CN102665294B - Vehicular sensor networks (VSN) event region detection method based on Dempster-Shafer (D-S) evidence theory - Google Patents

Abstract

The present invention provides a vehicle sensor network event region detection method based on DS evidence theory, which can effectively realize the detection of events in the absence of prior knowledge, high mobility network topology and complex and changeable road traffic scenarios. Occurrence area. The invention includes vehicle sensor network scene initialization and maintenance, road division sub-district event monitoring probability module, road division sub-district event occurrence probability module, event occurrence confidence degree module, evidence merge conflict calculation module, event area judgment module and event detection trigger module .

Description

Event Region Detection Method for Vehicle Sensor Networks Based on D-S Evidence Theory

technical field

The present invention relates to the field of collaborative data processing and event monitoring of vehicle-mounted self-organizing wireless sensor networks, and more specifically, relates to a new method that uses D-S evidence theory to describe the inconsistency of data fusion in vehicle-mounted sensor networks, thereby detecting the inconsistencies in the road condition environment. method of the event area.

Background technique

With the popularization of automobiles and the development of sensing technology, wireless communication and other technologies, by installing sensor node devices in vehicles driving on the road and interconnecting them through wireless communication, they are self-organized into wireless vehicle sensor networks; vehicle sensor networks can realize vehicle Collaborative perception, processing and transmission of information such as various road traffic conditions in urban areas is an important means and way to realize intelligent transportation. Event monitoring is one of the important applications of vehicle-mounted sensor networks for intelligent transportation. Event area detection is a key technology in event monitoring of vehicle-mounted wireless sensor networks. Through effective collaboration between vehicles and technologies such as data fusion processing, it is possible to effectively detect road conditions and locate event areas. , range, directly affect the efficiency and performance of road emergency handling applications. Event region detection in wireless sensor networks and ad hoc networks is one of the hot application-oriented issues, which have been discussed in technical literature and research papers in recent years.

Related literature: R.Nowak et al.Boundary Estimation in Sensor Network: Theory and Methods.In:Proc.IPSN 2003[c], 2003; K.Ren et al.Secure and Fault-tolerant Event Boundary Detection inWireless Sensor Networks.IEEE Transactions on Wireless Communications[J], 2008, 7(1); Cao Donglei et al. A fault-tolerant algorithm for event region detection in wireless sensor networks. Journal of Computers [J], 2007, 30(10); Zhang Shukui et al. Based on fusion tree Fault-tolerant algorithm for event region detection[J]. Journal of Communications, 2010, (09).

The realization of event area detection technology can be divided into three methods: statistical method, method based on classification and method based on image processing technology. Statistical methods mainly determine event nodes and non-event nodes by obtaining the data information of neighbor nodes and using statistical calculation methods. The classification-based method processes the data information collected by all nodes. Obviously, there is a big difference between the data information of internal nodes and external nodes in the event area. Using this difference, the network nodes can be classified to determine the event area nodes and non-event areas. node. The image processing technology-based approach modifies the image processing technology to adapt it to event region detection in sensor networks. Among them, the classification-based method has the characteristics of simple implementation and low complexity, and is the most suitable method for event region detection technology among the three methods.

The implementation method of event area detection technology should be related to specific applications. In the vehicle-mounted sensor network environment, due to the high mobility of vehicles and the complexity of urban road traffic conditions, the network topology changes rapidly, and event monitoring cannot be simply set. The sensor data threshold is realized; therefore, the general wireless ad hoc network and wireless sensor network event area detection methods cannot be well applied to this application scenario. Some researchers have proposed a method of using artificial intelligence for vehicle self-organizing networks, through local node collaboration, using machine learning, support vector machines, Bayesian neural networks, or using hidden Markov models, etc., to extract event features and Classify and judge the probability of event generation to realize event monitoring; these event monitoring methods can effectively monitor road vehicle-related events in the vehicle ad hoc network environment, but these methods need to give prior knowledge and other information in the specific environment of road vehicles in advance, these Since road traffic and vehicle driving are greatly affected by the natural environment, road terrain characteristics and human factors, the rationality of the value of prior knowledge directly affects the performance of the event monitoring system; in addition, these studies are mainly devoted to the judgment of events, rather than Events generate area extent and location information. In fact, the detection of event areas poses technical challenges due to the high mobility of vehicles in the vehicular network environment.

Related literature: J.Eriksson et al.The Pothole Patrol: Using a Mobile Sensor Network for RoadSurface Monitoring[c].In Proc.ACM MobiSys, 2008; V.D.Sanchez et al.Advanced support vectormachines and kernel me-thods[J]. Neurocomputing, 2003, (55); S Dipti.Evaluation of AdaptiveNeural Network Models for Freeway Incident Detection[J].IEEE Trans.On IntelligentTransportation Systems, 2004, 5(1); He Yi et al. Based on hidden Markov metric field model Vehicle monitoring and tracking [J]. Journal of Shanghai Jiaotong University, 2008 (2). Zhang Cunbao et al. Research on automatic identification method of highway traffic incidents based on floating vehicles [J]. Journal of Wuhan University of Technology (Transportation Science and Engineering Edition), 2006, (06); Zhang Jinglei et al. Research Progress in Event Detection Algorithms [J]. Journal of Wuhan University of Technology (Transportation Science and Engineering Edition), 2005, (02).

Contents of the invention

The present invention aims at the problems and deficiencies in the above-mentioned existing methods, and proposes a vehicle-mounted sensor network event area detection method that is oriented to intelligent transportation applications, does not require prior knowledge and effectively improves monitoring efficiency, and is dedicated to the dynamic topology environment of the vehicle-mounted ad hoc network , through inter-vehicle collaboration, using the data information fusion processing method, using evidence conflicts between incident areas and non-incident areas, to effectively detect incident areas.

The technical solution of the present invention is a vehicle-mounted sensor network event area detection method based on the D-S evidence theory, which is characterized in that: the event-monitored urban road area is divided into several sub-districts, and the vehicle-mounted sensor is provided in the event-monitored urban road area Vehicles are networked in a self-organizing manner to establish a vehicle-mounted sensor network graph. In the vehicle-mounted sensor network graph, vehicles equipped with vehicle-mounted sensors constitute vehicle nodes, and adjacent vehicle nodes are edges. When performing event area detection, the following steps are performed:

Step 1. From the distance from each vehicle node to the center of the sub-district, calculate the monitoring weight of each vehicle node to the sub-district; calculate the behavior factor of the vehicle node according to the direction and speed of the vehicle node; Calculate the rate of change of the perceived data based on the historical data of the perceived physical quantity of the community; and obtain the event occurrence probability of the sub-district where it is located according to the calculation results; the implementation method is as follows,

Step a, set the geometric distance from the coordinates (x _nk , y _nk ) of a certain vehicle node k to the coordinates (x _ci , y _ci ) of the center point of the sub-cell _ci where it is located, denoted as d _{k, i} =||( x _nk , _{y nk} )-(x _ci , y _ci )|, the monitoring weight w( _nk , ci ) from vehicle node k to the sub-cell _ci where it is located is calculated by the following formula,

$\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\frac{{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; r</span>}}& {\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; r</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& {\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; r</span>}\end{array}\right.$

Among them, r is the monitoring perception radius of the vehicle sensor;

Step b, set the observed time series as (t, t'), the velocity of vehicle node k at time t is v _k , the velocity of vehicle node k at time t' is v _k ', max(v _k , v _k ') is the larger value of the velocity v _{k and} v _k ', if max(v _k , v _k ')=0, the behavior factor of vehicle node k μ _k =0; if max(v _k , v _k ')≠ 0, the behavior factor μ _k of vehicle node k is calculated by the following formula,

Among them, α is the weight parameter, θ is the rate vector change to direction angle;

Step c, the vehicle node k obtains the historical data of the perceived physical quantity of the observation time series (t, t') of the sub-cell c _i where it is located, and the time interval of the observation time series (t, t') is denoted as Δt, and the time interval Δt is obtained The average value ave(p, Δt) of a certain physical quantity p within the time interval Δt = sum(p_data)/N(Δt), where sum(p_data) is the sum of the values obtained from the monitoring of the physical quantity p _by sub-communities within the time interval Δt, and N(Δt) is The data monitoring times of the sensor within Δt time;

The greater value between the average value ave(p, Δt) of the physical quantity p in the time interval Δt and the real-time monitoring value p_data(t') of the physical quantity p at the current moment t' is recorded as max(ave(p, Δt), p_data( t')), if max(ave(p, Δt), p_data(t'))=0, data change rate γ _ci =0, if max(ave(p, Δt), p_data(t'))≠0 , the data change rate γ _ci is calculated according to the following formula,

${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; ci</span>}}<span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; \_</span>\mathrm{<span\; class="notranslate">\; data</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; ave</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&Delta;t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}{\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; \_</span>\mathrm{<span\; class="notranslate">\; data</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; ave</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&Delta;t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}$

In step d, according to the monitoring weight w( _nk , _ci ) obtained in step a, the behavior factor μ _k obtained in step b, and the data change rate γ _ci obtained in step c, the event generation probability Pr( _ci ) in sub-cell _ci is obtained as follows:

$\mathrm{<span\; class="notranslate">\; PR</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&lambda;</span>}\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; K</span>}}<span\; class="notranslate">\; *</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; K</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&lambda;</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}_{\mathrm{<span\; class="notranslate">\; ci</span>}}$

Among them, λ is an adjustment factor, which is a constant between 0 and 1; K is the maximum number of vehicle nodes in the sub-cell c _i , and the value of k is 1, 2, ..., K;

Step 2. According to the D-S evidence theory, set the basic probability distribution function of each sub-cell event; the implementation method is as follows,

Let T represent that there is an event in a certain sub-district, and F represent that there is no event in a certain sub-district, then the target recognition framework is expressed as Θ={T, F}, and the total state set in the target recognition framework is 2 ^Θ ={Φ, T, F, Θ={T, F}}, use m _i (T) to represent the basic probability assignment of sub-cell c _i "has an event" state, m _i (F), _mi (Θ) represent sub-cell c The basic probability assignment of "no event" and "uncertain" state in _i ;

The basic probability distribution function m _i (T) of the sub-cell c _i is set according to the event generation probability Pr(c _i ) calculated in step 2, that is, m _i (T)≡Pr(c _i );

Step 3. According to the DS evidence theory, calculate the confidence function value Bel _i (T) of the events generated in each sub-cell on the observed time series (t, t'); the implementation method is as follows,

From the basic probability distribution function m _i (T) obtained in step 2, calculate the confidence function value Bel _i (T) of the sub-cell c _i on the observed time series (t, t') as follows,

Bel _i (T) = m _i (T)

Step 4. According to the D-S evidence theory, merge the confidence function values of each sub-cell and adjacent sub-cells to obtain the conflict value between the merged event evidence; the implementation method is as follows,

Assume that the neighboring sub-cells adjacent to each direction of sub-cell c _i form a set SN, and the confidence function value Bel _i (T) obtained in step 3 is used to calculate the combined evidence conflict function Con(Bel _i , Bel _{neigh(i )} ), to get the conflict value between the merged event evidence:

Con(Bel _i , Bel _neigh(i) )=log(1/η), $\mathrm{\&eta;}=\stackrel{\left|\mathrm{SN}\right|}{\underset{1}{\mathrm{\&Sigma;}}}{\mathrm{bel}}_i\left(T\right)*{\mathrm{bel}}_{\mathrm{neigh}\left(i\right)}\left(T\right)$

Among them, Bel _neigh(i) (T) is the confidence function value of each sub-cell neigh(i) adjacent to sub-cell c _i , neigh(i)∈SN; |SN| is the number of sub-cells in the set SN number;

Step 5, judging whether each sub-cell belongs to the event area, the implementation method is as follows,

If the conflict Con(Bel _i , Bel _neigh(i) ) between the merged event evidence obtained in step 4 is greater than or equal to the preset threshold Con _th , then it is determined that the sub-cell c _i is in the event area, otherwise it is determined that the sub-cell c _i is in the non-event area area.

Moreover, the event area detection is performed periodically, or the event area detection is triggered when the physical quantity update data collected by the vehicle sensor exceeds a certain preset value.

The technical solution provided by the invention is applicable to the dynamic vehicle-mounted self-organizing network topology environment and the complex and changeable characteristics of vehicle-mounted applications, without prior knowledge, making full use of the space-time correlation of monitoring data in the network, and adopting the D-S evidence conflict principle to identify network The inconsistency of regional monitoring information is used to classify incident areas and non-event areas by using the difference in evidence, so as to solve the problem that it is difficult to effectively monitor and identify incident areas in the vehicle sensor network environment, and improve the accuracy and effectiveness of incident monitoring and judgment.

Description of drawings

FIG. 1 is a schematic diagram of area division according to an embodiment of the present invention;

2 is a schematic diagram of a vehicle-mounted sensor network for road event monitoring according to an embodiment of the present invention;

Fig. 3 is the schematic diagram of the network topology randomly formed by 30 vehicle nodes in the embodiment of the present invention at time _t0 ;

Fig. 4 is the schematic diagram of the network topology and randomly generated events of 30 vehicle nodes at time _t1 according to the embodiment of the present invention;

FIG. 5 is a schematic diagram of network topology and continuous events of 30 vehicle nodes at time _t2 according to an embodiment of the present invention;

FIG. 6 is a detection result of an event region according to an embodiment of the present invention.

Fig. 7 is a structural diagram of an embodiment of the present invention.

Detailed ways

The classification-based event area detection method uses the difference of data information collected inside and outside the event area to classify and determine the scope of the event; the D-S evidence theory is used as an effective reasoning method for dealing with uncertainties. If T is used to represent the coverage road There is a monitoring event in a sub-cell in the map, F means no event, then the target recognition framework is: Θ={T, F}. D-S evidence conflict is used to reflect the degree of difference between specific space-time information evidence: when the D-S evidence conflict weight function value is ∞, it means that there is a complete conflict between the merged evidence, that is, there is a complete conflict between the merged evidence and other area evidence, indicating that the merged evidence area and Other areas belong to the event area and non-event area; when the D-S evidence conflict weight function value is 0, it means that there is no conflict in the merged evidence, that is, the merged evidence and other areas belong to the non-event area. According to the evidence inconsistency rules between the event area and its adjacent non-event area, this evidence conflict is used to identify and judge the event range.

Embodiments of the present invention are described in detail with reference to the accompanying drawings. This embodiment is carried out on the premise of the method of the present invention, and the detailed implementation manner and specific operation process are given, but the embodiments of the present invention are not limited to the following embodiments.

The present invention provides a vehicle sensor network event area detection method based on DS evidence conflict. For the sake of simplicity of description, 30 vehicle nodes arranged in a specific monitoring area are networked, and the ellipse randomly generated at time t ₁ after the implementation starts Take the shape event area as an example to detect the event area.

First, network initialization is performed, including event region initialization and vehicle network modeling.

The urban road area for event monitoring is divided into several sub-districts, and the vehicles equipped with on-board sensors in the urban road area for event monitoring are networked by self-organization to establish a network diagram of on-board sensors. The vehicles constitute the vehicle nodes, and the adjacent vehicle nodes are edges.

The event monitoring application monitors urban road conditions through vehicle driving coverage. The urban road area for event monitoring can be approximated as a planar rectangular area, and the planar map of the urban road area can be extracted, and the monitoring area can be divided into multiple small sub-cells with the same size C={c ₁ , c ₂ , .. .c _n }, as shown in Figure 1, can simply divide sub-cells by row and column. (X _ci , Y _ci ), (i=1,...n) can be used to uniquely identify the sub-cells _ci in the monitoring area, where X _ci is the row position index number, and Y _ci is the column position index number. The size parameters for dividing sub-cells depend on the monitoring accuracy requirements, are specified by the upper layer application and can be adjusted during the application process. After extracting the vehicles driving in the road area corresponding to the planar map, it is modeled as a vehicle sensor network diagram through self-organizing networking. Vehicles deployed and equipped with sensors are abstracted into nodes in the graph, and opportunistic connections are generated between vehicle nodes wirelessly, and these opportunistic communication links are abstracted into edges of the graph; except for the communication between vehicle nodes, the vehicle Nodes can also connect to the roadside access point wirelessly, and access the Internet through the roadside access point. Each vehicle node is assigned a unique ID number when the network is initialized. According to the GPS installed on the vehicle, the node can obtain vehicle information such as its position, speed, and direction in the driving area (generally, time, driving distance, driving time, and vehicle information). Acceleration, in addition, various types of sensors installed and placed inside and outside the vehicle, such as: temperature and humidity sensors, 3-axis acceleration sensors, etc., can be used to monitor various types of physical quantities in the driving environment of road vehicles, such as road humidity and road temperature. Vehicle node monitoring information can be obtained through multiple The hop mode is transmitted to the roadside access point through an opportunistic wireless link, so as to finally access the monitoring center for road traffic monitoring; the driving vehicle can also receive the global message issued by the monitoring center through the roadside access point. The communication between the side access point and the monitoring center can be realized through the Internet. After the network initialization is completed, the vehicle nodes update the network communication nodes and links through opportunistic message transmission and global messages issued by the roadside access point. Through event area detection, events can be judged, such as slippery road surface, road potholes, road congestion, etc.

Event area detection can be performed periodically, or triggered when the update data collected by the vehicle sensor exceeds a certain preset value.

When performing event region detection, perform the following steps:

Step 1. From the distance from each vehicle node to the center of the sub-district, calculate the monitoring weight of each vehicle node to the sub-district; calculate the behavior factor of the vehicle node according to the direction and speed of the vehicle node; Calculate the rate of change of the perceived data based on the historical data of the perceived physical quantity of the cell; and obtain the event occurrence probability of the sub-cell where it is located according to the calculation result.

The implementation of the embodiment is as follows:

Step a, set the geometric distance from the coordinates (x _nk , y _nk ) of a certain vehicle node k to the coordinates (x _ci , y _ci ) of the center point of the sub-cell _ci where it is located, denoted as d _{k, i} =||( x _nk , _{y nk} )-(x _ci , y _ci )|, the monitoring weight w( _nk , ci ) from vehicle node k to the sub-cell _ci where it is located is calculated by the following formula,

$w({\mathrm{no}}_k,c_i)=\left\{\begin{array}{cc}1-\frac{d_{k,i}}{r}& d_{k,i}\&le;r\\ 0& d_{k,i}>r\end{array}\right.$ Formula 1

Among them, r is the monitoring sensing radius of the vehicle sensor.

Step b, in the observation time series (t, t'), the velocity of vehicle node k at time t is v _k , the velocity of vehicle node k at time t' is v _k ', max(v _k , v _k ') is The larger value of velocity v _{k and} v _k ', if max(v _k , v _k ')=0, the behavior factor of vehicle node k μ _k =0; if max(v _k , v _k ')≠0 , the behavior factor μ _k of vehicle node k is calculated by the following formula,

Formula 2

The behavior of vehicle nodes is affected by road events, and corresponding behavior changes occur. For example, most vehicles choose to change their driving directions to avoid the event area. According to the observed time series (t, t'), the vehicle node k's speed v _k at time t and the node's speed v _k ' at time t', the behavior factor μ of each vehicle node is calculated as shown in formula 2. Among them, α is a weight parameter, which is used to adjust the weight of vehicle direction change and speed change for the evaluation of vehicle behavior factors; θ is the speed vector change to , max(v _k , v _k ') is the larger value (and should not be zero) of the moving speed of the vehicle node at time t and t', if max(v _k , v _k ')=0, that is When the vehicle node is in a static state, the node behavior factor is 0.

Step c, the vehicle node k obtains the historical data of the perceived physical quantity of the observation time series (t, t') of the sub-cell c _i where it is located, and the time interval of the observation time series (t, t') is denoted as Δt, and the time interval within the time interval Δt is obtained The average value ave(p, Δt) of a certain physical quantity p=sum(p_data)/N(Δt), where sum(p_data) is the sum of the values obtained from the monitoring of the physical quantity p by sub-cells c _i within the time interval Δt, and N(Δt) is Δt The number of sensor data monitoring times;

The greater value between the average value ave(p, Δt) of the physical quantity p in the time interval Δt and the real-time monitoring value p_data(t') of the physical quantity p at the current moment t' is recorded as max(ave(p, Δt), p_data( t')), if max(ave(p, Δt), p_data(t'))=0, data change rate γ _ci =0, if max(ave(p, Δt), p_data(t'))≠0 , the data change rate γ _ci is calculated according to the following formula,

$r_{\mathrm{ci}}=\frac{|p\_\mathrm{data}\left(t^{\&prime;}\right)-\mathrm{ave}(p,\mathrm{\&Delta;t})|}{\max (p\_\mathrm{data}\left(t^{\&prime;}\right),\mathrm{ave}(p,\mathrm{\&Delta;t}))}$ Formula 3

The vehicle node can obtain the historical data of the perceived physical quantity of the adjacent cell c _i through the roadside access point, based on which the average value of the physical quantity at the time interval Δt(t, t') can be calculated. During specific implementation, the data monitoring frequency of the sensor within Δt is related to the sampling frequency f of the sensor, N(Δt)=Δt*f. If max(ave(p, Δt), p_data(t)) is 0, that is, the region does not acquire sensory data, then its corresponding data change rate γ _ci is 0.

In step d, according to the monitoring weight w( _nk , _ci ) obtained in step a, the behavior factor μ _k obtained in step b, and the data change rate γ _ci obtained in step c, the event generation probability Pr( _ci ) in sub-cell _ci is obtained as follows:

$\mathrm{PR}\left(c_i\right)=\mathrm{\&lambda;}\frac{1}{K}*\underset{k=1}{\overset{K}{\mathrm{\&Sigma;}}}w({\mathrm{no}}_k,c_i){\mathrm{\&mu;}}_k+(1-\mathrm{\&lambda;}){\mathrm{\&gamma;}}_{\mathrm{ci}}$ Formula 4

Where λ is the adjustment factor, which is a constant between 0 and 1. This adjustment factor is used to adjust the influence weight of node behavior and data change rate on the calculation of sub-cell event monitoring probability results; K is the maximum number of vehicle nodes in sub-cell c _i , the value of k is 1, 2, ..., K.

Step 2. According to the D-S evidence theory, set the basic probability distribution function of each sub-cell event; the implementation method is as follows,

Let T represent that there is an event in a certain sub-district, and F represent that there is no event in a certain sub-district, then the target recognition framework is expressed as Θ={T, F}, and the total state set in the target recognition framework is 2 ^Θ ={Φ, T, F, Θ={T, F}}, use m _i (T) to represent the basic probability assignment of sub-cell c _i "has an event" state, m _i (F), _mi (Θ) represent sub-cell c The basic probability assignment of "no event" and "uncertain" state in _i ;

The basic probability distribution function m _i (T) of the sub-cell c _i is set according to the event generation probability Pr( _ci ) calculated in step 2, that is, the event generation probability function m _i (T)≡Pr of each sensor n _k to c _i cell (c _i ).

Step 3. According to the DS evidence theory, calculate the confidence function value Bel _i (T) of the events generated in each sub-cell on the observed time series (t, t'); the implementation method is as follows,

From the basic probability distribution function m _i (T) obtained in step 2, the confidence function value _Beli (T) of the sub-cell c _i occurrence event on the observed time series (t, t') is calculated according to the following formula.

Bel _i (T) = m _i (T) Formula 5

During specific implementation, the basic probability distribution functions obtained from the observation time series of multiple time intervals Δt can also be summed and averaged as the value of the confidence function to improve the accuracy of the results.

Step 4. According to the D-S evidence theory, merge the confidence function values of each sub-cell and adjacent sub-cells to obtain the conflict value between the merged event evidence; the implementation method is as follows,

Set the set of neighbor sub-cells adjacent to each direction of sub-cell ci as SN _{, and calculate the combined evidence conflict function Con(Bel i ,} Bel _{neigh( i)} ), the conflict value of event evidence merging in adjacent sub-cells is obtained:

$\mathrm{Con}({\mathrm{bel}}_i,{\mathrm{bel}}_{\mathrm{neigh}\left(i\right)}=\log (1/\mathrm{\&eta;}),\mathrm{\&eta;}=\underset{1}{\overset{\left|\mathrm{SN}\right|}{\mathrm{\&Sigma;}}}{\mathrm{bel}}_i\left(T\right)*{\mathrm{bel}}_{\mathrm{neigh}\left(i\right)}\left(T\right)$ Formula 6

Among them, Bel _neigh(i) (T) is the confidence function value of each sub-cell neigh(i) adjacent to sub-cell c _i , neigh(i)∈SN; |SN| is the number of sub-cells in the set SN number.

The D-S evidence theory is used as an effective reasoning method to deal with uncertainties, and the D-S evidence consistency is used to reflect the degree of combined difference between specific space-time information evidence: the value of the D-S evidence consistency function is in the interval (0, ∞) . By merging the event occurrence evidence of adjacent sub-cells, it is used to establish the evidence conflict attribute for the detection of event area and non-event area, and use the evidence conflict to classify according to the difference between the internal and external evidence of the event area, so as to determine the scope of the event.

Step 5, judging whether each sub-cell belongs to the event area. For a sub-cell with Con ≥ Con _th , it means that after the evidence of the critical area where it is merged, there is a large conflict with the non-event area, that is, the sub-cell falls within the scope of the event area, that is, it belongs to the event area. The implementation is as follows,

If the conflict Con(Bel _i , Bel _neigh(i) ) between the merged event evidence obtained in step 4 is greater than or equal to the preset threshold Con _th , then it is determined that the sub-cell c _i is in the event area, otherwise it is determined that the sub-cell c _i is in the non-event area area.

When each network cycle expires or when the data collected by new sensors reaches the set threshold, the event area detection process is restarted, and the above steps are repeated until the road event monitoring application ends, and real-time monitoring can be realized.

For the convenience of understanding the effect of the present invention, a section of specific detection process of the embodiment is provided in detail as follows:

After extracting and dividing the plane information of the specific monitoring road area in Figure 1, the vehicle sensor network is deployed on this basis. Figure 2 is a schematic diagram of the network networking of the vehicle sensor network for road event monitoring: On the basis of cell division, when the communication radius is 15 units, the 30 vehicle nodes in the monitoring area can randomly form the topology shown in Figure 3: that is, after the initialization of the vehicle network in the aforementioned monitoring area (time t ₀ ) , the node distribution and edges randomly formed by 30 vehicle nodes carrying sensors, the node coordinate distribution is as follows:

Node id: 1, X coordinate: 0.127567, Y coordinate: 23.5572

Node id: 2, X coordinate: 71.2967, Y coordinate: 7.43736

Node id: 3, X coordinate: 29.7266, Y coordinate: 36.0027

Node id: 4, X coordinate: 28.0816, Y coordinate: 98.529

Node id: 5, X coordinate: 35.081, Y coordinate: 93.4233

Node id: 6, X coordinate: 19.6487, Y coordinate: 85.757

Node id: 7, X coordinate: 3.83374, Y coordinate: 0.85757

Node id: 8, X coordinate: 68.9435, Y coordinate: 48.5885

Node id: 9, X coordinate: 30.0287, Y coordinate: 80.2179

Node id: 10, X coordinate: 10.061, Y coordinate: 44.8012

Node id: 11, X coordinate: 70.4172, Y coordinate: 97.3266

Node id: 12, X coordinate: 72.6966, Y coordinate: 79.0612

Node id: 13, X coordinate: 61.9608, Y coordinate: 4.0376

Node id: 14, X coordinate: 96.9546, Y coordinate: 6.83615

Node id: 15, X coordinate: 32.6472, Y coordinate: 98.5076

Node id: 16, X coordinate: 8.21131, Y coordinate: 1.80059

Node id: 17, X coordinate: 2.81991, Y coordinate: 56.7248

Node id: 18, X coordinate: 56.7574, Y coordinate: 64.8122

Node id: 19, X coordinate: 79.263, Y coordinate: 38.5052

Node id: 20, X coordinate: 41.8152, Y coordinate: 2.64595

Node id: 21, X coordinate: 99.1433, Y coordinate: 20.9906

Node id: 22, X coordinate: 94.7456, Y coordinate: 54.5885

Node id: 23, X coordinate: 106.072, Y coordinate: 62.8193

Node id: 24, X coordinate: 40.9961, Y coordinate: 94.9461

Node id: 25, X coordinate: 25.4396, Y coordinate: 52.0829

Node id: 26, X coordinate: 25.6108, Y coordinate: 89.7366

Node id: 27, X coordinate: 41.221, Y coordinate: 41.2152

Node id: 28, X coordinate: 81.8714, Y coordinate: 35.551

Node id: 29, X coordinate: 41.3889, Y coordinate: 21.9001

Node id: 30, X coordinate: 7.82525, Y coordinate: 58.7909

In this embodiment, the vehicle node can travel along four directions: horizontal left, horizontal right, vertical upward, and vertical downward, and the average speed is within the speed range of 0-10 unit length/unit time. After initialization, each vehicle node moves randomly in the monitoring area according to the vehicle driving mode according to the set route. When encountering the event area, the vehicle will avoid the area and detour or slow down.

Assuming that the topology shown in Figure ₄ is generated at time t1, and events are generated in the randomly generated oval area, the network topology information generated at time _t1 is: vehicle node coordinates, moving direction and speed are as follows.

Node id: 1, X coordinate: 1.77252, Y coordinate: 43.788 Movement direction: Horizontal to the right Movement rate: 3

Node id: 2, X coordinate: 46.6225, Y coordinate: 22.7485 Movement direction: vertical upward Movement rate: 2

Node id: 3, X coordinate: 53.5279, Y coordinate: 33.5154 Movement direction: Horizontal to the right Movement rate: 3

Node id: 4, X coordinate: 55.3273, Y coordinate: 58.5284 Movement direction: Vertical downward Movement rate: 8

Node id: 5, X coordinate: 99.3313, Y coordinate: 79.0612 Movement direction: vertical downward Movement rate: 9

Node id: 6, X coordinate: 55.1729, Y coordinate: 5.91754 Movement direction: Vertical downward Movement rate: 5

Node id: 7, X coordinate: 41.6474, Y coordinate: 75.1671 Movement direction: Horizontal to the right Movement rate: 5

Node id: 8, X coordinate: 23.5832, Y coordinate: 85.3359 Movement direction: Vertical downward Movement rate: 5

Node id: 9, X coordinate: 3.9613, Y coordinate: 14.6123 Movement direction: vertical downward Movement rate: 4

Node id: 10, X coordinate: 51.8998, Y coordinate: 13.538 Movement direction: Vertical downward Movement rate: 3

Node id: 11, X coordinate: 98.5827, Y coordinate: 83.9381 Moving direction: vertical upward Moving rate: 3

Node id: 12, X coordinate: 43.2621, Y coordinate: 90.0906 Movement direction: Vertical downward Movement rate: 0

Node id: 13, X coordinate: 45.1454, Y coordinate: 66.3472 Movement direction: Horizontal to the right Movement rate: 6

Node id: 14, X coordinate: 46.6594, Y coordinate: 48.8083 Moving direction: vertically upward Moving rate: 8

Node id: 15, X coordinate: 4.50179, Y coordinate: 30.0485 Movement direction: Horizontal to the left Movement rate: 0

Node id: 16, X coordinate: 89.3106, Y coordinate: 77.4346 Movement direction: Horizontal to the right Movement rate: 7

Node id: 17, X coordinate: 87.6556, Y coordinate: 29.9936 Movement direction: Vertical downward Movement rate: 9

Node id: 18, X coordinate: 74.5631, Y coordinate: 58.5437 Movement direction: Vertical upward Movement rate: 8

Node id: 19, X coordinate: 14.4151, Y coordinate: 1.26652 Movement direction: Vertical upward Movement rate: 1

Node id: 20, X coordinate: 37.5317, Y coordinate: 34.8949 Movement direction: Horizontal to the right Movement rate: 4

Node id: 21, X coordinate: 11.5381, Y coordinate: 72.1641 Movement direction: Vertical downward Movement rate: 2

Node id: 22, X coordinate: 5.62304, Y coordinate: 52.3179 Movement direction: Horizontal to the right Movement rate: 2

Node id: 23, X coordinate: 24.9596, Y coordinate: 75.161 Movement direction: Horizontal to the left Movement rate: 5

Node id: 24, X coordinate: 86.1382, Y coordinate: 83.2575 Movement direction: vertical upward Movement rate: 3

Node id: 25, X coordinate: 95.8232, Y coordinate: 83.8313 Movement direction: Vertical downward Movement rate: 3

Node id: 26, X coordinate: 28.1689, Y coordinate: 84.106 Movement direction: Horizontal to the left Movement rate: 4

Node id: 27, X coordinate: 75.0298, Y coordinate: 53.8743 Movement direction: Vertical downward Movement rate: 3

Node id: 28, X coordinate: 90.979, Y coordinate: 89.462 Movement direction: Vertical upward Movement rate: 9

Node id: 29, X coordinate: 95.9676, Y coordinate: 96.3683 Movement direction: Vertical downward Movement rate: 6

Node id: 30, X coordinate: 45.9746, Y coordinate: 96.411 Movement direction: Horizontal to the right Movement rate: 6

Calculate the monitoring weight from each vehicle node to the sub-cell according to formula 1, and take the sensing radius r equal to the radius of the sub-cell, and the results are as follows.

The monitoring weight of node 1 to the center position (5, 45) of the sub-cell where it is located is: 0.770162

The monitoring weight of node 2 to the center of the sub-cell (45, 25) is: 0.814986

The monitoring weight of node 3 to the center of the sub-cell (45, 35) is: 0.42292

The monitoring weight of node 4 to the center of the sub-cell (55, 55) is: 0.763764

The monitoring weight of node 5 to the center of the sub-cell (95, 75) is: 0.604165

The monitoring weight of node 6 to the center position (55, 5) of the sub-cell where it is located is: 0.937754

The monitoring weight of node 7 to the center position (35, 75) of the sub-cell where it is located is: 0.556701

The monitoring weight of node 8 to the center position (25, 85) of the sub-cell where it is located is: 0.902928

The monitoring weight of node 9 to the center position (5, 15) of the sub-cell where it is located is: 0.926086

The monitoring weight of node 10 to the center of the sub-cell (45, 15) is: 0.529802

The monitoring weight of node 11 to the center position (85, 85) of the sub-cell where it is located is: 0.0917224

The monitoring weight of node 12 to the center position (35, 95) of the sub-cell where it is located is: 0.35929

The monitoring weight of node 13 to the center of the sub-cell (45, 65) is: 0.909663

The monitoring weight of node 14 to the center position (45, 45) of the sub-cell where it is located is: 0.72306

The monitoring weight of node 15 to the center of the sub-cell (5, 35) is: 0.668235

The monitoring weight of node 16 to the center position (85, 75) of the sub-cell where it is located is: 0.66996

The monitoring weight of node 17 to the center position of the sub-cell (75, 25) is: 0.0929916

The monitoring weight of node 18 to the center position (65, 55) of the sub-cell where it is located is: 0.320095

The monitoring weight of node 19 to the center of the sub-cell (15, 5) is: 0.748065

Node 20 monitors the center position (35, 35) of the sub-cell where the weight is: 0.831077

The monitoring weight of node 21 to the center position (15, 75) of the sub-cell where it is located is: 0.701656

The monitoring weight of node 22 to the center position (5, 55) of the sub-cell is: 0.816431

The monitoring weight of node 23 to the center position (25, 75) of the sub-cell is: 0.988934

The monitoring weight of node 24 to the center position (75, 85) of the sub-cell where it is located is: 0.248423

The monitoring weight of node 25 to the center of the sub-cell (85, 85) is: 0.274257

The monitoring weight of node 26 to the center position (25, 85) of the sub-cell where it is located is: 0.780494

The monitoring weight of node 27 to the center position of the sub-cell (65, 55) is: 0.327152

The monitoring weight of node 28 to the center position (85, 85) of the sub-cell where it is located is: 0.502638

The monitoring weight of node 29 to the center position (85, 95) of the sub-cell where it is located is: 0.263159

The monitoring weight of node 30 to the center position (45, 95) of the sub-cell where it is located is: 0.885674

The time in the ellipse event area generated in this embodiment lasts within (t ₁ , t ₂ ), and the topological information at time t ₂ is node coordinates, moving direction and moving speed, as shown below.

Node id: 1, X coordinate: 4.77252, Y coordinate: 43.788 Moving direction: vertically upward Moving rate: 4

Node id: 2, X coordinate: 46.6225, Y coordinate: 20.7485 Movement direction: vertical upward Movement rate: 2

Node id: 3, X coordinate: 56.5279, Y coordinate: 33.5154 Movement direction: Horizontal to the right Movement rate: 3

Node id: 4, X coordinate: 55.3273, Y coordinate: 66.5284 Movement direction: vertical downward Movement rate: 1

Node id: 5, X coordinate: 99.3313, Y coordinate: 88.0612 Movement direction: vertical downward Movement rate: 10

Node id: 6, X coordinate: 55.1729, Y coordinate: 10.9175 Movement direction: Vertical downward Movement rate: 5

Node id: 7, X coordinate: 46.6474, Y coordinate: 75.1671 Movement direction: Horizontal to the right Movement rate: 5

Node id: 8, X coordinate: 23.5832, Y coordinate: 90.3359 Movement direction: Vertical downward Movement rate: 5

Node id: 9, X coordinate: 3.9613, Y coordinate: 18.6123 Movement direction: Vertical downward Movement rate: 4

Node id: 10, X coordinate: 51.8998, Y coordinate: 16.538 Movement direction: vertical downward Movement rate: 3

Node id: 11, X coordinate: 98.5827, Y coordinate: 80.9381 Movement direction: Vertical downward Movement rate: 4

Node id: 12, X coordinate: 43.2621, Y coordinate: 90.0906 Movement direction: Vertical downward Movement rate: 0

Node id: 13, X coordinate: 51.1454, Y coordinate: 66.3472 Movement direction: Horizontal to the right Movement rate: 1

Node id: 14, X coordinate: 46.6594, Y coordinate: 40.8083 Movement direction: Vertical upward Movement rate: 1

Node id: 15, X coordinate: 4.50179, Y coordinate: 30.0485 Movement direction: Horizontal to the left Movement rate: 1

Node id: 16, X coordinate: 96.3106, Y coordinate: 77.4346 Movement direction: Horizontal to the right Movement rate: 8

Node id: 17, X coordinate: 87.6556, Y coordinate: 38.9936 Movement direction: Vertical downward Movement rate: 9

Node id: 18, X coordinate: 74.5631, Y coordinate: 50.5437 Movement direction: Vertical upward Movement rate: 1

Node id: 19, X coordinate: 14.4151, Y coordinate: 0.266518 Moving direction: vertical upward Moving rate: 1

Node id: 20, X coordinate: 41.5317, Y coordinate: 34.8949 Movement direction: Horizontal to the right Movement rate: 4

Node id: 21, X coordinate: 11.5381, Y coordinate: 74.1641 Movement direction: vertical downward Movement rate: 2

Node id: 22, X coordinate: 7.62304, Y coordinate: 52.3179 Moving direction: vertically upward Moving rate: 3

Node id: 23, X coordinate: 19.9596, Y coordinate: 75.161 Movement direction: Horizontal to the left Movement rate: 5

Node id: 24, X coordinate: 86.1382, Y coordinate: 80.2575 Movement direction: Vertical downward Movement rate: 4

Node id: 25, X coordinate: 95.8232, Y coordinate: 86.8313 Movement direction: Vertical downward Movement rate: 4

Node id: 26, X coordinate: 24.1689, Y coordinate: 84.106 Movement direction: Horizontal to the left Movement rate: 4

Node id: 27, X coordinate: 75.0298, Y coordinate: 56.8743 Movement direction: Vertical downward Movement rate: 1

Node id: 28, X coordinate: 90.979, Y coordinate: 80.462 Movement direction: Vertical downward Movement rate: 10

Node id: 29, X coordinate: 95.9676, Y coordinate: 92 Movement direction: Vertical downward Movement rate: 6

Node id: 30, X coordinate: 51.9746, Y coordinate: 96.411 Movement direction: Horizontal to the right Movement rate: 6

Take the observation time series (t ₁ , t ₂ ), in order to determine whether there is a conflict in the observation time by calculating the DS conflict value, and infer whether there is an event and the event location, firstly, the topology at time t ₁ and t ₂ is calculated according to formula 2 The node behavior factor, the result is as follows (in this embodiment, α=0.5).

Behavior factor μ for node 1: 0.375

Behavior factor μ for node 2: 0

Behavior factor μ for node 3: 0

Behavior factor μ for node 4: 0.4375

Behavior factor μ for node 5: 0.05

Behavior factor μ for node 6: 0

Behavior factor μ for node 7: 0

Behavior factor μ for node 8: 0

Behavior factor μ for node 9: 0

Behavior factor μ for node 10: 0

Behavior factor μ for node 11: 0.625

Behavior factor μ for node 12: 0.5

Behavior factor μ for node 13: 0.416667

Behavior factor μ for node 14: 0.4375

Behavior factor μ for node 15: 0.5

Behavior factor μ for node 16: 0.0625

Behavior factor μ for node 17: 0

Behavior factor μ for node 18: 0.4375

Behavior factor μ for node 19: 0

Behavior factor μ for node 20: 0

Behavior factor μ for node 21: 0

Behavior factor μ for node 22: 0.416667

Behavior factor μ for node 23: 0

Behavior factor μ for node 24: 0.625

Behavior factor μ for node 25: 0.125

Behavior factor μ for node 26: 0

Behavior factor μ for node 27: 0.333333

Behavior factor μ for node 28: 0.55

Behavior factor μ for node 29: 0

Behavior factor μ for node 30: 0

In this embodiment, the sensory data value of the center point of the event occurrence area is set to be twice the value of the sensory data when there is no event, and it decreases linearly from the event center position in the event area with distance. On this basis, calculate the change rate γ _ci of each c _i (X _ci , Y _ci ) cell sensing data according to formula 3, where X _ci is the sub-cell row index number, the value range is 0-9, and Y _ci is the sub-cell column Index number, the value range is 0-10; the calculation result is the data change rate matrix γ relative to each plot c(X _ci , Y _ci ):

$\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; =</span>\left[\begin{array}{ccccccccccc}\mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0253</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0908</span>}& \mathrm{<span\; class="notranslate">\; 0.1888</span>}& \mathrm{<span\; class="notranslate">\; 0.2242</span>}& \mathrm{<span\; class="notranslate">\; 0.2007</span>}& \mathrm{<span\; class="notranslate">\; 0.1160</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0651</span>}& \mathrm{<span\; class="notranslate">\; 0.2242</span>}& \mathrm{<span\; class="notranslate">\; 0.3184</span>}& \mathrm{<span\; class="notranslate">\; 0.3556</span>}& \mathrm{<span\; class="notranslate">\; 0.3306</span>}& \mathrm{<span\; class="notranslate">\; 0.2476</span>}& \mathrm{<span\; class="notranslate">\; 0.1035</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1407</span>}& \mathrm{<span\; class="notranslate">\; 0.2945</span>}& \mathrm{<span\; class="notranslate">\; 0.3184</span>}& \mathrm{<span\; class="notranslate">\; 0.4487</span>}& \mathrm{<span\; class="notranslate">\; 0.4117</span>}& \mathrm{<span\; class="notranslate">\; 0.3184</span>}& \mathrm{<span\; class="notranslate">\; 0.1769</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1529</span>}& \mathrm{<span\; class="notranslate">\; 0.3064</span>}& \mathrm{<span\; class="notranslate">\; 0.4117</span>}& \mathrm{<span\; class="notranslate">\; 0.4783</span>}& \mathrm{<span\; class="notranslate">\; 0.4287</span>}& \mathrm{<span\; class="notranslate">\; 0.3306</span>}& \mathrm{<span\; class="notranslate">\; 0.1889</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1035</span>}& \mathrm{<span\; class="notranslate">\; 0.2593</span>}& \mathrm{<span\; class="notranslate">\; 0.3556</span>}& \mathrm{<span\; class="notranslate">\; 0.3965</span>}& \mathrm{<span\; class="notranslate">\; 0.3687</span>}& \mathrm{<span\; class="notranslate">\; 0.2827</span>}& \mathrm{<span\; class="notranslate">\; 0.1407</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1529</span>}& \mathrm{<span\; class="notranslate">\; 0.2476</span>}& \mathrm{<span\; class="notranslate">\; 0.2827</span>}& \mathrm{<span\; class="notranslate">\; 0.2593</span>}& \mathrm{<span\; class="notranslate">\; 0.1769</span>}& \mathrm{<span\; class="notranslate">\; 0.0253</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0780</span>}& \mathrm{<span\; class="notranslate">\; 0.1160</span>}& \mathrm{<span\; class="notranslate">\; 0.0908</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right]$

Substitute the aforementioned calculation results into formula 4 to calculate the sub-cell event occurrence probability. In this embodiment, the adjustment factor λ=0.5 is used, and the calculation result is the event occurrence probability matrix Pr relative to each cell c(X _ci , Y _ci ).

$\mathrm{<span\; class="notranslate">\; PR</span>}<span\; class="notranslate">\; =</span>\left[\begin{array}{ccccccccccc}\mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0253</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0908</span>}& \mathrm{<span\; class="notranslate">\; 0.1888</span>}& \mathrm{<span\; class="notranslate">\; 0.2242</span>}& \mathrm{<span\; class="notranslate">\; 0.2007</span>}& \mathrm{<span\; class="notranslate">\; 0.1160</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.1671</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0651</span>}& \mathrm{<span\; class="notranslate">\; 0.2242</span>}& \mathrm{<span\; class="notranslate">\; 0.3184</span>}& \mathrm{<span\; class="notranslate">\; 0.3556</span>}& \mathrm{<span\; class="notranslate">\; 0.3306</span>}& \mathrm{<span\; class="notranslate">\; 0.2476</span>}& \mathrm{<span\; class="notranslate">\; 0.1035</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.1444</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1407</span>}& \mathrm{<span\; class="notranslate">\; 0.2945</span>}& \mathrm{<span\; class="notranslate">\; 0.3174</span>}& \mathrm{<span\; class="notranslate">\; 0.4487</span>}& \mathrm{<span\; class="notranslate">\; 0.4117</span>}& \mathrm{<span\; class="notranslate">\; 0.3184</span>}& \mathrm{<span\; class="notranslate">\; 0.1769</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.1701</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1529</span>}& \mathrm{<span\; class="notranslate">\; 0.3064</span>}& \mathrm{<span\; class="notranslate">\; 0.4117</span>}& \mathrm{<span\; class="notranslate">\; 0.4062</span>}& \mathrm{<span\; class="notranslate">\; 0.2766</span>}& \mathrm{<span\; class="notranslate">\; 0.3306</span>}& \mathrm{<span\; class="notranslate">\; 0.1889</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1035</span>}& \mathrm{<span\; class="notranslate">\; 0.2593</span>}& \mathrm{<span\; class="notranslate">\; 0.3673</span>}& \mathrm{<span\; class="notranslate">\; 0.4287</span>}& \mathrm{<span\; class="notranslate">\; 0.3687</span>}& \mathrm{<span\; class="notranslate">\; 0.2827</span>}& \mathrm{<span\; class="notranslate">\; 0.1407</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1529</span>}& \mathrm{<span\; class="notranslate">\; 0.2476</span>}& \mathrm{<span\; class="notranslate">\; 0.2827</span>}& \mathrm{<span\; class="notranslate">\; 0.2593</span>}& \mathrm{<span\; class="notranslate">\; 0.1769</span>}& \mathrm{<span\; class="notranslate">\; 0.0336</span>}& \mathrm{<span\; class="notranslate">\; 0.0151</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0780</span>}& \mathrm{<span\; class="notranslate">\; 0.1160</span>}& \mathrm{<span\; class="notranslate">\; 0.0908</span>}& \mathrm{<span\; class="notranslate">\; 0.1553</span>}& \mathrm{<span\; class="notranslate">\; 0.1841</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0898</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right]$

According to the calculation result of sub-cell event occurrence probability calculated in step 3, the basic probability distribution function m _i (T)=Pr(ci ₎ of event occurrence in each sub-cell can be obtained from m _i (T)≡Pr(ci ₎ , Right now:

$\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left[\begin{array}{ccccccccccc}\mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0253</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0908</span>}& \mathrm{<span\; class="notranslate">\; 0.1888</span>}& \mathrm{<span\; class="notranslate">\; 0.2242</span>}& \mathrm{<span\; class="notranslate">\; 0.2007</span>}& \mathrm{<span\; class="notranslate">\; 0.1160</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.1671</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0651</span>}& \mathrm{<span\; class="notranslate">\; 0.2242</span>}& \mathrm{<span\; class="notranslate">\; 0.3184</span>}& \mathrm{<span\; class="notranslate">\; 0.3556</span>}& \mathrm{<span\; class="notranslate">\; 0.3306</span>}& \mathrm{<span\; class="notranslate">\; 0.2476</span>}& \mathrm{<span\; class="notranslate">\; 0.1035</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.14444</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1407</span>}& \mathrm{<span\; class="notranslate">\; 0.2945</span>}& \mathrm{<span\; class="notranslate">\; 0.3174</span>}& \mathrm{<span\; class="notranslate">\; 0.4487</span>}& \mathrm{<span\; class="notranslate">\; 0.4117</span>}& \mathrm{<span\; class="notranslate">\; 0.3184</span>}& \mathrm{<span\; class="notranslate">\; 0.1769</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.1701</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1529</span>}& \mathrm{<span\; class="notranslate">\; 0.3064</span>}& \mathrm{<span\; class="notranslate">\; 0.4117</span>}& \mathrm{<span\; class="notranslate">\; 0.4062</span>}& \mathrm{<span\; class="notranslate">\; 0.2766</span>}& \mathrm{<span\; class="notranslate">\; 0.3306</span>}& \mathrm{<span\; class="notranslate">\; 0.1889</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1035</span>}& \mathrm{<span\; class="notranslate">\; 0.2593</span>}& \mathrm{<span\; class="notranslate">\; 0.3673</span>}& \mathrm{<span\; class="notranslate">\; 0.4287</span>}& \mathrm{<span\; class="notranslate">\; 0.3687</span>}& \mathrm{<span\; class="notranslate">\; 0.2827</span>}& \mathrm{<span\; class="notranslate">\; 0.1407</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1529</span>}& \mathrm{<span\; class="notranslate">\; 0.2476</span>}& \mathrm{<span\; class="notranslate">\; 0.2827</span>}& \mathrm{<span\; class="notranslate">\; 0.2593</span>}& \mathrm{<span\; class="notranslate">\; 0.1769</span>}& \mathrm{<span\; class="notranslate">\; 0.0336</span>}& \mathrm{<span\; class="notranslate">\; 0.0151</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0780</span>}& \mathrm{<span\; class="notranslate">\; 0.1160</span>}& \mathrm{<span\; class="notranslate">\; 0.0908</span>}& \mathrm{<span\; class="notranslate">\; 0.1553</span>}& \mathrm{<span\; class="notranslate">\; 0.1841</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0898</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right]$

Calculate the confidence function of the occurrence of the sub-cell event on the specified observation time series by formula 5, the result of the confidence degree of the occurrence of the event on the (t ₁ , t ₂ ) time series calculated in this embodiment is as follows:

${\mathrm{<span\; class="notranslate">\; bel</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left[\begin{array}{ccccccccccc}\mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0253</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0908</span>}& \mathrm{<span\; class="notranslate">\; 0.1888</span>}& \mathrm{<span\; class="notranslate">\; 0.2242</span>}& \mathrm{<span\; class="notranslate">\; 0.2007</span>}& \mathrm{<span\; class="notranslate">\; 0.1160</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.1671</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0651</span>}& \mathrm{<span\; class="notranslate">\; 0.2242</span>}& \mathrm{<span\; class="notranslate">\; 0.3184</span>}& \mathrm{<span\; class="notranslate">\; 0.3556</span>}& \mathrm{<span\; class="notranslate">\; 0.3306</span>}& \mathrm{<span\; class="notranslate">\; 0.2476</span>}& \mathrm{<span\; class="notranslate">\; 0.1035</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.1444</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1407</span>}& \mathrm{<span\; class="notranslate">\; 0.2945</span>}& \mathrm{<span\; class="notranslate">\; 0.3174</span>}& \mathrm{<span\; class="notranslate">\; 0.4487</span>}& \mathrm{<span\; class="notranslate">\; 0.4117</span>}& \mathrm{<span\; class="notranslate">\; 0.3184</span>}& \mathrm{<span\; class="notranslate">\; 0.1769</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.1701</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1529</span>}& \mathrm{<span\; class="notranslate">\; 0.3064</span>}& \mathrm{<span\; class="notranslate">\; 0.4177</span>}& \mathrm{<span\; class="notranslate">\; 0.4062</span>}& \mathrm{<span\; class="notranslate">\; 0.2766</span>}& \mathrm{<span\; class="notranslate">\; 0.3306</span>}& \mathrm{<span\; class="notranslate">\; 0.1889</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1035</span>}& \mathrm{<span\; class="notranslate">\; 0.2593</span>}& \mathrm{<span\; class="notranslate">\; 0.3673</span>}& \mathrm{<span\; class="notranslate">\; 0.4287</span>}& \mathrm{<span\; class="notranslate">\; 0.3687</span>}& \mathrm{<span\; class="notranslate">\; 0.2827</span>}& \mathrm{<span\; class="notranslate">\; 0.1407</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1529</span>}& \mathrm{<span\; class="notranslate">\; 0.2476</span>}& \mathrm{<span\; class="notranslate">\; 0.2827</span>}& \mathrm{<span\; class="notranslate">\; 0.2593</span>}& \mathrm{<span\; class="notranslate">\; 0.1769</span>}& \mathrm{<span\; class="notranslate">\; 0.0336</span>}& \mathrm{<span\; class="notranslate">\; 0.0151</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0780</span>}& \mathrm{<span\; class="notranslate">\; 0.1160</span>}& \mathrm{<span\; class="notranslate">\; 0.0908</span>}& \mathrm{<span\; class="notranslate">\; 0.1553</span>}& \mathrm{<span\; class="notranslate">\; 0.1841</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0898</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right]$

In fact, multiple time intervals Δt may be selected for the observation time period in practical applications to improve the accuracy of the results.

On the basis of the above calculation results, the evidence conflict of each sub-cell event is calculated, and the evidence conflict factor of each sub-cell event is calculated by formula 6 as the following Con matrix:

$\mathrm{<span\; class="notranslate">\; Con</span>}<span\; class="notranslate">\; =</span>\left[\begin{array}{ccccccccccc}\mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0108</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0687</span>}& \mathrm{<span\; class="notranslate">\; 0.2067</span>}& \mathrm{<span\; class="notranslate">\; 0.2835</span>}& \mathrm{<span\; class="notranslate">\; 0.2100</span>}& \mathrm{<span\; class="notranslate">\; 0.0622</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.0244</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0439</span>}& \mathrm{<span\; class="notranslate">\; 0.2791</span>}& \mathrm{<span\; class="notranslate">\; 0.7289</span>}& \mathrm{<span\; class="notranslate">\; 1.1508</span>}& \mathrm{<span\; class="notranslate">\; 0.8706</span>}& \mathrm{<span\; class="notranslate">\; 0.3690</span>}& \mathrm{<span\; class="notranslate">\; 0.05757</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.0499</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1224</span>}& \mathrm{<span\; class="notranslate">\; 0.5649</span>}& \mathrm{<span\; class="notranslate">\; 1.1018</span>}& \mathrm{<span\; class="notranslate">\; 2.7480</span>}& \mathrm{<span\; class="notranslate">\; 1.8859</span>}& \mathrm{<span\; class="notranslate">\; 0.7699</span>}& \mathrm{<span\; class="notranslate">\; 0.1645</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0.0249</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1322</span>}& \mathrm{<span\; class="notranslate">\; 0.6898</span>}& \mathrm{<span\; class="notranslate">\; 2.0637</span>}& \mathrm{<span\; class="notranslate">\; 2.4614</span>}& \mathrm{<span\; class="notranslate">\; 0.9728</span>}& \mathrm{<span\; class="notranslate">\; 0.7662</span>}& \mathrm{<span\; class="notranslate">\; 0.2016</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.06027</span>}& \mathrm{<span\; class="notranslate">\; 0.4232</span>}& \mathrm{<span\; class="notranslate">\; 1.2422</span>}& \mathrm{<span\; class="notranslate">\; 2.2878</span>}& \mathrm{<span\; class="notranslate">\; 1.1235</span>}& \mathrm{<span\; class="notranslate">\; 0.4703</span>}& \mathrm{<span\; class="notranslate">\; 0.1275</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.1112</span>}& \mathrm{<span\; class="notranslate">\; 0.3726</span>}& \mathrm{<span\; class="notranslate">\; 0.5565</span>}& \mathrm{<span\; class="notranslate">\; 0.4939</span>}& \mathrm{<span\; class="notranslate">\; 0.2576</span>}& \mathrm{<span\; class="notranslate">\; 0.0272</span>}& \mathrm{<span\; class="notranslate">\; 0.0026</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0.0411</span>}& \mathrm{<span\; class="notranslate">\; 0.0846</span>}& \mathrm{<span\; class="notranslate">\; 0.0767</span>}& \mathrm{<span\; class="notranslate">\; 0.1170</span>}& \mathrm{<span\; class="notranslate">\; 0.0697</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right]$

The event occurrence area can be detected by the set Con threshold, that is, if Con _i >Con _th , all sub-cells c _i meeting the condition constitute the event occurrence area. In this embodiment, if Con _th =0.5, the detected event area is as shown in FIG. 6 .

When each monitoring cycle expires or when the sensing data exceeds the threshold, the above process will be triggered and the event occurrence area will be detected and updated in real time until the end of the monitoring task.

During concrete implementation, the present invention also can adopt software modularization technology to realize, as shown in Figure 7, can comprise the following parts:

(1) The network scene initialization and maintenance module of the vehicle sensor is used to model the monitored urban road area and the wireless self-organizing vehicle network composed of vehicles and updates, and obtain the sub-cell information and network topology of the monitoring area division Basic Information.

(2) The event monitoring probability module of the sub-cells is used to process the road sub-cell divisions obtained from the initialization of the vehicle sensor network scene and the sensor data collected by the vehicle nodes in each sub-cell, and calculate the behavior change factor and data change rate of the vehicle nodes, On this basis, the event monitoring probability of each sub-cell is obtained.

(3) The sub-cell event probability module is used to obtain the event occurrence probability of each road sub-cell according to the obtained event monitoring probability from the vehicle node to the sub-cell and the D-S evidence theory.

(4) The event occurrence confidence module is used to obtain the event occurrence confidence function result of the sub-cell in the time series according to the calculation result of the event occurrence probability function of the sub-cell in the continuous observation time series.

(5) Evidence merging conflict calculation module, which is used to merge the evidence obtained by each adjacent sub-cell, calculate its conflict attribute value, classify the degree of difference after the information evidence is merged, and obtain the conflict value of the merged evidence.

(6) The event area determination module is used to classify the evidence according to the conflict value calculated after the evidence is merged, determine that an event has occurred in the sub-cells greater than or equal to the conflict threshold, and obtain the event area composed of the sub-cell sequence.

(7) The event detection trigger module is used to trigger the event detection process according to the network cycle timer and the monitoring perception data when the timing period or the perception data exceeds the set threshold, and re-detect the event area.

Among them, the road modeling and network topology parameter sets output by the vehicle sensor network scene initialization and maintenance module are input to the event monitoring probability module and event detection trigger module of the sub-cell; the event detection module decides when to Start to use the method proposed in this paper to detect the event and its occurrence area; the event detection module uses the periodic detection mechanism of timing detection to trigger the event detection, or triggers the data collected by the sensor in the network and sets the threshold, that is: when the network If the sensor data of some areas in the road exceeds the preset safety value, the event detection process of the road area is started. The event monitoring probability from the vehicle node to each sub-cell calculated by the event monitoring probability module of the sub-cell is input to the event occurrence probability module of the sub-cell; the event occurrence probability of each sub-cell calculated by the event probability module is input into the event occurrence confidence degree module; the confidence degree of event occurrence on the observed time series of each sub-district calculated by the event occurrence confidence degree module is input to the evidence merge conflict calculation module; the evidence conflict value calculated by the evidence merge conflict calculation module is input to the event area judgment module , used to detect event regions. The event detection trigger command output by the event detection trigger module is input to the event monitoring probability module of the sub-cell for re-executing the event detection process.

Claims (2)

1. A vehicle-mounted sensor network event area detection method based on D-S evidence theory, is characterized in that: the urban road area of event monitoring is divided into some sub-districts, the vehicle that is provided with vehicle-mounted sensor in the urban road area of event monitoring passes through Networking in the form of organization, establishing a vehicle sensor network graph, in which vehicles equipped with vehicle sensors form vehicle nodes, and adjacent vehicle nodes are edges; when performing event area detection, perform the following steps: Step 1. From the distance from each vehicle node to the center of the sub-district, calculate the monitoring weight of each vehicle node to the sub-district; calculate the behavior factor of the vehicle node according to the direction and speed of the vehicle node; Calculate the rate of change of the perceived data based on the historical data of the perceived physical quantity of the community; and obtain the event occurrence probability of the sub-district where it is located according to the calculation results; the implementation method is as follows, Step a, set the geometric distance from the coordinates (x _nk , y _nk ) of a certain vehicle node k to the coordinates (x _ci , y _ci ) of the center point of the sub-cell _ci where it is located, denoted as d _k,i =||( x _nk ,y _nk )-(x _ci ,y _ci )||, the monitoring weight w( _nk ,ci ) from _the vehicle node k to the sub-cell _ci where it is located is calculated by the following formula, $<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\frac{{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; r</span>}}& {\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; r</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& {\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; r</span>}\end{array}\right.$ Among them, r is the monitoring perception radius of the vehicle sensor; Step b, set the observed time series as (t,t'), the velocity of vehicle node k at time t is v _k , the velocity of vehicle node k at time t' is v _k' , max(v _k ,v _k' ) is the larger value of the velocity v _k and v _k' , if max(v _k ,v _k' )=0, the behavior factor of vehicle node k μ _k =0; if max(v _k ,v _k' )≠0 , the behavior factor μ _k of vehicle node k is calculated by the following formula, Among them, α is the weight parameter, θ is the rate vector change to direction angle; Step c, the vehicle node k obtains the historical data of the perceived physical quantity of the observation time series (t, t') of the sub-cell c _i where it is located, the time interval of the observation time series (t, t') is recorded as Δt, and the time interval Δt The average value ave(p,Δt) of a certain physical quantity p=sum(p_data)/N(Δt), where sum(p_data) is the sum of the values obtained from monitoring the physical quantity p by sub-cells c _i within the time interval Δt, and N(Δt) is Δt The number of sensor data monitoring times; The larger value between the average value ave(p,Δt) of the physical quantity p in the time interval Δt and the real-time monitoring value p_data(t') of the physical quantity p at the current moment t' is recorded as max(ave(p,Δt),p_data( t')), if max(ave(p,Δt),p_data(t'))=0, data change rate γ _ci =0, if max(ave(p,Δt),p_data(t'))≠0 , the data change rate γ _ci is calculated according to the following formula, ${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; ci</span>}}<span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; \_</span>\mathrm{<span\; class="notranslate">\; data</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; ave</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&Delta;t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}{\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; \_</span>\mathrm{<span\; class="notranslate">\; data</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; ave</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&Delta;t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}$ Step d, according to the monitoring weight w( _nk , _ci ) obtained in step a, the behavior factor μ _k obtained in step b, and the data change rate γ _ci obtained in step c, the event occurrence probability Pr(ci ₎ in the sub-cell c _i is obtained as follows: $\mathrm{<span\; class="notranslate">\; PR</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&lambda;</span>}\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; K</span>}}<span\; class="notranslate">\; *</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; K</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; c</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&lambda;</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}_{\mathrm{<span\; class="notranslate">\; ci</span>}}$ Among them, λ is an adjustment factor, which is a constant between 0 and 1; K is the maximum number of vehicle nodes in the sub-cell c _i , and the value of k is 1, 2, ..., K; Step 2. According to the D-S evidence theory, set the basic probability distribution function of each sub-cell event; the implementation method is as follows, Let T represent that there is an event in a certain sub-cell, and F represent that there is no event in a certain sub-cell, then the target recognition framework is expressed as Θ = {T, F}, and the total state set in the target recognition framework is 2 ^Θ = {Ф, T, F, Θ={T, F}}, use m _i (T) to represent the basic probability assignment of sub-cell c _i "has an event" state, m _i (F), _mi (Θ) represent sub-cell c The basic probability assignment of "no event" and "uncertain" state in _i ; The basic probability distribution function m _i (T) of the sub-cell c _i is set according to the event generation probability Pr(c _i ) calculated in step 2, that is, m _i (T)≡Pr(c _i ); Step 3. According to the DS evidence theory, calculate the confidence function value Bel _i (T) of the event generated by each sub-cell on the observed time series (t, t'); the implementation method is as follows, From the basic probability distribution function m _i (T) obtained in step 2, calculate the confidence function value Bel _i (T) of the event generated by the sub-cell c _i on the observation time series (t, t') according to the following formula, Bel _i (T) = m _i (T) Step 4. According to the D-S evidence theory, merge the confidence function values of each sub-cell and adjacent sub-cells to obtain the conflict value between the merged event evidence; the implementation method is as follows, Assume that the neighboring sub-cells adjacent to each direction of the sub-cell c _i form a set SN, and the confidence function value Bel _i (T) obtained in step 3 is calculated as follows: Con(Bel _i ,Bel _{neigh(i )} ), to get the conflict value between the merged event evidence: Con(Bel _i ,Bel _neigh(i) )=log(1/η), $\mathrm{\&eta;}=\underset{1}{\overset{\left|\mathrm{SN}\right|}{\mathrm{\&Sigma;}}}{\mathrm{bel}}_i\left(T\right)*{\mathrm{bel}}_{\mathrm{neigh}\left(i\right)}\left(T\right)$ Among them, Bel _neigh(i) (T) is the confidence function value of each sub-cell neigh(i) adjacent to sub-cell ci, neigh(i)∈SN; |SN| is the number of sub-cells in the set SN ; Step 5, judging whether each sub-cell belongs to the event area, the implementation method is as follows, If the conflict Con(Bel _i ,Bel _neigh(i) ) between the merged event evidence obtained in step 4 is greater than or equal to the preset threshold Con _th , then it is determined that the sub-cell c _i is in the event area, otherwise it is determined that the sub-cell c _i is in the non-event area area. 2. The vehicle-mounted sensor network event area detection method based on D-S evidence theory as claimed in claim 1, characterized in that: the event area detection is performed periodically, or triggered when the physical quantity update data collected by the vehicle sensor exceeds a certain preset value Event area detection.