EP4639785A1

EP4639785A1 - Wireless sensing system with coverage extension using one or more relay nodes

Info

Publication number: EP4639785A1
Application number: EP23706568.5A
Authority: EP
Inventors: Leif Wilhelmsson; Athanasios STAVRIDIS
Original assignee: Telefonaktiebolaget LM Ericsson AB
Current assignee: Telefonaktiebolaget LM Ericsson AB
Priority date: 2022-12-23
Filing date: 2023-02-16
Publication date: 2025-10-29
Also published as: WO2024132224A1

Abstract

A wireless sensing system (400) for detecting changes in a physical environment based on Channel State Information, CSI is provided. The wireless sensing system (400) comprises a number of N network nodes (101, 102, 103, …10N). The first network node(101) is a receiving node configured to detect changes in the physical environment, the i- th network node is a relaying node, i=2, … N-1, and the N-th network node is a transmitting node. The i-th network node is configured to estimate CSI for a channel between the i-th and (i+1)-th network nodes based a signal received from the (i+1)-th network node and send information of the estimated CSI to the (i-1)-th network node. Thefirst network node (101) is configured to estimate CSI for a channel between the first and N-th network nodes via one or more relay nodes from the (N-1)-th network node to the second network node (102) based on the received information of the estimated CSI for the channel between the N-th and the second network nodes and the estimated CSI for the channel between the first and second network nodes, and determine whether there isa change in the physical environment based on the estimated CSI for the channel between the first and N-th network nodes. Publ.

Description

WIRELESS SENSING SYSTEM WITH COVERAGE EXTENSION USING ONE OR

MORE RELAY NODES

TECHNICAL FIELD

Embodiments herein relate to a wireless radio frequency (RF) sensing system and method therein for detecting changes in a physical environment. In particular, they relate to sensing coverage extension of the wireless radio frequency (RF) sensing system.

BACKGROUND

Wireless Radio Frequency (RF) sensing is an emerging market in the communication industry. Currently, RF sensing is under standardization in the ecosystem of the Institute of Electrical and Electronics Engineers IEEE 802.11 in the task group IEEE 802.11 bf which is a new task group on wireless local area network (WLAN) sensing within the IEEE 802.11 working group. In the 3rd Generation Partnership Project (3GPP), discussions on the incorporation of sensing in the evolution of the 5^th generation (5G) communication network also take place. At the same time, significant research is conducted in the field of Joint Communication and Sensing (JCAS) for 6G. The commercial motivation of sensing in communication networks is justified by a large number of use cases which extent from healthcare to security, safety, entertainment, and many others.

A possible approach for sensing is via classical radar measurements. In radar terminology, the terms ‘monostatic’ and ‘bi/m ulti-static’ are used. Monostatic means that the transmission and reception points are the same. That is, the transmit antenna(s) is first used to transmit a sensing signal, i.e. a ‘radar pulse’, the return echo of which is then received by the same antenna(s), or alternatively, the transmission/reception antennas are separate but co-located. Bi/m ulti-static means that the receiving antenna(s) is not the same as the transmitting antenna and that they are in different locations. In classical radar measurements, a mono/bi/m ulti-static radar provides estimates regarding the presence of a target along with its range and radial velocity. Even though the use of radar in Line-of- Sight (LoS) propagation environments is very effective, its deployment in environments with strong Non-Line-of-Sight (NLOS) characteristics becomes challenging. Another approach which has the potential to overcome some of these limitations is based on measurements of Channel State Information (CSI) of the channel between a transmitter and a receiver. In this approach, inference regarding the presence of a target which is not a device, and/or any other quantity of interest is done either through using the CSI directly provided so it is possible to relate different CSI to different events or through the change of CSI at different time instances. In literature, this form of sensing is called Device-Free- Sensing (DFS). Here, the word “free” is connected with the fact that the target is not a device, and it is just present in the physical space of interest.

Similar to classical radar deployments, DFS can be done in monostatic, bistatic, or multi-static setups. In a mono-static setup, a device transmits and receives packets from which it can extract changes of CSI from which sensing can be undertaken. In a bistatic setup, a device transmits packets which are received by another device. The received packets can be used for determining CSI and for tracking the changes of CSI between the two devices. In the multi-static case, initially, one or multiple devices are transmitting packages which are received by one or multiple devices. In the next stage, the receiving devices are forwarding the measured, compressed, or processed CSI to a central device which is able to make inferences about the environment from the changes of the received CSI.

Focusing on IEEE 802.11 bf, the devices involved in a sensing procedure can be Access Points Stations (AP STAs) or non-Access Point Stations (non-AP STAs), hereinafter sometimes simply referred to as APs and STAs, respectively. It is usual that the APs are more capable devices than the STAs in terms of e.g. bandwidth, processing, output power, and number of antennas. In fact, the introduction of Multi-User Multiple- Input Multiple-Output (MU-MIMO) in IEEE 802.1ac and Orthogonal Frequency Division Multiple Access (OFDMA) in IEEE 802.11 ax further differentiated the capabilities of APs and STAs as one AP usually needs to serve multiple STAs. Therefore, it becomes clear that, also in sensing, one will have to deal with environments where the involved devices in a sensing procedure typically have different capabilities. Note that even though the previous argument is presented in terms of the IEEE 802.11 standard, this holds also for most of other modern standards for wireless communication systems.

It is well-known that the performance of a CSI-based, but also any other, sensing procedure depends on the level of Signal-to-Noise-Ratio (SNR) of a received signal. This is the case irrespective of whether the considered scenario is monostatic, bistatic, or multi-static. Thus, due to the signal attenuation during propagation, the coverage area of a sensing procedure is limited to the area where the SNR is sufficient high for sensing which may be significantly smaller than the coverage area for communications. As a quantitative I example, communication may often be possible down to 0 dB SNR using robust modulation and powerful error correcting code, whereas sensing may require as much as 20-30 dB SNR. Consequently, sensing in a wider area, e.g., an area similar to the one where communication is possible, requires new technical solutions that overcome the coverage area limitation. In addition, even in the area where the SNR is sufficiently high, the accuracy of sensing is directly connected with the SNR of the received signal, as higher SNR results in enhanced accuracy. In addition, the resolution and/or accuracy of sensing in this area is not uniform. In fact, the resolution and/or accuracy of sensing decreases as the monitored activity is taking place in regions with lower signal power.

Therefore, technical solutions are needed for coverage extension, sensing resolution and accuracy improvement.

SUMMARY

It is therefore an object of embodiments herein to provide a sensing system and method therein for extending the coverage area and improving the sensing accuracy and resolution. Further, it is clear, for cost purposes, that it is desirable to obtain these improvements by reusing existing infrastructure as far as possible.

According to one aspect of embodiments herein, the object is achieved by a wireless sensing system and method therein for detecting changes in a physical environment based on CSI. The wireless sensing system comprises a number of N network nodes. The first network node is a receiving or sensing node configured to detect changes in the physical environment, an i-th network node is a relaying node, where i=2, ....N-1 , and the N-th network node is a transmitting node and configured to transmit a signal.

The i-th network node is configured to detect if any signal is transmitted from the (i+1 )-th network node. The i-th network node is further configured to estimate CSI for a channel between the i-th and (i+1 )-th network nodes based on a received signal from the (i+1)-th network node and send information of the estimated CSI to the (i-1)-th network node.

The first network node is configured to estimate CSI for the channel between the first and the second network nodes based on a received signal from the second network node.

The first network node 101 is further configured to estimate CSI for the channel between the first and N-th network nodes via one or more relay nodes from the (N-1 )-th network node to the second network node based on the received information of the estimated CSI for the channel between the N-th and the second network nodes and the estimated CSI for the channel between the first and second network nodes.

The first network node is further configured to determine whether there is a change in the physical environment based on the estimated CSI for the channel between the first and N-th network nodes.

According to some embodiments herein, the i-th network node may be configured to send information of the estimated CSI to the (i-1 )-th network node (101) by sending a signal explicitly containing the information of the estimated CSI for the channel between the i-th and (i+ 1 )-th network node or by sending a signal pre-coded by the estimated CSI for the channel between the i-th and (i+1 )-th network node.

In other words, embodiments herein provide an enhanced bistatic sensing system in terms of coverage area and sensing resolution. In more detail, by using existing infrastructure, i.e. by using additional existing network nodes which may or may not belong to the same network as the bistatic sensing system, the sensing coverage area of the bistatic sensing system is increased. Under the assumption of the existence of one or more additional unassociated or associated network nodes in the surrounding area of the bistatic sensing system of interest, Decode-and-Forward (DF) or Estimate-and-Forward (EF) relaying is used in a novel way in the one or more additional network nodes, e.g. in the 2nd to (N-1 )-th network nodes. That is a channel between the first and N-th network nodes may be established via one or more hops by the relay nodes from the (N-1 )-th network node to the 2nd network node. In the first hop, the (N-1 )-th network node does channel estimation and then forwards the channel estimation to the next network node. In the next hop, the network node does decoding, estimation and forwarding. In the following, EF and DF may be used interchangeably.

The i-th network node may be an all-time relaying node, i.e. the i-th network node may run EF/DF operation all the time. The i-th network node may also be a detecting and relaying node, i.e. the i-th network node may detect if any signal is transmitted from the (i+1 )-th network node by measuring a received signal strength and comparing the received signal strength with a threshold. If the received signal strength is larger than the threshold, the i-th network node starts EF/DF operation.

That is, the i-th network node is configured to send information of the estimated CSI to the (i-1 )-th network node either by explicitly signaling or by pre-coding to enhance sensing in terms of coverage area. The signal comprising the information of the estimated CSI is also used for estimating CSI. The sensing network node, i.e. the first network node, will in this way effectively estimate the channel from the sourcing network node, i.e. the N-th network node, to the sensing network node via one or more relaying nodes, so the sensing network node has enlarged the sensing area. In addition, when the additional nodes are associated with the bistatic system of interest, embodiments herein provide a solution based on EF/DF relaying which provides enhanced sensing resolution.

Some advantages of the embodiments herein are, but not limited to: Embodiments herein provide a sensing system with extended sensing coverage area.

Embodiments herein provide a sensing system with improved sensing resolution.

If used for dedicated sensing, rather than for communication, embodiments herein can reduce power consumption of the sensing system since the same coverage and accuracy may be obtained with a significantly reduced transmission power.

Therefore, the embodiments herein provide a sensing system with extended coverage area, improved sensing accuracy and resolution without cause extra cost by reusing existing infrastructure as far as possible.

BRIEF DESCRIPTION OF THE DRAWINGS

Examples of embodiments herein are described in more detail with reference to the attached drawings in which:

Figure 1 is a schematic block diagram illustrating a basic sensing system;

Figure 2 is a schematic block diagram illustrating a sensing system with extended sensing coverage area according to embodiment herein;

Figure 3 is a schematic block diagram illustrating a sensing system with improved sensing resolution according to embodiment herein;

Figure 4 is a schematic block diagram illustrating a sensing system with multi-hop relaying nodes according to embodiment herein; and

Figure 5 is a flow chart illustrating a method performed in a sensing system according to embodiments herein.

DETAILED DESCRIPTION

In a sensing procedure, it is often important to incorporate technical solutions which increase the SNR of a received signal. According to embodiments herein, a wireless sensing system and methods therein are proposed for extending the sensing coverage of a network node where a sensing procedure is based on CSI between a sensing transmitter and a sensing receiver directly or on the temporal changes of CSI between the sensing transmitter and sensing receiver.

Also, the presentation of the disclosure is done in terms of the IEEE 802.11bf standard. However, the extension to other standards is straightforward and considered obvious for a person of ordinary skill in the art.

Figure 1 shows a basic wireless RF sensing system 100. The wireless sensing system 100 comprises a sensing receiver, termed as a first network node 101. The first network node 101 aims to sense any change in the environment by measuring changes of CSI for a channel between itself and a sensing transmitter, termed as a second network node 102. It may be so that the first network node 101 is more capable than the second network node 102. For example, the first network node 101 may be an AP while the second network node 102 may be a STA. The first network node 101 needs to be able to receive, demodulate, and possibly decode a packet sent from the second network node 102. The packet may consist of a short training field (STF) for packet detection, a long training field (LTF) for measured SNR and MIMO channel estimation, and data symbols. It is emphasized that, due to the channel attenuation, the first network node 101 is able to establish a sensing coverage cell restricted in a specific physical area. For the purpose of abstract presentation, in Figure 1, this sensing coverage cell is represented as a circle 110, however, the actual shape and size of this sensing coverage cell depends on the SNR, environment, and the required resolution of the corresponding sensing operation. The first network node 101 senses the environment by conducting measurements on the packets transmitted from the second network node 102. The circle 110 represents the sensing coverage area of the first network node 101 , for which any physical change results in a CSI change which is measurable by the first network node 101.

Clearly, physical movements, presence, or absence of a target TO, which is physically located in the coverage cell 110 of the first network node 101 , can be identified by the first network node 101 via the usual CSI sensing measurements of the following received signal, y_x(t), in the first network node 101 : yi(t) = /ii,₂(t) * %i(t) + Wi(t) (1)

Where, * represents the convolution operator, h₁₂(t) is the channel from the second network node 102 to the first network node 101, x₂(t) is the signal transmitted from the second network node 102, and w^t) is the Gaussian thermal noise of the first network node 101. In contrast, for any other target, such as a target T1 , which is outside the sensing coverage cell 110 of the first network node 101 , has no or non-noticeable influence on the received signal y^t) of the first network node 101. This is the case even when the transmitted signal x₂(0 from the second network node 102 is able to reach T1. Thus, the first network node 101 is unable to perform any sensing on T1.

In order to overcome the sensing coverage limitation, a wireless sensing system 200 according to embodiments herein is proposed and shown in Figure 2, where another node, such as a third network node 103, which is in close proximity with the target T1 and the second network node 102 is included in the wireless sensing system 200. Signal propagation can occur between the second and third network nodes 102 and 103. Here, the third network node 103 may be a STA or an AP. Furthermore, it may be assumed that the third network node 103 has no association with the first and second network nodes 101 and 102, i.e. there is no dedicated signal transmission between the third and first network nodes 103 and 101, and no dedicated signal transmission between the third and second network nodes 103 and 102. Such a scenario is already common, but it will become even more common as the densification of networks will further increase in the future by adding more cell sites within the existing infrastructure to increase the amount of available capacity. As shown in Figure 2, the solid circle 210 represents the sensing coverage area of the first network node 101 , while the dashed circle 220 represents the area of all possible signal propagation paths between the second network node 102 and the third network node 103. Also, no signal propagation path exists between the third network node 103 and the first network node 101. Note that the target TO is removed from Figure 2 in order to simplify the presentation of the current disclosure.

The extension of the sensing coverage area is possible via estimate-and-forward of any signal reception in the second network node 102 from the third network node 103. The third network node 103 may be associated or not with the second network node 102. When the third network node 103 is associated with the second network node 102, sensing may be initiated on demand from the first network node 101 via an appropriated signaling. Whereas, when the third network node 103 is not associated, a sensing procedure can take place only during a transmission of a signal not intended for the second network node, like e.g. a beacon from the third network node 103, or preamble or any other known signal part, such as Long Training Field (LTF) contained in a commonly used packet. When the third network node 103 has no association with the second network node 102, no dedicated signal transmission between these nodes can be expected. However, the third network node 103 could be associated with another AP or STA to which it could transmit packets. For example, in IEEE 802.11, each packet is self-contained, and includes a preamble. A preamble includes predetermined fields which are known to the first network node 101 and the second network node 102, such as the LTF or other training fields. In addition, apart from the dedicated packet transmission, the third network node 103 transmits beacon frames periodically. In this disclosure, it is proposed to use these signal transmissions from the third network node 103 for CSI sensing purposes in the first network node 101.

For example, upon a request from the first network node 101, the second network node 102 may listen for any transmissions from the third network node 103, e.g. a beacon frame, a packet comprising a preamble or any known signal part transmitted to another network node. Once a transmission is detected from the third network node 103, the second network node 102 adopts an Estimate-and-Forward (EF) or Decode-and-Forward (DF) operation and relays the received signal to the first network node 101. In more detail, this operation is expressed as: y₂ (0 = ^2,3 (0 * 3 (0 + 2 (0 (2) where, h_{2 3} (t) is the channel from the third network node 103 to the second network node 102; x₃(t) is the signal transmitted from the third network node 103, and w₂(t) is the Gaussian thermal noise of the second network node 102. The second network node 102 will upon reception decode the signal x₃(t), which in particular implies that the second network node 102 will perform channel estimation in order to perform equalization and demodulation. Normally, when the EF/DF operation is used the signal is then re-encoded and transmitted to the first network node 101 , and the first network node 101 does not need to know anything about the channel conditions, i.e. the CSI for the channel between the third network node 103 and the second network node 102.

However, since the mere purpose is to determine the CSI for the channel between the third network node 103 and the second network node 102, it is essential that this information is relayed to the first network node 101. There are different alternatives to achieve this according to embodiments herein.

According to a first embodiment, the second network node 102 may relay CSI by explicit signaling. The second network node 102 in this case has certain sensing capabilities. According to this embodiment, the second network node 102 will estimate the CSI for the channel between the second network node 102 and the third network node 103. The estimated CSI will be forwarded to the first network node 101 by means of explicit signaling. How the CSI should be estimated is known by the first network node 101. That is, if for example the CSI consists of the amplitude function at a number of positions within the channel bandwidth, the number of points that should be used will be known by the first network node 101.

Also covered by this embodiment is the option that the signalled CSI may be an absolute CSI or a relative change in the CSI compared to an earlier signalled CSI. Specifically, the signalled CSI may be a change compared to the previous signalled CSI.

In addition, whether the absolute CSI or the relative CSI is signalled, as a means to reduce the amount of signaling needed, a threshold may be used to determine whether a detailed CSI signaling is needed or not. Specifically, the second network node 102 may compare the estimated CSI with an earlier estimated CSI and if the difference in CSI is determined to be below a threshold, the second network node 102 can indicate this with a single bit instead of signaling the full CSI. Alternatively, it may simply not send anything at all so that this information is implicitly signalled through the absence of a signal.

Optionally, the second network node 102 may also include information about the estimated quality of the CSI in addition to the CSI itself. As one example, the second network node 102 may also send information to the first network node 101 about what the SNR was when the CSI was determined.

The packet carrying information about the CSI for the channel between the third network node 103 and the second network node 102 is thus transmitted from the second network node 102 to the first network node 101 and used by the first network node 101 for estimating the CSI for the channel between the second network node 102 and the first network node 101. Thus, a characteristic feature of this embodiment is that a packet carrying the information on the CSI for one channel is used for determining CSI for another channel. That is, the same signal comprising information on the estimated CSI for the channel between the third and second network nodes 103 and 102 may be used to estimate the CSI for the channel between the second and first network nodes 102 and 101.

According to a second embodiment, the second network node 102 may relay CSI by means of pre-coding. In this embodiment, the second network node 102 estimates the CSI for the channel between the third and second network nodes 103 and 102. Then, when transmitting a signal or a packet intended for the CSI estimation to the first network node 101 , this signal or packet is pre-coded using the estimated CSI for the channel between the third and second network nodes 103 and 102.

The physical layer is based on OFDM as this allows for simple channel estimation and channel equalization in the frequency domain. When generating a packet or signal to be sent from the second network node 102 to the first network node 101 , the different symbols on the corresponding frequency bins at the input of an Inverse Fast Fourier Transform (IFFT) may be multiplied with the corresponding channel estimate for the channel between the third network node 103 and the second network node 102. In this way, when the first network node 101 is performing channel estimation it will effectively estimate the product of the channel between the third and second network nodes and the channel between the second and first network nodes.

Because the first network node 101 in this way effectively will estimate the channel from the third network node 103 to the first network node 101 via the second network node 102, the first network node 101 has enlarged the sensing area. In case of CSI variations, without any additional information, the first network node 101 will not be able to determine whether the CSI variations are due to variations in the channel between the third and second network nodes or due to variations in the channel between the second and first network nodes or possibly both. However, for many applications the goal may be to detect whether there is any change in the CSI or not and then this restriction is not a problem.

For some scenarios it is feasible to obtain more detailed information from the composite CSI. Specifically, one may e.g. use machine learning and supervised learning where one may perform training when it is known that the channel variations is in one of the four categories:

(1) Channel variations in the channel between the third and second network nodes.

(2) Channel variations in the channel between the second and first network nodes.

(3) Channel variations in the channel between the third and second network nodes as well as in the channel between the second and first network nodes.

(4) No channel variations in either of the two channels, i.e. the channel between the third and second network nodes and the channel between the second and first network nodes. After having done this training, the detected CSI in the first network node 101 may be determined or categorized as belong to one of the four categories above.

It is clear that the above analysis is still valid even if the third and second network nodes 103, 102 are associated. In this case, the EF/DF sensing approach described above, may be initiated on demand from the first network node 101 via the second network node 102. This results in a longer sensing period, and thus more accurate sensing, for the same time interval. This is because more or longer packets may be sent from the third network node 103 and be used by the second network node 102 for sensing, rather than only the preamble or beacon.

According to some embodiments herein, the sensing approaches described above may also apply to a scenario when the first, second and third network nodes 101, 102, 103 belong to the same network. This scenario is shown in Figure 3. In this figure, the solid circle 310 represents the communication coverage area of the first network node 101 , while the dashed circle 320 represents the sensing coverage area of the first network node 101. It is further assumed that the second network node 102 has a stronger channel towards the first network node 101, compared to the corresponding channel between the first and third network nodes 101 , 103. In this case, given that the achieved SNR is sufficiently high, both the AP and the STAs can establish communication links. However, in general, sensing with sufficient resolution requires higher SNR. There might be cases when the third network node 103 is close to the edge of the sensing coverage of the first network node 101 and the second network node 102 is placed somewhere in the middle of the propagation path between the third network node 103 and the first network node 101. In this case, the achieved sensing resolution might not be sufficient if sensing is undertaken using transmissions only from the second network node 102 or the third network node 103.

In this case the first network node 101 may be able to use the CSI for the direct channel between the third network node 103 and the first network node 101, and then in addition use the composite CSI corresponding to the channels between the third and second network nodes 103, 102 plus the channel between the second and first network nodes 102, 101. The two embodiments, relaying the CSI by explicit signaling or by pre-coding described above cover the situation with only one relaying node. The general case with an arbitrary number of relays is a straight-forward extension.

According to some embodiments herein, the approaches described above for extension the sensing coverage area may be extended to multi-hop sensing. Figure 4 shows a multi-hop wireless sensing system 400 according to embodiments herein for detecting changes in a physical environment based on CSI. The wireless sensing system 400 comprises a number of N network nodes 101, 102, 103, 10i, ...10N, wherein the first network node 101 is a receiving node or a sensing receiver configured to detect changes in the physical environment, an i-th network node is a relaying node, where i=2, ....N-1 , and the N-th network node is a transmitting node or sensing transmitter and configured to transmit a signal.

The i-th network node may be configured to detect if any signal is transmitted from the (i+1 )-th network node.

The i-th network node may be a detecting and relaying node, i.e. the i-th network node may detect if any signal is transmitted from the (i+ 1 )-th network node by measuring a received signal strength and comparing the received signal strength with a threshold. If the received signal strength is larger than the threshold, the i-th network node starts EF/DF operation.

The i-th network node may also be an all-time relaying node, i.e. the i-th network node may run EF/DF operation all the time without explicitly detect if there is a signal at its input. If there is no signal at its input, then there will be no signal relayed by the i-th network node.

With all-time relaying, the relaying node can have a simple design, while with detecting and relaying, the relaying node can have energy efficiency.

The i-th network node is further configured to estimate CSI for a channel between the i-th and (i+1 )-th network nodes based on a received signal from the (i+1 )-th network node and send information of the estimated CSI to the (i-1)-th network node.

The i-th network node may be further configured to receive a request from the first network node 101 or from the (i-1 )-th network node to start detecting if any signal is transmitted from the (i+1)-th network node.

The first network node 101 is configured to estimate CSI for the channel between the first and the second network nodes 101 , 102 based on a received signal from the second network node 102. The first network node 101 is further configured to estimate CSI for a channel between the first and N-th network nodes based on the received information of the estimated CSI for the channel between the N-th and the second network nodes and the estimated CSI for the channel between the first and second network nodes. The channel between the first and N-th network nodes is not a direct channel between the first network node and the N-th network node. The channel between the first and N-th network nodes may be established via multiple hops by intermediate or relay nodes from the (N-1 )-th network node to the second network node, i.e., via one or more i-th nodes, i=2,... N-1. In each hop, the intermediate or relay node is configured to perform the operations described above for the i-th network node. That is, the first network node 101 estimates CSI for the channel between the first and N-th network nodes via one or more relay nodes from the (N-1 )-th network node to the second network node based on the received information of the estimated CSI for the channel between the N-th and the second network nodes and the estimated CSI for the channel between the first and second network nodes.

The first network node 101 is further configured to determine whether there is a change in the physical environment based on the estimated CSI for the channel between the first and N-th network nodes.

The number of N network nodes may belong to the same network, i.e. all the network nodes are associated. Then the first network node 101 may be further configured to determine whether there is a change in the physical environment based on the estimated CSI for a direct channel between the first and N-th network nodes in addition to the estimated CSI for the channel between the first and N-th network nodes via one or more relay nodes from the (N-1 )-th network node to the second network node 102.

The first network node 101 to the (N-1 )-th network node may be associated, while the N-th network node 10N may be not associated. The solid circle 411 represents the sensing coverage cell of the first network node 101 , while the i-th circle 41 i, where i=2, ... .N-1 ,, represents the region from which the i-th network node can decode-and-forward to the (i-1 )-th network node a received signal from the (i+1 )-th network node. A sensing procedure in the coverage area of the first network node to the (N- 1 )-th network node may be initiated and undertaken at any time. Whereas a sensing procedure in the N-th network node may be initiated at any time from the first network node 101 , it can be completed only after the independent signal transmission of a known signal such as a beacon or LTF, from the N-th network node 10N. According to some embodiments herein, the N-th network node may be configured to transmit a signal upon a request received from the first network node 101 or any of the i-th network node, i=2,... N-1.

Regarding the first embodiment of relaying CSI by explicit signaling, this may simply be extended to send a signal that either contains explicit information of the estimated CSI of each one of the individual channels or contains explicitly information of the estimated CSI of all channels which are condensed or combined such that all the channels are treated as a single channel.

Regarding the second embodiment of relaying CSI by pre-coding, the channel variations for the individual channels may be concatenated and can be detected by the first network node 101. Also, here the approach for trying to identify more details for which one(s) of the channels that its CSI has changed if the composite CSI has changed can be based on machine learning in a similar fashion as is described above for the case with only one relaying node. Naturally, as the number of channels increases the number of classes will grow quickly.

Therefore, according to some embodiments herein, the first network node 101 may be further configured to determine whether channel variations are in any of the channels between the first network node 101 and the second network node 102, between the i-th network node and the (i+1 )-th network node, based on the estimated CSI for the channel between the first and N-th network nodes by machine learning.

A method performed in the wireless sensing system 200, 300, 400 for detecting changes in a physical environment based on CSI according to embodiments herein will be describe with reference to Figure 5.

As described above with reference to Figure 4, the wireless sensing system 200, 300, 400 may comprise a number of N network nodes 101, 102, 103, ... 10N, e.g. N=3. The first network node 101 is a receiving node configured to detect changes in the physical environment. The i-th network node is a relaying node, i=2,... N-1 , and the N-th network node is a transmitting node and configured to transmit a signal. The method comprises the following actions:

Action 510

The i-th network node 10i detects, if any signal is transmitted from the (i+1)-th network node. For example, when N=3, there are only three network nodes 101 , 102, 103 as shown in Figures 2 and 3, it is the second network node 102 detects if any signal is transmitted from the 3rd network node 103. The signal transmitted from the (i+1 )-th network node may be a packet comprising a preamble or any known signal part transmitted to another network node or a beacon frame if the (i+1)-th network node is not associated with the i-th network node.

The signal transmitted from the (i+1 )-th network node may be a dedicated packet transmission if the (i+1)-th network node is associated with the i-th network node.

If the N-th network node 10N is not associated with the (N-1 )-th network node, the signal transmitted from the N-th network node 10N may be a packet comprising a preamble or any known signal part transmitted to another network node or a beacon frame.

If the N-th network node 10N is associated with the (N-1 )-th network node, the signal transmitted from the N-th network node 10N is a dedicated packet transmission to the (N-1 )-th network node.

Action 520

The i-th network node 10i estimates CSI for a channel between the i-th and (i+1 )-th network nodes based on a received signal from the (i+ 1 )-th network node. For example, when N=3, the second network node 102 estimates CSI for a channel between the second and third network nodes based on a received signal from the third network node 103.

Action 530

The i-th network node 10i sends information of the estimated CSI for the channel between the i-th and (i+ 1 )-th network nodes to the (i-1 )-th network node. For example, when N=3, the second network node 102 sends information on the estimated CSI for the channel between the second and third network node to the first network node 101.

According to some embodiments herein, the information of the estimated CSI for the channel between the i-th and (i+1 )-th network nodes may comprise a relative change in a currently estimated CSI compared to a previously estimated CSI or an absolute value of the estimated CSI for the channel between the i-th and (i+ 1 )-th network nodes.

According to some embodiments herein, the information of the estimated CSI for the channel between the i-th and (i+1 )-th network nodes may further comprise information of the estimated quality of the CSI. For example, the i-th network node may also send information to the (i-1 )-th network node about what the SNR was when the CSI was determined.

According to some embodiments herein, to reduce the amount of signaling needed, the i-th network node may determine how to send the information of the estimated CSI based on whether a difference between a currently estimated CSI and a previously estimated CSI is above a predefined threshold. The predefined threshold may be used to determine whether a detailed CSI signaling is needed or not. Specifically, the i-th network node may compare the estimated CSI with an earlier estimated CSI and if the difference in CSI is determined to be below the predefined threshold, the i-th network node can indicate this with a single bit instead of signaling the full CSI.

According the first embodiment of relaying CSI by explicitly signaling, the i-th network node 10i may send a signal explicitly containing the information of the estimated CSI for the channel between the i-th and (i+1 )-th network node. That is the estimated CSI may be explicitly signaled from the i-th network node to the (i-1 )-th network node. The (i- 1 )-th network node then obtains the estimated CSI by demodulating the signal received from the i-th network node. For example, when N=3, the estimated CSI for the channel between the second and third network node may be explicitly signaled from the second network node 102 to the first network node 101. The first network node 101 then obtains the estimated CSI by demodulating the signal received from the second network node 102.

According to the second embodiment of relaying CSI by pre-coding, the i-th network node 10i may send a signal pre-coded by the estimated CSI for the channel between the i-th and (i+1 )-th network node. For example, the signal from the second network node 102 may be pre-coded by the estimated CSI for the channel between the second and third network node. The first network node 101 then obtains the estimated CSI for the channel between the second and third network nodes when estimating the channel between the first and the second network nodes using the received signal from the second network node 102.

Therefore, according to some embodiments herein, the i-th network node 10i may be further configured to generate a signal or packet pre-coded using the estimated CSI for the channel between the i-th and (i+1 )-th network nodes, and the i-th network node 10i is configured to send information of the estimated CSI to the (i-1 )-th network node by being configured to send the pre-coded packet or signal.

The pre-coded packet may be generated by multiplying different symbols on the corresponding frequency bins at an input of inverse discrete Fourier transform, IFFT, with the corresponding estimated CSI for the channel between the ith and (i+ 1 )th network nodes. For example, in the i-th node, the generated packet, i.e. OFDM symbol is given as: st = hj O x, where s is the transmitted packet, i.e. the OFDM symbol in the frequency domain, hj is the estimated CSI for the channel between the i-th node and the (i+1 )-th node in the frequency domain, is the reference signals in the frequency domain, and O is element- wise multiplication. That is the generated packet is the element-wise (Hadamard product) product of the estimated CSI and reference signals.

Action 540

The first network node 101 estimates CSI for a channel between the first and the second network nodes based on a received signal from the second network node 102. The signal from the second network node 102 may be pre-coded by the composite CSI for all channels between the i-th and (i+1 )-th network node from the N-th network node to the second network node. The signal from the second network node 102 may also contains explicitly the composite CSI for all channels between the i-th and (i+1 )-th network node from the N-th network node to the second network node.

Action 550

The first network node 101 estimates CSI for a channel between the first and N-th network nodes based on the received information of the estimated CSI for the channel between the N-th and the second network nodes and the estimated CSI for the channel between the first and second network nodes. Since there is no direct channel from the N- th network node to the first network node 101 , the channel between the first and N-th network nodes is established via hopping or relaying by one or more intermediate or relay nodes, i.e., the one or more i-th network nodes, from the (N-1 )-th network node to the second network node. In each hopping or relaying, the intermediate or relay node is configured to perform the actions 510-530 described above for the i-th network node. That is the first network node 101 estimates CSI for the channel between the first and N-th network nodes via one or more relay nodes from the (N-1 )-th network node to the second network node based on the received information of the estimated CSI for the channel between the N-th and the second network nodes and the estimated CSI for the channel between the first and second network nodes.

For example, when N=3, the first network node 101 estimates CSI for a channel between the first and third network nodes via the second network node 102 based on the received information of the estimated CSI for the channel between the second and third network nodes and the estimated CSI for the channel between the first and the second network nodes. If the estimated CSI is explicitly signaled from the second network node 102 to the first network node 101 , the first network node 101 may obtain the estimated CSI by demodulating the signal received from the second network node 102. The first network node 101 may also estimate the CSI for the channel between the first and second nodes based on the received signal from the second network node 102.

If the signal from the second network node 102 is pre-coded by the estimated CSI for the channel between the second and third network node, the first network node 101 may obtain the estimated CSI for the channel between the second and third network nodes when estimating the channel between the first and the second network nodes using the received signal from the second network node 102.

Then the network node 101 estimates CSI for the channel between the first and third network nodes via the second network node 102 based on the estimated CSI for the channel between the first and second network nodes and the estimated CSI for the channel between the second and third network nodes.

Action 560

The first network node 101 determines whether there is a change in the physical environment based on the estimated CSI for the channel between the first and N-th network nodes. For example, when N=3, the first network node 101 determines whether there is a change in the physical environment based on the estimated CSI for the channel between the first and third network nodes via the second network node 102.

To perform the method in the network nodes 101, 102,... 10i, ... 10N, the network node 101/10i/1 ON comprises modules as shown in Figure 6. The network node 101/10i/1 ON comprises a receiving module 610, a transmitting module 620, a determining module 630, a processing module 640, a memory 650 etc.

The network nodes 101, 102,... 10i, ... 10N are configured to perform corresponding method Actions 510-560 described above.

The i-th network node is configured to, by means of e.g., the receiving module 610 and determining module 630 being configured to, detect if any signal is transmitted from the (i+1 )-th network node.

The i-th network node is further configured to, by means of e.g., the processing module 640 being configured to, estimate CSI for a channel between the i-th and (i+1 )-th network nodes based on a received signal from the (i+ 1 )-th network node. The i-th network node is further configured to, by means of e.g., the transmitting module 620 being configured to, send information of the estimated CSI to the (i-1)-th network node.

The first network node 101 is configured to, by means of e.g., the processing module 640 being configured to, estimate CSI for the channel between the first and the second network nodes 101, 102 based on a received signal from the second network node 102.

The first network node 101 is further configured to, by means of e.g., the processing module 640 being configured to, estimate CSI for the channel between the first and N-th network nodes based on the received information of the estimated CSI for the channel between the N-th and the second network nodes and the estimated CSI for the channel between the first and second network nodes.

The first network node 101 is further configured to, by means of e.g., the determining module 630 being configured to, determine whether there is a change in the physical environment based on the estimated CSI for the channel between the first and N- th network nodes.

The method according to embodiments herein may be implemented through one or more processors together with computer program code for performing the functions and actions of the embodiments herein. The program code mentioned above may also be provided as a computer program product, for instance in the form of computer readable medium or a data carrier 680 carrying computer program code 670, as shown in Figure 6, for performing the embodiments herein when being loaded into the network node 101/10i/10N. One such carrier may be in the form of a CD ROM disc. It is however feasible with other data carriers such as a memory stick. The computer program code may furthermore be provided as pure program code on a server or a cloud and downloaded to the network node 101/10i/1 ON.

When using the word "comprise" or “comprising” it shall be interpreted as nonlimiting, i.e. meaning "consist at least of".

The embodiments herein are not limited to the above described preferred embodiments. Various alternatives, modifications and equivalents may be used. Therefore, the above embodiments should not be taken as limiting the scope of the invention, which is defined by the appended claims.

Claims

1 . A wireless sensing system (200, 300, 400) for detecting changes in a physical environment based on Channel State Information, CSI, wherein the wireless sensing system (200, 300, 400) comprises a number of N network nodes (101 , 102, 103, ... 10N), and wherein a first network node (101) is a receiving node configured to detect changes in the physical environment, an i-th network node is a relaying node, i=2, ... N-1 , and an N-th network node is a transmitting node, and wherein the i-th network node is configured to: detect if any signal is transmitted from the (i+ 1 )-th network node; estimate CSI for a channel between the i-th and (i+1 )-th network nodes based on a received signal from the (i+1)-th network node; send information of the estimated CSI to the (i-1 )-th network node; and the first network node (101) is configured to: estimate CSI for a channel between the first and the second network nodes based on a received signal from the second network node (102); estimate CSI for a channel between the first and N-th network nodes via one or more relay nodes from the (N-1 )-th network node to the second network node (102) based on the received information of the estimated CSI for the channel between the N-th and the second network nodes and the estimated CSI for the channel between the first and second network nodes; determine whether there is a change in the physical environment based on the estimated CSI for the channel between the first and N-th network nodes.

2. The wireless sensing system (200, 300, 400) according to claim 1 , wherein the i-th network node (102) is configured to send information of the estimated CSI to the (i- 1 )-th network node (101) by sending a signal explicitly containing the information of the estimated CSI for the channel between the i-th and (i+ 1 )-th network node.

3. The wireless sensing system (200, 300, 400) according to claim 1 , wherein the i-th network node (1 Oi) is configured to send information of the estimated CSI to the (i- 1 )-th network node (101) by sending a signal pre-coded by the estimated CSI for the channel between the i-th and (i+1 )-th network node.

4. The wireless sensing system (200, 300, 400) according to any one of claims 1-3, wherein the first to (N-1 )-th network nodes are associated and the N-th network node (10N) is not associated with the (N-1 )-th network node, and the signal transmitted from the N-th network node (10N) is a packet comprising a preamble or any known signal part transmitted to another network node or a beacon frame.

5. The wireless sensing system (200, 300, 400) according to according to any one of claims 1-3, wherein the N-th network node (10N) is associated with the (N-1 )-th network node, and the signal transmitted from the N-th network node (10N) is a dedicated packet transmission to the (N-1 )-th network node.

6. The wireless sensing system (200, 300, 400) according to any one of claims 1-5, wherein the information of the estimated CSI for the channel between the i-th and (i+1 )-th network nodes comprises a relative change in a currently estimated CSI compared to a previously estimated CSI.

7. The wireless sensing system (200, 300, 400) according to any one of claims 1-5, wherein the information of the estimated CSI for the channel between the i-th and (i+1 )-th network nodes comprises an absolute value of the estimated CSI for the channel between the i-th and (i+1 )-th network nodes.

8. The wireless sensing system (200, 300, 400) according to any one of claims 1-5, wherein the i-th network node is further configured to determine how to send the information of the estimated CSI based on whether a difference between a currently estimated CSI and a previously estimated CSI is above a predefined threshold.

9. The wireless sensing system (200, 300, 400) according to any one of claims 1-8, wherein the information of the estimated CSI for the channel between the i-th and (i+1 )-th network nodes further comprises information of the estimated quality of the CSI.

10. The wireless sensing system (200, 300, 400) according to any one of claims 1-9, wherein the i-th network node (1 Oi) is further configured to generate a signal precoded using the estimated CSI for the channel between the i-th and (i+1 )-th network nodes.

11. The wireless sensing system (200, 300, 400) according to any one of claims 1-10, wherein the number of N network nodes belong to the same network.

12. The wireless sensing system (200, 300, 400) according to claim 11 , wherein the first network node is further configured to determine whether there is a change in the physical environment based on the estimated CSI for a direct channel between the first and N-th network nodes in addition to the estimated CSI for the channel between the first and N-th network nodes via one or more relay nodes from the (N- 1 )-th network node to the second network node (102).

13. The wireless sensing system (200, 300, 400) according to any one of claims 1-12, wherein the i-th network node is further configured to receive a request from the first network node (101) or from the (i-1 )-th network node to start detecting if any signal is transmitted from the (i+1 )-th network node.

14. The wireless sensing system (200, 300, 400) according to any one of claims 1-13, wherein the first network node (101) is further configured to determine whether channel variations are in any of the channels between the first network node (101) and the second network node (102), between the i-th network node (102) and the (i+1 )-th network node, based on the estimated CSI for the channel between the first and N-th network nodes by machine learning.

15. The wireless sensing system (200, 300, 400) according to any one of claims 1-14, wherein the N-th network node is configured to transmit a signal upon a request received from the first network node (101) or any of the i-th network node, i=2, ... N- 1.

16. The wireless sensing system (200, 300, 400) according to any one of claims 1-15, wherein N=3 and i=2.

17. A method performed in a wireless sensing system (200, 300, 400) for detecting changes in a physical environment based on Channel State Information, CSI, wherein the wireless sensing system (200, 300, 400) comprises a number of N network nodes (101 , 102, 103, ... 10N), and wherein a first network node (101) is a receiving node configured to detect changes in the physical environment, an i-th network node is a relaying node, i=2,... N-1 , and an N-th network node is a transmitting node and configured to transmit a signal, the method comprising: detecting (510) by the i-th network node, if any signal is transmitted from the (i+1 )-th network node; estimating (520) by the i-th network node, CSI for a channel between the i-th and (i+1 )-th network nodes based on a received signal from the (i+1)-th network node; sending (530) by the i-th network node, information of the estimated CSI to the (i-1 )-th network node; estimating (540) by the first network node, CSI for a channel between the first and the second network nodes based on a received signal from the second network node (102); estimating (550) by the first network node, CSI for a channel between the first and N-th network nodes via one or more relay nodes from the (N-1 )-th network node to the second network node (102) based on the received information of the estimated CSI for the channel between the N-th and the second network nodes and the estimated CSI for the channel between the first and second network nodes; determining (560) by the first network node, whether there is a change in the physical environment based on the estimated CSI for the channel between the first and N-th network nodes.