Power line carrier low-power communication system based on edge computing module

Abstract

This invention relates to the field of power line carrier communication technology, and specifically discloses a low-power power line carrier communication system based on an edge computing module. The system includes: an edge computing node deployed at the branch box or meter box side of the distribution network, electrically connected to a carrier transceiver module; and a relay control module electrically connected to both the edge computing node and the carrier transceiver module. The edge computing node constructs a three-dimensional relationship model based on wake-up phase parameters, wake-up duty cycle parameters, and payload transmission success rate parameters, determines a three-dimensional optimized boundary surface, calculates the optimal approximation direction vector for the current operating state point, and generates wake-up phase adjustment and wake-up duty cycle adjustment amounts. The relay control module calculates the matching wake-up phase value based on the phase line difference between the current node and the target relay node, generates a relay wake-up command to control the target relay node to enter the receiving state at the matching moment. This invention achieves optimal power consumption at the distribution network level while ensuring real-time communication through three-dimensional optimization and cross-phase time slot compensation.

Description

Technical Field

[0001] This invention relates to the field of power line carrier communication technology, and in particular to a power line carrier low-power communication system based on an edge computing module. Background Technology

[0002] Power line carrier communication (PLC) is a technology that uses existing power lines as the transmission medium to achieve data communication. It is widely used in distribution network automation, smart meter reading, and electricity consumption information collection. At the end of the distribution network, PLC nodes are typically deployed in branch boxes or meter boxes, operating continuously and subject to strict power consumption requirements. To achieve low-power operation, existing PLC nodes generally employ a periodic sleep-wake mechanism, waking up once at a preset time interval to monitor the channel, and remaining in deep sleep mode the rest of the time. Simultaneously, to extend communication coverage, the PLC network uses a relay mechanism, where the source node transmits data hop-by-hop to the destination node through relay nodes. These periodic wake-up and relay mechanisms together constitute the basic operational framework of a distribution network PLC communication system.

[0003] However, existing periodic wake-up and relay forwarding mechanisms face the following fundamental technical defects: On the one hand, the wake-up window position and wake-up duty cycle of carrier nodes are usually fixed or simply adjusted based on single-dimensional traffic statistics, which cannot adapt to the non-uniform distribution characteristics of power line channel noise within the power frequency cycle and the fixed phase difference of noise windows between different phase lines. This results in the source node and relay node wake-up timing failing to form effective coordination, leading to low success rate of cross-phase relay communication and high network-level power consumption. On the other hand, when multiple source nodes compete for the same relay node, there is a lack of arbitration mechanism coordinated with the wake-up timing, resulting in time slot conflicts or delays in high-priority services. The essence of these defects lies in the fact that existing solutions treat single-node power consumption optimization, relay node selection, and time slot allocation as independent processes, lacking a unified architecture that jointly optimizes wake-up phase, wake-up duty cycle, transmission success rate, and cross-phase relay constraints. This results in the distribution network end-carrier communication system failing to achieve a dynamic optimal balance between low power consumption, high real-time performance, and high reliability.

[0004] Therefore, this invention proposes a low-power communication system based on an edge computing module using power line carrier. Summary of the Invention

[0005] This invention provides a low-power communication system for power line carrier based on an edge computing module. Through the synergistic effect of wake-up timing dynamic adjustment driven by three-dimensional optimized boundary surface and cross-phase time slot compensation based on phase line differences, it achieves significant reduction in single-node power consumption, a substantial increase in cross-phase relay communication success rate, and global optimization of power consumption at the distribution network end-carrier communication system without increasing hardware costs.

[0006] This invention provides a power line carrier low-power communication system based on an edge computing module, comprising: Edge computing nodes are deployed on the side of the distribution network branch box or meter box and are electrically connected to the carrier transceiver module. The relay control module is electrically connected to the edge computing node and the carrier transceiver module, and the relay control module is equipped with a relay time slot arbitration unit; Edge computing nodes are equipped with topology migration identification units and cross-station collaboration units; Edge computing nodes are used for: The wake-up phase parameter, wake-up duty cycle parameter and corresponding payload transmission success rate parameter of the carrier transceiver module in multiple power frequency cycles within the historical period are obtained. The wake-up phase parameter represents the phase offset value of the wake-up window of the carrier transceiver module relative to the zero crossing time of the power frequency voltage of the phase line. A three-dimensional relational model is constructed based on the wake-up phase parameter, wake-up duty cycle parameter and payload transmission success rate parameter. In the three-dimensional relational model, a three-dimensional optimized boundary surface composed of non-dominated state points is determined. Obtain the current wake-up phase parameters and current wake-up duty cycle parameters of the carrier transceiver module, determine the current working state point in the three-dimensional relational model, and calculate the optimal approximation direction vector from the current working state point to the three-dimensional optimized boundary surface; Based on the components of the optimal approximation direction vector in the wake-up phase dimension and wake-up duty cycle dimension, wake-up phase adjustment amount and wake-up duty cycle adjustment amount are generated respectively. The wake-up duty cycle adjustment amount is used to adjust the wake-up window duration of the carrier transceiver module in the next power frequency cycle. The relay control module is used for: Obtain the phase line identifier information of the phase line where the target relay node is located; Based on the first zero-crossing moment of the phase line where this node is located, the second zero-crossing moment of the phase line where the target relay node is located, and the wake-up phase adjustment amount, calculate the matching wake-up phase value of the target relay node; The relay wake-up command is generated based on the matching wake-up phase value. The relay wake-up command is used to control the target relay node to enter the receiving state at the time indicated by the matching wake-up phase value.

[0007] Furthermore, edge computing nodes are also used to construct the confidence interval boundaries of the 3D optimized boundary surface, and to correct the wake-up phase adjustment and wake-up duty cycle adjustment based on the confidence interval boundaries: In the three-dimensional relational model, the density of three-dimensional sample points in the preset spatial neighborhood around each non-dominated state point is calculated. Non-dominated state points with a 3D sample point density not lower than a preset density threshold are marked as high-confidence boundary points; Based on the high-confidence boundary points, a confidence-optimized boundary surface is refitted and generated, and the average offset vector between the confidence-optimized boundary surface and the original 3D optimized boundary surface is calculated. The average offset vector is used as the boundary confidence correction vector. Before generating the wake-up phase adjustment and wake-up duty cycle adjustment, the boundary confidence correction vector is superimposed on the optimal approximation direction vector.

[0008] Furthermore, edge computing nodes are also used to construct a cross-phase noise period distribution map and correct the matched wake-up phase value of the target relay node based on the cross-phase noise period distribution map: Obtain the noise energy sampling sequences of phases A, B, and C within this substation area in multiple power frequency cycles. The noise energy sampling sequence represents the noise energy value collected at a preset sampling interval within one power frequency cycle. The noise energy sampling sequences of phases A, B, and C are aligned according to their respective zero-crossing times and then superimposed to generate a three-phase composite noise envelope curve. Identify low-noise regions where the noise energy is below a preset noise threshold in the three-phase synthesized noise envelope curve; When the target relay node and this node are located on different phase lines, determine whether the matched wake-up phase value of the target relay node falls into the low noise range; If the matched wake-up phase value of the target relay node does not fall into the low noise range, the phase offset compensation between the matched wake-up phase value and the nearest low noise range boundary is calculated, and the matched wake-up phase value is corrected based on the phase offset compensation.

[0009] Furthermore, the relay control module is also used to adaptively adjust the preamble length of the relay wake-up command based on the cross-phase noise difference: Obtain the first noise energy estimate of the phase line where this node is located at the matched wake-up phase value; Obtain the second noise energy estimate at the matched wake-up phase value of the phase line where the target relay node is located; The ratio of the first noise energy estimate to the second noise energy estimate is calculated and used as the cross-phase noise difference coefficient. The target preamble length is obtained by querying the preset preamble length mapping table based on the cross-phase noise difference coefficient. The target preamble length is encapsulated in the relay wake-up command. After receiving the relay wake-up command, the target relay node sends the wake-up preamble sequence with the target preamble length.

[0010] Furthermore, the relay slot arbitration unit is used to allocate time slots when multiple source nodes simultaneously request the same target relay node: When it is detected that both the first source node and the second source node send a relay wake-up request to the same target relay node within a preset time window, the first data transmission urgency score of the first source node and the second data transmission urgency score of the second source node are obtained. Obtain the first wake-up duration requested by the first source node and the second wake-up duration requested by the second source node; A first competition weight vector is constructed based on a first data transmission urgency score and a first wake-up duration, and a second competition weight vector is constructed based on a second data transmission urgency score and a second wake-up duration. Calculate the weighted magnitude of the first and second competition weight vectors in the two-dimensional competition space, and determine the source node with the larger weighted magnitude as the arbitration winner node; Assign the current wake-up window of the target relay node to the winning node in the arbitration, and send a slot delay indication to the unwinning node in the arbitration. The slot delay indication contains a delay phase offset value that suggests re-requesting.

[0011] Furthermore, the relay slot arbitration unit is also used to perform preemptive slot allocation when the data transmission urgency score of the node that failed the arbitration exceeds a preset urgency threshold: The current wake-up window of the target relay node is divided into a basic time slot segment and an extended time slot segment, with the basic time slot segment allocated to the arbitration winner node. Calculate the minimum required wake-up time for the node that did not win the arbitration. If the length of the extended time slot is not less than the minimum required wake-up time, then allocate the extended time slot to the node that did not win the arbitration. Send a wake-up window extension command to the target relay node. The wake-up window extension command is used to control the target relay node to continue receiving after the basic time slot ends until the extended time slot ends.

[0012] Furthermore, the topology transfer recognition unit is used to detect topological changes in the transformer area and transfer the 3D optimized boundary surface: Periodically acquire neighbor node beacon reception records of all carrier transceiver modules within this station area. The neighbor node beacon reception records include the neighbor node identifier of the received beacon and the corresponding received signal strength indication value. Construct a sequence of adjacent area adjacency matrices based on neighbor node beacon reception records over multiple periods; Calculate the Frobenius norm difference of the adjacency matrix of adjacent transformer areas between adjacent cycles. When the Frobenius norm difference exceeds the preset topology change threshold, it is determined that a transformer area topology change has occurred. After determining that a topological change has occurred in the transformer area, the similarity deviation between the first three-dimensional optimized boundary surface before the topological change and the second three-dimensional optimized boundary surface reconstructed after the topological change is calculated. If the similarity deviation exceeds the preset surface similarity threshold, a weighted fusion is performed based on the first 3D optimized boundary surface and the second 3D optimized boundary surface to generate a migrated 3D optimized boundary surface. The migrated 3D optimized boundary surface replaces the first 3D optimized boundary surface, and the replaced 3D optimized boundary surface is used to calculate the optimal approximation direction vector of the current working state point.

[0013] Furthermore, the topology migration identification unit is also used to perform incremental optimized surface expansion when it is determined that a topology change has occurred in the transformer area and the type of topology change is the addition of a new node: Identify the access location of the new node in the transformer area topology and its adjacent existing nodes; Obtain the 3D optimized boundary surface of adjacent existing nodes as the reference boundary surface; Acquire initial sample points of wake-up phase parameters, wake-up duty cycle parameters, and payload transmission success rate parameters collected during the initialization phase of the newly added node; Based on the local curvature characteristics of the reference boundary surface, the initial sample points are projected onto the reference boundary surface along the normal vector direction to obtain the projection points; The projected points are weighted and averaged with the initial sample points to generate the initial 3D optimized boundary surface of the new node. The initial 3D optimized boundary surface of the new node is used to calculate the optimal approximation direction vector in subsequent runs.

[0014] Furthermore, edge computing nodes are also used to perform area-level time slot conflict detection and avoidance: Collect the wake-up phase parameters and wake-up duty cycle parameters of multiple carrier transceiver modules within this station area; A distribution map of wake-up time slots for a transformer area is constructed based on the wake-up phase parameters and wake-up duty cycle parameters of multiple carrier transceiver modules. Identify conflicting time slot regions in the wake-up time slot distribution map of the transformer area where the overlap of wake-up windows exceeds a preset overlap threshold; For multiple carrier transceiver modules whose wake-up windows overlap within the conflict time slot region, a wake-up phase fine-tuning amount is generated based on the data transmission priority of each carrier transceiver module, and the wake-up phase fine-tuning amount is sent to the corresponding carrier transceiver module.

[0015] Furthermore, the cross-regional coordination unit is used to coordinate the wake-up timing of boundary nodes in adjacent regions: Obtain the wake-up phase parameters and wake-up duty cycle parameters of the boundary nodes of this station area. The boundary nodes are nodes that have overlapping carrier signal coverage areas with the boundary nodes of adjacent stations. The adjacent wake-up phase parameters and adjacent wake-up duty cycle parameters of the boundary nodes of adjacent substations are obtained through the inter-substation communication link. Calculate the temporal overlap rate between the wake-up window of the boundary node of this transformer area and the wake-up window of the boundary node of the adjacent transformer area; When the time domain overlap rate exceeds the preset overlap tolerance threshold, a collaborative wake-up phase offset is generated based on the data transmission priority of the boundary node of this station area and the data transmission priority of the boundary node of the adjacent station area. The collaborative wake-up phase offset is sent to the boundary node on the side with lower data transmission priority. The boundary node on the side with lower data transmission priority adjusts the wake-up phase parameters for the next cycle based on the collaborative wake-up phase offset.

[0016] The beneficial effects of this invention compared to existing technologies are as follows: This invention constructs a three-dimensional relationship model based on the wake-up phase parameters, wake-up duty cycle parameters, and corresponding payload transmission success rate parameters of the edge computing node in multiple power frequency cycles within a historical period using the carrier transceiver module. Within this three-dimensional relationship model, a three-dimensional optimized boundary surface composed of non-dominated state points is determined, and the optimal approximation direction vector from the current working state point to the three-dimensional optimized boundary surface is calculated to generate wake-up phase adjustment and wake-up duty cycle adjustment amounts. Simultaneously, the relay control module calculates the matching wake-up phase value of the target relay node based on the first zero-crossing moment of the phase line where the current node is located, the second zero-crossing moment of the phase line where the target relay node is located, and the wake-up phase adjustment amount, and generates a relay wake-up command. This overcomes the fundamental defects of existing technologies, such as low cross-phase relay communication success rate and high network-level power consumption caused by the independent wake-up timing of the source node and relay node, which cannot be coordinated for optimization. It achieves the beneficial technical effects of significantly reducing single-node idle monitoring power consumption, greatly improving cross-phase relay communication success rate, and effectively reducing the overall network-level power consumption of the distribution area while ensuring end-to-end real-time communication.

[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a structural diagram of a power line carrier low-power communication system based on an edge computing module in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the construction process of the three-dimensional optimized boundary surface in an embodiment of the present invention. Figure 3 This is a flowchart of the cross-phase relay wake-up timing calculation in an embodiment of the present invention. Detailed Implementation

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] refer to Figure 1 , Figure 2 , Figure 3 This invention provides an embodiment of a power line carrier low-power communication system based on an edge computing module, comprising: Edge computing nodes are deployed on the side of the distribution network branch box or meter box and are electrically connected to the carrier transceiver module. The relay control module is electrically connected to the edge computing node and the carrier transceiver module, and the relay control module is equipped with a relay time slot arbitration unit; Edge computing nodes are equipped with topology migration identification units and cross-station collaboration units; Edge computing nodes are used for: The wake-up phase parameter, wake-up duty cycle parameter and corresponding payload transmission success rate parameter of the carrier transceiver module in multiple power frequency cycles within the historical period are obtained. The wake-up phase parameter represents the phase offset value of the wake-up window of the carrier transceiver module relative to the zero crossing time of the power frequency voltage of the phase line. A three-dimensional relational model is constructed based on the wake-up phase parameter, wake-up duty cycle parameter and payload transmission success rate parameter. In the three-dimensional relational model, a three-dimensional optimized boundary surface composed of non-dominated state points is determined. Obtain the current wake-up phase parameters and current wake-up duty cycle parameters of the carrier transceiver module, determine the current working state point in the three-dimensional relational model, and calculate the optimal approximation direction vector from the current working state point to the three-dimensional optimized boundary surface; Based on the components of the optimal approximation direction vector in the wake-up phase dimension and wake-up duty cycle dimension, wake-up phase adjustment amount and wake-up duty cycle adjustment amount are generated respectively. The wake-up duty cycle adjustment amount is used to adjust the wake-up window duration of the carrier transceiver module in the next power frequency cycle. The relay control module is used for: Obtain the phase line identifier information of the phase line where the target relay node is located; Based on the first zero-crossing moment of the phase line where this node is located, the second zero-crossing moment of the phase line where the target relay node is located, and the wake-up phase adjustment amount, calculate the matching wake-up phase value of the target relay node; The relay wake-up command is generated based on the matching wake-up phase value. The relay wake-up command is used to control the target relay node to enter the receiving state at the time indicated by the matching wake-up phase value.

[0022] In this embodiment, the edge computing node is deployed on the side of the distribution network branch box or meter box and is electrically connected to the carrier transceiver module. The edge computing node is a localized computing device deployed inside the branch box or meter box at the end of the distribution network, establishing a bidirectional electrical connection with the carrier transceiver module via a serial peripheral interface bus or a universal asynchronous transceiver. The carrier transceiver module is a power line carrier communication chip or module, directly coupled to the power line.

[0023] In this embodiment, the relay control module is electrically connected to the edge computing node and the carrier transceiver module, and the relay control module is configured with a relay time slot arbitration unit. The relay control module is a software functional module running inside the edge computing node, interacting with the edge computing node through an internal data bus and connected to the carrier transceiver module through control signal lines. The relay time slot arbitration unit is a sub-functional unit within the relay control module, responsible for making time slot resource allocation decisions when multiple source nodes simultaneously request the same target relay node.

[0024] In this embodiment, the edge computing node is configured with a topology migration identification unit and a cross-area coordination unit. The topology migration identification unit is a sub-functional unit within the edge computing node, responsible for monitoring changes in the topology of the carrier communication network within the area and adjusting the three-dimensional optimized boundary surface accordingly. The cross-area coordination unit is a sub-functional unit within the edge computing node, responsible for communicating with edge computing nodes in adjacent areas and coordinating the wake-up timing of boundary nodes in adjacent areas.

[0025] In this embodiment, the historical period refers to a continuous time period that has passed from the start of operation of the edge computing node to the current moment, with the length set to 24 hours, 48 ​​hours or 72 hours.

[0026] In this embodiment, the power frequency period refers to the time it takes for the AC power of the distribution network to complete one full sinusoidal change. For a 50 Hz power grid, it is 20 milliseconds, and for a 60 Hz power grid, it is 16.67 milliseconds.

[0027] In this embodiment, the wake-up phase parameter, wake-up duty cycle parameter, and corresponding payload transmission success rate parameter of the carrier transceiver module are acquired in multiple power frequency cycles within a historical period. The edge computing node extracts three data items corresponding to each successfully transmitted data communication event from the historical communication logs of the carrier transceiver module: the wake-up phase parameter, which is the phase offset value of the wake-up window relative to the zero-crossing moment of the power frequency voltage of the corresponding phase line, in milliseconds or electrical angles; the wake-up duty cycle parameter, which is the ratio of the wake-up window duration to the total duration of the power frequency cycle; and the corresponding payload transmission success rate parameter, which is the ratio of the successfully transmitted payload data amount to the total transmitted payload data amount. The edge computing node stores the three data items collected within the same power frequency cycle as a three-dimensional data group.

[0028] Furthermore, in one embodiment, to enable the 3D relational model to cover a more comprehensive range of operating states, the edge computing node is also used to perform parameter perturbation detection when preset conditions are met. For example, when the network load is below a preset threshold or a preset detection period is reached, the edge computing node controls the carrier transceiver module to temporarily deviate from the currently optimal wake-up phase parameter or wake-up duty cycle parameter, sending data with a tentative parameter combination and recording the corresponding payload transmission success rate. In this way, the boundaries of the 3D relational model can be continuously explored and updated without affecting normal communication, avoiding the model from getting trapped in local optima.

[0029] In this embodiment, the wake-up phase parameter represents the phase offset value of the wake-up window of the carrier transceiver module relative to the zero-crossing moment of the power frequency voltage of its respective phase line. The time interval between the start time of the carrier transceiver module switching from the sleep state to the listening state in each power frequency cycle and the positive zero-crossing moment of the power frequency voltage of its respective phase line is the wake-up phase parameter.

[0030] In this embodiment, a three-dimensional relationship model is constructed based on the wake-up phase parameter, wake-up duty cycle parameter, and payload transmission success rate parameter. Edge computing nodes map all three-dimensional data sets collected within a historical period as sample points onto a three-dimensional coordinate system. The first coordinate axis corresponds to the wake-up phase parameter, the second coordinate axis corresponds to the wake-up duty cycle parameter, and the third coordinate axis corresponds to the payload transmission success rate parameter. All sample points form a discrete point cloud distribution in the three-dimensional coordinate system, which constitutes the three-dimensional relationship model. The inputs to the three-dimensional relationship model are the wake-up phase parameter and the wake-up duty cycle parameter, and the output is the payload transmission success rate parameter.

[0031] In this embodiment, before constructing the three-dimensional relationship model, the edge computing node normalizes the collected wake-up phase parameter, wake-up duty cycle parameter, and payload transmission success rate parameter. The wake-up phase parameter is divided by the total duration of the power frequency cycle, and the wake-up duty cycle parameter and payload transmission success rate parameter are directly used with their original ratios, so that the values ​​of the three dimensions are all mapped to the range of zero to one.

[0032] In this embodiment, a non-dominated state point refers to a 3D sample point in the 3D relation model where no other 3D sample point is superior to it in all three dimensions. For two 3D sample points A and B, if A is closer to the preset ideal wake-up phase in the wake-up phase dimension, has a smaller wake-up duty cycle value in the wake-up duty cycle dimension, and a larger value in the payload transmission success rate dimension, then A is superior to B in all three dimensions. In other words, a non-dominated state point is a sample point in the 3D relation model where no other sample point can be superior to it in all three evaluation dimensions. All such non-dominated state points together constitute the basis of the 3D optimization boundary surface.

[0033] In this embodiment, a 3D optimized boundary surface composed of non-dominated state points is determined in the 3D relational model. The edge computing nodes employ a fast non-dominated sorting algorithm to hierarchically sort all 3D sample points in the 3D relational model, extracting all 3D sample points in the first non-dominated layer as the non-dominated solution set. Then, the Delaunay triangulation algorithm is used to connect the 3D sample points in the non-dominated solution set according to spatial proximity, forming a continuous 3D mesh surface composed of multiple triangular facets. This surface is the 3D optimized boundary surface.

[0034] In this embodiment, the current wake-up phase parameter and current wake-up duty cycle parameter of the carrier transceiver module are obtained, and the current working state point is determined in the three-dimensional relational model. The edge computing node reads the actual wake-up phase parameter configuration value and wake-up duty cycle parameter configuration value used by the carrier transceiver module in the current power frequency cycle, and locates the spatial coordinate point corresponding to the set of parameter values ​​in the three-dimensional coordinate system corresponding to the constructed three-dimensional relational model. This spatial coordinate point is the current working state point.

[0035] In this embodiment, the optimal approximation direction vector from the current working state point to the 3D optimized boundary surface is calculated. The edge computing node discretizes the 3D optimized boundary surface into multiple triangular patches, calculates the perpendicular distance from the spatial coordinates of the current working state point to each triangular patch, and finds the smallest value among all perpendicular distances. The foot of the perpendicular on the triangular patch corresponding to this smallest perpendicular distance is the target approximation point, and the directed line segment from the spatial coordinates of the current working state point to the target approximation point is the optimal approximation direction vector.

[0036] In this embodiment, wake-up phase adjustment and wake-up duty cycle adjustment are generated based on the components of the optimal approximation direction vector in the wake-up phase dimension and wake-up duty cycle dimension, respectively. The edge computing node projects the optimal approximation direction vector onto the coordinate axes of the wake-up phase dimension and wake-up duty cycle dimension, respectively, to obtain the projected lengths in the wake-up phase dimension and wake-up duty cycle dimension. The wake-up phase adjustment is obtained by multiplying the projected length in the wake-up phase dimension by a preset phase adjustment step size coefficient, and the wake-up duty cycle adjustment is obtained by multiplying the projected length in the wake-up duty cycle dimension by a preset duty cycle adjustment step size coefficient.

[0037] In this embodiment, the wake-up duty cycle adjustment amount is used to adjust the duration of the wake-up window of the carrier transceiver module in the next power frequency cycle. The edge computing node adds the current wake-up duty cycle parameter to the wake-up duty cycle adjustment amount to obtain the target wake-up duty cycle parameter for the next power frequency cycle, and writes the target wake-up duty cycle parameter into the wake-up configuration register of the carrier transceiver module.

[0038] In this embodiment, the phase line identification information of the phase line where the target relay node is located is obtained. When this node needs to forward data through the relay node, the edge computing node queries the phase line attribute field corresponding to the target relay node from the transformer area topology information table maintained by this node to obtain whether the phase line accessed by the target relay node is phase A, phase B, or phase C.

[0039] In this embodiment, the target relay node refers to an adjacent carrier communication node that is selected during carrier communication to receive the data frame sent by this node and forward the data frame to the destination node.

[0040] In this embodiment, "this node" refers to a carrier communication node that is currently performing edge computing node functions and needs to send a relay wake-up command to the target relay node.

[0041] In this embodiment, the first zero-crossing moment of the phase line where the current node is located refers to the moment when the power frequency voltage of the phase line where the current node is located crosses the zero voltage point from the negative half-cycle to the positive half-cycle within one power frequency cycle. The second zero-crossing moment of the phase line where the target relay node is located refers to the moment when the power frequency voltage of the phase line where the target relay node is located crosses the zero voltage point from the negative half-cycle to the positive half-cycle within one power frequency cycle. The edge computing node captures the zero-crossing moment of its own phase line through the zero-crossing detection circuit built into the carrier transceiver module, and looks up the zero-crossing moment of the target relay node's phase line through a preset three-phase zero-crossing moment correspondence table.

[0042] In this embodiment, the matching wake-up phase value of the target relay node is calculated based on the first zero-crossing time of the phase line where the current node is located, the second zero-crossing time of the phase line where the target relay node is located, and the wake-up phase adjustment amount. The edge computing node calculates the difference between the second zero-crossing time and the first zero-crossing time as the phase-to-phase zero-crossing offset, and adds the wake-up phase adjustment amount to the phase-to-phase zero-crossing offset to obtain the matching wake-up phase value of the target relay node. The preset three-phase zero-crossing time correspondence table is preset based on the principle that the phase-frequency voltage phases of phases A, B, and C in three-phase AC power are sequentially 120 degrees apart. After the edge computing node knows the phase line identifier where it is located, it can look up the zero-crossing time of any phase line through this table.

[0043] In this embodiment, a relay wake-up command is generated based on the matched wake-up phase value. This command controls the target relay node to enter the receiving state at the time indicated by the matched wake-up phase value. The relay control module encapsulates the matched wake-up phase value into a relay wake-up command data frame. The data frame contains the node identifier field of the target relay node and the wake-up time offset field. The carrier transceiver module sends the relay wake-up command data frame to the power line within the current wake-up window of the node. After receiving the data frame, the target relay node parses the value of the wake-up time offset field and, after the next zero-crossing of the power frequency on its own phase line, delays for the time length indicated by the value, switching the carrier transceiver module from the sleep state to the receiving state.

[0044] In this embodiment, after entering the receiving state, the target relay node starts a receiving state timer. The duration of the receiving state timer is equal to the duration of the wake-up window corresponding to the matched wake-up phase value. If the target relay node does not detect a valid carrier preamble before the receiving state timer expires, the target relay node exits the receiving state early and switches back to the sleep state after the receiving state timer expires.

[0045] Furthermore, edge computing nodes are also used to construct the confidence interval boundaries of the 3D optimized boundary surface, and to correct the wake-up phase adjustment and wake-up duty cycle adjustment based on the confidence interval boundaries: In the three-dimensional relational model, the density of three-dimensional sample points in the preset spatial neighborhood around each non-dominated state point is calculated. Non-dominated state points with a 3D sample point density not lower than a preset density threshold are marked as high-confidence boundary points; Based on the high-confidence boundary points, a confidence-optimized boundary surface is refitted and generated, and the average offset vector between the confidence-optimized boundary surface and the original 3D optimized boundary surface is calculated. The average offset vector is used as the boundary confidence correction vector. Before generating the wake-up phase adjustment and wake-up duty cycle adjustment, the boundary confidence correction vector is superimposed on the optimal approximation direction vector.

[0046] In this embodiment, the preset spatial neighborhood refers to a spherical spatial region defined in the three-dimensional coordinate system corresponding to the three-dimensional relationship model, centered on the spatial coordinates of a certain non-dominated state point and with a preset radius value. The preset radius value is set to 5% to 15% of the maximum range of values ​​for each coordinate axis.

[0047] In this embodiment, in the three-dimensional relationship model, the density of three-dimensional sample points within a preset spatial neighborhood around each non-dominated state point is calculated. For each non-dominated state point, the edge computing node uses the spatial coordinates of the non-dominated state point as the center and the radius of the preset spatial neighborhood as the search radius to count the total number of all three-dimensional sample points falling into the spherical spatial region. The total number is then divided by the volume of the sphere in the preset spatial neighborhood, and the quotient obtained is the three-dimensional sample point density.

[0048] In this embodiment, the preset density threshold is a three-dimensional sample point density determination threshold set in advance by the edge computing node, which is set to one-fifth to one-third of the number of three-dimensional sample points collected by the carrier transceiver module in a single power frequency cycle within the substation area.

[0049] In this embodiment, a confidence-optimized boundary surface is generated by refitting based on high-confidence boundary points. The edge computing node retains non-dominated state points whose 3D sample point density is not lower than a preset density threshold, and removes non-dominated state points whose 3D sample point density is lower than the preset density threshold. A local weighted regression scatter smoothing algorithm is used to fit the retained non-dominated state points to generate a confidence-optimized boundary surface.

[0050] In this embodiment, the average offset vector between the confidence-optimized boundary surface and the original 3D optimization boundary surface is calculated. Edge computing nodes divide the two-dimensional bottom surface, which is composed of the wake-up phase dimension and the wake-up duty cycle dimension, into uniform rectangular grids. For each grid intersection, the difference between the payload transmission success rate parameter value corresponding to the confidence-optimized boundary surface and the corresponding payload transmission success rate parameter value corresponding to the original 3D optimization boundary surface is calculated. The arithmetic mean of the differences corresponding to all grid intersections is then calculated. This average offset is transformed into a 3D vector pointing from the original 3D optimization boundary surface to the confidence-optimized boundary surface, thus obtaining the average offset vector.

[0051] In this embodiment, the average offset vector is used as the boundary confidence correction vector. Before generating the wake-up phase adjustment and wake-up duty cycle adjustment, the boundary confidence correction vector is superimposed on the optimal approximation direction vector. The edge computing node performs vector addition on the optimal approximation direction vector and the boundary confidence correction vector to obtain the corrected optimal approximation direction vector. Then, the corrected optimal approximation direction vector is projected onto the wake-up phase dimension coordinate axis and the wake-up duty cycle dimension coordinate axis, respectively, to generate the wake-up phase adjustment and the wake-up duty cycle adjustment.

[0052] Furthermore, edge computing nodes are also used to construct a cross-phase noise period distribution map and correct the matched wake-up phase value of the target relay node based on the cross-phase noise period distribution map: Obtain the noise energy sampling sequences of phases A, B, and C within this substation area in multiple power frequency cycles. The noise energy sampling sequence represents the noise energy value collected at a preset sampling interval within one power frequency cycle. The noise energy sampling sequences of phases A, B, and C are aligned according to their respective zero-crossing times and then superimposed to generate a three-phase composite noise envelope curve. Identify low-noise regions where the noise energy is below a preset noise threshold in the three-phase synthesized noise envelope curve; When the target relay node and this node are located on different phase lines, determine whether the matched wake-up phase value of the target relay node falls into the low noise range; If the matched wake-up phase value of the target relay node does not fall into the low noise range, the phase offset compensation between the matched wake-up phase value and the nearest low noise range boundary is calculated, and the matched wake-up phase value is corrected based on the phase offset compensation.

[0053] In this embodiment, "this distribution area" refers to the set of all power line carrier communication nodes covered by a low-voltage distribution network powered by the same distribution transformer.

[0054] In this embodiment, the preset sampling interval refers to the time interval between two adjacent samples when sampling noise energy within one power frequency cycle, and is set to forty microseconds, eighty microseconds, or one hundred microseconds.

[0055] In this embodiment, the noise energy value refers to the energy intensity of the background electromagnetic interference signal on the power line other than the useful carrier signal, as measured by the carrier transceiver module in the receiving state, and the unit is decibel milliwatt or millivolt.

[0056] In this embodiment, the zero-crossing moment of the power frequency refers to the moment when the power frequency voltage crosses the zero voltage point from the negative half-cycle to the positive half-cycle within one power frequency cycle. The carrier transceiver module has a built-in zero-crossing detection circuit. When the instantaneous value of the power frequency voltage changes from less than zero to greater than zero, it outputs a zero-crossing pulse signal. The moment corresponding to the rising edge of this zero-crossing pulse signal is the zero-crossing moment of the power frequency.

[0057] In this embodiment, the noise energy sampling sequences of phases A, B, and C are aligned according to their respective power frequency zero-crossing times and then superimposed to generate a three-phase composite noise envelope curve. Edge computing nodes obtain the noise energy sampling sequences reported by all carrier transceiver modules on phase A within their local area and calculate the arithmetic mean at the same sampling position to obtain the average noise energy sampling sequence of phase A. Similarly, the average noise energy sampling sequences of phases B and C are obtained. The first sampling point of the average noise energy sampling sequence of phase A is aligned to the power frequency zero-crossing time of phase A, the first sampling point of the average noise energy sampling sequence of phase B is aligned to the power frequency zero-crossing time of phase B, and the first sampling point of the average noise energy sampling sequence of phase C is aligned to the power frequency zero-crossing time of phase C. The envelope value of the noise energy values ​​of the three phase lines at the same time is taken to form the three-phase composite noise envelope curve.

[0058] In this embodiment, the preset noise threshold is a noise energy determination threshold used to determine whether the power line channel is in a low noise state, and is the sum of the receiving sensitivity value of the carrier transceiver module and the preset signal-to-noise ratio margin.

[0059] In this embodiment, noise energy refers to the instantaneous energy intensity value of the background electromagnetic interference signal on the power line, which has the same meaning as noise energy value.

[0060] In this embodiment, low-noise intervals with noise energy below a preset noise threshold are identified in the three-phase composite noise envelope curve. Edge computing nodes traverse all sampling points of the three-phase composite noise envelope curve within a complete power frequency cycle, marking time periods where the noise energy value is continuously below the preset noise threshold as candidate low-noise intervals. For each candidate low-noise interval, the duration of the interval is calculated, and candidate low-noise intervals whose duration is greater than the sum of the minimum time required for the carrier transceiver module to complete one complete data transmission and reception cycle and a preset protection margin are retained as low-noise intervals.

[0061] In this embodiment, the matched wake-up phase value of the target relay node refers to the time offset that the target relay node needs to wait for after the power frequency voltage of its phase line crosses zero, calculated according to the method described in the foregoing embodiment.

[0062] In this embodiment, the phase offset compensation between the matched wake-up phase value and the nearest low-noise interval boundary is calculated, and the matched wake-up phase value is corrected based on the phase offset compensation. The edge computing node determines whether the matched wake-up phase value of the target relay node falls within a certain low-noise interval. If it does not, the time difference between the matched wake-up phase value and the nearest low-noise interval boundary is calculated, and this time difference is taken as the phase offset compensation. The matched wake-up phase value and the phase offset compensation are added together to obtain the corrected matched wake-up phase value.

[0063] In this embodiment, the nearest low-noise interval boundary refers to the left or right boundary moment of the low-noise interval that is closest to the matched wake-up phase value of the target relay node on the time axis.

[0064] In this embodiment, cross-phase communication refers to carrier communication when the local node and the target relay node are connected to different phase lines. Because power line carrier signals need to pass through an inter-phase coupling path during cross-phase transmission, the signal attenuation is 10 to 20 decibels greater than in-phase communication. When the local node and the target relay node are on different phase lines, the edge computing node first determines whether the estimated link attenuation between them is less than a preset cross-phase communication attenuation threshold before calculating the matched wake-up phase value. If the estimated link attenuation is less than the preset cross-phase communication attenuation threshold, the subsequent matched wake-up phase value calculation continues; if the estimated link attenuation is not less than the preset cross-phase communication attenuation threshold, the current target relay node is abandoned and a new in-phase relay node is selected.

[0065] Furthermore, the relay control module is also used to adaptively adjust the preamble length of the relay wake-up command based on the cross-phase noise difference: Obtain the first noise energy estimate of the phase line where this node is located at the matched wake-up phase value; Obtain the second noise energy estimate at the matched wake-up phase value of the phase line where the target relay node is located; The ratio of the first noise energy estimate to the second noise energy estimate is calculated and used as the cross-phase noise difference coefficient. The target preamble length is obtained by querying the preset preamble length mapping table based on the cross-phase noise difference coefficient. The target preamble length is encapsulated in the relay wake-up command. After receiving the relay wake-up command, the target relay node sends the wake-up preamble sequence with the target preamble length.

[0066] In this embodiment, the first noise energy estimate of the phase line where the local node is located at the matched wake-up phase value is obtained. The edge computing node locates the time offset position corresponding to the matched wake-up phase value from the average noise energy sampling sequence of the phase line where the local node is located, extracts the noise energy values ​​of each preset number of sampling points before and after the time offset position, and calculates the arithmetic mean to obtain the first noise energy estimate.

[0067] In this embodiment, a second noise energy estimate is obtained at the matched wake-up phase value of the phase line where the target relay node is located. The edge computing node obtains the second noise energy estimate from the average noise energy sampling sequence of the phase line where the target relay node is located using the same method as obtaining the first noise energy estimate.

[0068] In this embodiment, the ratio of the first noise energy estimate to the second noise energy estimate is calculated as the cross-phase noise difference coefficient.

[0069] In this embodiment, the preset preamble length mapping table is a two-dimensional lookup table pre-stored by the edge computing node. The row index is the segmented interval of the phase noise difference coefficient, the column index is the baseline preamble length level, and the table entries store the recommended preamble length values, in bytes or milliseconds.

[0070] In this embodiment, the target preamble length is obtained by querying a preset preamble length mapping table based on the phase noise difference coefficient. The edge computing node determines the reference preamble length level used in the current communication and uses the table entry corresponding to the segment interval into which the phase noise difference coefficient falls as the target preamble length.

[0071] In this embodiment, the relay wake-up instruction refers to the data frame generated by the relay control module based on the matching wake-up phase value in the aforementioned embodiment, which is used to control the target relay node to enter the receiving state at a specified time. The data frame includes a node identifier field, a wake-up time offset field, and a preamble length indication field.

[0072] In this embodiment, the target preamble length is encapsulated within the relay wake-up command. Upon receiving the relay wake-up command, the target relay node sends a wake-up preamble sequence with the target preamble length. The relay control module writes the target preamble length into the preamble length indication field of the relay wake-up command data frame. After receiving the relay wake-up command, the target relay node parses the contents of the wake-up time offset field and the preamble length indication field, and sends the wake-up preamble sequence according to the length value indicated by the preamble length indication field when the wake-up time arrives.

[0073] Furthermore, the relay slot arbitration unit is used to allocate time slots when multiple source nodes simultaneously request the same target relay node: When it is detected that both the first source node and the second source node send a relay wake-up request to the same target relay node within a preset time window, the first data transmission urgency score of the first source node and the second data transmission urgency score of the second source node are obtained. Obtain the first wake-up duration requested by the first source node and the second wake-up duration requested by the second source node; A first competition weight vector is constructed based on a first data transmission urgency score and a first wake-up duration, and a second competition weight vector is constructed based on a second data transmission urgency score and a second wake-up duration. Calculate the weighted magnitude of the first and second competition weight vectors in the two-dimensional competition space, and determine the source node with the larger weighted magnitude as the arbitration winner node; Assign the current wake-up window of the target relay node to the winning node in the arbitration, and send a slot delay indication to the unwinning node in the arbitration. The slot delay indication contains a delay phase offset value that suggests re-requesting.

[0074] In this embodiment, the first source node refers to one of the carrier communication nodes that sends a relay wake-up request to the same target relay node within a preset time window. The second source node refers to another carrier communication node that sends a relay wake-up request to the same target relay node within the preset time window.

[0075] In this embodiment, the preset time window refers to the length of the time interval set by the relay time slot arbitration unit for collecting relay wake-up requests for the same target relay node, which is set to one-fifth to one-third of the wake-up cycle length of the target relay node.

[0076] In this embodiment, it is detected that both the first source node and the second source node send relay wake-up requests to the same target relay node within a preset time window. The relay time slot arbitration unit maintains a request registration table. When two or more relay wake-up requests for the same target relay node are registered within the same preset time window, it is determined that multi-source contention has occurred, and the time slot arbitration process is triggered.

[0077] In this embodiment, if the relay time slot arbitration unit receives only one relay wake-up request for the target relay node within a preset time window, the source node of that request is directly determined as the arbitration winner. If three or more relay wake-up requests for the same target relay node are received within the preset time window, the relay time slot arbitration unit pairs all requesting nodes and sequentially uses the aforementioned competition weight vector comparison method to determine the arbitration winner. If the relay time slot arbitration unit times out during the arbitration process, the first-come, first-served rule is used by default to determine the arbitration winner.

[0078] In this embodiment, a first data transmission urgency score for the first source node and a second data transmission urgency score for the second source node are obtained. The data transmission urgency score is a quantitative value calculated by the source node based on three factors: the service type of the data frame to be transmitted, the remaining validity period, and the maximum tolerable delay. The highest score is for distribution automation protection action data, followed by the score for electricity meter timed reading data, and the lowest score is for equipment status inspection data. The shorter the remaining validity period and the smaller the maximum tolerable delay, the higher the score. The edge computing node extracts the above three attribute information of the data frame to be transmitted from the data transmission queue of the source node and calculates the data transmission urgency score according to a preset weighted scoring formula. The first data transmission urgency score corresponds to the first source node, and the second data transmission urgency score corresponds to the second source node.

[0079] In this embodiment, the first wake-up duration requested by the first source node and the second wake-up duration requested by the second source node are obtained. The wake-up duration is a parameter attached by the source node when sending a relay wake-up request to the target relay node, representing the length of time the source node expects to occupy the target relay node's receiving state. The value of the wake-up duration is calculated by the source node based on the total length of the data frame to be transmitted and the current carrier communication rate. The first wake-up duration corresponds to the first source node, and the second wake-up duration corresponds to the second source node.

[0080] In this embodiment, a first contention weight vector is constructed based on a first data transmission urgency score and a first wake-up duration. The relay slot arbitration unit uses the first data transmission urgency score as the value of the horizontal axis of the two-dimensional contention space and the first wake-up duration as the value of the vertical axis of the two-dimensional contention space to construct a two-dimensional vector with the origin as the starting point and the horizontal and vertical coordinate values ​​as the ending points. This two-dimensional vector is the first contention weight vector.

[0081] In this embodiment, a second contention weight vector is constructed based on a second data transmission urgency score and a second wake-up duration. The relay slot arbitration unit uses the second data transmission urgency score as the value of the horizontal axis of the two-dimensional contention space and the second wake-up duration as the value of the vertical axis of the two-dimensional contention space to construct a two-dimensional vector with the origin as the starting point and the horizontal and vertical coordinate values ​​as the ending points. This two-dimensional vector is the second contention weight vector.

[0082] In this embodiment, the two-dimensional contention space refers to a two-dimensional Cartesian coordinate system consisting of a data transmission urgency score as the horizontal axis and a wake-up duration as the vertical axis. Each source node's contention request can be represented in this two-dimensional contention space as a vector originating from the origin. The direction of the vector reflects the relative proportion of urgency to occupancy duration, and the length of the vector reflects the overall contention intensity.

[0083] In this embodiment, the weighted magnitudes of the first and second competition weight vectors in the two-dimensional competition space are calculated, and the source node with the larger weighted magnitude is determined as the winning node in arbitration. The relay time slot arbitration unit calculates the magnitudes of the first and second competition weight vectors, respectively. The magnitude is equal to the square root of the sum of the squares of the horizontal and vertical coordinates. Then, the magnitude of the first competition weight vector is multiplied by a preset urgency weight coefficient to obtain the weighted magnitude of the first competition weight vector; the magnitude of the second competition weight vector is also multiplied by a preset urgency weight coefficient to obtain the weighted magnitude of the second competition weight vector. The two weighted magnitudes are compared, and the source node with the larger weighted magnitude is determined as the winning node in arbitration, while the source node with the smaller weighted magnitude is determined as the losing node in arbitration. If the two weighted magnitudes are equal, the source node with the higher data transmission urgency score is preferentially determined as the winning node in arbitration.

[0084] In this embodiment, the current wake-up window of the target relay node is allocated to the winning node in the arbitration, and a time slot delay indication is sent to the unsuccessful nodes in the arbitration. The time slot delay indication includes a delayed phase offset value suggesting a re-request. The relay time slot arbitration unit returns an arbitration success acknowledgment message to the winning node, allowing it to transmit data within the current wake-up window of the target relay node. Simultaneously, a time slot delay indication message is returned to the unsuccessful nodes in the arbitration. This message contains a delayed phase offset value suggesting a re-request, which is equal to the wake-up duration of the winning node plus a preset guard interval. After receiving the time slot delay indication, the unsuccessful node in the arbitration re-initiates a relay wake-up request to the same target relay node after a delay of the delayed phase offset value.

[0085] Furthermore, the relay slot arbitration unit is also used to perform preemptive slot allocation when the data transmission urgency score of the node that failed the arbitration exceeds a preset urgency threshold: The current wake-up window of the target relay node is divided into a basic time slot segment and an extended time slot segment, with the basic time slot segment allocated to the arbitration winner node. Calculate the minimum required wake-up time for the node that did not win the arbitration. If the length of the extended time slot is not less than the minimum required wake-up time, then allocate the extended time slot to the node that did not win the arbitration. Send a wake-up window extension command to the target relay node. The wake-up window extension command is used to control the target relay node to continue receiving after the basic time slot ends until the extended time slot ends.

[0086] In this embodiment, the current wake-up window of the target relay node is divided into a basic time slot segment and an extended time slot segment. The current wake-up window refers to the entire duration during which the target relay node switches from a sleep state to a receiving state after receiving a relay wake-up command. The duration of the basic time slot segment is determined based on the duration of the first wake-up request from the arbitrated winning node, and the duration of the extended time slot segment is equal to the total duration of the current wake-up window minus the duration of the basic time slot segment.

[0087] In this embodiment, the basic time slot is allocated to the winning node in arbitration. The winning node exclusively uses the receiving resources of the target relay node within the basic time slot.

[0088] In this embodiment, the minimum required wake-up time for the node that failed to win the arbitration is calculated. The minimum required wake-up time is calculated by the node that failed to win the arbitration based on the length of the highest priority critical data portion in the data frame to be transmitted and the current carrier communication rate.

[0089] In this embodiment, if the length of the extended time slot is not less than the minimum required wake-up time, the extended time slot is allocated to the node that did not win the arbitration. If the length of the extended time slot is less than the minimum required wake-up time, the relay time slot arbitration unit does not allocate the extended time slot and returns a time slot postponement indication to the node that did not win the arbitration.

[0090] In this embodiment, the wake-up window extension instruction is used to control the target relay node to continue its receiving state after the basic time slot segment ends until the extended time slot segment ends. The relay time slot arbitration unit generates a wake-up window extension instruction data frame, which includes a target relay node identifier field and an extended time slot duration field. After receiving the instruction, the target relay node will continue its receiving state for the extended time slot duration after the basic time slot segment ends.

[0091] Furthermore, the topology transfer recognition unit is used to detect topological changes in the transformer area and transfer the 3D optimized boundary surface: Periodically acquire neighbor node beacon reception records of all carrier transceiver modules within this station area. The neighbor node beacon reception records include the neighbor node identifier of the received beacon and the corresponding received signal strength indication value. Construct a sequence of adjacent area adjacency matrices based on neighbor node beacon reception records over multiple periods; Calculate the Frobenius norm difference of the adjacency matrix of adjacent transformer areas between adjacent cycles. When the Frobenius norm difference exceeds the preset topology change threshold, it is determined that a transformer area topology change has occurred. After determining that a topological change has occurred in the transformer area, the similarity deviation between the first three-dimensional optimized boundary surface before the topological change and the second three-dimensional optimized boundary surface reconstructed after the topological change is calculated. If the similarity deviation exceeds the preset surface similarity threshold, a weighted fusion is performed based on the first 3D optimized boundary surface and the second 3D optimized boundary surface to generate a migrated 3D optimized boundary surface. The migrated 3D optimized boundary surface replaces the first 3D optimized boundary surface, and the replaced 3D optimized boundary surface is used to calculate the optimal approximation direction vector of the current working state point.

[0092] In this embodiment, the neighbor node identifier and corresponding received signal strength indication value are received from the beacon. The carrier transceiver module periodically sends beacon frames to the power line, each carrying the node identifier of the sending node. Upon receiving the beacon frame, other carrier transceiver modules record the sending node's node identifier as a neighbor node identifier and simultaneously measure the received signal strength indication value in decibels and milliwatts. The carrier transceiver module periodically reports the neighbor node identifier and received signal strength indication value as a neighbor node beacon reception record to the edge computing node.

[0093] In this embodiment, all carrier transceiver modules within the local distribution area refer to all carrier communication modules that have been deployed and are in operation in a low-voltage distribution network powered by the same distribution transformer.

[0094] In this embodiment, a sequence of substation adjacency matrices is constructed based on beacon reception records of neighboring nodes over multiple periods. The topology migration identification unit constructs a substation adjacency matrix for each statistical period, typically set to 15, 30, or 60 minutes. The substation adjacency matrix is ​​a square matrix, with both the number of rows and columns equal to the total number of carrier transceiver modules within the substation. Row indices correspond to beacon receiving nodes, and column indices correspond to beacon transmitting nodes. For the element in the i-th row and j-th column of the matrix, if the i-th carrier transceiver module successfully receives the beacon frame transmitted by the j-th carrier transceiver module within the statistical period, the value of this element is the normalized value of the corresponding received signal strength indication; if the i-th carrier transceiver module does not receive the beacon frame transmitted by the j-th carrier transceiver module within the statistical period, the value of this element is zero. The topology migration identification unit arranges the substation adjacency matrices constructed over multiple consecutive statistical periods in chronological order to form a sequence of substation adjacency matrices.

[0095] In this embodiment, the Frobenius norm difference of the adjacency matrix of adjacent transformer stations is calculated. The topology migration identification unit calculates the difference matrix between the adjacency matrix of the next transformer station and the adjacency matrix of the previous transformer station, and calculates the Frobenius norm of the difference matrix. The Frobenius norm is equal to the square root of the sum of the squares of all elements in the difference matrix.

[0096] In this embodiment, the preset topology change threshold is a judgment threshold value pre-set by the topology migration recognition unit.

[0097] In this embodiment, a change in transformer area topology is determined. The topology migration identification unit compares the calculated Frobenius norm difference of the adjacency matrix of transformer areas between adjacent periods with a preset topology change threshold. If the Frobenius norm difference for three consecutive statistical periods exceeds the preset topology change threshold, it is determined that a substantial change has occurred in the transformer area topology.

[0098] In one embodiment, the similarity deviation is the Hausdorff distance. The Hausdorff distance is defined as the larger of the maximum value of the nearest distances from each point on the first 3D optimization boundary surface to the second 3D optimization boundary surface, and the maximum value of the nearest distances from each point on the second 3D optimization boundary surface to the first 3D optimization boundary surface.

[0099] In another embodiment, to reduce computational overhead, the similarity deviation can also be the root mean square or the maximum absolute value of the success rate difference between the two surfaces at multiple preset key points.

[0100] In this embodiment, the preset surface similarity threshold is a judgment threshold value pre-set by the topology migration recognition unit.

[0101] In this embodiment, a migration 3D optimized boundary surface is generated by weighted fusion of the first and second 3D optimized boundary surfaces. If the similarity deviation exceeds a preset surface similarity threshold, the topology migration recognition unit takes a uniform rectangular grid on the 2D bottom surface. For each grid intersection, it reads the corresponding payload transmission success rate parameter value on the first and second 3D optimized boundary surfaces and performs a weighted average according to a preset fusion weight coefficient. The fusion weight coefficient is dynamically determined based on the number of new sample points accumulated after the topology change.

[0102] In this embodiment, the first three-dimensional optimized boundary surface is replaced with a migrated three-dimensional optimized boundary surface. The replaced three-dimensional optimized boundary surface is used to calculate the optimal approximation direction vector of the current working state point. The topology migration identification unit writes the migrated three-dimensional optimized boundary surface into the three-dimensional relationship model, covering the original first and second three-dimensional optimized boundary surfaces.

[0103] Furthermore, the topology migration identification unit is also used to perform incremental optimized surface expansion when it is determined that a topology change has occurred in the transformer area and the type of topology change is the addition of a new node: Identify the access location of the new node in the transformer area topology and its adjacent existing nodes; Obtain the 3D optimized boundary surface of adjacent existing nodes as the reference boundary surface; Acquire initial sample points of wake-up phase parameters, wake-up duty cycle parameters, and payload transmission success rate parameters collected during the initialization phase of the newly added node; Based on the local curvature characteristics of the reference boundary surface, the initial sample points are projected onto the reference boundary surface along the normal vector direction to obtain the projection points; The projected points are weighted and averaged with the initial sample points to generate the initial 3D optimized boundary surface of the new node. The initial 3D optimized boundary surface of the new node is used to calculate the optimal approximation direction vector in subsequent runs.

[0104] In this embodiment, the topology change type is adding a node. Adding a node refers to the situation where a carrier transceiver module that was not originally in the node registration table of this substation area connects to the power line for the first time and begins to transmit beacon frames.

[0105] In this embodiment, the access location of the new node in the transformer area topology and its adjacent existing nodes are identified. The topology migration identification unit extracts all existing nodes that the new node can receive its beacon frames from the neighbor node beacon reception records initially reported by the new node, and finds all existing nodes that can receive the new node's beacon frames from the records reported by other existing nodes, and determines these existing nodes as adjacent existing nodes.

[0106] In this embodiment, the 3D optimized boundary surfaces of adjacent existing nodes are obtained as reference boundary surfaces. The topology migration identification unit reads the 3D optimized boundary surface data currently being used by adjacent existing nodes from the 3D relationship model storage area of ​​the edge computing node. If there are multiple adjacent existing nodes, the 3D optimized boundary surface of the adjacent existing node with the longest online runtime is selected as the reference boundary surface.

[0107] In this embodiment, the three-dimensional optimized boundary surface of adjacent existing nodes refers to the three-dimensional optimized boundary surface constructed and continuously updated and maintained by adjacent existing nodes based on the three-dimensional sample points collected within their own historical period according to the method described in claim 1.

[0108] In this embodiment, initial sample points are acquired from the wake-up phase parameters, wake-up duty cycle parameters, and payload transmission success rate parameters collected by the new node during the initialization phase. After accessing the substation area, the new node first enters the initialization phase, which lasts for 20 to 50 power frequency cycles. Communication is performed using preset fixed wake-up phase parameters and fixed wake-up duty cycle parameters, and the three parameters corresponding to each communication are recorded. After the initialization phase ends, the new node reports all the recorded three-dimensional sample points as initial sample points.

[0109] In this embodiment, the local curvature characteristics of the reference boundary surface refer to the geometric characteristics of the curvature degree and direction of the reference boundary surface in the region near the wake-up phase parameter and wake-up duty cycle parameter corresponding to each initial sample point of the newly added node, including Gaussian curvature and average curvature.

[0110] In this embodiment, the normal vector direction of the reference boundary surface refers to the direction vector perpendicular to the tangent plane at any point on the reference boundary surface.

[0111] In this embodiment, based on the local curvature features of the reference boundary surface, the initial sample point is projected onto the reference boundary surface along the normal vector direction to obtain the projection point. For each initial sample point, the topology transfer recognition unit finds the position on the reference boundary surface where the wake-up phase parameter and wake-up duty cycle parameter are the same, calculates the normal vector direction at that position, and projects the initial sample point onto the reference boundary surface along the normal vector direction; the intersection point is the projection point.

[0112] In this embodiment, the projected points and initial sample points are weighted and averaged to generate the initial 3D optimized boundary surface of the newly added node. The topology migration recognition unit weights and averages the payload transmission success rate parameter values ​​of the projected points and the initial sample points according to a preset initial fusion weight coefficient, and uses the Delaunay triangulation algorithm to connect the processed data points into a continuous 3D mesh surface. The weight coefficient of the projected points is dynamically determined according to the number of initial sample points of the newly added node; the fewer the number of initial sample points, the larger the weight coefficient of the projected points.

[0113] In this embodiment, the initial 3D optimized boundary surface of the newly added node is used by the newly added node to calculate the optimal approximation direction vector in subsequent operations. The topology migration identification unit sends the initial 3D optimized boundary surface data of the newly added node to the edge computing nodes of the newly added node, and the newly added node uses this surface as its own 3D optimized boundary surface.

[0114] Furthermore, edge computing nodes are also used to perform area-level time slot conflict detection and avoidance: Collect the wake-up phase parameters and wake-up duty cycle parameters of multiple carrier transceiver modules within this station area; A distribution map of wake-up time slots for a transformer area is constructed based on the wake-up phase parameters and wake-up duty cycle parameters of multiple carrier transceiver modules. Identify conflicting time slot regions in the wake-up time slot distribution map of the transformer area where the overlap of wake-up windows exceeds a preset overlap threshold; For multiple carrier transceiver modules whose wake-up windows overlap within the conflict time slot region, a wake-up phase fine-tuning amount is generated based on the data transmission priority of each carrier transceiver module, and the wake-up phase fine-tuning amount is sent to the corresponding carrier transceiver module.

[0115] In this embodiment, the wake-up phase parameters and wake-up duty cycle parameters of multiple carrier transceiver modules within the local area are collected. The edge computing node sends a parameter reporting request command to all registered carrier transceiver modules within the local area via power line carrier communication. Each carrier transceiver module encapsulates its currently effective wake-up phase parameter configuration value and wake-up duty cycle parameter configuration value in a parameter reporting response frame and returns it.

[0116] In this embodiment, a wake-up time slot distribution map of the distribution area is constructed based on the wake-up phase parameters and wake-up duty cycle parameters of multiple carrier transceiver modules. The edge computing nodes use the power frequency cycle time axis as the horizontal axis and the carrier transceiver module node identifier as the vertical axis. For each carrier transceiver module, the start position of the wake-up window is determined according to the wake-up phase parameters, and the duration of the wake-up window is determined according to the wake-up duty cycle parameters. Rectangular blocks are then drawn to form the wake-up time slot distribution map of the distribution area.

[0117] In this embodiment, the preset overlap threshold is a judgment threshold value pre-set by the edge computing node, which is set to 30% to 50%.

[0118] In this embodiment, conflicting time slot regions where the wake-up window overlap exceeds a preset overlap threshold are identified in the regional wake-up time slot distribution map. Edge computing nodes compare the wake-up windows of any two carrier transceiver modules pairwise, calculate the overlap time length, and divide the overlap time length by the duration of the shorter of the two wake-up windows to obtain the wake-up window overlap. If the wake-up window overlap exceeds the preset overlap threshold, the interval is marked as a conflicting time slot region.

[0119] In this embodiment, the wake-up window overlap refers to the ratio of the length of time during which the wake-up windows of the two carrier transceiver modules overlap on the power frequency cycle time axis to the shorter duration of the two wake-up windows.

[0120] In this embodiment, for multiple carrier transceiver modules whose wake-up windows overlap within a conflict time slot region, a wake-up phase fine-tuning amount is generated based on the data transmission priority of each carrier transceiver module, and the wake-up phase fine-tuning amount is sent to the corresponding carrier transceiver module. The edge computing node identifies the carrier transceiver module with the lowest data transmission priority as the object to be adjusted, calculates the time length of the overlapping part plus the preset guard interval as the wake-up phase fine-tuning amount, encapsulates it in a parameter adjustment command, and sends it to the corresponding carrier transceiver module. The carrier transceiver module adds the wake-up phase fine-tuning amount to its own wake-up phase parameters.

[0121] In this embodiment, the data transmission priority of each carrier transceiver module is determined according to the service type: Level 1 is distribution automation protection action data, Level 2 is remote fee control command data from electricity meters, Level 3 is real-time electricity consumption information collection data, and Level 4 is equipment status inspection and version upgrade data.

[0122] Furthermore, the cross-regional coordination unit is used to coordinate the wake-up timing of boundary nodes in adjacent regions: Obtain the wake-up phase parameters and wake-up duty cycle parameters of the boundary nodes of this station area. The boundary nodes are nodes that have overlapping carrier signal coverage areas with the boundary nodes of adjacent stations. The adjacent wake-up phase parameters and adjacent wake-up duty cycle parameters of the boundary nodes of adjacent substations are obtained through the inter-substation communication link. Calculate the temporal overlap rate between the wake-up window of the boundary node of this transformer area and the wake-up window of the boundary node of the adjacent transformer area; When the time domain overlap rate exceeds the preset overlap tolerance threshold, a collaborative wake-up phase offset is generated based on the data transmission priority of the boundary node of this station area and the data transmission priority of the boundary node of the adjacent station area. The collaborative wake-up phase offset is sent to the boundary node on the side with lower data transmission priority. The boundary node on the side with lower data transmission priority adjusts the wake-up phase parameters for the next cycle based on the collaborative wake-up phase offset.

[0123] In this embodiment, the wake-up phase parameters and wake-up duty cycle parameters of the local area boundary node are obtained. The boundary node is a node with a carrier signal overlap coverage area with the boundary node of the adjacent area. The edge computing node marks the local area carrier transceiver module that detects the beacon frame of the adjacent area carrier transceiver module within its communication range as the local area boundary node.

[0124] In this embodiment, the inter-transformer communication link refers to the data transmission channel connecting the edge computing node of this transformer area with the edge computing node of the adjacent transformer area. It is established by utilizing the existing optical fiber communication network, wireless public network communication module or power line carrier cross-transformer coupling communication method of the power distribution network.

[0125] In this embodiment, the adjacent wake-up phase parameters and adjacent wake-up duty cycle parameters of the boundary nodes of adjacent substations are obtained through the inter-substation communication link. The edge computing node of this substation sends a boundary node parameter query request to the edge computing nodes of the adjacent substations, and the edge computing nodes of the adjacent substations return the corresponding adjacent wake-up phase parameters and adjacent wake-up duty cycle parameters.

[0126] In this embodiment, the temporal overlap rate between the wake-up window of the local boundary node and the wake-up window of the adjacent boundary node is calculated. The edge computing node calculates the overlap time length of the two wake-up windows on the time axis, divides it by the duration of the shorter of the two wake-up windows, and obtains the temporal overlap rate.

[0127] In this embodiment, the preset overlap tolerance threshold is a judgment threshold value pre-set by the cross-station collaborative unit, which is set to 20% to 40%.

[0128] In this embodiment, the data transmission priority of the local area boundary node is determined by the local area edge computing node according to the service type, and the data transmission priority of the adjacent area boundary node is determined by the adjacent area edge computing node and returned in the query response.

[0129] In this embodiment, a collaborative wake-up phase offset is generated based on the data transmission priority of the boundary nodes of the local substation and the data transmission priority of the boundary nodes of adjacent substations. The cross-substation collaborative unit identifies the boundary node with the lower data transmission priority as the object to be adjusted, and calculates the overlap time length plus the preset inter-substation protection interval as the base value of the collaborative wake-up phase offset. If the boundary node with the lower data transmission priority is the local substation boundary node, the collaborative wake-up phase offset is positive; if the boundary node with the lower data transmission priority is the adjacent substation boundary node, the collaborative wake-up phase offset is negative.

[0130] In this embodiment, the collaborative wake-up phase offset is sent to the boundary node on the side with lower data transmission priority. The boundary node on the side with lower data transmission priority adjusts the wake-up phase parameters for the next cycle based on the collaborative wake-up phase offset. If the adjustment target is the boundary node of the same area, the cross-area collaborative unit directly sends the parameter adjustment command; if the adjustment target is the boundary node of an adjacent area, the cross-area collaborative unit sends it to the edge computing node of the adjacent area through the inter-area communication link, and the node then sends the parameter adjustment command.

[0131] 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 this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A power line carrier low-power communication system based on an edge computing module, characterized in that, include: Edge computing nodes are deployed on the side of the distribution network branch box or meter box and are electrically connected to the carrier transceiver module. The relay control module is electrically connected to the edge computing node and the carrier transceiver module, and the relay control module is equipped with a relay time slot arbitration unit; Edge computing nodes are equipped with topology migration identification units and cross-station collaboration units; Edge computing nodes are used for: The wake-up phase parameter, wake-up duty cycle parameter and corresponding payload transmission success rate parameter of the carrier transceiver module in multiple power frequency cycles within the historical period are obtained. The wake-up phase parameter represents the phase offset value of the wake-up window of the carrier transceiver module relative to the zero crossing time of the power frequency voltage of the phase line. A three-dimensional relational model is constructed based on the wake-up phase parameter, wake-up duty cycle parameter and payload transmission success rate parameter. In the three-dimensional relational model, a three-dimensional optimized boundary surface composed of non-dominated state points is determined. Obtain the current wake-up phase parameters and current wake-up duty cycle parameters of the carrier transceiver module, determine the current working state point in the three-dimensional relational model, and calculate the optimal approximation direction vector from the current working state point to the three-dimensional optimized boundary surface; Based on the components of the optimal approximation direction vector in the wake-up phase dimension and wake-up duty cycle dimension, wake-up phase adjustment amount and wake-up duty cycle adjustment amount are generated respectively. The wake-up duty cycle adjustment amount is used to adjust the wake-up window duration of the carrier transceiver module in the next power frequency cycle. The relay control module is used for: Obtain the phase line identifier information of the phase line where the target relay node is located; Based on the first zero-crossing moment of the phase line where this node is located, the second zero-crossing moment of the phase line where the target relay node is located, and the wake-up phase adjustment amount, calculate the matching wake-up phase value of the target relay node; A relay wake-up command is generated based on the matching wake-up phase value. The relay wake-up command is used to control the target relay node to enter the receiving state at the time indicated by the matching wake-up phase value.

2. The power line carrier low-power communication system based on an edge computing module according to claim 1, characterized in that, Edge computing nodes are also used to construct the confidence interval boundaries of the 3D optimized boundary surface, and to correct the wake-up phase adjustment and wake-up duty cycle adjustment based on the confidence interval boundaries: In the three-dimensional relational model, the density of three-dimensional sample points in the preset spatial neighborhood around each non-dominated state point is calculated. Non-dominated state points with a 3D sample point density not lower than a preset density threshold are marked as high-confidence boundary points; A confidence-optimized boundary surface is generated by refitting based on high-confidence boundary points, and the average offset vector between the confidence-optimized boundary surface and the original 3D optimized boundary surface is calculated. The average offset vector is used as the boundary confidence correction vector. Before generating the wake-up phase adjustment and wake-up duty cycle adjustment, the boundary confidence correction vector is superimposed on the optimal approximation direction vector.

3. The power line carrier low-power communication system based on an edge computing module according to claim 1, characterized in that, Edge computing nodes are also used to construct a cross-phase noise period distribution map and correct the matched wake-up phase value of the target relay node based on the cross-phase noise period distribution map: Obtain the noise energy sampling sequences of phases A, B, and C within this substation area in multiple power frequency cycles. The noise energy sampling sequence represents the noise energy value collected at a preset sampling interval within one power frequency cycle. The noise energy sampling sequences of phases A, B, and C are aligned according to their respective zero-crossing times at the power frequency and then superimposed to generate a three-phase composite noise envelope curve. Identify low-noise regions where the noise energy is below a preset noise threshold in the three-phase synthesized noise envelope curve; When the target relay node and this node are located on different phase lines, determine whether the matched wake-up phase value of the target relay node falls into the low noise range; If the matched wake-up phase value of the target relay node does not fall into the low noise range, the phase offset compensation between the matched wake-up phase value and the nearest low noise range boundary is calculated, and the matched wake-up phase value is corrected based on the phase offset compensation.

4. The power line carrier low-power communication system based on an edge computing module according to claim 1, characterized in that, The relay control module is also used to adaptively adjust the preamble length of the relay wake-up command based on the cross-phase noise difference: Obtain the first noise energy estimate of the phase line where this node is located at the matched wake-up phase value; Obtain the second noise energy estimate at the matched wake-up phase value of the phase line where the target relay node is located; The ratio of the first noise energy estimate to the second noise energy estimate is calculated and used as the cross-phase noise difference coefficient. The target preamble length is obtained by querying the preset preamble length mapping table based on the cross-phase noise difference coefficient. The target preamble length is encapsulated in the relay wake-up command. After receiving the relay wake-up command, the target relay node sends the wake-up preamble sequence with the target preamble length.

5. The power line carrier low-power communication system based on an edge computing module according to claim 1, characterized in that, The relay slot arbitration unit is used to allocate time slots when multiple source nodes simultaneously request the same target relay node. When it is detected that both the first source node and the second source node send a relay wake-up request to the same target relay node within a preset time window, the first data transmission urgency score of the first source node and the second data transmission urgency score of the second source node are obtained. Obtain the first wake-up duration requested by the first source node and the second wake-up duration requested by the second source node; A first competition weight vector is constructed based on a first data transmission urgency score and a first wake-up duration, and a second competition weight vector is constructed based on a second data transmission urgency score and a second wake-up duration. Calculate the weighted magnitude of the first and second competition weight vectors in the two-dimensional competition space, and determine the source node with the larger weighted magnitude as the arbitration winner node; Assign the current wake-up window of the target relay node to the winning node in the arbitration, and send a slot delay indication to the unwinning node in the arbitration. The slot delay indication contains a delay phase offset value that suggests re-requesting.

6. The power line carrier low-power communication system based on an edge computing module according to claim 5, characterized in that, The relay slot arbitration unit is also used to perform preemptive slot allocation when the data transmission urgency score of the node that did not win the arbitration exceeds a preset urgency threshold. The current wake-up window of the target relay node is divided into a basic time slot segment and an extended time slot segment, with the basic time slot segment allocated to the arbitration winner node. Calculate the minimum required wake-up time for the node that did not win the arbitration. If the length of the extended time slot is not less than the minimum required wake-up time, then allocate the extended time slot to the node that did not win the arbitration. Send a wake-up window extension command to the target relay node. The wake-up window extension command is used to control the target relay node to continue receiving after the basic time slot ends until the extended time slot ends.

7. The power line carrier low-power communication system based on an edge computing module according to claim 1, characterized in that, The topology migration recognition unit is used to detect topological changes in transformer areas and migrate 3D optimized boundary surfaces. Periodically acquire neighbor node beacon reception records of all carrier transceiver modules within this station area. The neighbor node beacon reception records include the neighbor node identifier of the received beacon and the corresponding received signal strength indication value. Construct a sequence of adjacency matrices for the transformer area based on the beacon reception records of neighboring nodes within multiple cycles; Calculate the Frobenius norm difference of the adjacency matrix of adjacent transformer areas between adjacent cycles. When the Frobenius norm difference exceeds the preset topology change threshold, it is determined that a transformer area topology change has occurred. After determining that a topological change has occurred in the transformer area, the similarity deviation between the first three-dimensional optimized boundary surface before the topological change and the second three-dimensional optimized boundary surface reconstructed after the topological change is calculated. If the similarity deviation exceeds the preset surface similarity threshold, a weighted fusion is performed based on the first 3D optimized boundary surface and the second 3D optimized boundary surface to generate a migrated 3D optimized boundary surface. The migrated 3D optimized boundary surface replaces the first 3D optimized boundary surface, and the replaced 3D optimized boundary surface is used to calculate the optimal approximation direction vector of the current working state point.

8. The power line carrier low-power communication system based on an edge computing module according to claim 7, characterized in that, The topology migration identification unit is also used to perform incremental optimization surface expansion when it is determined that a topology change has occurred in the transformer area and the type of topology change is the addition of a new node: Identify the access location of the new node in the transformer area topology and its adjacent existing nodes; Obtain the 3D optimized boundary surface of adjacent existing nodes as the reference boundary surface; Acquire initial sample points of wake-up phase parameters, wake-up duty cycle parameters, and payload transmission success rate parameters collected during the initialization phase of the newly added node; Based on the local curvature characteristics of the reference boundary surface, the initial sample points are projected onto the reference boundary surface along the normal vector direction to obtain the projection points; The projected points are weighted and averaged with the initial sample points to generate the initial 3D optimized boundary surface of the new node. The initial 3D optimized boundary surface of the new node is used to calculate the optimal approximation direction vector in subsequent runs.

9. The power line carrier low-power communication system based on an edge computing module according to claim 1, characterized in that, Edge computing nodes are also used to perform area-level time slot collision detection and avoidance: Collect the wake-up phase parameters and wake-up duty cycle parameters of multiple carrier transceiver modules within this station area; A distribution map of wake-up time slots for a transformer area is constructed based on the wake-up phase parameters and wake-up duty cycle parameters of multiple carrier transceiver modules. Identify conflicting time slot regions in the wake-up time slot distribution map of the transformer area where the overlap of wake-up windows exceeds a preset overlap threshold; For multiple carrier transceiver modules whose wake-up windows overlap within the conflict time slot region, a wake-up phase fine-tuning amount is generated based on the data transmission priority of each carrier transceiver module, and the wake-up phase fine-tuning amount is sent to the corresponding carrier transceiver module.

10. The power line carrier low-power communication system based on an edge computing module according to claim 1, characterized in that, The cross-regional coordination unit is used to coordinate the wake-up timing of boundary nodes in adjacent regions: Obtain the wake-up phase parameters and wake-up duty cycle parameters of the boundary nodes of this station area. The boundary nodes are nodes that have overlapping carrier signal coverage areas with the boundary nodes of adjacent stations. The adjacent wake-up phase parameters and adjacent wake-up duty cycle parameters of the boundary nodes of adjacent substations are obtained through the inter-substation communication link. Calculate the temporal overlap rate between the wake-up window of the boundary node of this transformer area and the wake-up window of the boundary node of the adjacent transformer area; When the time domain overlap rate exceeds the preset overlap tolerance threshold, a collaborative wake-up phase offset is generated based on the data transmission priority of the boundary node of this station area and the data transmission priority of the boundary node of the adjacent station area. The collaborative wake-up phase offset is sent to the boundary node on the side with lower data transmission priority. The boundary node on the side with lower data transmission priority adjusts the wake-up phase parameters for the next cycle based on the collaborative wake-up phase offset.

Images