WO2023045062A1

WO2023045062A1 - Method and apparatus for dividing time periods, and electronic device and storage medium

Info

Publication number: WO2023045062A1
Application number: PCT/CN2021/133148
Authority: WO
Inventors: 凌玮岑; 梅雨
Original assignee: 阿波罗智联(北京)科技有限公司
Priority date: 2021-09-27
Filing date: 2021-11-25
Publication date: 2023-03-30
Also published as: CN113851007B; CN113851007A

Abstract

A method and apparatus for dividing time periods, and an electronic device and a storage medium, which are applied to the field of artificial intelligence, and specifically relate to the field of intelligent transportation and the field of big data. A specific implementation solution of the method for dividing time periods is: on the basis of historical traffic data of an intersection, determining respective traffic values of n unit time periods that are comprised in a predetermined time period (S210); on the basis of the traffic values, dividing the n unit time periods by using an alternative division solution, so as to obtain alternative time period groups (S220); and determining a target time period group from among the alternative time period groups on the basis of the group similarities between the alternative time period groups and existing time period groups that are divided by using an existing division solution (S230), wherein n is an integer greater than 1.

Description

Method, device, electronic device and storage medium for dividing time periods

This application claims priority to a Chinese patent application with application number 202111139260.3 filed on September 27, 2021, the entire contents of which are incorporated herein by reference.

technical field

The present disclosure relates to the field of artificial intelligence, specifically to the field of intelligent transportation and big data, and more specifically to a method, device, electronic device and storage medium for dividing time periods.

Background technique

With the development of network technology, intelligent transportation has developed rapidly. As a branch of intelligent transportation, the time-segmented control of traffic signals at intersections is particularly important for alleviating traffic pressure and making full use of traffic resources. In order to realize the precise time-segmented control of traffic signals at intersections, it is necessary to improve the rationality and accuracy of time-segmentation.

Contents of the invention

Provided are a method, device, electronic equipment and storage medium for dividing time periods.

According to one aspect of the present disclosure, there is provided a method for dividing time periods, including: based on historical traffic data at intersections, determining the respective flow values of n unit periods included in a predetermined time period; based on the flow values, using an alternative division scheme to divide n unit time period, to obtain the alternative time period group; and based on the group similarity between the alternative time period group and the existing time period group obtained by using the existing division scheme, determine the target time period group in the alternative time period group, where n is an integer greater than 1.

According to another aspect of the present disclosure, there is provided a device for dividing periods, including: a flow value determination module, configured to determine the respective flow values of the n unit periods included in the predetermined period based on historical traffic data at intersections; alternative determination A module for dividing n unit time periods by using an alternative division scheme based on the flow value to obtain an alternative time period group; and a target group determination module for dividing an existing time period based on the alternative time period group and the existing division scheme The group similarity between groups determines the target time period group in the candidate time period group, where n is an integer greater than 1.

According to another aspect of the present disclosure, an electronic device is provided, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are Execution by at least one processor, so that at least one processor can execute the method for dividing time periods provided in the present disclosure.

According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method for dividing time periods provided in the present disclosure.

According to another aspect of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the method for dividing time periods provided in the present disclosure is implemented.

It should be understood that what is described in this section is not intended to identify key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will be readily understood through the following description.

Description of drawings

The accompanying drawings are used to better understand the present solution, and do not constitute a limitation to the present disclosure. in:

FIG. 1 is a schematic diagram of an application scenario of a method and device for dividing time periods according to an embodiment of the present disclosure;

FIG. 2 is a schematic flowchart of a method for dividing time periods according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of the principle of dividing n unit time periods by using an alternative division scheme to obtain an alternative time period group according to an embodiment of the present disclosure;

Fig. 4 is a schematic diagram of the principle of determining the loss with the jth unit period as an alternative division point according to an embodiment of the present disclosure;

Fig. 5 is a schematic diagram of the principle of determining a target time period group in an alternative time period group according to an embodiment of the present disclosure;

FIG. 6 is a structural block diagram of an apparatus for dividing time periods according to an embodiment of the present disclosure; and

FIG. 7 is a block diagram of an electronic device used to implement the method for dividing time periods according to an embodiment of the present disclosure.

Detailed ways

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and they should be regarded as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

The present disclosure provides a method for dividing time periods, which includes a flow determination phase, an alternative determination phase, and a target group determination phase. In the flow determination stage, based on the historical flow data of the intersection, the respective flow values of n unit periods included in the predetermined period are determined, wherein n is an integer greater than 1. In the alternative determination stage, based on the flow value, an alternative division scheme is used to divide n unit time periods to obtain an alternative time period group. In the stage of determining the target group, based on the group similarity between the candidate time period group and the existing time period group divided by the existing division scheme, the target time period group in the candidate time period group is determined.

The application scenarios of the method and device provided by the present disclosure will be described below with reference to FIG. 1 .

Fig. 1 is an application scenario diagram of a method and device for dividing time periods according to an embodiment of the present disclosure.

As shown in FIG. 1 , an application scenario 100 of this embodiment includes a monitoring personnel 110 , a terminal 120 , a communication base station 130 and a road traffic network. The road traffic network may include roads and intersections formed by intersections of roads. For example, in the scene 100 , the road traffic network includes at least an intersection 140 and

roads

151 , 152 , and 153 that meet to form the intersection 140 .

Exemplarily, as shown in FIG. 1 , there may be vehicles running on a road that merges to form an intersection 140 , and the vehicles may pass through the intersection according to the indication of a traffic light 160 .

In an embodiment, the application scene 100 can also set a video capture device at the intersection 140 . The video acquisition device may be a camera, which is used to photograph vehicles driving on the road. The video collection device can send the captured video to the terminal 120 via the communication base station 130 . In this way, the monitoring personnel can monitor the road conditions in real time. Alternatively, the terminal 120 may store the received video in a database. The terminal 120 can also periodically process the video stored in the database to obtain the traffic flow at the intersection in the historical period.

In an embodiment, most of the vehicles driving on the roads that meet to form an intersection may have integrated navigation systems, or the hand-held terminals of the drivers in the vehicles may be installed with navigation applications. The navigation system or navigation application can upload real-time data to the terminal 120 through the communication base station 130 . The terminal 120 can, for example, analyze the traffic flow at the intersection according to the real-time data. Alternatively, the terminal may store real-time data in a database. The terminal 120 can also regularly acquire historical data from the database, and obtain the traffic flow at the intersection in the historical period by analyzing the historical data.

In one embodiment, the terminal 120 can determine multiple time periods in a day for signal control of the traffic light 160 according to the analyzed traffic flow within a predetermined time period, and set a reasonable time period for each time period in the multiple time periods. control information, so as to obtain the timing scheme of the traffic light 160. Wherein, the control information may include, for example, the signal period and green signal ratio of the traffic light 160 .

In one embodiment, the terminal 120 can also communicate with the traffic light 160 through the network, so as to control the operation of the traffic light 160 according to the timing scheme, so that the operation of the traffic light 160 can better meet the needs of vehicles on the road. demand and effectively alleviate road congestion.

It should be noted that the method for dividing time periods provided in the present disclosure may generally be executed by the terminal 120 . Correspondingly, the device for dividing time periods provided in the present disclosure may be set in the terminal 120 . Alternatively, the method for dividing time periods provided in the present disclosure may also be executed by a server connected in communication with the terminal 120 . Correspondingly, the device for dividing time periods provided in the present disclosure may be set in a server connected in communication with the terminal 120 .

It should be understood that the numbers and types of terminals, roads, vehicles, communication base stations, and signal indicator lights in FIG. 1 are only schematic. According to the implementation requirements, there can be any number and types of terminals, roads, vehicles, communication base stations and signal lamps.

The method for dividing time periods provided by the present disclosure will be described in detail below with reference to FIG. 1 and FIG. 2 to FIG. 5 .

Fig. 2 is a schematic flowchart of a method for dividing time periods according to an embodiment of the present disclosure.

As shown in FIG. 2 , the method 200 for dividing time periods in this embodiment may include operation S210 to operation S230.

In operation S210, based on the historical traffic data of the intersection, the respective traffic values of n unit periods included in the predetermined period are determined.

According to an embodiment of the present disclosure, the length of the predetermined time period may be one day, that is, the predetermined time period is from 0:00 to 24:00. Alternatively, the predetermined time period may also be a time period that is more sensitive to the timing scheme, such as 6:00 to 22:00. Alternatively, the predetermined period may also be a certain sub-period in a longer period belonging to a certain type, so as to improve the accuracy of period division. For example, a certain type may be a weekday type, a holiday type, a holiday day type, a weekend type, and the like. For example, the predetermined time period may be an all-day period on weekdays, or may be an all-day period on holidays, weekends and the like.

According to an embodiment of the present disclosure, the historical traffic data of the intersection may include, for example, traffic data of one or more historical time periods corresponding to a predetermined time period. For example, if the predetermined time period is the whole day period of a Monday that belongs to a working day, the historical traffic data of the intersection may include the traffic data of one or more working days before the current date, and may also include one or more Mondays before the current date traffic data.

According to an embodiment of the present disclosure, the historical traffic data of the intersection may include the traffic data of each road in each unit period among multiple roads that meet to form the intersection. The flow data may include, for example, vehicle flow data, pedestrian flow data, non-motor vehicle flow data, and the like. Wherein, the length of the unit time period may be, for example, 5 min, 10 min, 15 min, or 30 min, which is set according to actual needs, which is not limited in the present disclosure. The historical traffic data can be obtained from the interface provided by the map navigation system or the data interface provided by the traffic control department.

After obtaining the historical flow data of the intersection, the average value of the flow data of all roads at the intersection in each unit period can be taken as the flow value of the intersection in each unit period. Or, if the historical traffic data of the intersection includes the traffic data of multiple historical periods corresponding to the predetermined time period, the traffic value of an intersection in each unit period can be obtained for each historical period, and for each unit period, the total can be obtained. flow value. Then, the average value of the plurality of flow values for each unit period is used as the flow value for each unit period.

According to an embodiment of the present disclosure, after obtaining the flow values of n unit periods, for example, unit conversion may be performed on the flow values of each unit period to obtain flow values in units of number/h. Wherein, n is an integer greater than 1.

In operation S220, based on the traffic value, the n unit time periods are divided using an alternative division scheme to obtain an alternative time period group.

According to an embodiment of the present disclosure, the alternative division scheme may be set with, for example, the number of time periods obtained by dividing n unit time periods. Alternatively, the alternative division scheme may set a threshold value for the number of unit periods included in each divided period. For example, the number of unit periods included in each period in the divided periods should be greater than a first predetermined value, and/or the number of unit periods included in each period in the divided periods should be smaller than a second predetermined value. Alternatively, the alternative division scheme may be set such that the difference of flow values between the unit periods included in each period cannot be greater than the difference threshold, so as to ensure the rationality of the divided alternative period groups. It can be understood that one or more division conditions may be set in the above alternative division scheme, and the number and type of the division conditions may be set according to actual requirements, which is not limited in the present disclosure.

According to an embodiment of the present disclosure, the alternative division scheme may also be set by the staff according to the obtained flow values of n unit periods.

By dividing n unit time periods by using an alternative division scheme, for example, one or more alternative time period groups can be obtained. Each candidate time period group includes a plurality of time periods, and each time period includes at least one unit time period.

Exemplarily, the n unit periods may be numbered according to chronological order to obtain n unit periods numbered 1 to n respectively. Multiple time periods included in each candidate time period may be arranged in chronological order. For two adjacent time periods in each candidate time period, the maximum code of at least one unit period included in the time period ranked earlier may be used as the identifier I _i representing the division point of the two adjacent time periods. In this way, an alternative time period group may include (k-1) dividing points, k is the number of multiple time periods included in the one candidate time period, and i is a value greater than or equal to 1 and less than or equal to (k-1) .

In operation S230, a target time period group in the candidate time period group is determined based on the group similarity between the candidate time period group and the existing time period group divided by using the existing division scheme.

According to an embodiment of the present disclosure, the existing division scheme may include a period division scheme in the timing scheme currently adopted by the traffic lights. The existing time period group divided by the existing division scheme also includes a plurality of time periods. Based on a principle similar to the principle that the candidate time period group includes division points, the existing division scheme also includes at least one division point.

According to an embodiment of the present disclosure, the group similarity between the candidate time period group and the existing time period group may be determined according to the distance between the division points included in the candidate time period group and the division points included in the existing time period group. In the case that the candidate time period group or the existing time period group includes multiple division points, the distance between each division point in the candidate time period group and each division point in the existing time period group may be calculated pair by pair. Subsequently, the group similarity is determined based on the average of all distances obtained. For example, the average value of the distance can be normalized, and the value obtained by the normalization can be used as the group similarity.

According to an embodiment of the present disclosure, if there are multiple candidate time period groups, the candidate time period group with the largest group similarity with the existing time period group may be selected from the multiple candidate time period groups as the target time period group. Multiple time periods in the target time period group are the division results of the method for dividing time periods.

Therefore, in the method of the embodiment of the present disclosure, by referring to the historical traffic data to determine the candidate time period group for screening, and by considering the similarity between the candidate time period group and the existing time period group to determine the target time period group as the division result, it can Effectively avoid missing the suboptimal solution that is close to the original time division scheme. Therefore, the accuracy and rationality of time division can be improved, and favorable conditions can be provided for alleviating traffic pressure and improving the utilization rate of traffic resources.

Fig. 3 is a schematic diagram of the principle of dividing n unit time periods by adopting an alternative division scheme to obtain an alternative time period group according to an embodiment of the present disclosure.

According to an embodiment of the present disclosure, the alternative division scheme may, for example, provide a plurality of alternative numbers, where the alternative number is the number of time periods obtained by dividing n unit time periods. In this embodiment, one candidate period group can be determined for each candidate number. In this way, when determining the target time period group, both the number of divided time periods and the division scheme can be selected. Therefore, the accuracy of the obtained target period group can be further improved.

For example, if the number of candidates is set to be k, then for each number of candidates, the n unit time periods described above can be divided into k time periods based on the flow value, so as to obtain a group of time periods for each candidate number. Wherein, k is an integer greater than or equal to 1 and less than or equal to n.

In an embodiment, a total of n integers from 1 to n may be used as n candidate numbers, and one candidate period group is obtained for each candidate number. In this way, the candidate time period group can cover the number of time periods that may be obtained in all division situations.

According to an embodiment of the present disclosure, when each candidate number of candidate time period groups is determined, according to the distribution of time periods in the time period groups, losses may be assigned to each time period group obtained by dividing n unit periods based on flow values. In this way, if n unit time periods are divided into k time periods to obtain p time period groups, then p losses can be obtained. Finally, the period group corresponding to the smallest loss among the p losses is used as the candidate period group for the candidate number k. In this way, the accuracy of the obtained candidate period groups can be improved.

Exemplarily, the loss allocated to each period group may be determined according to the deviation of dividing a unit period into one period. For example, for each time period in each time period group, the average value of the numbers of the unit time periods included in each time period may be calculated first. Then calculate the sum of the squares of the differences between the numbers of all unit periods included in each period and the average value. The sum of squares is used as the deviation for each period, that is, the unit period with the largest number in each period is used as the loss of the dividing point. Finally, the biases of all time periods in each time period group can be summed to obtain the assigned loss for each time period group.

In an embodiment, a dynamic programming algorithm or the like may be used to determine each candidate number of candidate period groups.

For example, as shown in FIG. 3 , in this embodiment 300 , after the respective flow values 301 of n unit periods are obtained, each candidate number of candidate period groups can be determined through operations S310 to S360 .

For example, in the embodiment 300, j may be assigned as k first. Subsequently, operation S310 is performed to determine the minimum loss for dividing the first (j-1) unit periods among the n unit periods into (k-1) periods as the first loss.

According to an embodiment of the present disclosure, when j is k, there is only one division scheme for dividing the previous (j-1) unit periods into (k-1) periods, then the first loss can be determined as dividing the previous Each of the (j-1) unit periods is divided into a period of loss. The first loss may be a deviation of dividing each unit period into a period.

While performing operation S310, operation S320 may be performed to determine a loss of dividing the last (n-j+1) unit periods of the n unit periods into one period as the second loss. Operation S320 may also be performed before operation S310, or may be performed after operation S310, which is not limited in the present disclosure.

According to an embodiment of the present disclosure, the second loss may be determined by using a deviation that divides the last (n-j+1) unit periods into one period. For example, the average value of (n-j+1) flow values of the next (n-j+1) unit periods may be determined first. Then, based on the respective deviations of the (n-j+1) flow values relative to the average value, the loss of dividing the next (n-j+1) unit periods into one period is determined.

Wherein, the deviations of the (n-j+1) flow values relative to the average value can be, for example, the square of the difference between the (n-j+1) flow values and the average value, or can be the The absolute value of the difference between the (n-j+1) flow values and the mean value. After obtaining the respective deviations of the (n-j+1) flow values relative to the average value, the sum of the deviations or the average value of the deviations can be used as the second loss.

For example, the i-th unit period to the j-th unit period arranged in chronological order is divided into a deviation of one period, for example, the following formula (1) can be used to express:

Among them, X _i is the flow value of the i-th unit period,

is the average value from the flow value of the i-th unit period to the flow value of the j-th unit period.

After performing operation S310 and operation S320, operation S330 may be performed, based on the first loss and the second loss, to determine a loss using the jth unit period among the n unit periods as a candidate division point.

According to an embodiment of the present disclosure, the sum of the first loss and the second loss may be used as the loss of the jth unit period as the candidate division point. Through this operation S330, the last division point for dividing n unit periods into k periods may be determined. Through the process of determining the first loss in operation S310, the first (k-2) dividing points for dividing n unit periods into k periods may be determined. In this way, through operations S310 to S330, a division scheme for dividing n unit time periods into k time periods can be obtained.

After operation S330 is performed, operation S340 may be performed first to determine whether j is equal to n. In this embodiment, k is used as the initial value of j, because in all schemes of dividing n unit periods into k periods, the most front position of the last division point of k periods is (k-1) . Therefore, if j is less than n, it means that all cases of the last division point of k time periods have not been searched. Then set j to j+1 through operation S350, and return to perform operation S310. That is to say, it is equivalent to moving the last division point of the k time periods backward by one unit time period, and continuing to search for unsearched situations.

If j is equal to n, it means that all cases of the last division point of k periods have been searched, and a total of (n-k+1) losses can be obtained. In this way, operation S360 may be performed to divide n unit time periods into k time periods based on the candidate division point corresponding to the smallest loss among (n-k+1) losses. According to the full text description, for each j, a division scheme can be obtained, and for each division scheme, a loss with the jth unit period as an alternative division point will be obtained. Therefore, after determining the minimum loss among the (n-k+1) losses, the optimal division scheme for the candidate number k can be obtained, and this embodiment will adopt the optimal division scheme for the candidate number k to obtain The period group of is used as the alternative period group for each alternative number.

According to an embodiment of the present disclosure, when j is greater than k, there may be multiple division schemes for dividing (j-1) unit time periods into (k-1) time periods, so that multiple division results may be obtained. In this embodiment, losses are allocated to each division result based on the foregoing principle of allocating losses to each candidate period group. Finally, the smallest loss among the multiple losses of the multiple division results is taken as the first loss.

According to an embodiment of the present disclosure, when the number of alternatives is 2, the implementation principle of the above operation S310 to operation S360 can refer to the following formula (2):

minL[c(n, 2)] = min _{j∈[2, n]} {SubL(1, j-1)+SubL(j, n)}; formula (2)

Among them, SubL(1, j-1) is the loss of dividing the first (j-1) unit periods into one period, and SubL(j, n) is the division of the j-th unit period to the n-th unit period into one time period loss. The SubL(1, j-1) and SubL(j, n) can be calculated by using the formula (1) described above. L[c(n, 2)] is the loss for dividing n unit periods into two periods.

By analogy, when the number of alternatives is k, the implementation principles of the above operation S310 to operation S360 can refer to the following formula (3):

minL[c(n,k)]=min _j∈[k,n] {minL[c(j-1,k-1)]+SubL(j,n)}; formula (3)

Among them, minL[c(j-1, k-1)] is the loss of dividing the previous (j-1) unit periods into (k-1) periods. Among them, in the case of k=2, since there is only one solution for dividing the first (j-1) unit periods into one period, in this case, formula (3) and formula (2) are equivalent formulas .

Fig. 4 is a schematic diagram of the principle of determining the loss with the jth unit period as an alternative dividing point according to an embodiment of the present disclosure.

According to an embodiment of the present disclosure, when determining the loss of the j-th unit period in the n unit periods as an alternative division point, for example, when the j-th unit period is used as an alternative division point, the alternative The similarity between the partition point and the existing partition scheme. In this way, the candidate time period group determined for each candidate number can be as close as possible to the existing time period group. And thus can further increase the probability of finding a suboptimal solution close to the original time division scheme.

For example, as shown in FIG. 4 , in this embodiment 400, when determining the loss 410 with the jth unit period as an alternative division point, in addition to considering the first loss 420 and the second loss 430, the jth unit period can also be considered The second similarity 440 between the unit time period and the existing time period group.

Wherein, the second similarity 440 may be determined, for example, according to the difference between the candidate division point and each division point in the existing time period group when the jth unit period is used as the candidate division point.

For example, using I _j 401 to represent the j-th unit period as an alternative division point, set each division point in the existing period group as

Then you can first determine I _j 401 and

The difference between, get a difference dif ₁ 404, and so on, determine I _j 401 with

The difference between, get a difference dif _m 405, get m differences in total. Then based on the m differences, a second degree of similarity is determined. Among them, m is the total number of dividing points in the existing time period group.

For example, the sum of the m differences may be used as the second similarity. Alternatively, the average of the m differences may be used as the second similarity. Alternatively, the largest difference among the m differences may be used as the second similarity.

Exemplarily, the following formula (4) may be used to determine the difference between the candidate division point and the division points in the existing time period group. And use the following formula (5) to determine the second similarity.

Among them, s(I _j , C ^o ) represents the similarity between the candidate division point I _j and the existing period group C ^o ,

Indicates the i-th dividing point in the existing period group C ^o .

According to an embodiment of the present disclosure, when determining the difference between the candidate division point and each division point in the existing period group, for example, a threshold may be set for the difference. In this way, the situation that the reference value for the similarity is low due to the large difference between the candidate division point and the previous division point in the existing time period group is avoided. Therefore, the accuracy of the determined second similarity and the loss of the jth unit period as a candidate division point can be improved.

For example, when determining the difference between the candidate division point and each division point in the existing period group, the absolute value of the difference between the candidate division point and each division point in the existing period group may be determined first, Get the first difference. For example, the first difference can be calculated by using the formula (4) described above. Then, the absolute value of the difference between the first difference and a predetermined threshold is determined to obtain a second difference. Finally, the smaller of the second difference and a predetermined threshold is determined as the difference between the candidate division point and each division point in the existing time period group.

For example, the following formula (6) can be used to calculate the difference between the candidate division point and the i-th division point in the existing time period group:

Wherein, T is a predetermined threshold, which may be, for example, 4 or other value set according to actual needs, which is not limited in the present disclosure.

Fig. 5 is a schematic diagram of the principle of determining a target time period group in an alternative time period group according to an embodiment of the present disclosure.

According to an embodiment of the present disclosure, before selecting a target time period group from multiple candidate time period groups, the group similarity between each time period group in the candidate time period groups and an existing time period group may be determined first.

Wherein, the similarity between each time period group and the existing time period group may be determined according to the similarity between each dividing point in each time period group and the existing time period group. For example, the similarity between each dividing point in each time period group and an existing time period group may be determined first to obtain at least one first similarity. The method for calculating the first similarity may adopt a method similar to the method for calculating the second similarity described above, which will not be repeated here. The number of the first similarities is the same as the number of time periods in each time period group.

Wherein, the average value of the at least one first similarity can be used as the group similarity between each time period group and the existing time period group. Alternatively, the average value of the at least one first similarity relative to the number of target time periods may be used as the group similarity between each time period group and an existing time period group. Wherein, the target number of time periods is the greater value between the number of time periods in each time period group and the number of time periods in an existing time period group. Through the process of calculating the average value, the comparability between different time period groups can be improved, and it is convenient to select the target time period group.

For example, the group similarity between each time period group and an existing time period group can be calculated using the following formula (7):

Among them, S(C, C ^o ) is the group similarity, k is the number of time periods in each time period group, k ^o is the number of time periods in the existing time period group. s(I _i , C ^o ) represents the similarity between the i-th dividing point in each time period group and the existing time period group.

Exemplarily, after the group similarity is obtained in the above manner, the group similarity can also be standardized to a value interval [0, 1], for example, through linear transformation.

According to an embodiment of the present disclosure, when determining the target time period group in the candidate time period group, in addition to considering the group similarity with the existing time period group, for example, the grouping deviation can also be considered. And based on the grouping deviation and group similarity, the grouping loss of each candidate time period group is determined. Finally, according to the grouping loss, the target period group is selected from the candidate period groups. In this way, the accuracy of the obtained target period group can be improved.

For example, as shown in Figure 5, in this embodiment 500, the alternative time period groups can be set to include time period group C ₁ 501 to time period group C _n 502, and for each time period group in the n time period groups, a grouping can be obtained respectively Deviation, so as to obtain the grouping deviation 511~grouping deviation 512. At the same time, by using the method for calculating group similarity described above, the group similarity between each time period group and the existing time period group C ^o 503 can be calculated, and the group similarity 521-522 can be obtained. Then, according to the group similarity 521 and the group deviation 511 , the group loss 531 of the period group C ₁ 501 can be obtained. By analogy, according to the group similarity 522 and the group deviation 512, the group loss 532 of the period group C _n 502 can be obtained. Finally, the target period group 540 in the period group C ₁ 501 -the period group C _n 502 can be determined according to the group loss 531 -the group loss 532 . For example, the target time period group 540 may be determined by comparing the magnitude relationship between the packet loss 531 to the packet loss 532 . For example, if the packet loss of time period group C ₁ 501 is the smallest, then this time period group C ₁ 501 may be determined as the target time period group 540 .

According to an embodiment of the present disclosure, the grouping deviation may be determined, for example, according to a sum of losses of dividing each time period in each time period group into one time period. In an embodiment, the grouping deviation can be represented by the ratio between the sum of the loss sum of dividing each time period in each time period group into one time period and the unsegmented loss when n unit time periods are not divided, so as to further In order to objectively reflect the relative size relationship of the group losses in different time periods.

For example, the formula (1) described above can be used to calculate the loss that each time period in each time period group is divided into one time period, and for each time period group, at least one segmentation loss can be obtained. In the case where at least one segment loss is multiple, the following formula (8) can be used to obtain the sum of the at least one segment loss:

Among them, SD(i, j) _v represents the loss that the v-th period among the k periods included in each period group is divided into one period. SDC is the sum of the at least one segment loss.

Similarly, the unsegmented loss when n unit periods are not divided can be calculated using the following formula (9), for example:

in,

is the average value of n flow values in n unit periods, and SDA is the unsegmented loss.

According to an embodiment of the present disclosure, grouping loss may be negatively correlated with group similarity and positively correlated with grouping deviation, for example. In this way, this embodiment can select the time period group with the smallest packet loss from the candidate time period groups as the target time period group. In this way, the obtained time period group can be most similar to the existing time period group, and at the same time, the grouping deviation can be minimized.

For example, the packet loss can be calculated using the following formula (10):

L=SDR-μS(C, C ^o ); formula (10)

Among them, SDR is the grouping deviation, μ is the similarity weight coefficient, and L is the grouping loss.

According to an embodiment of the present disclosure, in the case that the multiple candidate numbers are n candidate numbers whose values belong to an integer in the interval [1, n], after determining the packet loss of each period group in the candidate period group After that, the loss reduction range can be calculated for the grouping losses of the two time period groups of adjacent candidate numbers. Wherein, the two adjacent time period groups of the candidate number may be, for example, a time period group of the candidate number k and a time period group of the candidate number (k-1). In this manner, it can be determined whether the losses of the divided time period groups tend to converge as the number of divided segments increases. In this way, it is possible to avoid the situation that the number of optimal time divisions always takes the maximum value due to the fact that the more time periods obtained by division, the smaller the packet loss, which does not meet the actual demand.

For example, (n-1) candidate numbers whose values belong to the value interval [2, n] among the n candidate numbers may be sequentially searched based on the order of the candidate numbers from small to large, until the target period group is obtained. In the search process, first determine the loss difference between the packet loss for the candidate number (k-1) of the candidate period groups and the packet loss for the candidate number k of the candidate period groups. Then, based on the ratio between the loss difference and the packet loss for the candidate number (k-1) of the candidate period groups, the loss reduction for the candidate number k is determined. Finally, in the case that the loss reduction range for the candidate number k is less than the predetermined range, the candidate time period group for the candidate number (k-1) is determined as the target time period group.

For example, the ratio between the loss difference and the packet loss for the candidate number (k-1) of the candidate period groups may be used as the loss reduction range for the candidate number k. Alternatively, the sum of the ratio and a predetermined value may be used as the loss reduction range. Wherein, the predetermined range may be set according to actual needs, which is not limited in the present disclosure.

According to an embodiment of the present disclosure, when the number of periods obtained by dividing n unit periods is set in the alternative division scheme, this embodiment may also divide n unit periods into a predetermined number of periods based on the traffic value, and divide all The time period group obtained by the possible division scheme is used as the alternative time period group. Then, using a principle similar to that described in FIG. 5 , the target time period group in the candidate time period group is obtained.

Based on the method for dividing time periods provided in the present disclosure, the present disclosure also provides a device for dividing time periods. The device will be described in detail below with reference to FIG. 6 .

As shown in FIG. 6 , the apparatus 600 for dividing time periods in this embodiment may include a traffic value determination module 610 , an alternative determination module 620 and a target group determination module 630 .

The flow value determination module 610 is used to determine the respective flow values of the n unit periods included in the predetermined period based on the historical flow data of the intersection. Wherein, n is an integer greater than 1. In an embodiment, the flow value determination module 610 may be used to perform the operation S210 described above, which will not be repeated here.

The alternative determining module 620 is configured to divide n unit time periods by using an alternative division scheme based on the traffic value, to obtain an alternative time period group. In an embodiment, the candidate determining module 620 may be configured to perform the operation S220 described above, which will not be repeated here.

The target group determination module 630 is configured to determine a target time period group in the candidate time period groups based on the group similarity between the candidate time period groups and the existing time period groups divided by using the existing division scheme. In an embodiment, the target group determination module 630 may be used to perform the operation S230 described above, which will not be repeated here.

According to an embodiment of the present disclosure, the above-mentioned apparatus 600 for dividing time periods may further include a similarity determination module and a group similarity determination module. The similarity determination module is used to determine the similarity between each dividing point in each time period group and the existing time period group, and obtain at least one first similarity. The group similarity determining module is used to determine the average value of at least one first similarity relative to the number of target time periods as the group similarity between each time period group and an existing time period group. Wherein, the target number of time periods is the greater value between the number of time periods in each time period group and the number of time periods in an existing time period group.

According to an embodiment of the present disclosure, the above-mentioned alternative determination module 620 is configured to divide n unit time periods into k time periods based on flow values for each of the multiple alternative numbers, and obtain Alternative period groups for . Wherein, the plurality of alternative numbers are all integers greater than or equal to 1 and less than or equal to n, and k is the value of each alternative number.

According to an embodiment of the present disclosure, the above-mentioned candidate determination module 620 may include a first loss determination submodule and a period division submodule. The first loss determination sub-module is used to take k as the initial value of j, and take 1 as the step size, and perform the following operations in a loop until the value of j is n, and obtain (n-k+1) losses: based on the flow value, determine the The first (j-1) unit period of n unit periods is divided into the minimum loss of (k-1) periods, as the first loss; determine the last (n-j+1) of n unit periods The unit period is divided into a period of loss, as the second loss; and based on the first loss and the second loss, determine the jth unit period in the n unit period as the loss of the candidate division point, and set j to j+1. The period division sub-module is used to divide n unit periods into k periods based on the candidate division point corresponding to the minimum loss among (n-k+1) losses.

According to an embodiment of the present disclosure, the above-mentioned first loss determining submodule may include a similarity determining unit and a first loss determining unit. The similarity determination unit is configured to determine a second similarity between the candidate division point and the existing time period group based on the difference between the candidate division point and each division point in the existing time period group. The first loss determining unit is configured to determine a loss using the jth unit period among the n unit periods as a candidate division point based on the second similarity, the first loss, and the second loss.

According to an embodiment of the present disclosure, the above-mentioned similarity determination unit is configured to determine the difference between the candidate division point and the existing time period group based on the largest difference among the multiple differences between the candidate division point and the multiple division points in the existing time period group. The second similarity between.

According to an embodiment of the present disclosure, the first loss determining submodule further includes a first difference determining unit, a second difference determining unit, and a difference determining unit. The first difference determination unit is configured to determine the absolute value of the difference between the candidate division point and each division point in the existing time period group to obtain the first difference. The second difference determining unit is used to determine the absolute value of the difference between the first difference and the predetermined threshold to obtain the second difference. The difference determination unit is configured to determine the smaller of the second difference and a predetermined threshold as the difference between the candidate division point and each division point in the existing time period group.

According to an embodiment of the present disclosure, the first loss determination sub-module further includes a flow averaging unit and a second loss determination unit. The flow average unit is used to determine the average value of (n-j+1) flow values in the next (n-j+1) unit periods. The second loss determination unit is configured to determine the loss after dividing the last (n-j+1) unit periods into one period based on the deviations of the (n-j+1) flow values respectively relative to the average value.

According to an embodiment of the present disclosure, the above target group determination module 630 may include a second loss determination submodule and a target group determination submodule. The second loss determination submodule is used for each time period group in the alternative time period group: based on the group similarity between each time period group and the existing time period group and the grouping deviation of each time period group, determine each time period group group loss. The target group determination submodule is used to determine the target time period group in the candidate time period group based on the packet loss of each time period group in the candidate time period group.

According to an embodiment of the present disclosure, grouping loss and group similarity are negatively correlated with each other, and grouping loss and grouping deviation are positively correlated with each other.

According to an embodiment of the present disclosure, the plurality of candidate numbers includes n candidate numbers. The above-mentioned target group determination sub-module is used to use the following units, based on the order of the number of candidates from small to large, sequentially search for (n-1) of the n candidate numbers whose value belongs to the value interval [2, n] Select the number until the target period group is obtained: the loss difference determination unit is used to determine the difference between the grouping loss for the candidate period group of the alternative number (k-1) and the grouping loss for the candidate period group for the alternative number k The loss difference value; The decline range determination unit is used to determine the loss reduction range for the alternative number k based on the ratio between the loss difference value and the grouping loss for the alternative time period group of the alternative number (k-1); And a target group determining unit, configured to determine the candidate time period group for the candidate number (k-1) as the target time period group when the loss reduction range for the candidate number k is less than a predetermined range.

According to an embodiment of the present disclosure, the above-mentioned apparatus 600 for dividing time periods may further include a segmentation loss determination module, a non-segmentation loss determination module, and a grouping deviation determination module. The segmentation loss determination module is used for each time period group in the candidate time period group: based on the traffic value, determine the loss of dividing each time period in each time period group into a time period, and obtain at least one segmentation loss. The unsegmented loss determination module is used to obtain the unsegmented loss based on the deviations of the n flow values in n unit time periods relative to the average value of the n flow values. The grouping deviation determination module is used for determining the grouping deviation of each time period group based on the ratio between the sum of at least one segmentation loss and the non-segmentation loss.

According to an embodiment of the present disclosure, the above-mentioned candidate determining module 620 is configured to divide n unit time periods into a predetermined number of time periods based on the flow value to obtain a group of candidate time periods.

It should be noted that in the technical solution of the present disclosure, the acquisition, collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved are all in compliance with relevant laws and regulations, and do not violate public order and good customs.

According to the embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

FIG. 7 shows a schematic block diagram of an example electronic device 700 that can be used to implement the method for dividing a time period according to an embodiment of the present disclosure. Electronic device is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are by way of example only, and are not intended to limit implementations of the disclosure described and/or claimed herein.

As shown in FIG. 7, the device 700 includes a computing unit 701 that can execute according to a computer program stored in a read-only memory (ROM) 702 or loaded from a storage unit 708 into a random-access memory (RAM) 703. Various appropriate actions and treatments. In the RAM 703, various programs and data necessary for the operation of the device 700 can also be stored. The computing unit 701, ROM 702, and RAM 703 are connected to each other through a bus 704. An input/output (I/O) interface 705 is also connected to the bus 704 .

Multiple components in the device 700 are connected to the I/O interface 705, including: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a magnetic disk, an optical disk, etc. ; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 709 allows the device 700 to exchange information/data with other devices over a computer network such as the Internet and/or various telecommunication networks.

The computing unit 701 may be various general-purpose and/or special-purpose processing components having processing and computing capabilities. Some examples of computing units 701 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processing processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 executes the various methods and processes described above, such as the method of dividing time periods. For example, in some embodiments, the method of dividing time periods may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 708 . In some embodiments, part or all of the computer program may be loaded and/or installed on the device 700 via the ROM 702 and/or the communication unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the method for dividing time periods described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured in any other appropriate way (for example, by means of firmware) to execute the method for dividing time periods.

Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips Implemented in a system of systems (SOC), load programmable logic device (CPLD), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpreted on a programmable system including at least one programmable processor, the programmable processor Can be special-purpose or general-purpose programmable processor, can receive data and instruction from storage system, at least one input device, and at least one output device, and transmit data and instruction to this storage system, this at least one input device, and this at least one output device an output device.

Program codes for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special purpose computer, or other programmable data processing devices, so that the program codes, when executed by the processor or controller, make the functions/functions specified in the flow diagrams and/or block diagrams Action is implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer discs, hard drives, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

To provide for interaction with the user, the systems and techniques described herein can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user. ); and a keyboard and pointing device (eg, a mouse or a trackball) through which a user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and can be in any form (including Acoustic input, speech input or, tactile input) to receive input from the user.

The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., as a a user computer having a graphical user interface or web browser through which a user can interact with embodiments of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system can be interconnected by any form or medium of digital data communication, eg, a communication network. Examples of communication networks include: Local Area Network (LAN), Wide Area Network (WAN) and the Internet.

A computer system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other. Among them, the server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the problem of traditional physical host and VPS service ("Virtual Private Server", or "VPS" for short). ″), there are defects such as high management difficulty and weak business scalability. The server can also be a server of a distributed system, or a server combined with a blockchain.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, each step described in the present disclosure may be executed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution disclosed in the present disclosure can be achieved, no limitation is imposed herein.

The specific implementation manners described above do not limit the protection scope of the present disclosure. It should be apparent to those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made depending on design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

A method of dividing time periods comprising:

Based on the historical flow data of the intersection, determine the respective flow values of the n unit periods included in the predetermined period;

Based on the flow value, using an alternative division scheme to divide the n unit time periods to obtain an alternative time period group; and

determining a target time period group in the candidate time period group based on the group similarity between the candidate time period group and an existing time period group obtained by using an existing division scheme,

Wherein, n is an integer greater than 1.
The method according to claim 1, wherein the method further comprises, for each time period group in the candidate time period groups:

determining the similarity between each dividing point in each time period group and the existing time period group to obtain at least one first similarity; and

determining the average value of the at least one first similarity relative to the number of target time periods as the group similarity between each time period group and the existing time period group,

Wherein, the target number of time periods is the greater value of the number of time periods in each time period group and the number of time periods in the existing time period group.
The method according to claim 1 or 2, wherein, based on the flow value, the n unit periods are divided using an alternative division scheme, and the alternative period groups obtained include:

For each of the plurality of candidate numbers, dividing the n unit time periods into k time periods based on the flow value to obtain a group of candidate time periods for each of the candidate numbers,

Wherein, the plurality of candidate numbers are all integers greater than or equal to 1 and less than or equal to n, and k is the value of each candidate number.
The method according to claim 3, wherein dividing the n unit time periods into k time periods based on the flow value comprises:

Take k as the initial value of j, and take 1 as the step size, perform the following operations in a loop until the value of j is n, and get (n-k+1) losses:

Based on the flow value, determine the minimum loss for dividing the first (j-1) unit periods in the n unit periods into (k-1) periods, as the first loss;

Determining the loss of dividing the last (n-j+1) unit periods of the n unit periods into one period as the second loss;

Based on the first loss and the second loss, determine a loss with the jth unit period among the n unit periods as a candidate division point; and

Based on the candidate dividing point corresponding to the minimum loss among the (n-k+1) losses, divide the n unit time periods into the k time periods.
The method according to claim 4, wherein, based on the first loss and the second loss, determining the loss using the jth unit period among the n unit periods as a candidate division point comprises:

determining a second similarity between the candidate division point and the existing time period group based on the difference between the candidate division point and each division point in the existing time period group; and

Based on the second similarity, the first loss, and the second loss, determine a loss that uses the jth unit period among the n unit periods as a candidate division point.
The method according to claim 5, wherein determining the second similarity between the candidate dividing point and the existing time period group comprises:

Determine a second similarity between the candidate division point and the existing time period group based on the largest difference among the plurality of differences between the candidate division point and the existing time period group. Spend.
The method according to claim 5 or 6, wherein, based on the first loss and the second loss, determining the loss with the jth unit period in the n unit periods as a candidate division point further comprises :

determining the absolute value of the difference between the candidate division point and each division point in the existing period group to obtain a first difference;

determining an absolute value of the difference between the first difference and a predetermined threshold to obtain a second difference; and

The smaller of the second difference and the predetermined threshold is determined as the difference between the candidate division point and each division point in the existing period group.
The method according to claim 4, wherein determining the loss of dividing the last (n-j+1) unit periods of the n unit periods into one period comprises:

determining an average of (n-j+1) flow values for said subsequent (n-j+1) unit periods; and

Based on the respective deviations of the (n-j+1) flow values relative to the average value, the loss of dividing the last (n-j+1) unit periods into one period is determined.
The method according to any one of claims 1 to 3, wherein the determination of the candidate time period group is based on the group similarity between the candidate time period group and the existing time period group obtained by using the existing division scheme. The target period groups in the selection period group include:

For each time period group in the candidate time period group: based on the group similarity between each time period group and the existing time period group and the grouping deviation of each time period group, determine the the packet loss for the bucket group; and

A target time period group in the candidate time period groups is determined based on packet loss for each time period group in the candidate time period groups.
The method of claim 9, wherein the grouping loss and the group similarity are negatively correlated with each other, and the grouping loss and the grouping deviation are positively correlated with each other.
The method according to claim 9, wherein said plurality of candidate numbers comprises n candidate numbers; said determining said candidate time period based on the packet loss of each time period group in said candidate time period group The target time period group in the group includes: based on the order of the candidate numbers from small to large, sequentially search for (n-1) candidate numbers whose values belong to the value interval [2, n] among the n candidate numbers , until the target time period group is obtained:

determining the loss difference between the grouping loss for the candidate number (k-1) of the candidate period groups and the grouping loss for the candidate number k of the candidate period groups;

determining a loss reduction for the candidate number k based on a ratio between the loss difference and packet losses for the candidate number (k-1) of candidate time slot groups; and

In a case where the loss reduction range for the candidate number k is less than a predetermined range, determine the candidate time period group for the candidate number (k-1) as the target time period group.
The method of claim 9, further comprising:

For each time period group in the candidate time period group: based on the flow value, determine the loss that each time period in each time period group is divided into a time period, and obtain at least one segmentation loss;

Obtaining an unsegmented loss based on deviations of the n flow values of the n unit periods from the average value of the n flow values, respectively; and

A grouping bias for each group of time periods is determined based on a ratio between the sum of the at least one segmented loss and the unsegmented loss.
The method according to claim 1, wherein, based on the flow value, dividing the n unit periods by an alternative division scheme comprises:

Based on the flow value, the n unit time periods are divided into a predetermined number of time periods to obtain the candidate time period group.
A device for dividing time periods, comprising:

A flow value determination module, configured to determine the respective flow values of the n unit periods included in the predetermined period based on the historical flow data at the intersection;

An alternative determination module, configured to divide the n unit time periods using an alternative division scheme based on the flow value to obtain an alternative time period group; and

A target group determination module, configured to determine a target time period group in the candidate time period group based on the group similarity between the candidate time period group and the existing time period group obtained by using an existing division scheme,

Wherein, n is an integer greater than 1.
The apparatus of claim 14, further comprising:

A similar determination module, configured to determine the similarity between each division point in each time period group and the existing time period group, to obtain at least one first similarity; and

A group similarity determination module, configured to determine the average value of the at least one first similarity relative to the number of target time periods as the group similarity between each time period group and the existing time period group,

Wherein, the target number of time periods is the greater value of the number of time periods in each time period group and the number of time periods in the existing time period group.
The device according to claim 14 or 15, wherein the alternative determination module is used for:

For each of the plurality of candidate numbers, dividing the n unit time periods into k time periods based on the flow value to obtain a group of candidate time periods for each of the candidate numbers,

Wherein, the plurality of candidate numbers are all integers greater than or equal to 1 and less than or equal to n, and k is the value of each candidate number.
The apparatus according to claim 16, wherein the alternative determination module comprises:

The first loss determination sub-module is used to take k as the initial value of j and take 1 as the step size to perform the following operations in a loop until the value of j is n to obtain (n-k+1) losses:

Based on the flow value, determine the minimum loss for dividing the first (j-1) unit periods in the n unit periods into (k-1) periods, as the first loss;

Determining the loss of dividing the last (n-j+1) unit periods of the n unit periods into one period as the second loss;

Based on the first loss and the second loss, determine a loss with the jth unit period among the n unit periods as a candidate division point; and

The period division sub-module is configured to divide the n unit periods into the k periods based on the candidate division point corresponding to the minimum loss among the (n-k+1) losses.
The device according to claim 17, wherein the first loss determination submodule comprises:

a similarity determination unit, configured to determine a second division point between the candidate division point and the existing time period group based on the difference between the candidate division point and each division point in the existing time period group similarity; and

A first loss determining unit, configured to determine a loss using the j-th unit period among the n unit periods as a candidate division point based on the second similarity, the first loss, and the second loss .
The device according to claim 17, wherein the similarity determination unit is configured to:

Determine a second similarity between the candidate division point and the existing time period group based on the largest difference among the plurality of differences between the candidate division point and the existing time period group. Spend.
The device according to claim 18 or 19, wherein the first loss determination submodule further comprises:

A first difference determination unit, configured to determine the absolute value of the difference between the candidate division point and each division point in the existing time period group to obtain a first difference;

A second difference determining unit, configured to determine an absolute value of the difference between the first difference and a predetermined threshold to obtain a second difference; and

A difference determination unit, configured to determine the smaller of the second difference and the predetermined threshold as the difference between the candidate division point and each division point in the existing time period group.
The device according to claim 17, wherein the first loss determination submodule further comprises:

A flow averaging unit, configured to determine the average of (n-j+1) flow values of the latter (n-j+1) unit periods; and

The second loss determination unit is configured to determine to divide the last (n-j+1) unit periods into one period based on the deviations of the (n-j+1) flow values relative to the average value respectively Loss.
The device according to any one of claims 14-16, wherein the target group determination module comprises:

The second loss determination submodule is configured to, for each time period group in the candidate time period group: based on the group similarity between each time period group and the existing time period group and each time period group A grouping deviation of , determining a grouping loss for each time period group; and

A target group determination submodule, configured to determine a target time period group in the candidate time period groups based on the packet loss of each time period group in the candidate time period groups.
The apparatus of claim 22, wherein the grouping loss and the group similarity are negatively correlated with each other, and the grouping loss and the grouping bias are positively correlated with each other.
The device according to claim 22, wherein the plurality of candidate numbers includes n candidate numbers; the target group determination submodule is configured to adopt the following units, based on the order of the candidate numbers from small to large, in order Search for (n-1) candidate numbers whose values belong to the value interval [2, n] among the n candidate numbers, until the target period group is obtained:

Loss difference determining unit, for determining the grouping loss for the grouping loss of the candidate time period group of alternative number (k-1) and the loss difference value between the grouping loss for the grouping loss of the candidate time period group of candidate number k;

A reduction range determination unit configured to determine a loss reduction range for the candidate number k based on the ratio between the loss difference and the grouping loss for the candidate number (k-1) of the candidate period groups; and

A target group determining unit, configured to determine the candidate time period group for the candidate number (k-1) as the target time period group in the case that the loss reduction range for the candidate number k is less than a predetermined range.
The apparatus of claim 22, further comprising:

A segmentation loss determination module, for each time period group in the candidate time period group: based on the flow value, determine the loss that each time period in each time period group is divided into a time period, and obtain at least one segment loss;

The unsegmented loss determination module is used to obtain the unsegmented loss based on the deviations of the n flow values of the n unit periods relative to the average value of the n flow values; and

A grouping deviation determination module, configured to determine a grouping deviation for each time period group based on a ratio between the sum of the at least one segment loss and the non-segmentation loss.
The apparatus according to claim 14, wherein the alternative determination module is configured to:

Based on the flow value, the n unit time periods are divided into a predetermined number of time periods to obtain the candidate time period group.
An electronic device comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can perform any one of claims 1-13. Methods.
A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method according to any one of claims 1-13.
A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-13.