Device and method for detecting presence

Abstract

The present invention relates to a method for detecting presence in a network controlled by an access gateway and comprising a plurality of devices, the method comprising: - selecting, among the plurality of devices, at least one device to be considered, on the basis of information relating to a duration or a frequency of activity of the device on the network; - selecting, among the devices to be considered, on the basis of at least one identifier relating to a type of device, at least one mobile device; - selecting, among the mobile devices, at least one personal device, on the basis of the information relating to a duration or a frequency of activity of the device on the network; - detecting the presence of at least one user associated with the personal device, if the at least one personal device is connected or disconnected.

Description

presence detection device and method

[0001] The present invention relates to the field of presence detection.

[0002] Today there are a variety of presence detection systems in environments, for example homes but also factories, offices, schools or any place where presence detection may be necessary.

[0003] These solutions can be based on sensors such as motion sensors, surveillance cameras, body heat detection systems, pressure detection systems (such as sensitive floor mats) and sound or voice detection systems.

[0004] Presence detection can be applied in various fields to improve security, energy efficiency, home automation, and other aspects. It can be applied to residential and commercial security, automatic lighting, energy management, heating automation, access control, connected healthcare, traffic and smart parking management, and home automation.

[0005] These applications demonstrate the versatility of presence detection systems and their ability to improve various aspects of daily life.

[0006] However, these systems rely on sensors that are generally installed to perform this presence detection. This therefore requires the installation of specific sensors and modules, generating purchasing and installation costs for these sensors and often the complexity of installing these devices. This additional equipment consumes energy and requires a suitable data collection infrastructure.

[0007] The present invention provides a presence detection solution that does not require complex installation or a device dedicated to presence detection.

[0008] The present invention relates to a method for detecting presence in a network controlled by an access gateway and comprising a plurality of equipment, said method comprising - a selection, from said plurality of equipment, of at least one equipment to be considered, from information relating to a duration or a frequency of activity of said equipment on said network, - a selection, from said equipment to be considered, from at least one identifier relating to a type of equipment, of at least one mobile equipment, - a selection, from said mobile equipment, of at least one personal equipment, from said information relating to a duration or a frequency of activity of said equipment on said network, - a detection of presence of at least one user associated with said personal equipment, if said at least one personal equipment is connected.

[0009] According to certain embodiments, the selection of equipment to be considered comprises: a) a selection of at least one equipment whose last connection to said gateway dates back less than a first duration, b) a selection from the equipment selected in a) of at least one equipment whose connection frequency to said gateway by said first network is greater than a determined frequency, c) a selection from the equipment selected in b) of at least one equipment whose connections to said gateway have been observed for at least a certain duration.

[0010] According to certain embodiments, the method comprises: - receiving identification information associating said at least one device with said gateway with a user of said device - recording said identification information in said access gateway, - identifying, from said identification information, said user whose presence has been detected in said environment.

[0011] According to certain embodiments, a selection, among said mobile equipment, of personal equipment comprises - a determination and an exclusion of said equipment permanently connected to said access gateway among said mobile equipment, from the data exchanges observed between said gateway and said selected mobile equipment.

[0012] According to certain embodiments, the method comprises - a recording of at least one presence information over a time window, for at least one user, from said presence detection over said time window, - a presence prediction for said user over at least one future time window based on the at least one recorded presence information.

[0013] According to certain embodiments, the method comprises - putting at least one electrical device of said environment and / or said access gateway into standby mode as a function of said presence prediction. According to certain embodiments, the information relating to said at least one device connected or having been connected to said gateway comprises - the date and time of the connection of said device, - the MAC address of said device, - said type associated with said device - a volume of traffic sent or received by said device over a time interval.

[0014] The features presented in isolation in the present application in connection with certain embodiments of the method of the present application can be combined with each other according to other embodiments of the present method.

[0015] According to another aspect, the present invention also relates to a computer program comprising instructions for executing the steps of the method according to the invention, according to any one of its embodiments, when said program is executed by a computer.

[0016] According to another aspect, the present invention also relates to a computer-readable recording medium on which is recorded a computer program comprising instructions for carrying out the steps of the method according to the invention, according to any of its embodiments.

[0017] The invention also relates to an access gateway in a network comprising a plurality of devices connected to said gateway by said network, said gateway comprising one or more processors configured to- select, from said plurality of devices, at least one device to be considered, from information relating to a duration or a frequency of activity of said device on said network,- select, from said devices to be considered, from at least one identifier relating to a type of device, at least one mobile device,- select, from said mobile devices, at least one personal device, from said information relating to a duration or a frequency of activity of said device on said first network,- detect the presence of at least one user associated with said personal device, if said at least one personal device is connected.

[0018]

[0019] Other characteristics and advantages of the present invention will emerge from the description given below, with reference to the accompanying drawings which illustrate an exemplary embodiment thereof without any limiting character.

[0020] Represents a system according to certain embodiments of the invention,

[0021] Represents a method according to certain embodiments of the invention,

[0022] Larepresents an embodiment of detection of equipment to be considered according to the present invention,

[0023] The represents an embodiment of hardware architecture of a device according to the present invention,

[0024] The present invention relates to presence detection. The following description is based on presence detection in a domestic environment but can be applied to presence detection in various environments. It can be applied in a non-limiting manner to industrial environments, hospital environments, and offices.

[0025] Larepresents a home environment 1 in which this disclosure may be implemented.

[0026] A P1 gateway is used to control a local network (or several) and is configured to communicate with a plurality of D devices i , for example devices D1 to D7 shown in the. Devices D1 to D7 are also configured to communicate with gateway P1 and constitute a set of connected devices.

[0027] For example: - D1 is a tablet, - D2 is a connected watch, worn by a user, - D3 is a connected refrigerator, - D4 and D5 are laptops, - D6 and D7 are mobile phones.

[0028] Di devices can be devices that do not have a sensor-like function and are therefore not capable of providing data similar to those provided by sensors usually present in presence detection systems such as cameras, microphones, pressure sensors, volumetric sensors, opening detection sensors, smart bulbs. Di devices are devices whose main function is not associated with presence detection, even if, of course, a laptop or a mobile phone are equipped with a microphone and cameras.

[0029] The environment 1 may further comprise presence detection systems based on presence detection sensors. The present disclosure, as illustrated by the illustrated embodiments enet 3, excludes this type of device for detecting the presence of people in the environment.

[0030] This disclosure aims to detect the presence of people usually living in environment 1 or usually working in environment 1 or more generally the presence of people usually present in environment 1. It aims in particular to exclude the consideration of people considered to be people passing through environment 1.

[0031] Di devices can communicate with the P1 gateway using different communication protocols. Indeed, a smartwatch can communicate with the P1 gateway using, for example, a Bluetooth protocol, while tablets, computers, and mobile phones preferentially use a WIFI wireless network. In other embodiments, they can use a wired network or another communication protocol (Zigbee, cellular, etc.).

[0032] The gateway P1 is also connected to a network external to the environment 1, by a second network R2, which can for example be a cellular type wireless network (2G, 3G, 4G, 5G…, satellite) or any wired network such as ADSL, Fiber, RTC, Internet network (non-exhaustive).

[0033] The access gateway P1 can, for example, constitute an interface between environment 1 (or the devices present in environment 1 and connected to the gateway P1) and remote servers, in particular those present on the Internet network.

[0034] In accordance with the present disclosure, the access gateway P1 comprises one or more processors configured to- select, from the plurality of devices, at least one device to be considered, from information relating to a duration or a frequency of activity of the device on said network,- select, from the devices to be considered, from at least one identifier relating to a type of device, at least one mobile device,- select, from the mobile devices, at least one personal device, from the information relating to a duration or a frequency of activity of the device on the first network,- detect the presence of at least one user associated with the device.

[0035] The access gateway P1 can collect a plurality of information relating to the connected Di devices. In particular, it can receive or request this information continuously, for example every 15 minutes. This data can be recorded in a storage memory of the gateway P1.

[0036] In particular, it may collect, for Di devices, information relating to the date and time of connection to the gateway, the MAC address of the connected Di device, the type of device (mobile phone, laptop, tablet, etc.), the friendly name (for example, "phone_Gilles"), the operating system of the device, the manufacturer or brand of the device, and the volume of traffic sent and / or received by the device, for example per 5-minute interval. The friendly name may allow a device to be identified other than by its MAC address, in the event that its MAC address is modified.

[0037] The information recorded in the P1 gateway may also include information relating to the IP address of the device, an indicator relating to the signal strength, a signal-to-noise ratio, a maximum throughput supported by the device, a current throughput, a device version, an indicator called "active", for example binary, making it possible to determine whether the device is connected or not, a software version of the device, the average throughput of data received in Mbit / s, the average throughput of data transmitted in Mbit / s, the number of bytes received over a period, the number of bytes transmitted over a period.

[0038] The information thus collected can be processed by the access gateway P1, or transmitted to IR network infrastructures of an operator, in charge of implementing the presence detection method, from the information collected or at least part of the information collected. In such an embodiment, the method as disclosed is not implemented in the access gateway but in a server of a telecommunications operator to which the gateway P1 belongs and in charge of managing the network R2.

[0039] La represents an embodiment of a method according to certain embodiments of the present invention.

[0040] Data relating to the equipment in environment 1 connected to gateway P1 are collected regularly or not by gateway P1. For example, this data can be collected every 5 minutes. One of the objectives of the present method is to detect the presence of people present in environment 1, while excluding people occasionally coming to environment 1. In other words, the method can make it possible to detect the presence and identify people living in environment 1. It can then be used to predict the presence over subsequent time windows based on the detection over a past time window.

[0041] As mentioned previously, this data may include for Di devices, information relating to the date and time of connection to the gateway, the MAC address of the connected Di device, the type of device (mobile phone, laptop, tablet, etc.), the friendly name (for example, "phone_Gilles"), the operating system of the device, the manufacturer or brand of the device, and the volume of traffic sent and / or received by the device, for example per 5-minute interval, the IP address of the device, an indicator relating to the signal strength, a signal-to-noise ratio, a maximum throughput supported by the device, a current throughput, a device version, an indicator called "active", for example binary, to determine whether the device is connected or not, a software version of the device, the average throughput of data received in Mbit / s, the average throughput of data transmitted in Mbit / s, the number of bytes received over a period, the number of bytes transmitted over a period.

[0042] Thus, connection information (date, time, etc.) can be associated with a device from one of its identifiers, including its friendly name, or its MAC address when it is not modified. This information can be presented in the form of a table in the access gateway P1. When a new device connects for the first time to the access gateway P1 in environment 1, it is considered a new device and added to a list of devices known to the gateway P1 in environment 1.

[0043] During a step E1, at least one piece of equipment to be considered is selected from among the plurality of devices, or even equipment, Di, based on information relating to a duration or frequency of activity of said equipment on said network.

[0044] Step E1, or more generally the presence detection method, can be triggered regularly or can be triggered by an application which uses the presence information, such as the applications listed later in this disclosure.

[0045] It can be noted that the Di devices are devices known to the gateway P1. In other words, they are devices that are connected or have been connected to the gateway P1 in the environment 1. Thus, at a time t, and for example when collecting information relating to a duration or a frequency of activity, some of the Di devices are not connected to the gateway P1 and no information is therefore sent back to the gateway P1 or collected by the gateway P1.

[0046] The selection of at least one piece of equipment to consider takes into account information relating to a duration or frequency of activity.

[0047] Among the information relating to a duration or frequency of activity, we can cite in particular: - the duration, for example in days, since which the equipment has not been connected to the gateway P1, called Max_unseen, - a duration between the first connection of said device and the last observation of the same device, called Min_span - the frequency, over a given period, at which the equipment connects to the access gateway, called Average_pres.

[0048] This information can for example be obtained from the variation of the “active” indicator which represents the connectivity status of a device to the gateway P1. A time indicator can be associated with each change in status of the “active” indicator, making it possible to determine the information relating to a duration or frequency of activity of the device in the environment 1.

[0049] .

[0050] It may be noted that these values ​​may not be received directly by the P1 gateway but calculated by the P1 gateway from data relating to connection times or times / durations, which can be determined by the P1 gateway using the traffic observed on the network between the device and the gateway or the variation of the “active” indicator. In particular, certain durations are based on a daily observation. Thus, even if the connection data are retrieved more frequently than daily, they can be aggregated to determine per day whether a device has connected to the P1 gateway.

[0051] A plurality of combinations of these parameters can be considered to determine whether a Di device can be selected to be part of the equipment to be considered among the Di equipment.

[0052] In particular, we can compare the three values ​​listed above, Max_unseen, Min_span and Average_pres, to threshold values.

[0053] La represents an embodiment of a method for determining the equipment to be considered as implemented in step E1.

[0054] The information relating to a duration or frequency of activity, associated with the devices known to the access gateway, is retrieved, step E11, for example in storage means of the gateway P1. This information is associated with a device in the storage means, or table, by an identifier of the device, such as the MAC address, the friendly name.

[0055] This information may include:- the duration, for example in days, since which the equipment has not been connected to the P1 gateway, called Max_unseen,- a duration between the first connection of said device and the last observation of the same device, called Min_span,- the frequency, over a given period, at which the equipment connects to the access gateway, called Average_pres.

[0056] It should be noted that a device that has not been connected to the P1 gateway for more than 90 days is no longer one of the devices taken into account by this process.

[0057] During a step E12, the information Max_unseen, for each device Di, is compared to a first determined value S1. If Max_unseen is greater than S1, then it can be considered that the device is not part of the equipment to be considered since it has not been connected for a duration considered to be greater than that of an occupant considered to be regular of the environment 1. Such a device can for example be a device of a person passing through the environment 1, who has not returned there for a duration greater than the duration S1. For example, for information purposes, S1 can be set to 30 days. Thus, a device that has not been connected for 30 days to the access gateway P1 is considered to be a device that is not to be considered by said method among the set of devices Di known to the gateway P1.Step E12 can therefore make it possible to exclude devices that have not connected to the gateway P1 for a given duration, here for example S1. To do this, as mentioned previously, the gateway P1 determines, for each device, an indicator per day indicating whether the terminal has connected or not that day. This indicator can be compiled from a plurality of daily connection information for the device.

[0058] If the test in step E12 indicates that the device has been connected to the gateway P1 for less than a duration S1, for example 30 days as indicated above, then we move on to a second test for this device, step E13.

[0059] In step E13, we compare Average_pres to a second threshold S2. If Average_pres is lower than S2 then we can consider that the device is not part of the equipment to be considered since it has a connection frequency lower than the connection frequency S2 over a given period. The given period can be a sliding time window relative to the current date. Such a device can for example be that of a person coming occasionally to environment 1. For example, such equipment can be that of a childminder at home, a domestic employee, a friend. For example, we can, for information purposes, set S2 to 1.5 / 7 corresponding to 1.5 days of connection over a 7-day window.Thus, a device that has connected once in the last 7 days has an Average_pres value lower than the threshold S2 and is considered as a device that is not to be considered by said method among the set of devices Di known to the gateway P1. According to the method, a person occasionally coming into the environment 1 is not to be considered as a person belonging to the people living in the environment 1 and is therefore excluded. Step E13 can therefore make it possible to exclude the devices present occasionally in the environment 1. By occasionally, we can understand at a frequency lower than a given value, here S2.

[0060] If the test in step E13 indicates that the device has been connected to gateway P1 at a frequency greater than S2, for example 2 days per week, then we move on to a third test for this device, step E14.

[0061] In step E14, Min_span is compared to a third threshold S3. Min_span corresponds to a time window in days between the last connection and the first connection of a device.

[0062] If Min_span is less than this third threshold, then the device can be considered not to be part of the equipment to be considered. According to some embodiments, the threshold S3 can be taken as 15 days. Thus, if a device last connected 14 days ago, Min_span is less than S3 when S3 is 15 and the device is not part of the equipment to be considered. On the other hand, the next day, Min_span will be 15 and the device will be part of the equipment to be considered.

[0063] At the end of step E14, we obtain a set of equipment to consider.

[0064] With reference to the, following step E1, we move on to step E2. We determine the list of mobile equipment, a mobile equipment being associated with a person, the presence of a mobile equipment can indicate the presence of a person in environment 1.

[0065] To do this, we can exclude equipment that is not mobile equipment, for example connected household appliances, connected cameras.

[0066] In step E2, the mobile devices are selected from among the equipment to be considered obtained at the end of step E1, based on at least one identifier relating to a type of equipment. Among the type of equipment, a representative type of equipment, such as a mobile telephone, a laptop, a tablet, etc., may be cited. As mentioned previously, this information is collected by the access gateway during data exchanges with the devices Di. The access gateway therefore selects the mobile devices from among the devices to be considered, for example by comparing the type of each device to be considered with a list of representative types of mobile devices.

[0067] It may be noted that alternatively, step E2 could be implemented before step E1.

[0068] In a step E3, at least one personal device is selected from among the mobile devices selected in step E2, based on information relating to a duration or frequency of activity of the devices on the network. In other words, the devices permanently connected to the access gateway P1 are excluded from the list of mobile devices selected in step E2.

[0069] In this step, for example, we want to remove devices that remain permanently in environment 1 and which therefore cannot be associated with the presence of a person in environment 1. Among these devices, we can notably cite laptops which are family computers and which only leave environment 1 on an exceptional basis.

[0070] The selection is based on a duration or frequency of activity of the equipment on the network. It may include a determination and exclusion of said equipment permanently connected to said access gateway from among said mobile equipment, based on the data exchanges observed between said gateway and said selected mobile equipment.

[0071] The frequency or duration of equipment activity can be based on the hour, day, or month. For example, a device can be determined to be permanently connected if it is detected as connected every hour, over a 24-hour period, over a week, or over several weeks.

[0072] This selection can also take into account the "active" indicator to determine whether a device is permanently connected to the gateway, by determining its value over a period of time. It is assumed that a value of "1" for the indicator means that the device is connected, and a value of "0" means that the device is not connected.

[0073] If its value remains set to "1" for a specified period of time, for example two weeks, then the device can be considered to be permanently based in environment 1. For example, it could be a non-personal laptop.

[0074] If its value changes from "1" to "0" several times over a given period of time, for example two weeks, then the device can be considered a personal device.

[0075] In other words, according to certain embodiments of step E3, it is possible to select, from among the mobile equipment selected during step E2, at least one equipment, from a variation of an indicator relating to the connection of the equipment over a given period, for example a day.

[0076] At the end of step E3, a relevant list of personal devices is obtained, among the devices Di known to the gateway P1. These personal devices are devices determined through steps E1 to E3, as being able to be considered as devices belonging to a person and mobile. Their presence in the environment 1 is therefore representative of the presence of the person to whom they belong in the environment 1.

[0077] An optional step may allow a user name for that device to be associated with a device. This step may, for example, be implemented in the access gateway through a user interface that allows a device identifier, for example its Mac address or friendly name, to be associated with a user name. This may advantageously allow for more precise detection of which person is present rather than which device is present.

[0078] During a step E4, it is therefore possible to detect the presence of at least one user associated with a personal device as determined during steps E1 to E3, if the personal device is connected. Indeed, during steps E1 to E3, the list of personal devices that are representative of a person living in the environment 1 is obtained and can be used to detect the presence of this person in the environment from the presence of his or her personal device in the environment.

[0079] A presence or absence table can be updated during step E4 associating with each device identified as personal during steps E1 to E3, a presence indicator, such as present / absent, making it possible to know at any time whether the user of a device is present. The presence indicator can be a binary variable “0” if absent, “1” if present or vice versa. To do this, if the equipment is identified as personal equipment and if it is connected to the gateway P1, then this device has a “present” status, the user to whom this personal device belongs is considered present. If the device disconnects from the gateway P1, then the status of this device changes to “absent” status and the user to whom this device belongs is considered absent.

[0080] Steps E1 to E3 may constitute an update of the list of personal devices representative of the presence of individuals in an environment. This update may be done regularly or not, at the request of an application. However, since steps E1 to E3 may require a volume of data relating to a relatively long period of time, for example 30 days, steps E1 to E3 may advantageously be implemented at least daily.

[0081] In step E4, the "absent" or "present" status data can be recorded, either in the gateway P1 or in a network operator's infrastructure. This data can be recorded over a period of time, the duration of which can correspond to a daily, weekly, monthly, quarterly or annual duration, for example. From this recorded presence or absence data. This data can be recorded for each device, therefore each person to whom the device belongs or for environment 1.

[0082] The status of the environment is "absent", that is, there is no one present in environment 1 if the status of each personal device determined during step E3 is positioned as "absent". The status of the environment is "present" if at least the status of one of the personal devices determined during step E3 is positioned as "present".

[0083] From the recorded status data, predictions of presence or absence over a future period can be made, based on statistics or using time series prediction algorithms, for each person in environment 1 or for environment 1.

[0084] Thus, the present disclosure comprises, following the detection of presence of at least one user associated with the personal equipment, - a recording of status data relating to the detection, over at least one time period and - a prediction, for at least one of the users or for the environment, of presence over a subsequent time window from the at least one piece of status information.

[0085] In other words, the present disclosure comprises - a recording of at least one presence information over a time window, for at least one user, from the presence detection over the time window, - a presence prediction for the user over at least one future time window based on the at least one recorded presence information.

[0086] Of course, presence detection implies absence detection.

[0087] Presence detection can find many applications and in particular presence detection can be used to:

[0088] - Residential and commercial security: Presence detection systems are often used in security systems to detect any movement or unauthorized intrusion. This can include motion sensors, surveillance cameras with motion detection, etc.

[0089] - Automatic lighting: Occupancy sensors are often used to control lighting in residential, commercial, or industrial spaces. Lights automatically turn on when someone enters a room and turn off when the room is empty, helping to save energy.

[0090] - energy management: Presence detection systems can be integrated into energy management systems to regulate the use of electrical equipment based on the presence or absence of people.

[0091] - automation of heating, ventilation and air conditioning: Occupancy sensors can be used to automatically adjust air conditioning settings based on the presence of occupants, thereby optimizing comfort while saving energy.

[0092] - access control: Presence detection systems are often used in access control systems to authorize entry or exit based on the detection of authorized persons.

[0093] - connected health: Health tracking devices often use presence sensors to monitor individuals' physical activity, such as activity bracelets or tracking applications on smartphones.

[0094] - traffic management and smart parking: Presence sensors are used in traffic management systems to detect the presence of vehicles and adjust traffic lights accordingly. They are also used in smart parking systems to indicate available parking spaces.

[0095] - Home automation: Presence detection systems can be used to automate various functions in a home, such as automatically opening doors, starting household appliances, etc.

[0096] These applications demonstrate the versatility of presence detection systems and their ability to improve various aspects of daily life.

[0097] Advantageously, the present disclosure can make it possible to avoid the installation and use of sensors assisting in the detection of presence and in particular intrusive or energy-consuming sensors, such as motion sensors (volumetric sensors), cameras, microphones, sensors for detecting the opening of doors, windows, etc., as well as the implementation of complex remote monitoring applications.

[0098] La represents an example of hardware architecture 30 of an access gateway P1 allowing the implementation of a method according to the present invention and as represented for example in. This hardware architecture is that of a computer. Other hardware architecture elements are present in the gateway P1 and not represented here.

[0099] The hardware architecture 30 comprises one or more processors 31 (only one is shown in the) implementing a method according to the present disclosure, a read-only memory 32 (of the “ROM” type), a rewritable non-volatile memory 33 (of the “EEPROM” or “Flash NAND” type for example), a rewritable volatile memory 34 (of the “RAM” type), a communication interface 35 with the network R2 and a communication interface 36 (or several communication interfaces when the access gateway communicates via WIFI, Bluetooth, Zigbee) with the equipment of the environment 1).The read-only memory 32 constitutes a recording medium according to an exemplary embodiment of the invention, readable by the processor or processors 31 and on which is recorded a computer program Prog according to an exemplary embodiment of the invention comprising instructions for executing steps of the method according to one or more of the embodiments of the invention. Alternatively, the computer program Prog is stored in the rewritable non-volatile memory 33.

[0100] The computer program Prog may enable the gateway P1 to implement at least part of the method in accordance with the present disclosure and as illustrated for example in.

[0101] This computer program Prog can thus define functional and software modules, configured to implement the steps of a selection method in accordance with an exemplary embodiment of the invention, or at least part of these steps. These functional modules rely on or control the hardware elements 31, 32, 33, 34, 35 of the access gateway P1 cited above.

Claims

Method for detecting presence in a network controlled by an access gateway and comprising a plurality of equipment, said method comprising - a selection (E1, E2, E3), from said plurality of equipment, of at least one mobile and personal equipment, from information relating to a duration or a frequency of activity of said equipment on said network and at least one identifier relating to a type of equipment. - a presence detection (E4), of at least one user associated with said mobile and personal equipment, if said at least one personal equipment is connected. Method according to claim 1 in which the selection of equipment to be considered comprises: a) a selection (E12) of at least one equipment whose last connection to said gateway dates back less than a first duration (S1), b) a selection (E13) from among the equipment selected in a) of at least one equipment whose connection frequency to said gateway by said first network is greater than a determined frequency (S2), c) a selection (E14) from among the equipment selected in b) of at least one equipment whose duration between the first connection and the last connection is greater than a second determined duration (S3). Method according to one of the preceding claims comprising - a reception of identification information associating said at least one device of said network with said gateway with a user of said device - a recording of said identification information in said access gateway, - an identification, from said identification information, of said user whose presence has been detected in said environment. Method according to one of the preceding claims in which a selection, among said mobile equipment, of personal equipment comprises - a determination and an exclusion of said equipment permanently connected to said access gateway among said mobile equipment, from the data exchanges observed between said gateway and said selected mobile equipment. Method according to one of the preceding claims comprising- a recording of at least one presence information over a time window, for at least one user, from said presence detection over said time window,- a presence prediction for said user over at least one future time window based on the at least one recorded presence information. Method according to the preceding claim comprising - putting at least one electrical device of said environment and / or said access gateway on standby as a function of said presence prediction. Method according to one of the preceding claims in which the information relating to said at least one piece of equipment connected or having been connected to said gateway comprises - the date and time of the connection of said equipment, - the MAC address of said equipment, - said type associated with said equipment - a volume of traffic sent or received by said equipment over a time interval. Computer program comprising instructions for executing the steps of the method according to one of claims 1 to 7 when said program is executed by a computer. Computer-readable recording medium on which is recorded a computer program comprising instructions for carrying out the steps of the method according to one of claims 1 to 7. Access gateway in a network comprising a plurality of devices connected to said gateway by said network, said gateway comprising one or more processors configured to- select, from said plurality of devices, at least one mobile and personal device, from information relating to a duration or a frequency of activity of said device on said network and at least one identifier relating to a type of device,- detect the presence of at least one user associated with said mobile and personal device, if said at least one personal device is connected.

Images