A living body monitoring management method and system

Abstract

The embodiment of the present specification provides a kind of organism monitoring management method and system, the organism monitoring management method includes: at least one terminal equipment in target room is based on at least one first signal that at least one emission source emits is acquired;At least one second signal is extracted from at least one first signal;At least one second signal is based on, determine whether there is organism in target room;In response to the existence of organism in target room, the biological species of organism is determined based on at least one second signal;Based on biological species, determine the management scheme of target room.

Description

TECHNICAL FIELD

[0001] The present specification relates to the field of room management, and in particular, to a biological body monitoring management method and system. BACKGROUND

[0002] In practical applications, existing biological body monitoring often needs hardware support such as cameras and infrared sensors, and the installation and wiring costs of the equipment are high. Due to privacy factors and other reasons, the application of biological body monitoring in, for example, hotel room management is limited. Moreover, in the prior art, when the personnel in the room have a service demand, they usually need to make a phone call to the business to initiate a service request, which reduces the service efficiency. If the business initiates a phone inquiry or knocks on the door to inquire whether the customer needs service, there may be a situation where the customer is not in the room, thereby wasting the time and energy of the service personnel, and even causing disturbance to the customer, which reduces the customer experience.

[0003] Therefore, it is necessary to provide a biological body monitoring management method and system to achieve more efficient and accurate biological body monitoring management with low cost, so as to avoid wasting the service personnel, improve the service quality and efficiency, and thus improve the customer experience. SUMMARY

[0004] One of the embodiments of the present specification provides a biological body monitoring management method. The biological body monitoring management method comprises: acquiring at least one first signal emitted by at least one emission source based on at least one terminal device in a target room; extracting at least one second signal from the at least one first signal; determining whether a biological body exists in the target room based on the at least one second signal; in response to the existence of the biological body in the target room, determining a biological species of the biological body based on the at least one second signal; and determining a management scheme for the target room based on the biological species.

[0005] One of the embodiments of the present specification provides a biological body monitoring management system, comprising: an acquisition module configured to acquire at least one first signal emitted by at least one emission source based on at least one terminal device in a target room; an extraction module configured to extract at least one second signal from the at least one first signal; a determination module configured to determine whether a biological body exists in the target room based on the at least one second signal; in response to the existence of the biological body in the target room, determine a biological species of the biological body based on the at least one second signal; and determine a management scheme for the target room based on the biological species.

[0006] One of the embodiments of the present specification provides a computer readable storage medium, which stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the above-mentioned biological body monitoring management method. BRIEF DESCRIPTION OF DRAWINGS

[0007] The present specification will be further described in the manner of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. The embodiments are not restrictive, and in the embodiments, the same numbers refer to the same structures or operations, in which:

[0008] Figure 1 is a schematic diagram of an application scenario of a living body monitoring management system according to some embodiments of the present specification;

[0009] Figure 2 is an exemplary block diagram of a living body monitoring management system according to some embodiments of the present specification;

[0010] Figure 3 is an exemplary flowchart of a living body monitoring management method according to some embodiments of the present specification;

[0011] Figure 4A is an exemplary schematic diagram of a first prediction model according to some embodiments of the present specification;

[0012] Figure 4B is an exemplary schematic diagram of a second prediction model according to some embodiments of the present specification;

[0013] Figure 5 is an exemplary schematic diagram of a third prediction model according to some embodiments of the present specification;

[0014] Figure 6 is an exemplary schematic diagram of a fourth prediction model according to some embodiments of the present specification;

[0015] Figure 7 is an exemplary schematic diagram of a fifth prediction model according to some embodiments of the present specification; DETAILED DESCRIPTION

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can be applied to other similar scenarios without creative labor. Unless it is obvious from the language environment or otherwise stated, the same reference numbers in the drawings represent the same structure or operation.

[0017] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0018] As used in the specification and claims, the words "may" and "could" are used in the sense that they are not exclusive, but rather, they are used in a sense that they mean "possibly", "there is a possibility of", "it is possible that" and / or "it is possible, but not necessarily true, that". As used in the specification and claims, the words "include" and "comprise" are used in the sense of "including but not limited to". As used in the specification and claims, the words "a" and "an" are used in the sense that they mean "one or more" and / or "at least one".

[0019] Flowcharts are used in the specification to illustrate the operation of systems in accordance with embodiments of the specification. It will be understood that the acts shown in the figures are not necessarily performed in the order shown. Rather, the acts can be performed in any order, or in parallel, or in any combination. Other acts can also be added to, or removed from, the processes.

[0020] Figure 1 is a schematic diagram of an application scenario of a living body monitoring management system according to some embodiments of the specification.

[0021] In some embodiments, the application scenario of the living body monitoring management system can be used in various places for living body monitoring, such as service management of an apartment, a house or a hotel room, store monitoring, intrusion detection, old people care, intelligent shopping, etc. The living body monitoring management system can realize living body monitoring and management by implementing the living body monitoring management method disclosed in the present application.

[0022] In some embodiments, as shown in Figure 1 The living body monitoring management system 100 can include a transmitting source 110, a living body 120, a terminal device 130, a processing device 140, a network 150, and a storage device 160.

[0023] The transmitting source 110 can be a device for transmitting a signal. For example, the transmitting source 110 can be used to transmit a first signal. In some embodiments, the signal can include a Wi-Fi signal, etc. In some embodiments, the transmitting source 110 can include a Wi-Fi signal amplifier, a router, etc. In some embodiments, multiple transmitting sources can be included in an area, and the multiple transmitting sources can be arranged at multiple positions in the area. For example, one transmitting source can be arranged at each of the positions such as a front desk in a store, a warehouse, etc.

[0024] The living body 120 can be any living individual or object. In some embodiments, the living body 120 can include a mouse 120-1, a human 120-2, etc., or any combination thereof. In some embodiments, the living body 120 can also include other living beings, such as a cat, a dog, a bird, a snake, etc.

[0025] The terminal device 130 refers to one or more devices that can receive the signal transmitted by the transmission source 110. For example, the terminal device 130 can receive the first signal transmitted by the transmission source 110. In some embodiments, the terminal device 130 can include one or any combination of a mobile device 130-1, a tablet computer 130-2, a laptop computer 130-3, a desktop computer 130-4, and other devices with input and / or output functions. In some embodiments, the mobile device 130-1 can include a wearable device and a smart mobile device, or any combination thereof. In some embodiments, the smart mobile device can include a smart phone, a personal digital assistant (PDA), a game device, a navigation device, a handheld terminal (POS), or any combination thereof. In some embodiments, the terminal device 130 can also include a smart home appliance (not shown in the figure). For example, a smart refrigerator, a smart washing machine, a smart air conditioner, a smart sound, a smart projector, and other devices with wireless signal transmission functions.

[0026] It should be noted that the above examples are only used to illustrate the generality of the terminal device 130, but not to limit the scope thereof.

[0027] The processing device 140 can be used to process information and / or data related to the organism monitoring management system 100. The processing device 140 can execute program instructions based on the data, information, and / or processing results to perform one or more functions described in this specification. In some embodiments, the processing device 140 can obtain information and / or data sent by the terminal device 130 via the network 150. For example, the processing device 140 can obtain the first signal received by the terminal device 130 through the network 150. For another example, the processing device 140 can process the first signal to extract the second signal.

[0028] In some embodiments, the processing device 140 can include one or more sub-processing devices (e.g., single-core processing devices or multi-core multi-core processing devices). In some embodiments, the processing device 140 can be a single server or a group of servers. The group of servers can be a centralized group of servers connected to the network 150 via an access point, or a distributed group of servers connected to the network 150 via one or more access points, respectively.

[0029] The network 150 can connect the components of the system and / or connect the system with external resource parts. The network 150 can enable communication between the components and between other parts outside the system, facilitating exchange of data and / or information. In some embodiments, one or more components in the organism monitoring management system 100 (e.g., the terminal device 130 and the processing device 140) can exchange information and / or data through the network 150. For example, the terminal device 130 can send the first signal to the processing device 140 through the network 140.

[0030] In some embodiments, the network 150 can be any one or more of a wired network or a wireless network. In some embodiments, the network can be of various topologies or combinations of topologies, such as point-to-point, shared, hub-and-spoke, etc. In some embodiments, the network 150 can include one or more network access points. For example, the network 150 can include wired or wireless network access points, such as base stations and / or network switching points, through which one or more components of the organism monitoring management system 100 can connect to the network 150 to exchange data and / or information.

[0031] The storage device 160 can be used to store data and / or instructions. In some embodiments, the storage device 160 can store information and / or data acquired by the terminal device 130. For example, the storage device 160 can store the first signal acquired from the terminal device 130. In some embodiments, the storage device 160 can store data and / or instructions used by the processing device 140 to perform or use to complete the exemplary organism monitoring management method described in this specification. The storage device 160 can include one or more storage components, each of which can be a separate device or part of other devices. In some embodiments, the storage device 160 can include random access memory (RAM), read-only memory (ROM), mass storage, removable storage, volatile read-write memory, etc. or any combination thereof. In some embodiments, the storage device 160 can be implemented on a cloud platform.

[0032] It should be noted that the application scenarios are provided only for illustrative purposes and are not intended to limit the scope of this specification. Those of ordinary skill in the art can make various modifications or changes based on the description of this specification. For example, the application scenarios can also include storage devices. For another example, the application scenarios can be implemented on other devices to achieve similar or different functions. However, changes and modifications will not depart from the scope of this specification.

[0033] Figure 2 is an exemplary block diagram of an organism monitoring management system according to some embodiments of the present specification. In some embodiments, the organism monitoring management system 200 can include an acquisition module 210, an extraction module 220, and a determination module 230.

[0034] The acquisition module 210 is configured to acquire at least one first signal emitted by at least one emission source based on at least one terminal device in a target room.

[0035] The extraction module 220 is configured to extract at least one second signal from the at least one first signal.

[0036] More information about the first signal and the second signal can be found inFigure 3 and related descriptions.

[0037] The determining module 230 is configured to determine, based on the at least one second signal, whether a living body exists in the target room; in response to the living body existing in the target room, determine, based on the at least one second signal, a biological species of the living body; and determine, based on the biological species, a management scheme for the target room.

[0038] In some embodiments, to determine, based on the at least one second signal, whether a living body exists in the target room, the determining module 230 is configured to determine, based on a signal change rate of the at least one second signal, whether a living body exists in the target room. More details about determining whether a living body exists can be found in Figure 3 and related descriptions.

[0039] In some embodiments, to determine, in response to the living body existing in the target room, based on the at least one second signal, the biological species of the living body, the determining module 230 is configured to determine the biological species by processing the at least one second signal through a machine learning model. More details about determining the biological species based on the machine learning model can be found in Figure 4A , Figure 4B , Figure 5 , Figure 6 , Figure 7 and related descriptions.

[0040] It should be noted that the above description of the living body monitoring management system 200 and its modules is for the convenience of description, and cannot limit the scope of the embodiments. It can be understood that, for those skilled in the art, after understanding the principle of the system, any combination of the modules or connection of the modules to other modules can be made without departing from the principle. In some embodiments, Figure 2 The acquisition module 210, the extraction module 220 and the determining module 230 disclosed in the above embodiments can be different modules in a system, or one module can implement the functions of two or more modules. For example, the modules can share one storage module, or each module can have its own storage module. Such variations are within the scope of the present disclosure.

[0041] Figure 3 is an exemplary flowchart of a living body monitoring management method according to some embodiments of the present disclosure. The process 300 can be performed by the living body monitoring management system 100 or the living body monitoring management system 200. For example, the process 200 can be stored in the form of instructions in a storage medium (e.g., the storage device 160), and the process device 140 and / or Figure 2The modules in the memory 130 can execute the instructions, and the processing device 140 and / or the modules can be configured to perform the process 300 when executing the instructions. The operations of the process 300 shown below are for illustration purposes only. In some embodiments, the process 300 can be completed with one or more additional operations not described, and / or without one or more of the operations discussed. Additionally, Figure 3 The order of the operations of the process 300 shown in FIG. 3 and described below is not limiting.

[0042] At S310, at least one first signal transmitted by at least one transmission source is obtained based on at least one terminal device in the target room. In some embodiments, S310 can be performed by the obtaining module 210.

[0043] The target room is a space in which the user needs to perform the biological body monitoring. For example, the target room can be a residential room, a hotel room, a shop, etc.

[0044] The first signal can refer to a signal transmitted by a transmission source in the target room. For example, the first signal can be a Wi-Fi signal transmitted by a router in the target room. Each first signal can include a plurality of subcarriers. For example, the first signal A can be represented as A = [A1, A2, A3, …, An], where n is the number of subcarriers, and each A i (Ai∈[1, n]) represents the amplitude and phase of a subcarrier.

[0045] In some embodiments, the first signal can be a signal with path loss. Due to different propagation paths and distances between the transmission source and the terminal device (transmission source and terminal device), the path loss of the signal is different, and accordingly, the first signal received by different terminal devices is different.

[0046] In some embodiments, the first signal can be received by at least one terminal device in the target room. For example, the first signal can be received by a desktop computer in the target room. For another example, the first signal can be received by a smart sound in the target room. Each terminal device in the target room can receive a first signal. In some embodiments, the obtaining module 210 can communicate with each of the at least one terminal device in the target room to obtain the first signal received by each terminal device. For example, the obtaining module 210 can obtain the first signal received by the terminal device 130 via the network 150.

[0047] At S320, at least one second signal is extracted from the at least one first signal. In some embodiments, S320 can be performed by the extracting module 220.

[0048] The second signal can refer to a signal obtained after the first signal undergoes multipath fading during propagation. For example, the second signal can be a partial signal of the first signal that is attenuated due to encountering an obstacle (e.g., the signal is diffracted, reflected, scattered).

[0049] In some embodiments, the second signal can include a channel state information (CSI) signal, which is used to describe the combined effects of reflection, diffraction, scattering, and other signal fading effects in the signal channel. A complete CSI signal can be described by three dimensions: time, carrier frequency, and space, which correspond to the signal changes experienced by the channel with different time, carrier frequency, and spatial distribution, respectively. That is, a complete CSI signal can be represented by a three-dimensional signal matrix. For a terminal device at a fixed position, the CSI signal can be regarded as a two-dimensional matrix, which corresponds to the changes experienced by the wireless channel with different time and carrier frequency. Signal fading (e.g., including path loss and multipath fading) can be reflected by the amplitude, phase, and the like of the signal.

[0050] In some embodiments, the extraction module 220 can obtain the second signal in various ways.

[0051] For example, the extraction module 220 can extract at least one second signal from at least one first signal based on a channel impulse response (CIR). An exemplary channel impulse response can be represented by the following equation (1):

[0052] H(k) = ||H(k)||e j∠H(k) (1)

[0053] where H(k) represents the channel state information (CSI) of the kth subcarrier, ||H(k)|| represents the amplitude of the kth subcarrier, ∠H(k) represents the phase of the kth subcarrier, and j is a coefficient.

[0054] For another example, the extraction module 220 can extract at least one second signal from at least one first signal based on a relationship between a transmitting end (i.e., a transmitting source) and a receiving end (i.e., a terminal device). An exemplary relationship between the transmitting end (i.e., the transmitting source) and the receiving end (i.e., the terminal device) can be represented by the following equation (2):

[0055] Y = HX + N (2)

[0056] YX+H+N, wherein Y is a signal vector of the terminal device, X is a signal vector of the transmission source, H represents a signal matrix of the CSI signal, and N is a Gaussian white noise. That is, by substituting the signal vector of the terminal device, the signal vector of the transmission source, and the Gaussian white noise into the above formula (2), the second signal can be extracted from the first signal. The signal vector can be constructed based on the amplitude characteristics and the phase characteristics of the first signal.

[0057] In some embodiments, the extraction module 220 can extract one second signal from the first signal received by each terminal device, to obtain at least one second signal.

[0058] In some embodiments, the extraction module 220 can also perform low-pass filtering processing on the extracted second signal, to reduce environmental noise and original noise, and to improve the clarity and stability of the second signal.

[0059] S330, based on the at least one second signal, determining whether there is a living body in the target room. In some embodiments, S330 can be performed by the determination module 230.

[0060] In some embodiments, the determination module 230 can determine, by a preset rule, whether there is a living body in the target room based on the at least one second signal. In some embodiments, an exemplary preset rule can be: performing dimensionality reduction and noise reduction processing on the second signal by using a principal component analysis (PCA) method to obtain feature data of the second signal; then, based on the feature data of the second signal, matching with a reference vector library to determine a reference vector with the highest matching degree with the feature data of the second signal, and determining a reference result corresponding to the reference vector as the final result of whether there is a living body in the target room. The reference vector library includes a plurality of reference vectors corresponding to the feature data of the signals and reference results corresponding to the reference vectors (i.e., results of whether there is a living body). The reference vector library can be pre-set based on historical data.

[0061] In some embodiments, the determination module 230 can determine, based on a signal change rate of the at least one second signal, whether there is a living body in the target room.

[0062] The signal change rate of the second signal can be used to reflect the fluctuation level of the second signal in a unit of time. For example, the signal change rate of the second signal can be 10%. For another example, the signal change rate of the second signal includes (AC1, AC2, AC3), wherein AC1 is the signal change rate of the second signal C1, AC2 is the signal change rate of the second signal C2, and AC3 is the signal change rate of the second signal C3.

[0063] In some embodiments, the determining module 230 can determine the signal change rate of the second signal in various ways. For example, the signal change rate of the second signal can be determined by the following formula (3):

[0064]

[0065] wherein k is the signal change rate of the second signal, t1 is the time of the first occurrence (the first time), t2 is any time after the first time (the second time), represents the signal difference between the first time and the second time. The signal difference between the first time and the second time can include the difference between the amplitude of the second signal at the first time and the amplitude of the second signal at the second time, or the difference between the phase feature of the second signal at the first time and the phase feature of the second signal at the second time. The time interval is the time interval between the first time and the second time. The time interval of the first time and the second time can be a system default value, or can be set by a person.

[0066] In some embodiments, the signal change rate of the second signal can be inversely proportional to the matrix similarity between the signal matrix of the second signal at the first time (hereinafter referred to as the first signal matrix) and the signal matrix of the second signal at the second time (hereinafter referred to as the second signal matrix). For example, the higher the matrix similarity, the smaller the signal change rate of the second signal. The matrix similarity can be determined in various ways. For example, the matrix similarity can be determined by a similarity calculation model. For example, the similarity calculation model can be used to process the first signal matrix and the second signal matrix respectively to determine the matrix similarity of the two.

[0067] In some embodiments, the determining module 230 can determine whether there is a living body in the target room based on a sum of signal change rates of the at least one second signal. For example, the determining module 230 can determine a signal change rate of each of the at least one second signal, determine the sum of the signal change rates of the at least one second signal, and determine whether there is a living body in the target room according to a relationship between the sum of the signal change rates and a change rate threshold. For example, if the sum of the change rates exceeds the change rate threshold by 20%, it can be determined that there is a living body in the target room. In some embodiments, the determining module 230 can determine the sum of the signal change rates of the at least one second signal by weighting and summing the signal change rates of each of the at least one second signal by setting weights for the signal change rates of different second signals. The weights corresponding to the signal change rates of different second signals can be manually set according to the degree of environmental interference of the corresponding terminal device. For example, a terminal device near a window can be more susceptible to environmental interference such as wind, so the weight of the signal change rate of the second signal extracted from the first signal received by the terminal device can be lower (e.g., set to 0.3); the degree of environmental interference of a terminal device away from the window is smaller, so the weight of the signal change rate of the second signal extracted from the first signal received by the terminal device can be higher (e.g., set to 0.7).

[0068] In one or more embodiments of the present specification, the sum of the signal change rates is used to comprehensively determine whether there is a living body in the target room, which can eliminate the influence of environmental factors on individual signal change rates; and the weights of terminal devices in different positions are set according to the degree of environmental interference of the positions, which can further make the living body recognition result more accurate.

[0069] In some embodiments, the determining module 230 can also determine whether there is a living body in the target room based on a signal difference between the second signal and a standard signal.

[0070] The standard signal can refer to a second signal extracted from a first signal received by a terminal device in an empty room (i.e., a room containing no living body). Each target room can correspond to a standard signal. For example, for a target room A, a second signal can be extracted from a first signal received by a terminal device A in the target room A when there is no living body in the target room A, to obtain a standard signal corresponding to the target room A.

[0071] In some embodiments, the signal difference can include a phase difference and an amplitude difference of the signals. For example, the phase difference can be 2° and the amplitude difference can be 20A.

[0072] In some embodiments, the determining module 230 can determine whether there is a living body in the target room based on a relationship between the signal difference between the at least one second signal and the standard signal and a difference threshold. When the signal difference is greater than the difference threshold, the determining module 230 can determine that there is a living body in the target room. The difference threshold can be a system default value, an empirical value, a human pre-set value, or any combination thereof, and can be set according to actual needs, which is not limited in the specification.

[0073] In one or more embodiments of the specification, the determination of the presence of a living body in the target room through the signal difference between the second signal and the standard signal can avoid the problem that the living body cannot be identified when it is stationary in the room, and can further improve the accuracy and sensitivity of the living body monitoring.

[0074] S340, in response to the presence of a living body in the target room, determining the biological species of the living body based on the at least one second signal. In some embodiments, step S340 can be performed by the determining module 230.

[0075] In some embodiments, the biological species of the living body can include, but is not limited to, a human, a mouse, a cat, a dog, a bird, a snake, etc.

[0076] In some embodiments, the determining module 230 can determine the biological species of the current living body through the historical biological species corresponding to the signal change rate of the historical second signal and / or the historical biological species corresponding to the signal difference between the historical second signal and the standard signal. For example, assuming that the corresponding biological species is a mouse when the signal change rate of the historical second signal is 25%-30%, the biological species of the current living body can be determined to be a mouse when the signal change rate value of the current second signal is 26%. For another example, the corresponding biological species is a human when the phase difference between the historical second signal and the standard signal is 2°-5°, and the biological species of the current living body can be determined to be a human according to the phase difference between the current second signal and the standard signal being 4°.

[0077] In some embodiments, in response to the presence of a living body in the target room, the determining module 230 can process the at least one second signal through a machine learning model to determine the biological species. More details about determining the biological species based on the machine learning model can be found in the following content.

[0078] S350, determining a management scheme for the target room based on the biological species. In some embodiments, S350 can be performed by the determining module 230.

[0079] The management scheme for the target room refers to the action measures taken for the target room. For example, actively calling to inquire whether the room personnel need service, etc.

[0080] In some embodiments, the determining module 230 can determine the management scheme for the target room based on the biological species in various ways. For example, when the target room is a hotel room, if the determining module 230 identifies that the biological species of the organism is human, it can inquire whether room cleaning or meal service is needed; when the target room is a shop, if the determining module 230 identifies that the biological species of the organism is a mouse, the shop is managed to drive away the mouse, and if the determining module 230 identifies that the biological species of the organism is a cat, no action is taken.

[0081] The description of determining the management scheme in the embodiments of the present specification is only for the purpose of illustration and is not intended to limit the scope of the present specification. The management scheme for the target room can also be determined in other ways. For example, the above-mentioned organism monitoring and management system can also include an image recognition module, and the organism identified by the image recognition module is further used to determine whether the management scheme is to be implemented for the target room.

[0082] In some embodiments of the present specification, by determining whether an organism exists in the target room and determining the biological species based on the signals received by the terminal device in the target room, the management scheme for the target room can be determined according to the biological species, which can achieve more efficient and accurate, low-cost organism monitoring and management, and at the same time, the corresponding management scheme can be determined according to the biological species to provide targeted processing services for the corresponding target room.

[0083] Figure 4A is an exemplary schematic diagram of a first prediction model according to some embodiments of the present specification.

[0084] In some embodiments, the determining module 230 can process the signal change rate of the at least one second signal, the first position distribution of the at least one terminal device, and the second position distribution of the at least one emission source based on the first prediction model to determine the biological species. More information about the signal change rate of the second signal can be found in Figure 2 and the related description.

[0085] The first position distribution is the position distribution of the at least one terminal device relative to the target room. For example, the first position distribution can be represented as [R1(x1, y1, z1), R2(x2, y2, z2), R3(x3, y3, z3)]. Wherein (x1, y1, z1) is the position coordinate of the terminal device R1, (x2, y2, z2) is the position coordinate of the terminal device R2, and (x3, y3, z3) is the position coordinate of the terminal device R3.

[0086] The second position distribution is the position distribution of at least one transmitting source relative to the target room. For example, the second position distribution can be represented as [T1(x1', y1', z1'), T2(x2', y2', z2'), T3(x3', y3', z3')]. Where (x1', y1', z1') are the position coordinates of transmitting source T1, (x2', y2', z2') are the position coordinates of transmitting source T2, and (x3', y3', z3') are the position coordinates of transmitting source T3.

[0087] The first location distribution of at least one terminal device and the second location distribution of at least one transmitter can be obtained in various ways. For example, the methods for obtaining the location distribution may include, but are not limited to, proximity detection, centroid positioning, and polygon positioning. As an example, the center of the room can be determined as the coordinate origin using centroid positioning based on multiple second signals, and the location distribution of the terminal device and the transmitter relative to the room can be determined using proximity detection based on the coordinate origin and multiple second signals.

[0088] The primary predictive model can be a machine learning model used to determine the species of organism. For example, the type of primary predictive model can be a recurrent neural network (RNN) model, a deep neural network (DNN) model, a convolutional neural network (CNN) model, etc.

[0089] In some embodiments, such as Figure 4A As shown, the inputs to the first prediction model 420 may include the signal change rate 410-1 of at least one second signal, the first location distribution 410-2 of at least one terminal device, and the second location distribution 410-3 of at least one transmitter. The output may be the biological species 430 of the organism. In some embodiments, the output of the first prediction model may also be a letter code for the biological species; for example, output P represents the biological species as human, and output M represents the biological species as mouse. In some embodiments, the output of the first prediction model may also include the biological species. For example, the output of the first prediction model may be ([M,2], [P,3]), indicating that two mice and three people were detected in the target room.

[0090] In some embodiments, such as Figure 4A As shown, the input to the first prediction model 420 also includes location features 410-4.

[0091] Location features can reflect the positional relationships between various transmitters, terminal devices, and items within a room. For example, location features could be "table near the window" or "terminal device A near the table." Location features can be obtained through pre-recorded manual observation.

[0092] In some embodiments, the first prediction model 420 can be trained by a large number of first training samples 440-1 with first labels. An exemplary training process includes: inputting the first training samples 440-1 with the first labels into an initial first prediction model 440-2, constructing a loss function by the first labels and the results of the initial first prediction model 440-2, and iteratively updating the parameters of the initial first prediction model based on the loss function by gradient descent or other methods. When a preset condition is met, the model training is completed, and a trained first prediction model is obtained. The preset condition can be that the loss function converges, the number of iterations reaches a threshold, etc.

[0093] In some embodiments, the first training sample can be a signal change rate of the sample second signal, a sample first position distribution of the plurality of sample terminal devices, a sample second position distribution of the plurality of sample emission sources, and a sample position feature. The plurality of sample terminal devices and the plurality of sample emission sources are located in the same sample room. The first label can include the actual biological species in the sample room corresponding to the signal change rate of the sample second signal, the sample first position distribution of the plurality of sample terminal devices, the sample second position distribution of the plurality of sample emission sources, and the sample position feature.

[0094] In some embodiments, the first training sample and the first label can be obtained by simulating the movement of the sample biological body in different sample rooms using a simulation model. The sample room includes a plurality of sample terminal devices and a plurality of sample emission sources, and the first position distribution of the plurality of sample terminal devices and the second position distribution of the plurality of sample emission sources can be artificially preset. For example, a simulation model of a mouse can be made according to the contour, height, and volume of the mouse, and the simulation model can be moved in different sample rooms by remote control. The movement mode can include random movement or movement along each path. When the simulation model moves in a certain sample room, the second signal is obtained and the signal change rate of the second signal is determined. The obtained signal change rate of the second signal, the first position distribution of the plurality of sample terminal devices, and the second position distribution of the plurality of sample emission sources are used as the first training sample, and the biological species of the mouse is used as the first label of the first training sample.

[0095] In some embodiments, the first training sample and the first label can also be obtained by a simulation experiment with a real person. For example, a certain number of people can move in different sample rooms. When the certain number of people move in a certain sample room, the second signal is obtained and the signal change rate of the second signal is determined. The obtained signal change rate of the second signal, the first position distribution of the plurality of sample terminal devices, and the second position distribution of the plurality of sample emission sources are used as the first training sample, and the biological species of the people is used as the first label of the first training sample.

[0096] In some embodiments, the second signal can also be acquired when the sample room does not contain any organism, and the signal change rate of the acquired second signal is determined. The signal change rate of the acquired second signal, the first position distribution of the plurality of sample terminal devices, and the second position distribution of the plurality of sample emission sources are taken as the first training sample, and the biological species is taken as the first label of the first training sample. Accordingly, the first prediction model can be used to determine whether there is an organism in the target room. For example, when the model output is 0, it can be determined that there is no organism in the target room; when the model output is M, it can be determined that there is an organism in the target room, and the biological species is a mouse.

[0097] One or more embodiments of the present specification process the signal change rate of the second signal detected in the room, and the position distribution of the terminal device and the emission source in the room through the machine learning model, which is beneficial to quickly and accurately determine the biological species. At the same time, the training sample of the machine learning model is acquired through different simulation methods, and different methods can be used to acquire the training sample for different biological species, so that the training sample is more diverse, which provides flexible selection for acquiring the training sample. Through acquiring a large number of training samples, the model training result can be more accurate.

[0098] Figure 4B is an exemplary schematic diagram of a second prediction model according to some embodiments of the present specification.

[0099] In some embodiments, the determination module 230 can determine the biological species based on the second prediction model processing at least one second signal, at least one standard signal, a first position distribution of at least one terminal device, and a second position distribution of at least one emission source. More about the second signal and the standard signal can be referred to Figure 2 and the related description. More about the first position distribution and the second position distribution can be referred to Figure 4A and the related description.

[0100] The second prediction model can be a machine learning model for determining the biological species. For example, the type of the second prediction model can be an RNN model, a DNN model, a CNN model, etc.

[0101] In some embodiments, as Figure 4BAs shown, the input of the second prediction model 460 can be at least one second signal 450-1, at least one standard signal 450-2, at least one first position distribution of the terminal device 450-3, at least one second position distribution of the emission source 450-4, and the output can be the biological species 470 of the living body. Wherein, each second signal can correspond to one standard signal. Similar to the first prediction model, the output of the second prediction model can also be the letter code of the biological species, for example, the output P represents the biological species is human, and the output M represents the biological species is mouse.

[0102] In some embodiments, as shown in FIG. 4B, the input of the second prediction model 460 further includes the position feature 450-5. More details about the position feature can be found in Figure 4B Figure 4A

[0103] In some embodiments, the second prediction model 460 can be obtained by training a large number of second training samples 480-1 with second labels. An exemplary training process includes: inputting the second training sample 480-1 with the second label into the initial second prediction model 480-2, constructing a loss function by the second label and the result of the initial second prediction model 480-2, and iteratively updating the parameters of the initial second prediction model based on the loss function by gradient descent or other methods. When the preset condition is met, the model training is completed, and the trained second prediction model is obtained. Wherein, the preset condition can be that the loss function converges, the number of iterations reaches a threshold, etc.

[0104] In some embodiments, the second training sample can include a sample second signal, a sample standard signal corresponding to the sample second signal, a sample first position distribution of a plurality of sample terminal devices, a sample second position distribution of a plurality of sample emission sources, and a sample position feature. The second label can include the actual biological species in the sample room corresponding to the sample second signal, the sample standard signal, the sample first position distribution of the plurality of sample terminal devices, the sample second position distribution of the plurality of sample emission sources, and the sample position feature.

[0105] In some embodiments, the second training sample and the second label can be obtained by simulating the movement of the sample living body in different sample rooms by the simulation model. The acquisition method of the second training sample and the second label is similar to the acquisition method of the first training sample and the first label. More details can be found in Figure 4A

[0106] One or more embodiments of the present specification process the second signal detected in the room, the standard signal corresponding to the second signal, the position distribution of the terminal device and the emission source in the room, and the position feature by the machine learning model, which is beneficial to quickly and accurately determine the biological species. ​​​

[0107] Figure 5 is an exemplary schematic diagram of a third prediction model according to some embodiments of the present specification.

[0108] In different biological monitoring application scenarios, the terminal device of different target rooms is sometimes one and sometimes multiple. In some embodiments, in response to the terminal device being a single one, the determining module 230 can determine a differential sequence of the at least one second signal received by the single terminal device; and determine the biological species based on processing the differential sequence by the third prediction model.

[0109] The differential sequence of the second signal is a sequence composed of signal difference values of the second signal corresponding to multiple first times and second times. The description about the first times and the second times can be referred to Figure 2 . For example, the differential sequence of the second signal can be represented as (Δφ T2-T1 , Δφ T3-T2 , …). Wherein, in T2 and T1, T2 is the second time, and T1 is the first time; in T2 and T3, T3 is the second time, and T2 is the first time; Δφ T2-T1 represents the signal difference value of the second signal at T1 and T2, Δφ T3-T2 represents the signal difference value of the second signal at T2 and T3, and so on. In some embodiments, the differential sequence can be a multi-dimensional sequence, for example, one dimension can be used to represent the amplitude difference of the signal, one dimension can represent the phase difference of the signal, and so on.

[0110] The third prediction model can be another machine learning model for determining the biological species. For example, the type of the third prediction model can be a CNN model, an RNN model, a DNN model, and the like.

[0111] In some embodiments, as shown in Figure 5 , the input of the third prediction model 520 includes the differential sequence 510, and the output includes the biological species 530. In some embodiments, as shown in Figure 5 , the input of the third prediction model 520 can also include the room type 540 and the room layout 550 of the target room.

[0112] In some embodiments, the room type of the target room can include but is not limited to a hotel room, a store, a warehouse, a restaurant, and the like. In some embodiments, the room layout of the target room can include room features such as the floor area size, the floor height, the space occupation in the room, the size, the placement, and the material of the items in the room, and the like. The room type and the room layout of the target room can be determined by human input. For example, the room type of the target room A can be manually input as a hotel room, and the room layout can be manually input as a small house, the floor area size is 15m 2High 3.5m, including a table by the window, bed, wardrobe, TV, etc.

[0113] In some embodiments, the third prediction model can include a plurality of processing layers. For example, as shown in FIG. 5, the third prediction model 520 can include a first embedding layer 521 and a first prediction layer 523. Figure 5

[0114] The first embedding layer can be a machine learning model for processing the differential sequence of the second signal to determine the first feature vector. For example, the first embedding layer can be a CNN model. As shown in FIG. 5, the input of the first embedding layer 521 can include the differential sequence of the second signal 510, and the output can be the first feature vector 522. For example, the differential sequence of the second signal can be ([2, 2], [3, 2], [3, 2], [3, 2]), indicating that the signal difference value of the second signal at T1 time and T2 time (the time interval between T1 and T2 is 2 mins) is 2°, the signal difference value of the second signal at T2 time and T3 time (the time interval between T2 and T3 is 2 mins) is 3°, the signal difference value of the second signal at T3 time and T4 time (the time interval between T3 and T4 is 2 mins) is 3°, and the signal difference value of the second signal at T4 time and T5 time (the time interval between T3 and T5 is 2 mins) is 3°. Figure 5

[0115] The first feature vector can be used to digitally describe a set of features. The set of features can be measurable attributes or characteristics of the second signal. For example, the first feature vector can be represented as I = (3, 2, 1), indicating that the amplitude changes by 3 dB and the phase changes by 2° within 1 second in the time interval between the first time and the second time of the second signal. In some embodiments, the first feature vector can be represented as a row vector or a column vector. The first feature vector can correspond to an N-dimensional coordinate system. The N-dimensional coordinate system can be related to N signal characteristics of the second signal.

[0116] The first prediction layer can be a machine learning model for processing the first feature vector of the second signal to determine the biological species. For example, the first prediction layer can be a NN model. As shown in FIG. 5, the input of the first prediction layer 523 can include the first feature vector 522 output by the first embedding layer 521, and the output can include the biological species 530 of the biological body. In some embodiments, the input of the first prediction layer 523 can also include the room type 540 of the target room and the room layout 550 of the target room. Figure 5

[0117] In some embodiments, the output of the first embedding layer can be used as the input of the first prediction layer, and the first embedding layer and the first prediction layer can be jointly trained.

[0118] ​​​In some embodiments, the third training sample of the joint training can include a differential sequence of the sample second signal, a room type of the sample room, and a layout of the sample room, and the third label can include an actual biological species in the sample room corresponding to the differential sequence of the sample second signal, the room type of the sample target room, and the room layout. In some embodiments, the third training sample and the third label can be obtained by simulating the movement of the sample biological body in different sample rooms by using the simulation model. The third training sample and the third label can be obtained in a manner similar to the first training sample and the first label, and more details can be referred to Figure 4A and the related description, which will not be repeated here.

[0119] The exemplary training process includes: inputting the differential sequence of the sample second signal into the initial first embedding layer to obtain the first feature vector output by the initial first embedding layer; inputting the first feature vector as the training sample data, and the room type of the sample room and the room layout of the sample room into the first prediction model to obtain the biological species output by the first prediction model. A loss function is constructed based on the third label and the biological species output by the first prediction model, and the parameters of the initial first embedding layer and the initial first prediction layer are updated synchronously until the intermediate first embedding layer and the intermediate first prediction layer meet a preset condition, thereby obtaining the trained first embedding layer and the first prediction layer. The preset condition can be that the loss function is less than a threshold, converges, or the training period reaches a threshold.

[0120] In some embodiments of the present specification, when a single terminal device receives a signal, the first embedding layer of the third prediction model processes the differential sequence of the second signal, the room type of the target room, and the room layout to determine the biological species, fully considering the influence of the type of the target room, the information of the items in the space, and the like on the prediction of the biological species, which can make the predicted biological species more accurate. The parameters of the third prediction model are obtained through the above training method, which is beneficial to solve the problem of difficulty in obtaining labels when training the first embedding layer alone, and can also make the first embedding layer better reflect the characteristics of the second signal.

[0121] Figure 6 is an exemplary schematic diagram of a fourth prediction model according to some embodiments of the present specification.

[0122] In some embodiments, in response to the terminal device being multiple, the determination module 230 can determine a differential sequence of at least one second signal received by each of the multiple terminal devices to obtain multiple differential sequences; and process the multiple differential sequences based on the fourth prediction model to determine the biological species. More details about the differential sequence can be referred to Figure 5 and the related description.

[0123] The fourth prediction model can be another machine learning model for determining the biological species. For example, the fourth prediction model can be a RNN model, a DNN model, a CNN model, or the like.

[0124] In some embodiments, as shown in FIG. 6A, the input of the fourth prediction model 620 includes the plurality of differential sequences 610, and the output includes the biological species 630. Figure 6 In some embodiments, as shown in FIG. 6A, the input of the fourth prediction model 620 includes the plurality of differential sequences 610, and the output includes the biological species 630. Figure 6 In some embodiments, as shown in FIG. 6A, the input of the fourth prediction model 620 can further include the room type 640 of the target room, the room layout 650 of the target room, and the relative position relationship 660 of the at least one terminal device. For more information about the room type of the target room, the room layout, please refer to Figure 5 and the related description thereof.

[0125] In some embodiments, the relative position relationship of the at least one terminal device can include the distance between the at least one terminal device and the emission source, the height of the at least one terminal device from the ground, the position relationship of the at least one terminal device from the occlusion, and the like. In some embodiments, the relative position relationship of the at least one terminal device can be determined by the first position distribution of the terminal device and the second position distribution of the emission source. For example, the first position distribution of the terminal devices A, B, and C can be represented as [A(2, 8, 10), B(1, 4, 5), C(3, 5, 8)], and the second position distribution of the emission source D can be represented as (3, 5, 8). Then, the distance between the terminal device A and the emission source D can be calculated based on the first position distribution of the at least one terminal device and the second position distribution of the emission source as and the like.

[0126] In some embodiments, the fourth prediction model can include a plurality of processing layers. For example, as shown in FIG. 6A, the fourth prediction model 620 can include a second embedding layer 621 and a second prediction layer 623. Figure 6 In some embodiments, the fourth prediction model can include a plurality of processing layers. For example, as shown in FIG. 6A, the fourth prediction model 620 can include a second embedding layer 621 and a second prediction layer 623.

[0127] The second embedding layer can be a machine learning model for processing the plurality of differential sequences of the plurality of second signals to determine the second feature vector. For example, the second embedding layer can be a CNN model. As shown in FIG. 6A, the input of the second embedding layer 621 can include the plurality of differential sequences 610, and the output can be the second feature vector 622. Figure 6 The second embedding layer can be a machine learning model for processing the plurality of differential sequences of the plurality of second signals to determine the second feature vector. For example, the second embedding layer can be a CNN model. As shown in FIG. 6A, the input of the second embedding layer 621 can include the plurality of differential sequences 610, and the output can be the second feature vector 622.

[0128] The second feature vector is similar to the first feature vector. For more information, please refer to Figure 5 and the related description thereof.

[0129] The second prediction layer can be a machine learning model for processing the second feature vector to determine the biological species. For example, the second prediction layer can be one of a NN or a GNN model. As shown in FIG. 6A, the input of the second prediction layer 623 can include the second feature vector 622, and the output can be the biological species 630. Figure 6As shown, the input of the second prediction layer 623 can include the second feature vector 622 output by the second embedding layer 621, and the output can include the biological species 630 of the living body. In some embodiments, the input of the second prediction layer 623 can further include the room type 640 of the target room, the room layout 650 of the target room, and the relative position relationship 660 of the at least one terminal device.

[0130] In some embodiments, the output of the second embedding layer can be input to the second prediction layer, and the second embedding layer and the second prediction layer can be jointly trained.

[0131] In some embodiments, the fourth training sample of the joint training can include the differential sequence of the plurality of sample second signals, the room type and the room layout of the sample target room, and the relative position relationship of the sample terminal device, and the fourth label of the fourth training sample can include the actual biological species in the sample room corresponding to the differential sequence of the plurality of sample second signals, the room type and the room layout of the sample target room, and the relative position relationship of the sample terminal device. In some embodiments, the fourth training sample and the fourth label can be obtained by simulating the movement of the sample living body in different sample rooms using a simulation model. The fourth training sample and the fourth label can be obtained in a manner similar to the first training sample and the first label, and more details can be referred to Figure 4A and the related description, which will not be repeated here.

[0132] An exemplary training process includes: inputting the differential sequence of the plurality of sample second signals to the initial second embedding layer to obtain the second feature vector output by the initial second embedding layer; inputting the second feature vector as training sample data, and the room type of the sample room, the room layout of the sample room, and the relative position relationship of the sample terminal device to the second prediction model to obtain the biological species output by the second prediction model. A loss function is constructed based on the fourth label and the biological species output by the second prediction model, and the parameters of the initial second embedding layer and the initial second prediction layer are updated synchronously until the intermediate second embedding layer and the intermediate second prediction layer meet a preset condition, thereby obtaining the trained second embedding layer and the second prediction layer. The preset condition can be that the loss function is less than a threshold, converges, or the training period reaches a threshold.

[0133] In some embodiments of the present specification, when multiple terminal devices receive signals, the second embedding layer based on the fourth prediction model processes the differential sequence of the multiple second signals, the room type and room layout of the target room, and the relative position relationship of the at least one terminal device, so that the biological species can be determined according to the signals received by the multiple terminal devices, and the predicted biological species is more accurate. At the same time, the parameters of the fourth prediction model are obtained through the above training method, which is beneficial to solve the problem of difficulty in obtaining labels when training the second embedding layer alone in some cases, and can also make the second embedding layer better reflect the characteristics of the multiple second signals.

[0134] Figure 7 is an exemplary schematic diagram of a signal graph and a fifth prediction model according to some embodiments of the present specification.

[0135] In some embodiments, the determining module 230 can construct a signal graph based on the related information of the target room, the at least one terminal device, and the at least one emission source; process the at least one second signal and the constructed signal graph based on the fifth prediction model to determine the biological species.

[0136] The signal graph refers to a graph used to reflect the signal transmission relationship between each terminal device and the emission source in the target room. The graph is a data structure composed of nodes and edges, and the edges connect the nodes. The nodes and edges can have attributes.

[0137] In some embodiments, the nodes of the signal graph can correspond to the terminal devices or the emission sources in the target room. The node attributes can reflect the related characteristics of the corresponding terminal devices or emission sources. In some embodiments, the nodes of the signal graph can include two types, such as Figure 7 As shown in the figure, the nodes of the signal graph 710 can include signal emission nodes 720 and signal receiving nodes 730. The signal emission nodes 720 can include node A, node B, and node C. The signal receiving nodes 730 can include node A', node B', and node C'.

[0138] In some embodiments, the node attributes of the signal emission nodes can include the emission power, emission frequency, and emission position (which can be represented in three-dimensional coordinates, for example) of the emission source. The node attributes of the signal emission nodes can be represented by a three-dimensional vector. For example, the node attributes of the signal emission nodes can be (a, b, c), where a represents the emission power, b represents the emission frequency, and c represents the emission position.

[0139] In some embodiments, the node attributes of the signal receiving nodes can include the second signals received by the terminal devices and the receiving positions of the terminal devices (which can be represented in three-dimensional coordinates, for example).

[0140] The edges in a signal graph represent the signal transmission relationship between signal transmitting nodes and signal receiving nodes. In other words, edges exist between signal transmitting nodes and signal receiving nodes, such as... Figure 7 As shown, node C in signal transmitting node 720 and node C' in signal receiving node 730 have an edge 740, and node B in signal transmitting node 720 and node B' in signal receiving node 730 have an edge 740, etc. In some embodiments, the attributes of the edge may include the three-dimensional coordinates, material and volume of each object in the room located on the edge, as well as the length of the edge (i.e., the distance from the corresponding signal transmitting node to the signal receiving node).

[0141] The fifth prediction model can be another machine learning model used to determine the species of organism. For example, the fifth prediction model could be a GNN model, etc. The fifth prediction model can also be other graph models, such as a graph convolutional neural network model (GCNN), or a graph neural network model with added processing layers or modified processing methods.

[0142] In some embodiments, such as Figure 7 As shown, the input to the fifth prediction model 750 includes a signal spectrum 710, and the output is the biological species 790 of the organism. In some embodiments, such as Figure 7 As shown, the input to the fifth prediction model 750 may also include the room type 760 of the target room, the room layout 770 of the target room, and the second signal 780. Here, the edge outputs in the GNN correspond to the biological species of the organisms in the target room.

[0143] The fifth prediction model can be trained using the same or different processing devices based on training data. The training data includes a fifth training sample and a fifth label. For example, the fifth training sample can be a sample signal map, sample room layout, sample room type, and sample second signal of multiple target rooms determined based on historical data. The nodes and their attributes, edges and their attributes of the sample signal map are similar to those described above. The label can be the actual biological species in the target rooms corresponding to the sample signal map, sample room layout, sample room type, and historical sample second signal. The fifth label can be obtained based on manual annotation. In some embodiments, the fifth training sample and the fifth label can be obtained by simulating the movement of sample organisms in different sample rooms using a simulation model. The method of obtaining the fifth training sample and the fifth label is similar to the method of obtaining the first training sample and the first label; further explanation can be found in [link to documentation]. Figure 4A The details and related descriptions will not be repeated here.

[0144] In some embodiments, the fifth prediction model may include multiple processing layers. For example, such as Figure 7As shown, the fifth prediction model 750 can include a room feature layer 751, a distribution feature layer 753, and a species determination layer 755. The room feature layer 751 is configured to process the room type 760 and the room layout 770 of the target room to determine a room feature 752; the distribution feature layer 753 is configured to process the signal map 710 and the room feature 752 output by the room feature layer 751 to determine a location feature 754; and the species determination layer 755 is configured to process the at least one second signal 780 and the location feature 754 output by the distribution feature layer 753 to determine the biological species 790.

[0145] The room feature layer can be a machine learning model configured to determine a room feature of a target room. For example, the room feature layer can be a CNN model, etc. As shown in FIG. 7, the room feature layer 751 can include a room type layer 761, a room layout layer 763, and a room feature determination layer 765. The room type layer 761 is configured to process the room type 760 to determine a room type feature 762; the room layout layer 763 is configured to process the room layout 770 to determine a room layout feature 764; and the room feature determination layer 765 is configured to process the room type feature 762 and the room layout feature 764 to determine the room feature 752. Figure 7 As shown, the input of the room feature layer 751 can include the room type 760 and the room layout 770, and the output can include the room feature 752.

[0146] The room feature is information related to a room. For example, the room feature can include the floor area of the room, the shape of the room, and the indoor space occupancy, etc. The room feature can be represented by a feature vector. For example, A can be used to represent that the room type is a hotel room, B can be used to represent that the room type is a warehouse, etc., a specific numerical value can be used to represent the floor area of the room, 1 can be used to represent that the shape of the room is a cube, 2 can be used to represent that the shape of the room is a cuboid, etc., and a percentage can be used to represent the indoor space occupancy, etc. When the exemplary room feature is (A, 20, 2, 20%), it means that the room is a hotel room, the floor area is 2015 m 2 , the shape of the room is a cuboid, and the indoor space occupancy is 20%.

[0147] The distribution feature layer can be a machine learning model configured to determine a distribution feature of a target room. For example, the distribution feature layer can be a GNN model, etc. As shown in FIG. 7, the distribution feature layer 753 can include a signal map layer 757, a location feature determination layer 759, and a distribution feature determination layer 761. The signal map layer 757 is configured to process the signal map 710 to determine a signal map feature 758; the location feature determination layer 759 is configured to process the room feature 752 output by the room feature layer 751 and the signal map feature 758 to determine a location feature 754; and the distribution feature determination layer 761 is configured to process the location feature 754 to determine the distribution feature 754. Figure 7 As shown, the input of the distribution feature layer 753 can include the room feature 752 output by the room feature layer 751 and the signal map 710, and the output can include the distribution feature 754.

[0148] The distribution feature can reflect the positional relationship of each transmitting source and terminal device and the items in the room. For example, the distribution feature can be “the table is close to the window”, etc. The distribution feature can be represented by a feature vector. For example, a can be used to represent “the table is close to the window”, b can be used to represent “the TV is close to the wall”, etc. When the exemplary distribution feature is ([a, 0.1], [b, 0.2]), it means that the table is close to the window by 10 cm, and the TV is close to the wall by 20 cm.

[0149] The category determination layer can be a machine learning model for determining the biological category of the organism. For example, the category determination layer can be a DNN model. For example, the input of the category determination layer 755 can include the second signal 780 and the distribution feature 754 output by the distribution feature layer 753, and the output can include the biological category 790.

[0150] In some embodiments, the output of the room feature layer can be input to the distribution feature layer, and the output of the distribution feature layer can be input to the category prediction layer. The room feature layer, the distribution feature layer, and the category prediction layer can be jointly trained.

[0151] In some embodiments, the fifth prediction model can be obtained by jointly training a plurality of fifth training samples with fifth labels. The joint training process is similar to the joint training process of the fourth prediction model, which will not be described here.

[0152] In some embodiments of the present specification, the biological category of the organism is determined by combining the signal atlas with the machine learning model, which can deeply extract the room features and the distribution features of each object in the room, and combine the multiple information of the target room for calculation, so that the reliability of the finally obtained biological category is higher.

[0153] The above has described the basic concept, and it is obvious that the above detailed disclosure is only used as an example and does not limit the present specification. Although it is not explicitly stated here, those skilled in the art can make various modifications, improvements and corrections to the present specification. Such modifications, improvements and corrections are suggested in the present specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the present specification.

[0154] At the same time, the present specification uses specific words to describe the embodiments of the present specification. As "one embodiment", "an embodiment", and / or "some embodiments" means a certain feature, structure or characteristic related to at least one embodiment of the present specification. Therefore, it should be emphasized and noted that the "an embodiment" or "one embodiment" or "one alternative embodiment" mentioned in different places in the present specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present specification can be properly combined.

[0155] Furthermore, the order of the processing elements and sequences described in this specification are not intended to be construed as a limitation, unless specifically stated, but are included to provide a complete description of one or more embodiments of the present specification. Regardless of the particular sequence of processing elements and sequences, however, the description herein of a process should be understood to include any and all combinations of one or more elements, and sequences that can be perceived as either open-ended or specific.

[0156] Similarly, it is to be noticed that the term "comprising", used in the description, should not be interpreted as being restricted only to the means listed thereafter. It is to be understood that the term "comprising" means that any additional element, which is not specifically mentioned, is optionally present or can be added. In some embodiments, the description of an embodiment using the term "comprising" can also be interpreted as using the term "including" or "consisting of". Furthermore, the description herein of any particular embodiment of the present specification is intended to be illustrative only and is not intended to be limiting unless specifically stated. Thus, while the present specification has been described in terms of some embodiments, it is anticipated that alternatives, modifications, and equivalents will further occur to those skilled in the art. Accordingly, it is intended that the claims be interpreted as broadly as is reasonabl

[0157] Some embodiments use numerals to describe components, quantities of attributes. It should be understood that such numerals used in the description of the embodiments are, in some examples, modified by the adjectives "about", "approximately", or "generally". Unless otherwise stated, "about", "approximately", or "generally" indicates that the stated numerical value allows for a variation of ±20%. Accordingly, numerical parameters in the description and claims are approximations, and can vary depending upon the requirements of the particular embodiment. In some embodiments, numerical parameters are determined by the use of common rounding techniques. Although the numerical ranges and parameters setting forth the broadest scope of the embodiments of the specification are approximations, the numerical values set forth in the specific examples are reported as precisely as practicable. The numerical values set forth in the specific examples are provided to be as precise as reasonably possible. However, some variations may occur depending on the choice of input used to develop or derive the numerical values in the examples.

[0158] Each patent, patent application, publication, and other material cited in this specification is hereby incorporated by reference in its entirety for the purpose of describing and disclosing, for example, the compositions and methodologies described in such publications that might be used in connection with the embodiments of the present specification. However, the citation of any reference in this specification is not to be construed as an admission that such reference is available as prior art to the present specification. To the extent that any meaning or definition of a term in this specification conflicts with any meaning or definition of the same term in a document incorporated by reference, the meaning or definition assigned to that term in this specification shall govern. It is specifically intended that the present specification not be limited to the embodiments disclosed in the specification and that the scope of the specification be, therefore, limited only by the claims.

[0159] Finally, it should be understood that the embodiments described herein are only given by way of example and that other modifications can occur to persons skilled in the art. Therefore, the scope of the present description is not intended to be limited to the embodiments described herein but is only limited by the claims that follow.

Claims

1. A biological body monitoring management method characterized by, The method comprises: acquiring at least one first signal transmitted by at least one transmission source based on at least one terminal device in a target room, wherein the first signal comprises a Wi-Fi signal transmitted by a router in the target room; extracting at least one second signal from the at least one first signal, wherein the second signal comprises channel state information signals obtained after the first signal is subjected to multipath fading during propagation; determining whether a living body exists in the target room based on the at least one second signal; in response to the existence of the living body in the target room, constructing a signal graph based on relevant information of the target room, the at least one terminal device, and the at least one transmission source, wherein nodes of the signal graph correspond to the terminal devices or the transmission sources in the target room, the nodes comprise signal transmission nodes and signal receiving nodes, and edges of the signal graph represent signal transmission relationships between the signal transmission nodes and the signal receiving nodes; processing the at least one second signal and the signal graph based on a fifth prediction model to determine a biological species of the living body, wherein the fifth prediction model is a machine learning model; and determining a management scheme for the target room based on the biological species.

2. The method of claim 1, wherein, The determination of whether the living body exists in the target room based on the at least one second signal comprises: determining whether the living body exists in the target room based on a signal change rate of the at least one second signal.

3. The method of claim 2, wherein, The determination of the biological species of the living body based on the at least one second signal in response to the existence of the living body in the target room comprises: processing the at least one second signal by a machine learning model to determine the biological species.

4. The method of claim 3, wherein, The training sample of the machine learning model is obtained by simulating the movement of a sample living body in different sample rooms by a simulation model.

5. A living body monitoring management system characterized by comprising: The method comprises: an acquisition module configured to acquire at least one first signal transmitted by at least one transmission source based on at least one terminal device in a target room, wherein the first signal comprises a Wi-Fi signal transmitted by a router in the target room; an extraction module configured to extract at least one second signal from the at least one first signal, wherein the second signal comprises channel state information signals obtained after the first signal is subjected to multipath fading during propagation; a determination module configured to determine whether a living body exists in the target room based on the at least one second signal; in response to the existence of the living body in the target room, construct a signal graph based on relevant information of the target room, the at least one terminal device, and the at least one transmission source, wherein nodes of the signal graph correspond to the terminal devices or the transmission sources in the target room, the nodes comprise signal transmission nodes and signal receiving nodes, and edges of the signal graph represent signal transmission relationships between the signal transmission nodes and the signal receiving nodes; process the at least one second signal and the signal graph based on a fifth prediction model to determine a biological species of the living body, wherein the fifth prediction model is a machine learning model; and determine a management scheme for the target room based on the biological species. determine, based on the biological species, a management scheme for the target room.

6. The system of claim 5, wherein, To determine, based on the at least one second signal, whether a biological body exists in the target room, the determination module is configured to: determine, based on a signal change rate of the at least one second signal, whether the biological body exists in the target room.

7. The system of claim 6, wherein, To determine, based on the at least one second signal, a biological species of the biological body in response to the existence of the biological body in the target room, the determination module is configured to: determine the biological species by processing the at least one second signal through a machine learning model.

8. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions, and when a computer reads the computer instructions in the storage medium, the computer executes the method in any one of claims 1-4.

Images