Communication apparatus and communication method

Abstract

The invention provides a communication apparatus and a communication method. A communication device is a control unit and includes a communication unit that generates a frame including STA information, which is information about STAs, at least one of the STA information being probe NDP information for AI / ML, which is used for artificial intelligence / machine learning, and the NDP being a null data PPDU, which is a null data physical layer convergence process protocol data unit. The communication unit transmits the generated frame.

Description

Technical Field

[0001] The present disclosure relates to a communication device and a communication method. Background Art

[0002] The industry and academia are researching technologies for wireless LAN (Local Area Network) for artificial intelligence (hereinafter referred to as AI) or machine learning (hereinafter referred to as ML) (for example, Non-Patent Document 1).

[0003] In the IEEE (the Institute of Electrical and Electronics Engineers) 802.11 working group, a TIG (Topic Interest Group) related to AI / ML (hereinafter also referred to as AIML) was launched in July 2022 and research has begun (for example, Non-Patent Document 2).

[0004] As one of the technologies being researched for applications in wireless LAN, research on CSI compression aimed at reducing the feedback amount of channel state information (hereinafter referred to as CSI) is underway (for example, Non-Patent Document 3).

[0005] Prior Art Documents

[0006] Non-Patent Documents

[0007] Non-Patent Document 1: S. Szott, K. Kosek-Szott, P. Gawlowicz, J. T. Gomez, B. Bellalta, A. Zubow, F. Dressler, “WiFi Meets ML: A Survey on Improving IEEE 802.11 Performance with Machine Learning” in IEEE Communications Surveys & Tutorials, vol. 24, no. 3, pp. 1843 - 1893

[0008] Non-Patent Document 2: IEEE 802.11 - 22 / 847r3, AIML TIG July 2022 Agenda

[0009] Non-Patent Document 3: IEEE 802.11-22 / 950r2, Discussion on Connection between AI / ML&Wireless LAN

[0010] Non-Patent Document 4: M. Deshmukh, Z. Lin, H. Lou, M. Kamel, R. Yang, I. Guvenc, “Intelligent Feedback Overhead Reduction (iFOR) in Wi-Fi 7 and Beyond,” in Proceedings of 2022 VTC-Spring

[0011] Non-Patent Document 5: P. K. Sangdeh, H. Pirayesh, A. Mobiny, H. Zeng, “LB-SciFi: Online Learning-Based Channel Feedback for MU-MIMO in Wireless LANs,” in Proceedings of 2020 IEEE 28th ICNP Summary of the Invention

[0012] The NDPA (NDP Announcement) frame for specifying the probe null data packet (NDP) (null data PPDU (Physical Layer Convergence Procedure (PLCP) Protocol Data Unit (PPDU))) specification for AI / ML-oriented measurements has not been studied yet.

[0013] A communication device according to an embodiment of the present disclosure is a control unit and has a communication unit. The control unit generates a frame including information of each STA (also referred to as a station, terminal, or node), i.e., each STA information (STA Info), and at least one STA information among the each STA information is probe NDP information for artificial intelligence / machine learning, i.e., for AI / ML. The NDP is a null data physical layer convergence procedure protocol data unit, i.e., a null data PPDU. The communication unit transmits the generated frame.

[0014] In addition, these broad or specific embodiments can be implemented by a system, a device, a method, an integrated circuit, a computer program, or a recording medium, or can be implemented by any combination of a system, a device, a method, an integrated circuit, a computer program, and a recording medium.

[0015] A station (STA) can specify a probe NDP format for AI / ML for each STA. In addition, the STA can specify a mixture of multiple probe NDP formats. Brief Description of the Drawings

[0016] Figure 1 It is a diagram showing a non-trigger-based probing process example for AI / ML.

[0017] Figure 2 It is a diagram showing a trigger-based probing process example for AI / ML.

[0018] Figure 3 It is a diagram showing an interface example of the AIML measurement establishment phase.

[0019] Figure 4 It is a diagram showing an interface example of the non-trigger-based AIML measurement phase.

[0020] Figure 5 It is a diagram showing an interface example of the trigger-based AIML measurement phase.

[0021] Figure 6 It is a diagram showing an interface example of the AIML measurement termination phase.

[0022] Figure 7 It is a diagram showing the high-efficiency (HE) NDP announcement frame format.

[0023] Figure 8 It is a diagram showing the sounding dialog token.

[0024] Figure 9 It is a diagram showing the STA information field format in the extremely high throughput (EHT) NDP announcement frame.

[0025] Figure 10 It is a diagram showing the STA information field format in the AIML NDP announcement frame.

[0026] Figure 11 It is a diagram showing the NDP announcement frame variant encoding.

[0027] Figure 12 It is a diagram showing the AID11 subfield encoding in the NDP announcement frame.

[0028] Figure 13 It is a diagram showing the AIML type subfield encoding.

[0029] Figure 14 It is a diagram showing an example of AIML sounding NDP information in the case of AIML type = 0 (index-based CSI).

[0030] Figure 15This is a diagram showing an example of AIML sounding NDP information in the case where the AIML type = 1 (CSI Compressed Feedback Scheme).

[0031] Figure 16 This is a diagram showing an example of the format of an NDPA frame for AIML.

[0032] Figure 17 This is a diagram showing an example of the definition of the type value subfield and subtype value subfield of the Frame Control field.

[0033] Figure 18 This is a diagram showing an example of the definition of the Control Frame Extension subfield of the Frame Control field.

[0034] Figure 19 This is a diagram showing an example of the definition of the NDP announcement type.

[0035] Figure 20 This is a diagram showing an example of the format of a trigger frame for AIML.

[0036] Figure 21 This is a diagram showing an example of the structure of the AIML Sounding NDP Information Element.

[0037] Figure 22 This is a diagram showing the structure of the STA.

[0038] Figure 23 This is a diagram showing the structure of another communication device. Detailed implementation

[0039] According to an embodiment of the present disclosure, a node (STA) can specify and notify other nodes (STA) of subfields (AIML sounding NDP information) that define the format of the sounding NDP for AI / ML.

[0040] Figure 1 and Figure 2 shows an example of a sounding process for AI / ML. Figure 1 The process example of shows a non-Trigger-based (non-TB) sounding process example. Figure 2The process illustrates a trigger-based (TB) detection process example. These process examples are all composed of three phases: the AIML Measurement Setup Phase, the AIML Measurement Phase, and the AIML Measurement Termination Phase.

[0041] The AIML Measurement Setup Phase is the phase for setting up detections for AI / ML. The AIML Measurement Phase is the phase for sending NDPA (NDP advertisement) and performing measurements using NDP. The AIML Measurement Termination Phase is the phase for terminating the detection setup for AI / ML. In Figure 1 and Figure 2 Regarding the AIML Measurement Termination Phase, two processes are shown, namely, the process starting from the AIML initiating STA and the process starting from the AIML peer STA. In the AIML Measurement Termination Phase, only one of the processes needs to be used.

[0042] The SME (station management entity: terminal management entity) represents the terminal management function, and the MLME (MAC (Media Access Control) sublayer management entity) represents the MAC layer management function.

[0043] The AIML initiating STA is the STA that indicates detections for AIML, and the AIML peer STA is the STA that receives the indication of detections for AIML. The AIML initiating STA can perform the Figure 1 or Figure 2 described detection process for AI / ML.

[0044] Figures 3 - 6 Shows Figure 1 and Figure 2 the interface names (frame names) and functions in each of the phases of the detection process examples.

[0045] Describe the process of the AIML Measurement Setup Phase.

[0046] First, the SME of the AIML initiator STA sends an MLME-AIMLMSMTSETUP.request (S100, S200) to the MLME of the AIML initiator STA. MLME-AIMLMSMTSETUP.request is a frame requesting to send an AIML Measurement Setup Request frame to the peer STA.

[0047] The MLME of the AIML initiator STA that receives the MLME-AIMLMSMTSETUP.request sends an AIML Measurement Setup Request frame (S101, S201) to the MLME of the AIML peer STA. The AIML Measurement Setup Request frame is a frame indicating the settings of the AIML probe to the peer STA.

[0048] The MLME of the AIML peer STA that receives the AIML Measurement Setup Request frame sends an MLME-AIMLMSMTSETUP.indication (S102, S202) to the SME of the AIML peer STA. MLME-AIMLMSMTSETUP.indication is a frame indicating that the AIML Measurement Setup Request frame has been received.

[0049] The SME of the AIML peer STA that receives the MLME-AIMLMSMTSETUP.indication sends an MLME-AIMLMSMTSETUP.response (S103, S203) to the MLME of the AIML peer STA. MLME-AIMLMSMTSETUP.response is a response signal to the MLME-AIMLMSMTSETUP.indication and is a frame requesting the transmission of an AIML Measurement Setup Response frame.

[0050] The MLME of the AIML peer STA that receives the MLME-AIMLMSMTSETUP.response sends an AIML Measurement Setup Response frame (S104, S204) to the MLME of the AIML initiator STA. The AIML Measurement Setup Response frame is a frame that responds to the AIML initiator STA by indicating whether the peer STA has accepted the AIML probe settings.

[0051] The MLME of the AIML initiating STA that has received the AIML measurement establishment response frame sends MLME - AIMLMSMTSETUP.confirm (S105, S205) to the SME of the AIML initiating STA. MLME - AIMLMSMTSETUP.confirm is a signal indicating that the AIML measurement establishment response frame has been received.

[0052] The process of the AIML measurement phase will be described. Regarding the AIML measurement phase, Figure 1 the non - trigger - based AIML measurement phase shown Figure 2 has a different process from the trigger - based AIML measurement phase shown. First, the process of Figure 1 the non - trigger - based AIML measurement phase shown will be described.

[0053] First, the SME of the AIML initiating STA that has received MLME - AIMLMSMTSETUP.confirm in the AIML measurement establishment phase sends MLME - AIMLMSMTRQ.request (S110) to the MLME of the AIML initiating STA. MLME - AIMLMSMTRQ.request is a frame requesting the transmission of non - trigger - based AIML probes for the peer STA. The AIML probes are performed using, for example, NDPA and NDP.

[0054] The MLME of the AIML initiating STA that has received MLME - AIMLMSMTRQ.request sends NDPA (S111) to the MLME of the AIML peer STA. NDPA is a frame specifying the format of the probe NDP. Also, the MLME of the AIML initiating STA sends an NDP (S112) corresponding to the content specified by NDPA to the MLME of the AIML peer STA. NDP is a probe NDP frame. By specifying the probe NDP format for AI / ML for each STA in NDPA, it is possible to send NDP for AI / ML to each STA, and at each node, the received NDP can be used for measurement.

[0055] The MLME of the AIML peer STA that has received the NDP sends MLME - AIMLREPORT.indication (S113) to the SME of the AIML peer STA. MLME - AIMLREPORT.indication is a frame indicating that measurement has been performed using the probe NDP. Also, the MLME of the AIML peer STA sends the NDP (S114) to the MLME of the AIML initiating STA.

[0056] The MLME of the AIML initiator STA that has received the NDP sends an MLME-AIMLREPORT.indication (S115) to the SME of the AIML initiator STA. The MLME-AIMLREPORT.indication is a frame indicating that measurements have been made using non-trigger-based sounding NDPs.

[0057] The SME of the AIML peer STA that has received the MLME-AIMLREPORT.indication sends an MLME-AIMLREPORTRQ.request (S116) to the MLME of the AIML peer STA. The MLME-AIMLREPORTRQ.request is a frame requesting the transmission of an AIML Measurement Report frame.

[0058] The MLME of the AIML peer STA that has received the MLME-AIMLREPORTRQ.request sends an AIML measurement report frame (S117) to the MLME of the AIML initiator STA. The AIML measurement report frame is a frame reporting the measurement results of AIML sounding to the other STA. In addition, the MLME of the AIML peer STA sends an MLME-AIMLREPORTRQ.confirm (S119) to the SME of the AIML peer STA. The MLME-AIMLREPORTRQ.confirm is a frame indicating that the AIML measurement report frame has been sent.

[0059] The MLME of the AIML initiator STA that has received the AIML measurement report frame sends an MLME-AIMLMSMTRQ.confirm (S118) to the SME of the AIML initiator STA.

[0060] The SME of the AIML initiator STA that has received the MLME-AIMLMSMT.indication sends an MLME-AIMLREPORTRQ.request (S120) to the MLME of the AIML initiator STA.

[0061] The MLME of the AIML initiator STA that has received the MLME-AIMLREPORTRQ.request sends an AIML measurement report frame (S121) to the MLME of the AIML peer STA.

[0062] The MLME of the AIML peer STA that has received the AIML measurement report frame sends an MLME-AIMLREPORTRQ.confirm (S122) to the SME of the AIML peer STA.

[0063] Figure 1 Examples are shown in which the MLME notifies the measurement results to the SME via MLME - AIMLREPORT.indication, and in which the MLME notifies the received measurement result report to the SME via MLME - AIMLMSMTRQ.confirm. However, the measurement results can also be notified by an external application or a higher layer.

[0064] Figure 1 Examples are shown in which the measurement results are sent to the peer STA. However, it can also be that no sending is performed and an action using the measurement results is carried out, or it can be that both sending to the peer STA and an action using the measurement results at the sending - source STA are performed. It can be that the SME or the MLME includes an application that performs an action using the measurement results.

[0065] Next, Figure 2 the process of the trigger - based AIML measurement phase shown is described.

[0066] First, the SME of the AIML initiator STA that has received MLME - AIMLMSMTSETUP.confirm in the AIML measurement setup phase sends MLME - AIMLTBMSMTRQ.request to the MLME of the AIML initiator STA (S210). MLME - AIMLTBMSMTRQ.request is a frame that requests a trigger - based AIML probe (transmission of NDPA or NDP) for the peer STA. The AIML probe is performed using, for example, NDPA and NDP.

[0067] The MLME of the AIML initiator STA that has received MLME - AIMLTBMSMTRQ.request sends an AIML poll frame to the MLME of the AIML peer STA (S211). The AIML poll frame is a frame that polls for an AIML probe for the peer STA.

[0068] The MLME of the AIML peer STA that has received the AIML poll frame sends a clear - to - self (CTS - to - self) to the MLME of the AIML initiator STA (S212). The clear - to - self is a response frame to the AIML poll frame.

[0069] When the MLME of the AIML initiating STA that has received the MLME - AIMLTBMSMTRQ.request receives the self - permission to send from the AIML peer MLME, it sends an AIML NDPA frame (NDPA) to the MLME of the AIML peer STA that sent the self - permission to send (S213). The AIML NDPA frame is a frame that specifies the format of the AIML - used probe NDP.

[0070] Also, the MLME of the AIML initiating STA that has received the MLME - AIMLTBMSMTRQ.request and the self - permission to send sends an NDP corresponding to the content specified by the AIML NDPA frame to the MLME of the AIML peer STA (S214). The NDP is a probe NDP frame.

[0071] In addition, the MLME of the AIML initiating STA that has received the MLME - AIMLTBMSMTRQ.request and the self - permission to send sends an AIML Report Trigger frame to the MLME of the AIML peer STA (S217). The AIML Report Trigger frame is a frame that requests the sending of an AIML measurement report frame from the peer STA.

[0072] Furthermore, the MLME of the AIML initiating STA that has received the MLME - AIMLTBMSMTRQ.request and the self - permission to send sends an AIML Trigger frame to the MLME of the AIML peer STA (S221). The AIML Trigger frame is a frame that requests the sending of an NDP.

[0073] By specifying the probe NDP format for AI / ML for each STA in the NDPA, it is possible to perform NDP transmission for AI / ML for each STA, and at each node, the received NDP can be used for measurement.

[0074] The MLME of the AIML peer STA that has received the NDP sends an MLME - AIMLTBREPORT.indication to the SME of the AIML peer STA (S215). The MLME - AIMLTBREPORT.indication is a frame indicating that a measurement has been performed using the trigger - based probe NDP.

[0075] The SME of the AIML peer STA that has received the MLME - AIMLTBREPORT.indication sends an MLME - AIMLTBREPORTRQ.request (S216) to the MLME of the AIML peer STA. The MLME - AIMLTBREPORTRQ.request is a frame that requests the transmission of an AIML measurement report frame.

[0076] The MLME of the AIML peer STA that has received the MLME - AIMLTBREPORTRQ.request and the AIML report trigger frame sends an AIML measurement report frame (S218) to the MLME of the AIML initiating STA. The AIML measurement report frame is a frame that reports the measurement results of AIML probes to the other STA. In addition, the MLME of the AIML peer STA sends an MLME - AIMLTBREPORTRQ.confirm (S220) to the SME of the AIML peer STA. This is a frame that reports the result of the request to send the AIML measurement report frame. Regarding the transmission of the AIML measurement report frame, it can be transmitted not only when both the MLME - AIMLTBREPORTRQ.request and the AIML report trigger frame are received, but also when any one of them is received.

[0077] The MLME of the AIML initiating STA that has received the AIML measurement report frame sends an MLME - AIMLTBMSMTRQ.confirm (S219) to the SME of the AIML initiating STA.

[0078] The MLME of the AIML peer STA that has received the AIML trigger frame sends an NDP (S222) to the MLME of the AIML initiating STA. In Figure 2 the transmission of the NDP in S222, an example is shown where the measurement results are not sent to the other STA but the actions using the measurement results are performed.

[0079] The MLME of the AIML initiating STA that has received the NDP sends an MLME - AIMLTBREPORT.indication (S223) to the SME of the AIML initiating STA.

[0080] In Figure 2 the transmission of the NDP in S214, an example is shown where the measurement results are sent to the other STA, but it can also be that the actions using the measurement results are performed without sending, or that both the sending to the other STA and the actions using the measurement results are performed at the sending - source STA. It can be that the SME or the MLME contains an application program that performs the actions using the measurement results.

[0081] In Figure 2 the transmission of the NDP of S222, an example is shown in which the measurement result is not sent to the peer STA but the action using the measurement result is performed. However, the measurement result may also be sent to the peer STA, or it may be that both the transmission to the peer STA and the action of using the measurement result are performed in the transmitting STA. It may be that the SME or MLME includes an application program that performs the action of using the measurement result.

[0082] Regarding the AIML measurement termination phase, the processes starting from the AIML initiator STA and the processes starting from the AIML peer STA are shown.

[0083] When starting the AIML measurement termination phase from the AIML initiator STA, first, the SME of the AIML initiator STA sends MLME - AIMLMSMTTERMINATION.request (S130, S230) to the MLME of the AIML initiator STA. MLME - AIMLMSMTTERMINATION.request is a frame that requests the transmission of an AIML Measurement Setup Termination frame.

[0084] The MLME of the AIML initiator STA that has received MLME - AIMLMSMTTERMINATION.request sends an AIML measurement setup termination frame to the MLME of the AIML peer STA (S131, S231). The AIML measurement setup termination frame is a frame that indicates the release of the setting of the AIML probe.

[0085] The MLME of the AIML peer STA that has received the AIML measurement setup termination frame sends MLME - AIMLMSMTTERMINATION.indication (S132, S232) to the SME of the AIML peer STA. MLME - AIMLMSMTTERMINATION.indication is a frame indicating that the AIML measurement setup termination frame has been received. Also, the MLME of the AIML peer STA that has received the AIML measurement setup termination frame sends an Ack (S133, S233) to the MLME of the AIML initiator STA. Ack is an acknowledgment frame.

[0086] The MLME of the AIML initiator STA that has received an Ack sends MLME - AIMLMSMTTERMINATION.confirm (S134, S234) to the SME of the AIML initiator STA. MLME - AIMLMSMTTERMINATION.confirm is a frame indicating that the AIML measurement establishment termination frame has been received and the settings of the AIML probe have been released.

[0087] When starting the AIML measurement termination phase from the AIML peer STA, first, the SME of the AIML peer STA sends MLME - AIMLMSMTTERMINATION.request (S140, S240) to the MLME of the AIML peer STA.

[0088] The MLME of the AIML peer STA that has received MLME - AIMLMSMTTERMINATION.request sends an AIML measurement establishment termination frame (S141, S241) to the MLME of the AIML initiator STA.

[0089] The MLME of the AIML initiator STA that has received the AIML measurement establishment termination frame sends MLME - AIMLMSMTTERMINATION.indication (S142, S242) to the SME of the AIML initiator STA. Also, the MLME of the AIML initiator STA that has received the AIML measurement establishment termination frame sends an Ack (S143, S243) to the MLME of the AIML peer STA.

[0090] The MLME of the AIML peer STA that has received the Ack sends MLME - AIMLMSMTTERMINATION.confirm (S144, S244) to the SME of the AIML peer STA.

[0091] According to this embodiment, a node can specify a probe NDP format for AI / ML for each other node and can also make a specification mixed with the conventional probe NDP format.

[0092] Four embodiments are described regarding the NDPA frame for specifying the probe NDP format for AI / ML.

[0093] <Embodiment 1>

[0094] In Embodiment 1, for each STA information constituting the STA information list (STA info List) included in the NDPA frame format, it is defined that when the AID (Association Identifier) 11 subfield of the STA information is a specific value, it includes the AIML sounding NDP subfield. Thus, it is possible to specify and notify each node of the subfield (AIML sounding NDP subfield) that defines the sounding NDP format for AI / ML.

[0095] Figure 7 The format of the NDPA frame is shown. The NDPA frame includes a sounding dialog token field and a STA information list (STA Info List) field. Figure 8 Shows Figure 7 the format of the sounding dialog token field. Figure 9 Shows Figure 7 the format of the STA information field included in the STA information list field.

[0096] As Figure 8 shown, the sounding dialog token field consists of a 2-bit NDP announcement variant (NDP Announcement Variant) subfield and a 6-bit sounding dialog token number (Sounding Dialog Token Number) subfield.

[0097] Figure 11 The definition of the NDP announcement variant subfield is shown. For the 2-bit NDP announcement variant subfield, the following 4 types of NDP announcement frames have been defined: VHT (Very High Throughput) (NDP announcement variant = 00), Ranging (NDP announcement variant = 01), HE (High Efficiency) (NDP announcement variant = 10), EHT (Extremely High Throughput) (NDP announcement variant = 11). That is, there is no reservation (Reserved) in the NDP announcement variant subfield. Therefore, in Figure 11 the example shown, NDP announcement variant = 11 is defined to be used for both EHT and AIML (EHT / AIML). Therefore, other information is needed to identify whether it is EHT or AIML. NDP announcement variant = 11 means that the frame contains at least one NDP announcement for EHT / AIML.

[0098] Figure 9 The format of the STA information field as the information of the STA in the case of EHT is shown. As Figure 9As shown, there is an 11-bit AID11 subfield at the beginning of the STA information. It is defined that when the value of the AID11 subfield is a specific value, it is the AIML probe NDP subfield. Figure 12 shows an example of the definition of the AID11 subfield. In Figure 12 the example shown, 2042 which is undefined (reserved) is defined as the type of the NDP advertisement frame being AIML.

[0099] Figure 10 shows the format of the STA information field as the information of the STA in the case of AIML. For example, when the value of the AID11 subfield is 2042, as in the Figure 10 example of the format of the STA information field shown, the subfield after the AID11 subfield is defined to include the AIML probe NDP information (AIML sounding NDP info) as the information for the AI / ML probe NDP. In the AIML probe NDP information, for example, it may also include the AIML type (AIML Type) subfield as the type of AIML, or the AIML type-dependent information (AIML Type Dependent Info) as the information dependent on the AIML type.

[0100] Figure 13 shows an example of the definition of the AIML type subfield. As Figure 13 shown, as the type of AIML, for example, there are index-based CSI (Index-based CSI) (for example, refer to Non-Patent Document 4) and CSI compressed feedback scheme (Compressed CSI feedback scheme) (for example, refer to Non-Patent Document 5). The types of AIML can also be three or more. Figure 13 shows an example where the value of the AIML type subfield is defined as 0 when it is index-based CSI and 1 when it is the CSI compressed feedback scheme.

[0101] Figure 14 shows an example of the AIML probe NDP information when AIML type = 0 (index-based CSI). Index-based CSI is a method of optimizing (learning) the collected CSI, obtaining the CSI feedback vector with the least feedback information amount, and sending the index of this vector from each STA. In Figure 14In the example of the AIML detecting NDP information shown, the number of indexes specified by the Number of Index is assigned to each STA. In this example, the number is specified by the number of indexes, but it can also be specified by the starting index and the ending index, or the range of indexes can be specified by the starting index and the number of indexes. In addition, multiple indexes can be divided into multiple categories, and the indexes of the divided categories can be specified.

[0102] Figure 15 An example of the AIML detecting NDP information when the AIML type = 1 (CSI compression feedback scheme) is shown. The CSI compression feedback scheme is a scheme in which a neural network for CSI compression called "Autoencoder" is optimized in the training phase, and then it is divided into an encoder for compression and a decoder for decompression and configured on the transmitting side and the receiving side to operate. In Figure 15 In the example of the AIML detecting NDP information shown, in the specification of the compressed CSI feedback, it includes the compressed partial bandwidth information (Compressed Partial BW (BandWidth) Info), the compressed column number index (Compressed Nc Index), and the compressed feedback type and the number of groups, the codebook size (Compressed Feedback Type And Ng, Codebook Size) as the compressed information.

[0103] In Figure 9 the example, it is assumed that the AIML detecting NDP subfield is defined when the AID11 subfield has a specific value, but in the case where it is recognized at the start of the AIML measurement phase that the autoencoder has been configured, such as when the autoencoder has been configured by identifying the session ID (session ID) or the dialog token (Dialog Token), etc., it is also possible to switch to the CSI feedback specification after compression without explicitly sending the AIML NDP announcement frame. It can be that, as an identifier set in advance in the AIML measurement establishment phase, etc., the session ID or the dialog token, etc. is set to a specific value, and thus it is recognized as the AIML NDP announcement frame.

[0104] In addition, not limited to the format for reducing the CSI feedback amount using AIML, an extended CSI feedback specification corresponding to broadbanding and high-precision can also be specified. Regarding the extended CSI feedback, the value of the AIML type subfield can be defined as 2. In Embodiment 1, the sounding measurement using NDP is described, but not limited to the sounding using NDP. It can also be used for measurements using other frames, such as interference measurement of OBSS (Overlapping Basic Service Set). In this case, an AIML Measurement Request frame or an AIML Measurement Response frame can be used instead of NDP and the AIML measurement report.

[0105] In this way, by defining the AIML sounding NDP information when the AID11 subfield of the NDPA frame is a specific value, the sounding NDP format for AI / ML can be specified for each node. In addition, since the type of the sounding NDP can be specified by the AID11 of each STA information, a designation in which the sounding NDP format for AI / ML and the sounding NDP format not for AI / ML are mixed can be notified by one NDPA frame.

[0106] <Embodiment 2>

[0107] In Embodiment 2, it is defined that when the subtype value of the frame control field or the ControlFrame Extension value is a specific value (for example, 0001, 1100, etc. which are currently undefined values), it is an AIML sounding NDP frame. When the subtype value of the frame control field or the ControlFrame Extension value is a specific value, the AIML sounding NDP information is included in at least one of the respective STA information constituting the STA information list.

[0108] Figure 16 A format example of the NDPA frame for AIML is shown. The NDPA frame for AIML includes a frame control field, an NDP Announcement Type List field, and a Per STA Info List field. The frame control field includes a type value subfield and a subtype value subfield.

[0109] Figure 17 is shown in Figure 16 the definition example of the type value subfield and the subtype value subfield of the frame control field in Figure 17As shown, a frame with the subtype value of 0001 in the frame control field (which is an undefined value (reserved)) is defined as an AIML probe NDP frame. In Figure 17 an example of the subtype value of 0001 is shown, but it can also be defined for other undefined values.

[0110] Figure 18 An example definition of the control frame extension of the frame control field is shown. As Figure 17 shown, when the subtype value of the frame control field is 0110, it is the control frame extension. As Figure 18 shown, a frame with the control frame extension value of 1100 (which is an undefined value) in the frame control field is defined as an AIML probe NDP frame. In Figure 18 an example of defining the control frame extension value of the AIML probe NDP frame as 1100 (which is an undefined value) is shown, but it can also be defined as other undefined values.

[0111] As Figure 17 or Figure 18 shown, it can be defined that when the subtype value or the control frame extension value of the frame control field is a specific value, it includes an AIML probe NDP frame.

[0112] As Figure 16 shown in the example, the AIML probe NDP frame includes an NDP announcement type list. The NDP announcement type list includes the number of STAs (Number of STA) and a list of NDP announcement types (NDP announcement type list). The NDP announcement type list includes NDP announcement type fields with the number of STAs.

[0113] Figure 19 An example definition of the NDP announcement type is shown. As Figure 19 shown, as the NDP announcement type, in addition to VHT, ranging, HE, and EHT which are the types of NDP announcement frames, a definition for AIML is added. Each NDP announcement type contained in the NDP announcement type list represents the type of the NDP announcement frame of each STA. For example, the types of the NDP announcement frames of each STA can be different. For example, STA1 is EHT and STA2 is AIML.

[0114] The content of the STA information list including STA information 1 to n of each STA is determined according to the NDP advertisement type of each STA specified in the NDP advertisement type list. Here, n is the number of STAs. For example, when the NDP advertisement type of STA n is AIML, node ID, AIML probe NDP information, etc. are included in each STA information n. As the content of the AIML probe NDP information, the content described in Embodiment 1 can be included. For example, the NDP advertisement type of STAm may not be AIML either. Although it is described as "each STA information" or "STA information list", it may also be referred to as "User Info" or "UserInfo List".

[0115] In this way, it is defined that when the subtype value or control frame extension value of the frame control field is a specific value, it is an AIML probe NDP frame. By setting the NDP advertisement type list in the NDPA frame for AIML, it is possible to notify, by one NDPA frame, the specified probe NDP format for AI / ML or the specified probe NDP format not for AI / ML for each node, that is, it is possible to notify, by one NDPA frame, the specification in which different probe NDP formats for each node are mixed together.

[0116] <Embodiment 3>

[0117] In Embodiment 3, it is defined that when the trigger type of the trigger frame is a specific value (for example, 8 which is an undefined value, etc.), it is a frame including AIML probe NDP information.

[0118] Figure 20 An example of the format of the trigger frame for AIML is shown. The trigger frame includes a Common Info field and a STA information list field. The Common Info field includes a trigger type subfield and an NDP advertisement type list subfield. The trigger type subfield specifies the type of the trigger frame. The NDP advertisement type list includes an STA number field and a list of NDP advertisement types.

[0119] The content of the NDP advertisement type list is the same as that in Embodiment 2. The STA information list field includes each STA information 1 to n. The content of the STA information list is the same as that in Embodiment 2. The content of each STA information 1 to n is determined according to the NDP advertisement type of each STA specified in the NDP advertisement type list. Although it is described as "each STA information" or "STA information list", it may also be referred to as "user information" or "user information list". In addition, subfields indicating UL / DL (uplink / downlink) of the indicated NDP may be included in the NDP advertisement type list or the STA information list.

[0120] AsFigure 20 As shown, the case where the value of the trigger type subfield is 8 (which is an undefined value) is defined as the case including the AIML probe NDP information. In the case where the value of the trigger type subfield is 8, at least one of the STAs constituting the NDP advertisement type list includes the AIML probe NDP information. In Figure 20 , an example in which the AIML probe NDP information is defined at the undefined value 8 is shown, but it can also be defined at other undefined values.

[0121] Instead of Figure 1 and Figure 2 the NDPA or AIML NPDA frame in the probe process example for AI / ML shown, a trigger frame with the trigger type set to a specific value shown in Figure 20 is used.

[0122] For example, after the AP (Access Point) sends a PPDU including a trigger frame, for example, after a time of SIFS (Short Inter Frame Space), it sends another PPDU (for example, NDP) to the STA. The STA that receives the NDP can send an NDP or a report frame. The AP can include information (for example, the next PPDU subfield, the additional PPDU subfield, or the following PPDU subfield) indicating that another PPDU (for example, NDP) is sent after the trigger frame in the trigger frame.

[0123] As described above, according to Embodiment 3, in the case where the trigger type of the trigger frame is a specific value, the frame defined to include the AIML probe NDP information can specify the probe NDP format for AI / ML for each node by setting the NDP advertisement type list. That is, it is possible to notify, by one NDPA frame, the specification in which multiple types of probe NDP formats for each node are mixed together.

[0124] <Embodiment 4>

[0125] In Embodiment 4, an information element (AIML probe NDP information element) including the AIML probe NDP information is defined and notified by a management frame.

[0126] Figure 21 A structural example of the AIML probe NDP information element is shown. It can be defined that, in the case where the Element ID or the Element ID Extension is a specific value as an undefined value, this information element is the AIML probe NDP information element. In this case, it can be, asFigure 21 As shown, the AIML probe NDP information element adopts a structure including an NDP announcement type list and an STA information list. The content of these lists is the same as that in Embodiment 2. Although it is described as "per STA information" or "STA information list", it can also be referred to as "user information" or "user information list".

[0127] By including the AIML probe NDP information element configured in the above manner in management frames such as a Beacon frame, an Association Request / Response frame, a Probe Request / Response frame, and Action frames for notification, information including the AIML probe NDP information can be notified to each STA.

[0128] In this way, by notifying the information element (AIML probe NDP information element) including the AIML probe NDP information by the management frame, a probe NDP format for AI / ML is specified for each node, and thus, notification with multiple probe NDP formats mixed together can be performed.

[0129] The interface names (frame names), field names, or sub-field names described in Embodiments 1 to 4 may also be other names.

[0130] Figure 22 The structure of each STA2200 is shown. The control unit 2201 (for example, corresponding to a control circuit) controls the transmission unit 2202 and the reception unit 2203 based on the signal received by the reception unit 2203 or the signal input from an input unit (not shown) in the STA. Both the SME and the MLME are functions existing in the control unit. In another example, it may also be that the control unit 2201 includes the function of the MLME and a part of the function of the SME. In other words, a part or all of the function of the SME may be executed in another communication device 2300 outside the STA2200 (for example, another STA, another access point (AP), a wireless LAN controller, a multi-AP coordinator, a cloud server, software as a service (SaaS), a virtual machine (VM)). The transmission unit 2202 (for example, corresponding to a transmission circuit) transmits a signal to the other STA. The reception unit 2203 (for example, corresponding to a reception circuit) receives a signal from the other STA. The transmission unit 2202 and the reception unit 2203 may also be integrated to form a transceiver unit. The transmission unit 2202, the reception unit 2203, the transmission unit 2202 and the reception unit 2203, or the transceiver unit constitute a communication unit.

[0131] Figure 23The structure of another communication device 2300 is shown. Another communication device 2300 has a control unit 2301 and a communication circuit 2302. The control unit 2301 includes the functions of the SME. The communication circuit 2302 communicates with the other communication device (for example, STA 2200). The control unit 2301 and the communication circuit 2302 can communicate using the communication primitives (messages, signals) between the SME and the MLME described in the respective embodiments. The control unit 2301 may include a CPU 2311, a memory 2312, and a storage device 2313. The CPU 2311 executes the functions of the SME. The memory 2312 is used when the CPU 2311 executes processing. The storage device 2313 stores software related to the functions of the SME (for example, binary, executable file, program code, container image). It is possible that the CPU 2311 transfers the software stored in the storage device 2313 to the memory 2312 in order to execute the functions of the SME.

[0132] As described above, although the embodiments have been described with reference to the drawings, the present disclosure is not limited to this example. It is obvious that those skilled in the art can conceive of various modification examples or correction examples within the scope described in the claims. It should be understood that these modification examples or correction examples also belong to the technical scope of the present disclosure. In addition, the respective structural elements in the embodiments can be arbitrarily combined without departing from the gist of the present disclosure.

[0133] In the above embodiments, the expression "unit" used for each structural element can also be replaced with other expressions such as "… circuitry", "… assembly", "device", "unit", or "module".

[0134] The present disclosure can be implemented by software, hardware, or software in cooperation with hardware. Each functional block used in the description of the above embodiments can also be partially or wholly implemented as an LSI (Large Scale Integration, large-scale integrated circuit) as an integrated circuit. Each process described in the above embodiments can also be partially or wholly controlled by one LSI or a combination of LSIs. The LSI can be composed of individual chips or can be composed of one chip in a manner that includes part or all of the functional blocks. The LSI can also include input and output of data. Depending on the degree of integration, the LSI is sometimes referred to as "IC (Integrated Circuit)", "system LSI", "ultra-large LSI", or "extra-large LSI".

[0135] The method of integrating circuits is not limited to LSI, and can also be implemented by dedicated circuits, general-purpose processors, or dedicated processors. Additionally, an FPGA (Field Programmable Gate Array) that can be programmed after LSI manufacturing, or a reconfigurable processor that can reconfigure the connections or settings of circuit blocks inside the LSI, can also be used. The present disclosure can also be implemented as digital processing or analog processing.

[0136] Furthermore, if, with the progress of semiconductor technology or the derivation of other technologies, an integrated circuit technology that replaces LSI emerges, of course, this technology can also be used to achieve the integration of functional blocks. There is also the possibility of applying biotechnology, etc.

[0137] The present disclosure can be implemented in all types of devices, equipment, and systems with communication functions (collectively referred to as "communication devices"). A communication device may also include a wireless transceiver and a processing / control circuit. The wireless transceiver may also include a receiving unit and a transmitting unit, or perform the functions of these parts. The wireless transceiver (transmitting unit, receiving unit) may also include an RF (Radio Frequency) module and one or more antennas. The RF module may also include an amplifier, an RF modulator / demodulator, or devices similar to these. Non-limiting examples of communication devices include: telephones (mobile phones, smartphones, etc.), tablet computers, personal computers (PCs) (laptop computers, desktop computers, notebook computers, etc.), cameras (digital cameras, digital video cameras, etc.), digital players (digital audio / video players, etc.), wearable devices (wearable cameras, smartwatches, tracking devices, etc.), game consoles, e-book readers, remote health / telemedicine (remote healthcare / medical prescription) devices, transportation vehicles or means of transportation with communication functions (automobiles, airplanes, ships, etc.), and combinations of the above various devices.

[0138] Communication devices are not limited to portable or mobile devices, and also include all types of devices, equipment, and systems that are not portable or are fixed. For example, it includes: smart home devices (home appliances, lighting devices, smart meters or gauges, control panels, etc.), vending machines, and all "Things" that can exist on the IoT (Internet of Things) network.

[0139] Communication includes not only data communication through cellular systems, wireless LAN (Local Area Network) systems, communication satellite systems, etc., but also data communication through combinations of these systems.

[0140] In addition, the communication device also includes devices such as a controller or a sensor that are connected or linked to a communication device that executes the communication functions described in the present invention. For example, it includes a controller or a sensor that generates a control signal or a data signal used by the communication device that executes the communication functions of the communication device.

[0141] In addition, the communication device includes infrastructure devices that communicate with or control the above-mentioned various non-limiting devices, such as base stations, access points, and all other devices, equipment, and systems.

[0142] In addition, in recent years, in IoT (Internet of Things) technology, CPS (Cyber Physical Systems), a new concept that creates new added value through information cooperation between the physical space and the information space, has attracted much attention. In the above-mentioned embodiment, this CPS concept can also be adopted.

[0143] That is, as a basic configuration of CPS, for example, it is possible to connect an edge server configured in the physical space to a cloud server configured in the information space via a network, and perform distributed processing through processors mounted on both servers. Here, each piece of processing data generated in the edge server or the cloud server is preferably generated on a standardized platform. By using such a standardized platform, the efficiency of building a system including various sensor groups or IoT application software can be improved.

[0144] (1) The communication device according to an embodiment of the present disclosure is a control unit and has a communication unit. The control unit generates a frame including information of each station, that is, each STA information. At least one of the STA information is a detection NDP information for artificial intelligence / machine learning, that is, for AI / ML. The communication unit transmits the generated frame.

[0145] (2) In the communication device according to an embodiment of the present disclosure, when the AID11 subfield of the STA information is a specific value, the detection NDP information for AI / ML of the device in (1) is defined.

[0146] (3) In the communication device according to an embodiment of the present disclosure, when the subtype value or the control frame extension value of the frame control field is a specific value, the detection NDP information for AI / ML of the device in (1) is defined.

[0147] (4) In the communication device according to an embodiment of the present disclosure, when the trigger type of the trigger frame is a specific value, the detection NDP information for AI / ML of the device in (1) is defined.

[0148] In the communication device according to an embodiment of the present disclosure, the detection NDP information for AI / ML of the device in (1) is defined in a management frame.

[0149] (6) A communication method according to an embodiment of the present disclosure includes the steps of: generating a frame including information of each STA, that is, each STA information, at least one of the each STA information being detection NDP information for AI / ML; and transmitting the generated frame.

[0150] The disclosures of the specification, drawings, and abstract of the Japanese patent application No. 2022-181083 filed on November 11, 2022 are incorporated herein by reference in their entirety.

[0151] Industrial Applicability

[0152] An embodiment of the present disclosure is useful for a wireless device.

[0153] Description of Reference Numerals

[0154] 2201 Control Unit

[0155] 2202 Transmission Unit

[0156] 2203 Reception Unit

Claims

1. A communication device, which is a control unit and has a communication unit, The control unit generates a frame containing information of each station, i.e., STA information, and at least one STA information among the STA information is a probe NDP information for artificial intelligence / machine learning, i.e., for AI / ML, and the NDP is an empty data physical layer convergence protocol data unit, i.e., an empty data PPDU. The communication unit transmits the generated frame.

2. The communication device according to claim 1, wherein, The frame is a probe NDP announcement frame, i.e., a probe NDPA frame. The probe NDP information for AI / ML is defined when the AID11 subfield of the STA information has a specific value.

3. The communication device according to claim 1, wherein, The frame is a probe NDPA frame. The probe NDP information for AI / ML is defined when the subtype value or the control frame extension value of the frame control field has a specific value.

4. The communication device according to claim 1, wherein, The frame is a trigger frame. The probe NDP information for AI / ML is defined when the trigger type of the trigger frame has a specific value.

5. The communication device according to claim 1, wherein, The frame is a management frame. The probe NDP information for AI / ML is defined in the management frame.

6. A communication method, which includes the following steps: generating a frame containing information of each station, i.e., STA information, and at least one STA information among the STA information is a probe NDP information for artificial intelligence / machine learning, i.e., for AI / ML, and the NDP is an empty data physical layer convergence protocol data unit, i.e., an empty data PPDU; and transmitting the generated frame.

Images