Wireless network relay node deployment system and method based on directional antenna

Abstract

The invention discloses a wireless network relay node deployment system based on a directional antenna, which relates to the technical field of wireless communication, and at least comprises: an interference modeling module determines the service density of a link corresponding to each edge in a tree network topological graph of current iteration; the link quality evaluation module determines the outage probability of the link and determines the edge weight of the edge corresponding to the link so as to generate a tree-shaped weighted network topological graph of the current iteration; a deployment iteration optimization module reselects a candidate relay node with the lowest edge weight as a new father node for each service node according to the tree-shaped weighted network topological graph of the current iteration, and connects each new father node; and the output module outputs the deployment position of the relay node according to the tree-shaped weighted network topological graph of the last iteration. Through the scheme provided by the invention, a global optimal deployment scheme which is remarkably superior in the aspects of communication reliability, resource utilization efficiency and actual scene adaptability can be obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication, and in particular to a directional antenna-based wireless network relay node deployment system and method. BACKGROUND

[0002] In the field of wireless communication, relay node deployment can improve network coverage and connectivity. Related technologies usually abstract the deployment problem as a Steiner tree problem in graph theory, and minimize the number of relay nodes or path overhead under the premise of ensuring the connectivity of service nodes through approximation algorithms.

[0003] However, the traditional channel model used by related technologies cannot accurately depict the existence of direct diameters in directional antenna scenarios, resulting in a large deviation between the estimated outage and the actual outage, and the link layer signal quality evaluation is also not accurate. Therefore, it affects the reliability of the deployment decision while it is difficult to significantly improve the network capacity and throughput performance. In addition, related technologies are based on the deterministic assumption of environmental information, ignoring the dynamic changes of deployment environment and the uncertainty factors, resulting in insufficient robustness of the deployment scheme in the real environment, which cannot adapt to diversified scene requirements and cannot guarantee the quality of service. SUMMARY

[0004] Embodiments of the present application provide a directional antenna-based wireless network relay node deployment system and method, which aims to solve the problems in the above background technology.

[0005] To solve the above technical problems, the present application is implemented as follows: In a first aspect, the embodiments of the present application provide a directional antenna-based wireless network relay node deployment system, at least comprising: An interference modeling module is configured to determine the traffic density of each link corresponding to each edge in the tree network topology graph of the current iteration, and the traffic density of a link is the amount of traffic data transmitted between two nodes connected by the link in unit time. A link quality evaluation module is configured to determine the outage probability of each link according to the traffic density of each link in the tree network topology graph of the current iteration, and to combine the outage probability of each link with the node deployment cost of each node connected by the link as the edge weight of the edge corresponding to the link to generate a tree weighted network topology graph of the current iteration. A deployment iteration optimization module is configured to reselect a candidate relay node with the lowest edge weight for each service node as a new parent node according to the tree weighted network topology graph of the current iteration, and connect each new parent node to generate a tree network topology graph of the next iteration and return it to the interference modeling module for the next iteration using the tree network topology graph of the next iteration. an output module configured to output the deployment positions of the relay nodes according to the tree-shaped weighted network topology graph of the last iteration.

[0006] Optionally, the system further comprises: an environment perception module configured to determine a probabilistic evaluation of a channel state between each two nodes according to noise signals in the deployment environment, to determine whether a link between each two nodes is a line-of-sight path or a non-line-of-sight path; a link quality evaluation module configured to perform the following steps for each link in the tree-shaped network topology graph of the current iteration: determine a channel gain of the link by using a statistical channel model based on a Rayleigh fading channel if the link is a non-line-of-sight path, and determine a channel gain of the link by using a statistical channel model based on a Rician fading channel if the link is a line-of-sight path; calculate a signal strength of the link according to the channel gain of the link; determine an interference signal strength of the link according to a traffic density and a corresponding channel gain and a directional antenna gain of other interfering links except the link; determine an outage probability of the link according to the signal strength and the interference signal strength of the link.

[0007] Optionally, the system further comprises a network topology initialization module; the environment perception module is further configured to provide the network topology initialization module with node position information and communication range information, the node position information including deployment positions of each traffic node in the deployment environment obtained by the environment perception module, and positions of each candidate relay node determined by the environment perception module according to obstacle information of the deployment environment, the deployment positions of the each traffic node, and physical restriction conditions for deployment of the relay nodes, and the communication range constraint information including communication range constraints of the each traffic node and communication range constraints of the each candidate relay node; the network topology initialization module is configured to construct an initial tree-shaped network topology graph according to the node position information and the communication range information, select a nearest candidate relay node as a parent node for each traffic node according to the initial tree-shaped network topology graph, connect the parent nodes to generate the initial tree-shaped network topology graph, and provide the initial tree-shaped network topology graph to the interference modeling module for a first iteration.

[0008] Optionally, the link quality evaluation module is specifically configured to perform the following steps for an nth link in the tree-shaped network topology graph of the current iteration, by taking each link in the tree-shaped network topology graph of the current iteration except the nth link as an interfering link of the nth link: determine a service signal strength of the nth link according to a channel gain of the nth link and antenna gains of directional antennas of two nodes connected by the nth link; determine an interference signal strength of each interference link of the nth link according to a channel gain of the interference link and antenna gains of directional antennas of two nodes connected by the interference link; weight the interference signal strengths of the interference links of the nth link by service densities of the interference links of the nth link to obtain a total interference signal strength of the nth link; determine a signal-to-interference-and-noise ratio of the nth link according to the service signal strength, the noise signal strength and the total interference signal strength of the nth link; determine an outage probability of the nth link according to the signal-to-interference-and-noise ratio of the nth link.

[0009] Optionally, the link quality evaluation module is specifically configured to determine the deployment cost of the service node as zero. In a case where the deployment requirement scenario is a network performance maximization scenario, determine a node deployment cost of the candidate relay node as a first value. In a case where the deployment requirement scenario is a relay node quantity minimization scenario, determine a node deployment cost of the candidate relay node as a second value, the second value being greater than the first value. For the nth link in the tree network topology graph of the current iteration, perform the following steps: In a case where the deployment requirement scenario is a network performance maximization scenario and at least one of the two nodes connected by the nth link is a candidate relay node, determine a total node deployment cost of the nth link based on the first value. In a case where the deployment requirement scenario is a relay node quantity minimization scenario and at least one of the two nodes connected by the nth link is a candidate relay node, determine a total node deployment cost of the nth link based on the second value. fuse the outage probability and the total node deployment cost of the nth link as an edge weight of an edge corresponding to the nth link.

[0010] Optionally, the deployment iteration optimization module is specifically configured to: fuse the tree network topology graph of the next iteration generated and the tree network topology graphs of previous iterations based on an infinite impulse response filter mechanism to obtain an updated tree network topology graph of the next iteration, reduce the annealing temperature, and return the updated tree network topology graph of the next iteration to the interference modeling module to perform the next iteration based on the reduced annealing temperature and the updated tree network topology graph of the next iteration.

[0011] In a second aspect, an embodiment of the present application provides a directional antenna-based wireless network relay node deployment method applied to the system of the second aspect, comprising: determining a traffic density of a link corresponding to each edge in the tree network topology graph of the current iteration, the traffic density of a link being a traffic data volume transmitted between two nodes connected by the link in a unit of time; determining an outage probability of each link in the tree network topology graph of the current iteration according to the traffic density of the link, and fusing the outage probability of each link and a node deployment cost of each of the two nodes connected by the link to serve as an edge weight of an edge corresponding to the link to generate a tree weighted network topology graph of the current iteration; reselecting a candidate relay node with the lowest edge weight as a new parent node for each traffic node according to the tree weighted network topology graph of the current iteration, and connecting the new parent nodes to generate a tree network topology graph of the next iteration and return to the interference modeling module to perform the next iteration by using the tree network topology graph of the next iteration; outputting the deployment positions of the relay nodes according to the tree weighted network topology graph of the last iteration.

[0012] Optionally, the method further comprises: determining a channel state between each two nodes for probabilistic evaluation according to a noise signal in the deployment environment to determine whether a link between each two nodes is a line-of-sight path or a non-line-of-sight path; for each link in the tree network topology graph of the current iteration, performing the following steps: determining a channel gain of the link by using a statistical channel model based on a Rayleigh fading channel in a case where the link is a non-line-of-sight path, and determining the channel gain of the link by using a statistical channel model based on a Rician fading channel in a case where the link is a line-of-sight path; calculating a signal strength of the link according to the channel gain of the link; determining an interference signal strength of the link according to traffic densities and corresponding channel gains and directional antenna gains of other interference links other than the link; determining an outage probability of the link according to the signal strength and the interference signal strength of the link.

[0013] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, and a computer program stored on the memory and capable of running on the processor, wherein the computer program is executed by the processor to implement steps of a directional antenna-based wireless network relay node deployment method.

[0014] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement steps of a directional antenna-based wireless network relay node deployment method.

[0015] The technical solutions provided by the embodiments of the present application bring at least the following beneficial effects: The present application can make decisions based on real and complex network environment information by probabilistic evaluation of channel state by the environment perception module and calculation of traffic density according to dynamic routing and traffic patterns by the interference modeling module. The limitations of traditional methods relying on idealized distance thresholds or static models are overcome, and the generated deployment scheme is more adaptable to the dynamic changes and uncertainties of actual application scenarios. Further, the link quality evaluation module integrates the link outage probability and the adjustable node deployment cost to form the edge weight. The system can flexibly adjust the optimization objective according to specific application requirements, significantly enhancing the practicality and flexibility of the deployment scheme in different scenarios. It can be seen that the present application, through the deep cooperation of the interference modeling module and the link quality evaluation module and the iterative optimization of the deployment iterative optimization module, finally outputs a globally optimal deployment scheme that is significantly superior in communication reliability, resource utilization efficiency and actual scenario adaptability. BRIEF DESCRIPTION OF DRAWINGS

[0016] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not intended to limit the scope of the present application. Moreover, the same reference numerals are used throughout the same figures. In the drawings: Figure 1 is an architecture diagram of a directional antenna-based wireless network relay node deployment system provided by an embodiment of the present application; Figure 2 is a system application flow diagram based on a tree network topology provided by an embodiment of the present application; Figure 3 is an overall application flow diagram of a directional antenna-based wireless network relay node deployment system provided by an embodiment of the present application; Figure 4 is a step diagram of a directional antenna-based wireless network relay node deployment method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0017] With reference to the drawings and embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative work under the premise that the embodiments are within the scope of protection of the present application.

[0018] The terms "first", "second", and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be exchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / ", generally indicates that the front and rear associated objects are in a "or" relationship.

[0019] The system and method for deploying a relay node in a wireless network based on a directional antenna according to the embodiments of the present application will be described in detail below with reference to the drawings, specific embodiments and application scenarios.

[0020] Figure 1 FIG. 1 is a schematic diagram of an architecture of a system for deploying a relay node in a wireless network based on a directional antenna according to an embodiment of the present application, which at least includes: The interference modeling module is configured to determine the traffic density of each link corresponding to each edge in the tree network topology graph in the current iteration, and the traffic density of a link is the amount of traffic data transmitted between two nodes connected by the link in a unit of time.

[0021] In the embodiments of the present application, since the present application adopts iterative optimization, starting from the initial tree network topology graph, the connection relationship of the relay node is adjusted constantly, and the optimal deployment scheme is gradually approached. The current iteration refers to the computing cycle being performed in the system during the optimization deployment process.

[0022] The interference modeling module analyzes the traffic load of the tree network topology generated in the current iteration. The tree network topology refers to a loop-free network connection graph organized in a tree structure, representing the connection relationship between traffic nodes and relay nodes. The nodes of the tree network topology include traffic nodes and relay nodes. Traffic nodes are inherent terminal nodes in the network (such as sensors and user equipment), and relay nodes are candidate deployment nodes for forwarding data and enhancing connectivity. The edges represent the communication links between the nodes. The tree network topology is updated in each iteration to reflect the latest relay node connection selection. Specifically, the interference modeling module calculates the traffic density of each link corresponding to an edge in the tree network topology of the current iteration based on the preset traffic pattern and routing scheme. Traffic density is defined as the amount of traffic data transmitted between two nodes connected by a link in a unit of time, reflecting the intensity of data traffic on the link. In a real network, the traffic pattern can include traffic nodes generating data in an equal probability manner and randomly selecting other traffic nodes as data transmission targets. The routing scheme can generate a routing table based on the shortest path principle or other optimization strategies. The interference modeling module analyzes these routing information and calculates the traffic volume carried on each link to assign a traffic density value to each link. This process not only considers the static traffic distribution but also adapts to the dynamic changes in network traffic, ensuring the accuracy of the interference evaluation.

[0023] The link quality evaluation module determines the outage probability of each link based on the traffic density of each link in the tree network topology of the current iteration, and combines the outage probability of each link with the node deployment cost of the two nodes connected by the link as the edge weight of the edge corresponding to the link to generate the tree weighted network topology of the current iteration.

[0024] The link quality evaluation module quantifies the reliability of each link in the tree network topology of the current iteration. The link quality evaluation module first calculates the outage probability of each link based on the traffic density data provided by the interference modeling module. The outage probability is an indicator of the communication reliability of the link under random channel fluctuations and interference environment.

[0025] Figure 2 is the system application flowchart based on the tree network topology in an embodiment of the application, please see Figure 2 After obtaining the outage probability, the link quality evaluation module further combines it with the node deployment cost to generate the edge weight corresponding to each edge. The calculation of the edge weight integrates the node deployment cost of both ends of the link and the bidirectional outage probability of the link, forming a comprehensive indicator reflecting the balance between deployment economy and communication reliability. Finally, the link quality evaluation module outputs the tree weighted network topology of the current iteration, providing a quantitative basis for deployment optimization.

[0026] Specifically, the calculation formula of the edge weight is:

[0027] wherein, is the weight of the edge ; and are the deployment costs of the nodes and , respectively (the business node cost is , and the relay node cost is an adjustable parameter ); and are the interruption probabilities of the links from to and from to , respectively.

[0028] In an optional implementation, the link quality evaluation module is specifically configured to determine the deployment cost of the business node as zero.

[0029] The link quality evaluation module first assigns different deployment cost attributes to different types of nodes in the network. Specifically, the deployment cost of the business node (i.e., the inherent sensor or user node in the network) is set to zero. Based on the following considerations: the business node is usually regarded as an infrastructure component of the network, and its location is often determined before deployment, without additional site selection or installation costs. Therefore, its cost factor is not considered in the optimization process, and the optimization focus is concentrated on the deployment decision of the relay node.

[0030] In the case of a network performance maximization scenario, the node deployment cost of the candidate relay node is determined as a first value.

[0031] For the candidate relay node, its deployment cost is not fixed. In order to enable the system described in the present application to accurately adapt to different deployment requirement scenarios, the link quality evaluation module presets at least two typical deployment requirement scenarios and configures corresponding cost strategies for them: In the network performance maximization scenario, the priority goal is to improve the overall communication quality of the network, such as maximizing throughput and minimizing interruption probability. In this scenario, the link quality evaluation module sets the node deployment cost of the candidate relay node to a relatively low first value. The first value means that in the optimization algorithm, the "cost penalty" brought by enabling the relay node is smaller, thereby encouraging the algorithm to more actively deploy relay nodes in order to improve link reliability, even if this means an increase in the number of relay nodes.

[0032] In the case that the deployment requirement scenario is the scenario of minimizing the number of relay nodes, the node deployment cost of the candidate relay node is determined as a second value, which is greater than the first value.

[0033] In the scenario of minimizing the number of relay nodes, the priority is to reduce the number of relay nodes as much as possible under the premise of ensuring basic connectivity, so as to reduce the hardware cost, energy consumption and maintenance complexity. In this scenario, the link quality evaluation module sets the node deployment cost of the candidate relay node as a second value which is relatively higher than the first value. The second value means that the "cost" of enabling the relay node is high, which will guide the optimization algorithm to tend to deploy the relay node only in an absolutely necessary case, thereby effectively controlling the total number of relay nodes.

[0034] For the nth link in the tree network topology graph of the current iteration, the following steps are performed: In the case that the deployment requirement scenario is the scenario of maximizing network performance, and at least one of the two nodes connected by the nth link is a candidate relay node, the total node deployment cost of the nth link is determined based on the first value.

[0035] First, the link quality evaluation module identifies the types of the two nodes connected by the nth link. The two ends of the link can be connected to two service nodes, two candidate relay nodes, or one service node and one candidate relay node.

[0036] Next, the link quality evaluation module calculates the total node deployment cost of the nth link according to the currently activated deployment requirement scenario.

[0037] If the current scenario is the scenario of maximizing network performance, and at least one of the two nodes connected by the nth link is a candidate relay node, the link quality evaluation module will calculate the total node deployment cost based on the first value set for this scenario. Specifically, the total cost is the sum of the deployment costs of the nodes at both ends of the link. Since the service node cost is zero, if the link connects one service node and one relay node, the total cost is the first value; if it connects two relay nodes, the total cost is twice the first value.

[0038] In the case that the deployment requirement scenario is the scenario of minimizing the number of relay nodes, and at least one of the two nodes connected by the nth link is a candidate relay node, the total node deployment cost of the nth link is determined based on the second value.

[0039] If the current scenario is the scenario of minimizing the number of relay nodes, and at least one of the two nodes connected by the nth link is a candidate relay node, the link quality evaluation module determines the total node deployment cost of the link based on the higher second value set for this scenario using the same summation logic.

[0040] It should be noted that if both nodes connected by the nth link are business nodes, then the total node deployment cost is zero regardless of the deployment scenario, because the cost of business nodes is constant at 0.

[0041] The outage probability of the nth link is integrated with the total node deployment cost as the edge weight of the edge corresponding to the nth link.

[0042] The link quality assessment module obtains the outage probability of the nth link. The outage probability is calculated by comprehensively considering various factors such as the channel state information (such as the line-of-sight probability) provided by the environment perception module, the gain pattern of the directional antenna, the interference level weighted by the traffic density provided by the interference modeling module, and applying advanced probability and statistics methods (such as interference approximation based on central limit theorem), which reflects the communication reliability of the link in the real wireless environment. The calculation of the outage probability is described in detail below.

[0043] In an optional embodiment, the link quality assessment module is specifically configured to, for the nth link in the tree network topology graph of the current iteration, take each link in the tree network topology graph of the current iteration except the nth link as an interference link of the nth link, and perform the following steps: According to the channel gain of the nth link and the antenna gain of the directional antenna of each of the two nodes connected by the nth link, the traffic signal strength of the nth link is determined.

[0044] In each iteration process, the link quality assessment module performs a set of calculation procedures that comprehensively consider the co-channel interference in the network for each link (referred to as the nth link in the embodiment of the application) in the tree network topology graph of the current iteration to determine the outage probability. This embodies the deep understanding and accurate modeling capability of the present application for the cumulative effect of interference in real wireless communication environment.

[0045] The link quality assessment module first calculates the strength of the expected signal carried by the nth link, i.e. the traffic signal strength. The calculation is based on two key parameters: The first is the channel gain of the nth link, which is determined by the line-of-sight probability provided by the environment perception module. If the link is determined to be a line-of-sight path, the channel gain is calculated using the Rician fading channel model; if it is determined to be a non-line-of-sight path, the channel gain is calculated using the Rayleigh fading channel model. The channel gain comprehensively reflects the large-scale path loss, shadow fading and small-scale fading effects experienced by the signal during transmission.

[0046] Secondly, the antenna gain of the directional antennas of the two nodes connected by the nth link. Since the directivity of directional antennas is fully considered in the present application, the link quality assessment module calculates the gain of the signal in the main lobe direction of the transmitting node and the gain of the signal in the main lobe direction of the receiving node according to the relative orientation of the two communication nodes. The high gain of the main lobe of the directional antenna directly acts on the useful signal here, significantly improving the reception strength of the service signal.

[0047] By integrating the above-mentioned channel gain and antenna gain, the link quality assessment module calculates the effective service signal strength observed at the receiving end of the nth link from the transmitting end thereof.

[0048] According to the channel gain of each interference link of the nth link and the antenna gain of the directional antennas of the two nodes connected by the interference link, the interference signal strength of the interference link is determined.

[0049] In the wireless network, other links in the same frequency band will cause co-frequency interference to the nth link. The link quality assessment module regards each other link in the tree network topology diagram of the current iteration except the nth link as a potential interference link of the nth link.

[0050] For each such interference link, the link quality assessment module calculates the interference signal strength thereof reaching the receiving end of the nth link. The calculation is also based on two factors: Firstly, the channel gain of the interference link, that is, the channel gain between the transmitting node of the interference link and the receiving node of the nth link is calculated. This process also adaptively selects the Rician or Rayleigh fading model according to the line-of-sight probability evaluation of the environment perception module for the interference path.

[0051] Secondly, the antenna gain of the directional antennas of the two nodes connected by the interference link. Here, the directional antennas have a dual role. On the one hand, it is considered whether the main lobe direction of the antenna of the transmitting node of the interference link is aligned with the receiving node of the nth link, which determines the transmission strength of the interference. On the other hand, it is considered the gain of the antenna pattern of the receiving node of the nth link in the direction of the interference source, which determines the sensitivity of the receiving node to the interference from the direction. The sidelobe suppression characteristic of the directional antenna plays a key role in this link, which can effectively attenuate the interference from the undesired direction.

[0052] By performing the above calculation for each interference link, the link quality assessment module obtains independent interference signal strength values, each value representing the potential interference level of a specific interference source to the nth link.

[0053] The interference signal strengths of the interference links of the nth link are weighted with the service density of each interference link of the nth link as the weight, to obtain the total interference signal strength of the nth link.

[0054] Different interference links carry different traffic flows in the network, and their "activity" or "duty cycle" of interference also varies. The link quality assessment module introduces traffic density as a weight to reflect this difference. The link quality assessment module performs the following operations: The traffic density of each interference link of the nth link is weighted. The traffic density is provided by the interference modeling module and represents the average data flow of the interference link per unit time.

[0055] The interference signal strength of each interference link of the nth link is weighted and averaged or summed.

[0056] Through this weighting process, those interference links with heavy traffic and large data flow are given higher weights, and their contribution to the total interference is amplified; while those interference links with idle traffic, their contribution is correspondingly reduced. Finally, the module calculates the total interference signal strength of the nth link, which is a time-averaged interference power value that is closer to the actual running state of the network.

[0057] According to the traffic signal strength and the noise signal strength of the nth link and the total interference signal strength of the nth link, the signal-to-interference-and-noise ratio of the nth link is determined.

[0058] After obtaining all the above parameters, the link quality assessment module calculates the signal-to-interference-and-noise ratio of the nth link. The signal-to-interference-and-noise ratio is used to measure the received signal quality, and the signal-to-interference-and-noise ratio is equal to the traffic signal strength of the nth link divided by the sum of the total interference signal strength of the nth link and the inherent noise signal strength in the environment. The noise signal strength is the power of the background noise such as thermal noise inherent to the communication receiver. The higher the signal-to-interference-and-noise ratio value, the better the signal quality and the more reliable the communication.

[0059] According to the signal-to-interference-and-noise ratio of the nth link, the outage probability of the nth link is determined.

[0060] Finally, the link quality assessment module determines the outage probability of the nth link according to the calculated signal-to-interference-and-noise ratio. The outage probability is defined as: the probability that the instantaneous signal-to-interference-and-noise ratio of the link is lower than the minimum threshold value necessary to maintain reliable communication under random channel fading and interference fluctuations.

[0061] The link quality assessment module establishes a mapping relationship from the signal-to-interference-and-noise ratio to the outage probability based on the statistical properties of the specific fading channel model (Ricean or Rayleigh) used. For the Ricean channel, the relationship is determined by the Ricean factor and the signal-to-interference-and-noise ratio threshold; for the Rayleigh channel, it is represented as a negative exponential relationship of the signal-to-interference-and-noise ratio threshold. Through this mapping, the link quality assessment module finally outputs a quantitative outage probability value, reflecting the risk of communication failure of the nth link under the current network topology, traffic load and channel conditions.

[0062] deploying an iteration optimization module to reselect a candidate relay node with the lowest edge weight as a new parent node for each service node according to the tree-weighted network topology graph of the current iteration Figure 2 and connecting the new parent nodes to generate a tree network topology graph of the next iteration and return to the interference modeling module to use the tree network topology graph of the next iteration for the next iteration.

[0063] Figure 3 is a schematic diagram of the overall application process of a directional antenna-based wireless network relay node deployment system according to an embodiment of the present application, as shown in Figure 3 The deployment iteration optimization module uses an iterative mechanism to continuously improve the deployment scheme of the relay node. Based on the tree-weighted network topology graph of the current iteration, the deployment iteration optimization module first reselects a candidate relay node with the lowest edge weight as a new parent node for each service node. This selection process aims to minimize the comprehensive cost of the local link (including the deployment cost and the outage probability), thereby improving the overall network performance. Subsequently, the deployment iteration optimization module connects these new parent nodes to construct a new tree network topology graph (the new tree network topology graph is shown in Figure 2 ), ensuring that all service nodes are connected through relay nodes to achieve full network connectivity.

[0064] The newly generated tree network topology graph is fed back to the interference modeling module to recalculate the traffic density and interference model of the link, thereby starting the next iteration. This closed-loop optimization mechanism enables the system to dynamically adapt to the interference distribution adjustment caused by the network topology change.

[0065] In an optional embodiment, the deployment iteration optimization module is specifically used for: based on the infinite impulse response filter mechanism, fusing the generated tree network topology graph of the next iteration with the tree network topology graphs of the previous iterations to obtain an updated tree network topology graph of the next iteration, reducing the annealing temperature, and returning the updated tree network topology graph of the next iteration to the interference modeling module to use the updated tree network topology graph of the next iteration for the next iteration based on the reduced annealing temperature.

[0066] In the fusion process, the deployment iteration optimization module uses the infinite impulse response filter mechanism to update the tree-weighted network topology graph, and the formula is as follows:

[0067] wherein, is the updated tree-weighted network topology graph of the n th iteration; ​is the tree-weighted network topology graph corresponding to the current record optimal solution; is the tree-weighted network topology graph of the first iteration; is the tree-weighted network topology graph of the n-th iteration; is the adjustable fusion parameter, the value range is .

[0068] It should be noted that in the deployment of the iterative optimization module, when the tree-weighted network topology graph is updated using the infinite impulse response filter mechanism, the tree-weighted network topology graph corresponding to the current record optimal solution is dynamically evaluated and recorded according to the tree network topology graph of each iteration. In one embodiment, the tree-weighted network topology graphs of the first iteration to the n-th iteration are compared, and the tree-weighted network topology graph with the best performance (such as the smallest total edge weight) in the iterations is taken as the tree-weighted network topology graph corresponding to the current record optimal solution.

[0069] After completing an iteration and generating the tree network topology graph of the next iteration, the deployment iterative optimization module starts the information fusion and update process. Specifically, the deployment iterative optimization module processes the generated tree network topology graph of the next iteration using the infinite impulse response filter mechanism. The essence of the infinite impulse response filter mechanism is to use feedback to fuse the tree network topology graph of the next iteration with the tree-weighted network topology graph corresponding to the current record optimal solution . It should be noted that it is not simply replacing the old scheme with the new scheme, but fusing the tree network topology graph of the next iteration with the tree-weighted network topology graph corresponding to the current record optimal solution The synthesis is performed with certain weights, so that an updated tree network topology of the next iteration is obtained, which carries the optimized historical experience. The fusion process plays a smoothing role, effectively filtering out short-term fluctuations or poor temporary solutions caused by random disturbances in the optimization process, making the optimization path more stable, and helping to suppress the algorithm from oscillating around or falling into local optimal solutions, thereby significantly improving the quality of the final solution and the convergence speed of the optimization process. At the same time, the deployment iterative optimization module follows the basic framework of the simulated annealing algorithm and performs annealing temperature reduction. The annealing temperature metaphorically represents the "willingness" or "probability" of accepting a suboptimal solution in the optimization process. In the early stage of optimization, the temperature is set high, and the system has a high probability of accepting a new solution that is slightly worse than the current solution, which is beneficial to the algorithm to jump out of the local optimal trap and explore the entire solution space. As the iteration proceeds, the deployment iterative optimization module gradually reduces the annealing temperature according to the preset cooling strategy (such as equal ratio cooling or linear cooling). The reduction of temperature means that the system's willingness to accept a poor solution gradually weakens, and the optimization process gradually shifts from extensive exploration in the early stage to fine mining in the most promising area, and finally tends to stabilize around the global optimal solution or a high-quality near-optimal solution.

[0070] After the above information fusion and temperature reduction operations are completed, the deployment iterative optimization module returns the updated tree network topology of the next iteration to the interference modeling module. This step constitutes the key feedback link of the optimization closed loop. Based on the new tree network topology that is fused with historical information, more smooth and rational, and the reduced annealing temperature, the next iteration cycle begins. The interference modeling module will recalculate the traffic density and interference distribution in the network based on the new topology; the link quality evaluation module updates the link outage probability and edge weight; then the deployment iterative optimization module again performs parent node selection, tree weighted network topology construction and acceptance criterion judgment based on the new weighted graph and the current (lower) temperature. The criterion judgment is used to determine whether to accept the newly generated tree network topology as the current solution. Simply put, in the simulated annealing algorithm, the acceptance criterion is based on the Metropolis criterion, which allows the simulated annealing algorithm to accept a poor solution with a certain probability to avoid falling into a local optimum. For details, please refer to the relevant technology, which will not be repeated here.

[0071] The output module is configured to output the deployment position of the relay node according to the tree weighted network topology of the last iteration.

[0072] At the end of the iterative optimization process, the output module processes and formats the tree-shaped weighted network topology generated in the last iteration. The output module extracts the specific deployment location information of the relay nodes and converts it into a user-readable coordinate output form. The output results can be directly used in network deployment practice, such as guiding the physical installation location of the relay nodes, to ensure the implementability of the deployment scheme in the actual environment.

[0073] In an alternative embodiment, further comprising: An environment perception module is configured to determine the channel state between each two nodes based on the noise signals in the deployment environment, and to perform a probabilistic evaluation to determine whether the link between each two nodes is a line-of-sight path or a non-line-of-sight path.

[0074] The environment perception module not only processes macro geographic information such as the predetermined positions of the service nodes, the topography, and the distribution of buildings, but also focuses on various noise signals that inevitably exist in the deployment environment. These noises may come from measurement errors, environmental electromagnetic interference, or other unpredictable random factors. The core function of the environment perception module is to perform a probabilistic evaluation of the channel state between any two nodes in the network based on these perception information containing noise.

[0075] Specifically, the environment perception module does not simply give a deterministic, "black or white" channel judgment, but outputs a probability value representing the likelihood of the link being a line-of-sight path. Line-of-sight path means that there is an unobstructed direct propagation path between the transmitting node and the receiving node; otherwise, it is a non-line-of-sight path, and the signal mainly relies on reflection, diffraction, scattering, etc. for propagation. Through probabilistic evaluation, the environment perception module effectively quantifies the channel state uncertainty in the real world due to incomplete information and noise interference, providing more realistic and robust input information for subsequent processing.

[0076] Please refer to Figure 2 As shown in (1), the environment perception module reads and processes the physical information in the deployment scenario, including the predetermined positions of the service nodes (the circular pattern in (1) represents the service nodes), building or terrain obstacle elements, and physical limitations of relay node deployment. Based on this information, the environment perception module generates a set of candidate relay node positions that meet the actual conditions (the square pattern in (1) represents the candidate relay nodes). On this basis, considering that the perception information such as buildings and terrain may be incomplete or noisy, the environment perception module performs a probabilistic evaluation of the channel state between any two nodes and outputs the line-of-sight probability of each link, providing more realistic and robust environmental input for subsequent modules.

[0077] The link quality evaluation module is configured to perform the following steps for each link in the tree-shaped network topology of the current iteration: The link quality assessment module receives the line-of-sight probability assessment result from the environment perception module, and assesses the quality of each link in the tree network topology graph of the current iteration accordingly. The calculation process includes the following steps: Please refer to Figure 2 In step (4), for each link to be assessed, the link quality assessment module first selects the most suitable statistical channel model according to the probabilistic judgment provided by the environment perception module to calculate the channel gain.

[0078] For a link determined to be a non-line-of-sight path, the link quality assessment module uses a statistical channel model based on a Rayleigh fading channel to determine the channel gain of the link. The Rayleigh fading model is particularly suitable for describing the typical multipath fading environment where the signal undergoes a large number of scatterers during transmission, and there is no dominant direct path. Under the Rayleigh fading model, the fluctuation characteristics of the channel gain are described by the Rayleigh distribution, which can accurately reflect the random variation of signal strength under non-line-of-sight transmission conditions.

[0079] For a link determined to be a line-of-sight path, the link quality assessment module uses a statistical channel model based on a Rician fading channel to determine the channel gain of the link. The Rician fading model is suitable for transmission scenarios where there is a stable direct signal component, as well as multiple weak scattered components. The Rician fading model characterizes the relative strength of the direct path component and the scattered path component through the Rician factor, which can more accurately describe the statistical characteristics of the channel gain in the presence of a direct path, making up for the shortcomings of the traditional Rayleigh model in such scenarios.

[0080] That is, the link quality assessment module selects different statistical channel models to calculate the channel gain according to the line-of-sight probability provided by the environment perception module. For a line-of-sight path, the Rician fading channel model is used:

[0081] wherein, is the channel gain between nodes and ; is the large-scale average channel power gain, including path loss and shadowing attenuation; is the small-scale attenuation coefficient, whose square term obeys a non-central chi-square distribution with a Rician factor , a mean of 1, and a degree of freedom of 2, representing the fading characteristics in the presence of a direct path. For a non-line-of-sight path, the Rayleigh fading channel model is used:

[0082] wherein, is the channel gain of the non-line-of-sight path; is the small-scale attenuation coefficient, whose square term Subject to an exponential distribution with mean 1, representing the fading characteristics without line-of-sight.

[0083] According to the channel gain of the link, the signal strength of the link is calculated.

[0084] The link quality assessment module further calculates the signal strength of the link according to the channel gain of the link. The signal strength here specifically refers to the power strength of the desired signal at the receiving end. The calculation process takes into account the transmission power, the channel gain calculated via the corresponding fading channel model (which includes the effects of path loss and fading), and the high gain of the directional antenna in the main lobe direction. Since the nodes of the present invention are equipped with directional antennas, when the transmitting node and the receiving node are within the main lobe coverage of each other's antennas, the main lobe gain of the antenna will significantly enhance the power of the desired signal reaching the receiving end. Therefore, the final calculated signal strength is a quantitative value that reflects the advantages of directional transmission.

[0085] According to the traffic density and corresponding channel gain, directional antenna gain of other interfering links except for the link, the interference signal strength of the link is determined.

[0086] Considering that in the network, when other links are simultaneously transmitting data, they will cause co-frequency interference to the current link. The link quality assessment module first identifies all other interfering links in the network that can potentially interfere with the current link. For each interfering source link, its interference contribution is not equal. The link quality assessment module weights the transmission of other interfering links using the traffic density provided by the interference modeling module. A high traffic density link means that it transmits a large amount of data per unit time, and its average interference power as an interference source is correspondingly stronger. For each path from each other interfering link to the receiving node of the link, the link quality assessment module also calculates its channel gain. The calculation of channel gain also depends on the line-of-sight / non-line-of-sight judgment of the environment perception module for the interference path, and selects the statistical channel model based on Rayleigh fading channel or the statistical channel model based on Rician fading channel.

[0087] It is also considered that the gain of the receiving node's directional antenna of the current link (i.e. the link) is directional. If the interference signal comes from outside the main lobe direction of the receiving antenna, especially from the side lobe or null region, then the antenna gain will be very low, thus significantly suppressing the strength of the interference signal. The link quality assessment module applies the corresponding antenna gain value based on the relative relationship between the direction of arrival of the interference signal and the pointing direction of the receiving antenna.

[0088] All the interference powers generated by other interfering links, after being weighted by traffic density, attenuated by interference path channel, and suppressed by receiving antenna directionality, are accumulated to finally obtain the total interference signal strength suffered by the link.

[0089] Based on the signal strength and the interference signal strength of the link, a probability of interruption of the link is determined.

[0090] Based on the signal strength and the interference signal strength, and considering the receiver background noise, the link quality evaluation calculates a signal-to-interference-and-noise ratio of the link under the current network state. The probability of interruption is defined as the probability that the signal-to-interference-and-noise ratio of the link is lower than a minimum threshold value required to ensure reliable communication. Since both the signal strength and the interference strength contain random components introduced by the fading channel (reflected in the randomness of the channel gain), the signal-to-interference-and-noise ratio is a random variable. By using the statistical characteristics of the signal and the interference (for example, under Rayleigh fading, the signal power obeys an exponential distribution, and after the central limit theorem approximation, the total interference can be modeled as a Gaussian distribution, etc.), the probability that the signal-to-interference-and-noise ratio is lower than the threshold, i.e., the probability of interruption of the link, is obtained by a probability calculation method.

[0091] After obtaining the channel gain of the link, the link quality evaluation module further integrates the traffic density, the channel gain, and the antenna gain to finally calculate the probability of interruption of the link. The parameters required for calculating the probability of interruption are briefly described as follows: The traffic density of the link is provided by the interference modeling module, and reflects the intensity of the data flow carried on the link, which directly affects the level of co-channel interference.

[0092] As described in the above step, the channel gain of the link is calculated by an adaptive fading channel model, and reflects the attenuation and fluctuation of the signal under specific channel conditions.

[0093] The present application fully considers the characteristics of nodes equipped with directional antennas. The directivity pattern of the antenna, including the high gain in the main lobe direction and the suppression ability in the side lobe direction, is taken into account in the calculation. The high gain in the main lobe improves the strength of the desired signal, and the side lobe suppression effectively reduces the interference from the undesired direction, which jointly affect the final signal-to-interference-and-noise ratio.

[0094] In calculating the probability of interruption, the link quality evaluation module comprehensively considers the channel state, the antenna gain, and the interference data. Specifically, the present application adopts a directional antenna model, and the antenna gain formula can refer to related technologies.

[0095] The link quality evaluation module integrates all the factors, the interference weighted by the traffic density, the channel gain calculated by the adaptive model, and the gain and interference suppression effect brought by the directional antenna, into a unified probability calculation framework. Through a series of processing, the link quality evaluation module finally outputs a probability of interruption value for each link, which can accurately reflect the communication reliability of the link in the real and uncertain wireless environment, and provides a basis for subsequent construction of a tree-shaped weighted network topology graph and finally driving an iterative optimization algorithm to find an optimal relay deployment scheme.

[0096] In one optional implementation, the system further includes: a network topology initialization module; The environment perception module is also used to provide node location information and communication range information to the network topology initialization module. The node location information includes: the deployment location of each service node in the deployment environment obtained by the environment perception module, and the location of each candidate relay node determined by the environment perception module based on the obstacle information of the deployment environment, the deployment location of each service node, and the physical constraints of the relay node deployment. The communication range constraint information includes: the communication range constraints of each service node and the communication range constraints of each candidate relay node.

[0097] To automate and expedite the optimized deployment of relay nodes, this embodiment of the invention further includes a network topology initialization module. This module works in conjunction with the environment awareness module to construct the network infrastructure, laying a solid foundation for subsequent iterative optimization processes.

[0098] The environment awareness module goes beyond mere environment analysis; it extends to providing input data for the network topology initialization module. First, the environment awareness module acquires the deployment locations of each service node in the deployment environment. These service nodes are the terminals of network services, and their location information is the fundamental basis for constructing the network topology.

[0099] Secondly, the environmental perception module demonstrates its intelligent decision-making capabilities. It comprehensively analyzes and processes information about obstacles in the deployment environment, the deployment locations of each service node, and the physical constraints on relay node deployment. Specifically, the obstacle information includes physical entities that may obstruct the direct propagation of wireless signals, such as buildings and terrain undulations. The deployment locations of each service node refer to the terminal node distribution obtained above. Physical constraints on relay node deployment include areas where equipment cannot be deployed due to permission, power supply, or security reasons.

[0100] like Figure 2 As shown in (2), the network initialization module initializes the network in a given set of service nodes. and the pre-selected set of relay node locations Based on this, construct a tree-structured network topology diagram for the communication network. Define the communication range of the service node as follows: The communication range of the relay node is , where the set of nodes edge set It contains edges that satisfy the following condition: for a business node and a node ,side If and only if For two relay nodes , edge iff Meanwhile, the network initialization module selects the nearest candidate relay node to each business node as its initial parent node, and applies the Steiner tree approximation algorithm to connect the parent nodes, constructing an initial tree network topology that connects all sensor nodes (the darker square pattern in (2) represents the connected relay nodes).

[0101] Based on the comprehensive analysis of the above information, the environmental perception module determines the positions of each candidate relay node. These positions are not randomly assigned, but are the result of taking into account bypassing obstacles, optimizing coverage, and meeting actual deployment constraints, forming a candidate relay node set that not only conforms to physical reality but also has optimization potential.

[0102] In addition, the environmental perception module also provides the network topology initialization module with communication range constraint information, including the communication range constraints of each business node (due to power consumption and antenna limitations, the communication range is small) and the communication range constraints of each candidate relay node (equipped with stronger transmission power or antennas, with larger communication range). These range constraints are the fundamental criteria for determining whether any two nodes can establish a direct communication link.

[0103] The network topology initialization module is configured to construct an initial tree network topology graph based on the node position information and the communication range information, and to select a nearest candidate relay node as a parent node for each business node based on the initial tree network topology graph, and to connect each parent node to generate an initial tree network topology graph, and to provide the initial tree network topology graph to the interference modeling module for the first iteration.

[0104] After receiving complete information from the environmental perception module, the network topology initialization module starts the automated network topology construction process. First, the network topology initialization module constructs an initial tree network topology graph based on the node position information and the communication range information. The initial tree network topology graph is an undirected graph, with all business nodes and candidate relay nodes as vertices. The generation of edges strictly follows the communication range constraints: if the geometric distance between two nodes is less than or equal to the smaller communication range of the two, a direct communication link is established between them, indicating the existence of a potential direct communication link. This step constructs an initial connected graph that reflects physical reachability.

[0105] Subsequently, the network topology initialization module selects a nearest candidate relay node to each service node as its parent node. "Nearest" refers to the nearest in geometric distance, and the purpose is to preliminarily associate the service node to the potential relay node, and to prepare for forming a network structure with the relay node as the backbone. This process establishes an initial, local connection relationship for each service node.

[0106] Next, the initial tree network topology graph connects various parent nodes to generate an initial tree network topology graph. Here, the initial tree network topology graph uses the Steiner tree approximation algorithm in graph theory. All candidate relay nodes selected as parent nodes are regarded as "terminal nodes" that must be connected, while other candidate relay nodes are regarded as optional intermediate nodes (Steiner points). The algorithm aims to find a minimum-cost connected subgraph connecting all parent nodes in the initial tree network topology graph. The final generated initial tree network topology graph ensures that all service nodes (through their parent nodes) are connected in a tree structure, which guarantees network connectivity while preliminarily optimizing path overhead, forming a basic available and well-structured initial network.

[0107] Finally, the network topology initialization module provides the constructed initial tree network topology graph to the interference modeling module, marking the completion of the system initialization phase and starting the first iteration cycle of the entire optimization system. As shown in Figure 2 As shown in (3), the interference modeling module calculates the initial link traffic density according to the preset service mode and routing scheme. In the embodiment, the service mode is that the service node generates data with equal probability and randomly selects other service nodes to transmit data. According to the service mode, the traffic density of the On the link associated with it , the traffic density is used to weight the interference of the interference node , and the interference is averaged in time. In the embodiment, the routing scheme is to use the shortest path algorithm to construct the routing table with distance as the weight.

[0108] Figure 4 is a step diagram of a directional antenna-based wireless network relay node deployment method provided by an embodiment of the present application, applied to the system as described above, please refer to Figure 4 , including: Step S11, determine the traffic density of the link corresponding to each edge in the tree network topology graph of the current iteration. The traffic density of a link is the amount of service data transmitted between the two nodes connected by the link in unit time.

[0109] ​Based on the preset traffic pattern and routing scheme, the traffic density of each link corresponding to an edge in the tree network topology of the current iteration is calculated. The traffic density is defined as the amount of traffic data transmitted between two nodes connected by the link in unit time, reflecting the intensity of data flow on the link. In the actual network, the traffic pattern can include data generated by traffic nodes in an equal probability manner and randomly selected other traffic nodes as data transmission targets; the routing scheme can generate a routing table based on the shortest path principle or other optimization strategies. The interference modeling module analyzes these routing information and counts the amount of traffic carried on each link, thereby assigning a traffic density value to each link. This process not only considers the static traffic distribution, but also adapts to the dynamic changes of network traffic, ensuring the accuracy of interference evaluation.

[0110] In step S12, the outage probability of each link is determined according to the traffic density of each link in the tree network topology of the current iteration, and the outage probability of each link is combined with the node deployment cost of each node connected by the link to generate the edge weight of the edge corresponding to the link, thereby generating the tree weighted network topology of the current iteration.

[0111] The reliability of each link in the tree network topology of the current iteration is quantified. First, based on the traffic density data provided by the interference modeling module, the outage probability of each link is calculated. The outage probability is an index for measuring the communication reliability of the link in the random channel fluctuation and interference environment.

[0112] After obtaining the outage probability, it is further combined with the node deployment cost to generate the edge weight corresponding to each edge. The calculation of the edge weight integrates the node deployment cost at both ends of the link and the bidirectional outage probability of the link, forming a comprehensive index reflecting the balance between deployment economy and communication reliability. Finally, the tree weighted network topology of the current iteration is output, providing a quantitative basis for deployment optimization.

[0113] In step S13, according to the tree weighted network topology of the current iteration, a candidate relay node with the lowest edge weight is selected as a new parent node for each traffic node, and the new parent nodes are connected to generate the tree network topology of the next iteration and returned to the interference modeling module, so that the tree network topology of the next iteration is used for the next iteration.

[0114] An iterative mechanism is adopted to continuously improve the deployment scheme of the relay nodes. Based on the tree-shaped weighted network topology of the current iteration, a candidate relay node with the lowest edge weight is first selected as a new parent node for each service node. This selection process aims to minimize the comprehensive cost of the local link (including deployment cost and outage probability), thereby improving the overall network performance. Subsequently, these new parent nodes are connected to build a new tree-shaped network topology, ensuring that all service nodes are connected through relay nodes to achieve full network connectivity.

[0115] The newly generated tree-shaped network topology is fed back to the interference modeling module to recalculate the traffic density and interference model of the link, thereby starting the next iteration. This closed-loop optimization mechanism enables the system to dynamically adapt to the interference distribution adjustment caused by changes in network topology.

[0116] In step S14, the deployment positions of the relay nodes are output according to the tree-shaped weighted network topology of the last iteration.

[0117] After the iterative optimization process is completed, the tree-shaped weighted network topology generated in the last iteration is processed and formatted. The specific deployment position information of the relay nodes is extracted and converted into a user-readable coordinate output form. The output result can be directly used for network deployment practice, such as guiding the physical installation position of the relay nodes, to ensure the implementability of the deployment scheme in the actual environment.

[0118] In an optional embodiment, it further includes: In step S21, the channel state between each two nodes is probabilistically evaluated based on the noise signals in the deployment environment to determine whether the link between each two nodes is a line-of-sight path or a non-line-of-sight path.

[0119] The present application not only processes macro geographic information such as the predetermined position of the service node, topography, building distribution, etc., but also focuses on various noise signals that inevitably exist in the deployment environment. These noises may come from measurement errors, environmental electromagnetic interference, or other unpredictable random factors. Based on these perception information containing noise, the channel state between any two nodes in the network is probabilistically evaluated.

[0120] Specifically, the above probabilistic evaluation does not simply give a certain, "black or white" channel judgment, but outputs a probability value representing the likelihood of the link being a line-of-sight path. Line-of-sight path refers to the existence of an unobstructed direct propagation path between the transmitting node and the receiving node; otherwise, it is a non-line-of-sight path, and the signal mainly relies on reflection, diffraction, scattering, etc. Through probabilistic evaluation, the uncertainty of the channel state in the real world due to incomplete information and noise interference is effectively quantified, providing more realistic and robust input information for subsequent processing.

[0121] Step S22, for each link in the tree network topology of the current iteration, the following steps are performed: In the case that the link is a non-line-of-sight path, the channel gain of the link is determined by using a statistical channel model based on Rayleigh fading channel, and in the case that the link is a line-of-sight path, the channel gain of the link is determined by using a statistical channel model based on Rician fading channel.

[0122] For the link determined as a non-line-of-sight path, the channel gain of the link is determined by using a statistical channel model based on Rayleigh fading channel. The Rayleigh fading model is particularly suitable for describing the typical multipath fading environment where the signal experiences a large number of scatterers during propagation, and there is no dominant direct path. Under the Rayleigh fading model, the fluctuation characteristics of the channel gain are characterized by the Rayleigh distribution, which can accurately reflect the random variation of signal strength under non-line-of-sight propagation conditions.

[0123] For the link determined as a line-of-sight path, the channel gain of the link is determined by using a statistical channel model based on Rician fading channel. The Rician fading model is suitable for propagation scenarios where there is a stable main direct signal component, and there are multiple weak scattering components superimposed. The Rician fading model characterizes the relative strength of the direct path component and the scattering path component through the Rician factor, which can more accurately describe the statistical characteristics of the channel gain when there is a direct path, and makes up for the shortcomings of the traditional Rayleigh model in such scenarios.

[0124] Step S23, according to the channel gain of the link, the signal strength of the link is calculated.

[0125] Further according to the channel gain of the link, the signal strength of the link is calculated. The signal strength here specifically refers to the power strength of the expected signal at the receiving end. The calculation process takes into account the transmission power, the channel gain calculated through the corresponding fading channel model (the channel gain includes the effects of path loss and fading), and the high gain of the directional antenna in the main lobe direction. Since the nodes of the present application are equipped with directional antennas, when the transmitting node and the receiving node are located within the main lobe coverage range of each other's antennas, the main lobe gain of the antenna will significantly improve the power of the expected signal reaching the receiving end. Therefore, the finally calculated signal strength is a quantitative value that can reflect the advantages of directional transmission.

[0126] Step S24, according to the traffic density and corresponding channel gain and directional antenna gain of the other interfering links except the link, the interference signal strength of the link is determined.

[0127] Considering that in the network, when other links are transmitting data at the same time, it will cause co-channel interference to the current link. First, identify all other interference links in the network that can interfere with the current link. For each interference source link, its interference contribution is not equal to be seen. The transmission of other interference links is weighted by traffic density. The link with high traffic density means that the amount of data transmitted per unit time is large, and the average interference power generated by the interference source is also correspondingly stronger. For each other interference link to the link receiving node, its channel gain is also calculated. The calculation of channel gain is also based on the line-of-sight / non-line-of-sight judgment of the environment perception module for the interference path, and the statistical channel model based on Rayleigh fading channel or the statistical channel model based on Rician fading channel is selected.

[0128] Also considering that the gain of the receiving node's directional antenna of the current link (i.e. the link) is directional. If the interference signal comes from outside the main lobe direction of the receiving antenna, especially the side lobe or null region, then the antenna gain will be very low, thereby significantly suppressing the strength of the interference signal. Based on the relative relationship between the direction of arrival of the interference signal and the pointing direction of the receiving antenna, the corresponding antenna gain value is applied.

[0129] The interference power generated by all other interference links after being weighted by traffic density, channel attenuation of interference path, and directional suppression of receiving antenna is accumulated, and the total interference signal strength borne by the link is finally obtained.

[0130] Step S25, according to the signal strength and the interference signal strength of the link, the outage probability of the link is determined.

[0131] Based on the signal strength and the interference signal strength, and considering the receiver background noise, the link quality evaluation calculates the signal-to-interference-and-noise ratio of the link under the current network state. The outage probability is defined as the probability that the signal-to-interference-and-noise ratio of the link is lower than the minimum threshold value required to ensure reliable communication. Since both the signal strength and the interference strength contain random components introduced by the fading channel (reflected in the randomness of the channel gain), the signal-to-interference-and-noise ratio is a random variable. By using the statistical characteristics of signal and interference (for example, under Rayleigh fading, the signal power obeys exponential distribution, and after central limit theorem approximation, the total interference can be modeled as Gaussian distribution, etc.), the probability that the signal-to-interference-and-noise ratio is lower than the threshold, i.e. the outage probability of the link, is obtained by probability calculation method.

[0132] After obtaining the channel gain of the link, further integrate the traffic density, channel gain and antenna gain, and finally calculate the outage probability of the link. The parameters required for calculating the outage probability have been described in the foregoing, and will not be repeated here.

[0133] All the above factors, the interference weighted by the service density, the channel gain calculated by the adaptive model and the gain and interference suppression effect brought by the directional antenna are integrated into a unified probability calculation framework. Through a series of processes, an outage probability value that can accurately reflect the communication reliability of each link in the real and uncertain wireless environment is finally output, which provides a basis for subsequent construction of a tree-shaped weighted network topology graph and finally driving an iterative optimization algorithm to find an optimal relay deployment scheme.

[0134] The embodiment of the present application also provides an electronic device, including a processor, a memory and a computer program stored in the memory and capable of running on the processor, which realizes each process of the above-mentioned relay node deployment method based on a directional antenna wireless network embodiment and achieves the same technical effects. To avoid repetition, details are not described here.

[0135] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize each process of the above-mentioned relay node deployment method based on a directional antenna wireless network embodiment and achieve the same technical effects. To avoid repetition, details are not described here. Each embodiment in the specification is described in a progressive manner, and each embodiment mainly describes the difference from other embodiments. The same and similar parts of each embodiment can be referred to.

[0136] Although the preferred embodiments of the embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to the embodiments once they know the basic creative concept. Therefore, the appended claims are intended to include the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present application.

[0137] Finally, it should also be noted that in this document, relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or terminal device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or terminal device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or terminal device including the element.

[0138] The embodiments of the present application are described above with reference to the accompanying drawings, but the present application is not limited to the above-described specific embodiments, and the above-described specific embodiments are merely illustrative, but not restrictive, and a person of ordinary skill in the art can make many forms under the inspiration of the present application without departing from the purpose of the present application and the scope protected by the claims, and these all belong to the protection of the present application.

Claims

1. A wireless network relay node deployment system based on a directional antenna, comprising at least: The interference modeling module is used to determine the service density of the links corresponding to each edge in the tree network topology graph of the current iteration. The service density of a link is the amount of service data transmitted between the two nodes connected by the link per unit time. The link quality assessment module is used to determine the interruption probability of each link based on the service density of each link in the tree network topology graph of the current iteration, and to integrate the interruption probability of each link with the node deployment cost of each of the two nodes connected to the link as the edge weight of the corresponding edge of the link, so as to generate the tree weighted network topology graph of the current iteration. The deployment iteration optimization module is used to select a candidate relay node with the lowest edge weight as a new parent node for each service node based on the tree-weighted network topology graph of the current iteration, and connect each new parent node to generate the tree-structured network topology graph of the next iteration and return it to the interference modeling module so that the tree-structured network topology graph of the next iteration can be used for the next iteration. The output module is used to output the deployment locations of relay nodes based on the tree-weighted network topology graph of the last iteration.

2. The system of claim 1, wherein, Also includes: The environment awareness module is used to perform a probabilistic assessment of the channel state between each two nodes based on noise signals in the deployment environment, in order to determine whether the link between each two nodes is a line-of-sight path or a non-line-of-sight path. The link quality assessment module performs the following steps for each link in the current iteration of the tree network topology: When the link is a non-line-of-sight path, the channel gain of the link is determined using a statistical channel model based on Rayleigh fading channels; and when the link is a line-of-sight path, the channel gain of the link is determined using a statistical channel model based on Rice fading channels. Calculate the signal strength of the link based on its channel gain; The interference signal strength of the link is determined based on the traffic density of other interfering links besides the link itself and the corresponding channel gain and directional antenna gain. The probability of link interruption is determined based on the signal strength and interference signal strength of the link.

3. The system of claim 2, wherein, The system also includes: a network topology initialization module; The environment perception module is also used to provide node location information and communication range information to the network topology initialization module. The node location information includes: the deployment location of each service node in the deployment environment obtained by the environment perception module, and the location of each candidate relay node determined by the environment perception module based on obstacle information in the deployment environment, the deployment location of each service node, and the physical constraints of relay node deployment. The communication range constraint information includes: the communication range constraints of each service node and the communication range constraints of each candidate relay node. The network topology initialization module is used to construct an initial tree-like network topology based on the node location information and the communication range information, and to select the nearest candidate relay node as the parent node for each service node based on the initial tree-like network topology, and to connect each parent node to generate the initial tree-like network topology, and to provide the initial tree-like network topology to the interference modeling module for the first iteration.

4. The system of claim 2, wherein, The link quality assessment module is specifically used to, for the nth link in the current iteration of the tree network topology, treat each link in the current iteration of the tree network topology, excluding the nth link, as an interfering link of the nth link, and perform the following steps: The service signal strength of the nth link is determined based on the channel gain of the nth link and the antenna gains of the directional antennas of the two nodes connected to the nth link. The interference signal strength of the interference link is determined based on the channel gain of each interference link of the nth link and the antenna gain of the directional antennas of the two nodes connected to the interference link. Using the service density of each interfering link of the nth link as the weight, the interference signal strength of each interfering link of the nth link is weighted to obtain the total interference signal strength of the nth link. The signal-to-interference-plus-noise ratio (SIR) of the nth link is determined based on the service signal strength, noise signal strength, and total interference signal strength of the nth link. The probability of interruption of the nth link is determined based on the signal-to-interference-plus-noise ratio of the nth link.

5. The system according to claim 1, characterized in that, The link quality assessment module is specifically used to: determine the deployment cost of business nodes as zero; In the scenario where the deployment requirement is to maximize network performance, the node deployment cost of the candidate relay node is determined as the first value. In the scenario where the deployment requirement is to minimize the number of relay nodes, the node deployment cost of the candidate relay nodes is determined to be a second value, which is greater than the first value. For the nth link in the tree network topology graph of the current iteration, perform the following steps: In the scenario where the deployment requirement is to maximize network performance, and at least one of the two nodes connected by the nth link is a candidate relay node, the total node deployment cost of the nth link is determined based on the first value. In the scenario where the deployment requirement is to minimize the number of relay nodes, and at least one of the two nodes connected by the nth link is a candidate relay node, the total node deployment cost of the nth link is determined based on the second value. The probability of interruption of the nth link is combined with the total node deployment cost and used as the edge weight of the edge corresponding to the nth link.

6. The system according to claim 1, characterized in that, Deploy the iterative optimization module, specifically for: Based on the infinite impulse response filter mechanism, the generated tree network topology graph for the next iteration is fused with the tree network topology graphs from previous iterations to obtain the updated tree network topology graph for the next iteration. At the same time, the annealing temperature is reduced, and the updated tree network topology graph for the next iteration is returned to the interference modeling module so that the next iteration can be performed based on the reduced annealing temperature and the updated tree network topology graph for the next iteration.

7. A method for deploying wireless network relay nodes based on directional antennas, characterized in that, Applied to the system as described in any one of claims 1-6, comprising: Determine the service density of each link corresponding to each edge in the tree network topology graph of the current iteration. The service density of a link is: the amount of service data transmitted between the two nodes connected by the link per unit time. Based on the service density of each link in the tree network topology graph of the current iteration, the interruption probability of the link is determined, and the interruption probability of each link is combined with the node deployment cost of the two nodes connected to the link, which is used as the edge weight of the corresponding edge of the link to generate the tree weighted network topology graph of the current iteration. Based on the tree-weighted network topology of the current iteration, a candidate relay node with the lowest edge weight is selected as the new parent node for each service node, and each new parent node is connected to generate the tree-based network topology of the next iteration and returned to the interference modeling module so that the tree-based network topology of the next iteration can be used for the next iteration. Based on the tree-weighted network topology from the last iteration, output the deployment locations of the relay nodes.

8. The system according to claim 7, characterized in that, Also includes: Based on the noise signals in the deployment environment, the channel state between each pair of nodes is probabilistically evaluated to determine whether the link between each pair of nodes is a line-of-sight path or a non-line-of-sight path. For each link in the tree network topology graph of the current iteration, perform the following steps: When the link is a non-line-of-sight path, the channel gain of the link is determined using a statistical channel model based on Rayleigh fading channels; and when the link is a line-of-sight path, the channel gain of the link is determined using a statistical channel model based on Rice fading channels. Calculate the signal strength of the link based on its channel gain; The interference signal strength of the link is determined based on the traffic density of other interfering links besides the link itself and the corresponding channel gain and directional antenna gain. The probability of link interruption is determined based on the signal strength and interference signal strength of the link.

9. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the steps of the method as described in claim 7 or 8.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the steps of the method as described in claim 7 or 8.

Images