A Traffic Signal Control Method Based on Roadside Edge

Abstract

This invention discloses a traffic signal control method based on the edge of a roadside, comprising: acquiring the traffic flow of each intersection at the current moment; calculating the proportion of the traffic flow of each intersection to the total traffic flow of the intersection, obtaining first data; processing the first data through a Logistic mapping to predict second data; calculating the Jacobian matrix based on the second data, and obtaining the Lyapunov exponent based on the Jacobian matrix; if the Lyapunov exponent is positive, adjusting the noise suppression intensity based on the Lyapunov exponent to obtain a noise suppression factor; processing the first data based on the noise suppression factor to obtain denoised first data; processing the denoised first data based on chaos theory to dynamically adjust the signal cycle and green light time, and then calculating pollutant emissions; and adjusting the signal cycle and green light time again based on the target pollutant emissions. This invention improves road traffic efficiency and vehicle throughput, and reduces parking time and pollutant emissions.

Description

Technical Field

[0001] This invention relates to the field of electronic digital data processing. Specifically, it relates to a traffic signal control method based on the edge of a roadside. Background Technology

[0002] In recent years, with the rapid development of intelligent transportation technology, unified traffic signal control at the intelligent edge roadside has gradually come into the public eye. This technology, based on advanced technologies such as the Internet of Things, artificial intelligence, and big data analytics, combines traffic signal control systems with roadside equipment to achieve real-time data collection, information exchange, and intelligent decision-making.

[0003] However, traditional signal control methods are usually based on linear system models, while traffic systems are often nonlinear and complex. Real-time changes in traffic flow require signal control systems to adjust in a timely manner to adapt to different traffic conditions. Traditional signal control is usually based on fixed timing or reference tables and cannot be flexibly adjusted according to actual traffic flow. This may lead to unnecessary stops, idling, and emissions, which limits the efficiency and smoothness of road traffic.

[0004] Therefore, there is an urgent need for a signal control method to improve traffic efficiency. Summary of the Invention

[0005] This invention is proposed based on the aforementioned needs of the prior art. The technical problem to be solved by this invention is to provide a traffic signal control method based on the edge roadside to achieve dynamic adjustment of signal timing and duration, thereby improving road traffic efficiency and vehicle throughput, and reducing parking time and pollutant emissions.

[0006] To solve the above problems, the present invention is implemented using the following technical solution:

[0007] A traffic signal control method based on the edge roadside is provided. The method includes: acquiring the traffic flow at each intersection at the current time; calculating the proportion of traffic flow at each intersection to the total traffic flow at the intersection, obtaining first data; processing the first data using a Logistic mapping to predict the proportion of traffic flow at each intersection to the total traffic flow at multiple times, as second data; calculating a Jacobian matrix based on the second data, and obtaining the Lyapunov exponent based on the Jacobian matrix; determining whether the Lyapunov exponent is positive; if it is positive, adjusting the noise suppression intensity based on the Lyapunov exponent to obtain a noise suppression factor, the expression of which is: Where μ(t) represents the noise suppression factor, λ′ represents the adjustment factor, and λ(t) represents the Lyapunov exponent; the first data is processed based on the noise suppression factor to obtain the denoised first data, the expression of which is: Where y(t) represents the first data after denoising, x t Indicates the first data. The first signal cycle and the first green light time are dynamically adjusted based on the first denoised data. The first signal cycle and the first green light time are obtained. The pollutant emission is calculated based on the first signal cycle and the first green light time. If the pollutant emission exceeds the preset value, the signal cycle and the green light time of the intersection are adjusted again based on the target pollutant emission of the intersection to obtain the second signal cycle and the second green light time.

[0008] Optionally, it also includes: reconstructing the Jacobian matrix based on Takens' theorem to obtain a reconstructed matrix; suppressing the noise trajectory of the reconstructed matrix to obtain a suppressed Jacobian matrix; and calculating the Williams entropy based on the suppressed Jacobian matrix.

[0009] Optionally, the first data is processed using a Logistic mapping to predict the proportion of traffic flow at each intersection to the total traffic flow at the intersection at multiple times, which is then used as the second data, expressed as: x t+1 =r·x t ·(1-x t ), where x t+1 This represents the second data point at time t+1, where r is a parameter reflecting the influence of external factors, and x... t This indicates the first data point.

[0010] Optionally, the Lyapunov exponent is obtained from the Jacobian matrix, and its expression is: Where λ(t) represents the Lyapunov exponent, n represents the number of the second data points, and f′(x) t+i ) represents the eigenvalue of the Jacobian matrix at time t+1.

[0011] Optionally, dynamically adjusting the signal period and green light duration based on the denoised first data to obtain the first signal period and first green light duration includes: calculating the original signal period based on the denoised first data, the expression of which is: Where C represents the original signal period, L represents the intersection loss time, and y j Let C represent the first denoised data for lane j, where j represents the lane identifier and m represents the number of lanes at the intersection. The original signal period is dynamically adjusted based on the Lyapunov exponent to obtain the adjusted signal period, expressed as: C new =C·(1+α·λ(t)), where, C new The adjusted signal period is represented by α, and the sensitivity coefficient is represented by α. The green light time for the corresponding lane is calculated based on the original signal period and the ratio of the denoised first data of each lane to the total denoised first data of all lanes. The expression for this is: Among them, G i y represents the green light time for lane i. i (t) represents the first denoised data for lane i; based on the current green light time of the current lane, the traffic flow at the current moment, and the traffic flow at the next moment, the updated green light time of the current lane is obtained, and its expression is: in, β represents the green light time after lane i is updated, and β represents the adjustment coefficient. Q represents the predicted traffic flow at the next moment. t This indicates the traffic flow at the current moment.

[0012] Optionally, noise trajectories are suppressed on the reconstructed matrix to obtain the suppressed Jacobian matrix, which is expressed as: X i (t)=X(t+(i-1)τ), Where X′(t) represents the suppressed Jacobian matrix, φ(t) represents the time evolution factor, X(t) represents the reconstruction matrix, and C i X is the weighting coefficient. i X(t) represents the i-th dimension of the phase space generated by the reconstruction matrix X(t) during the reconstruction process, τ represents the delay time, and X(t+(i-1)τ) represents the signal component with a delay of (i-1)τ.

[0013] Optionally, it also includes: obtaining the lane density, vehicle speed, and vehicle type of each lane based on the adjusted signal cycle and green light duration; and calculating the pollutant emissions based on the lane density, vehicle speed, and vehicle type of each lane, expressed as: Where Q represents pollutant emissions, L′ represents road length, k(x, t) represents vehicle density at time t, v represents vehicle speed, T represents vehicle type, and E represents emission factor.

[0014] Optionally, based on the overall target emissions at the intersection, the signal cycle and green light time at the intersection may be adjusted again, including: Among them, C′ new Indicates the second signal period, C new This represents the first signal period, α represents the sensitivity coefficient, and Q represents the pollutant emission rate. target Indicates the target pollutant emissions. This indicates the second green light time for lane i. Indicates the first green light time for lane i, k i k represents the vehicle density in lane i. j Let m represent the vehicle density of lane j, and m represent the number of lanes at the intersection.

[0015] Compared with existing technologies, this invention dynamically adjusts signal timing and duration based on real-time data, thereby improving road traffic efficiency and vehicle throughput, reducing traffic congestion, shortening travel time, and providing a more convenient and comfortable traffic environment for road users. At the same time, it optimizes vehicle flow by adjusting the signal cycle and green light duration at intersections based on calculated pollutant emissions, reducing vehicle stopping and idling time in traffic congestion, and lowering emissions. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of a traffic signal control method based on the edge roadside provided in this embodiment. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] To facilitate understanding of the embodiments of the present invention, further explanations and descriptions will be provided below with reference to the accompanying drawings and specific embodiments. These embodiments do not constitute a limitation on the scope of protection of the present invention.

[0020] This embodiment provides a traffic signal control method based on the edge of the roadside, the process of which is as follows: Figure 1 As shown, it includes:

[0021] S1 obtains the traffic flow at each intersection at the current time.

[0022] Real-time traffic flow data at intersections and on roads is collected using sensors, monitoring equipment, and intelligent traffic cameras.

[0023] S2 calculates the proportion of traffic flow at each intersection to the total traffic flow at the intersection, thus obtaining the first data.

[0024] S3 processes the first data using a Logistic mapping to predict the proportion of traffic flow at each intersection to the total traffic flow at the intersection at multiple times, which is then used as the second data.

[0025] The Logistic mapping describes the chaotic behavior in a one-dimensional dynamical system. Overall, it simulates the changes in traffic flow at different time steps, reflecting the variation of traffic flow under different external influences. Its expression is as follows:

[0026] x t+1 =r·x t ·(1-x t )

[0027] Where, x t+1 This represents the second data point at time t+1, where r is a parameter reflecting the influence of external factors, such as road conditions and vehicle density, which determines the dynamic behavior of the system. t This represents the proportion of traffic flow at time t to the total traffic flow at the intersection, i.e., the first data point.

[0028] Similarly, using Logistic mapping based on x t+1 Get x t+2 Through multiple iterative calculations, the second data x at n time points is obtained. t+1 x t+2 , ..., x t+i , ..., x t+n , where x t+n This represents the second data point at time t+n.

[0029] In this embodiment, the estimated parameter r is calibrated using the historical proportion of traffic flow at each intersection to the total traffic flow at the intersection, and the current proportion of traffic flow at each intersection to the total traffic flow at the intersection is used as the initial condition for the Logistic prediction system to simulate the dynamic changes of traffic flow under different external conditions.

[0030] S4 calculates the Jacobian matrix based on the second data, and obtains the Lyapunov index based on the Jacobian matrix.

[0031] The Lyapunov exponent is used to evaluate the sensitivity of a system to initial conditions, measuring the sensitivity of the system's trajectory in phase space. The specific process is as follows:

[0032] For each second data x t+i Calculate the Jacobian matrix f′(x) t+iFor each Jacobian matrix, its eigenvalues, i.e., local eigenvalues, are calculated. The logarithm of each eigenvalue is taken, and the average of these eigenvalues ​​is calculated to obtain the Lyapunov exponent. As the number of iterations n approaches infinity, the Lyapunov exponent approaches a stable limit. The sign of the Lyapunov exponent is determined: a positive Lyapunov exponent indicates sensitivity to initial conditions, meaning that small initial differences can lead to exponential growth of the system trajectory, resulting in chaotic behavior. A negative Lyapunov exponent indicates insensitivity to initial conditions, and the system trajectory tends towards stability, i.e., it tends towards convergence or periodic behavior.

[0033] The expression for calculating the Lyapunov exponent is:

[0034]

[0035] Where λ represents the Lyapunov exponent, f′(x t+i ) represents the eigenvalue of the Jacobian matrix at time t+1.

[0036] S5 determines whether the Lyapunov exponent is positive. If it is positive, the noise suppression strength is adjusted based on the Lyapunov exponent to obtain the noise suppression factor.

[0037] The noise suppression strength is dynamically adjusted based on the Lyapunov exponent. A larger Lyapunov exponent increases the noise suppression strength; conversely, a smaller exponent reduces noise interference. Specifically, the noise suppression factor is adjusted using the following formula:

[0038]

[0039] Here, μ(t) represents the noise suppression factor, and λ′ represents the adjustment factor. As the Lyapunov exponent increases, the strength of noise suppression increases, thereby reducing the system's sensitivity to noise.

[0040] S6 processes the first data based on the noise suppression factor to obtain the denoised first data.

[0041] μ(t) is used as a weighting factor to balance the original data and known noise. The expression for this step is:

[0042]

[0043] Where y(t) represents the first denoised data, and μ(t) represents the weighting factor. This represents the first historical denoised data.

[0044] If the Lyapunov exponent is large, μ(t) will increase, thus relying more on noise suppression. If the Lyapunov exponent is small, the strength of noise suppression will decrease, relying more on the original data.

[0045] The first data is smoothed and denoised, and then the denoised first data is used as the input to the chaotic traffic signal control system.

[0046] S7 dynamically adjusts the signal period and green light time based on the first data processing for noise reduction, thus obtaining the first signal period and the first green light time.

[0047] After predicting traffic flow for a future time period, the original signal period is calculated based on the denoised first data, and its expression is:

[0048]

[0049] Where C represents the original signal period, L represents the intersection loss time, including the time for vehicle start-up and stop, in seconds, yj represents the first denoised data for lane j, j represents the lane identifier, and m represents the number of lanes at the intersection.

[0050] The Lyapunov exponent is used to assess the chaos level of the traffic system in real time, and then the signal cycle is dynamically adjusted accordingly. When the Lyapunov exponent is highly sensitive to initial conditions, the cycle length is increased to improve stability; when the Lyapunov exponent is low, the cycle is decreased to improve traffic efficiency. The signal cycle update formula is as follows:

[0051] C new =C·(1+α·λ(t))

[0052] Among them, C new This represents the first signal period, and α represents the sensitivity coefficient, which is used to control the degree of response to the Lyapunov exponent. It can be set according to different scenarios.

[0053] The green light time for the corresponding lane is calculated based on the ratio of the original signal period and the denoised first data of each lane to the total denoised first data of all lanes. The expression is as follows:

[0054]

[0055] Among them, G i y represents the original green light time for lane i. i (t) represents the first denoised data for lane i;

[0056] To avoid phase conflicts caused by prolonged green lights, a minimum and maximum green light duration range is set: G min ≤G i ≤G max Among them, Gmin G represents the minimum time for each green light cycle. max This indicates the maximum duration of each green light cycle. The green and red light phases are adjusted by comparing current and predicted traffic flow.

[0057] Furthermore, by combining traffic flow prediction and chaos sensitivity, the green light time for the next cycle can be set. Specifically, based on the current lane's green light time, the current traffic flow, and the next traffic flow, the updated green light time for the current lane is obtained, expressed as:

[0058]

[0059] in, This represents the first green light time for lane i, and β represents the adjustment coefficient, which is used to control the impact of predicted traffic flow on the green light duration. The specific value can be set according to the traffic scenario. Q represents the predicted traffic flow at the next moment. t This indicates the traffic flow at the current moment.

[0060] In summary, by combining chaos theory, and based on predicting future traffic flow, the signal cycle and green light duration can be dynamically adjusted to achieve efficient traffic signal control.

[0061] S8 calculates pollutant emissions based on the first signal cycle and the first green light time.

[0062] Specifically: based on the adjusted signal cycle and green light duration, obtain the lane density, vehicle speed, and vehicle type for each lane; calculate the pollutant emissions based on the lane density, vehicle speed, and vehicle type for each lane.

[0063] Vehicles produce different emissions depending on their type and current speed, and the emission factor varies with time and vehicle speed. The emission factor E represents the mass of pollutants emitted by a unit vehicle per unit time, reflecting the emissions to traffic flow on a road segment. The emission factor is related to vehicle speed v and vehicle type T. Let the relationship between the emission factor and vehicle speed be expressed as:

[0064]

[0065] E0 is the basic emission factor, and α is an index related to vehicle speed.

[0066] The pollutant emission Q is obtained by integrating the vehicle density k and the emission factor E, and its expression is as follows:

[0067]

[0068] Where Q represents pollutant emissions, L′ represents road length, k(x, t) represents vehicle density at time t, v represents vehicle speed, T represents vehicle type, and E represents emission factor.

[0069] The above expressions can be used to obtain pollutant emissions under different traffic flow and control strategies, thereby measuring the emission indicators of traffic flow models and estimating pollutant emissions more accurately. Based on emission factors, unified monitoring and control of traffic flow and corresponding roadside pollutant emissions from traffic systems can be implemented, vigorously promoting sustainability and environmental protection, and making traffic signal systems more environmentally friendly.

[0070] Furthermore, after completing the Logistic mapping and Lyapunov index calculations, the evolution results of the Logistic mapping simulation are evaluated to verify the rationality of the evolution method. Specifically, actual traffic flow data is used for verification, and the predicted traffic flow data is adjusted based on the verification results. Assume there is a set of actual traffic flow data {X}. t+1 X t+2 , ..., X t+n}, by minimizing the predicted traffic flow data and actual traffic flow data X obtained from the Logistic mapping. N+1 The error between them.

[0071] This embodiment uses the least squares method to fit the parameter r, that is, to find the parameter value that minimizes the sum of squared errors. The loss function L(r) of the least squares method is:

[0072]

[0073] Where, x t+i Let X be the second data point corresponding to time t+i. t+i This is actual observation data.

[0074] After constructing the loss function, the optimal parameters are obtained by solving the following optimization problem:

[0075]

[0076] in, This represents the value of variable r that makes the L(r) function reach its minimum.

[0077] If the pollutant emissions exceed the preset value, S9 will readjust the signal cycle and green light time of the intersection based on the target pollutant emissions at the intersection to obtain a second signal cycle and a second green light time.

[0078]

[0079]

[0080] Among them, C′ new Indicates the second signal period, C new This represents the first signal period, α represents the sensitivity coefficient used to adjust the period adjustment speed, and Q represents the pollutant emission rate. target Indicates the target pollutant emissions. This indicates the second green light time for lane i. Indicates the first green light time for lane i, k i k represents the vehicle density in lane i. j Let m represent the vehicle density of lane j, and m represent the number of lanes at the intersection.

[0081] By adjusting the signal cycle and green light duration at intersections, vehicle flow is optimized, speed control schemes are updated, and traffic flow dynamics are dynamically allocated, reducing idling time and emissions. Idling time refers to the time the engine runs in neutral; although the vehicle is not moving, the engine still consumes fuel.

[0082] Furthermore, Williams entropy is used to evaluate the complexity and predictability of the Logistic mapping model, and the model is adjusted and iterated accordingly.

[0083] Before calculating the Williams entropy, if the Lyapunov exponent is positive, the Jacobian matrix is ​​adjusted, including:

[0084] Step 1: Reconstruct the Jacobian matrix based on Takens' theorem to obtain the reconstructed matrix.

[0085] Phase space reconstruction based on Takens' theorem maps input data to a higher dimension. This higher-dimensional structure and dynamic behavior are used for noise filtering and data correction, compensating for spurious trajectories or instabilities ignored by methods like weighted averaging that directly control noise. Specifically, using Takens' theorem, a phase space reconstruction is constructed from a time series of traffic flow data. By selecting appropriate delay times and embedding dimensions, the reconstructed phase space reveals the underlying nonlinear dynamic characteristics of the data. This process helps to better capture the intrinsic structure of chaotic systems and reduce noise interference with the original data.

[0086] Step 2: Suppress the noise trajectory of the reconstructed matrix to obtain the suppressed Jacobian matrix.

[0087] By introducing a synchronization trajectory suppression method, synchronization trajectories can be identified and filtered in high-dimensional phase space, especially for spurious synchronization phenomena generated against a noisy background. The synchronization trajectory refers to a trajectory exhibiting similar evolutionary paths in a multi-dimensional state space. If a synchronization trajectory appears, it is considered a "noise pattern" and therefore needs to be suppressed. The following formula is used to suppress noisy trajectories:

[0088] X i (t)=X(t+(i-1)τ)

[0089]

[0090] Among them, X i (t) represents the i-th dimension in the phase space generated during the reconstruction of the original time series X(t), where each X... i (t) corresponds to the signal components with different time delays. X1(t) corresponds to the signal component X(t) in the original time series, X2(t) corresponds to the signal component X(t+τ) delayed by τ, and so on, until X i (t) corresponds to the signal component X(t+(i-1)τ) delayed by (i-1)τ. The above method is used to reconstruct a multidimensional dynamic trajectory reflecting traffic flow changes in phase space; τ represents the delay time, used to determine how to decompose the time series data into multidimensional vectors; X′(t) is the suppressed Jacobian matrix; X(t) represents the reconstruction matrix; C i φ(t) is the weighting coefficient, and φ(t) is the time evolution factor. By adjusting φ(t), the influence of noise in the synchronization trajectory is reduced, thereby extracting a more stable signal.

[0091] The Williams entropy is calculated based on the suppressed Jacobian matrix to assess the model's complexity. The calculation formula is as follows:

[0092]

[0093] Where K represents the Williams entropy, p(x) t ) indicates the first data x t The probability density function at point f′(x) t ) represents the suppressed Jacobian matrix.

[0094] In traffic flow forecasting, Williams entropy is used to evaluate a model's ability to describe traffic flow dynamics. Specifically, lower Williams entropy indicates a more regular and predictable system, with a relatively simple and stable traffic flow evolution pattern. Higher Williams entropy indicates higher complexity and uncertainty in the system, meaning the traffic flow evolution pattern is diverse and difficult to predict. Lower Williams entropy values ​​may suggest that the model is better suited to capturing periodic changes, while higher values ​​may indicate that the model is better able to cope with the impact of unforeseen events. By comprehensively considering the model's Williams entropy value and other indicators, and iterating the model's parameters accordingly, while introducing nonlinear terms to increase the system's complexity and nonlinear characteristics, a comprehensive simulation of traffic flow evolution is achieved, and the model's performance and effectiveness are comprehensively evaluated. By comparing with actual data, it is determined that the model can predict the overall trend of traffic flow relatively well, and in some cases, it can capture the periodic changes and the impact of unforeseen events, thus realizing the mapping construction of a traffic flow model under chaos theory.

[0095] Furthermore, edge devices are used for local data processing, intelligent decision-making, distributed data storage, edge caching, and load balancing scheduling to reduce communication latency, improve the system's real-time performance, intelligence, and security, optimize traffic flow, reduce congestion, and improve road traffic efficiency.

[0096] Local data processing: Edge devices collect real-time data such as traffic flow, vehicle speed, and pedestrian numbers by installing sensors or cameras. Assume data processing takes time T. p Data transmission to the central server takes time T. t Then, through local data processing, the total latency can be reduced to T. p .

[0097] Intelligent decision-making: Edge devices can formulate intelligent traffic signal control strategies in real time based on locally processed data and preset algorithms. Assume the algorithm execution time is T. a Then, through intelligent decision-making, the total delay can be reduced to T. a .

[0098] Distributed data storage: Deploying a distributed data storage system on edge devices allows multiple edge devices to share data and make collaborative decisions through edge computing, enabling data to be stored and accessed closer to the data source. Assume the data access time is T. d Data transmission to the central server takes time T. t Then, through distributed data storage, the total latency can be reduced to T. d .

[0099] Edge caching technology: Edge devices can use caching technology to cache frequently used data or processing results locally, building an edge intelligent sensing network that enables traffic signal equipment to communicate and exchange information with surrounding vehicles, pedestrians, and other traffic participants in real time. Assume the time it takes to retrieve data from the cache is T. c Data transmission to the central server takes time T. t Then, by using edge caching technology, the total latency can be reduced to T. c .

[0100] Load balancing and proximity scheduling: Edge computing systems can distribute scheduling control commands to the nearest edge devices for processing based on the load status of edge devices and the location of traffic data sources. Assuming proximity scheduling can assign tasks to the nearest edge device, the total latency can be significantly reduced. Total latency = T p +T a +T d +T c .

[0101] Furthermore, local data encryption and data integrity verification are performed based on edge devices.

[0102] Edge devices use local data encryption technology to encrypt sensitive traffic data, ensuring that even if the data is intercepted during transmission, it cannot be stolen or tampered with. Local data encryption typically employs symmetric or asymmetric key encryption algorithms, ensuring that only authorized devices can decrypt the data. A symmetric key is generated on the edge device for encrypting and decrypting traffic data. When traffic data is ready to be transmitted, the edge device uses the generated key to encrypt the data. The encrypted data is transmitted over the network to the receiving end. Upon receiving the encrypted data, the receiving end uses the same key to decrypt it, obtaining the original traffic data. Because the key is stored only on the edge device, and only authorized users have access to the device, unauthorized access cannot decrypt the encrypted data, ensuring data security.

[0103] Edge devices implement a data integrity verification mechanism to ensure that traffic data is not tampered with or damaged during transmission. Specifically, when preparing to transmit data, the edge device uses a hash function to calculate a digest of the data and transmits the digest along with the data to the receiving end. The receiving end recalculates the digest of the received data based on the same hash function and compares it with the transmitted digest to verify the data integrity.

[0104] The hash function calculation and judgment process is as follows:

[0105] Hash functions are used to verify data integrity, ensuring that data has not been tampered with or corrupted during transmission. In this embodiment, the hash function is SHA-256, which converts an input of arbitrary length into a 256-bit hash value.

[0106] The process of calculating the hash value of data can be represented as follows:

[0107] H(D) = H(d1, d2, ..., d) n )

[0108] Where d1, d2, ..., d n These are the various parts of data D, which may be bytes, bits, or other data units.

[0109] Use the SHA-256 hash function to calculate the digest of the data:

[0110] H(D) = SHA-256(D)

[0111] SHA-256 takes the data D as input and generates a 256-bit hash value as output.

[0112] The data D and its corresponding hash value H(D) are transmitted together to the receiving end, and the transmitted data is D||H(D), where || denotes a join operation. After receiving the transmitted data, the receiving end recalculates the hash value H(D′) of the data and compares it with the transmitted hash value H(D). If they match, it means that the data integrity has not been compromised; if they do not match, it means that the data may have been tampered with.

[0113] Furthermore, through the comprehensive implementation of edge intelligent sensing networks and the comprehensive optimization of roadside edge control systems, their efficiency and performance are improved, better supporting the intelligence and real-time operation of traffic signal control systems. Edge intelligent sensing networks include local data processing and summary transmission, as well as edge intelligent device distribution strategies.

[0114] Local data processing refers to using aggregation algorithms on edge devices to summarize and process locally sensed data, generating summary information, and then transmitting this summary information to the central server. This effectively reduces communication overhead and the risk of privacy leaks. This approach preserves the privacy of the sensed data while providing sufficient information for the central server to formulate traffic control strategies, thereby optimizing the performance of the edge intelligent sensing network. This reduces communication overhead and potential privacy leaks.

[0115] The edge intelligent device distribution strategy involves dynamically distributing data on edge devices. During traffic congestion or peak hours, priority is given to transmitting critical data related to traffic flow and conditions to ensure the traffic control system can promptly capture and respond to real-time traffic conditions. During periods of smooth traffic flow, data related to key indicators such as traffic safety is prioritized to optimize the performance and efficiency of the traffic control system. Based on the importance of the data and real-time changes in traffic conditions, the priority and method of data transmission to the central server are dynamically adjusted to maximize the real-time performance and accuracy of the perceived data, reduce system communication overhead, and optimize the performance of the edge intelligent sensing network.

[0116] This embodiment proposes a traffic signal control method based on the edge roadside, which can better capture the complex dynamics of the traffic system, achieve real-time signal adjustments more quickly and flexibly, accurately optimize the intersection signal control system under multiple scenarios, and enhance the system's stability in the face of external interference such as accidents and emergencies, making signal control more robust. Introducing edge computing reduces the latency of data transmission to the central server, and real-time performance allows traffic signals to adapt more quickly to changes in traffic conditions, improving signal control efficiency. Furthermore, based on real-time monitored traffic flow data and environmental information, the cycle, phase duration, and signal timing of traffic lights are dynamically adjusted; simultaneously, based on calculated pollutant emissions, the signal cycle and green light duration at the intersection are adjusted again to optimize vehicle flow, reduce vehicle stopping and idling time under traffic congestion, and lower emissions.

[0117] 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 description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for edge roadside-based traffic signal control, the method comprising: The method comprises the following steps: acquiring the traffic flow of each intersection in the intersection at the current time; calculating the proportion of the traffic flow of each intersection in the total traffic flow of the intersection to obtain first data; processing the first data through a Logistic mapping to predict the proportion of the traffic flow of each intersection in the total traffic flow of the intersection at multiple times as second data; calculating a Jacobian matrix based on the second data and obtaining a Lyapunov exponent based on the Jacobian matrix; determining whether the Lyapunov exponent is a positive number, and if the Lyapunov exponent is a positive number, adjusting the noise suppression strength based on the Lyapunov exponent to obtain a noise suppression factor, and an expression of the noise suppression factor is: wherein, μ(t) represents the noise suppression factor, λ' represents an adjustment factor, and λ(t) represents the Lyapunov exponent; The first data is processed based on the noise suppression factor to obtain denoised first data, and an expression thereof is: wherein y(t) represents the denoised first data, x(t) represents the first data, represents an average value of the historical denoised first data; dynamically adjusting a signal period and a green light time based on the denoised first data to obtain a first signal period and a first green light time, comprising: calculating an original signal period based on the denoised first data, and the expression is: wherein C represents a signal cycle, L represents a loss time of the intersection, y j (t) represents the first denoised data of the lane j, j represents an identification number of the lane, and m represents a number of lanes of the intersection. dynamically adjusting the original signal period according to the Lyapunov exponent to obtain an adjusted signal period, and the expression is: C new = C · (1 + a · l(t)) where C new denotes the first signal period, and a denotes the sensitivity coefficient. calculating the green light time of each lane based on the original signal period and the ratio of the denoised first data of each lane to the total lane denoised first data, and the expression is: wherein G i represents the original green light time of the lane i, y i (t) represents the denoised first data of the lane i; obtaining the updated green light time of the current lane based on the green light time of the current lane, the traffic flow at the current time and the traffic flow at the next time, and the expression is: wherein, represents the first green light time of the lane i, and β represents an adjustment coefficient, represents the predicted traffic flow at the next time, Q t represents the traffic flow at the current time; calculating the pollutant emission based on the first signal period and the first green light time; if the pollutant emission exceeds a preset value, adjusting the signal period and the green light time of the intersection again based on the target pollutant emission of the intersection to obtain a second signal period and a second green light time, comprising: where C' new represents the second signal cycle, C new represents the first signal cycle, a represents a sensitivity coefficient, Q represents a pollutant emission amount, Q target represents a target pollutant emission amount, represents a second green time of the lane i, k i represents a vehicle density of the lane i, k j represents a vehicle density of the lane j, and m represents the number of lanes of the intersection.

2. The edge-based roadside traffic signal control method of claim 1, wherein, adjusting the signal period and the green light time of the intersection again based on the overall target emission of the intersection: further comprising: reconstructing the Jacobian matrix based on the Takens theorem to obtain a reconstructed matrix; suppressing the noise trajectory of the reconstructed matrix to obtain a suppressed Jacobian matrix; 3. The edge-based roadside traffic signal control method of claim 1, wherein, calculating the Williams entropy according to the suppressed Jacobian matrix. x t+1 = r x t (1 - x t ) wherein x t+1 represents the second data at t+1, and r is a parameter reflecting the influence of external factors.

4. The edge-based roadside traffic signal control method of claim 3, wherein, processing the first data through a Logistic mapping to predict the proportion of the traffic flow of each intersection in the total traffic flow of the intersection at multiple times as second data, and the expression is: where λ(t) represents a Lyapunov exponent, n represents the number of second data, and f'(x t+i ) represents an eigenvalue of the Jacobian matrix at time t+i.

5. The edge roadside-based traffic signal control method of claim 2, wherein, obtaining the Lyapunov exponent based on the Jacobian matrix, and the expression is: X i (t) = X(t + (i - 1)τ) where X'(t) represents the inhibited Jacobian matrix, φ(t) represents the time evolution factor, X(t) represents the reconstructed matrix, C i is the weight coefficient, X i (t) is the i-th dimension of the reconstructed matrix X(t) in the phase space generated in the reconstruction process, τ represents the delay time, and X(t + (i - 1)τ) represents the signal component delayed by (i - 1)τ.

6. The edge-based roadside traffic signal control method of claim 1, wherein, suppressing the noise trajectory of the reconstructed matrix to obtain a suppressed Jacobian matrix, and the expression is: calculating the pollutant emission based on the adjusted signal period and the green light time, comprising: obtaining the lane density, vehicle speed and vehicle type of each lane based on the adjusted signal period and the green light time; Q = ∫0 L′ k(x, t) · E[v(k(x, t), T)] dx calculating the pollutant emission according to the lane density, vehicle speed and vehicle type of each lane, and the expression is: wherein Q represents the pollutant emission, L' represents the road length, k(x, t) represents the vehicle density at time t, v represents the vehicle speed, T represents the vehicle type, and E represents the emission factor.

Images