JP2024035413A - Detection system, sensor device, and detection method - Google Patents

Abstract

To detect and monitor radio events even in radio communication systems in which CSI information cannot be obtained.SOLUTION: A detection system detects state events. A radio analysis unit for obtaining a delay profile representing the correlation between a received radio signal and a reference signal includes: a radio receiving unit that receives the radio signal; a reference signal input unit that outputs the reference signal which is a fixed pattern portion extracted from the received radio signal; and a delay profile acquisition unit that outputs a delay profile by calculating the correlation between the received radio signal and the reference signal. A detection unit for detecting the state events from the delay profile includes a clustering analysis unit that outputs a first clustering result generated by classification of the delay profile, and a determination unit that uses the first clustering result to detect the state events that indicate real-world information.SELECTED DRAWING: Figure 3

Description

FIELD OF THE INVENTION The present invention relates to methods, apparatus and systems for state event detection and monitoring.

In order to detect and monitor state events using radio, a technique for detecting events based on classification of channel state information is described in Patent Document 1.

Patent Document 1 describes a configuration related to a device for detecting an event. The apparatus described in US Pat. No. 5,001,202 includes a processor and, when executed by the processor, classifies channel state information (CSI) and detects events based on the classification of the CSI obtained during a monitoring phase. and a storage device identifying instructions for causing the processor to train a classifier for the processor. The training of the classifier is based on the training CSI of a wireless multipath channel between a wireless transmitter and a wireless receiver in the venue for each known event to be detected. obtaining a training CSI derived from one or more probing signals transmitted to a receiver during a period during which known events occur at the venue; and and training a classifier based on the .

International Publication 2017/100706

Patent Document 1 describes a classifier for detecting events based on the classification of channel state information (CSI) of radio waves emitted by a wireless communication system. There is no consideration given to how to detect and monitor.

The present invention has been made in view of the above-mentioned circumstances, and aims to provide a system, device, and method that can detect and monitor wireless events even in a wireless communication system in which CSI information cannot be acquired.

A typical example of the invention disclosed in this application is as follows. That is, the detection system detects a state event, and includes a radio analysis section that obtains a delay profile representing a correlation between a received radio signal and a reference signal, and a detection section that detects the state event from the delay profile. , the radio analysis unit includes a radio reception unit that receives a radio signal, a reference signal input unit that outputs a reference signal in which a fixed pattern portion is extracted from the received radio signal, and a reference signal input unit that outputs a reference signal in which a fixed pattern portion is extracted from the received radio signal; a delay profile acquisition unit that performs a correlation calculation with a reference signal and outputs a delay profile; the detection unit includes a clustering analysis unit that outputs a first clustering result generated by classifying the delay profile; The present invention is characterized by having a determination unit that detects a state event indicating real-world information using the results of clustering.

According to one aspect of the present invention, wireless events can be accurately detected and monitored. Problems, configurations, and effects other than those described above will be made clear by the description of the following examples.

1 is a diagram illustrating a wireless communication environment according to an embodiment of the present invention. 1 is a diagram showing a configuration example of a wireless event detection system according to the present embodiment. 1 is a diagram illustrating a configuration example of a wireless sensor according to the present embodiment. FIG. 2 is a diagram showing an example of the hardware configuration of a wireless sensor according to the present embodiment. It is a flowchart of the process which the wireless monitoring sensor concerning this embodiment performs. 7 is a flowchart of a process for creating a reference signal according to the present embodiment. It is a flowchart of the process which acquires the delay profile based on this embodiment. FIG. 3 is a diagram showing an example of a reference signal according to the present embodiment. FIG. 3 is a diagram showing an example of a reference signal according to the present embodiment. FIG. 3 is a diagram showing an example of a delay profile according to the present embodiment. FIG. 3 is a diagram showing an example of a wireless signal from which a delay profile according to the present embodiment is calculated. It is a flowchart of the process of associating a radio signal and an event and learning by machine learning according to the present embodiment. It is a flowchart of the process of estimating an event by machine learning concerning this embodiment. FIG. 3 is a diagram showing the configuration of a learning table according to the present embodiment. FIG. 2 is a sequence diagram in a wireless communication system capable of acquiring channel state information, which is a comparative example of the present invention. FIG. 2 is a sequence diagram in a wireless communication system in which channel state information cannot be acquired according to the present embodiment.

In the following embodiments, the description will be divided into a plurality of sections or embodiments when necessary for convenience. In the following embodiments, when referring to the number of elements (including numbers, numerical values, amounts, ranges, etc.), unless specifically specified or clearly limited to a specific number in principle, It is not limited to the specific number, and may be more than or less than the specific number. Note that, in the following embodiments, the constituent elements (including processing steps, etc.) are not necessarily essential, except when explicitly stated or when it is considered to be clearly essential in principle.

Further, each configuration, function, processing unit, processing means, etc. in the following embodiments may be partially or entirely realized as, for example, an integrated circuit or other hardware. Further, each configuration, function, processing unit, processing means, etc. described below may be realized as a program executed on a computer. That is, it may be realized as software. Information such as programs, tables, files, etc. that realize each configuration, function, processing unit, processing means, etc. is stored in storage devices such as memory, hard disk, SSD (Solid State Drive), and storage media such as IC cards, SD cards, and DVDs. can be stored in.

Hereinafter, embodiments of the present invention will be described in detail based on the drawings. In addition, in all the drawings for explaining the embodiment, members having the same function are given the same or related reference numerals, and repeated explanation thereof will be omitted. Furthermore, in the following embodiments, descriptions of the same or similar parts will not be repeated in principle unless particularly necessary.

[System configuration]
A wireless event detection system 200 according to an embodiment of the present invention includes a wireless base station 101, a wireless sensor 103, and a server 107. Wireless base station 101 communicates with wireless terminal 102 . Hereinafter, a wireless event detection system 200 according to an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a diagram illustrating a wireless communication environment 100 according to an embodiment of the invention. The wireless communication environment 100 may be, for example, an indoor environment such as a distribution warehouse, factory, or office, or an outdoor environment such as a construction site. Wireless communication environment 100 includes a wireless base station 101 , a wireless terminal 102 , a wireless sensor 103 , and a monitoring target 104 . The monitoring target 104 is a physical object that does not have a communication function and exists within the communication range of the wireless communication system, and may be a physical obstacle that blocks radio waves, such as a physical object that is transported in a distribution warehouse. This can be a movable object such as a luggage rack. Further, the monitoring target 104 may or may not have a wireless communication function, and may be, for example, the wireless terminal 102 or an object that is not under the control of the wireless event detection system 200.

Note that in this embodiment, the wireless terminal 102 is a device that transmits or receives user data, and the wireless sensor 103 is a device that does not transmit or receive user data but transmits or receives radio waves for observation. 102 may have the function of the wireless sensor 103. Further, only the wireless sensor 103 may be installed without installing the wireless terminal 102, or only the wireless terminal 102 may be installed without installing the wireless sensor 103.

The wireless base station 101 is connected by wire to a network hub 105, and the network hub 105 is connected to a server 107 via an external network 106. The server 107 may be provided outside or inside the wireless communication environment 100 such as a distribution warehouse.

The radio base station 101 sends control commands such as movement instructions to the radio terminal 102 by transmitting and receiving radio waves to and from the radio terminal 102 . In addition to the radio waves that propagate to the wireless terminal 102 as direct waves, the radio waves also propagate as multipath waves and reach the wireless sensor 103 . At this time, the multipath waves are influenced by the monitoring target 104 and are repeatedly reflected to reach the wireless sensor 103. Depending on the position of the monitoring target 104, the state of the multipath wave (for example, delay time, intensity) changes. Wireless sensor 103 receives these multipath waves and analyzes channel state information. The channel state information analyzed by the wireless sensor 103 is, for example, a delay profile expressed by received signal strength and propagation delay time. Among the radio waves transmitted by the radio base station 101, an example will be described in which the radio waves addressed to the wireless terminal 102 and the radio waves received by the wireless sensor 103 are the same radio waves, but they are different (for example, the frequency is different). (different) radio waves may also be used.

Next, a configuration example of the wireless event detection system 200 will be described with reference to FIG. 2. Wireless event detection system 200 includes a server 107 , an external network 106 , a network hub 105 , a wireless base station 101 , a wireless terminal 102 , and a wireless sensor 103 . Network hub 105 and wireless base station 101 are connected via a network (eg, Ethernet). The wireless base station 101 communicates with the wireless terminal 102 using radio waves 208.

The server 107 is a computer that detects events by analyzing the radio waves 208 received by the wireless sensor 103, and is configured by a computer having an arithmetic unit, a memory, and a communication interface. The computing device executes a program stored in memory. The functions provided by the server 107 are realized by the arithmetic unit executing various programs. Note that the arithmetic device may include a hardware arithmetic device (eg, ASIC, FPGA, etc.) in addition to a processor that executes a program. Memory includes ROM, which is a nonvolatile storage element, and RAM, which is a volatile storage device. The ROM stores unchangeable programs (eg, BIOS) and the like. RAM is a high-speed and volatile storage device such as DRAM (Dynamic Random Access Memory), and temporarily stores programs executed by an arithmetic unit and data used during execution of the programs.

The program executed by the arithmetic device is provided to the server 107 via a removable medium (CD-ROM, flash memory, etc.) or a network, and is stored in a non-volatile auxiliary storage device that is a non-temporary storage medium. For this reason, the server 107 preferably has an interface for reading data from removable media.

The server 107 is a computer system that is physically configured on one computer or on one or more logically or physically configured computers, and is constructed on multiple physical computer resources. It may also run on a virtual machine. For example, each process executed by an arithmetic device may be executed on a separate physical or logical computer, or a plurality of processes may be combined and executed on one physical or logical computer.

The processing content of the wireless sensor 103 will be described later with reference to subsequent figures.

Next, a configuration example of the wireless sensor 103 will be described with reference to FIG. 3. The wireless sensor 103 includes a wireless analysis unit 301 that analyzes wireless signals, a learning unit 302 that executes learning processing that learns by associating wireless signals with events, and a detection unit that executes detection processing that detects events using learning results. 303. The radio analysis section 301 includes radio signal reception sections 304 and 308, a reference signal input section 306, a training signal acquisition section 310, and a delay profile acquisition section 311.

Radio signal receiving units 304 and 308 convert radio waves captured by antenna 403 shown in FIG. 4 into baseband signals. The reference signal input section 306 outputs the input reference signal to the training signal acquisition section 310 and the delay profile acquisition section 311.

The learning section 302 includes a learning table 312, a clustering analysis section 313, and an event detection threshold updating section 317.

In the learning process, the wireless signal receiving unit 304 receives a wireless signal at the time of a known event to be detected and outputs a training signal 305. Further, the reference signal input section 306 outputs the input reference signal 307. Events include the position of transport vehicles in the warehouse, the movements of people (standing, sitting, movements of limbs, gestures, etc.), the material of objects that reflect radio waves, the soil quality (moisture content), etc. It may be detected as an event. The reference signal 307 is radio signal data having a known pattern, the details of which will be described later. The training signal acquisition unit 310 calculates the cross-correlation between the training signal 305 and the reference signal 307, and outputs a delay profile representing the calculated cross-correlation to the learning unit 302.

In the learning unit 302, a clustering analysis unit 313 classifies the delay profiles into arbitrary groups and outputs a clustering result 314 generated. The event detection threshold updating unit 317 records the clustering results 314 in association with events in the learning table 312, updates the detection threshold, and uses the clustering results 314 to learn the correlation between events.

The detection unit 303 includes a clustering analysis unit 315 and a determination unit 318.

In the detection process, the wireless signal receiving unit 308 receives an unknown signal 309 associated with an unknown event. The delay profile acquisition unit 311 calculates the cross-correlation between the reference signal 307 and the unknown signal 309, and outputs a delay profile representing the calculated cross-correlation to the clustering analysis unit 315.

In the detection unit 303, a clustering analysis unit 315 outputs a clustering result 316 generated by classifying the delay profile. The determining unit 318 refers to the learning table 312 and detects an event corresponding to the clustering result 316. If the certainty (confidence) of event detection exceeds the detection threshold, it is determined that the event recorded in the learning table 312 is detected (320), and if the certainty is smaller than the detection threshold, it is determined that the corresponding event does not exist. (319). A wireless event can be detected by the processing described above.

Note that in the configuration described above, the wireless sensor 103 has the wireless analysis section 301, the learning section 302, and the detection section 303; however, the wireless sensor 103 has the wireless analysis section 301, and the server 107 has the learning section 302 and the detection section 303. It may have.

Next, an example of the hardware configuration of the wireless sensor 103 will be described with reference to FIG. 4. The wireless sensor 103 includes a wireless processing section 401 and an information processing section 402. The wireless processing section 401 includes an antenna 403, an analog high frequency circuit section 404, and a digital signal processing section 405. Analog high frequency circuit section 404 converts the radio waves captured by antenna 403 into baseband signals. The digital signal processing section 405 demodulates the baseband signal output from the analog high frequency circuit section 404 and generates a digital signal.

Further, the information processing unit 402 includes a CPU 406, a memory 407, a storage unit 408, an external IF 409, a network IF 410, and a data bus 411 connecting these. The CPU 406 performs information processing by executing a predetermined program stored in the memory 407. The storage unit 408 is composed of a nonvolatile storage device, and stores programs executed by the CPU 406 and data used when executing the programs. The external IF 409 is an interface to which an external storage medium such as a flash memory can be connected. The network IF 410 may be, for example, a wired Ethernet connection interface device, a wireless WiFi module, or the like that connects the wireless sensor 103 to another device via a network. The network IF 410 may be a communication interface different from the wireless processing unit 401 or a communication interface connected to the wireless processing unit 401.

Furthermore, the wireless processing section 401 and the information processing section 402 may be in the form of separate devices, the same device, or one IC chip.

Next, with reference to FIG. 5, the overall processing operation of the wireless event detection system 200 will be described. In FIG. 5, after the wireless event detection system 200 is powered on (501), a reference signal is created and input to the wireless event detection system 200 (502). The process of creating a reference signal in step 502 will be described later with reference to FIG. The input reference signal may be created separately or may be created within the wireless event detection system 200. Next, the training signal 305 is learned (503). After the wireless event detection system learns to detect a predetermined event, it performs wireless signal measurement and sensing processing in actual operation (504). If the system is not powered off (No in 505), the process returns to step 504 and continues processing. On the other hand, when the system is powered off (Yes in 505), the process ends and the wireless event detection system 200 stops. Through the processing described above, the steps of creating a reference signal, learning an event, and detecting an event are realized. Details of each step will be explained using the drawings described later.

FIG. 6 is a flowchart of the process of creating a reference signal. The reference signal is a signal that has a known pattern in a wireless signal that is transmitted by a wireless system to detect a wireless event. The reference signal is held in the recording medium as data. First, while the target wireless system is in operation, a wireless signal is transmitted from a wireless transmitter (601). Next, the wireless sensor 103 for recording the reference signal is installed in an environment where there is no or little influence of noise, shielding, etc. (602). For example, the reference signal may be used in an environment where there is no reflection measured in an anechoic chamber, in an environment where the wireless transmitter and the wireless receiver (wireless signal receiving section 304 of the wireless sensor 103) are directly connected by a cable, or in a field where there is no detection target. Good to measure. Next, the wireless signal received by the wireless signal receiving unit 304 of the wireless sensor 103 is recorded in the recording unit (603). Next, a portion having a fixed pattern is extracted from the recorded wireless signal (604). Next, the extracted portion is held in the internal memory as a reference signal, and is outputted to an external recording medium as necessary (605). The reference signal can be obtained by the procedure described above. Note that the reference signal may be generated from a wireless signal received by each wireless sensor 103, or may be a reference signal generated from a wireless signal received by another wireless sensor 103 and input via a recording medium or a network. good.

FIG. 7 is a flowchart of the process of acquiring a delay profile. To obtain a delay profile, the wireless sensor 103 starts receiving measurement signals (701). When the signal is received (Yes in 702), the cross-correlation with the reference signal is calculated (703). If no signal is received (No at 702), the process returns to step 701 and waits for a measurement signal. As a result of the correlation calculation, if the correlation peak value exceeds the threshold (704), the generated delay profile is output (Yes in 705). On the other hand, if the correlation peak value does not exceed the threshold (No in 705), the process returns to step 701 and waits for a measurement signal. A delay profile can be obtained by the procedure described above.

8A and 8B are diagrams showing examples of reference signals. In FIGS. 8A and 8B, the vertical axis represents frequency, the horizontal axis represents time, and the power intensity of the reference signal is represented by color shading. The reference signal has a known pattern and is a fixed pattern that does not change depending on the communication content. An example of a signal that can be used as a reference signal is a reference signal 800 with a preamble pattern shown in FIG. 8A. A preamble pattern is a known pattern for detecting the arrival of a signal, and is defined for digital wireless signals in most wireless communication standards currently in use. Alternatively, if the wireless communication standard has periodicity in signal transmission timing in addition to the preamble pattern, the periodicity may be regarded as a fixed pattern and used as a reference signal. For example, the reference signal 801 shown in FIG. 8B is a frequency hopping signal with a fixed frequency hopping pattern, and this period may be used as the reference signal.

FIG. 9A is a diagram showing an example of a delay profile 900, and FIG. 9B is a diagram showing an example of a wireless signal 901 from which the delay profile 900 is calculated. In FIG. 9A, a delay profile 900 is shown with relative signal strength on the vertical axis and time on the horizontal axis. In FIG. 9B, the vertical axis represents frequency, the horizontal axis represents time, and the color shading represents the signal strength of the wireless signal 901. When such a radio signal 901 arrives, by calculating the cross-correlation with the reference signal 801, a correlation peak shown in the delay profile 900 can be obtained. The shape of the delay profile 900 changes due to the effect of fading caused by the overlap of the measurement signal and the multipath reflected wave. A wireless event can be detected by classifying a change in the shape of the delay profile 900 using machine learning, which will be described later, and associating it with an event.

FIG. 10 is a flowchart of a process of associating wireless signals with events and learning by machine learning. For learning, real-world information and a delay profile calculated from a wireless signal are simultaneously obtained (1001). At this time, examples of real-world information include a situation in which a person is in a specific position, a situation in which a person is performing a specific gesture, and video data from a camera, which may affect the radio wave propagation environment. If the situation is such that the situation is not limited to the above-mentioned example. Next, a state event ID is generated for classifying the acquired real world information using the learning table 312 (1002). Next, a delay profile ID corresponding to the state event ID is generated (1003). Then, the generated state event ID and delay profile ID are associated and stored in the learning table 312 (1004). The learning that associates one training signal 305 with one event is thus completed, and the next training signal 305 is awaited (1005). Although the learning process for generating the learning table 312 has been described in FIG. 10, the learning table 312 obtained by learning of another wireless sensor 103 may be introduced via a recording medium or a network. In this case, the wireless sensor 103 may not include the learning section 302, and the detecting section 303 may be provided with a learning table 312.

FIG. 11 is a flowchart of a process of measuring a wireless signal in the actual operation and estimating an event by machine learning based on the learning table 312. In the process of event detection, first, the real-world event associated with the wireless signal that initiates the reception of the wireless signal is unknown. Then, a delay profile calculated by the delay profile acquisition process shown in FIG. 7 is acquired from the received radio signal (1103). Then, the acquired delay profile is classified by machine learning, and the presence or absence of a corresponding event is estimated (1104). If there is a corresponding event (Yes in 1105), a result indicating that the event has been detected is output to the outside (1106). Through the processing described above, event detection using wireless signals can be realized.

FIG. 12 is a diagram showing the configuration of the learning table 312. The learning table 312 includes channel state information ID 1201, delay profile data 1202, and state event ID 1203. In the learning process shown in FIG. 10, a delay profile ID 1201 for classifying delay profiles calculated from wireless signals and a state event ID 1203 for classifying real-world information are stored, and wireless signals are associated with real-world events. be able to. In the detection process of FIG. 11, an event can be detected by referring to the learning table 312 and estimating the event from the measured wireless signal.

FIG. 13 is a sequence diagram in a wireless communication system capable of acquiring channel state information (CSI) according to a comparative example of the present invention. A wireless system that can acquire channel state information is, for example, a wireless communication standard such as IEEE802.11ac. With reference to FIG. 13, the manner in which the wireless base station 101, wireless terminal 102, wireless sensor 103, and server 107 operate in cooperation will be described. The wireless base station 101 transmits a CSI request to the wireless terminal 102 at an arbitrary timing. Upon receiving the CSI request, the wireless terminal 102 transmits CSI information to the wireless base station 101. At this time, the wireless sensor 103 intercepts the signal from the wireless terminal 102 to acquire CSI information, and transmits the acquired channel state information to the server 107. The server 107 classifies the received channel state information using machine learning, estimates events, and detects events. The wireless base station 101 and the wireless terminal 102 continue to transmit and receive data, ACK signals, and the like. CSI information is transmitted at a timing predetermined by communication standards. Therefore, there is a problem in that it is difficult to immediately obtain sensing results at the required timing in accordance with a request from the wireless sensor 103 side.

FIG. 14 is a sequence diagram in a wireless communication system in which channel state information (CSI) cannot be acquired according to an embodiment of the present invention. For example, in IEEE802.11n, CSI information cannot be obtained without using an IC chip and software provided by a specific manufacturer. Furthermore, in communication standards that do not perform MIMO communication, CSI information is not transmitted and received between wireless devices. With the configuration shown in FIG. 14, wireless sensing is possible in a wireless communication system in which channel state information cannot be acquired. With reference to FIG. 14, the manner in which the wireless base station 101, the wireless terminal 102, the wireless sensor 103, and the server 107 operate in cooperation will be described. Radio base station 101 transmits data to radio terminal 102 at arbitrary timing. Upon receiving the data, the wireless terminal 102 transmits an ACK signal to the wireless base station 101 to notify that the data signal has been received. At this time, the wireless sensor 103 intercepts the data transmitted from the wireless base station 101, extracts a known pattern such as a preamble signal included in the data, and obtains a delay profile. Then, the delay profile is sent to the server 107. The server 107 classifies the received delay profile using machine learning, estimates events, and detects events. In the configuration shown in FIG. 14, since signals such as data signals and ACK signals that can be transmitted and received at arbitrary timings are used, sensing requests can be transmitted from the wireless sensor side at arbitrary timings. In addition, since data other than channel state information and signals containing other fixed patterns can be used for sensing, sensing can be done at any timing or the frequency of sensing can be increased, making it possible to realize sensing that requires more real-time performance. .

Although FIG. 14 describes a configuration in which the wireless sensor 103 acquires delay profiles and the server 107 classifies the delay profiles and detects events, the role of the wireless sensor 103 and server 107 is not limited to this. The wireless sensor 103 may classify delay profiles and detect events, or the wireless sensor 103 may only receive wireless signals, and the server 107 may execute all processing after creating the delay profile.

With the configuration and processing described above, the wireless event detection system 200 according to the embodiment of the present invention can detect a wireless event even in a wireless communication system in which CSI information cannot be acquired. Furthermore, compared to a sensing system that uses CSI information, sensing can be performed at arbitrary timing and the frequency of sensing can be increased, making it possible to realize highly immediate sensing that requires more real-time performance. As a result, it can be applied to systems that detect dangerous conditions and sense more complex events.

Note that the present invention is not limited to the embodiments described above, and includes various modifications and equivalent configurations within the scope of the appended claims. For example, the embodiments described above have been described in detail to explain the present invention in an easy-to-understand manner, and the present invention is not necessarily limited to having all the configurations described. Further, a part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Further, the configuration of one embodiment may be added to the configuration of another embodiment. Further, other configurations may be added, deleted, or replaced with a part of the configuration of each embodiment.

Further, each of the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole by hardware, for example by designing an integrated circuit, and a processor realizes each function. It may also be realized by software by interpreting and executing a program.

Information such as programs, tables, files, etc. that realize each function can be stored in a storage device such as a memory, a hard disk, or an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

Furthermore, the control lines and information lines shown are those considered necessary for explanation, and do not necessarily show all the control lines and information lines necessary for implementation. In reality, almost all configurations can be considered interconnected.

100 Wireless communication environment 101 Wireless base station 102 Wireless terminal 103 Wireless sensor 104 Monitoring target 105 Network hub 106 External network 107 Server 200 Wireless event detection system 208 Wireless radio waves 213 Learning table 301 Wireless analysis unit 302 Learning unit 303 Detection unit 304 Wireless signal reception Section 305 Training signal 306 Reference signal input section 307 Reference signal 308 Radio signal reception section 309 Signal 310 Training signal acquisition section 311 Delay profile acquisition section 312 Learning tables 313, 315 Clustering analysis sections 314, 316 Clustering result 317 Event detection threshold update section 318 Determination unit 401 Wireless processing unit 402 Information processing unit 403 Antenna 404 Analog high frequency circuit unit 405 Digital signal processing unit 406 CPU
407 Memory 408 Storage unit 409 External IF
410 Network IF
411 data bus

Claims (9)

A detection system for detecting a state event, the detection system comprising:
comprising: a radio analysis unit that acquires a delay profile representing a correlation between a received radio signal and a reference signal; and a detection unit that detects the state event from the delay profile;
The radio analysis unit includes a radio reception unit that receives a radio signal, a reference signal input unit that outputs a reference signal in which a fixed pattern portion is extracted from the received radio signal, and a reference signal input unit that outputs a reference signal in which a fixed pattern portion is extracted from the received radio signal. and a delay profile acquisition unit that performs a correlation calculation with the signal and outputs a delay profile,
The detection unit includes a clustering analysis unit that outputs a first clustering result generated by classifying the delay profile, and a determination unit that uses the first clustering result to detect a state event indicating real-world information. A detection system comprising: The detection system according to claim 1,
comprising a learning unit that learns by associating the state event and the delay profile,
The learning unit includes a clustering analysis unit that outputs a second clustering result generated by classifying the delay profile, and an event detection threshold updating unit that learns by associating the second clustering result with the state event. A detection system characterized by: The detection system according to claim 1,
The detection system is characterized in that the reference signal input section outputs the reference signal extracted from the wireless signal to a recording medium. The detection system according to claim 1,
Detection characterized in that the delay profile acquisition unit outputs the generated delay profile when a correlation peak value is larger than a predetermined threshold as a result of correlation calculation between the received radio signal and the reference signal. system. The detection system according to claim 1,
The detection unit includes:
obtain the delay profile;
Estimating the presence or absence of a state event corresponding to the obtained delay profile,
A detection system characterized by outputting an event detection result to the outside if there is a corresponding state event. 3. The detection system according to claim 2,
The learning department is
Acquire real-world information and training signals, which are wireless signals, at the same time.
Adding state event identification information for classifying the acquired real world information,
adding delay profile identification information to a second delay profile generated from a training signal corresponding to the state event identification information;
A detection system characterized in that the state event identification information and the delay profile identification information are learned in association with each other. 7. The detection system according to claim 6,
The learning department is
The learning table has a learning table in which the delay profile identification information and the state event identification information are recorded, and can associate the wireless signal with the state event indicating the real world information,
A detection system characterized in that the given delay profile identification information and the state event identification information are stored in association with each other in the learning table. A sensor device for detecting a state event, the sensor device comprising:
comprising: a radio analysis unit that acquires a delay profile representing a correlation between a received radio signal and a reference signal; and a detection unit that detects the state event from the delay profile;
The radio analysis unit includes a radio reception unit that receives a radio signal, a reference signal input unit that outputs a reference signal in which a fixed pattern portion is extracted from the received radio signal, and a reference signal input unit that outputs a reference signal in which a fixed pattern portion is extracted from the received radio signal. and a delay profile acquisition unit that performs a correlation calculation with the signal and outputs a delay profile,
The detection unit includes a clustering analysis unit that outputs a first clustering result generated by classifying the delay profile, and a determination unit that uses the first clustering result to detect a state event indicating real-world information. A sensor device comprising: A detection method wherein the detection system detects a state event, the detection method comprising:
The detection system includes a radio analysis unit that acquires a delay profile representing a correlation between a received radio signal and a reference signal, and a detection unit that detects the state event from the delay profile,
The detection method includes:
The wireless analysis unit receives a wireless signal,
The radio analysis unit calculates a reference signal in which a fixed pattern portion is extracted from the received radio signal,
the radio analysis unit performs a correlation calculation between the received radio signal and the reference signal and outputs a delay profile;
The detection unit calculates a first clustering result generated by classification of the delay profile,
A detection method characterized in that the detection unit detects a state event indicating real-world information using the first clustering result.

Images