High fill foundation construction communication monitoring system based on internet of things

Abstract

The application relates to the field of civil engineering intelligent monitoring and Internet of Things application, and particularly discloses a high fill foundation construction communication monitoring system based on the Internet of Things. The system comprises a distributed sensing node array, a wireless communication network, a geotechnical-channel coupling analysis server, a routing dynamic optimization unit and a remote monitoring center; the sensing nodes are buried in the foundation at different depths to collect settlement, displacement, soil pressure and moisture content data, the wireless network synchronously records signal strength and uploads physical layer characteristics, the geotechnical-channel coupling analysis server fuses geotechnical parameters and a radio wave propagation model, constructs a four-dimensional coupling relationship between soil body characteristics and signal attenuation, the routing dynamic optimization unit predicts communication blind areas and adjusts the forwarding path accordingly, and the remote monitoring center realizes state visualization and construction decision support. The application realizes the deep integration of monitoring and communication, can maintain the sensing capability when part of the sensors fail, and guarantees the continuity and reliability of data transmission in a complex construction environment.

Description

Technical Field

[0001] This invention belongs to the field of intelligent monitoring and Internet of Things (IoT) application technology in civil engineering, specifically relating to an IoT-based communication and monitoring system for high embankment foundation construction. Background Technology

[0002] With the widespread application of IoT technology in civil engineering, sensor network-based monitoring systems have become a core tool for ensuring the quality and safety of high-fill foundation construction. By deploying large-scale sensor nodes inside the fill structure, construction teams can acquire key physical parameters such as foundation settlement, horizontal displacement, and stress changes in real time. In the construction of large airports, dams, and other infrastructure projects, continuous and accurate monitoring data plays a crucial role in assessing construction risks, dynamically optimizing the filling schedule, and ensuring the long-term stability of the foundation.

[0003] The communication monitoring technology for high-fill foundations based on wireless sensor networks aims to construct a data transmission link covering the entire construction area. This technology utilizes communication nodes embedded within the fill layer to aggregate environmental data sensed at each location via a wireless link to a monitoring center, enabling remote sensing of the foundation's internal condition. Its basic principle relies on a stable radio wave transmission path to establish a transparent data exchange channel in the complex field construction site, ensuring that monitoring indicators can promptly reflect changes in the geotechnical mechanical properties during construction.

[0004] Existing technologies face several challenges in monitoring high-fill foundation construction. Due to the intense compaction caused by heavy machinery at construction sites, precision sensors embedded within the soil are highly susceptible to physical damage, leading to frequent monitoring point failures and low survival rates. Traditional systems isolate wireless communication from the soil and rock environment, failing to effectively utilize the physical layer characteristics of the communication link itself to perceive environmental changes, resulting in complete loss of monitoring functionality after sensor damage. Furthermore, as the fill layer thickens and soil compaction changes, the propagation environment of wireless signals dynamically evolves. Existing technologies struggle to predict these resulting communication blind spots, preventing the system from proactively adjusting routing strategies based on terrain changes and impacting the continuity and reliability of monitoring data under complex conditions.

[0005] Therefore, a communication and monitoring system for high-fill foundation construction based on the Internet of Things is desired. Summary of the Invention

[0006] The purpose of this invention is to provide a communication and monitoring system for the construction of high embankment foundations based on the Internet of Things, which can solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] The IoT-based communication and monitoring system for high-fill foundation construction includes a distributed sensor node array, a wireless communication network, a geotechnical-channel coupling analysis server, a dynamic routing optimization unit, and a remote monitoring center, among which:

[0009] The distributed sensor node array is buried at different depths and in different areas of the high embankment foundation. It is configured to collect data on settlement, horizontal displacement, soil pressure and moisture content inside the foundation, and transmit the collected data to the wireless communication network through the built-in wireless transceiver module.

[0010] The wireless communication network consists of multiple relay nodes, forming multiple transmission links covering the entire filling area. It is used not only to forward sensor data, but also to continuously record the wireless signal strength and link quality indication information of each link, and synchronously upload the physical layer characteristic data to the geotechnical-channel coupling analysis server.

[0011] The geotechnical-channel coupling analysis server receives physical layer feature data from the wireless communication network and geotechnical parameters from the distributed sensor node array. Based on the fused geotechnical constitutive model and radio wave propagation model, it constructs a four-dimensional coupling relationship between soil density, water content, dielectric constant and wireless signal attenuation, and thereby infers the impact of changes in the internal structure of the filling body on the communication channel.

[0012] The routing dynamic optimization unit identifies potential communication blind spots in advance during the compaction of the filling layer based on the channel evolution prediction results output by the geotechnical-channel coupling analysis server, and dynamically adjusts the data forwarding paths of each relay node in the wireless communication network to ensure the continuity and reliability of monitoring data transmission.

[0013] The remote monitoring center is communicatively connected to the geotechnical-channel coupling analysis server and the routing dynamic optimization unit, and is used to display the foundation condition assessment results, communication link health and routing topology in real time, and to support construction management personnel in making dynamic decisions on filling processes.

[0014] Preferably, each sensor node in the distributed sensor node array adopts a pressure-resistant encapsulation structure, which is wrapped with flexible buffer material on the outside. It can maintain basic communication functions during the operation of large rolling machinery. Even if the internal sensing elements fail, its wireless transceiver module can still continue to work as a channel sensing unit.

[0015] Furthermore, the geotechnical-channel coupling analysis server is embedded with a finite element analysis algorithm, which is used to divide the fill body into multiple calculation units. Combined with the real-time updated soil density and moisture content distribution, the algorithm simulates the changes in dielectric properties of each unit and inversely retrieves the propagation path loss and multipath effect characteristics of wireless signals inside the fill body.

[0016] Furthermore, the relay nodes in the wireless communication network are equipped with an adaptive power adjustment function, which can automatically increase the transmission power or switch to a backup frequency band when the signal attenuation in a local area is predicted to increase, in order to maintain link stability, according to the instructions of the routing dynamic optimization unit.

[0017] Preferably, the routing dynamic optimization unit adopts a hierarchical response mechanism based on prediction confidence. When the channel quality degradation trend exceeds a preset threshold but has not yet caused link interruption, the adjacent redundant path is activated first. When the degradation trend rises sharply and approaches the communication failure critical point, the global routing reconstruction strategy is triggered to replan the optimal transmission path from the source node to the aggregation node.

[0018] Furthermore, even when some sensors fail, the geotechnical-channel coupling analysis server can still reconstruct the estimated foundation state of the missing area by utilizing the correlation between the wireless signal attenuation patterns of the remaining effective nodes and historical geotechnical parameters, and by using interpolation and pattern matching methods, thus maintaining the integrity of the overall sensing capability of the system.

[0019] Furthermore, the remote monitoring center is equipped with a visual interface that can overlay the compaction evolution process of the filling body with the prediction results of communication blind spots to form a joint "terrain-channel" situation map, which helps construction personnel to predict monitoring blind spots and adjust sensor deployment schemes before filling operations.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] 1. The IoT-based communication and monitoring system for high-fill foundation construction provided by this invention overcomes the limitations of traditional monitoring technologies that separate communication and sensing functions. For the first time, it incorporates the fill itself into the wireless channel modeling framework, achieving cross-domain coupling of geotechnical mechanical properties and electromagnetic wave propagation characteristics. Even if some sensors are physically damaged due to mechanical crushing, the system can still continuously acquire environmental status information through the signal characteristics fed back by its remaining wireless modules, improving the survival rate of monitoring nodes and the robustness of the system.

[0022] 2. The system can predict the formation trend of communication blind spots in advance based on the changes in density and moisture content during soil compaction, and drive the dynamic evolution of the routing structure, truly realizing the proactive adaptability of "communication evolving with the terrain". This technology not only ensures the continuity and reliability of monitoring data in complex construction environments, but also provides a new perception dimension and decision-making basis for intelligent construction of high embankment projects. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall technical solution architecture according to the present invention;

[0024] Figure 2This is a schematic diagram of the core principle framework of the geotechnical-channel coupling analysis server according to the present invention;

[0025] Figure 3 This is a flowchart illustrating the dynamic routing optimization logic based on prediction confidence according to the present invention.

[0026] Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow between distributed sensing nodes, wireless communication network and coupling analysis server according to the present invention.

[0027] Figure 5 This is a schematic diagram illustrating the principle of ground condition estimation and reconstruction using wireless signal attenuation modes in the event of partial sensor failure, according to the present invention. Detailed Implementation

[0028] Example 1: To make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments.

[0029] The IoT-based communication and monitoring system for high-fill foundation construction includes a distributed sensor node array, a wireless communication network, a soil-channel coupling analysis server, a dynamic routing optimization unit, and a remote monitoring center.

[0030] The distributed sensor node array is configured to be buried at different depths and in different areas of the high fill foundation, and includes multiple sensor units with independent sensing and communication capabilities. Each sensor unit is constructed with a modular structure, integrating high-precision settlement monitoring components, horizontal displacement detection components, earth pressure sensing components, and moisture content measurement components. The settlement monitoring component uses the principles of hydrostatic leveling or electromagnetic induction to capture real-time vertical displacement data of the soil caused by the weight of the fill material and mechanical compaction. The horizontal displacement detection component uses a combination of a built-in triaxial accelerometer and a gyroscope to achieve vectorized sensing of lateral slippage within the soil. The earth pressure sensing component uses piezoresistive or string-type pressure transmitters, deployed at the interfaces of different fill layers, to quantitatively analyze the spatial distribution characteristics of fill pressure. The moisture content measurement component uses frequency domain reflectometry or time domain reflectometry to obtain the percentage of water content between soil particles in real time, and uses this parameter as a key input for calculating the dielectric constant. Each of the sensing units is equipped with a low-power microprocessor and a wireless transceiver module electrically connected to it. The microprocessor is configured to perform analog-to-digital conversion, data preprocessing, and data packaging of various sensor signals according to a preset time sampling period, and then transmit the data packets carrying geotechnical mechanics parameters to the surrounding communication nodes through the wireless transceiver module.

[0031] The wireless communication network consists of multiple relay nodes deployed at and around the construction site, forming multiple self-organizing wireless transmission topologies covering the entire high-fill area. These relay nodes not only possess traditional data forwarding functions but are also tasked with sensing physical layer characteristics. During each data forwarding cycle, each relay node is configured to synchronously record key physical layer characteristic data such as received signal strength indication, link quality indication, signal-to-noise ratio, and bit error rate. The wireless communication network employs a hierarchical networking strategy, where lower-level nodes are responsible for short-range communication with the distributed sensor node array, while aggregation layer nodes are responsible for summarizing data from local areas and transmitting the data over long distances to the geotechnical-channel coupling analysis server via high-gain directional antennas. To cope with the complex electromagnetic environment and terrain evolution at the high-fill construction site, the wireless communication network supports multi-band adaptive switching, dynamically hopping frequencies between multiple preset working channels based on interference intensity to ensure the real-time performance and integrity of monitoring data transmission.

[0032] The geotechnical-channel coupling analysis server, serving as the intelligent core of the entire system, is configured to receive and process all uplink data from the wireless communication network, including raw geotechnical sensing parameters and physical layer characteristic data generated during transmission. Internally, the geotechnical-channel coupling analysis server operates a deeply integrated cross-domain analysis engine, which achieves heterogeneous fusion of the geotechnical constitutive model and the radio wave propagation model through software logic. The server first utilizes the received water content, soil pressure, and settlement displacement data, combined with a pre-set soil property parameter library, to construct the current real-time density distribution field of the fill material. The server's embedded dielectric property conversion module maps the physical density and water content of the soil to the real and imaginary parts of the dielectric constant in the electromagnetic dimension, based on the complex dielectric constant model. The server can transform traditional geotechnical calculation units into waveguide media with specific electromagnetic loss characteristics. The server uses finite element analysis algorithms to perform three-dimensional meshing of the entire filling area. Within each meshed unit, the attenuation characteristics of the wireless signal are superimposed to invert and deduce the comprehensive impact of the internal structural evolution of the filling body on the multipath distribution, path loss, and phase shift characteristics of the wireless signal. The soil-channel coupling analysis server is also configured to store historical monitoring data. Through long short-term memory networks or similar sequence analysis algorithms, it learns the correlation between soil settlement trends and signal attenuation curves, providing support for subsequent failure prediction.

[0033] The dynamic routing optimization unit interacts with the geotechnical-channel coupling analysis server at high speed. Its core function is to proactively intervene in network transmission paths based on current channel evolution prediction results. The dynamic routing optimization unit includes a routing decision machine configured to maintain a global network weight topology table in real time. In this table, the weight of each communication link depends not only on the physical distance between nodes but also on the expected link loss, calculated by the geotechnical-channel coupling analysis server and affected by soil compaction. As the fill layer thickens with construction, the dynamic routing optimization unit can identify potential blind spots that may cause communication interruptions due to increased dielectric loss. Once the link reliability in a certain area is identified as falling below a preset confidence threshold, the unit immediately calculates the alternative optimal path and generates a corresponding route reconstruction command, which is then sent to each relay node in the wireless communication network. This dynamic optimization mechanism based on physical evolution prediction allows the system to smoothly evolve its communication structure before significant changes occur in the foundation environment.

[0034] The remote monitoring center, configured as the system's integrated presentation and interactive terminal, is typically deployed at the construction command center or cloud management platform. Equipped with a multi-dimensional visualization engine, it displays the digital twin model output by the geotechnical-channel coupling analysis server as a three-dimensional image. On the display interface, the internal stress cloud map and settlement contour lines of the fill body are overlaid with the field strength thermal map of the wireless channel. Through this visualization, managers can intuitively observe communication signal attenuation caused by excessive local soil compaction or moisture accumulation. The remote monitoring center also integrates a construction management decision-making system, which automatically generates construction guidance suggestions based on the foundation stability assessment results and the health status of the communication link, such as adjusting the number of compaction passes, changing the fill material ratio, or adding temporary communication relays. The remote monitoring center also supports historical traceability, allowing users to review the "soil-channel" coupling data of specific construction stages, providing evidence for project quality acceptance.

[0035] In the aforementioned system, the distributed sensor node array exhibits high environmental adaptability. The pressure-resistant encapsulation structure of the sensing unit is constructed as a rigid shell with a specific geometric configuration. This rigid shell is made of high-strength composite material, capable of withstanding the instantaneous static and dynamic pressures of hundreds of kilopascals generated by a large road roller. Between the rigid shell and the internal precision sensing components, a flexible buffer material of a predetermined thickness is filled. This flexible buffer material is configured to possess nonlinear damping characteristics, capable of absorbing and dissipating the impact energy generated during compaction, preventing brittle fracture of the components.

[0036] The wireless transceiver module and main control unit of the sensing unit are located in the core protected area of ​​the encapsulation structure. Even if the external displacement probe or pressure sensor suffers mechanical damage or electrical failure due to uneven foundation settlement, the wireless transceiver module can still maintain its working state as long as the power supply circuit remains intact. At this time, the failed sensor node will change its role, and its main function will change from data acquisition to a simple channel detection unit. It continuously sends specific detection frames, cooperating with the peer node in the wireless communication network to record the signal attenuation characteristics in the damaged area. The geotechnical-channel coupling analysis server then uses these signal characteristics from the failed node, combined with the data of surrounding effective nodes, and through pattern matching and interpolation algorithms, can still infer the change in soil density in the area even in the absence of direct mechanical readings. This design solves the pain point of low monitoring survival rate in high embankment construction and realizes the leap from single sensor perception to group channel perception.

[0037] The finite element analysis algorithm embedded in the geotechnical-channel coupling analysis server is configured to execute the following logical process during actual operation: First, the system divides the three-dimensional high embankment model into millions of tetrahedral or hexahedral computational units; then, based on the real-time collected geotechnical parameters, each computational unit is assigned corresponding mechanical and electromagnetic properties. In the computational logic, the server is configured to establish a mapping function describing the proportional relationship between soil density and dielectric constant, and simultaneously establish an algebraic model describing the correlation between water content and the tangent of the medium loss angle. By performing numerical solutions to the wave equation, the server can simulate the refraction, reflection, and diffraction paths of electromagnetic waves in non-uniform soil media. By comparing the simulated path loss values ​​with the link quality data actually transmitted back by the wireless communication network, the server can reverse-correct parameter deviations within the foundation, achieving high-precision inversion of the foundation state.

[0038] When relay nodes in the wireless communication network perform tasks, their adaptive power adjustment function is configured as a hierarchical feedback control logic. Each relay node stores a preset transmit power level table. When the routing dynamic optimization unit detects that the signal attenuation rate of a certain transmission path exceeds a preset linear threshold, it indicates that the soil layer traversed by the path is undergoing a rapid compaction process, and the dielectric loss is increasing sharply. At this time, the routing dynamic optimization unit generates a power boost command. The relay node receiving this command will precisely adjust the output power of its RF front-end through its internal digital variable gain amplifier according to the gain decibel value specified in the command. If the link quality still fails to recover to the preset communication baseline after the power boost, the relay node is configured to activate the frequency band migration mechanism, switching from the center frequency band with severe interference or drastic attenuation to a low-frequency backup frequency band with stronger penetration capability, thereby ensuring that the monitoring link can still maintain a minimum level of data synchronization during the extreme phase of high-fill operations.

[0039] The routing dynamic optimization unit employs a hierarchical response mechanism based on predicted confidence levels. Its specific execution logic is as follows: The system calculates a numerical indicator representing the probability of link interruption within a preset time period based on the channel evolution trend output by the geotechnical-channel coupling analysis server. When this indicator is within a first preset interval, it is defined as a low-risk state, and the routing dynamic optimization unit only maintains close monitoring of the link. When the indicator rises to a second preset interval, exceeding a preset warning threshold but not yet reaching the disconnection critical point, the system initiates a first-level response, i.e., searching for and preheating nearby redundant paths in the existing topology, without changing the current main data flow direction. When the indicator further climbs and enters a third preset interval, i.e., approaching the communication failure critical point, the system triggers a second-level response, i.e., performing global route reconstruction. During reconstruction, the system uses a weighted shortest path algorithm, employing the predicted attenuation value of each link as the edge weight, to calculate a new transmission path that can bypass high-attenuation areas. This hierarchical response strategy balances network configuration stability and response speed to environmental changes, avoiding additional energy overhead and data latency caused by frequent path switching.

[0040] The visualization interface configured in the remote monitoring center employs layered rendering technology to achieve joint "terrain-channel" situational awareness display. The base layer is updated in real time based on the latest UAV mapping data or construction logs to generate a digital terrain model, displaying the macroscopic geometry of the fill layer. The middle layer renders the stress distribution and deformation trend within the foundation based on settlement and displacement data transmitted from the distributed sensor node array, typically using color depth to represent numerical values. The top layer is covered with a semi-transparent channel quality heatmap, where the brightness of the color blocks represents the coverage strength of the wireless signal. When a signal "black hole" occurs in a certain area due to abnormal soil density, the system highlights the area on the visualization interface and simultaneously pops up a related dialog box, informing management personnel of potential construction quality hazards, such as soft and elastic roadbed phenomena caused by excessively high local moisture content. This provides a three-dimensional holographic view from macro to micro, from physical to electromagnetic, for precise intervention at the construction site.

[0041] Example 2: As another embodiment of the present invention, the Internet of Things-based high embankment foundation construction communication monitoring system adopts a system architecture that combines distributed edge computing with centralized cloud decision-making, aiming to further improve the data processing efficiency and real-time response of ultra-large-scale high embankment projects.

[0042] In this embodiment, the system also includes a distributed sensor node array, a wireless communication network, a geotechnical-channel coupling analysis server, a dynamic routing optimization unit, and a remote monitoring center. However, unlike Embodiment 1, the relay nodes in the wireless communication network are upgraded to edge processing nodes. These edge processing nodes are configured with more powerful computing capabilities and integrate lightweight geotechnical mechanics evaluation algorithms.

[0043] The interaction between the distributed sensor node array and the edge processing nodes is not limited to the transmission of raw data. The edge processing nodes are configured to perform preliminary feature extraction on the signals collected by each sensor node locally. For example, they can extract slope features from settlement monitoring data and fluctuation frequency features from moisture content data. This localized processing mechanism aims to filter out a large amount of redundant data, uploading only information with changing characteristics to subsequent analysis stages, thereby reducing the transmission pressure on the backbone network.

[0044] The geotechnical-channel coupling analysis server is constructed in a cluster mode within this architecture, distributed across different server rooms at the construction site. Each server sub-cluster is responsible for maintaining the local geotechnical-channel coupling model for its assigned area. To achieve global consistency, each server sub-cluster synchronizes its clock and model via high-speed fiber optic links. The geotechnical-channel coupling analysis server is further configured to include a cross-domain correlation inference module, which is capable of processing multi-source unstructured data. When a distributed sensor node array in a certain area experiences a high failure rate due to extensive construction compaction, the server sub-cluster initiates a virtual data compensation logic based on a generative adversarial network. Specifically, the server utilizes the physical layer characteristics fed back from the residual wireless communication links, combined with the geological survey report from before construction in that area, to simulate and generate a set of virtual sensor readings that conform to current physical and mechanical laws. These readings are used to fill in the gaps in the foundation assessment model, ensuring that the system's overall judgment of foundation stability is not biased due to localized physical damage.

[0045] In this distributed architecture, the dynamic routing optimization unit evolves into a peer-to-peer network optimization strategy based on a distributed consensus algorithm. Instead of relying on centralized command issuance, each edge processing node exchanges channel health information through a neighbor discovery protocol based on its perceived local channel state. The dynamic routing optimization unit includes a local decision microkernel configured to perform game theory-based optimal forwarding strategy optimization. When a sudden change in soil density in a local area causes a simultaneous increase in signal attenuation on multiple existing links, the affected edge nodes spontaneously negotiate through the microkernel in multiple rounds to jointly determine a temporary routing path that maximizes the overall system throughput. This decentralized design enhances the system's self-healing capability against large-scale hardware failures.

[0046] At the hardware implementation level, the distributed sensor node array employs an enhanced energy self-sufficiency design. In addition to conventional chemical battery power, each sensor unit's casing is covered with a piezoelectric energy harvesting patch. This patch is configured to capture the mechanical energy generated during heavy rolling or vibratory compaction of the fill material and convert it into electrical energy stored in an internal supercapacitor. This feature ensures that the sensor nodes will not cease operation due to power depletion during high-fill construction cycles lasting several years. Furthermore, the wireless transceiver module of the sensor unit employs spread spectrum modulation technology with enhanced multipath resistance, enabling stable channel synchronization in highly scattering environments such as soil.

[0047] In this embodiment, the finite element analysis algorithm running in the soil-channel coupling analysis server further integrates the dynamic boundary element method. During the simulation of the construction evolution of the fill material, the server is configured to adjust the model's topology in real time to simulate the continuous stacking of new fill layers. Whenever construction machinery completes a new layer of earthwork and initiates compaction, the system automatically senses the upward movement of the physical boundary. At this time, the coupling analysis server dynamically adds a computational unit to the top of the model and sets initial moisture migration parameters based on the current ambient temperature and humidity. This dynamic evolution simulation method allows the soil-channel coupling model to follow the construction progress in real time, enabling dynamic prediction of foundation conditions and channel quality.

[0048] In this embodiment, the remote monitoring center is configured to support augmented reality (AR) display. Managers wearing specialized visual devices can directly view a virtual profile superimposed on the ground surface within the construction site's field of view. Through this AR technology, construction workers can visually identify areas where signal shielding zones are forming due to excessive moisture content within the soil, and, combined with real-time settlement rate warnings, quickly pinpoint potential landslide or collapse hazards. The monitoring center is also configured with a remote intervention interface, supporting online adjustment of the sampling rate of sensor nodes in specific areas via wireless links, enabling the activation of a high-frequency encrypted monitoring mode upon detection of anomalies.

[0049] Example 3: As another embodiment of the present invention, an IoT-based communication and monitoring system for high embankment foundation construction is specifically optimized for high embankment projects with extreme terrain changes or complex material compositions. In such conditions, the fill material may contain large-diameter boulders or special chemically modified soil, which can cause highly uneven scattering of wireless signals.

[0050] In this embodiment, the system's components are endowed with deeper physical adaptability. Each sensing unit in the distributed sensing node array is constructed as a composite sensor with "electronic tag" attributes. Before leaving the factory, each sensing unit has its antenna's original radiation pattern in free space precisely measured and recorded. Once these nodes are buried in the soil, the soil-rock coupling analysis server compares the actual received field distortion to derive the dielectric non-uniformity index of the soil within a 3-meter radius around the node. This microscopic sensing capability enables the system to identify whether large cavities or uncompacted rock accumulation areas exist within the fill material.

[0051] In this embodiment, the wireless communication network incorporates cooperative multipoint transmission technology. In this configuration, multiple relay nodes are organized into a cooperative transmission group. When a sensing unit located deep underground transmits data, multiple surrounding relay nodes are configured to simultaneously acquire the signal and converge the received multipath components to a designated cooperative processor via a microsecond-level synchronization protocol. The cooperative processor uses a maximum ratio combining algorithm or interference cancellation algorithm to reconstruct the original data stream from the attenuated and distorted signal. This cooperative transmission mechanism enhances the signal penetration capability in ultra-deep and ultra-thick embankments, ensuring that critical monitoring data can still be reliably uploaded even under extreme conditions where signal attenuation exceeds 100 dB.

[0052] The geotechnical-channel coupling analysis server employs stochastic medium theory when handling high-fill models with such complex materials. The server is configured to no longer treat the soil as a continuous, homogeneous medium, but rather as a stochastic complex composed of discrete particles and pore fluids. Under this model, the propagation of wireless signals is described as a multiple scattering process among randomly distributed scatterers. The coupling analysis logic within the server predicts how the energy flow of radio waves disperses within the complex rock-fill structure by solving the Beth-Salpité equations or applying radiative transfer theory. This statistical physics-based analysis method can more accurately describe the rapid signal fading phenomenon in large rock foundations, providing a more scientific reference for routing decisions.

[0053] In this embodiment, the dynamic routing optimization unit integrates a reinforcement learning-based routing policy seeker. Since the environmental evolution under complex conditions is difficult to describe with simple analytical expressions, the policy seeker is configured to automatically learn the optimal forwarding path through continuous interaction with the environment. The system defines channel quality, data latency, node remaining energy, and foundation stability risk together as the reward function. During construction, the routing policy seeker continuously tries different link combinations and updates its internal neural network weights based on the actual transmission effects. As construction time increases, the system exhibits increasingly stronger environmental adaptability, automatically matching the topology with the strongest anti-interference capability for different filling stages.

[0054] The remote monitoring center further integrates an AI-based risk warning expert system. This expert system is configured not only to monitor numerical alarm thresholds but also to focus on the co-evolution patterns between the foundation's multiphysics field and the communication physical layer. For example, when the expert system detects an increase in settlement in a certain area accompanied by a significant increase in the number of wireless signal multipaths, the system will automatically perform correlation analysis to conclude that the area may be experiencing the loss of fine-grained soil, leading to increased internal porosity. This in-depth analysis based on cross-domain logic enables the system to identify early damage characteristics of the foundation's microstructure before visible cracks or macroscopic deformation occur.

[0055] Furthermore, to address sudden electromagnetic interference from construction machinery on wireless communication, the wireless communication network is also equipped with cognitive radio functionality. The relay nodes can monitor the surrounding electromagnetic spectrum occupancy in real time, automatically identifying the characteristics of interference signals generated by construction walkie-talkies, engineering vehicle dispatch systems, and other construction equipment. The dynamic routing optimization unit directs nodes in the affected area to automatically avoid these interfered frequency bands, or improves the signal capture resistance by adjusting the spreading factor. This keen awareness and rapid response to the dynamic spectrum environment ensures the robustness of the monitoring system as the last line of defense for construction safety.

[0056] The IoT-based communication and monitoring system for high-fill foundation construction described in the above embodiments of this invention, by deeply integrating geotechnical constitutive models and electromagnetic wave propagation theory, changes the isolated mode of "sensing is sensing, transmission is transmission" in traditional engineering monitoring. The system fully utilizes the massive physical medium of the fill, transforming it from an obstacle to signal transmission into a medium for sensing environmental changes. This "soil-sensing and co-evolution of soil and information" technical approach not only ensures the continuity of monitoring functions in harsh construction environments where hardware damage is frequent, but also provides a novel technical means to reverse-verify construction quality by analyzing communication link characteristics. For national-level projects with high requirements for foundation stability, such as large hydropower hubs, high-standard civil aviation airports, and approach roads of cross-sea bridges, the system provided by this invention not only reduces the full life-cycle maintenance cost of the monitoring system, but also lays a solid technical foundation for realizing the digital transformation and intelligent decision-making in high-fill construction.

[0057] The collaborative logic among the components of this invention is highly consistent. Whether it is the underlying data acquisition of distributed sensing nodes or the path reconstruction of the dynamic routing optimization unit, it all revolves around the core objective of improving monitoring reliability and foundation assessment accuracy. Through in-depth mining of physical layer characteristics and precise simulation of mechanical evolution processes, the system achieves comprehensive control over the complex, dynamic, and hidden internal world of high embankments, filling the gap in existing technologies in the field of interdisciplinary sensing, and possessing engineering application value and broad market prospects.

[0058] Example 4: This example is a customized implementation for the construction scenario of high embankment foundation for civil airport runways. The fill material is silty clay, the filling layer thickness is 30cm, the number of compaction passes is 6-8, the monitoring depth is 0-15m, the specific sensor component model is selected, the core threshold parameters are quantified, and a practical data processing case is provided for the entire process from raw data acquisition to early warning output from the monitoring center.

[0059] In this embodiment, the system still includes a distributed sensor node array, a wireless communication network, a geotechnical-channel coupling analysis server, a dynamic routing optimization unit, and a remote monitoring center. Each component adopts a deterministic configuration with quantified parameters, as detailed below:

[0060] 4.1 Specific configuration of the distributed sensor node array:

[0061] The sensing unit selects a fixed-type sensing component combination, and specifies the hardware parameters, sampling rules, and preprocessing formulas, abandoning the approach of only describing directionality. The specific configuration is as follows:

[0062] Settlement monitoring component: adopts the hydrostatic leveling principle, with a range of 0-500mm, an accuracy of ±0.1mm, and a sampling cycle of 10s;

[0063] Horizontal displacement detection component: adopts a triaxial accelerometer + gyroscope, with a range of ±16g / ±2000° / s, an accuracy of ±0.01g / ±0.1° / s, and a sampling period of 5s;

[0064] Earth pressure sensing component: adopts a string pressure transmitter with a range of 0-2MPa and an accuracy of ±0.5%FS. It is deployed at the interface of the fill layer and has a sampling cycle of 10s.

[0065] Moisture content measurement component: Employs time-domain reflectometry, with a measurement range of 0-100%, accuracy of ±1%, and a sampling period of 30 seconds;

[0066] Core hardware: Low-power microprocessor, wireless transceiver module (operating frequency band 433MHz, transmit power adjustable from 5-20dBm).

[0067] Sensing unit data preprocessing formula (executed by the microprocessor, input is the raw analog signal, output is standardized geotechnical parameters):

[0068] Water content to dielectric constant conversion formula:

[0069] ;

[0070] Soil moisture content, The dielectric constant of the soil is used as input for subsequent channel coupling analysis.

[0071] Earth pressure correction formula, eliminating temperature effects, and correction of raw data for string transmitters:

[0072] ;

[0073] To correct the earth pressure, The original pressure collected by the sensor. Temperature correction factor (take) ), The soil temperature at the time of collection. The calibration temperature is 20℃.

[0074] Horizontal displacement vectorization calculation:

[0075] ;

[0076] This represents the horizontal displacement in the X / Y directions. For the horizontal acceleration collected by the accelerometer, The initial velocity is 0 (taken as 0, initial state of rest). Sampling time.

[0077] The sensor unit is encapsulated in a pressure-resistant manner: the rigid shell is made of glass fiber reinforced plastic (FRP) with a compressive strength of ≥300MPa, and the flexible buffer material is polyurethane foam (thickness 5cm, damping coefficient 0.35). The core protection zone can withstand an instantaneous rolling pressure of 200kPa, which meets the requirements of heavy roller operation in airport construction.

[0078] 4.2 Quantization Parameters and Physical Layer Feature Acquisition for Wireless Communication Networks:

[0079] In this embodiment, the wireless communication network adopts a LoRa+4G hierarchical networking approach. The relay nodes are configured with deterministic thresholds and power adjustment parameters, and the physical layer feature acquisition is based on quantified values ​​rather than merely describing feature types.

[0080] Network topology: The bottom layer nodes (LoRa) communicate with the sensor nodes at a distance of 0-500m, while the aggregation layer nodes (4G) enable long-distance transmission at a rate of 1Mbps.

[0081] Physical layer feature acquisition items: Received Signal Strength Indicator (RSSI), Link Quality Indicator (LQI), Signal-to-Noise Ratio (SNR), Bit Error Rate (BER). The acquisition period is synchronized with the sensor node sampling period, and the data forwarding period is 30s.

[0082] Core threshold parameters:

[0083] RSSI warning threshold: ≤-110dBm (signal attenuation warning).

[0084] LQI communication baseline: ≥80 (out of 100; a value below this is considered a failure to meet link quality standards).

[0085] Minimum SNR threshold: ≥-5dB (signal interference exists below this value);

[0086] BER failure threshold: ≥10 -3 (A value higher than this indicates a link transmission failure).

[0087] Adaptive power adjustment parameters: Transmit power is divided into 5 levels (5 / 10 / 15 / 18 / 20dBm). When RSSI≤-110dBm, the power is increased by 5dBm. If LQI is still <80 after the increase, it switches to the low-frequency spare band (315MHz, with improved penetration).

[0088] Signal attenuation calculation formula (executed at the wireless communication network end, input is raw RSSI data, output is link signal attenuation value, providing physical layer feature input for coupling analysis):

[0089] ;

[0090] This represents the signal attenuation (dB) caused by the soil medium. Transmit power (dBm) The transmit / receive antenna gain is 2dBi. The wavelength of electromagnetic waves (m, corresponding to 433MHz) ), This represents the physical distance between nodes.

[0091] 4.3 Specific calculation model and threshold for the geotechnical-channel coupling analysis server:

[0092] In this embodiment, the coupling analysis server constructs a specific formula for the four-dimensional coupling relationship (soil density) for high fills of silty clay. Moisture content Dielectric constant Signal attenuation The document clarifies the simplified parameters for finite element analysis, the prediction formula for link interruption probability, and quantifies all analysis thresholds, specifically:

[0093] Soil density calculation formula (input is earth pressure) Moisture content (The output is the real-time density of the soil).

[0094] ;

[0095] 1.65 represents the dry density of the soil, and 1.65 represents the initial dry density of the silty clay. To correct the earth pressure.

[0096] The core formula for four-dimensional coupling (integrating the soil and rock constitutive model and the radio wave propagation model, which is the core coupling formula of this invention, with input as follows) The output is the predicted signal attenuation. ):

[0097] ;

[0098] To predict signal attenuation, The embedment depth of the sensing nodes is 0.85 / 0.52 / 0.38 / 0.015, which are the fitting coefficients for silty clay (obtained through indoor tests and field calibration).

[0099] Finite element analysis parameters: The 0-15m monitoring area is divided into 10cm×10cm×10cm hexahedral computational elements (a total of 15×100×100=150000 elements). The numerical solution of the wave equation adopts the finite difference method, with a time step of 1e-6s and a spatial step of 0.05m.

[0100] Link interruption probability prediction formula (input is the predicted signal attenuation) The output is the probability of link interruption in the next 30 minutes. (to provide a basis for decision-making in route optimization)

[0101] Core threshold for coupling analysis:

[0102] Dielectric constant normal range: 3.4 to 8.0;

[0103] Predicted signal attenuation warning threshold: ;

[0104] Long Short-Term Memory (LSTM) network training parameters: Input layer dimension 4 ( ), 2 hidden layers, 64 neurons, 32 training batches, 100 iterations.

[0105] 4.4 Specific thresholds and algorithms for hierarchical response of the dynamic routing optimization unit:

[0106] In this embodiment, the routing dynamic optimization unit adopts a hierarchical response mechanism based on prediction confidence to clearly define the link interruption probability. The three-level interval threshold is given, and the specific weight calculation method of the weighted shortest path algorithm is given without fuzziness. Specifically:

[0107] Three-level threshold for link interruption probability:

[0108] First preset range (low risk): Simply maintain monitoring without adjusting the routing;

[0109] Second pre-set interval (medium risk / Level 1 response): Preheat adjacent redundant paths, with a number of redundant paths ≥ 2;

[0110] Third preset interval (high risk / level 2 response): Perform a global route reconstruction.

[0111] Weighted shortest path algorithm weight formula (input is physical distance) Predicted signal attenuation The output is the link weight. (The smaller the weight, the better the path)

[0112] ;

[0113] The maximum physical distance for networking is 500m. The maximum predicted signal attenuation is 30dB, and 0.3 / 0.7 are weighting coefficients (prioritizing the impact of signal attenuation).

[0114] Route reconfiguration execution time: ≤5s; after reconfiguration, the link LQI must be ≥85, and the BER must be ≤5×10. -4 .

[0115] 4.5 Early Warning Thresholds and Output Formats of Remote Monitoring Center:

[0116] In this embodiment, the remote monitoring center clearly defines the quantitative early warning threshold for foundation stability and provides a linked output format of "geotechnical parameters - channel characteristics - early warning results" to achieve a closed loop from data input to decision output. The specific early warning threshold is as follows:

[0117] Settlement warning threshold: single settlement ≥ 5 mm / 30 min, cumulative settlement ≥ 50 mm;

[0118] Horizontal displacement warning threshold: single horizontal displacement ≥3mm / 30min, cumulative horizontal displacement ≥30mm;

[0119] Earth pressure warning threshold: Earth pressure at the interface of the fill layer ≥ 150 kPa (adapted value for silty clay with 30cm layer compaction).

[0120] Moisture content warning threshold: (The optimal moisture content for silty clay is 18%-22%; if it exceeds 25%, the roadbed is prone to becoming soft and elastic.)

[0121] Warning output rules: When a single parameter reaches the warning threshold, a yellow warning will pop up on the interface; when ≥2 parameters reach the warning threshold and are accompanied by signal attenuation... When this happens, a red warning pops up on the interface, and construction decision suggestions are automatically generated, such as stopping compaction, allowing the soil to dry, and adding intermediate nodes.

[0122] 4.6 Case study of end-to-end data processing from input to output:

[0123] Taking the 5m deep silty clay fill layer of the airport high embankment construction in this embodiment as an example, the complete numerical processing process of a certain sensor node (number: S5-03, buried at a depth of 5m, the 16th layer of the fill layer) from the original data acquisition to the output of the remote monitoring center is given. All data are actual simulated values ​​collected on site, and the formulas all adopt the above-mentioned deterministic formulas.

[0124] Step 1: Acquisition of raw data from sensor nodes (input)

[0125] Acquisition time: 30s, Acquisition environment: Soil temperature 25℃, Physical distance between nodes d=100m, Transmission power =10dBm, the original collected data is as follows:

[0126] Hydrostatic level sensor: Original settlement simulation signal → Corresponding original settlement value =6.2mm;

[0127] MPU6050: , Collection time ;

[0128] String-type pressure transmitter: raw pressure =160kPa;

[0129] TDR350: Original moisture content measurement value =26%.

[0130] Step 2: Data preprocessing at the sensor unit microprocessor end

[0131] The standardized geotechnical parameters are calculated using the above preprocessing formula and output as follows:

[0132] Moisture content correction: TDR data were used directly. Substituting into the dielectric constant formula, we get:

[0133] ;

[0134] Earth pressure temperature correction:

[0135] ;

[0136] Horizontal displacement vectorization calculation:

[0137] ;

[0138] Settlement: Corrected single settlement (No temperature effect, use directly).

[0139] Preprocessing output: , , , , , The data is transmitted to the wireless communication network via the LoRa module.

[0140] Step 3: Physical Layer Feature Acquisition and Signal Attenuation Calculation of Wireless Communication Network

[0141] The relay node collects raw physical layer characteristic data and calculates and outputs physical layer characteristic parameters using the signal attenuation formula.

[0142] Raw data acquisition: RSSI = -112dBm, LQI = 75, SNR = -3dB, BER = 1.2 × 10⁻⁶ -3 ;

[0143] Signal attenuation calculation:

[0144] ;

[0145] Power adjustment judgment: RSSI=-112dBm≤-110dBm, trigger power increase, transmit power is increased from 10dBm to 15dBm. After the increase, LQI=78, still <80, trigger frequency band switching, switch from 433MHz to 315MHz.

[0146] Communication layer output: RSSI=-112dBm, LQI=75, power increased to 15dBm, switched to 315MHz, this data and geotechnical parameters are uploaded synchronously to the geotechnical-channel coupling analysis server.

[0147] Step 4: Geotechnical-channel coupling analysis server calculation

[0148] The server receives soil and rock parameters and physical layer characteristics, calculates using density formulas, four-dimensional coupling formulas, and link interruption probability formulas, and outputs coupling analysis results:

[0149] Soil dry density calculation:

[0150] ;

[0151] Calculation of predicted signal attenuation (buried depth h=5m):

[0152] ;

[0153] Link interruption probability calculation ( ≤10dB <20dB range):

[0154] ;

[0155] Coupling analysis determination: (Warning threshold) (Upper limit of normal range), soil density (Initial density) indicates insufficient soil compaction, high moisture content, and increased signal attenuation.

[0156] Coupling analysis output: , , The channel evolution prediction result is medium risk, and this result is sent to the routing dynamic optimization unit.

[0157] Step 5: Dynamic route optimization unit path adjustment

[0158] The unit determines and performs route optimization based on the link interruption probability, and outputs route adjustment instructions:

[0159] Risk assessment: ,belong (Second preset interval), initiate Level 1 response;

[0160] Redundant path warm-up: Select two adjacent redundant paths in the existing topology. Path 1: S5-03→R05→R08→convergence layer, Path 2: S5-03→R06→R09→convergence layer.

[0161] Weight calculation (taking path 1 as an example, d=120m, =14.2dB):

[0162] ;

[0163] Command issuance: Send a preheating command to relay nodes R05, R08, R06, and R09 to keep the redundant paths active while the main path continues to transmit normally.

[0164] Route optimization output: Warm up 2 redundant paths, path 1 weight 0.403, path 2 weight 0.421, no global route reconstruction.

[0165] Step 6: Visualized output and early warning from the remote monitoring center (final output)

[0166] The monitoring center receives the coupling analysis results and route optimization results, combines them with the foundation stability early warning threshold to make a judgment, and outputs visualization results, early warnings, and construction suggestions:

[0167] Single parameter determination:

[0168] Settlement amount: 6.2mm / 30min ≥ 5mm / 30min (settlement warning threshold), triggered;

[0169] Horizontal displacement: 13.5mm / 30min ≥ 3mm / 30min (horizontal displacement warning threshold), triggered;

[0170] Earth pressure: 159.832 kPa ≥ 150 kPa (earth pressure warning threshold), triggered;

[0171] Moisture content: 26% ≥ 25% (moisture content warning threshold), triggered;

[0172] Coupling determination: All four geotechnical parameters triggered the early warning, and (Signal attenuation warning threshold) triggers a red alert;

[0173] Routing status display: A 3D visualization interface overlaid with settlement contour lines (settlement in this area is 6.2mm, marked with a red block), channel field strength heat map (signal attenuation in this area is 52.9dB, marked with a dark block), and routing topology (main path + 2 preheating redundant paths marked with blue lines).

[0174] Construction decision recommendations: Automatically generate on-site construction adjustment recommendations, which state: "The soil moisture content in the S5-03 area at a depth of 5m is too high (26%) and the compaction is insufficient, resulting in settlement / horizontal displacement exceeding the limit. It is recommended to immediately stop the rolling operation in this area, allow the soil to dry until the moisture content is below 22%, and re-compact it twice with a light roller. At the same time, add a relay node (R10) in this area to strengthen the communication link."

[0175] Step 7: Emergency Handling of Failed Nodes

[0176] If the settlement probe of sensor node S5-03 is mechanically damaged due to compaction by a road roller (sensor element failure, power supply circuit intact), the following emergency handling procedure shall be followed:

[0177] Node role switching: The wireless transceiver module switches from data transmission to channel detection unit, continuously sending detection frames (frame frequency 1Hz, frame length 128 bytes).

[0178] Communication network data acquisition: Surrounding relay nodes acquire the RSSI of this probe frame (-115dBm, LQI=70) and calculate the signal attenuation. ;

[0179] Coupled analysis inversion: The server combines the geotechnical parameters of three surrounding valid nodes (S5-02, S5-04, S4-03) The soil and rock parameters of the failure node region were inverted using the Kriging interpolation algorithm:

[0180] ;

[0181] Inversion of soil moisture content; For soil dry density inversion;

[0182] The monitoring center outputs: The inverted data fills the blank areas of the model, and the interface is marked "The sensor element of this node has failed. The soil and rock parameters have been inverted through channel characteristics. The red warning is still in effect. The construction recommendations remain unchanged."

[0183] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention.

[0184] Any modifications, equivalent substitutions, or improvements made by those skilled in the art to the technical solutions disclosed in this invention without departing from the spirit and principles of the invention should be included within the scope of protection of this invention. In the description of this specification, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, those skilled in the art can combine and integrate different embodiments or examples and features of different embodiments or examples described in this specification without contradiction. Moreover, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified. All the textual mathematical expressions used in this invention are intended to clearly define the logical relationships between the variables, and their interpretation should follow the recognized physical laws and logical principles in this field.

Claims

1. A communication and monitoring system for high embankment foundation construction based on the Internet of Things, characterized in that, include: A distributed sensor node array is configured to be buried at different depths and in different areas of a high embankment foundation to collect soil and rock parameters inside the foundation and transmit the collected data to a wireless communication network via a built-in wireless transceiver module; the distributed sensor node array contains multiple sensor units with independent sensing capabilities. The wireless communication network consists of multiple relay nodes deployed at and around the construction site, forming multiple transmission links covering the entire filling area. The wireless communication network is used to forward the geotechnical parameters and continuously record the physical layer characteristic data of each transmission link, and synchronously upload the physical layer characteristic data to the geotechnical-channel coupling analysis server. A geotechnical-channel coupling analysis server is configured to receive the physical layer feature data and the geotechnical parameters, and based on the fused geotechnical constitutive model and radio wave propagation model, construct a four-dimensional coupling relationship between soil density, water content, dielectric constant and wireless signal attenuation, thereby inferring the impact of changes in the internal structure of the filling body on the communication channel. A dynamic routing optimization unit, which is communicatively connected to the geotechnical-channel coupling analysis server, is used to dynamically adjust the data forwarding path of each relay node in the wireless communication network based on the channel evolution prediction results. A remote monitoring center is used to display the foundation status assessment results, communication link health, and routing topology in real time. The wireless transceiver module and the main control unit of the sensing unit are arranged in the core protected area of ​​the encapsulation structure. When the external displacement probe or pressure sensor is mechanically damaged or electrically failed due to uneven settlement of the foundation, if the power supply circuit is intact, the wireless transceiver module is transformed into a channel detection unit. By continuously sending detection frames, it works with the wireless communication network to record the signal attenuation characteristics in the damaged area. The geotechnical-channel coupling analysis server uses signal characteristics from failed nodes, combined with data from surrounding valid nodes, to infer changes in soil density in the area through pattern matching and interpolation algorithms. The dynamic routing optimization unit includes a routing decision machine, which is configured to maintain a global network weight topology table in real time. In the global network weight topology table, the weight of each communication link is determined not only by the physical distance between nodes, but also by the expected link loss calculated by the geotechnical-channel coupling analysis server. The routing dynamic optimization unit adopts a hierarchical response mechanism based on prediction confidence. The system calculates a numerical index of the probability of link interruption within a preset time period based on the channel evolution trend. When the index is in the first preset interval, it is defined as a low-risk state, and the routing dynamic optimization unit keeps monitoring it. When the indicator is in the second preset range, that is, when it exceeds the preset warning threshold but has not yet reached the disconnection critical point, the system initiates the first level response, which searches for and warms up adjacent redundant paths in the existing topology. When the indicator enters the third preset interval, that is, when it is close to the communication failure critical point, the system starts the second level response, performs global route reconstruction, and uses the weighted shortest path algorithm to use the predicted attenuation value of each link as the edge weight to calculate a new transmission path that bypasses the high attenuation area. The core formula for four-dimensional coupling relationship: ； To predict signal attenuation, The embedment depth of the sensing nodes is given, and 0.85 / 0.52 / 0.38 / 0.015 are the fitting coefficients for silty clay. For soil density, Where is the dielectric constant. This refers to the moisture content.

2. The IoT-based communication monitoring system for high embankment foundation construction according to claim 1, characterized in that: The sensing unit is constructed in a modular structure, which integrates a settlement monitoring component, a horizontal displacement detection component, an earth pressure sensing component, and a moisture content measurement component. The settlement monitoring component uses the principle of hydrostatic leveling or electromagnetic induction to capture the vertical displacement data of the soil caused by the self-weight of the fill and mechanical compaction. The horizontal displacement detection component uses a combination of a built-in triaxial accelerometer and a gyroscope to achieve vectorized perception of the lateral slippage within the soil. The soil pressure sensing component uses a piezoresistive pressure transmitter or a string pressure transmitter, and is deployed at the interface of different filling layers to quantitatively analyze the spatial distribution characteristics of filling pressure. The moisture content measurement component obtains the percentage of water content between soil particles based on frequency domain reflectance technology or time domain reflectance technology, and uses soil and rock parameters as input values ​​for dielectric constant calculation. Each of the sensing units is equipped with a low-power microprocessor and a wireless transceiver module electrically connected to it. The microprocessor is configured to perform analog-to-digital conversion, data preprocessing, and packetization of various sensor signals according to a preset time sampling period, and then transmit the data packets carrying the soil and rock parameters to the surrounding relay nodes through the wireless transceiver module.

3. The IoT-based communication monitoring system for high embankment foundation construction according to claim 2, characterized in that: The sensing unit adopts a pressure-resistant encapsulation structure, which includes a rigid shell with a geometric configuration. The rigid shell is made of high-strength composite material and is configured to withstand the instantaneous static and dynamic pressure generated by construction machinery. A flexible buffer material of a predetermined thickness is filled between the rigid outer shell and the internal sensing components. The flexible buffer material has nonlinear damping characteristics and is used to absorb and dissipate the impact energy during the rolling process.

4. The IoT-based communication monitoring system for high embankment foundation construction according to claim 1, characterized in that: The relay nodes in the wireless communication network synchronously record physical layer characteristic data in each data forwarding cycle. The physical layer characteristic data includes received signal strength indication, link quality indication, signal-to-noise ratio, and bit error rate. The wireless communication network adopts a hierarchical networking strategy, in which the bottom layer nodes are responsible for short-range communication with the distributed sensor node array, the aggregation layer nodes are responsible for summarizing the data of the local area, and transmitting the data to the soil-channel coupling analysis server over long distance through a high-gain directional antenna. The wireless communication network supports multi-band adaptive switching and can dynamically hop frequencies between multiple preset working channels based on interference intensity.

5. The IoT-based communication monitoring system for high embankment foundation construction according to claim 4, characterized in that: The relay node is configured with an adaptive power adjustment function, which executes hierarchical feedback control logic; The relay node stores a preset transmit power level table. When the dynamic routing optimization unit detects that the signal attenuation rate of a certain transmission path exceeds a preset linear threshold, the dynamic routing optimization unit generates a power boost command. The relay node that receives the power boost command adjusts the output power of its RF front-end through its internal digital variable gain amplifier according to the gain decibel value specified in the command. If the link quality still fails to recover to the preset communication baseline after the power is increased, the relay node will activate the frequency band migration mechanism to switch from the current operating frequency band to a low-frequency backup frequency band with stronger penetration capability.

6. The IoT-based communication monitoring system for high embankment foundation construction according to claim 1, characterized in that: The geotechnical-channel coupling analysis server runs a cross-domain analysis engine. The cross-domain analysis engine first uses the received water content, soil pressure and settlement displacement data, combined with a preset soil physical property parameter library, to construct a real-time density distribution field of the fill body. The geotechnical-channel coupling analysis server is embedded with a dielectric property conversion module. Based on the complex dielectric constant model, the dielectric property conversion module maps the physical density and water content of the soil to the real and imaginary parts of the dielectric constant in the electromagnetic dimension, and transforms the geotechnical calculation unit into a waveguide medium with electromagnetic loss characteristics. The geotechnical-channel coupling analysis server is also configured to store historical monitoring data and learn the correlation between geotechnical settlement trends and signal attenuation curves through a long short-term memory network to predict communication failure risks.

7. The IoT-based communication monitoring system for high embankment foundation construction according to claim 6, characterized in that: The geotechnical-channel coupling analysis server is embedded with a finite element analysis algorithm. The finite element analysis algorithm divides the three-dimensional high fill model into millions of tetrahedral or hexahedral calculation units and assigns corresponding mechanical and electromagnetic properties to each calculation unit based on the real-time collected geotechnical parameters. The server establishes a mapping function describing the proportional relationship between soil density and dielectric constant, and establishes an algebraic model describing the correlation between water content and the tangent of medium loss angle. By performing numerical calculations of the wave equation, the server simulates the refraction, reflection, and diffraction paths of electromagnetic waves in non-uniform soil media, and compares the simulated path loss values ​​with the link quality data actually transmitted back by the wireless communication network, thereby correcting the parameter deviations inside the foundation and realizing the inversion of the foundation state. During the simulation of the construction evolution of the filling body, the server dynamically adjusts the topology of the model according to the construction progress, adds a computing unit at the top of the model, and sets the moisture migration parameters according to the current ambient temperature and humidity.

8. The IoT-based communication monitoring system for high embankment foundation construction according to claim 1, characterized in that: The geotechnical-channel coupling analysis server is equipped with a cross-domain correlation inference module. When a large proportion of the distributed sensor node array in a certain area fails, the server starts the virtual data compensation logic based on generative adversarial network. Using the physical layer characteristics fed back by the residual wireless communication link and the geological survey report before construction in the area, it simulates and generates a set of virtual sensor readings that conform to the current physical and mechanical laws to fill the blank areas in the foundation assessment model. The routing dynamic optimization unit evolves into a peer-to-peer network optimization strategy based on a distributed consensus algorithm under a distributed architecture. Each edge processing node exchanges channel health information through a neighbor discovery protocol based on its perceived local channel state, and executes an optimal forwarding strategy based on game theory to determine a temporary routing path that maximizes the overall system throughput.

9. The IoT-based communication monitoring system for high embankment foundation construction according to claim 1, characterized in that: The remote monitoring center is equipped with a multi-dimensional visualization engine, which is configured to display the digital twin model output by the geotechnical-channel coupling analysis server in the form of a three-dimensional image, and to co-construct and superimpose the internal stress cloud map and settlement contour lines of the filling body with the field strength thermal map of the wireless channel on the display interface. The visualization engine uses layered rendering technology. The base layer displays the macroscopic geometry of the filling layer based on the digital terrain model. The middle layer renders the stress distribution and deformation trend inside the foundation based on the settlement and displacement data transmitted back by the distributed sensor node array. The top layer is covered with a semi-transparent channel quality heat map, in which the brightness of the color blocks represents the coverage strength of the wireless signal. The remote monitoring center also integrates an AI-based risk warning expert system. This expert system monitors the collaborative evolution pattern between the foundation's multiphysics field and the communication physical layer. When it detects an increase in settlement in a certain area accompanied by an increase in the number of wireless signal multipaths, it automatically performs correlation analysis to arrive at an assessment conclusion that fine-grained soil loss has led to increased internal porosity in that area, and highlights the potential hazard area on the interface.

Images