Active user detection method, device, computer equipment and medium

Abstract

The embodiments of the present application provide a method, device, computer equipment and medium for detecting active users, which belongs to the field of communication technology. The method includes: obtaining multiple access point information and feature sequences of devices to be detected; constructing an indicator variable function according to the active state and feature sequence of the devices to be detected; obtaining a first objective function according to the indicator variable function; conditionally converting the first objective function to obtain a second objective function; for any access point information, performing collaborative detection on the access point information to obtain a local objective function; locally optimizing the local objective function, and iteratively updating the Lagrangian equation to obtain an optimization function and a closed-form solution of the optimization function; gradient updating the dual variable based on the coordinate descent algorithm, the optimization function and the closed-form solution to determine the target active user. The embodiments of the present application can avoid the interference of poor local optimal points and improve the detection accuracy of active users.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a method, device, computer equipment and medium for detecting active users. Background Art

[0002] With the development of communication technology, large-scale machine communication has become the main way to connect massive IoT terminals. In the process of scheduling-free random access, active users directly transmit data on the allocated wireless communication resources. Each user has a unique feature sequence. The access point detects which users are sending feature sequences in the received signal to detect active devices. However, due to the existence of massive IoT devices, the feature sequences of different devices are non-orthogonal. Therefore, it is difficult to achieve accurate active user detection when the access point is interfered by the device, and since Internet devices are usually equipped with cheap crystal oscillators, it is difficult to keep the feature sequence transmission of different users synchronized. In order to solve the problem of active user detection, there are currently two solutions. One is to model the active user detection problem as a compressed sensing problem using the same time slot; the other is to use channel statistical characteristics for active user detection. Both methods can greatly improve the detection performance of active users. However, although the above two methods point out the technical direction for the access method of large-scale machine communication, in reality, since IoT devices are usually equipped with low-cost crystal oscillators, it is difficult to ensure synchronous transmission between different devices. This makes the above research work based on the assumption of ideal synchronous transmission not applicable to the actual asynchronous transmission scenario. Therefore, how the device can improve the detection efficiency of active users in the case of asynchronous transmission is an important research issue at present. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to provide a method, apparatus, computer device and medium for detecting active users, which can avoid interference from poor local optimal points and improve the accuracy of detecting active users.

[0004] To achieve the above objective, a first aspect of an embodiment of the present application provides a method for detecting active users, the method comprising:

[0005] Acquire multiple access point information and feature sequences of devices to be detected, wherein the access point information corresponds to at least one device to be detected;

[0006] constructing an indicator variable function according to the predetermined active state of the device to be detected and the feature sequence;

[0007] Performing maximum likelihood estimation on the indicator variable function to obtain a first objective function of the device to be detected;

[0008] Conditionally transforming the first objective function according to a preset penalty weight to obtain a second objective function;

[0009] For any one of the access point information, collaboratively detect the access point information according to the second objective function to obtain a local objective function of the access point;

[0010] Locally optimizing the local objective function to obtain the Lagrangian equation and the dual variable;

[0011] Iteratively updating the Lagrangian equation according to preset approximate terms to obtain an optimization function and a closed-form solution of the optimization function;

[0012] The dual variable is gradient updated based on a coordinate descent algorithm, the optimization function, and the closed-form solution, and a target active user is determined from the device to be detected according to the updated dual variable.

[0013] In some embodiments, the constructing the indicator variable function according to the predetermined activity state of the device to be detected and the feature sequence includes:

[0014] Obtaining a delay value corresponding to the characteristic sequence;

[0015] Obtaining an equivalent characteristic sequence of the device to be detected according to the characteristic sequence and the time delay value;

[0016] The indicator variable function is constructed according to the equivalent feature sequence and the active state.

[0017] In some embodiments, the access point information includes auxiliary variables; and the collaborative detection of the access point information according to the second objective function to obtain a local objective function of the access point includes:

[0018] Determine a local indicator variable and a constraint condition of an access point according to the auxiliary variable and the equivalent feature sequence;

[0019] Local detection is performed on the plurality of access point information according to the second objective function and the constraint condition to obtain a local objective function of the access point.

[0020] In some embodiments, the local objective function is locally optimized to obtain the Lagrangian equation and the dual variable, including:

[0021] Performing a Lagrangian transformation on the local objective function;

[0022] The local objective function after Lagrangian transformation is constrained according to the local indicator variable and the preset penalty parameter to obtain the Lagrangian equation and the dual variable.

[0023] In some embodiments, the iterative updating of the Lagrangian equation according to the preset approximation term to obtain the optimization function and the closed-form solution of the optimization function includes:

[0024] Adding the preset approximate term to the Lagrangian equation to obtain the optimization function;

[0025] The optimization function is iteratively updated according to the preset approximate term to obtain a closed-form solution of the optimization function.

[0026] In some embodiments, after iteratively updating the Lagrangian equation according to the preset approximation term to obtain the optimization function and the closed-form solution of the optimization function, the method further includes:

[0027] The local indicator variable is decomposed in parallel according to the iteratively updated optimization function to obtain a plurality of sub-parallel formulas.

[0028] In some embodiments, the step of performing gradient updating on the dual variable based on the coordinate descent algorithm, the optimization function, and the closed-form solution, and determining the target active user from the device to be detected according to the updated dual variable, includes:

[0029] Iteratively updating the sub-parallel formula according to a coordinate descent algorithm to obtain an updated local indicator variable;

[0030] Performing gradient update on the dual variable according to the closed-form solution and the updated local indicator variable to obtain a target dual variable;

[0031] A target active user is determined from the device to be detected according to the target dual variable and the local objective function.

[0032] A second aspect of an embodiment of the present application provides a device for detecting active users, the device comprising:

[0033] An information acquisition module, used to acquire information of multiple access points, wherein the access point information includes information of at least one device to be detected;

[0034] A function construction module, used to construct an indicator variable function according to the activity state and delay value of the device to be detected;

[0035] A function calculation module, used for performing maximum likelihood estimation on the indicator variable function to obtain a first objective function of the device to be detected;

[0036] A conditional conversion module, used to perform conditional conversion on the first objective function according to a preset penalty weight to obtain a second objective function;

[0037] a collaborative detection module, configured to perform collaborative detection on the access point information according to the second objective function to obtain a local objective function of the access point;

[0038] A local optimization module, used for locally optimizing the local objective function to obtain the Lagrangian equation and the dual variable;

[0039] An iterative updating module, used for iteratively updating the Lagrangian equation according to preset approximate terms to obtain an optimization function and a closed-form solution of the optimization function;

[0040] A gradient updating module is used to perform gradient updating on the dual variable based on a coordinate descent algorithm, the optimization function and the closed-form solution, and to determine a target active user from the device to be detected according to the updated dual variable.

[0041] A third aspect of an embodiment of the present application provides a computer device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor is used to execute the active user detection method as described in any one of the embodiments of the first aspect of the present application.

[0042] The fourth aspect of the embodiments of the present application proposes a storage medium, which is a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by a computer, the computer is used to execute the active user detection method as described in any one of the embodiments of the first aspect of the present application.

[0043] The active user detection method, device, computer device and medium proposed in the embodiments of the present application first obtain multiple access point information and feature sequences of devices to be detected, and construct an indicator variable function according to the predetermined active state and feature sequence of the devices to be detected, so as to facilitate subsequent collaborative detection, and then perform maximum likelihood estimation on the indicator variable function to obtain a first objective function of the device to be detected, and conditionally convert the first objective function according to a preset penalty weight to obtain a second objective function, thereby avoiding the discontinuous constraint problem caused by asynchronous transmission, and for any access point information, perform collaborative detection on the access point information according to the second objective function to obtain the access point information. The local objective function of the entry point is obtained to realize parallel distributed collaborative detection, and then the local objective local function is locally optimized to obtain the Lagrangian equation and the dual variable, and the Lagrangian equation is iteratively updated according to the preset approximate terms to obtain the optimization function and the closed-form solution of the optimization function, so as to avoid the interference of the poor local optimal point while satisfying the constraints. Finally, the dual variable is gradient updated based on the coordinate descent algorithm, the optimization function and the closed-form solution, and the target active user is determined from the device to be detected according to the updated dual variable, thereby reducing the burden in asynchronous machine communication, reducing the computational complexity, and improving the detection accuracy of active users. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flow chart of a method for detecting active users provided by an embodiment of the present application;

[0045] Figure 2 yes Figure 1 Specific flow chart of step S120;

[0046] Figure 3 yes Figure 1 Specific flow chart of step S150;

[0047] Figure 4 yes Figure 1 Specific flow chart of step S160;

[0048] Figure 5 yes Figure 1 Specific flow chart of step S170;

[0049] Figure 6 is a flow chart of a method for detecting active users provided by another embodiment of the present application;

[0050] Figure 7 yes Figure 1 Specific flow chart of step S180;

[0051] Figure 8 is a schematic diagram of a method for detecting active users provided by a specific example of the present application;

[0052] Fig. 9 is a schematic diagram of a method for detecting active users provided by another specific example of the present application;

[0053] Fig.10 is a schematic diagram of the structure of an active user detection device provided in an embodiment of the present application;

[0054] Fig.11 It is a schematic diagram of the hardware structure of the computer device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0058] The active user detection method provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer or a smart watch, etc.; the server side can be configured as an independent physical server, or a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN) and big data and artificial intelligence platforms; the software can be an application that implements the above method, etc., but is not limited to the above forms.

[0059] Embodiments of the present application can be used in numerous general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer computer devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0060] Please refer to Figure 1 , Figure 1 1 is a flowchart of a specific method of the active user detection method provided in an embodiment of the present application. In some embodiments, the active user detection method includes but is not limited to steps S110 to S180.

[0061] Step S110, obtaining information of multiple access points and characteristic sequences of devices to be detected;

[0062] It should be noted that the access point information corresponds to at least one device to be detected.

[0063] In step S110 of some embodiments, information of multiple access points and feature sequences of devices to be detected are obtained to distribute computing tasks to each access point, thereby reducing the burden of collaborative detection of multiple access points in asynchronous large-scale machine communications.

[0064] It can be understood that each device to be detected has a unique feature sequence. For example, in this embodiment, there are k devices to be detected, and the feature sequence of the device to be detected is recorded as Where L is the length of the feature sequence.

[0065] Step S120, constructing an indicator variable function according to the predetermined activity state and feature sequence of the device to be detected;

[0066] In step S120 of some embodiments, according to the predetermined activity state and feature sequence of the device to be detected Construct indicator variable function b k,t .

[0067] Step S130, performing maximum likelihood estimation on the indicator variable function to obtain a first objective function of the device to be detected;

[0068] In step S130 of some embodiments, the indicator variable function b k,t Perform maximum likelihood estimation to obtain the first objective function f(b) of the device to be detected.

[0069] It can be understood that the first objective function f(b) can be derived based on the maximum likelihood criterion of the covariance, and this implementation does not impose any specific limitation on this.

[0070] It should be noted that the detection of active devices by access points in asynchronous large-scale machine communication can be mathematically equivalent to the detection of b k,t ∈{0,1}. Specifically, when b of the device to be detected k,t When it is detected as 1, it means that device k is in an active state and its delay is T. Therefore, this embodiment can model the active user detection problem in asynchronous large-scale machine communication as shown in the following formula (3):

[0071]

[0072] in, is the constraint condition that the first objective function f(b) needs to satisfy.

[0073] Step S140, conditionally transforming the first objective function according to a preset penalty weight to obtain a second objective function;

[0074] In step S140 of some embodiments, due to the constraint condition || b k The norm constraint of ||0≤1 is discontinuous and non-convex, so the solution of formula (3) is very challenging. It is necessary to conditionally transform the first objective function f(b) according to the preset penalty weight ρ to obtain the second objective function, where the formula of the second objective function is shown as follows (4):

[0075]

[0076] It should be noted that the penalty weight ρ is preset to be greater than 0. When the value of ρ is a sufficiently large finite value, solving formula (4) is equivalent to solving formula (3). Therefore, by solving formula (4), the solution of the original formula (3) can be finally obtained.

[0077] Step S150, for any access point information, collaborative detection is performed on the access point information according to the second objective function to obtain a local objective function of the access point;

[0078] In step S150 of some embodiments, for any access point information, collaborative detection is performed on the access point information according to the second objective function to obtain a local objective function f of the access point. m (xm ), thereby reducing the computational burden caused by multi-access point cooperative detection in asynchronous large-scale machine communications.

[0079] Step S160, locally optimizing the local objective function to obtain the Lagrangian equation and the dual variable;

[0080] In step S160 of some embodiments, the local objective function is locally optimized to obtain the Lagrangian equation And the dual variable This ensures the convergence of the local objective function.

[0081] Step S170, iteratively updating the Lagrange equation according to preset approximate terms to obtain an optimization function and a closed-form solution of the optimization function;

[0082] In step S170 of some embodiments, according to the preset approximation term For the Lagrange equation An optimization function is obtained by iteratively updating the optimization function, and the optimization function is iteratively updated according to a preset approximate term to obtain a closed-form solution of the optimization function, so as to facilitate the subsequent determination of target active users in the device to be detected.

[0083] Step S180 , performing gradient update on the dual variables based on the coordinate descent algorithm, the optimization function and the closed-form solution, and determining the target active user from the devices to be detected according to the updated dual variables.

[0084] In some embodiments, in step S110 to step S180, first, multiple access point information and feature sequences of devices to be detected are obtained, and an indicator variable function is constructed according to the predetermined active state and feature sequence of the device to be detected, so as to facilitate subsequent collaborative detection. Then, the indicator variable function is estimated by maximum likelihood to obtain a first objective function of the device to be detected, and the first objective function is conditionally converted according to a preset penalty weight to obtain a second objective function, thereby avoiding the discontinuous constraint problem caused by asynchronous transmission. For any access point information, the access point information is collaboratively detected according to the second objective function to obtain a local Objective function, thereby realizing parallel distributed collaborative detection, and then locally optimizing the local objective local function to obtain the Lagrangian equation and the dual variable, and iteratively updating the Lagrangian equation according to the preset approximate terms to obtain the optimization function and the closed-form solution of the optimization function, so as to avoid the interference of the poor local optimal point while satisfying the constraints, and finally, based on the coordinate descent algorithm, the optimization function and the closed-form solution, the dual variable is gradient updated, and the target active user is determined from the device to be detected according to the updated dual variable, thereby reducing the burden in asynchronous machine communication, reducing the computational complexity, and improving the detection accuracy of active users.

[0085] Please refer to Figure 2 , Figure 2 1 is a specific flow chart of step S120 provided in an embodiment of the present application. In some embodiments, step S120 specifically includes but is not limited to step S210 and step S230.

[0086] Step S210, obtaining a delay value corresponding to the characteristic sequence;

[0087] Step S220, obtaining an equivalent feature sequence of the device to be detected according to the feature sequence and the delay value;

[0088] Step S230: construct an indicator variable function according to the equivalent feature sequence and the active state.

[0089] In some embodiments, in steps S210 to S220, the characteristics of asynchronous transmission cause the transmission of the characteristic sequence to be accompanied by an unknown delay. Therefore, firstly, the characteristic sequence is obtained. The corresponding delay value t k ∈{0,…,T}, where T represents the maximum possible delay. Then, according to the feature sequence and the delay value, the equivalent feature sequence of the device to be detected is obtained. The equivalent feature sequence is shown in the following formula (1):

[0090]

[0091] In step S230 of some embodiments, an indicator variable function is constructed according to the equivalent feature sequence and the active state of the device to be detected. When the device to be detected k is in the active state, it is recorded as a k =1, otherwise a k = 0, where the indicator variable function b k,t As shown in the following formula (2):

[0092]

[0093] It should be noted that b k,t = 1. Since each device to be detected can have at most one possible delay during each transmission process,

[0094] Please refer to Figure 3 , Figure 3 1 is a specific flow chart of step S150 provided in an embodiment of the present application. In some embodiments, step S150 specifically includes but is not limited to step S310 and step S320.

[0095] It should be noted that the access point information includes the auxiliary variable x m .

[0096] Step S310, determining the local indicator variables and constraint conditions of the access point according to the auxiliary variables and the equivalent feature sequence;

[0097] Step S320: Perform local detection on information of multiple access points according to the second objective function and the constraint conditions to obtain a local objective function of the access point.

[0098] In some embodiments, in step S310 to step S320, according to the auxiliary variable x m and equivalent feature sequences Determine the local indicator variable and constraint conditions of the access point, and perform local detection on multiple access point information according to the second objective function and the constraint conditions to obtain the local objective function f of the access point m (x m ).

[0099] It can be understood that there are M access points in the large-scale machine communication scenario. First, an auxiliary variable x is introduced for the mth access point. m As a local indicator variable, the cooperative detection problem of multiple access points can be described as m (x m ) is shown in the following formula (5):

[0100]

[0101] Among them, f m (x m ) is the local objective function at access point m, f m (x m ) can be derived according to the maximum likelihood criterion based on covariance, st is the constraint condition that needs to be satisfied. Since it is a local function at access point m and only depends on the local received signal of access point m, this makes formula (5) update x when b is fixed. m It is a local optimization problem.

[0102] Please refer to Figure 4 , Figure 4 4 is a specific flow chart of step S160 provided in an embodiment of the present application. In some embodiments, step S160 specifically includes but is not limited to step S410 and step S420.

[0103] Step S410, performing Lagrangian transformation on the local objective function;

[0104] Step S420, constraining the local objective function after Lagrangian transformation according to the local indicator variable and the preset penalty parameter to obtain the Lagrangian equation and the dual variable.

[0105] In some embodiments, in step S410 to step S420, in order to perform a parallel distributed solution to formula (5), firstly Lagrangian transformation is performed on formula (5), and then according to the local indicator variable x m And the preset penalty parameter μ constrains the local objective function after Lagrangian transformation, and the Lagrangian equation is obtained And the dual variable In order to ensure the convergence of formula (5), the specific Lagrangian equation is shown in the following formula (6):

[0106]

[0107] Understandably, Each is an equality constraint x m =b is the dual variable corresponding to , μ>0 is the penalty parameter used to penalize the constraint condition in formula (5).

[0108] Please refer to Figure 5 , Figure 5 1 is a specific flow chart of step S170 provided in an embodiment of the present application. In some embodiments, step S170 specifically includes but is not limited to step S510 and step S520.

[0109] Step S510, adding a preset approximate term to the Lagrange equation to obtain an optimization function;

[0110] Step S520, iteratively update the optimization function according to the preset approximate terms to obtain a closed-form solution of the optimization function.

[0111] In some embodiments, in step S510 to step S520, the preset approximation term Adding to the Lagrange equations The optimization function is obtained, and the optimization function is iteratively updated according to the preset approximate term to obtain a closed-form solution of the optimization function. The specific process of iteratively updating the optimization function according to the preset approximate term is shown in the following formula (7):

[0112]

[0113] It is understandable that since formula (5) is non-smooth and non-convex, the classic ADMM algorithm (Alternating Direction Method of Multipliers) cannot guarantee its convergence to formula (5). Therefore, this application uses the Lagrange equation Adding preset approximations The problem of the optimization function regarding b at the i-th iteration can be expressed as shown in formula (7), where the superscript i-1 is the update value at the i-1th iteration, which is used to control the convergence of the algorithm. Finally, the properties of the infinite norm are used to derive the global optimal closed-form solution of formula (7).

[0114] Please refer to Figure 6 , Figure 6 This is a flowchart of a specific method of the active user detection method provided by another embodiment of the present application, which specifically includes but is not limited to step S610.

[0115] Step S610, performing parallel decomposition on the local indicator variables according to the iteratively updated optimization function to obtain a plurality of sub-parallel formulas.

[0116] In step S610 of some embodiments, the local indicator variable is decomposed in parallel according to the iteratively updated optimization function to obtain multiple sub-parallel formulas, as shown in the following formula (8):

[0117]

[0118] It can be understood that at the i-th iteration, The problem can be decomposed into M parallel sub-problems, thereby reducing the burden caused by multi-access point collaborative detection in asynchronous large-scale machine communications, while ensuring the detection capability and reducing the computational complexity.

[0119] Please refer to Figure 7 , Figure 7 It is a specific flow chart of step S180 provided in an embodiment of the present application. In some embodiments, step S180 specifically includes but is not limited to step S710 and step S730.

[0120] Step S710, iteratively updating the sub-parallel formula according to the coordinate descent algorithm to obtain an updated local indicator variable;

[0121] Step S720, performing gradient update on the dual variable according to the closed-form solution and the updated local indicator variable to obtain the target dual variable;

[0122] Step S730: determining a target active user from the devices to be detected according to the target dual variable and the local objective function.

[0123] In some embodiments, in steps S710 to S730, the sub-parallel formula, that is, formula (8), is iteratively updated according to the coordinate descent algorithm to obtain an updated local indicator variable, and then the dual variable is updated according to the updated local indicator variable and the closed-form solution. Perform gradient update to obtain the target dual variable Finally, the target active user is determined from the device to be detected based on the target dual variable and the local objective function, where the dual variable Perform gradient update to obtain the target dual variable The formula is shown in formula (9):

[0124]

[0125] It is understandable that regarding the primal and dual variables The alternating updating process realizes the parallel and distributed solution of large-scale problems.

[0126] In some embodiments, this embodiment embeds constraints into the local optimization process to find an equivalent penalty function problem. By solving the transformed equivalent problem, the detection results of active users are obtained. This allows the algorithm to avoid bad local optimal points in a smoother way while satisfying the constraints, thereby improving the detection results. At the same time, in order to reduce the burden caused by multi-access point collaborative detection in asynchronous large-scale machine communications, this embodiment reduces the computational complexity while ensuring the detection capability by parallelizing the distributed collaborative detection mechanism.

[0127] In order to more clearly illustrate the process of the active user detection method, a specific example is given below for illustration.

[0128] Example 1:

[0129] Assume that there is a square area of ​​1 square kilometer, with M APs and K IoT devices evenly distributed in the area, and the ratio of active devices to the total number of devices is 0.1. The large-scale path loss conforms to the model {-30.5-36.7log 10 D k,m +Ψ k,m}, where D k,m Represents the distance between user k and access point m. k,m The shadow fading represents a complex Gaussian distribution with a mean of 0 and a variance of 4. In the embodiment of the present application, the relaxed problem is solved by a coordinate descent algorithm, and finally the obtained relaxed solution is projected onto the constraint condition.

[0130] refer to Figure 8 , Figure 8 is an optional schematic diagram of a method for detecting active users provided by a specific example of the present application;

[0131] In some embodiments, the relationship between the probability of missed detection and false alarm of active users under different algorithms is first compared. In this embodiment, the coordinate descent algorithm and the central algorithm are used. Figure 8It can be seen that the algorithm proposed in this embodiment is much better than the baseline algorithm. The reason is that in this embodiment, the delay constraint condition is embedded in the optimization process, and by gradually increasing the penalty factor and solving the corresponding optimization problem, the obtained solution can avoid the bad local optimal point.

[0132] refer to Fig. 9 , Fig. 9 is an optional schematic diagram of a method for detecting active users provided by another specific example of the present application;

[0133] In some embodiments, Fig. 9 It is used to characterize the corresponding relationship between the number of iterations of different algorithms and the probability of false detection. Fig. 9 It can be seen that the effectiveness of the distributed algorithm proposed in the embodiment, with the increase of the number of iterations, the detection ability of the distributed algorithm gradually approaches the centralized algorithm.

[0134] See also Fig.10 The present application also provides an active user detection device, which can implement the above-mentioned active user detection method, and the device includes:

[0135] The information acquisition module 810 is used to acquire information of multiple access points, wherein the access point information includes information of at least one device to be detected;

[0136] A function construction module 820, configured to construct an indicator variable function according to the activity state and delay value of the device to be detected;

[0137] The function calculation module 830 is used to perform maximum likelihood estimation on the indicator variable function to obtain a first objective function of the device to be detected;

[0138] A condition conversion module 840, configured to perform condition conversion on the first objective function according to a preset penalty weight to obtain a second objective function;

[0139] A collaborative detection module 850, configured to perform collaborative detection on the access point information according to the second objective function to obtain a local objective function of the access point;

[0140] A local optimization module 860 is used to locally optimize the local objective function to obtain the Lagrangian equation and the dual variable;

[0141] An iterative updating module 870, used for iteratively updating the Lagrange equation according to a preset approximate term to obtain an optimization function and a closed-form solution of the optimization function;

[0142] The gradient updating module 880 is used to perform gradient updating on the dual variables based on the coordinate descent algorithm, the optimization function and the closed-form solution, and to determine the target active user from the device to be detected according to the updated dual variables.

[0143] The active user and data detection device of the embodiment of the present application is used to execute the active user detection method in the above embodiment. Its specific processing process is the same as the active user detection method in the above embodiment, and will not be repeated here.

[0144] An embodiment of the present application further provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor is used to execute the active user detection method in the above embodiment of the present application.

[0145] Reference Fig.11 , Fig.11 It is a schematic diagram of the hardware structure of the computer device provided in the embodiment of the present application.

[0146] Combine the following Fig.11 The hardware structure of the computer device is described in detail. The computer device includes: a processor 910 , a memory 920 , an input / output interface 930 , a communication interface 940 and a bus 950 .

[0147] The processor 910 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0148] The memory 920 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 920 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 920, and the processor 910 calls and executes the active user detection method of the embodiment of this application;

[0149] Input / output interface 930, used to implement information input and output;

[0150] The communication interface 940 is used to realize the communication interaction between the present device and other devices, and the communication can be realized by wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.); and the bus 950 is used to transmit information between the various components of the device (such as the processor 910, the memory 920, the input / output interface 930 and the communication interface 940);

[0151] The processor 910 , the memory 920 , the input / output interface 930 , and the communication interface 940 are connected to each other in communication within the device via a bus 950 .

[0152] An embodiment of the present application further provides a storage medium, which is a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by a computer, the computer is used to execute the active user detection method in the above embodiment of the present application.

[0153] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0154] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0155] It can be understood by those skilled in the art that Figures 1 to 7 The technical solutions shown in the figure do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than those shown in the figure, or a combination of certain steps, or different steps.

[0156] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0157] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0158] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0159] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0160] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0161] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0162] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0163] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0164] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for detecting active users, characterized in that: The method comprises: Acquire multiple access point information and feature sequences of devices to be detected, wherein the access point information corresponds to at least one device to be detected; constructing an indicator variable function according to the predetermined active state of the device to be detected and the feature sequence; Performing maximum likelihood estimation on the indicator variable function to obtain a first objective function of the device to be detected; Conditionally transforming the first objective function according to a preset penalty weight to obtain a second objective function; For any one of the access point information, collaboratively detect the access point information according to the second objective function to obtain a local objective function of the access point; Locally optimizing the local objective function to obtain the Lagrangian equation and the dual variable; Iteratively updating the Lagrangian equation according to preset approximate terms to obtain an optimization function and a closed-form solution of the optimization function; The dual variable is gradient updated based on a coordinate descent algorithm, the optimization function, and the closed-form solution, and a target active user is determined from the device to be detected according to the updated dual variable.

2. The method for detecting active users according to claim 1, characterized in that: The constructing the indicator variable function according to the predetermined active state of the device to be detected and the feature sequence includes: Obtaining a delay value corresponding to the characteristic sequence; Obtaining an equivalent characteristic sequence of the device to be detected according to the characteristic sequence and the time delay value; The indicator variable function is constructed according to the equivalent feature sequence and the active state.

3. The method for detecting active users according to claim 2, characterized in that: The access point information includes auxiliary variables; and the collaborative detection of the access point information according to the second objective function to obtain a local objective function of the access point includes: Determine a local indicator variable and a constraint condition of an access point according to the auxiliary variable and the equivalent feature sequence; Local detection is performed on the plurality of access point information according to the second objective function and the constraint condition to obtain a local objective function of the access point.

4. The method for detecting active users according to claim 3, characterized in that: The local objective function is locally optimized to obtain the Lagrangian equation and the dual variable, including: Performing a Lagrangian transformation on the local objective function; The local objective function after Lagrangian transformation is constrained according to the local indicator variable and the preset penalty parameter to obtain the Lagrangian equation and the dual variable.

5. The method for detecting active users according to claim 4, characterized in that: The iterative updating of the Lagrangian equation according to the preset approximate term to obtain the optimization function and the closed-form solution of the optimization function includes: Adding the preset approximate term to the Lagrangian equation to obtain the optimization function; The optimization function is iteratively updated according to the preset approximate term to obtain a closed-form solution of the optimization function.

6. The method for detecting active users according to claim 3, characterized in that: After the Lagrangian equation is iteratively updated according to the preset approximate term to obtain the optimization function and the closed-form solution of the optimization function, the method further includes: The local indicator variable is decomposed in parallel according to the iteratively updated optimization function to obtain a plurality of sub-parallel formulas.

7. The method for detecting active users according to claim 6, characterized in that: The step of performing gradient updating on the dual variable based on the coordinate descent algorithm, the optimization function, and the closed-form solution, and determining a target active user from the device to be detected according to the updated dual variable, includes: Iteratively updating the sub-parallel formula according to a coordinate descent algorithm to obtain an updated local indicator variable; Performing gradient update on the dual variable according to the closed-form solution and the updated local indicator variable to obtain a target dual variable; A target active user is determined from the device to be detected according to the target dual variable and the local objective function.

8. An active user detection device, characterized in that: The device comprises: An information acquisition module, used to acquire multiple access point information and a characteristic sequence of a device to be detected, wherein the access point information corresponds to at least one device to be detected; A function construction module, used for constructing an indicator variable function according to the predetermined active state of the device to be detected and the feature sequence; A function calculation module, used for performing maximum likelihood estimation on the indicator variable function to obtain a first objective function of the device to be detected; A conditional conversion module, used to perform conditional conversion on the first objective function according to a preset penalty weight to obtain a second objective function; a collaborative detection module, configured to perform collaborative detection on the access point information according to the second objective function to obtain a local objective function of the access point; A local optimization module, used for locally optimizing the local objective function to obtain the Lagrangian equation and the dual variable; An iterative updating module, used for iteratively updating the Lagrangian equation according to preset approximate terms to obtain an optimization function and a closed-form solution of the optimization function; A gradient updating module is used to perform gradient updating on the dual variable based on a coordinate descent algorithm, the optimization function and the closed-form solution, and to determine a target active user from the device to be detected according to the updated dual variable.

9. A computer device, characterized in that: The computer device includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor is configured to execute the active user detection method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and when the computer program is executed by a computer, the computer is used to execute the active user detection method according to any one of claims 1 to 7.

Images