetermined reference energy threshold; add one to the ratio of the average energy of external interference to the reference energy threshold and take the natural logarithm; multiply the obtained natural logarithm value, the external interference channel occupancy rate and the preset interference intensity adjustment coefficient to obtain the external interference intensity quantification index. Step S203: Read the number of co-frequency correlation peaks corresponding to multiple consecutive communication cycles from the local historical channel evaluation record, and calculate the average number of concurrent nodes corresponding to the multiple consecutive communication cycles; when the estimated number of energy meters to be responded to is greater than the preset dense scene judgment threshold, determine the ratio of the estimated number of energy meters to be responded to to the average number of concurrent nodes as the load factor; when the estimated number of energy meters to be responded to is not greater than the preset dense scene judgment threshold, set the load factor to one; encapsulate the external interference intensity quantification index, load factor, estimated number of energy meters to be responded to, and collision source priority in the channel evaluation information into the environmental parameter information.

4. The low-power data communication method for smart meters as described in claim 3, characterized in that, In step S202, the step of determining the reference energy threshold includes: during the deployment and initialization phase of the smart energy meter, selecting a silent period that does not contain wake-up signals and data transmission signals; collecting multiple air interface noise energy samples within the silent period according to a preset noise sampling period; comparing each air interface noise energy sample with a preset abnormal noise threshold, and removing air interface noise energy samples with energy values ​​greater than the preset abnormal noise threshold; calculating the arithmetic mean of the air interface noise energy samples retained after the removal process, and determining the arithmetic mean as the reference energy threshold.

5. A low-power data communication method for smart meters as described in claim 3, characterized in that, Step S30, which involves performing a contention window determination task based on the environmental parameter information using priority compensation and contention window limiting calculation, and outputting the final contention window, specifically includes: Step S301: Extract the external interference intensity quantification index from the environmental parameter information. Loading factor and the estimated number of electricity meters to be responded to And read the historical collision probabilities stored locally. The basic competition window is calculated according to the following formula. : in, To preset the minimum competition window, To preset the maximum competition window, The preset load window gain is expressed in terms of the number of wake-up access slots. To preset the interference adjustment coefficient, This indicates rounding to the nearest integer. Indicates the variable Limited to the lower limit and upper limit Within the closed interval formed; Step S302: Obtain the local service priority, the urgency of the data to be transmitted, and the remaining local battery power. Input the urgency of the data to be transmitted and the remaining local battery power into a preset two-dimensional compensation mapping table to obtain the priority window compensation amount. ; Step S303: Extract the collision source priority from the environmental parameter information; when the local service priority is high and the collision source priority is high, calculate the final contention window according to the following formula. : When the local service priority is not high priority, or the collision source priority is not high priority, the final contention window is calculated according to the following formula. : in, To pre-set the window reduction amount, The increment of the preset yield window is used as the basis for calculating the basic contention window, the final contention window, the priority window compensation amount, the preset preemption window reduction amount, and the preset yield window increment amount. All of these are expressed in terms of the number of wake-up access slots and are non-negative integers.

6. A low-power data communication method for smart meters as described in claim 5, characterized in that, Step S40, which involves initializing the backoff counter and generating the wake-up request frame based on the truncated negative exponential distribution sampling using historical collision probability feedback within the final contention window, and outputting the wake-up request frame to be sent, specifically includes: Step S401: Obtain the preset maximum retreat scale parameter Calculate the retreat scale parameters according to the following formula. : The final competition window As the upper limit of the interval truncated by the negative exponential distribution, the initial value of the backoff counter is generated according to the following formula. : Wherein, the historical collision probability The preset maximum retreat scale parameter is greater than or equal to zero and less than 1. The backoff metric parameter is represented by the number of wake-up access slots and is greater than zero. Expressed in terms of the number of wake-up access slots, A random number generated uniformly within an interval greater than or equal to zero and less than 1. Represents the natural logarithm function. Represents an exponential function. This indicates rounding down; the initial value of the backoff counter. greater than or equal to zero and less than the final competition window Integers; Step S402: Decrement the initial value of the backoff counter sequentially according to the wake-up access time slot; when the decremented backoff counter returns to zero, obtain the local service priority, select the perturbation phase sequence corresponding to the local service priority from the locally pre-stored priority phase rotation codebook, and use the perturbation phase sequence to perform phase modulation on the locally pre-stored Zadoff-Chu base sequence to obtain the composite pilot sequence. Step S403: Obtain the short identifier of the local device, and write the short identifier of the local device and the composite pilot sequence into the wake-up request frame to obtain the wake-up request frame to be sent.

7. A low-power data communication method for smart meters as described in claim 1, characterized in that, In step S50, the steps of sending the wake-up request frame to be sent, performing collision intensity determination, historical collision probability update, and retransmission contention window determination tasks based on the base station's response to the wake-up request frame to be sent, and outputting retransmission control parameters specifically include: Step S501: Send the wake-up request frame to be sent, and detect the response frame returned by the base station within a preset response timeout period; when no response frame is detected within the preset response timeout period, it is determined that a wake-up access collision has occurred, pilot correlation peaks are extracted from the locally received signal, and the number of pilot correlation peaks with different delays is determined as the collision intensity index; when the response frame is detected within the preset response timeout period, the collision intensity index is set to zero. Step S502: Read the historical collision probability and preset forgetting factor stored locally before the update, divide the conflict intensity index by the sum of the conflict intensity index and 1 to obtain the collision characterization value of the current communication cycle; use the preset forgetting factor as the weight of the historical collision probability before the update, and use the difference between 1 and the preset forgetting factor as the weight of the collision characterization value, and perform a weighted summation of the historical collision probability before the update and the collision characterization value to obtain the updated historical collision probability; Step S503: When the conflict intensity index is greater than zero, multiply the conflict intensity index, the preset window enhancement coefficient, and the final competition window to obtain the candidate window increase; calculate the difference between the preset maximum competition window and the final competition window to obtain the increaseable window amount; determine the smaller value between the candidate window increase and the increaseable window amount as the target window increase; add the target window increase to the final competition window to obtain the retransmission competition window; when the conflict intensity index is equal to zero, determine the final competition window as the retransmission competition window; encapsulate the updated historical collision probability and the retransmission competition window into the retransmission control parameters.

8. A low-power data communication system for smart meters, applied to the low-power data communication method for smart meters as described in any one of claims 1 to 7, characterized in that, The low-power data communication system for smart meters includes: The channel evaluation module is used to acquire the wake-up confirmation frame broadcast by the base station carrying the estimated number of energy meters to be responded to, and to perform channel evaluation and collision source priority identification tasks based on the wake-up confirmation frame using Zadoff-Chu sequence sliding correlation detection, and output channel evaluation information. The environmental parameter determination module is used to perform the environmental parameter determination task based on the channel evaluation information by calculating the external interference occupancy rate and the concurrent load classification, and output the environmental parameter information. The contention window determination module is used to perform the contention window determination task based on the environmental parameter information by using priority compensation and contention window limiting calculation, and output the final contention window. The wake-up request frame generation module is used to perform backoff counter initialization and wake-up request frame generation tasks based on the final contention window and the truncated negative exponential distribution sampling with historical collision probability feedback, and output the wake-up request frame to be sent. The retransmission control module is used to send the wake-up request frame to be sent, perform collision intensity determination, historical collision probability update and retransmission contention window determination tasks according to the base station's response to the wake-up request frame to be sent, and output retransmission control parameters.

9. A low-power data communication device for smart energy meters, characterized in that, The low-power data communication device for smart meters includes: a memory, a processor, and a low-power data communication program for smart meters stored in the memory and executable on the processor. When the low-power data communication program for smart meters is executed by the processor, it implements a low-power data communication method for smart meters according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a low-power data communication program for smart meters, which, when executed by a processor, implements a low-power data communication method for smart meters according to any one of claims 1 to 7.

Images