A method for improving detection performance in a distributed RF sensing system

Abstract

A method (500) for improving radio frequency, RF, based motion detection in a wireless network (100) comprising a plurality of nodes (200a, 200b, 200c, 200d); the method (500) comprising a node out of the plurality of nodes carrying out steps of: receiving (S501) messages from one or more neighboring nodes; deriving (S502), for a first neighboring node, a moving average of received signal strength of received messages; calculating (S503) a variation in received signal strength of a newly received sensing message from the first neighboring node as compared to the moving average; detecting (S504) motion when the variation is negative and the magnitude of the variation is larger than a first threshold or when the variation is positive and the magnitude of the variation is larger than a second threshold; with the second threshold is different from the first threshold.

Description

[0001] A method for improving detection performance in a distributed RF sensing system

[0002] FIELD OF THE INVENTION

[0003] The invention relates to the field of a radio frequency based sensing system. More particularly, various methods, apparatus, and systems are disclosed herein related to improving robustness to RF noise in a distributed RF sensing system.

[0004] BACKGROUND OF THE INVENTION

[0005] Radio frequency (RF) based sensing is a technique widely used for motion detection, which involves one or more transmitters arranged for transmitting RF signals on a regular basis, such as according to a certain transmission frequency, and one or more receivers for receiving the RF signals. These RF signals, which may also be used for data communication, when passing through a sensing volume, are affected by presence / movement of a person within the sensing volume. By monitoring such variations of RF signals, it is possible to infer presence or motion of the person. RF based sensing may also be extended to other applications such as location detection, fall detection, gesture detection, or vital signs detection, which are also based on monitoring how RF signals are affected in the sensing volume.

[0006] The processing of the RF signals for presence or motion detection may be either performed locally e.g., at the receiving node, or it can be performed externally at an external device. For external processing, a collector node is generally used to collect RF signals and transmit it to the external device. With the external processing approach, it is necessary to commission different roles for the nodes in the system, which leads to additional commissioning work for an installer, and when the nodes for performing the various RF sensing functions are not selected properly some motions may even not be detected at all. Furthermore, the traffic resulted from collecting RF signals at the collector node and forwarding the RF signals from the collector node to the external device also adds to the traffic load of the network, especially when the network is used also for data communication. Therefore, it is considered to be beneficial to implement RF sensing in a distributed manner, such that each node in a particular sensing group sends RF sensing messages and itself listens to all RF messages coming from its own neighbour nodes in the same sensing group for presence or motion detection.

[0007] The RF sensing system may make use of existing wireless communication technologies. For example, ZigBee, Thread and Bluetooth Mesh are commonly used technologies targeted at loT applications such as lighting and building automation. They provide a low latency, low-rate service that enables messages to be passed between, for example, a light switch and one or more luminaires. To enable messages to be routed correctly, each node on the wireless network is assigned a local network unicast address and may be addressed either directly as an individual node or, via a group address, as a member of a group. Once configured, such networks are typically expected to operate autonomously.

[0008] For large installations in an RF sensing system, many nodes may be configured to send RF sensing messages in a regular pattern, which results in a significant use of the available bandwidth and also causes RF interference to each other. This could lead to RF sensing messages being dropped due to collisions (e.g. with messages from neighbouring networks) or data messages not being delivered reliably. What is even worse is that when messages interfere only partially with each other, or if the interference does not lead to the loss of an entire sensing message, there is a possibility that the RF power of the sensing packet is changed due to the interference from a colliding packet. This situation is known as the ‘hidden terminal’ problem, which can cause the RS SI of the received packet to change while it still possible to interpret the packet correctly. For an RF sensing system that relies on the detection of changes in the RSSI over time, such a coincidence of two packets coming from different nodes could result in a change in the RSSI on the receiving node, which would lead to a false positive detection.

[0009] Booranawong Apidet et al: “Adaptive filtering methods for RSSI signals in a device-free human detection and tracking system” is related to an adaptive filtering method for RSSI signals in a device-free human detection and tracking system. To improve the detection accuracy, the measured RSSI inputs are filtered with different filtering levels adaptively.

[0010] Bojan Mrazovac et al: “A human detection method for residential smart energy systems based on Zigbee RSSI changes” is related to a method for device-free presence detection by analyzing and quantifying radio signal strength variations.

[0011] Wisanmongkol Juthatip et al: “Multipath mitigation for RSSI-based Bluetooth Low Energy localization” is related to a 2-stage multipath mitigation technique comprising RSSI filtering and RSSI combining. SUMMARY OF THE INVENTION

[0012] It is recognized by the inventor that it is beneficial to implement additional measures to alleviate the impact on RF sensing performance due to RF interference. More particularly, the goal of this invention is achieved by a method as claimed in claim 1, by a node as claimed in claim6, by a wireless network as claimed in claim 10, and by a computer program as claimed in claim 13.

[0013] In accordance with a first aspect of the invention a method is provided. A method for improving radio frequency, RF, based motion detection in a wireless network comprising a plurality of nodes; the method comprising a node out of the plurality of nodes carrying out steps of: receiving messages from one or more neighboring nodes out of the plurality of nodes; deriving, for a first neighboring node out of the one or more neighboring nodes, a moving average of received signal strength of received messages; calculating a variation in received signal strength of a newly received sensing message from the first neighboring node as compared to the moving average of received signal strength of previously received messages from the first neighboring node; detecting motion when the variation is negative and the magnitude of the variation is larger than a first threshold or when the variation is positive and the magnitude of the variation is larger than a second threshold; updating for the first neighboring node the moving average of received signal strength with the newly received sensing message when no motion is detected; wherein the second threshold is larger than the first threshold.

[0014] The messages may be sensing messages dedicated for RF based motion detection. Alternatively, the messages may also be data messages or messages for network control purposes.

[0015] The plurality of nodes may be within a direct communication range of each other.

[0016] The plurality of nodes may belong to a same sensing group, while there is another group of nodes belong to a different sensing group located in the proximity. The division of different sensing groups may be based on locations, such as different groups for nodes located in different rooms or on different floors, different control functions, or different capabilities of nodes. Each node of the group may ignore sensing messages received from nodes belong to a different sensing group. The motion detection may be used for building automation to control sensors and actuators integrated in or co-located with the one or more nodes.

[0017] The messages or sensing messages may be transmitted by each node via broadcasting. Alternatively, the messages may also be transmitted by each node of a sensing group via multicasting to other nodes in the same group, such that the sensing messages comprise a group address for a particular sensing group.

[0018] To detect the presence or motion in the sensing area timely, preferably the messages sensing messages may be transmitted by each node on a regular basis. The group of nodes belonging to a same sensing group may use a same duty cycle or transmission frequency for sending sensing messages repeatedly. It may also be an option that each node in the same sensing group also adopts a different duty cycle or transmission frequency for sending sensing messages.

[0019] According to a conventional motion detection approach, each node out of the plurality of nodes calculates and keeps track of a running mean and variance value for the received signal strength of messages that it receives from each node in the surrounding or each of the other nodes in the sensing group and compares this to a threshold value. And motion detection is confirmed when the variance is larger than the threshold value.

[0020] There is also situation that a decreased received signal strength of the message is measured when a link line between a transmitter and a receiver is obstructed by a person or object. Hence, it is important to monitor both positive and negative variations of the received signal strength.

[0021] However, with regard to the interference situations discussed above, when a message is only interfered partially by an interference, such that it can still be interpreted by the receiving node, but the received signal strength of the packet is changed due to the interference. In most cases, since the interference and the message are non-coherent, and the energy is simply added up, leading to an increased received signal strength. Consequently, a false positive detection may be derived based on such increased received signal strength.

[0022] Therefore, it is important to apply different thresholds when checking the positive and negative variations on the received signal strength respectively.

[0023] The received signal strength may be calculated via a Received Signal Strength Indication, RSSI.

[0024] Motion detection may be related to a binary presence indication, and a more advance detection regarding a type of the motion, such as gesture detection, activity detection, vital sign detection, etc, may be followed. The type of the motion may be detected by classification, or a pattern recognition based on a comparison against an established baseline.

[0025] The motion or presence detection may be used to trigger a control command to one or more electronic devices connected to or controlled by each node of the group of nodes. It may also be possible that each node out of the group is integrated in an electronic device. The electronic device may be a lighting device, a wall switch, a bridge, a speaker, a television, a thermostat, a wireless sensor, a power outlet plug, a doorbell, a smart home assistant device, and / or a smart home appliance. For example, the control command may be related to a turn on or power on command to the one or more electronic devices. The control command may be related to changing a setting or an operation mode of the one or more electronic devices.

[0026] The motion detection may not be limited to presence sensing of a people in the detection area, which may also be used to identify and classify the motion, such as for gesture detection, fall detection, sleep monitoring, vital sign detection by means of heart rate detection, breathing detection etc.

[0027] The group of nodes may follow a wireless communication standard for transmitting the sensing messages. The wireless communication standard may be related to a standard of Zigbee, Thread, or Bluetooth Low Energy.

[0028] Preferably, the second threshold is larger than the first threshold.

[0029] Since the interference usually only results in an increased received signal strength or a positive variation of the received signal strength, it is beneficial to have the second threshold larger than the first threshold. In that case, the chance of a false detection triggered by a large positive variation is reduced with the relatively larger second threshold. By keeping the first threshold lower as compared to the second threshold, it also helps to reduce the chance of miss detection on the negative variations of the received signal strength.

[0030] Beneficially, the first threshold and the second threshold are either preconfigured or determined adaptively based on the moving average of received signal strength.

[0031] The first threshold and the second threshold may be fixed values defined by a user or an application. It may also be the case that the first threshold and the second threshold are defined in a relative unit, such as in decibel (dB), based on the moving average of received signal strength.

[0032] In one example, the method further comprises the steps of: repeating the steps in processing messages received from the first neighboring node to process messages received from a further neighboring node out of the one or more neighboring nodes; confirming motion detection only when motion is detected by processing messages from the first neighboring node and by processing messages from the further neighboring node respectively.

[0033] To further reduce the chance of false detection, it is also beneficial that motion detection is confirmed only when motion is detected based on messages from both the first neighboring node and the further neighboring node. This is especially beneficial when the nodes are deployed in a high density, and the presence of a person or object may have impact on more than one link lines, with each link line identifying a pair of transmitter and receiver.

[0034] The motion detections based on messages from the first neighboring node and messages from the further neighboring node may be confirmed within a certain time interval to guarantee the correlation between the two detections. The time interval may be related to the frequencies on receiving messages from the first neighboring node and from the further neighboring node.

[0035] Advantageously, the first threshold and the second threshold are determined for the first neighboring node and the further neighboring node respectively.

[0036] Each pair of a sending node and a receiving node distinguishes a signal propagation path for the sensing messages from the sending node to the receiving node. Therefore, each pair of a sending node and a receiving node may confront a different path loss. Even if the same type of sensing messages and same transmission power are used by the nodes for sending the messages, different received signal strengths may be experienced by different receiving nodes. It is thus beneficial to determine the first threshold and the second threshold for the first neighboring node and the further neighboring node respectively.

[0037] In another example, the method further comprises the steps of: receiving a message from a further neighboring node out of the one or more neighboring nodes with the message comprising an indication of motion detection by the further neighboring node; confirming motion detection only when motion is detected by processing messages from the first neighboring node and by receiving the message from the further neighboring node respectively.

[0038] In this example, a more reliable motion detection is achieved by motion detection by the node itself based on processing messages from the first neighboring node and the received indication of motion detection at the further neighboring node side. Similar to the previous example, the reception of the indication from the further neighboring node and the motion detection based on messages from the first neighboring node may happen within a certain time interval to guarantee the correlation between the two detections.

[0039] In one example, the messages are either InterP AN messages according to a Zigbee standard, or Bluetooth beacons according to a Bluetooth Low Energy standard.

[0040] In accordance with a second aspect of the invention a node is provided. A node out of a plurality of nodes comprised in a wireless network for radio frequency, RF, based motion detection; the node comprising: a receiver configured to receive messages from one or more neighboring nodes; and a controller configured to: o derive, for a first neighboring node out of the one or more neighboring nodes, a moving average of received signal strength of received messages; o calculate a variation in received signal strength of a newly received sensing message from the first neighboring node as compared to the moving average of received signal strength of previously received messages from the first neighboring node; o detect motion when the variation is negative and the magnitude of the variation is larger than a first threshold or when the variation is positive and the magnitude of the variation is larger than a second threshold; and o update for the first neighboring node the moving average of received signal strength with the newly received sensing message when no motion is detected. wherein the second threshold is larger than the first threshold.

[0041] The motion detection may be used for building automation to control sensors and actuators integrated in or co-located with the one or more nodes.

[0042] The node may be integrated in an electronic device, and the electronic device is controlled via RF based motion or presence detection. It may also be possible that the node is connected to or attached to an electronic device to control the electronic device via motion or presence detection. The electronic device may be a lighting device, a wall switch, a bridge, a speaker, a television, a thermostat, a wireless sensor, a power outlet plug, a doorbell, a smart home assistant device, and / or a smart home appliance.

[0043] The node may be further configured to carry out data communication in addition to RF sensing.

[0044] The node may follow a wireless communication standard for sending the sensing messages. The wireless communication standard may be related to a standard of Zigbee, Thread, or Bluetooth Low Energy. For example, the sensing messages may be InterP AN messages according to a Zigbee standard. Alternatively, the sensing messages may be related to Bluetooth beacons according to a Bluetooth Low Energy standard.

[0045] Preferably, the second threshold is larger than the first threshold.

[0046] Beneficially, the first threshold and the second threshold are either preconfigured or determined by the controller adaptively based on the moving average of received signal strength.

[0047] In one example, the controller is further configured to: repeat the steps in processing messages received from the first neighboring node to process messages received from a further neighboring node out of the one or more neighboring nodes; confirm motion detection when motion is detected by processing messages from the first neighboring node and by processing messages from the further neighboring node respectively.

[0048] Advantageously, the controller is further configured to confirm motion detection when motion detection based on messages from the first neighboring node and motion detection based on messages from the further neighboring node are detected within a predetermined time interval.

[0049] The predetermined time interval helps to determine the temporal correlations between the motion detection based on messages from the first neighboring node and motion detection based on messages from the further neighboring node, such that they are triggered by a same event. The predetermined time interval may also be related to the transmission frequency of the messages. For example, the predetermined time interval may be at least the same duration as one cycle of the transmission frequency. A motion detection is confirmed when a further motion event is detected based on a message from a further node in a previous cycle or in a next cycle. In accordance with a third aspect of the invention a wireless network is provided. A wireless network comprising a plurality of nodes configured to perform radio frequency, RF, based motion detection; the wireless network comprising: a node out of the plurality of nodes according to the present invention; and one or more neighboring nodes out of the plurality of nodes with each comprising a transmitter configured to send messages for the node to carry out motion detection.

[0050] Beneficially, one or more nodes out of the plurality of nodes are integrated in or attached to a luminaire.

[0051] The wireless system may be used for both RF sensing and lighting control. The communication within the wireless system is based on a Zigbee network, which is widely adopted in home automation and lighting control applications.

[0052] Beneficially, the motion detection may be used to trigger a control command for lighting control, such as to switch on / off a lamp or to change colour temperature of the lamp, etc.

[0053] In one example, the wireless network is further configured to carry out data communication among the plurality of nodes.

[0054] For example, the messages may be data messages used for both RF sensing and data communication. Alternatively, there are dedicated RF sensing messages for motion detection and data messages for data communication. The RF sensing messages and data messages share the same wireless radio resource in the wireless network.

[0055] BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In the drawings, like reference characters generally refer to the same parts throughout the different figures. Also, the drawings are not necessarily to scale, emphasis instead generally being placed upon illustrating the principles of the invention.

[0057] Fig. 1 illustrates a wireless system comprising a group of nodes according to the present invention;

[0058] Fig. 2 illustrates a basic block diagram of a node;

[0059] Fig. 3 shows an example of ARSSI2observed in all node-node combinations for a plurality of nodes as function of time;

[0060] Fig. 4 shows an example of ARSSI2observed in all node-node combinations for a plurality of nodes by filtering out positive steps; Fig. 5 shows an example of number of node-node combinations reporting a ARSSI2above threshold in the last cycle without filtering; and

[0061] Fig. 6 shows a flow chart of a method according to the present invention.

[0062] DETAILED DESCRIPTION OF EMBODIMENTS

[0063] The embodiments set forth below represent information to enable those skilled in the art to practice the embodiments. Upon reading the following description in light of the accompanying drawings, those skilled in the art will understand the concepts of the disclosure and will recognize applications of these concepts not particularly addressed herein. It should be understood that these concepts and applications fall within the scope of the disclosure.

[0064] FIG. 1 illustrates a wireless system 100 comprising a group of nodes 200a, 200b, 200c, 200d according to the present invention. The wireless system 100 is configured to perform radio frequency, RF, based motion detection with the group of nodes. The group of nodes may belong to a same sensing group, while there is another group of nodes belong to a different sensing group located in the proximity. Hence, the wireless system 100 may comprise a plurality of sensing groups (not shown in the figure) with each sensing group carry out the motion detection either independently or collaboratively, and the division of different sensing groups may be based on locations, such as different groups for nodes located in different rooms or on different floors, different control functions, or different capabilities of nodes. Each node of the group may ignore sensing messages received from nodes belong to a different sensing group. The motion detection may be used for building automation to control sensors and actuators integrated in or co-located with the one or more nodes.

[0065] Each node out of the group may be integrated in, attached to, or connected to an electronic device to be controlled mainly or partially via motion detection. The electronic device may be a lighting device, a wall switch, a bridge, a speaker, a television, a thermostat, a wireless sensor, a power outlet plug, a doorbell, a smart home assistant device, and / or a smart home appliance. For example, a control command to be generated upon motion detection may be related to a turn on or power on command to the one or more electronic devices. The control command may be related to changing a setting or an operation mode of the one or more electronic devices. With one or more nodes out of the group of nodes integrated in or attached to a luminaire, the wireless system may also be a lighting control system and is configured to carry out data communication in addition to RF sensing.

[0066] FIG. 2 illustrates a basic block diagram of a node 200. A node 200 out of a plurality of nodes 200a, 200b, 200c, 200d comprised in a wireless network 100 for radio frequency (RF) based motion detection. As a basic setup, the node comprises at least a receiver 201 and a controller 202. The receiver 201 is configured to receive messages from one or more neighboring nodes. The controller 202 is configured to: o derive, for a first neighboring node out of the one or more neighboring nodes, a moving average of received signal strength of received messages; o calculate a variation in received signal strength of a newly received sensing message from the first neighboring node as compared to the moving average of received signal strength of previously received messages from the first neighboring node; o detect motion when the variation is negative and the magnitude of the variation is larger than a first threshold or when the variation is positive and the magnitude of the variation is larger than a second threshold; and o update for the first neighboring node the moving average of received signal strength with the newly received sensing message when no motion is detected. wherein the second threshold is larger than the first threshold.

[0067] The controller 202 may be further configured to: repeat the steps in processing messages received from the first neighboring node to process messages received from a further neighboring node out of the one or more neighboring nodes; confirm motion detection when motion is detected by processing messages from the first neighboring node and by processing messages from the further neighboring node respectively.

[0068] The controller 202 may be further configured to confirm motion detection when motion detection based on messages from the first neighboring node and motion detection based on messages from the further neighboring node are detected within a predetermined time interval.

[0069] The predetermined time interval helps to establish the temporal correlations between the motion detection based on messages from the first neighboring node and motion detection based on messages from the further neighboring node, such that they are triggered by a same event. The predetermined time interval may also be related to the transmission frequency of the messages. For example, the predetermined time interval may be at least the same duration as one cycle of the transmission frequency. A motion detection is confirmed when a further motion event is detected based on a message from a further node in a previous cycle or in a next cycle.

[0070] Optionally, the node may also comprise a transmitter 203, which is configured to send out messages to other nodes. In one example, the transmitter 203 is configured to send out sensing messages to the neighboring nodes periodically. In another option, the transmitter and receiver may be comprised in a single module, such as a transceiver.

[0071] For large installations where RF sensing is used, many (or even all) nodes are sending sensing messages (InterP AN messages in case of a Zigbee network) in a regular pattern, which results in a significant use of the available bandwidth. This could lead to RF sensing messages being dropped due to collisions (e.g. with messages from neighboring networks) or other messages not being delivered reliably. This in itself is not a large problem, if there are multiple combinations of nodes that send and receive messages from each other. Even if half of the messages would get lost in this way, there will likely be enough transmitter-receiver combinations left that will detect motion.

[0072] However, if messages interfere only partially with each other, or if the interference does not lead to packets that cannot be interpreted by the receiving node, there is the possibility that the RF power of the packet is changed due to the interference with the colliding packet. This situation is known as the ‘hidden terminal’ problem, which can cause the RSSI of the received packet to change while it still possible to interpret the packet correctly.

[0073] Since RF sensing in a Zigbee network relies on the detection of changes in the RSSI over time, such a coincidence of two packets coming from different nodes could result in a change in the RSSI on the receiving node, which would lead to a false positive detection. With long-term measurements in a large RF sensing network such situations are indeed recognized. FIG. 3 shows an example of ARSSI2observed in all node-node combinations for a plurality of nodes as a function of time. In this example, it can be seen that the maximum ARSSI2= (RSSI - RSSImean)2in all channels plotted against time in a parking garage application with 26 nodes (hence 26x25 combinations of transmitter and receiver nodes).

[0074] Most of the time these is a low noise band which corresponds to steps in the received RSSI of 0 or ±1. However, occasionally larger steps up of 2,3 or 4 are observed, leading to ARSSI2of 4, 9 and even 16. If the threshold for detection is not set sufficiently high, these random peaks would lead to false positive detections. It was also observed that these peaks are always caused by positive steps in the RSSI value. After filtering the above results for only those cases where a negative RSSI step occurred, the results are shown in FIG. 4. Thus, it can be concluded that due to the non-coherence between interference and the message, collision between packets always leads to a positive step in the received signal strength.

[0075] It is thus proposed in the present invention to put a check in the ARSSI2-based detection algorithm, which interprets the positive and negative steps in the RSSI differently. This allows for setting the ARSSI2threshold much lower when checking the negative steps as compared to the ARSSI2threshold used when checking the positive steps.

[0076] For example, we could require that small steps in RSSI for triggering a motion detection must be negative, while allowing for larger positive steps in RSSI for triggering a motion detection, e.g. ARSSI > 6 or 7. For example, a practical value for the first threshold for ARSSI2might be 9 (so a step of -3 w.r.t. the running average RSSI), and a practical value for the second threshold might be 36 (a step of +6 w.r.t. the running average RSSI). Sometimes in very noise environments, it may be beneficial to use even higher values for the second threshold (e.g., a step +7 or +8) to reduce the number of false positives due to the hidden node problem.

[0077] Moreover, if the node receives InterP AN messages from all other nodes (as is the case in the distributed implementation), it is not to be expected that the ‘motion event’ is going to be seen in any of the other RS Sis that the node can receive (these will occur at a different time as each node fires its InterP AN in a kind of ‘Round-Robin’ fashion). Therefore, as a secondary measure it is proposed to verify whether a motion event was also detected as a RSSI step in the detection of InterP AN messages from other nodes in the previous cycle or will be detected in the next cycle.

[0078] FIG. 5 shows an example of number of node-node combinations reporting a ARSSI2above threshold in the last cycle without filtering out positive steps. It can be seen that it’s also possible to limit the occurrence of false positive detections in RF sensing due to the ‘hidden terminal’ problem by requiring that any detection that takes place must be seen in the RSSI of at least two transmitting nodes. In a practical embodiment, either of these two measures, or a combination of the two could be used to avoid the occurrence of random spikes in the RSSI resulting in a false positive detection.

[0079] Moreover, the minimum number of node-node combinations could be set at 2, 3 or more if the network is large enough, in order to require more redundancy in the detection before concluding that there is a motion event. Typically, using two pairs of node-node combinations would be sufficient to have a reliable detection.

[0080] FIG. 6 shows a flow chart of a method 500 according to the present invention.

[0081] A method 500 for improving radio frequency, RF, based motion detection in a wireless network 100 comprising a plurality of nodes 200a, 200b, 200c, 200d; the method 500 comprising a node out of the plurality of nodes carrying out steps of: receiving S501 messages from one or more neighboring nodes out of the plurality of nodes 200a, 200b, 200c, 200d; deriving S502, for a first neighboring node out of the one or more neighboring nodes, a moving average of received signal strength of received messages; calculating S503 a variation in received signal strength of a newly received sensing message from the first neighboring node as compared to the moving average of received signal strength of previously received messages from the first neighboring node; detecting S504 motion when the variation is negative and the magnitude of the variation is larger than a first threshold or when the variation is positive and the magnitude of the variation is larger than a second threshold; updating S505 for the first neighboring node the moving average of received signal strength with the newly received sensing message when no motion is detected; wherein the second threshold is larger than the first threshold.

[0082] The method 500 may further comprise the steps of: repeating the steps in processing messages received from the first neighboring node to process messages received from a further neighboring node out of the one or more neighboring nodes; confirming motion detection only when motion is detected by processing messages from the first neighboring node and by processing messages from the further neighboring node respectively.

[0083] The method according to the present invention may be implemented on a computer as a computer implemented method, or in dedicated hardware, or in a combination of both.

Claims

CLAIMS:

1. A method (500) for improving radio frequency, RF, based motion detection in a wireless network (100) comprising a plurality of nodes (200a, 200b, 200c, 200d); the method (500) comprising a node out of the plurality of nodes carrying out steps of: receiving (S501) messages from one or more neighboring nodes out of the plurality of nodes (200a, 200b, 200c, 200d); deriving (S502), for a first neighboring node out of the one or more neighboring nodes, a moving average of received signal strength of received messages; calculating (S503) a variation in received signal strength of a newly received sensing message from the first neighboring node as compared to the moving average of received signal strength of previously received messages from the first neighboring node; detecting (S504) motion when the variation is negative and the magnitude of the variation is larger than a first threshold or when the variation is positive and the magnitude of the variation is larger than a second threshold; updating (S505) for the first neighboring node the moving average of received signal strength with the newly received sensing message when no motion is detected; wherein the second threshold is larger than the first threshold.

2. The method (500) of claim 1, wherein the first threshold and the second threshold are either preconfigured or determined adaptively based on the moving average of received signal strength.

3. The method (500) of claim 1 or 2 further comprising the steps of: repeating the steps in processing messages received from the first neighboring node to process messages received from a further neighboring node out of the one or more neighboring nodes; confirming motion detection only when motion is detected by processing messages from the first neighboring node and by processing messages from the further neighboring node respectively.

4. The method (500) of any one of previous claims 1-2 further comprising the steps of: receiving a message from a further neighboring node out of the one or more neighboring nodes with the message comprising an indication of motion detection by the further neighboring node; confirming motion detection only when motion is detected by processing messages from the first neighboring node and by receiving the message from the further neighboring node respectively.

5. The method (500) of any one of previous claims, wherein the messages are either InterP AN messages according to a Zigbee standard, or Bluetooth beacons according to a Bluetooth Low Energy standard.

6. A node (200) out of a plurality of nodes (200a, 200b, 200c, 200d) comprised in a wireless network (100) for radio frequency, RF, based motion detection; the node comprising: a receiver (201) configured to receive messages from one or more neighboring nodes; and a controller (202) configured to: o derive, for a first neighboring node out of the one or more neighboring nodes, a moving average of received signal strength of received messages; o calculate a variation in received signal strength of a newly received sensing message from the first neighboring node as compared to the moving average of received signal strength of previously received messages from the first neighboring node; o detect motion when the variation is negative and the magnitude of the variation is larger than a first threshold or when the variation is positive and the magnitude of the variation is larger than a second threshold; and o update for the first neighboring node the moving average of received signal strength with the newly received sensing message when no motion is detected.wherein the second threshold is larger than the first threshold.

7. The node (200) of claim 6, wherein the first threshold and the second threshold are either preconfigured or determined by the controller adaptively based on the moving average of received signal strength.

8. The node (200) of claim 6-7, wherein the controller (202) is further configured to: repeat the steps in processing messages received from the first neighboring node to process messages received from a further neighboring node out of the one or more neighboring nodes; confirm motion detection when motion is detected by processing messages from the first neighboring node and by processing messages from the further neighboring node respectively.

9. The node (200) of claim 8, wherein the controller is further configured to confirm motion detection when motion detection based on messages from the first neighboring node and motion detection based on messages from the further neighboring node are detected within a predetermined time interval.

10. A wireless network (100) comprising a plurality of nodes (200a, 200b, 200c, 200d) configured to perform radio frequency, RF, based motion detection; the wireless network (100) comprising: a node (200) out of the plurality of nodes (200a, 200b, 200c, 200d) according to claim 6; and one or more neighboring nodes out of the plurality of nodes (200a, 200b, 200c, 200d) with each comprising a transmitter configured to send messages for the node to carry out motion detection.

11. The wireless network (100) of claim 10, wherein one or more nodes out of the plurality of nodes (200a, 200b, 200c, 200d) are integrated in or attached to a luminaire.

12. The wireless network (100) of claim 10 or 11 is further configured to carry out data communication among the plurality of nodes (200a, 200b, 200c, 200d).

13. A computer program product comprising instructions which, when the program is executed by a controller, cause the controller to control a node according to claim 6 to carry out the steps of the method (500) of any one of claims 1-5.

Images