CN113316251A

CN113316251A - Positioning method and device based on wireless signal, electronic equipment and storage medium

Info

Publication number: CN113316251A
Application number: CN202110865837.2A
Authority: CN
Inventors: 夏浩; 沈国斌; 张延�
Original assignee: Lazas Network Technology Shanghai Co Ltd; Zhejiang Koubei Network Technology Co Ltd
Current assignee: Lazas Network Technology Shanghai Co Ltd; Zhejiang Koubei Network Technology Co Ltd
Priority date: 2021-07-29
Filing date: 2021-07-29
Publication date: 2021-08-27

Abstract

The present specification provides a positioning method based on wireless signals, comprising: acquiring a target characteristic value of a wireless signal detected by a target client in the execution process of the current distribution service, and determining a matching relation between a variation trend of the target characteristic value and each standard variation trend, wherein the standard variation trend is a variation trend of the characteristic value of the wireless signal in each characteristic subset in a signal characteristic set, the signal characteristic set is used for recording the characteristic value of the wireless signal detected by a client of a distribution party in the execution process of at least one historical distribution service, distribution articles corresponding to at least one historical distribution service come from the same pick-up place, and each characteristic subset corresponds to each time period in the execution process; and determining the current time period of the target client according to the matching relation so as to determine the position relation between the target client and the pickup place according to the current time period.

Description

Positioning method and device based on wireless signal, electronic equipment and storage medium

Technical Field

One or more embodiments of the present disclosure relate to the field of communications technologies, and in particular, to a positioning method and apparatus, an electronic device, and a storage medium based on a wireless signal.

Background

With the development of internet technology, services which need to be provided by adopting a distribution mode are increasing, and the business volume of distribution business is rapidly increased.

The instant delivery is a delivery mode which depends on social inventory and can meet the delivery requirement in a short time, and is in a logistics form in response To O2O (Online To Offline). In the field of instant delivery, the accurate time for a delivery person to arrive at and leave a pick-up place of delivered articles is obtained, and the method has great value for delivery businesses.

Disclosure of Invention

In view of this, one or more embodiments of the present disclosure provide a positioning method and apparatus, an electronic device, and a storage medium based on wireless signals.

To achieve the above object, one or more embodiments of the present disclosure provide the following technical solutions:

according to a first aspect of one or more embodiments of the present specification, there is provided a wireless signal based positioning method, including:

acquiring a target characteristic value of a wireless signal detected by a target client in the execution process of the current distribution service, and determining a matching relationship between the variation trend of the target characteristic value and each standard variation trend, the standard variation trend is a variation trend of the characteristic values of the wireless signal in each characteristic subset in the signal characteristic set, the signal feature set is used for recording feature values of wireless signals detected by a client of a distributor in the execution process of at least one historical distribution service, the delivered items corresponding to the at least one historical delivery service come from the same pick-up place, each characteristic subset corresponds to each time period in the execution process, the time period comprises an article pickup stage between the delivery side client terminal arriving at the pickup place and leaving the pickup place in the execution process and a time period when the delivery side client terminal is positioned at other places different from the pickup place in the execution process;

and determining the current time period of the target client according to the matching relation so as to determine the position relation between the target client and the pickup place according to the current time period.

Alternatively to this, the first and second parts may,

selecting a characteristic subset corresponding to each time period in the execution process from the signal characteristic set according to first time information, wherein the first time information is the time information corresponding to an article picking stage uploaded by the client of the delivery party;

or, the feature subsets corresponding to the respective time periods in the execution process are selected from the signal feature sets according to second time information, wherein the second time information is time information under the condition that the wireless signals detected by the delivery side client are matched with the wireless signals transmitted from the pick-up place;

or, the feature subsets corresponding to the respective time periods in the execution process are selected from the signal feature sets according to third time information, where the third time information is time information in a case where the wireless signal detected by the delivery side client matches the wireless signal detected by the reference client located at the pickup location.

Optionally, the feature values of the wireless signals in each feature subset are labeled with tag information indicating whether the wireless signals are located at the pickup location within a corresponding time period; wherein the standard variation trend of the characteristic values of the wireless signals in the respective characteristic subsets in the corresponding time period is determined by the following method: inputting the characteristic values of the wireless signals in each characteristic subset and the labeled label information into a time sequence machine learning model for training to obtain a positioning model;

the determining the current time period of the target client according to the matching relationship to determine the position relationship between the target client and the pickup location according to the current time period includes: and inputting the target characteristic value of the wireless signal detected by the target client into the positioning model so as to determine the position relation between the target client and the pickup place according to the output result of the positioning model.

Optionally, the positioning model is deployed in the target client.

Optionally, the determining a position relationship between the target client and the pickup location according to the current time period includes:

and under the condition that the current time period belongs to an article pickup stage, judging that the target client arrives at the pickup place.

Optionally, the method further includes:

after the target client is judged to be located at the pickup place, if the current time period of the target client is determined not to be an article pickup stage, the target client is judged to leave the pickup place.

According to a second aspect of one or more embodiments of the present specification, there is provided a feature extraction method of a wireless signal, including:

acquiring a signal characteristic set of a wireless signal detected by a client of a distribution party in the execution process of at least one historical distribution service, wherein distributed articles corresponding to the at least one historical distribution service come from the same pick-up place, and the signal characteristic set is used for recording a characteristic value of the wireless signal;

identifying a subset of features in the set of signal features that correspond to respective time periods in the performance process, and determining a standard variation trend of the characteristic values of the wireless signals in the respective characteristic subsets in the corresponding time periods, the time period comprises an article pickup stage between the delivery side client terminal arriving at the pickup place and leaving the pickup place in the execution process and a time period when the delivery side client terminal is positioned at other places different from the pickup place in the execution process, the standard variation trend is used to match a variation trend of a target characteristic value of the wireless signal detected by the target client during the execution of the current distribution service, and determining the current time period of the target client according to the matching result, and determining the position relation between the target client and the pickup place according to the current time period.

Optionally, the identifying a subset of features in the signal feature set corresponding to each time period in the execution process includes:

acquiring first time information which is uploaded by the client of the delivery party and corresponds to an article picking stage, and selecting a feature subset corresponding to each time period from the signal feature set according to the first time information;

or acquiring second time information under the condition that the wireless signal detected by the client of the delivery party is matched with the wireless signal transmitted from the pick-up place, and selecting a feature subset corresponding to each time period from the signal feature set according to the second time information;

or, acquiring third time information in a case where the wireless signal detected by the delivery side client matches the wireless signal detected by the reference client located at the pickup location, and selecting a feature subset corresponding to each time period from the signal feature set according to the third time information.

Optionally, the feature values of the wireless signals in each feature subset are labeled with tag information indicating whether the wireless signals are located at the pickup location within a corresponding time period;

the determining of the standard variation trend of the characteristic values of the wireless signals in the respective characteristic subsets in the corresponding time periods includes: inputting the characteristic values of the wireless signals in each characteristic subset and the labeled label information into a time sequence machine learning model for training to obtain a positioning model;

determining a positional relationship between the target client and the pickup location by: and inputting the target characteristic value of the wireless signal detected by the target client into the positioning model so as to determine the position relation between the target client and the pickup place according to the output result of the positioning model.

Optionally, the positioning model is deployed in the target client.

Optionally, the characteristic values recorded by the signal characteristic set are obtained by detecting each wireless signal according to a preset detection period; the method further comprises the following steps:

calculating the variation range of the characteristic value of each wireless signal detected in the adjacent detection period;

and deleting the characteristic value in the corresponding detection period when the variation range exceeds the range threshold.

Optionally, the wireless signal includes a WiFi signal and/or a bluetooth signal;

in the case where the wireless signal includes a WiFi signal, detecting a dimension of the eigenvalue of the wireless signal includes a multipath structure of the WiFi signal and/or a received signal strength of the WiFi signal;

where the wireless signal comprises a bluetooth signal, the dimension comprises a signal strength of the bluetooth signal.

According to a third aspect of one or more embodiments of the present specification, there is provided a positioning apparatus based on wireless signals, including:

an obtaining unit, configured to obtain a target feature value of a wireless signal detected by a target client during execution of a current distribution service, and determine a matching relationship between a variation trend of the target feature value and each standard variation trend, where the standard variation trend is a variation trend of feature values of the wireless signal in each feature subset in a signal feature set, the signal feature set is used to record feature values of the wireless signal detected by a client of a distributor during execution of at least one historical distribution service, a distribution item corresponding to the at least one historical distribution service is from a same pickup location, each feature subset corresponds to each time period in the execution process, and the time period includes an item pickup stage between arrival of the client of the distributor at the pickup location and departure from the pickup location during the execution process and a time period when the client of the distributor is located at another location different from the pickup location during the execution process A time period;

and the determining unit is used for determining the current time period of the target client according to the matching relation so as to determine the position relation between the target client and the pickup place according to the current time period.

Alternatively to this, the first and second parts may,

the determining unit is specifically configured to: and inputting the target characteristic value of the wireless signal detected by the target client into the positioning model so as to determine the position relation between the target client and the pickup place according to the output result of the positioning model.

Optionally, the positioning model is deployed in the target client.

Optionally, the determining unit is specifically configured to:

Optionally, the determining unit is further configured to:

According to a fourth aspect of one or more embodiments of the present specification, there is provided a feature extraction apparatus for a wireless signal, including:

the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a signal characteristic set of a wireless signal detected by a client of a delivery party in the execution process of at least one historical delivery service, delivered articles corresponding to the at least one historical delivery service come from the same pick-up place, and the signal characteristic set is used for recording a characteristic value of the wireless signal;

an identification unit that identifies a subset of features in the set of signal features that correspond to respective time periods in the execution process, and determining a standard variation trend of the characteristic values of the wireless signals in the respective characteristic subsets in the corresponding time periods, the time period comprises an article pickup stage between the delivery side client terminal arriving at the pickup place and leaving the pickup place in the execution process and a time period when the delivery side client terminal is positioned at other places different from the pickup place in the execution process, the standard variation trend is used to match a variation trend of a target characteristic value of the wireless signal detected by the target client during the execution of the current distribution service, and determining the current time period of the target client according to the matching result, and determining the position relation between the target client and the pickup place according to the current time period.

Optionally, the identification unit is specifically configured to:

the identification unit is specifically configured to: inputting the characteristic values of the wireless signals in each characteristic subset and the labeled label information into a time sequence machine learning model for training to obtain a positioning model;

Optionally, the positioning model is deployed in the target client.

Optionally, the characteristic values recorded by the signal characteristic set are obtained by detecting each wireless signal according to a preset detection period; the acquisition unit is further configured to:

According to a fifth aspect of one or more embodiments herein, there is provided an electronic device, comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor implements the method as described in any of the above embodiments by executing the executable instructions.

According to a sixth aspect of one or more embodiments of the present specification, there is provided a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the method as in any one of the above-described embodiments.

In the technical solution of the present specification, a client of a delivery side detects a signal characteristic of a wireless signal in a process of executing a delivery service, where the executing process includes a time period for going to a pickup location of a delivered article, a time period for picking up the delivered article after arriving at the pickup location, and a time period for delivering the delivered article after leaving the pickup location in a time dimension. Meanwhile, based on the limitation of the wireless signal on the signal range (beyond a certain range, the wireless signal cannot be detected), the client of the distribution side can only detect the wireless signal within a certain range near the position of the client. Then, in the stage of going to and leaving the pickup location, the wireless signal detected by the client of the distribution side is not the wireless signal corresponding to the pickup location, and in the stage of picking up the article, the client of the distribution side is in the pickup location and can detect the wireless signal corresponding to the pickup location.

Based on the characteristics, the change trend of the characteristic values of the signal characteristics in each time period in the execution process of the distribution service can be mined aiming at the signal characteristic set acquired from the same pickup place, and the change trend is used as the standard change trend for representing the characteristic value change characteristics in each time period. After the standard change trend is obtained, the target characteristic value of the wireless signal detected by the target client is also obtained for the distribution service currently executed by the target client, and then the current time period of the target client is determined according to the matching relation between the change trend of the target characteristic value and each standard change trend, so that the position relation between the target client and the pickup place can be determined according to the relation between the current time period and the pickup place.

On the one hand, at some time points during the transmission of the wireless signal, there may be a case where the transmission of the wireless signal is unstable, and then the detected signal characteristic is represented as an unstable characteristic value. The positioning is carried out by excavating the change trend of the characteristic value of the wireless signal in each time period of the distribution service, compared with an isolated time point, the change trend of the characteristic value in the whole time period is more accurate and stable, and even if the wireless signal fluctuates, the change trend of the characteristic value in the whole time period cannot be greatly influenced, so that the subsequent positioning based on the change trend can be more accurate.

On the other hand, the process of acquiring the signal feature set in the specification does not require that a distributor adds extra operation in the process of using the client of the distributor for distribution, and the client of the distributor uploads the detected feature values in real time, so that the normal distribution process of the distributor is not affected, and the condition that the distribution service is not affected and executed is avoided, so that the cost is reduced, and the detection efficiency is improved. Moreover, as long as there is delivery service in the pickup location, the technical scheme of the present specification can be used for positioning, and all pickup locations with delivery service can be covered, thereby improving the coverage rate.

Drawings

Fig. 1 is a flowchart of a method for feature extraction of a wireless signal according to an exemplary embodiment.

Fig. 2 is a flowchart of a positioning method based on wireless signals according to an exemplary embodiment.

Fig. 3 is a schematic structural diagram of an electronic device according to an exemplary embodiment.

Fig. 4 is a block diagram of a feature extraction apparatus for wireless signals according to an exemplary embodiment.

Fig. 5 is a block diagram of a wireless signal based positioning apparatus according to an exemplary embodiment.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with one or more embodiments of the present specification. Rather, they are merely examples of apparatus and methods consistent with certain aspects of one or more embodiments of the specification, as detailed in the claims which follow.

It should be noted that: in other embodiments, the steps of the corresponding methods are not necessarily performed in the order shown and described herein. In some other embodiments, the method may include more or fewer steps than those described herein. Moreover, a single step described in this specification may be broken down into multiple steps for description in other embodiments; multiple steps described in this specification may be combined into a single step in other embodiments.

The instant delivery is a delivery mode which depends on social inventory and can meet the delivery requirement in a short time, and is in a logistics form corresponding to O2O. In the field of instant delivery, the accurate time for a delivery person to arrive at and leave a pick-up place of delivered articles is obtained, and the method has great value for delivery businesses. For example, in an outdoor scene, the judgment of the time points when the rider arrives at the store and leaves the store is helpful for improving the accuracy of the estimated meal time. For another example, the user who initiates the distribution service can be informed whether the current distributor arrives at the store or leaves the store, thereby facilitating the user to know the distribution progress. The present specification aims to provide a feature extraction scheme for wireless signals, which determines whether a distributor arrives at a pick-up location for distributing articles based on a feature value of a wireless signal acquired by a client of a distributor in the process of executing distribution business.

Referring to fig. 1, fig. 1 is a flowchart illustrating a method for extracting features of a wireless signal according to an exemplary embodiment. As shown in fig. 1, the method applied to the server may include the following steps:

step 102, a signal feature set of a wireless signal detected by a client of a distribution party in the execution process of at least one historical distribution business is obtained, distributed articles corresponding to the at least one historical distribution business come from the same pickup place, and the signal feature set is used for recording feature values of the wireless signal.

In the present embodiment, the feature extraction scheme of the present specification is implemented for each pickup location. The wireless signals to which this specification is directed may include WiFi signals and/or bluetooth signals. Wherein, under the condition that the detected wireless Signal includes a WiFi Signal, the dimension of the characteristic value of the detected wireless Signal includes a multi-path structure of the WiFi Signal and/or a Received Signal Strength of the WiFi Signal, and the Received Signal Strength is RSS (Received Signal Strength) of an AP (Access Point) of the WiFi Signal. In the case where the detected wireless signal includes a bluetooth signal, the dimension of the characteristic value of the detected wireless signal includes a signal strength of the bluetooth signal. Of course, any other near field communication technology can be used as long as the characteristic of 'range limitation' is provided. For example, Communication technologies such as IrDA (Infrared Data Association) Infrared Data transmission, ZigBee, NFC (Near Field Communication), UWB (Ultra WideBand), DECT (Digital Enhanced Cordless communications), and the like may be used.

Due to the fact that the distribution service of the same pickup location can be carried by different distributors, the signal feature set can be added by the method by means of the fact that the multiple feature data (namely the multiple detections for the same pickup location) can be obtained for the same pickup location, and therefore the signal feature set is accurate and comprehensive.

Step 104, identifying the feature subsets in the signal feature set corresponding to the time periods in the execution process, and determining a standard variation trend of the characteristic values of the wireless signals in the respective characteristic subsets in the corresponding time periods, the time period comprises an article picking stage between the delivery party arriving at the picking place and leaving the picking place in the execution process and a time period when the delivery party is positioned at other places different from the picking place in the execution process, the standard variation trend is used to match a variation trend of a target characteristic value of the wireless signal detected by the target client during the execution of the current distribution service, and determining the current time period of the target client according to the matching result, and determining the position relation between the target client and the pickup place according to the current time period.

In this embodiment, the user can place an order for the goods on the shelf on the e-commerce platform. In some cases, the items ordered by the user need to be delivered from a physical store or warehouse to the location of the user. Therefore, a delivery service platform (e.g., an e-commerce platform or other delivery platform cooperating therewith) is required to generate a corresponding delivery service for delivering the item, and distribute the delivery service to the deliverer. After receiving the delivery service, the delivery person goes to a pick-up location (such as the aforementioned physical store or warehouse) for picking up the delivery item, and delivers the delivery item to the user location after the pick-up is successful.

For example, in an outsourcing scene, a user places an order to a certain physical store on the takeout platform through a user client, the takeout platform generates a corresponding takeout order and distributes the takeout order to a delivery party client (in this case, a client device used by a rider), and the rider goes to the physical store (i.e., a pickup location for delivering an article) to pick up the takeout and delivers the takeout to a location designated by the user. For another example, in an express scene, for a delivered item stored in a warehouse, an express platform generates a corresponding express order and distributes the express order to a delivery side client (in this case, a client device used by a courier), so that the courier goes to the warehouse (i.e., a pickup location of the delivered item) to pick up the item and delivers the item to a location where a recipient is located.

Therefore, the execution process of the distribution service can be divided into an article pickup stage from the arrival of the distribution side client at the pickup location to the departure from the pickup location (a time period from the arrival of the distribution side client at the pickup location to the departure from the pickup location, that is, a time period from the arrival of the distribution side client at the pickup location to the pickup of the distribution articles) and a time period when the distribution side client is located at another location different from the pickup location. The other time periods different from the item phase may be further divided into a time period in which the dispensing client (the dispenser carrying the dispensing client) goes to a pickup location for dispensing the item (hereinafter, referred to as a go-to phase) and a time period in which the dispensing item is dispensed after leaving the pickup location (hereinafter, referred to as a dispensing phase). At the same time, there is a corresponding wireless signal at the pickup location. Taking the WiFI signal as an example, the AP is configured at the pickup location, and on the premise of obtaining the authorization of the distributor (the client of the distributor is authorized to detect the WiFI signal in the process of executing the distribution service), the client of the distributor can detect the WiFI signal emitted from the pickup location within a certain range. Taking bluetooth signals as an example, an AP is configured at a pickup location, a distribution party is configured with an entity bluetooth beacon, and a distribution party client can detect the bluetooth signals broadcasted from the pickup location within a certain range on the premise of obtaining the authorization of a distribution operator (the distribution party client is authorized to detect the bluetooth signals in the process of executing distribution services).

Then, the distribution client may detect the characteristic value of the wireless signal during the process of executing the distribution service, and based on the limitation of the wireless signal on the signal range, the distribution client may only detect the wireless signal within a certain range near the location of the distribution client. Then, in the going-to stage and the distribution stage, the wireless signal detected by the distribution side client is not the wireless signal corresponding to the pickup location, and in the article pickup stage, the distribution side client is located at the pickup location and can detect the wireless signal corresponding to the pickup location.

Based on the above characteristics, the client of the distribution party can detect the signal characteristics of the wireless signals in the execution process of the distribution service according to the preset detection period to obtain the characteristic values of the wireless signals, and the characteristic values are recorded in the signal characteristic set. Since the execution process of the distribution service covers the forward stage, the item picking stage and the distribution stage in the time dimension, the characteristic values (i.e., the characteristic values recorded in the signal characteristic set) detected by the client of the distribution party in the whole execution process also cover the forward stage, the item picking stage and the distribution stage in the time dimension. In other words, the signal feature set may be divided into feature subsets in the time dimension, wherein the feature subsets respectively correspond to the heading stage, the item picking stage and the distribution stage, and each feature subset includes the feature values detected by the distribution client in the corresponding stage.

For how to identify the feature subsets corresponding to each time period in the signal feature set, modes such as active reporting by a client of a distribution party, matching with a wireless signal at a pickup location, matching with a wireless signal detected by a client at the pickup location, and the like can be referred. The following are detailed below.

In one case, the first time information corresponding to the item pickup stage may be actively uploaded by the distribution side client during the execution of the distribution service, and then the first time information corresponding to the item pickup stage uploaded by the distribution side client may be acquired, so as to select the feature subset corresponding to each time period from the signal feature set according to the first time information. For example, the time information corresponding to the item pickup stage uploaded by the distribution side client may include a time of arrival at the pickup location and a time of departure from the pickup location, which are respectively uploaded by the distributor through the distribution side client. Taking a take-out scene as an example, after a rider receives an order, the rider can manually click a store-to trigger control and a store-leaving trigger control on the rider client, the store-to trigger control is used for triggering the rider client to report a store-to event (store-to time is recorded), and the store-leaving trigger control is used for triggering the rider client to report a store-leaving event (store-leaving time is recorded). Then, the time period from the store arrival time to the store departure time is an article pickup period, the time period before the store arrival time (for example, from the time when the rider takes an order to the store arrival time) is a travel period, and the time period after the store departure time (for example, from the store departure time to the delivery time) is a delivery period.

In another case, second time information in a case where the wireless signal detected by the delivery-side client matches the wireless signal transmitted from the pickup site may be acquired, so that the subset of the features corresponding to each time period is selected from the signal feature set based on the second time information. Also taking a take-away scenario as an example, a physical bluetooth beacon may be deployed in a physical store to periodically broadcast a bluetooth beacon, and when the bluetooth beacon is detected by a rider client, it may be determined that a rider using the rider client arrives at the store, and then the rider client may post a store event when the bluetooth beacon is detected. Further, when the bluetooth beacon is subsequently changed from being detectable to being undetectable, it can be determined that the rider using the rider client leaves the store, and then the rider client can report the exit event. Similarly, each time segment is divided according to the arrival time recorded in the arrival event and the departure time recorded in the departure event, and then the feature subset corresponding to each time segment is selected from the signal feature set.

In another case, third time information in a case where a wireless signal detected by the distributor client matches a wireless signal detected by a reference client located at the pickup site may be acquired, so that a subset of features corresponding to each time period is selected from the signal feature set based on the third time information. Also taking the take-away scenario as an example, a shop owner of the physical store uses a mobile phone (i.e., a reference client) to detect a WiFi signal in the store to obtain a feature sequence, and uploads the feature sequence detected by the mobile phone to the background server. The signature sequence is in the form of WiFi _ id rssi. The WiFi _ id is a signal identifier of the detected WiFi signal, for example, a mac address of a hardware device that transmits the WiFi signal may be used as the signal identifier; rssi is the received signal strength. Meanwhile, the mobile phone of the rider also detects the characteristic sequence of the WiFi signal in real time and uploads the characteristic sequence to the background server. The background server compares the characteristic sequence detected by the mobile phone of the rider with the characteristic sequence uploaded by the mobile phone of the shop owner, and if the characteristic sequence detected by the mobile phone of the rider is matched with the characteristic sequence uploaded by the mobile phone of the shop owner, the rider is judged to be in the shop; otherwise, it is judged that the rider is not in the store. Similarly, other ways of obtaining the store arrival time, the store departure time and the subsequent partition feature subset are similar to those described above, and are not described herein again.

In the process of detecting the wireless signals, the client of the distribution side may detect a plurality of different wireless signals at the same time (i.e. there are a plurality of different wireless signals at the same location), and the detected signal characteristic is a characteristic sequence, which includes characteristic values of all the wireless signals detected at the same time in the characteristic dimension. Accordingly, since the signal feature set covers the execution process of the whole distribution service, the signal feature set can be divided into feature subsets in the time dimension, wherein the feature subsets respectively correspond to the going-to stage, the item picking stage and the distribution stage, and each feature subset comprises a feature sequence detected by the client of the distribution party in the corresponding stage. The wireless signal detected by the client of the distribution party is not the wireless signal corresponding to the pick-up place in the going-to stage and the distribution stage, and the wireless signal corresponding to the pick-up place can be detected when the distribution party is at the pick-up place in the article pick-up stage. Further, for the going-to stage and the distribution stage, since the client of the distribution party is moving continuously and the moving range is large in the two stages, the characteristic value of the wireless signal detected by the client of the distribution party is also changed frequently, for example, the signal identifier of the wireless signal changes frequently, and the characteristic value of the same wireless signal also changes greatly. In the article pickup stage, the client of the delivery party is located in the pickup location, and does not need to move continuously and has a small moving range, so that the characteristic value of the wireless signal detected by the client of the delivery party is relatively stable, for example, the signal identifier of the wireless signal is not changed (i.e., the wireless signal that can be detected is not changed), and the characteristic value of each wireless signal is relatively stable. In summary, compared with the proceeding stage and the distribution stage, the characteristic sequence detected in the item picking stage is more stable, that is, the variation trend of the characteristic value in each time period in the execution process of the whole distribution business has a difference, and the difference and the time period can establish an association relationship. Based on the characteristics, after the characteristic subsets corresponding to the time periods in the execution process of the distribution business in the signal characteristic set are identified, the change trend of the characteristic values of the signal characteristics in the characteristic subsets can be mined, and the change trend is used as the standard change trend for representing the change characteristics of the characteristic values in the corresponding time periods. After the standard change trend is obtained, the target characteristic value of the wireless signal detected by the target client is also obtained for the distribution service currently executed by the target client, and then the current time period of the target client is determined according to the matching relation between the change trend of the target characteristic value and each standard change trend, so that the position relation between the target client and the pickup place can be determined according to the relation between the current time period and the pickup place. For example, after the target client starts to execute the distribution service, if the current time period of the target client is determined to be an article pickup stage, the target client may be determined to be located at a pickup location; otherwise, it may be determined that the target client is not located at the pickup location.

Of course, the influence caused by wireless signal fluctuation can be further avoided. Based on the fact that the client of the distribution party detects the signal characteristics of each wireless signal according to a preset detection period, a range threshold value can be set for measuring whether the change range is reasonable or not according to the change range of the characteristic values of each wireless signal detected in the adjacent detection period. Therefore, the variation range of the characteristic value of each wireless signal detected in the adjacent detection period can be calculated, and if the variation range exceeds the range threshold, the variation range is shown to have large fluctuation, and then the characteristic value in the corresponding detection period can be deleted.

For example, the signature sequences detected in each detection cycle are shown in table 1:

detection period	Characteristic sequence
		1	WiFi_1:-80,WiFi_2:-70
2	WiFi_1:-70,WiFi_2:-60
		3	WiFi_1:-20,WiFi_2:-30
4	WiFi_1:-50,WiFi_2:-40
		5	WiFi_1:-40,WiFi_2:-30
……	……

TABLE 1

As can be seen, the variation range of WiFi _1 between the 1 st detection period and the 2 nd detection period is 10, and the variation range of WiFi _2 is 10; the variation range of WiFi _1 between the 2 nd detection period and the 3 rd detection period is 50, and the variation range of WiFi _2 is 30; the variation range of WiFi _1 between the 3 rd detection period and the 4 th detection period is-30, and the variation range of WiFi _2 is-10; the variation range of WiFi _1 between the 4 th detection period and the 5 th detection period is 10, and the variation range of WiFi _2 is 10. Assuming that the range threshold is 15, it can be found that the variation ranges of WiFi _1 and WiFi _2 are abnormal between the 2 nd detection period and the 3 rd detection period, and the variation ranges of WiFi _1 and WiFi _2 are abnormal between the 3 rd detection period and the 4 th detection period. Therefore, the signature sequence in the 3 rd detection cycle can be deleted.

In this embodiment, a machine learning technique may be employed to mine the standard trend of the feature values of the signal features in the respective feature subsets. Machine learning techniques may utilize algorithms to learn from existing data to make decisions and decisions about real-world conditions. Machine learning techniques include supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, and the like. Aiming at the training process of supervised learning, input sample data is called as a training set, the sample data in the training set has a definite identification or result (namely a sample label), when a predictive model is established by using a supervised learning algorithm, the supervised learning algorithm establishes a learning process, the predictive result is compared with the actual result of the training set, and the predictive model is continuously adjusted until the predictive result of the model reaches an expected accuracy rate. Therefore, the characteristic values of the wireless signals in each characteristic subset can be labeled to obtain sample data, and the label information of the sample data is whether the sample data is located at the pickup place in the corresponding time period. For example, a label "1" indicates being located at the pickup location, and a label "0" indicates not being located at the pickup location. Then, the characteristic value of the wireless signal within the characteristic subset corresponding to the item pickup phase may be labeled as "1", and the characteristic value of the wireless signal within the characteristic subset corresponding to the go-to phase and the delivery phase may be labeled as "0", thereby obtaining sample data. Then, a supervised learning algorithm is adopted to train the sample data to obtain a positioning model, namely the characteristic values of the wireless signals in each characteristic subset and the labeled label information are input into a machine learning model to train to obtain the positioning model, and the training process is to learn the incidence relation between the change trend of the characteristic values of the wireless signals in each characteristic subset in the sample data in a corresponding time period and whether the characteristic values are located in a pickup place. After the training is completed, the model parameters in the positioning model can reflect the standard variation trend and the association relationship.

Further, since the feature extraction process in this specification focuses on the variation trend of the feature values in the whole time period, that is, there is a correlation before and after each sequence data in the sample data, in order to enable the training process to accurately mine the variation trend to learn the correlation, a time series machine learning model may be used to train the sample data.

For example, a Recurrent Neural Network (RNN) is a type of neural network for processing sequence data, and can be used to solve the problem associated with the sequence data. In terms of network structure, RNN network structure can be divided into an input layer (input layer), hidden layers (also called middle layers), and output layers (output layers). The input layer and the output layer are used for processing the input and the output of data, the hidden layer is used for calculating and predicting the data, and the method is characterized in that each unit can be processed according to the input sequence, the output of the previous process is used as the input of the next process to be calculated together with the current input, the calculation result of the previous input is integrated with the current input to be predicted together, namely, the RNN memorizes the information before the sequence data and utilizes the previous information to influence the output of the following unit. Of course, LSTM (Long Short-Term Memory) may also be used.

As another example, both RNN and LSTM can only predict the output at the next time based on the sequence data at the previous time, but in some cases, the output at the current time is not only related to the previous sequence data, but may also be related to future sequence data. Then, BRNN (Recurrent Neural Network) may be employed. Of course, the above-mentioned timing machine learning model is only an exemplary example, and timing machine learning models such as BLSTM (Bidirectional Long Short-Term Memory), bert (Bidirectional Encoder retrieval from transformations), gru (gated recovery Unit recovery Neural networks), and the like may also be used, and the description is not limited thereto.

After the location model is trained, the location model can be used to predict the location of the client of the distributor (which can represent the location of the distributor using the client of the distributor) performing the current distribution service. Specifically, for the target client, the target characteristic value of the wireless signal detected by the target client may be input into the positioning model to determine the position relationship between the target client and the pickup location according to the output result of the positioning model. For example, when the output result of the localization model is "1", it may be determined that the target client is located at the pickup location, and when the output result of the localization model is "0", it may be determined that the target client is not located at the pickup location.

It should be noted that the feature extraction scheme provided in this specification is for each pickup location separately, i.e. the signal feature sets are for the same pickup location. Under the condition that a signal feature set is trained by adopting a time sequence machine learning model, the positioning model obtained by training is used for predicting whether the target client is located at the pickup location corresponding to the signal feature set. In other words, for each pickup location, there is a corresponding positioning model, i.e., a "one-to-one" relationship. Based on the characteristics, the model parameters of the positioning model corresponding to each pickup location are small, so that the positioning model can be deployed in the target client, and the target client inputs the target characteristic value of the wireless signal detected in the execution process of the current distribution service into the locally deployed positioning model, thereby improving the positioning efficiency. For example, after detecting the characteristic value, the target client directly inputs the locally deployed positioning model to determine whether the target client is located at the pickup site, so as to report the corresponding event to the background server without uploading the detected characteristic value to the background server for determination. Of course, the positioning scheme may also be implemented by deploying the positioning model in a background server, which is not limited in this specification.

Based on the above feature extraction scheme, the present specification further provides a positioning scheme based on wireless signals, which is described below with reference to fig. 2.

Referring to fig. 2, fig. 2 is a flowchart illustrating a positioning method based on wireless signals according to an exemplary embodiment. As shown in fig. 2, the method may include the steps of:

step 202, obtaining a target characteristic value of a wireless signal detected by a target client in the execution process of a current distribution service, and determining a matching relationship between a variation trend of the target characteristic value and each standard variation trend, where the standard variation trend is a variation trend of the characteristic value of the wireless signal in each characteristic subset in a signal characteristic set, the signal characteristic set is used for recording the characteristic value of the wireless signal detected by a client of a distributor in the execution process of at least one historical distribution service, a distribution article corresponding to the at least one historical distribution service is from the same pick-up location, each characteristic subset corresponds to each time period in the execution process, and the time period includes an article pick-up stage from the arrival of the client of the distributor at the pick-up location to the departure from the pick-up location in the execution process and a time when the client of the distributor is located at another location different from the pick-up location in the execution process And (4) section.

And 204, determining the current time period of the target client according to the matching relation, and determining the position relation between the target client and the pickup place according to the current time period.

As described above, the feature subsets corresponding to the time periods in the execution process in the signal feature set are selected from the signal feature set according to the first time information, where the first time information is the time information corresponding to the item pickup stage uploaded by the client of the distribution party.

Or, the feature subsets corresponding to the respective time periods in the execution process are selected from the signal feature sets according to second time information, where the second time information is time information in a case where the wireless signal detected by the delivery side client matches the wireless signal transmitted from the pickup place.

As described above, the feature values of the wireless signals in each feature subset are labeled with the tag information of whether the wireless signals are located at the pickup location in the corresponding time period; wherein the standard variation trend of the characteristic values of the wireless signals in the respective characteristic subsets in the corresponding time period is determined by the following method: inputting the characteristic values of the wireless signals in each characteristic subset and the labeled label information into a time sequence machine learning model for training to obtain a positioning model; the determining the current time period of the target client according to the matching relationship to determine the position relationship between the target client and the pickup location according to the current time period includes: and inputting the target characteristic value of the wireless signal detected by the target client into the positioning model so as to determine the position relation between the target client and the pickup place according to the output result of the positioning model.

As previously described, the positioning model is deployed in the target client. In this case, the execution subject of the positioning scheme shown in the present embodiment is the target client.

As described above, in the case that the current time period belongs to the item pickup stage, it may be determined that the target client arrives at the pickup location. Further, after the target client is determined to be located at the pickup location, if it is determined that the current time period of the target client is not the article pickup stage, it is determined that the target client leaves the pickup location.

For example, after the target client starts to execute the distribution service, if the current time period of the target client is determined to be an article pickup stage, the target client may be determined to be located at a pickup location; otherwise, it may be determined that the target client is not located at the pickup location.

It should be noted that the description related to the embodiment of the feature extraction scheme may also be applied to the embodiment of the positioning scheme, and this description is not repeated here.

Corresponding to the method embodiment, the specification also provides a corresponding device embodiment.

FIG. 3 is a schematic block diagram of an apparatus provided in an exemplary embodiment. Referring to fig. 3, at the hardware level, the apparatus includes a processor 302, an internal bus 304, a network interface 306, a memory 308, and a non-volatile memory 310, but may also include hardware required for other services. One or more embodiments of the present description may be implemented in software, such as by processor 302 reading a corresponding computer program from non-volatile storage 310 into memory 308 and then executing. Of course, besides software implementation, the one or more embodiments in this specification do not exclude other implementations, such as logic devices or combinations of software and hardware, and so on, that is, the execution subject of the following processing flow is not limited to each logic unit, and may also be hardware or logic devices.

Referring to fig. 4, the feature extraction apparatus for wireless signals may be applied to the device shown in fig. 3 to implement the technical solution of the present specification. Wherein, the feature extraction device of the wireless signal may include:

an obtaining unit 41, configured to obtain a signal feature set of a wireless signal detected by a client of a distribution party during execution of at least one historical distribution service, where distributed items corresponding to the at least one historical distribution service are from the same pickup location, and the signal feature set is used to record a feature value of the wireless signal;

an identification unit 42 identifying a subset of features of the set of signal features corresponding to respective time periods in the execution process, and determining a standard variation trend of the characteristic values of the wireless signals in the respective characteristic subsets in the corresponding time periods, the time period comprises an article pickup stage between the delivery side client terminal arriving at the pickup place and leaving the pickup place in the execution process and a time period when the delivery side client terminal is positioned at other places different from the pickup place in the execution process, the standard variation trend is used to match a variation trend of a target characteristic value of the wireless signal detected by the target client during the execution of the current distribution service, and determining the current time period of the target client according to the matching result, and determining the position relation between the target client and the pickup place according to the current time period.

Optionally, the identification unit 42 is specifically configured to:

the identification unit 42 is specifically configured to: inputting the characteristic values of the wireless signals in each characteristic subset and the labeled label information into a time sequence machine learning model for training to obtain a positioning model;

Optionally, the positioning model is deployed in the target client.

Optionally, the characteristic values recorded by the signal characteristic set are obtained by detecting each wireless signal according to a preset detection period; the obtaining unit 41 is further configured to:

Referring to fig. 5, the positioning apparatus based on wireless signals may be applied to the device shown in fig. 3 to implement the technical solution of the present specification. Wherein the wireless signal based positioning apparatus may include:

an obtaining unit 51, configured to obtain a target characteristic value of a wireless signal detected by a target client during execution of a current distribution service, and determine a matching relationship between a trend of the target characteristic value and standard change trends, where the standard change trend is a trend of a characteristic value of the wireless signal in each characteristic subset in a signal characteristic set, the signal characteristic set is used to record characteristic values of the wireless signal detected by a client of a distributor during execution of at least one historical distribution service, a distribution item corresponding to the at least one historical distribution service is from a same pickup location, each characteristic subset corresponds to each time period in the execution process, the time period includes an item pickup stage from the arrival of the client of the distributor at the pickup location to the departure from the pickup location during the execution process and an item pickup stage from the arrival of the client of the distributor at the pickup location to the departure of the client from the pickup location during the execution process, and the client of the distributor at another location different from the pickup location during the execution process A time period;

the determining unit 52 determines the current time period of the target client according to the matching relationship, so as to determine the position relationship between the target client and the pickup location according to the current time period.

Alternatively to this, the first and second parts may,

the determining unit 52 is specifically configured to: and inputting the target characteristic value of the wireless signal detected by the target client into the positioning model so as to determine the position relation between the target client and the pickup place according to the output result of the positioning model.

Optionally, the positioning model is deployed in the target client.

Optionally, the determining unit 52 is specifically configured to:

Optionally, the determining unit 52 is further configured to:

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. A typical implementation device is a computer, which may take the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email messaging device, game console, tablet computer, wearable device, or a combination of any of these devices.

In a typical configuration, a computer includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used in one or more embodiments of the present description to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of one or more embodiments herein. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.

The above description is only for the purpose of illustrating the preferred embodiments of the one or more embodiments of the present disclosure, and is not intended to limit the scope of the one or more embodiments of the present disclosure, and any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the one or more embodiments of the present disclosure should be included in the scope of the one or more embodiments of the present disclosure.

Claims

1. A method for wireless signal based positioning, comprising:

2. The method of claim 1,

3. The method of claim 1, wherein the eigenvalues of the radio signals within each subset of characteristics are labeled with tag information whether or not they are located at the pickup location within the corresponding time period; wherein the standard variation trend of the characteristic values of the wireless signals in the respective characteristic subsets in the corresponding time period is determined by the following method: inputting the characteristic values of the wireless signals in each characteristic subset and the labeled label information into a time sequence machine learning model for training to obtain a positioning model;

4. The method of claim 3, wherein the positioning model is deployed in the target client.

5. The method of claim 1, wherein the determining the location relationship between the target client and the pickup location according to the current time period comprises:

6. The method of claim 5, further comprising:

7. A method for extracting features of a wireless signal, comprising:

8. The method of claim 7, wherein identifying the subset of features in the set of signal features that correspond to respective time periods in the execution process comprises:

9. The method of claim 7, wherein the eigenvalues of the radio signals within each subset of characteristics are labeled with tag information whether or not they are located at the pickup location within the corresponding time period;

10. The method of claim 9, wherein the positioning model is deployed in the target client.

11. The method according to claim 7, wherein the characteristic value recorded by the signal characteristic set is obtained by detecting each wireless signal according to a preset detection period; the method further comprises the following steps:

12. The method of claim 7, wherein the wireless signals comprise WiFi signals and/or bluetooth signals;

13. A wireless signal based positioning apparatus, comprising:

14. An apparatus for extracting a feature of a wireless signal, comprising:

15. An electronic device, comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor implements the method of any one of claims 1-12 by executing the executable instructions.

16. A computer-readable storage medium having stored thereon computer instructions, which, when executed by a processor, carry out the steps of the method according to any one of claims 1-12.