DK202400084Y3 - Electronically controlled lock and sensor arrangements - Google Patents

Abstract

An installation is provided which includes an electronically controlled lock mounted on a door, the electronically controlled lock including a first detection arrangement in the form of an accelerometer or a magnetometer, the installation comprising a unit, separate from the lock, mounted on the door or a door frame, which receives the door, the unit including a second detection arrangement, the second detection arrangement comprising an accelerometer or a magnetometer, the installation including digital processor circuitry configured to receive signals from the first and second detection arrangements and to process the signals to detect a potential intrusion attempt based on sensor data recorded simultaneously from different detection positions relative to the door and/or door frame.

Description

DK 202400084 Y3

Electronically controlled load and sensor arrangements

Technical area

The present invention relates to electronically controlled locks and sensor arrangements for a door in a building, associated security monitoring installations that include one or more such locks, and associated systems and subsystems. The electronically controlled lock may be in the form of, or include, an electronic lock driver, with an electric motor or solenoid, to drive the mechanism of a mechanical lock, rather than being a fully integrated smart lock.

Background

Locks for securing doors in buildings, to prevent unauthorized opening thereof and entry into the building (access control), have been known for hundreds if not thousands of years, the general idea being to prevent anyone not in possession of a suitable key from gaining access to the protected building protected by the door. However, for as long as there have been locks, there have been malicious individuals who will attempt to open the lock without a suitable key, for example by using a key-picking or “bumping” key. Over the past decade, electromechanical locks, known as smart locks, have been introduced in an attempt to improve security and possibly user convenience. Such smart locks typically respond to the presentation of an electronic token, or a command signaled electromagnetically, which —o replaces the usual key and keyhole mechanism, to cause a locking or unlocking of a locking mechanism. Such locks can operate with access cards or other tokens, or active devices such as phones that can generate RF signals or be read by

RF signals (e.g. Bluetooth, NFC, RFID or similar), or magnetic, and/or a biometric interface (such as an iris, fingerprint or facial recognition interface). Many smart lock designs lack a keyhole, so that lock activation can only be accomplished using some electromagnetic method, many home users seem to find comfort in the continued presence of a keyhole, which cooperates with a tangible key that can be carried by a user.

Smart locks can be used as stand-alone security measures, and such a lock may include an alarm sound that is activated if the lock detects a tampering attempt. However, smart locks can also be used as part of or in conjunction with a building security monitoring system, and also in conjunction with smart home installations, often with the ability to signal any detected tampering or disturbance attempt. Such a smart lock, which can also be referred to as an electronically controlled lock, is

, DK 202400084 Y3 described in the applicant's previous patent application first published as WO2023/227258. This smart lock is designed with the aim of meeting the requirements of ENS0131 Class 2, so that the smart lock can be incorporated into a security monitoring system that otherwise meets the requirements of ENS0131 Class 2 or above.

Some smart locks are designed to work with conventional lock cylinder hardware, providing either a convenient upgrade to an existing lock installation or a way to enjoy both the convenience of electronic lock control and the ability to still use mechanical keys. However, such smart locks can be just as susceptible to “bumping” attempts as the mechanical lock cylinders on which they are based.

In this application, the terms “smart lock” and “electronically controlled lock” are to be interpreted as embracing conventional (mechanical) locks to which an electronic lock driver has been added. Such electronic lock drivers may include an electric motor or solenoid that is mechanically coupled to a mechanical lock so that electronics in the lock driver can be used to control operation of (“manipulate”) the lock. The terms “smart lock” and “electronically controlled lock” also encompass an electronic lock driver for mounting on a mechanical locking device.

Although there are many smart lock designs, there is a need for an improved smart lock, particularly one suitable for installation in homes. It would also be desirable for such a smart lock to meet the requirements of EN50131 Class 2, so that the smart lock could be incorporated into a security monitoring system that otherwise meets the requirements of EN50131 Class 2 or above.

Summary

Various aspects of the invention are described herein. These aspects may be used independently, or any two or more aspects may be used together.

According to a first aspect, an electronically controlled lock for a door in a building is provided, the lock comprising a processor, and an accelerometer or magnetometer is coupled to the processor, wherein: the accelerometer or magnetometer is configured to function as a vibration sensor, the processor and the accelerometer or magnetometer together — function as a vibration detector to generate a warning signal if an attempt is made to pry or tamper with the lock; and the processor is configured to first use a first classifier to detect anomalies in signals received from the accelerometer or magnetometer, during or as a result of an event, and if an anomaly is detected, the processor uses a second classifier to classify

2 DK 202400084 Y3 the irregularity as belonging to an identified one of a plurality of classes of events, and optionally where the processor is configured to send a warning signal, using a transceiver of the lock, if an attempt is made to pry or tamper with the lock.

According to another aspect, there is provided a smart lock comprising: an accelerometer or magnetometer configured as a shock detector, a processor configured to detect an anomaly in a signal received from the accelerometer or magnetometer, the processor configured to implement an autoencoder, the autoencoder configured to classify the received signal according to pre-trained non-anomaly signal data and to re-evaluate the signal therefrom, and the processor configured to detect an anomaly depending on a difference between the received signal and the re-evaluated signal. Optionally, the processor may be configured (e.g., programmed) to detect an anomaly according to an energy difference between the received signal and the re-evaluated signal.

According to a third aspect, an electronically controlled smart lock for a door in a building is provided, the lock comprising an accelerometer or a magnetometer, a processor coupled to the accelerometer or magnetometer and configured to use a first classifier to detect anomalies in signals received from the accelerometer with respect to an event, and upon detection of an anomaly to use a second classifier to attempt to classify the event into one of a plurality of event classes, at least one of the classes > denotes an alarm event, and at least one of the classes corresponds to a non-alarm event, the processors being configured to signal an alarm event if the classifier determines that an event is an alarm event, and optionally also if an event cannot be classified as a non-alarm event.

A fourth aspect of the invention provides a vibration sensor (e.g., for a door or window and configured to detect intrusion), the vibration sensor including a processor configured to use a machine learning model to process sensor data to determine the presence of anomalies in the data, and if an anomaly is detected/determined to generate an output. The vibration sensor may include an accelerometer and/or a magnetometer, plus optionally a microphone. The microphone may be located within the lock and adapted to detect sounds of attacks on the lock, such as prying. A further microphone may be provided within the lock, or within another detection arrangement, to detect ambient sounds, and in particular ambient sounds on the exposed or at-risk side of the door — which sounds may be used to help distinguish between false alarms and real alarm events (e.g., as the result of signals,

1 DK 202400084 Y3 representing such sounds which are made available to and processed by a device that monitors the installation of which the lock and/or the other detection arrangement is a part.

Any microphone forming part of the lock or other detection arrangement is preferably located within the housing of the lock or other detection arrangement in such a way that the microphone is shielded from direct attack through any opening, port or grille through which ambient sound reaches the microphone — e.g. by the microphone being offset relative to any such opening or grille, for example by being mounted transversely thereto, so that the microphone is not exposed to damage if a foreign object (such as a screwdriver or drill bit) is inserted through the hole, opening or — grille. A microphone intended to detect the sound of knocking may be coupled by an open passage to a keyhole in the lock, but is preferably again offset, e.g. mounted transversely to the axis of the bore of the keyhole, so that the microphone is not easily attacked through the keyhole.

In some embodiments, the processor is programmed to act as an autoencoder to process sensor data.

For example, in some embodiments, the autoencoder is configured to adapt and/or classify a received signal according to pre-trained non-anomaly signal data and to re-evaluate the signal therefrom, and the processor is configured to detect an anomaly depending on a difference between the received signal and the re-evaluated signal. The processor may be configured to detect an anomaly in accordance with an energy difference between the received signal and the re-evaluated signal.

In some embodiments, if the processor determines the existence of an anomaly, the processor is programmed to analyze sensor data, or processed sensor data, to allocate a detected event to one of several event types, the event types including at least one threat class and at least one non-threat class.

The threat class may include at least one of lock picking, lock bumping, and lock snapping.

A fifth aspect of the invention provides an electronically controlled lock that includes a vibration sensor as previously described, and optionally wherein the smart lock comprises a smart electronic lock driver operatively coupled to a mechanical lock.

The lock may include a keyhole, and the machine learning model may be trained to recognize as anomalous events at least one of lock picking, lock bumping, and lock snapping. A lock with a keyhole may include a microphone, for example, adapted to preferentially detect sound from a keyhole in the lock and/or from the vicinity of a lock cylinder, and the digital processing circuit may be adapted to receive signals or data from

. DK 202400084 Y3 the microphone in addition to signals received from the accelerometer or magnetometer; optionally, the lock may further or alternatively include a pressure sensor and/or an insertion detector (possibly based on mechanical displacement of a detection element) for detecting insertion of an element (such as a key, bump key or dirk) into an external keyhole of the lock, and the digital processor circuit (or a remotely located processor) may be arranged to receive signals or data from any such sensor(s) in addition to signals received from the accelerometer or magnetometer or simply as an additional sensor whose signals may contribute to the determination of an anomalous event and possibly to its classification.

The lock may further comprise one or more electronic detection arrangements for performing hand-on-handle or hand-on-lock detection, for example capacitive detection, and/or key-in-lock detection (e.g., by using impedance detection), and the digital processor circuit is configured to use data from the one or more electronic detection arrangements to classify events into specific attack modes such as — lock picking, lock snapping and lock bumping.

A sixth aspect of the invention provides an electronically controlled lock that includes a vibration sensor as previously described, wherein the processor is configured to implement an autoencoder, the autoencoder is configured to classify a signal received from the accelerometer or magnetometer according to pre-trained non- —> irregular signal data and to re-evaluate the signal therefrom, and the processor is further configured to detect an irregularity depending on a difference between the received signal and the re-evaluated signal. The processor may be configured to detect an irregularity according to an energy difference between the received signal and the re-evaluated signal.

According to a seventh aspect, an electronically controlled lock for a door in a building is provided, the lock comprising a processor, and an accelerometer or magnetometer is coupled to the processor, wherein: the accelerometer or magnetometer is configured to function as a vibration sensor, the processor and the accelerometer or magnetometer together functioning as a vibration detector to generate a warning signal if an attempt is made to pry or tamper with the lock; the processor is configured to first detect irregularities in signals received from the accelerometer or magnetometer, during or as a result of an event, and in the event of detection of an irregularity, the processor is configured to use a classifier to classify the irregularity as belonging to an identified one of a plurality of classes of events, and optionally wherein

3 DK 202400084 The Y3 processor is configured to send a warning signal, using a transceiver of the lock, if an attempt is made to pry or tamper with the lock.

Locks according to the seventh aspect may include a keyhole and may further comprise one or more additional sensors in the form of (1) a microphone, for example adapted to preferentially detect sound from a keyhole in the lock and/or from the vicinity of a lock cylinder, and the digital processor circuit may be adapted to receive signals or data from the microphone in addition to signals received from the accelerometer or magnetometer; (11) a pressure sensor; and/or (ii) an insertion detector (possibly based on mechanical displacement of a detection element) for detecting insertion of an element (such as a key, bump key or dirk) into an external keyhole of the lock: and the processor (or a remotely located processor) may be adapted to receive signals or data from any such additional sensor in addition to signals received from the accelerometer or magnetometer, the digital processor circuit (or a remotely located processor) being programmed or configured to use signals or data from any such additional sensor in determining and/or classifying an anomalous event.

According to an eighth aspect, there is provided a building security monitoring installation, comprising a control unit for controlling operation of the security monitoring installation, the installation including: a window or door vibration sensor incorporating a first detector, the first detector comprising an accelerometer or a magnetometer; an electronically controlled lock according to any variant of the seventh aspect incorporating a second detector, the second detector comprising an accelerometer or a magnetometer; the window or door vibration sensor and the smart lock being configured to send signal data from the first and second detectors to the control unit; and the control unit being configured/programmed to classify an event based on the signal data received from the first and second detectors and optionally based on signal data from one or more of the aforementioned additional sensors in the lock.

A ninth aspect of the invention provides a building security monitoring installation comprising a control unit for controlling operation of the security monitoring installation, the installation including: an electronically controlled lock according to any variant of the seventh aspect coupled to a door in the building, the electronically controlled lock incorporating a first detector, the first detector comprising a

, DK 202400084 Y3 accelerometer or a magnetometer; a door vibration sensor incorporating a second detector, the second detector comprising an accelerometer or a magnetometer, coupled to the same door at a position remote from the lock; the door vibration sensor and the smart lock being configured to send signal data from the first and second detectors to the control unit; and — the control unit being configured/programmed to classify an event based on the signal data received from the first and second detectors and optionally based on signal data from one or more of the aforementioned additional sensors in the lock.

A tenth aspect of the invention provides an installation which includes an electronically controlled lock mounted on a door, the electronically controlled lock being according to any variant of the seventh aspect and including a first detection arrangement in the form of a first accelerometer and/or magnetometer, the installation comprising a unit, separate from the lock, mounted on the door or a door frame which receives the door, the unit including a second detection arrangement separate from the first detection arrangement, the second detection arrangement comprising one or more of an accelerometer, a magnetometer, — the installation including digital processor circuitry configured to receive signals from the first and second detection arrangements and optionally from one or more of the aforementioned additional sensors in the lock and to process the signals to detect a potential attempt at intrusion based on data recorded simultaneously from different detection positions relative to the door and/or door frame.

An eleventh aspect of the invention provides an electronically controlled lock that includes a vibration sensor as previously described, wherein the processor is configured to use a first classifier to detect irregularities in signals received from the accelerometer or magnetometer in relation to an event, and upon detection of an irregularity to use a second classifier to attempt to classify an event into — one of a plurality of event classes, at least one of the classes representing an alarm event, and at least one of the classes corresponding to a non-alarm event, the processor being configured to signal an alarm event if the classifier determines that an event is an alarm event, and optionally also if an event cannot be classified as a non-alarm event.

A twelfth aspect of the invention provides a building security monitoring installation which includes at least one vibration sensor with any of the previously described, and or an electronically controlled lock as previously described.

2 DK 202400084 Y3

A thirteenth aspect of the invention provides a building security monitoring installation, optionally as above, comprising a control unit for controlling operation of the security monitoring installation, the installation including: a window or door vibration sensor incorporating a first detector, the first detector comprising an accelerometer or a magnetometer; an electronically controlled lock, optionally as defined by any of the features above, incorporating a second detector, the second detector comprising an accelerometer or a magnetometer; the window or door vibration sensor and the smart lock being configured to send signal data from the first and second detectors to the control unit; and the control unit being configured/programmed to classify an event based on the signal data received from the first and second detectors.

A fourteenth aspect of the invention provides a building security monitoring installation optionally as defined above, comprising a control unit for — controlling operation of the security monitoring installation, the installation including: an electronically controlled lock, optionally as defined by any of the features above, coupled to a door in the building, the electronically controlled lock incorporating a first detector, the first detector comprising an accelerometer or a magnetometer; a door vibration sensor incorporating a second detector, the second detector comprising an accelerometer or a magnetometer, coupled to the same door at a position remote from the lock; the door vibration sensor and the smart lock being configured to send signal data from the first and second detectors to the control unit; and the control unit being configured/programmed to classify an event based on — the signal data received from the first and second detectors.

A fifteenth further aspect of the invention provides a building security monitoring installation, optionally as previously described, wherein the control unit includes a processor configured to use a machine learning model, such as an autoencoder, to process received signal data to determine the presence of anomalies in the received signal data.

A sixteenth aspect of the invention provides an installation which includes an electronically controlled lock mounted on a door, the electronically controlled lock including a first detection arrangement in the form of an accelerometer and/or magnetometer, the installation

2 DK 202400084 Y3 comprises a unit, separate from the lock, mounted on the door or a door frame which receives the door, the unit including a second detection arrangement separate from the first detection arrangement, the second detection arrangement comprising an accelerometer and/or a magnetometer (such that the first and second detection arrangements can each —> comprise the same detector type or different detector types of both), the installation including digital processor circuitry configured to receive signals from the first and second detection arrangements and to process the signals to detect a potential intrusion attempt based on data recorded simultaneously from different detection positions relative to the door and/or door frame.

The processor circuit may include a processor programmed to function as an autoencoder to process data from the first and second detection arrangements.

In some embodiments, if the processor determines the existence of an anomaly, the processor circuitry is programmed to analyze the sensor data, or processed sensor data, to allocate a detected event to one of a plurality of event types, wherein the event types include at least one threat class and at least one non-threat class.

The threat class may optionally include at least one of lock picking, lock bumping, and lock snapping.

In some embodiments, the processor circuit includes a processor that is a component of the electronically controlled lock.

The installation may further comprise a security monitoring system for buildings of which the door is a part, and optionally at least part of the processor circuit may comprise a processor of a controller of the security monitoring system. The controller of the security monitoring system may optionally be provided by a video doorbell device.

The security monitoring system may have at least one operating mode configured to notify alarm events to a remote monitoring unit. For example, the security monitoring system may have at least one operating mode configured to notify the remote monitoring unit of anomalous events detected in the sensor data.

A seventeenth aspect of the invention provides a building security monitoring installation, which includes a controller for controlling operation of the security monitoring installation, the security monitoring installation further comprising: a door vibration sensor incorporating a first accelerometer and/or magnetometer, the vibration sensor being coupled to a door in the building; an electronically controlled lock incorporating a second accelerometer, the electronically

0 DK 202400084 Y3 controlled lock is coupled to the device in the building; wherein the vibration sensor and the electronically controlled lock are configured to send signal data obtained from their accelerometer(s) and/or magnetometer(s) to the control unit; and wherein the control unit is configured/programmed to classify an event based on the signal data obtained from the first and second vibration sensors.

The controller may include a processor configured to use a machine learning model, such as an autoencoder, to process received signal data to determine the presence of anomalies in the received signal data.

The controller may be configured/programmed to provide a quantized output whose value is selected from at least three levels (i.e., a non-binary output).

Non-limiting features and advantages of the invention include the ability to detect irregular vibrations, for example, compared to vibrations that can be detected under normal operating conditions of the smart lock and door. Where a machine learning algorithm is used that is trained only with, or by prioritizing, training data from “normal” conditions, it is possible to detect any suspicious vibrations as an anomaly, even if these correspond to an attempted intrusion or attack on the lock that the system has not been specifically trained to recognize (e.g., a new type of attempted intrusion or attack on the lock). Where multiple detection arrangements are used, one at the lock and another at another location spaced from the lock, it is possible to derive additional context by detecting the same vibration event substantially simultaneously from or at multiple detection locations. This can enable automatic distinction between an intrusion attempt that focuses on the lock and another intrusion attempt that focuses on another part of the door. It can also help reduce the incidence of — false alarms compared to using only a single detection arrangement at the door.

Further aspects of the invention are defined in the claims. Independent protection is also claimed in the claims for any novel feature and/or novel idea described herein and/or in the drawings, whether or not emphasis is placed on it.

Brief description of the figures

Embodiments of the invention will now be described, by way of example only, with reference to the accompanying figures, in which:

Figure 1 schematically shows a building with a smart lock installation, which may include video surveillance when approaching the smart lock.

] DK 202400084 Y3

Figure 2 schematically illustrates the main elements of a smart lock according to one aspect of the invention;

Figure 3A shows a mortise lock of a type with which aspects of the invention may be used, - installed in a door;

Figure 3B shows a device that includes the mechanism and electronics of an electronic lock according to aspects of the invention;

Figure 3C schematically shows a longitudinal section through the device of Figure 3B;

Figure 4 schematically illustrates the operation of devices according to aspects of the invention; - Figure 5 schematically illustrates the elements constituting an autoencoder that can be used in aspects of the invention;

Figure 6 schematically illustrates how a trained autoencoder, according to aspects of the invention, can distinguish between “routine” events and aberrant events;

Figure 7 schematically illustrates how an autoencoder may be integrated into a smart lock — according to one aspect of the invention;

Figure 8 schematically illustrates the main elements of a video doorbell according to one aspect of the invention;

Figure 9 schematically illustrates a home security monitoring system built around a video doorbell; and

Figure 10 is generally similar to Figure 9, but also includes the benefits of professional monitoring using a remote alarm receiving center.

Specific description

Figure 1 schematically shows a home (“building”) with a smart lock installation, which may include video surveillance upon approaching the smart lock. In this case, a video doorbell 100 may be provided adjacent the main entrance door 102, in this case the front door, of the house 104. The front door 102 is equipped with a “smart” lock 106 that can be locked or unlocked by electrical control signals received, for example, from a personal communication device 108 of the owner of the house. For example, the video doorbell 100 may cooperate with a backend service 110 that sends push notifications to a user’s personal communication device 108 when the doorbell is pressed (or optionally when someone approaches the doorbell 100 or other video surveillance device). To support such communications, the video doorbell may use Wi-Fi to connect to a Wi-Fi router 112 (or a controller of a security monitoring installation in the building), which

2 DK 202400084 Y3 has a broadband connection 114 to the Internet 115, as the backend service 110 sends push notifications out to a PLMN (public land mobile network) 116, with which notifications can be delivered to the user's personal communication device 108. The same communication path can be used to stream live video (and audio) from the video doorbell 100 — to the user. A reverse communication path may also be supported to allow the user to speak to a person at the doorbell 100, and also allows the user to activate (i.e., open) the lock 106 to allow the person at the door to open the door 102. The signals to activate (open) the lock may be conveyed to the lock 106 directly from the router 112 (or the controller of the security monitoring system), or they may be conveyed via the video doorbell 100, which in turn communicates with the smart lock 106. The smart lock 106 has been designed to meet

Class 2 of EN50131 by incorporating a shock sensor (more generally a vibration sensor) with a tamper detection function.

Any or all of the smart lock, the video doorbell, and the controller of the security monitoring system may additionally or alternatively be provided with at least one transceiver that supports NB-IOT (narrowband IoT supported by 3GPP), LTE-M (LTE machine-type communication), or the like, which are LPWAN technologies that provide secure connections using licensed spectrum in operator-controlled networks.

Figure 2 schematically illustrates the major elements of a smart lock 106, such as that shown in Figure 1, according to embodiments of the invention. The lock 106 is shown installed in a door 102, with a door frame 240 that houses a cooperating striker plate 242 that receives at least one bolt 244. The lock 106 may be configured as a mortise lock, in which case the lock will typically include a mortise lock set or a latch (not shown) in addition to the bolt 244.

Providing the lock 106 as a mortise lock is particularly advantageous if the lock is to be pre-installed in a door (or if the door is to be manufactured with mortises provided for easy installation of the lock on site) rather than retrofitted to an already hung door as a replacement for an existing lock. However, it is generally desired, particularly when a surface-mounted lock is to be replaced, to surface-mount the lock 106 (on the protected, typically interior, side of the door 102) because installation is typically much easier and quicker if a cavity (mortise) is not required to be formed in the door to receive the new lock. In either case, the bolt 224 may be received in a cavity in the door, but with a surface-mounted lock the bolt may also be outside the door. The lock 106 is provided with an actuator 246, which may be a solenoid or a motor, which either drives the bolt into the locked or unlocked position, or which is engaged with or disengaged from a clutch which engages or disengages a handle or lever 248 on the exposed side of the

1 DK 202400084 Y3 the door, and by means of which the bolt can be pulled into the unlocked position, allowing the door to be opened via the exposed (unprotected) side. Typically, if the lock 106 is provided with a coupled handle, the actuator will be a solenoid rather than a motor. The lock 106 may of course be provided with a second handle or lever (not shown), on the protected side of the door 102, for operating the bolt 244 to enable opening and closing of the door 102 from the protected (generally the inside) side of the door 102.

While Figure 2, for ease of illustration and explanation, illustrates a locking installation with only a single bolt 244 and corresponding single striker plate 242, locks according to embodiments of the invention may include or be incorporated into or used with — multi-point locking installations/doors, where in fact more than one bolt is used to secure a door.

The lock 106 includes a power supply 250, which preferably includes a rechargeable battery 252, and preferably also a charging device 254 to enable on-site charging of the rechargeable battery 252. The charging device may be connected to a power supply from a mains power supply, but for ease and simplicity it may be preferable to include — an interface 256 (such as a micro USB port, a USB-C port, a “Lightning” connector, or the like) to receive a low voltage (e.g., SV or 20V) supply (e.g., from a power bank or external charger), which, in the absence of a mains power supply (or in the event of a mains power supply failure) can be used to recharge the rechargeable battery 252 on-site, rather than having to remove the battery for recharging, so that the lock 106 can always be — powered. In an alternative embodiment, the lock may include a coil or other device to allow inductive charging from a suitably placed transmitter connected to an external power source (such as a power bank or power supply from a mains power supply).

The actuator 246 is operatively coupled to and controlled by a processor 258 having an associated memory 260 storing program instructions (“software”) that, when executed on the processor 258, control the operation of the lock 106. The processor 258 may be, for example, a microcontroller or a microprocessor. At least one transceiver is also coupled to and controlled by the processor 258, which transceiver is configured to receive control signals from, and transmit event information to, for example, a controller in the security monitoring system (e.g., from the video doorbell 100 or the controller in the security monitoring installation in the — embodiment illustrated in Figure 1). These may be relatively low bandwidth transmissions (e.g., compared to video transmissions from the video doorbell 100), so that a low bandwidth channel is appropriate — meaning that a transceiver with low power consumption may be used, thereby enabling a satisfactorily long battery life. For example, the transceiver 262 may be configured to operate using an appropriately allocated

DK 202400084 Y3 frequency in the ISM band (for industry, science and medicine) — such as (in Europe) 868MHz. Preferably, the communications to and from the lock are encrypted. As previously noted, the transceiver 262 may additionally or alternatively also support an LPWAN technology such as NB-IOT or LTE-M, with which one can communicate with one or more => user devices (WTRUs) 108 and/or a remote system back-end or remote monitoring station or an alarm receiving center (ARC).

The lock 106 may also include a ringer or speaker (more generally, an audio output device, e.g., an electromechanical ringer or buzzer) 263 operatively coupled to the processor 258. This speaker may be used instead of, or in addition to, a separate — the ringer (e.g., located in an entryway or elsewhere in the building) that sounds when a ring trigger device of the video doorbell is activated — with a processor of the video doorbell 100 optionally configured to use a transceiver to send a ring activation signal tagged with a flag, to which the lock processor 258 and/or the processor of a separate ring sounder device (not shown) responds by using an internal audio output device to ring in response to the activation of the ring trigger device.

The lock 106 may be provided with a housing 259 through which the bolt 244 and any mortise lock set may freely extend, the handle or grip 248 being of course external to the housing 259. Preferably, the housing is made of a structural plastic material or a non-ferromagnetic material so as not to interfere with the operation of one or more magnetometers —> optionally provided as part of the lock 106.

The lock 106 may further include a magnetometer 264, which may be configured to sense and monitor a first magnetic field provided by an optional first magnet 266, which may be fixedly mounted relative to the frame 240 or other support against which the door 102 closes. The same magnetometer 264, or an additional magnetometer (not shown), may also be configured to sense and monitor a second magnetic field provided by a second magnet 270, which may be attached to the door, preferably mounted in the door—i.e., between the protected surface and the unprotected surface of the door, although it is also possible to attach (e.g., adhere with adhesive and/or use a mechanical attachment such as a screw) a magnet 270 to the surface of the door, (the surface on the inside (e.g., indoor side) of the door). This second magnet 270, if present, is external to the lock 106 and is provided such that a magnetometer configured to sense the magnetic field of the second magnet 270 can generate a tamper signal to indicate an attempt to remove the lock 106 from the door 102. At the same time, a magnetometer 264 can be configured to sense and monitor the first magnetic field provided by a first magnet 266 and can provide a

DK 202400084 Y3 signal to indicate the status of the door, e.g. whether the door 102 is open or closed. The processor 258 may be configured to generate a warning signal that is sent to a controller of the security monitoring system, e.g. an amplified video doorbell according to some aspects of the invention, for onward reporting to a monitoring service and/or 5 — the owner/occupant of the house, if the opening of the door is detected without the lock having been unlocked and/or the alarm having been disarmed — since such a combination of circumstances may indicate that the door 102 has been forced open. It will be appreciated that a magnetometer can act as a shock or vibration sensor, as it may be able to respond to short-term interference or disturbances of the magnetic field, from an associated magnet for example, as a result of the magnetometer being displaced relative to the source of the magnetic field: for example, a magnetometer mounted on or in a door may detect a shock caused by something hitting the door (e.g. a criminal trying to get in, or a football kicked hard against the door) as a result of small changes in the separation between the magnetometer and the source of the magnetic field. It is thus possible to use a magnetometer instead of an accelerometer in constructing a vibration sensor. In general, embodiments of the present invention may be implemented using an accelerometer or a magnetometer, or both. Depending on the application and the precise details of any installation, an accelerometer may be more sensitive to small disturbances, and therefore it may be better to use an accelerometer than a magnetometer. > However, in other situations, a magnetometer may be preferred due to its relative insensitivity to minor disturbances.

The lock 106 may further include an accelerometer 275, which is also coupled to the processor 258, and which is configured to function as a shock sensor (or more generally a vibration sensor) to provide the processor with signals to enable the processor to detect attempts to “pick” or “bump” the lock, drill out the lock cylinder, and other mechanical attacks on the lock, as well as provide signals in the event of attempts to, for example, break the door (or force open the lock) with a sledgehammer or a battering ram.

The processor 258 preferably uses machine learning, e.g. deep learning, and/or other AI techniques to distinguish between accelerometer signals generated as a result of — “everyday events” such as a normal “I’m here” knock on the door, the door being unlocked with an appropriate key, an unlocked door being slammed, an accidental hit from a football, or the like from irregular signals generated as a result of criminal activity such as attempted lock picking or bumping, or attempts to force a door open or smash the door.

The application of AI techniques to this problem of distinguishing between signals that

6 DK 202400084 Y3 represents irregular events and "everyday" or "non-irregular" or "ordinary" events are described and discussed later in this application. The provision of the accelerometer 275 also provides a potential starting point for certifying the lock as a shock detector (for an alarm system of at least Class 2 or Class 3) according to EN —50131-2-8 (also referred to herein as a "peripheral alarm device"). The accelerometer may be a three-axis device, although a simpler device may also be used to provide satisfactory results. As already noted, in some situations it may be the case that the relevant sensor is a magnetometer rather than an accelerometer.

A suitable accelerometer is available from STMicroelectronics as the LIS2DTW12, — although there are of course many similar alternative devices available from other manufacturers.

Conveniently, the accelerometer may be configured to be passive when not exposed to a certain level of activity for more than a predetermined period of time (to reduce power consumption, thereby extending battery life in battery-powered devices), in which case the accelerometer may be further configured to be activated rapidly when exposed to a certain amount of stimulation — such as an RMS signal value of a threshold level or more (for example, on any of the detection axes) (possibly an increasing level of activity). The STMicro device in question is an "always on" device, but has, in sleep mode, a sampling rate of 12.5 Hz, which means a sample > every 80ms. Activation is fast, and the device can switch to a sampling rate of up to 1600 Hz upon activation (although the optional 800 Hz sampling rate is likely to be fast enough for this application), so it is generally expected that an accelerometer that has been activated will likely continue to experience stimulation occurring from the activation event. This stimulation will cause an analyzable output signal. The STMicro device mentioned (like likely its counterparts from other manufacturers) can also generate an interrupt signal upon activation so that the processor can also be notified of the activity.

Thus, the processor 258 may be configured to process first sensor data from at least one first sensor, for example from a first sensor, such as the accelerometer 275 or a magnetometer, to determine the presence of anomalies represented by the data, and if an anomaly is detected/determined, generate an output. Where a detection device has a sleep mode, such as the STMicro accelerometer just mentioned, it may be preferable not to consider the fact that the device being activated is itself an irregular event—although activity following activation of the device may well

7 DK 202400084 Y3 constitute an irregular event and be treated as such by the processor. An attack event that activates a detection device may thus constitute an irregular event, even though the simple act of activating the detection device would not in itself be considered an irregular event.

The processor 258 may also be configured to process other sensor data provided by at least one other sensor (e.g., a magnetometer 264 or another accelerometer or magnetometer optionally located elsewhere (i.e., outside the lock and optionally remote from the lock) on the door 102 and/or on the door frame 240) and to consider both the first sensor data and the second sensor data in determining whether the data indicates the occurrence of an anomaly and, if so, to provide an output. In either case, the output provided by the processor 258 may be sent to a remote receiver (such as a user device, e.g., a WTRU, a system back-end, a remote monitoring station, or a

ARC, a control unit of a security monitoring installation in the building, a “smart doorbell” as described elsewhere in this application). Additionally or alternatively, the output provided by the processor 258 may be subjected to a classification process (for example, by using a classifier (or autoencoder) running on the processor 258 or on another processing device in the lock or elsewhere) to classify events into particular threat types (e.g., lock bumping, lock picking, lock picking, door slamming, etc.), and optionally to classify events into particular everyday categories, and optionally to classify an event as being between a threat and an everyday event.

Based on such classification, the processor may, for example, be configured to indicate in signals with warnings and "tampering detected" the nature of the attack that resulted in the sending of the signal. For example, it is desirable to distinguish signals resulting from mechanical impact as a result of a forceful attack on the door or door frame from those caused by attempts to pick the lock, and those caused by attempts to break the lock from the door (removal of the lock from its mounting surface). Such information may be useful in distinguishing between genuine attacks (genuine intrusion), tampering, and false alarms, i.e., in verifying that a security alert is one that can be forwarded to the police for intervention.

Optionally, the lock may include a microphone 276, for example, adapted to preferentially detect sound from the keyhole and/or from the vicinity of the lock cylinder, and the processor 258 may be adapted to receive signals or data from the microphone as one or more of the other sensors. As previously described, the microphone is preferably protected from direct attack by

DK 202400084 Y3 the keyhole, for example by being offset relative to the keyhole, or arranged orthogonally relative to the axis of the keyhole. Optionally, the lock may further or alternatively include a pressure sensor 278 and/or an insertion detector 279 (possibly based on mechanical displacement of a detection element) for detecting insertion of an element 5 (such as a key, bump key or dirk) into an external keyhole of the lock, and the processor 258 (or a remotely located processor) may be arranged to receive signals or data from any such sensor(s) as one or the other sensor or simply as an additional sensor whose signals may contribute to the determination of an anomalous event and possibly to the classification thereof. The lock may also optionally include a microphone for — detecting ambient sounds, particularly those from the side of the door exposed to risk. A hole or aperture may be provided in the locking body to enable the microphone to detect such ambient sounds, and the opening or aperture may be provided with a grille or the like to protect the microphone from attack. As previously described, the microphone is preferably further protected from attack through the hole or aperture (e.g. from attack by — the insertion of an object through the hole or aperture, by being offset from, or mounted orthogonally to an axis of the hole/aperture.

Optionally, the processor is configured to indicate, in the signal reporting an event, the identity of the sensor(s) causing the signal to be sent, i.e. whether it was the accelerometer 275 or the magnetometer 264, (and indicating which magnetometer if the lock contains more than one), the external accelerometer, the external magnetometer, the microphone, the pressure sensor, the displacement detector, or a combination of these — and, at least in the case of a single magnetometer, the nature of the change in the observed magnetic environment, e.g. whether the magnetometer detects a change in the magnetic field from a door magnet 270 or from a frame magnet, and whether — the change is a reduction in magnetic field strength, and an increase in magnetic field strength, or a change in polarity — both of which suggest a possible attempt to fool the system by adding a fake magnet to trick the magnetometer ("masking"). This information can also be useful in distinguishing between genuine intrusion attempts, tampering, and false alarms, for example, distinguishing between vibrations caused by an intrusion or vibrations caused by road or train traffic or the impact of a kicked soccer ball.

While the foregoing description was in the context of the electronics of the lock being provided in a lock case together with the main mechanical elements of the lock, the invention also contemplates alternative arrangements where the electronics of the smart lock are provided in a unit other than the lock case. For example, with a mortise lock, where the

9 DK 202400084 Y3 main mechanical parts of the lock are provided in a mortise in a door (preferably contained in a lock case in the door), a separate unit may be provided (preferably) on the protected (e.g. internal) surface of the door to house the electronics of a smart lock, the separate unit being mechanically coupled to the mechanical parts of the lock in the door. Such a device — may be used to enable smart lock functionality to be "retrofitted" to a previously installed lock, for example by removing the existing lock cylinder and replacing it with a new cylinder which provides mechanical connection to the new separate unit by means of which the lock can be locked or unlocked remotely. If the locking device provides a latch device, this may also be coupled to the new separate unit to allow — a motor or other actuator (e.g., a solenoid) to drive both the locking bolt and the latch, potentially enabling contactless opening of the door. This method is perhaps best understood with reference to Figure 3.

Figure 3A shows a mortise lock 300 installed in a mortise (cavity) formed in a door 302, viewed from the protected, e.g., interior, side of the door. The lock includes a lock case 304, — shown as a phantom drawing, in which the bolt 306 and a latch 308 are received. The bolt 306 is shown here retracted so that only the free end of the bolt is visible at the front end 310 of the lock, in which condition the door 302 can be opened from the closed position. The lock 300 is secured in the mortise by screws 312, 312°, which pass through the front end 310 into the material of the door. In the illustrated example, the lock is lockable and unlockable due to the key cylinder 314, here shown away from its hole 315 in the lock towards the interior side (the protected side) of the door.

The cylinder shown here is a Euro Cylinder, in which a key operates a pin and tumbler assembly (other cylinder types are of course used, and the invention can also be used with locks employing such other cylinder types, as well as with other locking and latching devices such as multi-point devices and other single-point locking devices). A keyhole 316 is provided at each end of the conventional cylinder, and by inserting a properly coded key into a keyhole 316 at either end of the cylinder, it is possible to rotate a rotating cam 318 which has an integral tongue 320 which, when installed in the lock, is effective to move the bolt 306 in and out of the lock case 304, thereby unlocking and locking the door 302 to an associated door frame 240 (not shown). The free end of the bolt — engages an opening or “holder” in the door frame 240 to lock the door in the closed position.

The cylinder 314 is secured in position in the lock by a retaining screw, not shown, which in use passes through an opening 322 in the front end of the lock. It should be understood that the lock is shown without the usual cover plate which would normally cover the front end 310 of the lock and the screws 322 and the front end of the opening 322. The latch 308 can be opened or closed by a

> DK 202400084 Y3 latch driver which engages a typical square opening 324 in the latch (or in a latch driver in the lock). The latch may be coupled to a handle or lever (one on each side of the door) by means of which the latch can be retracted. Latches are often spring-loaded so that they return to the position shown, where the latch tongue projects from the lock, when pressure on the handle or lever is released. Sometimes the latch is coupled to be driven by the cylinder so that additional rotation of the cylinder in one direction retracts the latch, and rotation in the opposite direction does the opposite — in which case no latch handle is provided, or a latch handle is provided only on the protected side of the door.

In Figure 3A, two Euro cylinders are shown: on the left, a conventional cylinder 314 with a nail hole 316 at each end; on the right, a new cylinder 314", where instead of the keyhole on the protected side of the door, an elongated tongue 326 projects. The tongue 326 will engage the mechanism of a device that includes the mechanism and electronics of an electronic lock. This device, which will be described with reference to Figures 3B and 3C, will be mounted in place of the key bracket 328 that encloses the hole 315 of the cylinder 314".

The modified cylinder 314' is also shown schematically as including a pressure sensor 278 coupled to a hole communicating with the keyhole on the exposed outer end of the cylinder. The pressure sensor is adapted to detect pressure changes following the insertion of a key (or the like) into the keyhole. Also shown schematically is a displacement sensor 279 which may be provided in addition to or in place of the pressure sensor 278 which is coupled to a displaceable member which will be displaced upon insertion of an elongated object (such as a key or dirk) into the keyhole. A microphone 276 may additionally or alternatively be provided at a suitable location on the cylinder 314' to detect sounds of interaction with the lock, e.g. sounds of key insertion, dirk insertion, bump key insertion etc.

In Figure 3B, a device including the mechanism and electronics of an electronic lock is contained within a handle, or more generally a handle, 330 which is secured to the door 302 by means of a securing plate 332 which has itself been mounted to the door in place of the key holder 328. The securing plate 332 is secured to the interior surface of the door by means of a pair of securing screws 334. The securing plate includes engagement members, here shown in the form of projections 336, which may be in the form of flanges which mate with and lock into engagement with corresponding formations in the handle 330. The securing plate 332 and the handle may be held locked together via one or more concealed retaining screws each of which is concealed in respective holes 338 through the (here) curved surface of a first portion 340 of the handle 330. In the example shown, the majority of the length of the handle forms a body 339, which here extends over and

1 DK 202400084 Y3 receives the first (or inner) part 340, although other arrangements are of course possible within the scope of the invention.

Also shown in Figure 3B is a set of one or more visual indicators 341, for example in the form of RGB LEDs, by means of which a status of the electronic lock and/or a mode or condition of the electronic lock can be indicated. Finally, Figure 3B shows one or more openings 342 formed in the main surface of the attachment plate 332. Each opening 342 provides a window through which the magnetic field of a magnet 270 can pass unhindered.

The magnet(s) 270, if used, may be surface mounted on the door in the window provided by an opening 342, or the mounting plate may be used as a template (or jig) to enable a recess or hole (e.g., drilled, machined, or cut) to be formed in the structure of the door so that a magnet may be inserted into the body of the door. If the recess or hole is deep enough, a magnet may be embedded in the door and concealed under a wooden or plastic covering, possibly prior to securing the mounting plate to the door. As noted, one or more magnets 270 may be mounted on the door in this manner — to cooperate with the magnetometer(s) 264 of the electronic lock, as described with reference to Figure 2. It is clear that the position of the aperture(s) 342 relative to the engagement members 336, and the position of the complementary features of the device relative to the accelerometer used to detect the magnetic field of magnet 270, are chosen such that when the device is mounted on the door, the magnetometer 264 and magnet 270 — (positioned by the mounting plate used as a jig) are aligned for optimum sensitivity.

Figure 3C schematically shows a longitudinal section through the device 330, the mounting plate 332, the lock case 304, the door 302 and the door frame 240. The device includes the handle 330 which is mounted over an inner portion 340. The inner portion 340 includes —> the engagement portions 350 which mate and lock into engagement with the engagement portions 336 of the mounting plate 332. As shown, approximately half of the length of the handle overlaps the inner portion 340, with the inner portion effectively providing an enclosure or box for the electronics of the smart lock, although it will be appreciated that many other possible configurations are possible within the scope of the invention. The tongue 326 of the cylinder 314' extends through an opening in a PCB 251 (or between a pair of PCBs) which carries the majority of the electronic components of the smart lock, which include the processor 258, memory 260, RF transceiver 262, magnetometers 264 and accelerometer 275. As illustrated here, two magnetometers may be used, one for tamper detection using a magnetic field from a door-mounted magnet 270, and the other 264' for detecting door open.

> DK 202400084 Y3 status (and thus potentially an intrusion) by using a magnetic field from a frame-mounted magnet 266, although the same functionality can be provided by using just a single magnetometer. To the processor, but possibly mounted away from

The PCB 251, the speaker/sounder 263 and the indicators 341 are connected.

The device includes a battery power supply 252, which includes a charging device 254, optionally together with a charging connector 256 (preferably on what in use would be the underside of the handle, so as to be hidden under normal circumstances).

In the example shown in Figure 3C, the locking mechanism includes an electrically controlled actuator 246 (e.g., a motor or solenoid) and optionally an associated mechanism 352, which together are — configured to apply a torque between the inner portion 340 of the handle and the tongue 326, to rotate the cylinder 314' of the lock — to retract the bolt 244 into the lock housing 304 or to extend the bolt 244 from the lock housing 304. A force transmission device 354 may be provided between the cam tongue 320 and the bolt. Similarly, the lock may be configured such that rotation (e.g., over-rotation) of the cam 318 — also drives the latch 308.

Although Figure 3 shows devices suitable for a conventional door, those skilled in the art will appreciate that the same device and modified cylinder can equally be used with a door (or French window, patio door, etc. or the like) with a multi-point locking system (which typically has a minimum of 3 locking points, all of which lock simultaneously when the mechanism is engaged).

These typically operate in conjunction with a cylinder lock (e.g. a Euro cylinder lock), although locking typically involves manually activating a door handle to engage the various locking points with their counterparts before the lock will lock. Although this engagement phase may require the application of too much torque for sensible use of the described electronic locking add-on, (if for no other reason than due to short battery life), the locking and unlocking steps, distinct from the engagement/disengagement actions, could potentially be performed using the system just described.

In an alternative arrangement, the smart lock is provided with a coupled handle or grip on each side of the door, the mechanism of the smart lock being configured to control the coupling device to, in the unlocked condition, allow the bolt to be retracted, and in the locked position to disengage the handles/grips from the bolt so that they are unable to retract the bolt to unlock the door. It also goes without saying that rather than providing the additional device in the form of a handle, a larger body which is not in the form of a handle or other handle may be used (although potentially including a handle or other handle for

’s DK 202400084 Y3 to enable the door to be pulled open), to accommodate the added mechanism — and such a device may be configured to include a powered latch actuator.

The smart lock in Figure 3 and developments thereof as discussed with reference to Figures 4-7 may be designed and configured as an EN certifiable shock sensor for use with —>=EN certified building security systems (intruder alarm systems) classified in

Class 2 or higher (EN50131), which includes the necessary tamper detection function, which will trigger a tamper warning if an attempt is made to open the device, remove its battery, or remove the lock from the door on which it is mounted.

Figure 4 schematically illustrates the operation of devices according to aspects of the invention. In particular, the figure illustrates detecting the presence of an anomaly in signals generated by one or more shock or vibration sensors, such as one included in a smart lock as just described. An event is considered to be an anomaly if the relevant sensor data is outside the usual range for "normal" use.

In operation, a device receives an input at 402 from a shock or vibration sensor, which may be a shock or vibration sensor incorporated into a smart lock, for example based on an accelerometer or a magnetometer. The received input is then subjected to an anomaly detection process 404.

The anomaly detection process 404 may involve both anomaly detection 406 and classification 410 (possibly as a two-step process), or may involve > anomaly detection 406 but no classification 410 (a one-step process). In the latter (one-step) case, if an anomaly is detected in the input 402, an output may be provided at 408. If the operation includes classification 410, this may be as an additional step in a multi-step process — for example, as the second step after a first step of anomaly detection, in a two-step process, or it may be performed as part of a > processing step that involves both anomaly detection and classification. A result of classification may be provided as an output 412.

Using a two-step process, where a separate anomaly detection step 406 is performed prior to an (optional) classification step, rather than attempting to classify the input 402 directly, has the advantage that an input 402 that represents a — deviation compared to “usual” behavior can be detected as a potentially suspicious event, even if it is of a type that has not previously been encountered or modeled. So new types of attacks can be detected, even if they are not specifically modeled.

They may be difficult to classify afterwards, but at least the system can detect

> DK 202400084 Y3 the occurrence of something unusual. This could future-proof the system, making it more robust than single-step classification.

The anomaly detection process 404 may be based solely on output provided by a shock or vibration detection device of the smart lock, such that the output 412 is derived solely from the output provided by the shock or vibration detection device of the smart lock. Alternatively, the output 412 may be the result of processing the input 402 provided by the shock or vibration detection device of the smart lock together with one or more optional additional inputs 414. Such optional additional inputs 414 may be provided by one or more sensors optionally external to the smart lock but — coupled to the door or door frame, optionally at one or more locations remote from the smart lock.

For example, a sensor, such as a magnetic open/close detector, which is adapted to detect the open/close state of the door on which the smart lock is mounted. Similarly, a sensor which is capable of detecting the open/close state of the door by detecting rotation of the door relative to an axis (for example, which is capable of encoding — the states closed, opening, open, e.g. as -1/0/1), or which is capable of detecting rotation speed (normalized or not) or angular orientation change (e.g., by using an accelerometer or a magnetometer or the like).

An additional input 414 may also be provided by another sensor in or associated with the smart lock, for example, a microphone in or coupled to the lock, for example coupled to a cylinder of the lock, an inductive or capacitive sensor coupled to the lock or cylinder to detect human interaction with the lock (e.g., touch or close contact), or an insertion detector (such as a microswitch) to detect insertion of a key or dirk into the keyhole of the smart lock.

Anomaly detection 404 or 406 may be performed using a — machine learning algorithm.

The machine learning algorithm may be trained with training data that includes several examples (e.g., at least 1000, at least 5000, at least 10,000, or more) of sensor output data for “normal” events (such as key insertion, door opening, door closing, door unlocking, door knocking, door slamming, etc.) and optionally for each class of “violent” or “attack” behavior (e.g., lock picking, lock bumping, lock drilling, lock or cylinder snapping, pulling or cylinder removal, attempted burglary). The training data is labeled according to its “violent” class. Training data for “normal” events may be labeled to indicate “normality,” or they may be unlabeled, and thus indistinguishable from data representing “violent” events. The training data preferably includes field-

DK 202400084 Y3 25 collected data rather than just "laboratory-collected" data, although it may be necessary to obtain a lot of "violent" data from attacks performed on test equipment set up in laboratories or other environments that simulate the real world. The training data set does not need to be a balanced data set, i.e. it is not necessary that there are the same number of data examples for normal events as for "violent" cases.

For two-stage anomaly detection, the autoencoder may be trained only with, and/or by prioritizing, normal data (where “attack” data is omitted and/or penalized). When the signal is regenerated by the autoencoder, anomaly detection depends on how well the autoencoder was able to “fit” the original input signal to a “normal” state, and regenerate it. A large error (measured in energy) means that the autoencoder did a poor job of fitting it to any of the normal states, so this suggests an anomaly compared to the “normal” training data. This generates the detection.

But the “violent” (or “attack”) data can be used to train a classifier, which is then able to “indicate” or categorize a violent event — for example, an — event as a lock-digging or lock-bumping attack.

The anomaly detection process can be performed using unsupervised learning with GMM (gaussian mixture models) or alternatively by using an autoencoder. Autoencoders are models based on neural networks that are trained via unsupervised learning to efficiently compress input data to form an encoded representation and then to reconstruct the compressed representation to produce an output that is as close as possible to the original input. An autoencoder can be considered to be a multilayer perceptron (MLP).

Classification can be performed using any suitable classifier, such as a convolutional recurrent neural network (CRNN) architecture, which can be based on combining a convolutional neural network (CNN) architecture with a gated recurrent unit (GRU) architecture.

Figure 4B schematically illustrates the principle of a processor 450 that makes a determination based both on data received from a sensor 452, e.g., a shock or vibration sensor such as an accelerometer or a magnetometer, of a smart lock such as 106, and — from an additional sensor 454 that is not part of the smart lock. The processor may also be part of the smart lock. The additional sensor 454 may, for example, be a door contact sensor that detects the state (open, closed, opened or closed, degree of rotation, etc.) of the door to which the smart lock 106 is attached, e.g., a door contact sensor based on magnetic detection, such as a magnetometer, a Hall sensor, or other magnetically switched device. Alternatively, the additional sensor 454 may be an accelerometer (one, two or three axes) which is not contained in or part of the smart lock, the additional sensor being coupled to (e.g. mounted on or in) the door or on the door frame in which the door is mounted. Alternatively, the additional sensor 454 may be an inductive or capacitive sensor, a pressure sensor, a motion or presence sensor, a microphone, a pressure pad, a radar arrangement, an RF-based presence detection system, a camera (video, e.g. the camera in a video doorbell).

The processor 450 may implement a classifier and/or an anomaly detection algorithm, optionally based on an AI method. The processor 450 may be configured to process sensor data, from sensors 452 and 454, to identify the presence of — anomalies (e.g., deviations) that may indicate an anomalous event — and in response thereto, the processor may provide an output (e.g., sending a signal to an associated security or monitoring system) to alert the existence of an anomalous event.

Thus, a smart lock is provided which includes an accelerometer and/or a magnetometer, plus a device, separate from the lock, mounted on the door or door frame, which itself includes an additional sensor, such as an accelerometer and/or a magnetometer. Such an arrangement may make it easier to perceive the source/location of a shock or vibration, (e.g. whether the lock, the door close to or distant from the lock, etc.). The provision of differentiated information could be useful in making it easier to locate the source of a — disturbance/vibration/shock.

The smart lock (or door sensor) could include a processor to handle the calculations based on data from both sources, but data could also be sent, for example, to a control unit in a security monitoring system so that the control unit can perform the necessary calculations.

A shock or vibration sensor may be provided with a processor configured to process sensor data to determine the presence of anomalies in the data, and if an anomaly is detected/determined, the processor may be configured to generate an output. In an optional second processor, the processor may be arranged to analyze the sensor data (or the processed sensor data) to perform classification to allocate a detected event to one of a plurality of event types, the event types including at least one threat class and at least one non-threat class.

Figure 5A schematically illustrates the elements that make up an autoencoder 500.

Essentially, the autoencoder comprises an encoder part 502, whose job is to compress input data to form an encoded representation 504, and a decoder part whose job is to transform the encoded

>7 DK 202400084 Y3 representation 504 to reproduce, as closely as possible, the original input data. Input data comes to the encoder input 508, and the reconstructed output is provided at the decoder output 510. Figure 5B illustrates the layered structure of the autoencoder 500. In this example, the autoencoder comprises 6 fully connected layers. The encoder side applies dimensionality reduction by using fewer nodes for each layer between the input and the encoded portion, for example from 32 nodes at layer X0 to 16 nodes at XI to 8 nodes at X2. The layers of the decoder portion 506 similarly utilize an increasing number of nodes as we move from the encoded representation 504 (or Z), with the number of nodes per layer increasing from 8 to 16 to 32. It should be understood that this structure is only an example, and that embodiments of the invention may utilize encoders where the number of layers, and the number of nodes in each layer of the encoder and decoder portions, differ from those set forth as this example.

Figure 6 illustrates schematically how a trained autoencoder can distinguish between “routine” events — here designated as “in-grouping”, and anomaly events (which may represent attacks on the smart lock or the door on which it is mounted), here designated as “out-grouping”. An appropriate threshold t can be chosen, for example based on calibration tests, to mark the distinction between “in-grouping” and “out-grouping” events. Exceeding the threshold can be used to trigger the reporting of an event (more simply the provision of an output signal) or the triggering of an alarm or the like.

Figure 7 schematically illustrates how an autoencoder may be integrated into a smart lock 700 according to one aspect of the invention. The autoencoder 702, which may correspond to the autoencoder 500, may run on a processor 704, which may correspond to the processor 258. The processor may be a microprocessor but may instead be an MCU such as a SiliconLabs EFR32xG4 Cortex-

M4, which incorporates an ARM Cortex-M4 core. A memory 706, which may correspond to memory 260, stores program instructions for controlling the processor 704, and may particularly store code for implementing the autoencoder 702 and the classifier 708. The smart lock 700 further includes an RF transceiver 710, which may correspond to transceiver 262, a power supply 712, corresponding to power supply 250, and a lock actuator 714, which may correspond to actuator 246, all of which are operatively connected to the processor 704.

Also coupled to a processor is a first sensor 716, which may be the accelerometer previously referred to as element 275. Optionally, as described above, the smart lock 700 may include at least one second sensor 718, which may be in the form of a pressure sensor 278, an insertion detection device 279, or a microphone 276.

9 DK 202400084 Y3

Outside the lock 700, one or more other sensors 720, 722 may be provided and configured to communicate (wirelessly or via a wired connection) with the processor 704. The other sensor may, as already described, be a door contact sensor (possibly based on the use of magnetism) that detects the state (open, closed, opening) of the door to which the lock 700 is coupled, or it may be another accelerometer coupled to the door or door frame in which the door hangs, or some other sensor capable of providing data relevant to determining the state of the door and/or the occurrence of an attack or other event, or some combination thereof. Typically, the external sensor(s) 720/722 may include an RF transceiver for communicating with

The RF transceiver 710 of the lock, although the external sensor may be coupled to the processor 704 in another manner (e.g., through a coupled connection).

In one embodiment, the encoder 702 can be run as a model using, for example, Tensorflow Lite. The models (autoencoder and classifier, if present) can be quantized to int8 from the float32 representation of the weights used during training. Data can be — received from each sensor (e.g., 716, 718, 720, 722, or any combination thereof) in 128-sample bursts, which then provide the input of the encoder program.

Each 128-sample packet is then divided into 4 subgroups of 32 samples each, and the group of 4 subgroups is then loaded into the encoder model. A sample size of 128 was chosen during initial studies, which found it to correspond to the most appropriate duration for a 200Hz sampling rate used by the lock's 714/275 accelerometer. 128 samples gives a duration of 640 ms, while 32 samples gives a duration of 160 ms. In some embodiments of the invention, these samples, and their groups, do not overlap, and no sliding window is used.

Tanh activation can be used for the neural network of the autoencoder, but in other —> implementations, Relu or leakyRelu can be used instead of Tanh activation.

In some embodiments, it is possible to form a satisfactory neural network with a network with a total weight count (sum of weight counts for all layers in the network) of 7,624 parameters. This may correspond to a file size, including weights and states, of approximately 17kB. Such a small model can easily run on an MCU, despite the memory limitations typical of MCU implementations. When the sensor is a three-axis accelerometer, the input to the autoencoder 702 may be a vector of samples each with a dimension of 3 in size, representing x, y, z coordinates. The sample format may be int8, with a length selected from 16, 32, 64, or 128, for example 32. The autoencoder output may be a single number, which is the detection result, with a length of 1, and with an int8

9 DK 202400084 Y3 format (with values from 0-127) mapped to values from 0 to 100 (i.e. normalized to values within this range). The main mechanism of the model may be to calculate expected values (for normal behavior). Expected inputs should result in low values, for example close to zero with a maximum in practice of between, for example, 10 and 30 (to give some margin of error). The “out of grouping” events should give significantly higher values. An alarm threshold can therefore be set somewhere in the range from 60 to 90 (within a total range of 0-100). The threshold value can be set based on analysis of the model’s output values once the network has been trained with training data that includes everyday and out of grouping events.

If a classifier 708 is used to identify out-of-group events, e.g. to denote an irregular event such as lock-digging, lock-bumping, etc., an appropriate model such as one with a CRNN architecture can be selected to run on processor 704. The classifier can, if the sensor is a 3-axis accelerometer, receive an input vector of samples each with a dimension 3 in size, representing x-, y-, z- coordinates. As with the autoencoder, the sample format can be int8, with a length selected from the set 16, 32, 64, and 128, e.g. 32. The output of the classifier can be a vector of classification result for labels/classes, optionally with a format of uint8 (values 0-100) with a length of 12.

The classification model can be based on supervised learning using deep learning, provided that a well-prepared dataset is available for training the model. With a smaller dataset, it can help to use cross-validation, stratified k-fold (k=5) to prevent overfitting to specific data and to avoid leading the model towards wrong decisions. Such an approach can still be useful even with larger datasets (e.g. tens of thousands or more events). It can also be useful to train the classification model in advance with all the data to couple iteration and later have iteration with a properly balanced dataset. As an alternative, one can use SVM with some features extracted from the signal.

That would be a more classic approach. It has some limitations, primarily the need to find and select the best features.

However, the use of unsupervised learning, with the autoencoder, is likely to yield better results than supervised learning if there is a lack of real-world and labeled data — which will generally be the case for at least some of the attack scenarios. Unsupervised learning is aimed at selecting outliers, and this means that it should still be possible to provide an appropriate output when previously unseen

20 DK 202400084 Y3 data is pushed to the input of the anomaly detection stage — so that even new attack types can be expected to trigger an alarm (or warning) notification.

If signals from a second sensor (e.g., exterior sensor 720 or 722) are used in addition to those from the lock's first sensor, the signals from the two (or more) sensors can be combined to provide a long vector (making the input wider). For example, if the second sensor is a door state sensor, the longer vector could be in the form of [lock accelerometer |state] rotation].

Alternatively, the state/rotation data could be represented as positional encoding (numbers modulated as sin/cos, and added after the first layer) as described in the article available at the following URL: https://medium.com/(Øhunter-j-phillips/positional-encoding-7a93db4109e6

These signals could then be used as an input to the anomaly detection model and/or to the classification model, if used.

Alternatively, a decision maker could be implemented in the firmware for the detection model — that it only triggers an alarm event when the door on which the smart lock is mounted is closed and no lock rotation occurs.

It may be possible to improve the overall classification results by using a different sensor type in addition to using data from the accelerometer (or magnetometer) in the smart lock, because the more data points in general, the better the overall m— classification.

In practice, there is always a way to normalize data from different sensors, and to fit them together as well. In general, adding more information can help improve the accuracy of the classification, e.g. from 3 to 30% relative. However, in some specific use cases, the accuracy of the classification can be significantly improved, for example, hand on handle or hand on lock detection (using capacitive sensing, for example), and key in lock detection (using impedance sensing) to help classify specific attack methods such as lock picking, lock snapping, and lock bumping.

As previously noted, when using multiple sensors, all sensor data can be combined into a single input, concatenated to form a vector as a long array. However, it is also possible to consider certain sensor outputs separately.

When using a microphone as a second sensor (either in the lock or as an external sensor), it would make sense to perform another form of feature extraction, possibly pre-filtering, and possibly using FFT, possibly using MFCC (medium frequency cepstral coefficient) or CQCC (Constant Q Cepstral Coefficients) to reduce the range of data, and highlight only some frequencies (very low < 50Hz, probably in the range from 2-10Hz).

The smart lock 106 can be used as part of a security monitoring system, for example one classified as Class 2 or higher according to EN50131, although it can also be used as a separate component that can communicate with one or more user devices and/or a network (e.g. cloud) based backend (possibly via a local

RF connection to a broadband router and thus via a wired and/or wireless connection, or via an LPWAN (e.g. via NB-IOT or LTE-M). In one example, such a security surveillance system may be controlled by a video doorbell controller, such as that shown as 100 in Figure 1, which will now be described.

Figure 8 schematically illustrates the main components of an example of such a video doorbell 100 suitable for use in embodiments of the invention in conjunction with a smart lock as described with reference to Figures 3 to 7. A video doorbell 100 — includes a processor 800, which may be a microcontroller (MCU), with an associated memory 802 that stores program instructions (“software”) that, when executed on the processor 400, control the operation of the video doorbell 100. A pair of transceivers 804, 806 may be provided. A first of the transceivers 804 may support Wi-Fi communication with a

Wi-Fi access point (AP), such as Wi-Fi router 112 in Figure 1, for the communication of — video signals from video camera 808 and audio signals from microphone 810. The second transceiver 806 may support a lower bandwidth channel for the transmission of control signals to, and event notifications from, peripheral alarm devices such as smart lock 106, external video cameras, door/window switches (which detect the state — open or closed, of windows or doors), etc. Typically, a low bandwidth channel or channels may be — provided, which utilizes an appropriately allocated frequency or frequencies in the ISM band (for industry, science, and medicine — such as (in Europe) 868MHz. Preferably, communication on the low bandwidth channel(s) is encrypted. The primary reason for using a low bandwidth transmission channel is to enable the use of transceivers (in peripheral alarm devices) that have low power consumption (and in particular lower power consumption than — experience shows power-hungry Wi-Fi transceivers), so that the battery-powered peripheral alarm devices can achieve the required minimum battery life of 3-5 years. The transceivers 804 and 806 are connected to an antenna arrangement 812, which typically includes (multiple) separate antenna elements for the two transceivers 804 and 806. One or both of the transceivers 804, 806, may support other transmission bands/protocols, such as BlueTooth (RTM) or BLE for

2 DK 202400084 Y3 example. The video doorbell 100 can also support communication via an LPWAN (e.g. via NB-IOT or LTE-M), which may include device-to-device communication (D2D).

The video doorbell 100 also includes a ring trigger device 814, which may comprise a mechanical switch or "bell push", and preferably one that provides tactile feedback upon activation. In known manner, the ring trigger device 814 may be co-located with a lens of the video camera 808, but may equally be provided separately so that the video camera lens need not move when the ring trigger device 814 is activated.

The video doorbell 100 also preferably includes (or may be connected to) a keypad 816 for entering numbers (such as codes or PINs). The keypad 816 is preferably a mechanical device that provides tactile feedback upon activation.

The video doorbell 100 also includes a motion detection device 817, which can be used to trigger the video camera to record motion-triggered video recording and marking (notification) of motion — although the video camera can be arranged to always — be on to provide continuous monitoring. The motion detection device can include one or more passive infrared sensors (e.g., PIR or T-MOS sensors).

Additionally or alternatively, the motion detection device may use radar and/or ultrasound to determine the distance of an object (e.g., a person) from the video doorbell, and this may be combined with passive infrared presence detection and/or other forms of presence detection. The video doorbell may have an associated app or online tool to allow a user to set activity zones, intervals, configure sensitivities, etc., so that notifications regarding detected motion or presence are not, for example, transmitted regarding motion detected in public areas (e.g., of people walking on a sidewalk outside the building, or vehicles driving past along a road outside the protected building).

The video doorbell 100 may also include lighting 818 (in the form of one or more light sources), particularly infrared lighting, but optionally also lighting to provide illumination with visible light.

The video doorbell 100 may also include an optical display 819 for displaying messages and user feedback. While it would be possible to include the functionality of the keypad 816 within the display 819, using a touch-sensitive display, it is preferable to keep these functions separate and to use a separate mechanical keypad 816, both for reasons of durability and power consumption.

The video doorbell 100 may also include an audio output device such as a speaker.

23 DK 202400084 Y3 819 to reproduce speech from a user at user device 108, for messages from processor 800, and optionally to function as an alarm sounder or siren in the event of an alarm event — as will be explained later. The provision of at least one audio output device (two or more may be provided) together with at least one microphone (two or more may be provided) enables two-way audio between a person at the video doorbell and another person, for example a designated user via a user device (e.g., smartphone with a video doorbell app or a security monitoring system app).

The presence of one or more microphones 810 also enables the use of voiceprint identification as a form of access control — so that, for example, a — resident of the building, or a trusted guest (such as a nurse, caregiver, cleaning lady, etc.) who has stored a voiceprint in the system, can gain access to the house by speaking to the video doorbell, which compares the given voice sample with known stored voice samples to identify the speaker. If the identified speaker is on an approved list of people, the video doorbell can send an instruction to a smart lock on the relevant — access door to allow the identified speaker to open the door to gain access to the protected building.

Optionally, the video doorbell 100 also includes a near-field communication (NFC) antenna 820 to enable near-field communication with an NFC device such as a dongle, smartwatch, NFC tag, or suitably equipped mobile phone—for example, to authenticate a user and possibly unlock an associated smart lock 106 to gain access to the protected building.

The video doorbell 100 is provided with an outer housing or enclosure 821 to protect the components of the doorbell. The housing 821 is preferably provided with one or more tamper detection arrangements 813 which operate to generate a tamper warning if the housing is removed from the video doorbell 100. Such tamper warning is preferably processed by the controller of the security monitoring system (where the controller may be provided either by the video doorbell itself or by a separate device), regardless of the connected state of the security monitoring system.

Users (or the owner(s)) of the security monitoring system may be equipped with an app by which the security monitoring system controller may be warned in advance of a scheduled event that would otherwise be treated as an alarm event, so that the security monitoring system controller may be configured to actually ignore the scheduled event when it occurs. For example, it may be necessary to remove the housing 821 of the video doorbell to replace a defective battery 826. In

21 DK 202400084 Y3 using the app, a user or service technician can prevent the security monitoring system controller from treating the removal of the housing as a tamper event, which should be reported as an alarm event. The same principle applies to the lock 106 and its tamper detection.

The video doorbell 100 is preferably provided with a mains power supply 822 connected to a power supply 824, which includes a rechargeable battery 826 (as a backup power supply for use in the event of a power failure) and a charging arrangement 828 for handling charging of the rechargeable battery 826. As shown, the video doorbell 100 preferably also includes an interface 830 (such as a micro-USB port, a USB-C port or a — “Lightning” connector or the like) for receiving a low voltage (e.g., SV or 20V) supply (e.g., from a power bank or external charger), which, in the absence of a mains supply (or in the event of a mains supply failure) can be used to recharge the rechargeable battery 826 on site, rather than the battery having to be removed for recharging, so that the video doorbell can always be powered — which is an important factor when the video doorbell — also functions as the controller of a security monitoring system. Alternatively, the video doorbell can be configured to receive wireless (e.g. inductive) charging from an external power source.

The function of the video doorbell as a controller of a security surveillance system will now be considered in the context of Figure 9.

Figure 9 schematically illustrates an embodiment of a home security monitoring (home alarm) system 900 built around a video doorbell 100 that functions as the controller of the system 900. In the illustrated example, the system 900 includes a smart lock 106 as described with reference to Figures 2 through 7, such that the system user (e.g., the homeowner) is able to remotely unlock the door 102 to which the lock 106 is attached to — admit a guest, where the guest may first be screened using the video camera in the video doorbell 100.

The system 900 includes a peripheral alarm device such as a door switch 902, for example on the back door 904 of the house (and optionally also on the front door to which the lock 106 is attached), which may be a magnetic detection switch (such as a reed relay) or a magnetometer or a Hall effect sensor that detects whether the door 904 is open or closed, or a shock or vibration sensor based on an accelerometer, and which transmits an alert that is received by the second transceiver 806 (e.g. 868MHz transceiver) of the video doorbell 100 and processed by the processor 800 to generate, if the system is connected, an alarm event signal that is addressed to a remote (e.g. cloud-based)

25 DK 202400084 Y3 alarm reporting service 110 and transmitted by the first (e.g. Wi-Fi) transceiver 804, (the speaker in the video doorbell may also be immediately activated as a siren or to provide some form of other sound deterrent). The transmitted alarm event signal may be received by the Wi-Fi access point (AP) 112 in the building and sent via a => broadband connection to the Internet 115 and thus to the cloud-based alarm reporting service 110 (or may be transmitted via LTE-M, or NBIOT or the like). If a door sensor 902 is provided for the door to which the lock 106 is attached, the sensor 902, which may be accelerometer-based or responsive to a magnetic field, may act as (one of) the second sensor 454 and/or be the source of the optional additional input 414. An alarm event signal may be transmitted if the smart lock 106 determines the existence of an anomalous event (as described with reference to any of FIGS. 4 to 7 ), the smart lock may optionally determine the existence of an anomalous event even if the door sensor 902 does not itself detect a stimulus that meets its own alarm event threshold: the alarm event may be signaled by the smart lock to the video doorbell for — onward transmission, (although transmission from the smart lock 106 directly to the alarm reporting service 110 rather than via the video doorbell is also contemplated).

At the cloud-based alarm reporting service 110, the identity of the security monitoring system (such as the video doorbell 100 or smart lock 106) included in the alarm event signal can be checked against a database of registered system identities to retrieve an address (e.g., a SIP address, a mobile phone number, an email address) corresponding to the registered alarm system identity. The alarm reporting service 110 can then send a notification to the registered address, such that a user, for example, is presented with a notification (e.g., a native notification) on their smartphone 108 or other personal electronic device, and/or an automated phone call providing details of an alarm event — e.g., the identity/type/location of the sensor (e.g., back door contact sensor, or smart lock 106) that has been triggered. A user receiving such a notification may be able to request streaming of video or transmission of images from cameras 906, 906’ installed in or at the building (e.g., house) of the security surveillance system installation: for example, a push notification may include one or more “action buttons” that, upon activation by the user, cause the user’s device 108 to transmit an instruction to the alarm reporting service 110, in order for the alarm reporting service 110 to send an instruction to the controller of the security surveillance system 900 (i.e., the processor of the video doorbell) to transmit activation signals to wake up the cameras 906, 906’, causing them to record and

26 DK 202400084 Y3 send images and/or video. The controller of the security monitoring system 900 (i.e. the processor of the video doorbell) may receive these video transmissions using the first (Wi-Fi) transceiver 804, and the same transceiver may be used to further transmit the images/video to the alarm reporting service 110 using the > access point 112 and the Internet 115. The alarm reporting service 110 may further transmit the images/video to the user's device 108. The alarm reporting service 110 may be configured to store, (if the user has paid the required subscription), images/video recorded in this manner.

Each of the video cameras 906, 906° may in fact be an available (“off the shelf”) — video camera (as a way to keep costs down) that includes only one transceiver, and that transceiver may support Wi-Fi but not a secondary, low bandwidth channel. In such a case, the video doorbell in its form of a system controller of the security monitoring station must communicate with, and receive communications from, the cameras solely by means of Wi-Fi — which uses the first transceiver 804. — Preferably, however, the video cameras 906, 906° support both Wi-Fi and a relevant second channel for communication with the video doorbell’s second transceiver 806 — and this functionality is preferably provided by the inclusion of a second transceiver in each of the cameras 906, 906’. Optionally, the video cameras each support an LPWAN technology such as LTE-M and/or NB-IoT.

The user's device may also be programmed with a complementary app that allows the user to send instructions to the video doorbell (which in this embodiment acts as the controller of the security monitoring system) to cause activation of selected video cameras, so that the user can "check in" to the home while away from home, and also to arm/disarm the alarm system. Images and video from the cameras — 906, 906’ can be transmitted over Wi-Fi between the camera 906, 906” and the video doorbell 100 — and thus these images and video will be received by the first transceiver 804, which is also used for further transmission to the access point 112. However, optionally, in some embodiments, the system can also be configured to send lower resolution images, and/or lower frame rate video using the low bandwidth channel — (e.g., 868MHz) received by the second transceiver, for use as a backup if the wider bandwidth channel (e.g., Wi-Fi) is interrupted, with the processor of the video doorbell configured to forward the first image/video that arrives, and then subsequently to forward the corresponding higher resolution/higher frame rate image/video if and when it arrives. Similarly,

27 DK 202400084 Y3 The cameras may be arranged to use an LPWAN such as LTE-M and/or NB-I0T for the transmission of video images and possibly of duplicate images with different resolutions and/or frame rates as just described.

Likewise, notifications are also preferably provided to the alarm reporting service — 110 (and further to the user device 108) regarding alerts received from the smart lock 106 (such as upon determining the existence of an irregular event) or other peripheral alarm devices such as motion-triggered cameras 906, 906°, and a window contact sensor, (which may also or instead function as a window impact or vibration sensor) 908, which is associated with a window 810. Similarly, an — irregularity report or a tamper alert may be received from the door lock 106 as a result of, for example, an attempt to bump the lock, to break the lock 106 away from the door 102, or as a result of a magnetometer of the lock detecting opening of the door when the lock has not been unlocked, or the detection of an irregular event (such as an attempted lock picking or other attack — such as the detection of a drilling attack - on the lock) are detected - and reported by the lock processor based on signals from the accelerometer 275 of the smart lock (e.g. as described with reference to any of Figures 4 to 7), and may be handled by the processor 800 of the video doorbell (in this embodiment) and thus reported to the user via the alarm reporting service 110. Event notifications from the peripheral alarm device may generally be provided over a low bandwidth channel, for reception by the second transceiver 806 of the video doorbell, since most of the simple peripheral devices (e.g. door contacts, motion sensors, shock or vibration sensors) do not require a high bandwidth channel (e.g. do not require something like Wi-Fi) but must be inexpensive and have very long battery life - so that they will generally each contain only a single transceiver, and are configured to communicate with the second transceiver 806 (both LTEM, or NB-IoT also supports low power consumption, especially at low data rates, and support long range communication, making them attractive for this role as well). As previously mentioned, peripheral alarm devices in the form of video cameras can benefit from the provision of a high bandwidth channel, such as Wi-Fi, because timely transmission of good quality video images (in color, with high resolution and an acceptably high frame rate) generally requires a fairly high bandwidth, and peripheral video camera devices will almost always include a Wi-Fi transceiver and will consequently almost always be primarily powered by the mains (preferably with battery backup). While the Wi-Fi transceiver could be used for the transmission of event notification signals — to be received by the video doorbell using the first transceiver 804, it is preferred (if

29 DK 202400084 Y3 budget permitting), that video cameras are provided which can report events using a low bandwidth channel (for reception by the second transceiver 806), and which preferably include an additional transceiver to allow this, (although a single, multi-function transceiver could of course be used instead). By equipping the peripheral — video camera units with both narrowband and wideband communication capability, we make it easier and more convenient to upgrade a system such as that of Figure 9 to the enhanced system of Figure 10, as will be described below.

The video doorbell’s ability to detect presence (for example, by using one or more of a PIR sensor, a TMOS sensor, Wi-Fi detection, radar, or using the — camera) allows the video doorbell itself to act as a peripheral alarm device of the system. When connected to the system, the video doorbell can in this way communicate an alert, just as it does when another peripheral alarm device is triggered. While this may only be appropriate in certain circumstances (particularly circumstances dictated by the likelihood of false alarms caused by “safe” actors), it is a potentially valuable option.

The processor 800 of the video doorbell 100 in its form as a controller of the security monitoring system may also be configured to check the status of the peripheral devices, including the quality of the communication channel, the status of the peripheral battery, etc.

It is contemplated that a video doorbell with the functionality of the security surveillance > system controller, in addition to being retrofitted to existing homes, will also be installed in newly constructed homes, possibly in conjunction with the smart lock 106 as described with reference to any one of Figures 2 to 7, together with at least one peripheral alarm device, possibly in the form of a motion detection video camera, such as 906, 906", and preferably another peripheral alarm device in the form of a door switch to detect the state — (open or closed) of the front door (equipped with the smart lock 106), the back door (or another secondary access door to the building). Such a system could be installed at relatively low cost, and the smart lock 106 and its door magnet(s) 270 (and optionally frame magnet(s) 266) could, for example, be factory installed — or at least the door could be factory prepared for on-site installation of the smart lock 106 and door (and frame) magnet(s) — 266 and 270. The first occupant of the home would then benefit from the enhanced security that comes from a professionally installed alarm, albeit one with a subset of all the possible features of such a system — and notably lacking a professional monitoring service. The design and configuration of the video doorbell, and the peripheral alarm devices — which include the smart lock 106, are such that once a

29 DK 202400084 Y3 security monitoring system based on their application, such as the one illustrated schematically in Figure 9, has been installed, upgrading the system to include professional monitoring by means of a remote ARC (alarm receiving centre) can be achieved quite easily. Such an upgrade may of course also be attractive to a resident who has already installed a system such as that shown in Figure 9. Such an upgrade will now be described with reference to Figure 10.

Figure 10 is generally similar to Figure 9 in that it is based on the assumption that a system such as that illustrated in Figure 9 has been upgraded to include the benefits of professional monitoring using a remote alarm receiving center 1002—although—it is also contemplated that security monitoring systems such as that shown in Figure 10 may be installed from the outset. The security monitoring system 1000 differs from the system 900 of Figure 9 in that it includes a separate alarm controller 1004 that is configured to control the security monitoring system 1000, rather than relying on the video doorbell to act as the controller. The video doorbell 1006 of the system 1000 may be, and preferably is, the video doorbell 100 that controlled the system 900, but with its processor reconfigured (e.g., through appropriate reprogramming or system updates) to function (at least in part) as a peripheral device that reports to the alarm controller 1004 rather than continuing to function as the controller of the system 900.

The present disclosure thus contemplates, as an option, an initial installation comprising a video doorbell as previously described, acting as the controller of a security monitoring system, which preferably includes a lock, such as the lock previously described with reference to any one of Figures 2 to 7, and at least one other installed peripheral alarm device comprising either a door/window switch for detecting the status of a door or window, a door and/or window impact or vibration sensor for detecting physical attacks on the relevant door or window, or a separate motion-triggered video camera, the peripheral alarm device including a transceiver for communicating with the first transceiver of the video doorbell. And the present disclosure further contemplates a mode of operation involving upgrading such an initial installation by introducing a separate alarm controller 1004 configured to control the security monitoring system 1000 and reconfiguring the video doorbell such that its processor is reconfigured to function (at least in part) as a peripheral device reporting to the alarm controller 1004 rather than continuing to function as the controller of the system 900. Once the system is upgraded, the video doorbell can then function as the system's control interface such that a user (e.g., a resident) can send

10 DK 202400084 Y3 arm/disarm and lock/unlock commands from it, which are received and followed by the separate alarm control unit 1004, which functions as the system controller.

The alarm controller 1004, like the video doorbell described with reference to Figure 8, includes a processor 1010, which may be a microcontroller, with an associated memory 1012 that stores program instructions (“software”) that, when executed on the processor 1010, control the operation of the alarm controller 1004 and of the security monitoring system 1000. At least one pair of transceivers 1014, 1016 may be provided. A first of the transceivers 1014 may support Wi-Fi communication, the alarm controller being configured to function as an access point for a Wi-Fi network (with a first SSID — which is dedicated to the security monitoring system), which may operate in parallel with a second

Wi-Fi network (with a different SSID), which can support Wi-Fi needs in a household such as supporting video streaming services (e.g. Netflix, Amazon Prime, RTM's), homework, etc.. The alarm controller 1004 is configured to receive video signals from video cameras 906, 906' using the first of the transceivers 1014. The second transceiver 916 can support a lower bandwidth channel for the transmission of control signals to, and event notifications from, peripheral alarm devices such as the video doorbell 1006, the smart lock 106, the video cameras 906, 906', (if they contain the relevant functionality, otherwise event notifications, etc. receive via Wi-Fi) door/window switches such as 902 and 908, etc. The low bandwidth channel or channels can typically be provided by using an appropriately allocated frequency in the ISM band (for industry, science and medicine) — such as (in Europe) 868MHz. Communication on the low bandwidth channel(s) is preferably encrypted. The primary reason for using a low bandwidth transmission channel is to enable the use of transceivers (in the peripheral alarm units) with low power consumption (and in particular lower power consumption than the power-hungry Wi-Fi transceivers that have been shown to be used), so that the battery-powered peripheral alarm units can achieve the required minimum battery life of 3 to 5 years. The transceivers 1014 and 1016 are coupled to a suitable antenna arrangement (not shown), which typically comprises (multiple) separate antenna elements for the two transceivers 1014 and 1016. One or both of the transceivers 1014, 1016 may support other transmission bands/protocols, such as a BlueTooth (RTM) or BLE, etc.. There is — preferably provided a transceiver that is configured or is configurable to communicate using 4G or 5G (or using any other appropriate PLMN data protocol) to communicate with the ARC 1002, for example in the event of a power failure in the building or loss of wired connection to the Internet 115 — for example, as the result of a malicious act by a dishonest person.

mn DK 202400084 Y3

The alarm controller 1004 is preferably configured to operate in at least one “connected away” mode, where the security of the perimeter of the building is monitored by door and window sensors such as door sensor 902, smart lock 106, and window sensors 908, and where at least one presence sensor monitors the interior of the building. For example, the internal presence sensor may be a passive infrared sensor (or T-MOS sensor), which may or may not be associated with a video camera. Additionally or alternatively, internal presence may be detected using so-called Wi-Fi detection, where the alarm controller runs a Wi-Fi detection algorithm, and the video doorbell and/or some other static Wi-Fi node (or access point) acts as a light emitter for the Wi-Fi detection. — The alarm controller 1004 is preferably also configured to operate in a disarmed mode, where internal presence is not monitored, and where warning signals received from door and window sensors are largely ignored by the alarm controller 604. However, the alarm controller 1004 is preferably configured to respond to tamper signals from, for example, the lock 106 (and also reports of an anomalous event determination), or other peripheral — alarm devices, even in the disarmed mode — because rogue individuals and other malicious actors may exploit the disarmed mode to attempt to sabotage or disable peripheral devices and other components of the system 1000. The alarm controller 1004 is preferably also configured to operate in at least one “connected home” mode, where internal presence is generally not monitored (or at least warnings received with — regard to detected presence by the alarm controller 1004 are ignored), but where security of the perimeter or a portion of the perimeter is monitored, such that that event signals are notified regarding detected door openings and detected window openings and possibly also reported as alarm events (possibly after giving a user at a location time to alert the detected event as a false alarm, for example by entering a security code on a user interface, or presenting a dongle, token or mobile phone, for example to an NFC interface).

The alarm controller 1004 is operatively coupled to a remote alarm receiving center (ARC) 1002 preferably via both a wired connection (e.g., broadband) and one or more wireless (e.g., SigFox (RTM), PLMN utilizing, e.g., GPRS, LTE, 4G, or 5G) — connections (e.g., utilizing one or more of NB-IOT (narrowband IoT supported by 3GPP), LTE-M (LTE machine-type communication). Notifications received by the alarm controller 1004 that match the profile of a potential “alarm event” (based on one or more of the identity of the sensor providing the notification, the nature of the notification, the connection state of the alarm system 1000, the time of day, the day of the week,

2 DK 202400084 Y3 etc.) may be reported to the ARC 1002 together with the identity of the alarm installation 1000. Such a notification may be generated or caused by the smart lock 106 detecting or determining the existence of an anomalous event. If the notification is associated with images or video, for example because the triggered sensor was a camera or was connected to a camera, then images/video may also be sent to the ARC. Typically, persons operating the ARC will review the reported event and will attempt to verify the event as an alarm event requiring action, based for example on the content of the images/videos, sounds received from microphones in the building, interactions with a user on site (for example via a user interface in the monitored building). If a — reported incident is verified as an incident requiring action (e.g., a burglary, theft, fire, etc., or a medical emergency), the local police and/or other emergency personnel/service (e.g., fire department, ambulance personnel, etc.) are alerted. The ARC may further contact a registered user or other personal contact (family member, support person, caregiver, etc.) by sending a push (or other) notification to a — user device 108, for example, using a technique as previously described.

In a system as illustrated in Figure 10, the video doorbell 1006 can communicate events to, and receive control signals from, the alarm controller 1004, which preferably uses a low bandwidth channel via the second transceiver 1016. The wider bandwidth of a Wi-Fi channel, provided by the first transceiver, is preferably used to transmit images/video from the video doorbell to the controller 1004, the controller relaying the images/video to the ARC 1002 and optionally to a user device such as 108 via a backend service 110, which sends push notifications to a user's personal communication device 108 when the doorbell is pressed (or optionally when someone approaches the doorbell 100). To support such communications, the controller 1004 —>o may utilize its broadband connection 114 to the Internet 115, the backend service 110 which sends push notifications to a PLMN (public land mobile network) 116, by which notifications are delivered to the user's personal communication device 108. The same communication path may be used to stream live video (and audio) from the video doorbell 100 to the user. A reverse communication path may also be supported to allow the user to speak to a person at the doorbell 100, and also to allow the user to activate (i.e., release) the lock 106 to allow the person at the door to open the door 102. The signals to activate (release) the lock are sent to the lock 106 via the controller 1004, which communicates with the lock 106.

13 DK 202400084 Y3

The video camera, processor, and software of the video doorbell are preferably together configured to record and stream (on demand, or as needed) full HD (e.g., 1080p) video.

The controller of the system, whether provided by the video doorbell 100 or by the alarm controller 1004, is preferably configured to respond to the entry of a “duress code” on the keypad of the video doorbell by alerting either the ARC 1002, in the case that the controller is the controller 904, or by alerting designated persons via calls (e.g., SIP calls) or notifications, preferably in each case providing images or video recorded at the time the duress code was entered (and perhaps from 30 seconds or more before the duress code was entered). A duress code is a code (e.g., a PIN) that can be used to disarm the security monitoring system, or gain access to the building, or both, but which is otherwise a “normal” access code. The idea is that under normal circumstances a normal code will be used, and the use of such a code does not cause a special notification. However, in the event that a user is threatened (for example, by an — abusive partner or ex-partner, a thief, or other type of malicious person) and forced to enter the protected building, the user enters the duress code rather than the normal code, thereby signaling the existence of a duress condition. Upon receipt of a duress notification, ARC staff may contact the police or other security personnel to facilitate an intervention. Similarly, if the duress notification is received by a friend, > neighbor, or family member (for example, in the case where the video doorbell acts as the controller of the security monitoring system), that person may intervene or involve the police so that they can intervene.

The controller of the system, whether provided by the video doorbell 100 or by the alarm controller 604, may be configured to enable access to the protected building (e.g., by unlocking the lock 106 and either disarming the system or at least temporarily ignoring a door open signal in relation to the relevant access door) based on

FacelD — i.e. recognition of the face of a visitor as corresponding to the face of a stored identity (where the controller stores, or has access to, a store of facial recognition data for one or more known identities to whom access has been granted).

The controller of the system, whether provided by the video doorbell 100 or by the alarm controller 1004, may be configured to enable access to the protected building (e.g., by unlocking the lock 106 and either disarming the system or at least temporarily ignoring a door open signal in relation to the relevant access door) based on recognition of a QR code, e.g., a dynamic QR code, which is displayed to the camera on the

2 DK 202400084 Y3 video doorbell — for example, by a visitor (or resident) who displays a QR code on the screen of a phone or other wearable device (such as a smart watch). Any QR code generated by or in conjunction with the system (e.g., by a relevant back-end system or a user's video doorbell or system app) preferably has a limited — duration of 24 hours or less (shorter durations, e.g., 12 hours, 6 hours, 4 hours, can also be used advantageously), to improve system security.

It is clear that the foregoing description is merely illustrative of examples of the invention, without limiting the scope thereof.

Claims (14)

45 DK 202400084 Y3 Requirements 1. Installation including an electronically controlled lock mounted on a door, the electronically controlled lock including a first detection arrangement in the form of an accelerometer or a magnetometer, the installation including a unit, separate from the lock, mounted on the door or on a door frame, which receives the door, which unit includes a second detection arrangement, the second detection arrangement comprising an accelerometer or a magnetometer, the installation including a digital processor circuit configured to receive signals from the first and second detection arrangements and to process the signals to detect a potential intrusion attempt based on sensor data detected simultaneously from different detection positions relative to the door and/or door frame. 2. The installation according to claim 1, wherein the lock includes a keyhole and further comprises one or more additional sensors in the form of (i) a microphone, preferably adapted to detect sound from a keyhole in the lock and/or from the vicinity of a cylinder in the lock; (11) a pressure sensor; and/or (iii) an insertion detector, optionally based on mechanical displacement of a detection element, for detecting insertion of an element such as a key, bump key or dirk into the keyhole of the lock: and wherein the digital processor circuit is adapted to receive signals or data from any such additional sensor in addition to signals received from the first detection arrangement, the digital processor circuit (or a remotely located processor) being programmed or configured to use signals or data from any such additional sensor for the determination and/or classification of an anomalous event. 3. The installation according to claim 1 or claim 2, wherein the lock further comprises one or more electronic detection arrangements for performing hand-on-handle or hand-on-lock detection, for example capacitive detection, and key-in-lock detection (e.g. using impedance detection), and the digital processor circuit is configured to use data from the one or more electronic detection arrangements for classifying events into specific attack modes such as lock picking, lock snapping and lock bumping. 16 DK 202400084 Y3 4. The installation according to any one of the preceding claims, wherein the digital processor circuit comprises a processor programmed to function as an autoencoder for processing the sensor data. 5. The installation according to any one of the preceding claims, wherein in the event that the digital processor circuit determines the presence of an anomaly, the processor circuit is programmed to analyze sensor data, or processed sensor data, to allocate a detected event to one of a plurality of event types, the event types including at least one threat class and at least one non-threat class. 6. The installation of claim 5, wherein the threat class includes at least one of lock picking, lock bumping, and lock snapping. 7. The installation according to any one of the preceding claims, wherein the processor circuitry comprises a processor which is a component of the electronically controlled lock. 8. The installation according to any one of the preceding claims, further comprising a security monitoring system for buildings of which the door is a part, and optionally at least a part of the processor circuit comprising a processor of a controller of the security monitoring system. The installation according to claim 8, wherein the controller of the security monitoring system is provided by a video doorbell arrangement. The installation according to claim 8 or claim 9, wherein the security monitoring system has at least one operating mode in which it is arranged to notify alarm events to a remote monitoring unit. The installation according to claim 10, wherein the security monitoring system has at least one operating mode in which it is arranged to notify the remote monitoring unit of anomalous events detected in the sensor data. 4 DK 202400084 Y3 12. A building security monitoring installation including a control unit for controlling operation of the security monitoring installation, the security monitoring installation further comprising: a door vibration sensor incorporating a first detector in the form of an accelerometer or a magnetometer, the vibration sensor being coupled to a door in the building; an electronically controlled lock incorporating a second detector, the second detector being in the form of an accelerometer or a magnetometer, the electronically controlled lock being coupled to the door in the building; the vibration sensor and the electronically controlled lock being configured to send signal data obtained from their detectors to the control unit; and the control unit being configured/programmed to classify an event based on the signal data obtained from the first and second detectors. 13. The building security monitoring installation of claim 12, wherein the control unit — includes a processor configured to use a machine learning model, such as an autoencoder, to process received signal data to determine the presence of anomalies in the received signal data. 14. — Building security monitoring installation according to claim 12 or 13, wherein — the control unit is configured/programmed to provide a quantized output, the value of which is selected from at least three levels (i.e. non-binary output).