Adaptive Wireless Connection Optimization Method for Intelligent Environment Perception

Abstract

The present invention discloses an adaptive wireless connection optimization method for intelligent environment perception, which relates to the technical field of wireless connection. By using sensors and scanning devices in step S1 to construct a perception data set GW, the wireless device can obtain the structural semantic information of the propagation environment for the first time, thereby constructing an extended path loss model that is more accurate than the traditional model in S2. Compared with the existing fixed power configuration scheme, this mechanism has the capabilities of environmental self-perception and path self-correction, avoiding the energy waste and uneven signal coverage caused by blind transmission. Based on the communication stability index Γs calculated in real time, it is compared with the target channel stability threshold Γtar, and the current transmission power PT that minimally satisfies the communication quality in the current area is deduced inversely, and the actual transmission power PTadj is obtained by combining the perturbation fine-tuning mechanism. This set of mechanisms realizes the two-way coupling control between power and stability, no longer configures redundant power globally, and can effectively control the upper limit of power consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless connection, and specifically to an adaptive wireless connection optimization method for intelligent environment perception. Background Art

[0002] As the backbone of modern communication networks, wireless communication systems undertake the main function of data interaction between terminal devices, and have been widely applied in scenarios such as Wi-Fi, Internet of Things, smart home, and intelligent office. With the increasing complexity of the physical space layout, the interaction relationship between wireless communication and the built environment has gradually become a research hotspot. Among them, the dynamic behavior modeling of signal propagation in the structural environment and the device power regulation and control problems are particularly crucial. Therefore, how to achieve the device's cognition and adaptive control of the current environment based on spatial topology structure perception information has become an important research direction for improving communication performance and energy efficiency ratio.

[0003] Most existing wireless communication devices adopt a fixed transmit power mechanism, that is, the device sets a unified signal transmit intensity during initialization and maintains it unchanged throughout the operation process. However, there are a large number of channel perturbation factors in the actual environment, such as differences in wall thickness, frequent changes in furniture layout, and different reflectivity of floor materials. These factors will significantly affect the signal path attenuation and reflection angle distribution.

[0004] The limitation of the fixed transmit power strategy lies in its inability to perceive environmental changes and thus unable to make timely adjustments. When environmental perturbations such as furniture movement, personnel flow, and door and window opening and closing occur frequently, the original channel model is broken, and the propagation path changes significantly at the physical level. However, the communication device still maintains the initial power parameters, resulting in a disconnect between the transmit power and the actual requirements. This disconnect will cause a series of abnormal consequences: in high-interference areas, the signal cannot reach the receiving end due to severe penetration attenuation, resulting in unstable connections, reduced rates, and increased delays; in low-obstruction areas, excessive power is likely to cause side effects such as co-channel interference, device overheating, and increased energy consumption. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention provides an adaptive wireless connection optimization method for intelligent environment perception, which solves the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: An adaptive wireless connection optimization method for intelligent environment perception, including the following steps:

[0007] S1. Use sensors and scanning devices to collect information in the propagation environment, fit it into the original data set SW, and perform preprocessing to obtain the perception data set GW;

[0008] S2. By using the perception dataset GW, construct an extended path loss model and calculate the total signal attenuation Lt on different propagation paths;

[0009] S3. Combine the obtained total signal attenuation Lt with the received power Pr of the target area and the topology perturbation data CL on-site, evaluate the communication stability index Γs of each area in the space, and draw a spatial energy efficiency map;

[0010] S4. Combine the obtained communication stability index Γs with the preset target channel stability threshold Γtar, inversely deduce the current required transmission power PT in the current environment, and calculate and obtain the actual transmission power PTadj;

[0011] S5. According to the obtained actual transmission power PTadj and communication stability index Γs, conduct a comprehensive efficiency evaluation of the global area to obtain the evaluation index TΦ;

[0012] S6. According to the obtained evaluation index TΦ, topology perturbation data CL, and communication stability index Γs, after detecting dynamic topology changes such as furniture position changes and personnel movements, adjust the degree of perturbation of the signal environment to obtain the power adjustment value ΔPT required for real-time compensation.

[0013] Preferably, S1 includes S11 and S12;

[0014] S11. Scan and sense the environment using sensor devices, collect information data in the propagation environment, including the wall dielectric constant change rate QD, average reflection angle Er, surface material reflection coefficient Pm, furniture interference factor Hf, topology perturbation data CL, and received power Pr of the target area, and construct the original dataset SW;

[0015] Among them, the wall dielectric constant change rate QD is collected by a millimeter-wave radar;

[0016] The average reflection angle Er is collected by a ToF depth camera;

[0017] The surface material reflection coefficient Pm is collected by a multispectral infrared scanner;

[0018] The furniture interference factor Hf is collected by a high-frequency sonar;

[0019] The topology perturbation data CL is collected by an IMU dynamic attitude sensor;

[0020] The received power Pr of the target area is collected by a received signal monitor;

[0021] S12. Clean and normalize the obtained original dataset SW to obtain the perception dataset GW;

[0022] Data cleaning includes using the statistical outlier detection method to remove outliers from the original dataset SW, removing the noise data higher than the mean value, and obtaining the data cleaning set CW;

[0023] Normalization processing includes normalizing the data cleaning set CW, mapping the data uniformly to the interval [0, 1], and obtaining the perception dataset GW;

[0024] The perception dataset GW is obtained through the following formula:

[0025] ;

[0026] In the formula, GMb represents the b-th data in the perception dataset GW, CWb represents the b-th data in the data cleaning set CW, minCWb represents the valley value of the b-th data in the data cleaning set CW, and maxCWb represents the peak value of the b-th data in the data cleaning set CW.

[0027] Preferably, S2 includes S21 and S22;

[0028] S21. For the obtained perception dataset GW, path feature extraction is performed, including the material structure correction factor QPD, the angle multipath attenuation factor Qer, the environmental interference factor HCf, and the path length exponent correction factor DMp, and they are fitted into the path dataset LF for model training to construct a multi-factor extended path loss model;

[0029] The material structure correction factor QPD is obtained through the following formula:

[0030] ;

[0031] The angle multipath attenuation factor Qer is obtained through the following formula:

[0032] ;

[0033] In the formula, cos represents the cosine function, and π represents the pi;

[0034] The environmental interference factor HCf is obtained through the following formula:

[0035] ;

[0036] The path length exponent correction factor DMp is obtained through the following formula:

[0037] ;

[0038] In the formula, CD represents the path loss factor;

[0039] S22. Calculate the total signal attenuation Lt on different propagation paths through the obtained path dataset LF and the multi-factor extended path loss model.

[0040] The total signal attenuation Lt is obtained through the following formula:

[0041] ;

[0042] In the formula, Lo represents the free space reference path loss, log represents the logarithmic function, respectively represent the preset weight values of the material structure correction factor QPD, the angular multipath attenuation factor Qer, and the environmental interference factor Hcf, and d represents the physical length of the propagation path.

[0043] Preferably, S3 includes S31 and S32;

[0044] S31. Combine the obtained total signal attenuation Lt with the received power Pr of the target area and the topological perturbation data CL on-site to calculate the communication stability index Γs of the area;

[0045] The communication stability index Γs is obtained through the following formula:

[0046] ;

[0047] In the formula, X represents the perturbation sensitivity coefficient, with a set value of 0.1 - 0.5, indicating the sensitivity control of the structural perturbation.

[0048] Preferably, S32. Construct the regional signal energy efficiency index CKs according to the obtained communication stability index Γs, in combination with the actual space topological coordinate system;

[0049] The energy efficiency index CKs is obtained through the following formula:

[0050] ;

[0051] In the formula, ck represents a constant;

[0052] Combine the spatial points (x, y, z) in the three-dimensional environment with the energy efficiency index CKs to form an energy efficiency mapping color gradient diagram in the three-dimensional space;

[0053] The gradient diagram is obtained through the following matching method:

[0054] Red area: 0 < energy efficiency index CKs < 0.3, indicating unstable signal and low transmission energy efficiency;

[0055] Yellow area: 0.3 ≤ energy efficiency index CKs ≤ 0.7, indicating signal coverage acceptance;

[0056] Green area: 0.7 < energy efficiency index CKs, indicating stable communication and efficient power usage.

[0057] Preferably, S4 includes S41 and S42;

[0058] S41. First, establish the relationship between theoretical received power and transmitted power;

[0059] Recover the received power Pr = PT - Lt in the target area from the channel loss model, substitute it into the acquisition formula of the communication stability index Γs, and obtain the theoretical stability expression:

[0060] ;

[0061] In the formula, Γtheo represents the theoretical communication stability index;

[0062] Reverse solve the target power formula: According to the communication stability index Γs, inversely deduce the current required transmitted power PT in the current environment;

[0063] The current transmitted power PT is obtained through the following formula:

[0064] ;

[0065] In the formula, Γtar represents the preset target channel stability threshold.

[0066] Preferably, by introducing a perturbation response fine-tuning mechanism and constructing a perturbation response compensation model to adjust the current transmitted power PT, the actual transmitted power PTadj is obtained, eliminating the interference caused by the dynamic perturbation characteristics of the environment, such as personnel movement and furniture movement, to ensure the stable operation of the system under slightly changing conditions;

[0067] Obtain the perturbation change rate ΔCL through the perturbation response compensation model:

[0068] The perturbation change rate ΔCL is obtained through the following formula:

[0069] ;

[0070] In the formula, ∂ represents the partial derivative;

[0071] According to the obtained perturbation change rate ΔCL, define the dynamic perturbation response function ϕ(CL) to correct the transmitted power:

[0072] The dynamic perturbation response function ϕ(CL) is obtained through the following formula:

[0073] ;

[0074] In the formula, β represents the perturbation response coefficient;

[0075] The actual transmission power \(P_{Tadj}\) is obtained by the following formula:

[0076] .

[0077] Preferably, S5 includes S51 and S52;

[0078] S51. Construct the energy efficiency index \(Q_{us}\) of the area through the actual transmission power \(P_{Tadj}\) and the communication stability index \(\Gamma_s\);

[0079] The energy efficiency index \(Q_{us}\) is obtained by the following formula:

[0080] ;

[0081] In the formula, \(Q_{us}(i)\) represents the energy efficiency index of the \(i\)-th area, \(\Gamma_s(i)\) represents the communication stability index of the \(i\)-th area, \(P_{Tadj}(i)\) represents the actual transmission power of the \(i\)-th area, \(B\) represents a real number to prevent division by 0, and is usually set to 0.01.

[0082] Preferably, S52. Integrate the energy efficiency indexes \(Q_{us}\) of all areas, conduct a comprehensive efficiency evaluation of the global area, obtain the evaluation index \(T_{\Phi}\), and judge the overall signal distribution state;

[0083] The evaluation index \(T_{\Phi}\) is obtained by the following formula:

[0084] ;

[0085] In the formula, \(N\) represents the number of evaluation areas, \(e\) represents a constant, \(k\) represents the distance attenuation coefficient, and \(D(i)\) represents the distance from the \(i\)-th area to the signal source;

[0086] The overall signal distribution state is obtained by matching in the following way:

[0087] When \(0 \lt\) evaluation index \(T_{\Phi} \lt 0.5\), it means that there are signal blind spots and power waste;

[0088] When \(0.5 \leq\) evaluation index \(T_{\Phi} \leq 0.8\), it means that the communication is generally acceptable, but there is power redundancy;

[0089] When \(0.8 \lt\) evaluation index \(T_{\Phi} \lt 1.0\), it means that the overall system configuration is reasonable, the power is used efficiently, and the communication is stable.

[0090] Preferably, S6. Extract the characteristics of the perturbation change rate \(\Delta CL\) obtained according to the topology perturbation data \(CL\) to obtain the perturbation trend factor \(\lambda_p\);

[0091] The perturbation trend factor \(\lambda_p\) is obtained by the following formula:

[0092] ;

[0093] In the formula, tanh represents the hyperbolic tangent function, and λa represents the variation sensitivity adjustment coefficient;

[0094] According to the obtained perturbation trend factor λp, combined with the evaluation index TΦ and the communication stability index Γs, calculate the additional power adjustment value ΔPT required, and judge the state of the wireless power;

[0095] The power adjustment value ΔPT is obtained through the following formula: as shown in Table 1 specifically:

[0096] ;

[0097] In the formula, represents the power adjustment sensitivity factor;

[0098] The state of the wireless power is obtained through the following matching method:

[0099] When 0 < power adjustment value ΔPT < 0.5, it means that the power is stable and no additional compensation is required;

[0100] When 0.5 ≤ power adjustment value ΔPT < 1, it means that the power is abnormal and power adjustment is required. Combine the obtained power adjustment value ΔPT with the actual transmission power PTadj to obtain the final power output value PTfin;

[0101] The final power output value PTfin is obtained by adding the actual transmission power PTadj and the power adjustment value ΔPT.

[0102] The present invention provides an adaptive wireless connection optimization method for intelligent environment perception, which has the following beneficial effects:

[0103] (1) Using sensors and scanning devices to construct the perception data set GW, enabling wireless devices to obtain the structural semantic information of the propagation environment for the first time, thereby constructing a more accurate extended path loss model than traditional models in S2. Compared with the existing fixed power configuration scheme, this mechanism has the capabilities of environmental self-perception and path self-correction, avoiding energy waste and uneven signal coverage caused by blind transmission.

[0104] Based on the real-time calculated communication stability index Γs, compare it with the target channel stability threshold Γtar, and inversely deduce the current transmission power PT that meets the minimum communication quality in the current area. Combine it with the perturbation fine-tuning mechanism to obtain the actual transmission power PTadj. This set of mechanisms realizes the two-way coupling control between power and stability, that is, instead of configuring redundant power globally, it is "quantified according to demand", effectively controlling the power consumption upper limit.

[0105] (2) Introduce a variety of heterogeneous sensors to perform directional acquisition and classification processing on key physical variables in the wireless signal propagation environment, including millimeter-wave radar to obtain the change rate QD of the wall dielectric constant, ToF depth camera to obtain the average reflection angle Er, and infrared scanner to obtain the material reflection coefficient Pm, and model the channel environment from multiple dimensions such as material characteristics, reflection behavior, and structural perturbation. Compared with the traditional model with RSSI or distance as a single variable, this embodiment significantly enhances the system's spatial resolution ability and propagation semantic understanding ability of the real physical environment.

[0106] Clean and standardize the original dataset SW, uniformly map data with different dimensions and distribution characteristics to the [0,1] interval, and particularly use a statistical outlier removal algorithm to eliminate high-noise points and environmental perturbation pseudo-values in the sensing data, finally obtaining a highly reliable sensing dataset GW. This not only improves the consistency of the model input and data reliability but also lays a high-stability and highly reliable basic data support for subsequent path modeling and power regulation.

[0107] (3) Use the obtained total signal attenuation Lt, received power Pr in the target area, and topological perturbation data CL, and combine with the perturbation sensitivity coefficient X to establish a calculation mechanism for the communication stability index Γs. This approach enables the system to perform partitioned and dynamic quantitative perception of the communication quality in different regions. Compared with the traditional single judgment method relying on RSSI, it is more robust and has higher spatial accuracy, effectively solving the key problems of "inability to determine signal blind spots and unawareness of excessive signal waste".

[0108] In step S32, by coupling the communication stability index Γs with factors such as device power, the system constructs a regional energy efficiency index CKs for the first time, realizing a two-layer index linkage evaluation from communication stability to energy efficiency utilization. This mechanism not only measures the "strength" of the signal but also measures the "effect produced by unit power", making the wireless connection no longer only focus on whether the signal is "covered", but more on whether it is "covered reasonably and used efficiently", improving the overall system perspective of energy management and quality of service control.

[0109] (4) Construct a regional-level energy efficiency index through the combined relationship between the actual transmission power and the communication stability index, quantifying the communication quality return brought by unit power in each spatial region. This mechanism realizes an evaluation method in the dimension of "output power cost performance", breaking through the evaluation limitations of traditional wireless systems that only rely on static indicators such as signal strength or reception success rate, providing a more scientific and comparable optimization basis for power regulation, and making energy efficiency control measurable and traceable.

[0110] In S52, through the fusion processing of the energy efficiency data of all regions, a system-level comprehensive efficiency index is formed to judge the rationality of the overall signal distribution and the effectiveness of power allocation. Compared with the traditional method of only local optimization, this embodiment realizes the transformation from "single-point optimization" to "global optimization", enabling the system to comprehensively grasp whether there are blind spots, signal waste, power overload and other problems in the current configuration, possessing the system-level efficiency intelligent judgment ability, and improving the balance and stability of the overall network deployment. BRIEF DESCRIPTION OF THE DRAWINGS

[0111] Figure 1 It is a schematic flow chart of the steps of the adaptive wireless connection optimization method for intelligent environment perception of the present invention;

[0112] Figure 2 It is a schematic block diagram flow chart of the power adjustment of the present invention;

[0113] Figure 3 It is a line graph of the power adjustment value of the present invention;

[0114] Figure 4 It is an area graph of the power adjustment value of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0115] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0116] Embodiment 1

[0117] The present invention provides an adaptive wireless connection optimization method for intelligent environment perception. Please refer to Figures 1 to 4 , including the following steps:

[0118] S1. Use sensors and scanning devices to collect information in the propagation environment, fit it into the original data set SW, and perform preprocessing to obtain the perception data set GW;

[0119] S2. By using the perception data set GW, construct an extended path loss model and calculate the total signal attenuation Lt on different propagation paths;

[0120] S3. Combine the obtained total signal attenuation Lt with the target region received power Pr received on-site and the topology perturbation data CL, evaluate the communication stability index Γs of each region in the space, and draw a space energy efficiency map;

[0121] S4. Combine the obtained communication stability index Γs with the preset target channel stability threshold Γtar, inversely deduce the current required transmission power PT in the current environment, and calculate and obtain the actual transmission power PTadj;

[0122] S5. Based on the obtained actual transmission power PTadj and communication stability index Γs, conduct a comprehensive effectiveness evaluation of the global area to obtain the evaluation index TΦ;

[0123] S6. Based on the obtained evaluation index TΦ, topology perturbation data CL, and communication stability index Γs, after detecting dynamic topology changes such as furniture position changes and personnel movements, adjust the degree of perturbation of the signal environment to obtain the power adjustment value ΔPT required for real-time compensation.

[0124] In this embodiment, by using sensors and scanning devices in step S1 to construct the perception data set GW, the wireless device can obtain the structural semantic information of the propagation environment for the first time, thereby constructing a more accurate extended path loss model than the traditional model in S2. Compared with the existing fixed power configuration scheme, this mechanism has the capabilities of environmental self-awareness and path self-correction, avoiding energy waste and uneven signal coverage caused by blind transmission.

[0125] Through steps S3 to S4, the system compares the communication stability index Γs calculated in real time with the target channel stability threshold Γtar, inversely deduces the current transmission power PT that minimally satisfies the communication quality in the current area, and combines the perturbation fine-tuning mechanism to obtain the actual transmission power PTadj. This set of mechanisms realizes the two-way coupling control between power and stability, that is, instead of configuring redundant power globally, it is "quantified according to demand", effectively controlling the power consumption upper limit.

[0126] In S5, the present invention first introduces a global comprehensive evaluation index TΦ based on unit stability and unit power, and through exponential decay combined with distance factors, forms a comprehensive perception of the signal efficiency distribution. This mechanism enables the system to not only look at local effects, but also judge from the system level whether the power configuration is reasonable and whether the signal distribution is balanced, providing data support for subsequent adjustments and expansions.

[0127] Innovatively introduce the topological perturbation time derivative in S6, which can quickly identify the perturbation trend for sudden spatial topology changes such as furniture movement and personnel walking, and automatically adjust the power adjustment value ΔPT, realizing the closed-loop ability of truly "adjusting as the environment changes". This adaptive mechanism significantly enhances the adaptability and stability of the system to complex scenarios.

[0128] Embodiment 2

[0129] This embodiment is an explanatory description based on Embodiment 1. Please refer to Figure 1, specifically: S1 includes S11 and S12;

[0130] S11, by using a sensor device to scan and perceive the environment, collect information data in the propagation environment, including the wall dielectric constant change rate QD, average reflection angle Er, surface material reflection coefficient Pm, furniture interference factor Hf, topological perturbation data CL, and target area received power Pr, and construct an original data set SW;

[0131] Among them, the wall dielectric constant change rate QD is collected by a millimeter-wave radar;

[0132] The average reflection angle Er is collected by a ToF depth camera;

[0133] The surface material reflection coefficient Pm is collected by a multispectral infrared scanner;

[0134] The furniture interference factor Hf is collected by a high-frequency sonar;

[0135] The topological perturbation data CL is collected by an IMU dynamic attitude sensor;

[0136] The target area received power Pr is collected by a received signal monitor;

[0137] S12, perform data cleaning and normalization processing on the obtained original data set SW to obtain a perception data set GW;

[0138] Data cleaning includes using a statistical outlier detection method to remove outliers from the original data set SW, removing noise data higher than the mean value, and obtaining a data cleaning set CW;

[0139] Normalization processing includes normalizing the data cleaning set CW, uniformly mapping the data to the [0, 1] interval, and obtaining a perception data set GW;

[0140] The perception data set GW is obtained through the following formula:

[0141] ;

[0142] In the formula, GMb represents the b-th data in the perception data set GW, CWb represents the b-th data in the data cleaning set CW, minCWb represents the valley value of the b-th data in the data cleaning set CW, and maxCWb represents the peak value of the b-th data in the data cleaning set CW.

[0143] S2 includes S21 and S22;

[0144] S21. Extract path features from the obtained perception data set GW, including the material structure correction factor QPD, the angular multipath attenuation factor Qer, the environmental interference factor HCf, and the path length exponent correction factor DMp, fit them into the path data set LF, perform model training, and construct a multi-factor extended path loss model;

[0145] The material structure correction factor QPD is obtained through the following formula:

[0146] ;

[0147] The angular multipath attenuation factor Qer is obtained through the following formula:

[0148] ;

[0149] In the formula, cos represents the cosine function, and π represents the pi;

[0150] The environmental interference factor HCf is obtained through the following formula:

[0151] ;

[0152] The path length exponent correction factor DMp is obtained through the following formula:

[0153] ;

[0154] In the formula, CD represents the path loss factor;

[0155] S22. Calculate the total signal attenuation Lt on different propagation paths through the obtained path data set LF and the multi-factor extended path loss model;

[0156] The total signal attenuation Lt is obtained through the following formula:

[0157] ;

[0158] In the formula, Lo represents the free space reference path loss, log represents the logarithmic function, respectively represent the preset weight values of the material structure correction factor QPD, the angular multipath attenuation factor Qer, and the environmental interference factor HCf, and d represents the physical length of the propagation path.

[0159] In this embodiment, in S11, a variety of heterogeneous sensors are introduced for the first time to directionally collect and classify key physical variables in the wireless signal propagation environment, including millimeter-wave radar to obtain the change rate QD of the wall dielectric constant, ToF depth camera to obtain the average reflection angle Er, and infrared scanner to obtain the material reflection coefficient Pm, and the channel environment is modeled from multiple dimensions such as material characteristics, reflection behavior, and structural disturbance. Compared with the traditional model with RSSI or distance as a single variable, this embodiment significantly enhances the system's spatial resolution ability and propagation semantic understanding ability of the real physical environment.

[0160] Through the cleaning and standardization of the original data set SW in step S12, data with different dimensions and distribution characteristics are uniformly mapped to the [0,1] interval. In particular, the statistical outlier removal algorithm is used to eliminate high-noise points and environmental disturbance pseudo-values in the sensed data, and finally a highly reliable sensed data set GW is obtained. This not only improves the consistency and data reliability of the model input, but also lays a high-stability and highly credible basic data support for subsequent path modeling and power adjustment.

[0161] In S21 and S22, this embodiment no longer relies on the traditional path loss model, but introduces a four-dimensional loss factor structure based on the sensed data set GW: material structure correction factor QPD, angular multipath attenuation factor Qer, environmental interference factor Hcf, and path length exponent correction factor DMp. By fitting into the path data set LF and combining the actual path distance ddd, a multi-factor extended path loss model is finally constructed to calculate the total signal attenuation Lt. This enables the system to more accurately predict the signal strength attenuation trend in a non-ideal spatial structure, effectively solving the problems of large estimation deviation and rough path evaluation of traditional models in complex building environments.

[0162] This embodiment clearly quantifies the contribution relationship of each factor by introducing multiple mathematical formulas, and at the same time sets the weight parameters in the loss model, retains the adjustability and learning ability of the model, adapts to the propagation mechanisms in different spaces, structures, and density scenarios, and enhances the system's cross-scenario generalization ability and deployment flexibility.

[0163] By completing the integrated data-to-model construction process from S1 to S2, a complete process from environmental perception → data conversion → path mapping → attenuation prediction is established, providing high-reliability and high-expression model input support for subsequent steps, and solving the problems of large power regulation fluctuations and strong prediction deviations caused by inaccurate model bases in traditional methods.

[0164] Embodiment 3

[0165] This embodiment is an explanatory description based on Embodiment 2. Please refer to Figure 1 and Figure 2 , specifically: S3 includes S31 and S32;

[0166] S31. Combine the obtained total signal attenuation Lt with the received target area power Pr and the topological perturbation data CL at the scene to calculate the communication stability index Γs of the area;

[0167] The communication stability index Γs is obtained through the following formula:

[0168] ;

[0169] In the formula, X represents the perturbation sensitivity coefficient.

[0170] S32. According to the obtained communication stability index Γs, combined with the actual space topological coordinate system, construct the regional signal energy efficiency index CKs;

[0171] The energy efficiency index CKs is obtained through the following formula:

[0172] ;

[0173] In the formula, ck represents a constant;

[0174] Combine the spatial points (x, y, z) in the three-dimensional environment with the energy efficiency index CKs to form an energy efficiency mapping color gradient diagram in the three-dimensional space;

[0175] The gradient diagram is obtained through the following matching method:

[0176] Red area: 0 < energy efficiency index CKs < 0.3, indicating unstable signal and low transmission energy efficiency;

[0177] Yellow area: 0.3 ≤ energy efficiency index CKs ≤ 0.7, indicating signal coverage acceptance;

[0178] Green area: 0.7 < energy efficiency index CKs, indicating stable communication and efficient power use.

[0179] In this embodiment, by using the obtained total signal attenuation Lt, the received target area power Pr, and the topological perturbation data CL, and combining the perturbation sensitivity coefficient X, a calculation mechanism for the communication stability index Γs is established. This approach enables the system to perform partitioned and dynamic quantitative perception of the communication quality in different regions. Compared with the traditional single judgment method relying on RSSI, it is more robust and has higher spatial accuracy, effectively solving the key problems of "inability to determine signal blind spots and unawareness of excessive signal waste".

[0180] In step S32, by coupling the communication stability index Γs with factors such as device power, the regional energy efficiency index CKs is constructed. For the first time, the system realizes a two-layer index linkage evaluation from communication stability to energy efficiency utilization. This mechanism not only measures the "strength" of the signal, but more importantly, measures the "effect generated by unit power", making the wireless connection no longer only focus on whether the signal is "covered", but more on whether it is "covered reasonably and used efficiently", enhancing the overall system perspective of energy management and service quality control.

[0181] By mapping the energy efficiency index CKs to the spatial coordinate system (x, y, z) and visualizing it in the form of a color gradient map (red-yellow-green partition), the system realizes an intuitive presentation of the spatial communication energy efficiency distribution. Compared with the traditional parameters that are limited to background analysis, this method can:

[0182] Enable the system to automatically identify weak signal areas (red areas) and locate blind spots;

[0183] Accurately mark the energy efficiency balance areas (yellow areas) for maintenance;

[0184] Highlight the power excess areas (green areas) for subsequent energy-saving adjustment;

[0185] This not only improves the interactivity and intuitiveness of spatial decision-making, but also significantly enhances the system's engineering visualization and real-time configuration assistance capabilities, especially suitable for large-scale multi-region deployment scenarios such as smart buildings, industrial workshops, and intelligent warehouses.

[0186] By incorporating the topology perturbation data CL into the stability and energy efficiency calculation paths, the system can identify the dynamic changes in the propagation environment caused by furniture movement, personnel activities, device migration, etc., and capture the communication stability fluctuations caused by them in space. This ability constitutes a "perception-determination-mapping" type of spatial dynamic recognition mechanism. Compared with the traditional static model, it can track the changes in connection quality in real time and effectively prevent problems such as sudden disconnection, power redundancy lag, and increased channel interference.

[0187] Example 4

[0188] This example is an explanatory note for Example 3. Please refer to Figure 1 , specifically: S4 includes S41 and S42;

[0189] S41. First, establish the relationship between theoretical received power and transmitted power;

[0190] Recover the received power Pr = PT - Lt in the target area from the channel loss model, substitute it into the acquisition formula of the communication stability index Γs, and obtain the theoretical stability expression:

[0191] ;

[0192] In the formula, Γtheo represents the theoretical communication stability index;

[0193] Backward solve the target power formula: According to the communication stability index Γs, inversely deduce the current transmit power PT required in the current environment;

[0194] The current transmit power PT is obtained through the following formula:

[0195] ;

[0196] In the formula, Γtar represents the preset target channel stability threshold.

[0197] By introducing a perturbation response fine-tuning mechanism and constructing a perturbation response compensation model to adjust the current transmit power PT, the actual transmit power PTadj is obtained;

[0198] Obtain the perturbation change rate ΔCL through the perturbation response compensation model:

[0199] The perturbation change rate ΔCL is obtained through the following formula:

[0200] ;

[0201] In the formula, ∂ represents the partial derivative;

[0202] According to the obtained perturbation change rate ΔCL, define the dynamic perturbation response function ϕ(CL) to correct the transmit power:

[0203] The dynamic perturbation response function ϕ(CL) is obtained through the following formula:

[0204] ;

[0205] In the formula, β represents the perturbation response coefficient;

[0206] The actual transmit power PTadj is obtained through the following formula:

[0207] .

[0208] In this embodiment, in the transmit power control strategy, it no longer depends on empirical settings or fixed values, but is based on the communication stability target. By establishing the mathematical relationship between the received power and the channel attenuation, and further combining the communication stability index for theoretical modeling and target matching inverse deduction. This mechanism realizes the inverse deduction of the minimum transmit power required in the current environmental conditions starting from the "communication quality target", so that the device transmit power can be dynamically adjusted strictly around the actual requirements of the system, significantly improving the scientificity, accuracy and energy efficiency level of the transmit power configuration.

[0209] In traditional wireless systems, when the environmental structure changes rapidly due to factors such as furniture movement and personnel flow, it is often difficult for the device transmission power to respond in a timely manner, resulting in problems such as short-term connection interruption and sudden drop in channel stability. In this embodiment, by introducing a perturbation response compensation model, the change rate of topological perturbation is obtained in real time and converted into a power adjustment factor, so as to realize the instant perception of the perturbation in the burst scenario and the dynamic fine-tuning of the transmission power. This mechanism effectively alleviates the problem of the communication system being "insensitive" to the dynamic environment and improves the steady-state operation ability and anti-interference performance of the system.

[0210] This embodiment designs through two-level control paths; on the one hand, based on the theoretical model, the preliminary power configuration is deduced backwards to ensure that the communication stability reaches the expected level; on the other hand, combined with the actual environmental perturbation change trend, power fine-tuning is carried out through the compensation mechanism to further precisely control the energy output. While ensuring the connection quality, this strategy effectively avoids the energy consumption waste and co-channel interference caused by excessive transmission, and is a power adjustment architecture integrating "demand-driven + perception response".

[0211] In traditional methods, the transmission power control often lacks basis and is not traceable, and it is difficult for system operation and maintenance to judge whether the power configuration is reasonable. In this embodiment, by explicitly modeling the theoretical power calculation logic and the perturbation response relationship, the power adjustment path has a clear and interpretable logical support and can be transparently traced and adjusted according to factors such as time, environment, and communication quality. This not only helps to realize the intelligent control of the system, but also facilitates the visual management and policy update after deployment.

[0212] Embodiment 5

[0213] This embodiment is an explanatory description based on Embodiment 4. Please refer to Figure 3 and Figure 4 , specifically: S5 includes S51 and S52;

[0214] S51. Construct the energy efficiency index Qus of the area through the actual transmission power PTadj and the communication stability index Γs;

[0215] The energy efficiency index Qus is obtained through the following formula:

[0216] ;

[0217] In the formula, Qus(i) represents the energy efficiency index of the i-th area, Γs(i) represents the communication stability index of the i-th area, PTadj(i) represents the actual transmission power of the i-th area, and B represents a real number.

[0218] S52. Integrate the energy efficiency indexes Qus of all areas, conduct a comprehensive efficiency evaluation on the global area, obtain the evaluation index TΦ, and judge the overall signal distribution state;

[0219] The evaluation index TΦ is obtained through the following formula:

[0220] ;

[0221] Where N represents the number of evaluation regions, e represents a constant, k represents the distance attenuation coefficient, and D(i) represents the distance from the i-th region to the signal source;

[0222] The overall signal distribution state is obtained by matching in the following way:

[0223] When 0 < evaluation index TΦ < 0.5, it indicates the existence of signal blind spots and power waste;

[0224] When 0.5 ≤ evaluation index TΦ ≤ 0.8, it indicates that the communication is generally acceptable, but there is power redundancy;

[0225] When 0.8 < evaluation index TΦ < 1.0, it indicates that the overall system configuration is reasonable, the power usage is efficient, and the communication is stable.

[0226] S6. Extract features from the perturbation change rate ΔCL obtained according to the topological perturbation data CL to obtain the perturbation trend factor λp;

[0227] The perturbation trend factor λp is obtained through the following formula:

[0228] ;

[0229] Where tanh represents the hyperbolic tangent function, and λa represents the change sensitivity adjustment coefficient;

[0230] According to the obtained perturbation trend factor λp, combined with the evaluation index TΦ and the communication stability index Γs, calculate the additional power adjustment value ΔPT required, and judge the state of the wireless power;

[0231] The power adjustment value ΔPT is obtained through the following formula:

[0232] ;

[0233] Where represents the power adjustment sensitivity factor; = 0.7;

[0234] Specific example:

[0235] Table 1: Dynamic power adjustment table;

[0236]

[0237] The state of the wireless power is obtained by matching in the following way:

[0238] When 0 < power adjustment value ΔPT < 0.5, it indicates that the power is stable and no additional compensation is required;

[0239] When 0.5 ≤ power adjustment value ΔPT < 1, it indicates that the power is abnormal and power adjustment is required. By combining the obtained power adjustment value ΔPT with the actual transmission power PTadj, the final power output value PTfin is obtained;

[0240] The final power output value PTfin is obtained by adding the actual transmission power PTadj and the power adjustment value ΔPT.

[0241] In this embodiment, through the combined relationship between the actual transmission power and the communication stability index, an energy efficiency index at the regional level is constructed to quantify the communication quality return brought by unit power in each spatial region. This mechanism realizes the evaluation method in the dimension of "output power cost performance", breaks through the evaluation limitations of traditional wireless systems that only rely on static indicators such as signal strength or reception success rate, provides a more scientific and comparable optimization basis for power adjustment, and enables the energy efficiency control to have a measurable and traceable foundation.

[0242] In S52, through the fusion processing of all regional energy efficiency data, a system-level comprehensive efficiency index is formed to judge the rationality of the overall signal distribution and the effectiveness of power allocation. Compared with the traditional local optimization method only, this embodiment realizes the transformation from "single-point optimization" to "global optimization", enables the system to comprehensively grasp whether there are problems such as blind spots, signal waste or power overload in the current configuration, has the system-level efficiency intelligent judgment ability, and improves the balance and stability of the overall network deployment.

[0243] Through the multi-factor coupling analysis of the evaluation index, communication stability index and disturbance trend factor, this embodiment designs a dynamic calculation and response mechanism for the power adjustment value, which can judge whether compensation adjustment is required according to the current system state, combine the adjustment value with the current transmission power, and calculate the final output power. This closed-loop mechanism ensures that the system has the ability to adjust on demand, respond dynamically and correct quickly, and truly realizes the qualitative change from the traditional static power strategy to the intelligent, self-adjusting and self-sensing direction.

[0244] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirits of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An adaptive wireless connection optimization method based on intelligent environment perception, characterized in that: The following steps are involved: S1. Use sensors and scanning equipment to collect information in the propagation environment, fit it into the original data set SW, and perform preprocessing to obtain the perception data set GW; S2. By using the sensing data set GW, an extended path loss model is constructed and the total signal attenuation Lt on different propagation paths is calculated; S3, combining the total signal attenuation Lt obtained with the target area receiving power Pr and the topology disturbance data CL received on site, evaluating the communication stability index Γs of each area in the space, and drawing a space energy efficiency diagram; S4. By combining the acquired communication stability index Γs with the preset target channel stability threshold Γtar, the current transmission power PT required in the current environment is reversed, and the actual transmission power PTadj is calculated; S5. Perform a comprehensive performance evaluation on the global region based on the actual transmission power PTadj and the communication stability index Γs, and obtain an evaluation index TΦ; S6. According to the obtained evaluation index TΦ, topology disturbance data CL and communication stability index Γs, after detecting the dynamic topological changes of furniture position changes and personnel movement, the disturbance degree of the signal environment is adjusted to obtain the power adjustment value ΔPT required for real-time compensation.

2. The method for optimizing the intelligent environment-aware adaptive wireless connection according to claim 1, characterized in that: S1 includes S11 and S12; S11, by using sensor equipment to scan and perceive the environment, collect information data in the propagation environment, including the wall dielectric constant change rate QD, the average reflection angle Er, the surface material reflection coefficient Pm, the furniture interference factor Hf, the topological disturbance data CL and the target area receiving power Pr, and construct the original data set SW; Among them, the wall dielectric constant change rate QD is obtained by millimeter wave radar collection; The average reflection angle Er is acquired through the ToF depth camera; The surface material reflectance coefficient Pm is acquired through a multi-spectral infrared scanner; The furniture interference factor Hf is obtained through high-frequency sonar collection; The topological disturbance data CL is acquired through the IMU dynamic attitude sensor; The target area receiving power Pr is acquired through the receiving signal monitor; S12, performing data cleaning and normalization processing on the acquired original data set SW to obtain the perception data set GW; Data cleaning includes using statistical outlier detection methods to remove outliers from the original data set SW, remove noise data above the mean, and obtain the data cleaning set CW; The normalization process includes normalizing the data cleaning set CW, uniformly mapping the data to the interval [0, 1], and obtaining the perception data set GW; The perception data set GW is obtained by the following formula: ； Wherein, GMb represents the b-th data in the perception data set GW, CWb represents the b-th data in the data cleaning set CW, minCWb represents the valley value of the b-th data in the data cleaning set CW, and maxCWb represents the peak value of the b-th data in the data cleaning set CW.

3. The method for optimizing the intelligent environment-aware adaptive wireless connection according to claim 2, characterized in that: S2 includes S21 and S22; S21. Extract path features from the acquired perception data set GW, including material structure correction factor QPD, angular multipath attenuation factor Qer, environmental interference factor HCf, and path length index correction factor DMp, and fit them into the path data set LF, perform model training, and construct a multi-factor extended path loss model; The material structure correction factor QPD is obtained by the following formula: ； The angular multipath attenuation factor Qer is obtained by the following formula: ； In the formula, cos represents the cosine function, and π represents pi; The environmental interference factor HCf is obtained by the following formula: ； The path length index correction factor DMp is obtained by the following formula: ； Where CD represents the path loss factor; S22, calculating the total signal attenuation Lt on different propagation paths by using the acquired path data set LF and the multi-factor extended path loss model; The total signal attenuation Lt is obtained by the following formula: ； Where Lo represents the free space reference path loss, log represents the logarithmic function, They respectively represent the preset weight values ​​of the material structure correction factor QPD, the angular multipath attenuation factor Qer and the environmental interference factor HCf, and d represents the physical length of the propagation path.

4. The method for optimizing the intelligent environment-aware adaptive wireless connection according to claim 2, characterized in that: S3 includes S31 and S32; S31, combining the acquired total signal attenuation Lt with the target area receiving power Pr and topology disturbance data CL received on site, and calculating the communication stability index Γs of the area; The communication stability index Γs is obtained by the following formula: ； Where X represents the disturbance sensitivity coefficient.

5. The method for optimizing the intelligent environment-aware adaptive wireless connection according to claim 4, characterized in that: S32, constructing a regional signal energy efficiency index CKs based on the acquired communication stability index Γs and the actual space topology coordinate system; The energy efficiency index CKs is obtained by the following formula: ； In the formula, ck represents a constant; Combine the spatial points (x, y, z) in the three-dimensional environment with the energy efficiency indicators CKs to form an energy efficiency mapping color gradient map in the three-dimensional space; The gradient map is obtained by matching in the following way: Red area: 0<energy efficiency index CKs<0.3, indicating unstable signal and low transmission efficiency; Yellow area: 0.3≤energy efficiency index CKs≤0.7, indicating signal coverage acceptance; Green area: 0.7 < energy efficiency index CKs, indicating stable communication and efficient power use.

6. The method for intelligent environment-aware adaptive wireless connection optimization according to claim 5, characterized in that: S4 includes S41 and S42; S41, first establish a theoretical receiving power-transmitting power relationship; Recover the target area received power Pr=PT-Lt from the channel loss model, substitute it into the formula for obtaining the communication stability index Γs, and obtain the theoretical stability expression: ； Where Γtheo represents the theoretical communication stability index; Reversely solve the target power formula: According to the communication stability index Γs, reverse the current transmission power PT required in the current environment; The current transmit power PT is obtained by the following formula: ； Where Γtar represents the preset target channel stability threshold.

7. The method for optimizing the intelligent environment-aware adaptive wireless connection according to claim 6, characterized in that: By introducing a disturbance response fine-tuning mechanism and building a disturbance response compensation model, the current transmit power PT is adjusted to obtain the actual transmit power PTadj; The disturbance change rate ΔCL is obtained through the disturbance response compensation model: The disturbance change rate ΔCL is obtained by the following formula: ； In the formula, ∂ represents the partial derivative; According to the obtained disturbance change rate ΔCL, the dynamic disturbance response function φ(CL) is defined to correct the transmission power: The dynamic disturbance response function φ(CL) is obtained by the following formula: ； Where, β represents the disturbance response coefficient; The actual transmit power PTadj is obtained by the following formula: 。 8. The method for intelligent environment-aware adaptive wireless connection optimization according to claim 7, characterized in that: S5 includes S51 and S52; S51, constructing the regional energy efficiency index Qus through the actual transmission power PTadj and the communication stability index Γs; The energy efficiency index Qus is obtained by the following formula: ； Where Qus(i) represents the energy efficiency index of the i-th region, Γs(i) represents the communication stability index of the i-th region, PTadj(i) represents the actual transmission power of the i-th region, and B represents a real number.

9. The method for optimizing the intelligent environment-aware adaptive wireless connection according to claim 8, characterized in that: S52, integrating the energy efficiency index Qus of all regions, performing a comprehensive performance evaluation on the global region, obtaining the evaluation index TΦ, and judging the overall signal distribution state; The evaluation index TΦ is obtained by the following formula: ； Where N represents the number of evaluation areas, e represents a constant, k represents the distance attenuation coefficient, and D(i) represents the distance from the i-th area to the signal source; The overall signal distribution status is obtained by matching: When 0<evaluation index TΦ<0.5, it means there is a signal blind area and power waste; When 0.5≤evaluation index TΦ≤0.8, it means that the communication is generally acceptable, but there is power redundancy; When 0.8<evaluation index TΦ<1.0, it means that the overall configuration of the system is reasonable, the power usage is efficient, and the communication is stable.

10. The method for optimizing the intelligent environment-aware adaptive wireless connection according to claim 9, characterized in that: S6, performing feature extraction on the disturbance change rate ΔCL obtained according to the topological disturbance data CL to obtain a disturbance trend factor λp; The disturbance trend factor λp is obtained by the following formula: ； In the formula, tanh represents the hyperbolic tangent function, λa represents the change sensitivity adjustment coefficient; According to the obtained disturbance trend factor λp, combined with the evaluation index TΦ and the communication stability index Γs, the power adjustment value ΔPT to be added is calculated, and the state of the wireless power is judged; The power adjustment value ΔPT is obtained by the following formula: ； In the formula, represents the power adjustment sensitivity factor; The wireless power status is obtained by matching: When 0<power adjustment value ΔPT<0.5, it means that the power is stable and no additional compensation is required; When 0.5≤power adjustment value ΔPT<1, it indicates power abnormality and power adjustment is required. The final power output value PTfin is obtained by combining the obtained power adjustment value ΔPT with the actual transmit power PTadj; The final power output value PTfin is obtained by adding the actual transmission power PTadj and the power adjustment value ΔPT.

Images