WO2022142573A1 - 一种信道接入方法和装置 - Google Patents

一种信道接入方法和装置 Download PDF

Info

Publication number
WO2022142573A1
WO2022142573A1 PCT/CN2021/122982 CN2021122982W WO2022142573A1 WO 2022142573 A1 WO2022142573 A1 WO 2022142573A1 CN 2021122982 W CN2021122982 W CN 2021122982W WO 2022142573 A1 WO2022142573 A1 WO 2022142573A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
moment
time
message
time period
Prior art date
Application number
PCT/CN2021/122982
Other languages
English (en)
French (fr)
Inventor
刘鹏
郭子阳
郭宇宸
罗嘉俊
杨讯
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP21913318.8A priority Critical patent/EP4258793A1/en
Publication of WO2022142573A1 publication Critical patent/WO2022142573A1/zh
Priority to US18/342,468 priority patent/US20230345536A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W74/00—Wireless channel access
    • H04W74/08—Non-scheduled access, e.g. ALOHA
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W74/00—Wireless channel access
    • H04W74/08—Non-scheduled access, e.g. ALOHA
    • H04W74/0808—Non-scheduled access, e.g. ALOHA using carrier sensing, e.g. carrier sense multiple access [CSMA]
    • H04W74/0816—Non-scheduled access, e.g. ALOHA using carrier sensing, e.g. carrier sense multiple access [CSMA] with collision avoidance
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W74/00—Wireless channel access
    • H04W74/002—Transmission of channel access control information
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/092—Reinforcement learning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/12—Wireless traffic scheduling
    • H04W72/1263—Mapping of traffic onto schedule, e.g. scheduled allocation or multiplexing of flows

Definitions

  • the present solution relates to the field of communications, and in particular, to a channel access method and device.
  • channels are shared, that is, multiple nodes in the wireless network use the same shared channel for communication.
  • no central node In a distributed scenario, there is no central node to uniformly schedule and coordinate all nodes in the network, and the nodes directly decide whether to access the shared channel.
  • nodes can usually access the shared channel randomly or obtain a large amount of data to predict the result of accessing the shared channel and decide whether to access the channel. This method has a low probability of successfully accessing the channel and low channel throughput.
  • Embodiments of the present application provide a channel access method and device.
  • the first node before sending the message to the second node through the shared channel, the first node can input the obtained network state information at each moment in the target period into the target neural network, and then send the obtained first moment to the second node.
  • the prediction result of the message, and finally, whether to send the message to the second node through the shared channel is determined according to the prediction result.
  • the target period is a preset time period before the first moment including the first moment;
  • the network status information may include the first time period, the second time period, and the busy and idle status of the shared channel monitored by the first node.
  • Status and the total number of interfering nodes, the first time period at the second time is used to indicate the time interval between the time when the first node successfully sent a packet through the shared channel last time before the second time and the second time, the second time
  • the second time period is used to indicate the time interval between the time when the interfering node successfully sends a message through the shared channel last time before the second time and the second time;
  • the interfering node is the monitored data of the second node except the first node The node that sends the message through the shared channel; the second moment is any moment in the target period.
  • the node can decide whether to access the shared channel according to the network state in the past period of time, and the method can improve the success rate of the node accessing the channel and increase the throughput of the channel.
  • an embodiment of the present application provides a channel access method, characterized in that the method includes:
  • the first node acquires network status information at each moment in the target period, where the target period is a time period with a preset duration before the first moment including the first moment; the network status information includes the first period and the second period.
  • Time period, the first time period at the second time is used to indicate the time interval between the time when the first node successfully sent a message through the shared channel last time before the second time and the second time, the second time at the second time
  • the segment is used to indicate the time interval between the time when the interfering node successfully sends a message through the shared channel last time before the second time and the second time;
  • the node sending the message; the second moment is any moment in the target cycle;
  • the first node inputs the network state information at each moment into the target neural network to obtain a prediction result.
  • the prediction result includes a first prediction value and a second prediction value.
  • the first prediction value is used to instruct the first node to pass the shared channel at the first moment.
  • the probability that the message is successfully sent to the second node, and the second predicted value is used to indicate the probability that the first node fails to send the message to the second node through the shared channel at the first moment;
  • the first node When the first predicted value is greater than the second predicted value, the first node sends the message to be sent to the second node through the shared channel.
  • the network status information further includes the busy and idle status of the shared channel and the total number of interfering nodes monitored by the first node.
  • the first node acquires network status information at each moment in the target period, including: acquiring a second time period at a second moment;
  • Get the second time period of the second moment including:
  • the first node When the first node monitors the response information at the second time according to the first indication information, it determines that the second time period of the second time is 0, and the second time is the time after the first node monitors the first duration of the first message , the header of the first packet includes first indication information, where the first indication information is used to instruct the node that receives the first packet to send response information after a first duration.
  • the first node acquires network status information at each moment in the target period, including: acquiring a second time period at a second moment;
  • Get the second time period of the second moment including:
  • the first node determines that the second time period at the second moment is 0, and the second indication information is used to indicate the packet corresponding to the response information including the second indication information The sending result is successful;
  • the first node monitors the response information including the third indication information at the second moment, it determines that the second time period of the second moment is the difference between the second time period of the previous moment of the second moment in the target period and the first time interval. And, the first time interval is the time interval between the last time and the second time, and the third indication information is used to indicate that the sending result of the packet corresponding to the response information including the second indication information is a failure.
  • the first node obtains network status information at each moment in the target period, including: obtaining the total number of interfering nodes at the second moment;
  • the first node receives the first broadcast frame broadcast by the second node
  • the first node determines the total number of interfering nodes parsed from the first broadcast frame as the total number of interfering nodes at the second moment.
  • the first node acquires network status information at each moment in the target period, including: acquiring a second time period at a second moment;
  • Get the second time period of the second moment including:
  • the first node When the first node receives the second broadcast frame at the second moment, it determines that the second time period at the second moment is 0.
  • the second broadcast frame is broadcast by the second node when it monitors that the interfering node successfully sends the message through the shared channel. to the first node.
  • the second broadcast frame includes a first time point, and the first time point is the time when the second node monitors that the interfering node successfully sends the message through the shared channel.
  • determining that the second time period at the second moment is 0, including:
  • the first node determines the time interval between the first time point and the second time as the first time interval
  • the method further includes:
  • the first node obtains the sending result of the message to be sent
  • the first node takes the network state information at each moment of the target period as input and updates the target neural network with the sending result of the message to be sent as the label.
  • the first node takes the network state information of the target period as input and uses the sending result of the message to be sent as the label to update the target neural network, including:
  • the target neural network is updated according to the loss, which includes the error between the predicted result and the sending result of the message to be sent.
  • the loss further includes a reward value at the first moment, and the reward value is determined according to the prediction result and the busy-idle status of the shared channel monitored by the first node at the first moment.
  • the determined reward value is the ratio of the first time period to the second time period
  • the reward value is the negative value of the total number of interfering nodes at the first moment.
  • the return value is determined as the ratio of the second time period at the first moment to the first time period at the first moment;
  • the first node When the first node detects that the shared channel has response information indicating that the message transmission fails at the first moment, it determines that the return value is M, and M is a positive number;
  • the first node determines that the return value is 0 when monitoring that there is no response information on the shared channel at the first moment.
  • a second aspect discloses a channel access device, the device comprising:
  • the obtaining unit is used to obtain the network status information at each moment in the target period, where the target period is a time period with a preset duration before the first moment including the first moment; the network status information includes the first period and the second period Time period, the first time period at the second time is used to indicate the time interval between the time when the first node successfully sent a message through the shared channel last time before the second time and the second time, the second time at the second time.
  • the segment is used to indicate the time interval between the time when the interfering node successfully sends a message through the shared channel last time before the second time and the second time; The node sending the message; the second moment is any moment in the target cycle;
  • the prediction unit is used to input the network state information at each moment into the target neural network to obtain a prediction result, the prediction result includes a first prediction value and a second prediction value, and the first prediction value is used to instruct the first node to pass through at the first moment The probability that the shared channel successfully sends the message to the second node, and the second predicted value is used to indicate the probability that the first node fails to send the message to the second node through the shared channel at the first moment;
  • a decision unit configured to send the to-be-sent message to the second node by the first node through the shared channel when the first predicted value is greater than the second predicted value.
  • the acquiring unit is further configured to acquire the busy-idle status of the shared channel and the total number of interfering nodes monitored by the first node.
  • the obtaining unit is specifically configured to obtain the second time period of the second moment, wherein:
  • the first node When the first node monitors the response information at the second time according to the first indication information, it determines that the second time period of the second time is 0, and the second time is the time after the first node monitors the first duration of the first message , the header of the first packet includes first indication information, where the first indication information is used to instruct the node that receives the first packet to send response information after a first duration.
  • the obtaining unit is specifically configured to obtain the second time period of the second moment, wherein:
  • the first node determines that the second time period at the second moment is 0, and the second indication information is used to indicate the packet corresponding to the response information including the second indication information The sending result is successful;
  • the first node monitors the response information including the third indication information at the second moment, it determines that the second time period of the second moment is the difference between the second time period of the previous moment of the second moment in the target period and the first time interval. And, the first time interval is the time interval between the last time and the second time, and the third indication information is used to indicate that the sending result of the packet corresponding to the response information including the second indication information is a failure.
  • the obtaining unit is specifically configured to obtain the total number of interfering nodes at the second moment, wherein:
  • the first node receives the first broadcast frame broadcast by the second node
  • the first node determines the total number of interfering nodes parsed from the first broadcast frame as the total number of interfering nodes at the second moment.
  • the obtaining unit is specifically configured to obtain the second time period of the second moment, wherein:
  • the first node When the first node receives the second broadcast frame at the second moment, it determines that the second time period at the second moment is 0.
  • the second broadcast frame is broadcast by the second node when it monitors that the interfering node successfully sends the message through the shared channel. to the first node.
  • the second broadcast frame includes a first time point, and the first time point is the time when the second node monitors that the interfering node successfully sends the message through the shared channel.
  • the acquiring unit is specifically configured to execute:
  • the first node determines the time interval between the first time point and the second time as the first time interval
  • the channel access apparatus further includes a training unit, and the training unit is configured to perform:
  • the first node obtains the sending result of the message to be sent
  • the first node takes the network state information at each moment of the target period as input and updates the target neural network with the sending result of the message to be sent as the label.
  • the channel access apparatus further includes a training unit, and the training unit is specifically configured to perform: updating the target neural network according to the loss, where the loss includes the error between the prediction result and the sending result of the to-be-sent message.
  • the loss further includes a reward value at the first moment, and the reward value is determined according to the prediction result and the busy-idle status of the shared channel monitored by the first node at the first moment.
  • the channel access apparatus further includes a training unit, and the training unit is further configured to perform:
  • the reward value is the negative value of the total number of interfering nodes at the first moment.
  • the channel access apparatus further includes a training unit, and the training unit is further configured to perform:
  • the first node monitors the shared channel at the first moment
  • the return value is determined as the ratio of the second time period at the first moment to the first time period at the first moment;
  • the first node When the first node detects that the shared channel has response information indicating that the message transmission fails at the first moment, it determines that the return value is M, and M is a positive number;
  • the first node determines that the return value is 0 when monitoring that there is no response information on the shared channel at the first moment.
  • a third aspect discloses a channel access device
  • the channel access device may include a processor, a memory, an input interface and an output interface, where the input interface is configured to receive channel access from other channels other than the channel access device information of the device, the output interface is used to output information to other channel access devices than the channel access device, and when the processor executes the computer program stored in the memory, the processor executes the first A channel access method disclosed in one aspect or any embodiment of the first aspect.
  • a fourth aspect discloses a channel access network, where the channel access network includes the channel access device of the second aspect and the channel access device of the third aspect.
  • Fig. 1 is a kind of effect comparison schematic diagram that the embodiment of this application provides
  • FIG. 2 is a schematic diagram of a channel access network architecture according to an embodiment of the present application.
  • FIG. 3 is a schematic diagram of another channel access network architecture provided by an embodiment of the present application.
  • FIG. 4 is a flowchart of a channel access method provided by an embodiment of the present application.
  • FIG. 5 is a schematic diagram of a first time period and a second time period provided by an embodiment of the present application
  • 6A is a schematic diagram of acquiring a second time period according to an embodiment of the present application.
  • 6B is a schematic diagram of a response message provided by an embodiment of the present application.
  • FIG. 7 is a schematic diagram of a target neural network provided by an embodiment of the present application.
  • FIG. 8 is a flowchart of another channel access method provided by an embodiment of the present application.
  • 9A is a schematic diagram of an information unit of a broadcast frame provided by an embodiment of the present application.
  • 9B is a schematic diagram of an information unit of another broadcast frame provided by an embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of a channel access apparatus 400 according to an embodiment of the present application.
  • FIG. 11 is a schematic structural diagram of another channel access apparatus 500 according to an embodiment of the present application.
  • the first node is a node that sends a packet to the second node
  • the second node is a node that receives a packet from the first node. Since the first node, the second node and the interfering node all communicate through a shared channel, if the interfering node sends a message through the above-mentioned shared channel while the first node accesses the shared channel, it will affect the first node to send the message to the first node. Section two sends the message. Therefore, before the first node sends a message to the second node, the first node may execute the channel access method provided by the embodiment of the present application to determine whether to access the above-mentioned shared channel.
  • Embodiments of the present application provide a channel access method and device.
  • the first node before sending the message to the second node through the shared channel, the first node can input the obtained network state information at each moment in the target period into the target neural network, and then send the obtained first moment to the second node.
  • the prediction result of the message, and finally, whether to send the message to the second node through the shared channel is determined according to the prediction result.
  • the target period is a preset time period before the first moment including the first moment;
  • the network status information may include the first time period, the second time period, and the busy and idle status of the shared channel monitored by the first node.
  • Status and the total number of interfering nodes, the first time period at the second time is used to indicate the time interval between the time when the first node successfully sent a packet through the shared channel last time before the second time and the second time, the second time
  • the second time period is used to indicate the time interval between the time when the interfering node successfully sends a message through the shared channel last time before the second time and the second time;
  • the interfering node is the monitored data of the second node except the first node The node that sends the message through the shared channel; the second moment is any moment in the target period.
  • the node can decide whether to access the shared channel according to the current network state, and the method can improve the success rate of the node accessing the channel and increase the throughput of the channel.
  • FIG. 1 is a schematic diagram of an effect comparison provided by an embodiment of the present application.
  • the technical solution increases the throughput of the channel; as shown in (B) in FIG. 1 , the technical solution reduces the average delay and the maximum delay of communication between nodes.
  • FIG. 2 is a schematic diagram of a channel access network architecture provided by an embodiment of the present application.
  • the network architecture includes a first node 10, a second node 20 and an interference node 30, wherein the interference node 30 may include multiple nodes, and the first node 10, the second node 20 and the interference node 30 are all Communicate over the same shared channel. in:
  • the first node 10 may input the acquired network status information at each moment in the target period into the target neural network, and obtain the information sent to the second node at the first moment. 20 sends the prediction result of the message, and finally, the first node 10 determines whether to send the message to the second node 20 through the shared channel according to the prediction result.
  • the first node 10 may acquire network status information at each moment in the target period, wherein the target period is a time period with a preset duration before the first moment including the first moment; the network status information may include the first moment.
  • the busy-idle status, the first time period, the second time period, and the total number of interfering nodes of the shared channel monitored by the node 10, and the first time period at the second moment is used to instruct the first node 10 to pass the shared channel before the second moment.
  • the time interval between the time when the last time the packet was successfully sent and the second time, the second time period at the second time is used to indicate the time between the time when the interfering node 30 successfully sent the last message through the shared channel before the second time and the first time.
  • the time interval between the two moments, the second moment is any moment in the target period; the interfering node 30 is the node monitored by the second node 20 , except the first node 10 , that sends messages through the shared channel. Further, the first node 10 inputs the network state information at each moment into the target neural network to obtain a prediction result.
  • the prediction result includes a first prediction value and a second prediction value, and the first prediction value is used to indicate that the first node 10 is in the first The probability that the message is successfully sent to the second node 20 through the shared channel at the moment, and the second prediction value is used to indicate the probability that the first node 10 fails to send the message to the second node 20 through the shared channel at the first moment; When the value is greater than the second predicted value, the first node 10 sends the message to be sent to the second node 20 through the shared channel.
  • any one of the first node 10 , the second node 20 , and the interfering node 30 in the embodiment of the present application may be an access point AP or a station STA.
  • the station STA may refer to user equipment, access terminal, subscriber unit, subscriber station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user equipment.
  • the station may also be a cellular telephone, a cordless telephone, a session initiation protocol (SIP) telephone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a wireless communication capable handheld devices, computing devices or other processing devices connected to wireless modems, in-vehicle devices, wearable devices, terminal devices in future 5G networks or terminals in future evolved public land mobile networks (PLMN) equipment, etc., which are not limited in this embodiment of the present application.
  • SIP session initiation protocol
  • WLL wireless local loop
  • PDA personal digital assistant
  • An access point AP may be a device used to communicate with stations.
  • the access point can be any device with a wireless transceiver function or a chip that can be provided in the device, and the device includes but is not limited to: an evolved Node B (evolved Node B, eNB), a radio network controller (radio network controller, RNC), Node B (Node B, NB), base station controller (BSC), base transceiver station (base transceiver station, BTS), home base station (for example, home evolved NodeB, or home Node B, HNB), base band unit (BBU), access point (AP), wireless relay node, wireless backhaul node, transmission point (transmission point) in wireless fidelity (wireless fidelity, WIFI) system , TP) or transmission and reception point (TRP), etc., can also be 5G, such as NR, gNB in the system, or, transmission point (TRP or TP), one or one of the base stations in the 5G system A group (including a
  • FIG. 3 is a schematic diagram of another channel access network architecture provided by an embodiment of the present application.
  • the channel access network architecture includes a first node, a second node and an interfering node, wherein the second node is an AP, and both the first node and the interfering node are STAs connected to the second node. It is understandable that the second node can obtain the responses of all STAs in the same network for sending packets.
  • the second node may generate a broadcast frame according to the moment of obtaining the response when obtaining the response of all STAs under the same network for sending the message, and broadcast the above-mentioned broadcast frame to all STAs under the same network. It can be understood that when the STA that receives the broadcast frame executes the channel execution method of the embodiment of the present application, the second time period can be updated according to the broadcast frame. For details, see the following related to the channel access method of the embodiment of the present application. content.
  • the technical solutions of the present application can be applied not only to the communication between the AP and the AP, but also to the communication between the AP and the STA, and also to the communication between the STA and the STA.
  • the following takes the communication between the AP and the STA as an example for description, which does not constitute a limitation to the embodiments of the present application.
  • channel access network architectures in FIG. 2 and FIG. 3 are only exemplary implementations of the embodiments of the present application, and the channel access network architectures in the embodiments of the present application include but are not limited to the above channel access network architectures. .
  • FIG. 4 is a flowchart of a channel access method provided by an embodiment of the present application.
  • the method may be applied to the channel access network shown in FIG. 2 or FIG. 3.
  • the method may include some or all of the following: step.
  • the first node acquires network status information at each moment in a target period, where the target period is a time period with a preset duration before the first moment including the first moment, and the network status information includes the first period and second time period.
  • the first node may acquire network status information at each moment in the target period.
  • the target period is a preset time period before the first moment including the first moment;
  • the network status information may include a first time period and a second time period, and the first time period at the second moment is used to indicate The time interval between the time when the first node successfully sends a packet through the shared channel before the second time and the second time.
  • the second time period at the second time is used to instruct the interfering node to pass the shared channel before the second time.
  • the time interval between the time when the last successful message was sent and the second time; the interfering node is the node monitored by the second node that sends the message through the shared channel except the first node; the second time is the time in the target cycle. any moment.
  • FIG. 5 is a schematic diagram of a first time period and a second time period provided by an embodiment of the present application.
  • Node0 (Node 0) represents the first node in the embodiment of the present application
  • Nodei ⁇ 0 represents the interfering node in the embodiment of the present application
  • the response information (Acknowledge character, ACK) represents the response time of the node successfully sending the message, such as As shown in Figure 5, the time from node 0 to the last successful transmission That is, the first time period in this embodiment of the present application, the time from the last successful transmission of any node that interferes with the receiving segment That is, the second time period in this embodiment of the present application.
  • step S101 includes some or all of the following steps.
  • the first node acquires the first time period of each moment in the target period.
  • the first node may receive a response to the message, and according to the response may determine the first time period of the moment when the response is received.
  • the first node When receiving a response that the message is sent successfully, the first node can determine that the first time period of the time when the response is received is 0; when receiving a failure to send the message, the first node can determine the time when the response is received.
  • the first time period of is the time interval from the last time the first node is cleared to the moment when the above response is received. For example, if the first node receives the response that the above-mentioned message is successfully sent at the first moment, the first time period at the first moment is 0; the first node receives the response that the above-mentioned message is not sent at the first moment, and the first time period is 0. If the time interval from the last time the second time period is cleared to zero is t, the first node determines that the first time period at the first moment is t.
  • the second time period of the time when the first node does not send the message is the time interval from the time when the first node successfully sends the message before the time to the time.
  • the first node acquires the second time period at each moment in the target period.
  • the first node may monitor the message sent by the interfering node, determine the time when the interfering node successfully sent the message last time according to the foregoing message, and obtain the second time period of the current moment according to the foregoing time.
  • the first node can demodulate all received packets, and according to the ID (MAC address) and type of the packets, determine whether the packet comes from an interfering node of the second node and whether the packet is response information , the method has high complexity and high power consumption.
  • ID MAC address
  • type of the packets determine whether the packet comes from an interfering node of the second node and whether the packet is response information , the method has high complexity and high power consumption.
  • the following describes three implementation manners for obtaining the second time period at each moment in the target period provided by the embodiments of the present application.
  • the foregoing implementation manners are an immediate response method, which can reduce complexity and power consumption.
  • the data packet sent by the interfering node includes first indication information, where the first indication information is used to instruct the node receiving the data packet to send a response after receiving the data packet for a first duration information.
  • the first node may monitor data packets of a first target range, where the first target range may be the maximum range in which the first node can monitor packets.
  • the first node may monitor whether there is response information after the first time period.
  • the first node may determine that the data packet is successfully sent, and then determine the second time period at the current moment as 0.
  • FIG. 6A is a schematic diagram of acquiring a second time period according to an embodiment of the present application.
  • the first indication information may be on the physical layer header of the message, and the first node may monitor the response after a short interframe space (Short interframe space, SIFS).
  • SIFS short interframe space
  • the first node may determine the second time period when the response message is monitored according to the indication information in the monitored response message.
  • the indication information in the response message may be indication information indicating that the message corresponding to the response message fails to be sent, or may be indication information indicating that the message corresponding to the response message is successfully sent.
  • the first node monitors the response information including the second indication information at the second moment, it determines that the second time period at the second moment is 0, and the second indication information is used to indicate that the response information including the second indication information corresponds to The sending result of the message is successful; when the first node monitors the response information including the third indication information at the second moment, it determines that the second time period collected at the second moment is the second time period of the previous collection moment at the second moment.
  • the sum of the second time period and the first time interval, the first time interval is the time interval between the last moment of the second moment and the second moment in the target period, and the third indication information is used to indicate that the response information including the second indication information corresponds to The sending result of the message is failed.
  • FIG. 6B is a schematic diagram of a response message provided by an embodiment of the present application.
  • the indication information in the above-mentioned response message may be that the packet header of the response packet carries special bits, phases, and domain values, etc., and the packet headers carry special bits, phases, and domain values. Rules for distinguishing the sending structure of the message corresponding to the response message.
  • the first node determines the second time period at each moment according to the broadcast frame sent by the second node.
  • the first node when receiving the second broadcast frame at the second moment, may determine that the second time period of the second moment is 0, and the second broadcast frame is the second broadcast frame that is monitored by the second node through the sharing of the interfering node Broadcast to the first node when the channel successfully sends a message.
  • the second node is an AP and other nodes are STAs
  • the second node can generate a second broadcast frame when it obtains a response that the interfering node successfully sends the message
  • the second broadcast frame is broadcast to every STA in the network.
  • each STA in the network can execute the channel access method provided by the embodiment of the present application, so each STA can determine the second time period at each moment by using the second broadcast frame broadcast by the second node.
  • the method can effectively use the data obtained by the second node, and at the same time, it improves the speed at which each node in the network obtains network state information, thereby shortening the time for each node in the network to decide to access the channel, and improving the speed of the network. communication speed.
  • the second broadcast frame includes a first time point, and the first time point is the time when the second node monitors that the interfering node successfully sends the message through the shared channel.
  • the first node may determine the second time period according to the second time point in the second broadcast frame. Specifically, the first node may determine the time interval between the last second time period at the second time and the second time interval as the second time interval; when the first time interval is smaller than the second time interval, the first time interval The node determines that the second time period of the second moment is 0. It can be understood that this method can solve the problem that when the second broadcast frame is lost or delayed, the first node receives the second broadcast frame earlier than the time when the second time period was cleared last time, and clears the second broadcast frame again. time period issue.
  • first node may also obtain the second time period at any moment in other ways, which is not limited here.
  • the first node inputs the network state information at each moment into the target neural network to obtain a prediction result, where the prediction result includes a first prediction value and a second prediction value.
  • the first node may input the first time period and the second time period of each moment in the target period into the target neural network model to obtain the first predicted value and the second predicted value.
  • the first predicted value is used to indicate the probability that the first node successfully sends a message to the second node through the shared channel at the first moment
  • the second predicted value is used to indicate that the first node is in the first The probability of failure to send a message to the second node through the shared channel at a moment.
  • the above-mentioned target neural network may be obtained by taking the network state information at each moment in the sample period as the input and the result of sending the message at the sample moment as the label, and the sample period is the sample including the sample moment.
  • the network state information at each moment in the sample period includes a first time period and a second time period at each moment.
  • the target neural network can be a deep neural network (Deep Neural Networks, DNN), a gated recurrent unit (Gate Recurrent Unit, GRU), or a long short-term memory network (Long Short-Term Memory, LSTM) and other neural networks.
  • DNN Deep Neural Networks
  • GRU Gate Recurrent Unit
  • LSTM Long Short-Term Memory
  • FIG. 7 is a schematic diagram of a target neural network provided by an embodiment of the present application.
  • x represents the network state at each moment.
  • the target cycle has T moments
  • the first node inputs the network state at T moments into the target neural network
  • x t represents the network state information at the first moment.
  • the output at the first moment is y t .
  • the first node can use the loss function shown below to train the target neural network model:
  • r t represents the reward value at the first moment
  • ⁇ represents the discount factor
  • E represents the experience pool
  • NE represents the number of experience e t in the experience pool
  • Q(s t+1 , a'; ⁇ - ) represents the output of the target neural network
  • ⁇ - is the parameter of the target neural network
  • Q(s t , a t ; ⁇ ) represents the output of the target neural network
  • ⁇ is the parameter of the target neural network.
  • the training process can use the mini-batch gradient descent method to update the parameters of the target neural network.
  • C can be 100.
  • an optimization parameter h will be generated in each training process. As shown in Figure 7, h t-1 will be generated in the training where the input is x t-T+1 , and h t-1 and h t-1 will be generated in the next training. x t-T+1 is used as input for training, resulting in h t .
  • the first node sends the message to be sent to the second node through the shared channel.
  • the first node can compare the size of the first predicted value and the second predicted value, and when the first predicted value is greater than the second predicted value, the first node can share the The channel sends the message to be sent to the second node; when the first predicted value is smaller than the second predicted value, the first node does not send the message to be sent to the second node through the shared channel.
  • the first node may determine whether to send the message to be sent to the second node according to the function value of the state-action value function of sending and not sending obtained in S202. Specifically, if Q(s t ,0)>Q(s t ,1), the first node sends the message to be sent to the second node, otherwise, the first node does not send the message to be sent to the second node.
  • FIG. 8 is a flowchart of another channel access method provided by an embodiment of the present application.
  • the method can be applied to the channel access network shown in FIG. 2 or FIG. 3 , and the method can include some or all of the following steps.
  • the first node acquires network status information at each moment in a target period, where the target period is a time period with a preset duration before the first moment including the first moment, and the network status information includes the network status information monitored by the first node.
  • the first node obtains network status information at each moment in the target period during the postgraduate entrance examination.
  • the target period is a time period with a preset duration before the first time including the first time;
  • the network status information includes the busy and idle state of the shared channel monitored by the first node, the first time period, and the second time period and the total number of interfering nodes
  • the first time period at the second time is used to indicate the time interval between the time when the first node successfully sent a message through the shared channel last time before the second time and the second time
  • the second time is used to indicate the time interval between the time when the interfering node successfully sends a message through the shared channel last time before the second time and the second time
  • the interfering node is the second node except the first node monitored by the second node.
  • the node that sends the message through the shared channel; the second moment is any moment in the target period.
  • step S201 includes some or all of the following steps:
  • the first node acquires the busy and idle status of the shared channel at each moment in the target period.
  • the first node may acquire the busy and idle status of the shared channel at each moment of the target period through the carrier sense mechanism.
  • the first node when the first node detects that other nodes are using the shared channel to transmit messages, it can know that the shared channel at that moment is in a busy state; when the first node detects that no other nodes are using the shared channel to transmit messages , it can be obtained that the shared channel at this moment is in an idle state.
  • the first node may record the busy and idle state of the shared channel according to the preset regulation.
  • the method for the first node to acquire the busy and idle state of the shared channel is not limited here.
  • the first node acquires the total number of interfering nodes at each moment in the target period.
  • the second node may acquire the total number of interfering nodes at each moment, and when acquiring the total number of interfering nodes, generate a broadcast frame including the total number of interfering nodes, and broadcast the broadcast frame to the first node, corresponding , the first node can parse the broadcast frame to the number of interfering nodes at each moment.
  • the first node receives the first broadcast frame broadcast by the second node, where the first broadcast frame includes the total number of interfering nodes; the first node parses the total number of interfering nodes from the first broadcast frame Determined as the total number of interfering nodes at the second instant.
  • FIG. 9A is a schematic diagram of an information unit of a broadcast frame provided by an embodiment of the present application.
  • the information element may include an element identifier (Element ID), a length (Length), an element identifier extension (Element ID Extension), and the total number of interfering STAs (Number of interfering STAs).
  • the network architecture may be as shown in FIG. 3 , wherein the second node is an AP, and the first node and the interfering node are STAs related to the AP. Specifically, when acquiring the total number of interfering nodes, the second node may broadcast the total number of interfering nodes to each STA under the same network architecture through a beacon. Correspondingly, each of the above-mentioned STAs (including the first node) can receive the broadcast frame, and parse the broadcast frame to obtain the total number of the above-mentioned interfering nodes.
  • the second node may broadcast the total number of interfering nodes to the first node through a probe request frame.
  • FIG. 9B is a schematic diagram of another information unit of a broadcast frame provided by an embodiment of the present application.
  • the total number of interfering nodes may exist in a field (A-control subfield) in the HT control field (HT control field).
  • HT control field HT control field
  • Control List includes multiple Controls.
  • One of the Controls may include Control ID and Control information. Number of interfering STAs can exist in the above Control information.
  • the detailed content of the first time period and the second time period obtained by the first node at each moment may refer to the relevant content in step S101, and details are not repeated here.
  • the first node inputs the network state information at each moment into the target neural network to obtain a prediction result, where the prediction result includes a first prediction value and a second prediction value.
  • the first node can input the busy-idle status of the shared channel, the first time period, the second time period and the total number of interfering nodes monitored by the first node at each moment in the target period into the target neural network model , get the first predicted value and the second predicted value.
  • the first prediction value is used to indicate the probability that the first node successfully sends a message to the second node through the shared channel at the first moment
  • the second prediction value is used to indicate the first node to the second node through the shared channel at the first moment. The probability that the two nodes fail to send packets.
  • the above-mentioned target neural network may be obtained by taking the network state information at each moment in the sample period as the input and the result of sending the message at the sample moment as the label, and the sample period is the sample including the sample moment.
  • the network status information at each moment in the sample period includes the busy and idle status of the shared channel, the first time period, the second time period, and the total number of interfering nodes monitored by the first node at each moment.
  • FIG. 7 is a schematic diagram of a target neural network provided by an embodiment of the present application.
  • x represents the network state at each moment.
  • the target cycle has T moments.
  • the first node inputs the network state information at T moments into the target neural network, and the output y t at the first moment can be obtained. .
  • the network status information at the first moment is: Among them, o t-1 represents the carrier sense result at the previous moment, represents the first time period of the first moment, represents the second time period at the first moment, and N represents the number of interfering nodes at the first moment.
  • Q(s t ,0) and Q(s t ,1) represent the function values of the state-action value function (state-action value function) that are sent and not sent, respectively, where Q(s t ,0) is the first The first predicted value at the moment, Q(s t ,1) is the second predicted value at the first moment.
  • the first node sends the to-be-sent message to the second node through the shared channel.
  • the first node can compare the size of the first predicted value and the second predicted value, and when the first predicted value is greater than the second predicted value, the first node can share the The channel sends the message to be sent to the second node; when the first predicted value is smaller than the second predicted value, the first node does not send the message to be sent to the second node through the shared channel.
  • the first node may determine whether to send the message to be sent to the second node according to the function value of the state-action value function of sending and not sending obtained in S202. Specifically, if Q(s t ,0)>Q(s t ,1), the first node sends the message to be sent to the second node, otherwise, the first node does not send the message to be sent to the second node.
  • the channel access method further includes some or all of the following steps:
  • the first node may wait for a response after sending the to-be-sent message to the second node through the shared channel, and then, according to the response, determine the sending result of the to-be-sent message. For example, when a response with indication information indicating success is received, it is determined that the transmission result of the message to be sent is successful: when a response with indication information indicating failure is received, it is determined that the transmission result of the message to be sent is fail.
  • the first node takes the network state information of the target period as an input and uses the sending result of the message to be sent as a label to update the target neural network.
  • the first node can update the target neural network according to the loss, and the loss includes the error between the prediction result and the sending result of the to-be-sent message.
  • the first node can obtain the loss according to the sending result of the message to be sent and the prediction result of the target neural network, and then update the target neural network according to the loss,
  • the loss may also include a reward value at the first moment, and the reward value is determined according to the sending result of the to-be-sent message and the busy-idle status of the shared channel monitored by the first node at the first moment.
  • the reward value is determined according to the sending result of the to-be-sent message and the busy-idle status of the shared channel monitored by the first node at the first moment.
  • the first node can use the loss function shown below to train the target neural network model:
  • r t represents the reward value at the first moment
  • ⁇ represents the discount factor
  • E represents the experience pool
  • NE represents the number of experience e t in the experience pool
  • Q(s t+1 , a'; ⁇ - ) represents the output of the target neural network
  • ⁇ - is the parameter of the target neural network
  • Q(s t , a t ; ⁇ ) represents the output of the target neural network
  • ⁇ is the parameter of the target neural network.
  • the following introduces a method for calculating a reward value provided by an embodiment of the present application.
  • the decision result is that the first node sends a message to the second node at the first moment. Furthermore, the first node may determine the reward value at the first moment according to the sending result of the message.
  • the first node may determine that the above-mentioned reward value is the ratio of the first time period to the second time period when the sending result of the to-be-sent message is successful;
  • the reward value is the negative value of the total number of interfering nodes at the first moment.
  • the decision result is that the first node does not send a packet to the second node at the first moment. Then, the first node can monitor the busy-idle state of the shared channel at the first moment, and determine the reward value at the first moment.
  • the first node may determine that the above-mentioned reward value is the ratio of the second time period at the first moment to the first time period at the first moment when monitoring the shared channel at the first moment that there is response information indicating that the packet is sent successfully ;
  • the first node listens at the first moment that the shared channel has response information indicating that the message transmission fails, it determines that the above-mentioned return value is M, where M is a positive number, for example, N can be 1;
  • M is a positive number, for example, N can be 1
  • N can be 1
  • FIG. 10 is a schematic structural diagram of a channel access apparatus 400 according to an embodiment of the present application. in:
  • the obtaining unit 410 is configured to obtain network status information at each moment in a target period, where the target period is a time period with a preset duration before the first moment including the first moment; the network status information includes The first time period and the second time period, the first time period at the second moment is used to indicate the time when the first node successfully sends a packet through the shared channel last time before the second moment and the second moment The second time period at the second time is used to indicate the time interval between the time when the interfering node successfully sends a packet through the shared channel last time before the second time and the second time. time interval; the interfering node is a node that is monitored by the second node and sends packets through the shared channel except the first node; the second moment is any moment in the target period ;
  • the prediction unit 420 is configured to input the network state information at each moment into the target neural network to obtain a prediction result, where the prediction result includes a first prediction value and a second prediction value, and the first prediction value is used to indicate the the probability that the first node successfully sends a packet to the second node through the shared channel at the first moment, and the second predicted value is used to indicate that the first node passes the the probability that the shared channel fails to send the message to the second node;
  • the decision unit 430 is configured to send, by the first node, a message to be sent to the second node through the shared channel when the first predicted value is greater than the second predicted value.
  • the obtaining unit 410 is further configured to obtain the busy and idle states of the shared channel and the total number of the interfering nodes monitored by the first node.
  • the obtaining unit 410 is specifically configured to obtain the second time period of the second moment, wherein:
  • the first node listens to the response information at the second time according to the first indication information, it determines that the second time period of the second time is 0, and the second time is when the first node listens to the first time.
  • the packet header of the first packet includes the first indication information, and the first indication information is used to indicate that the node receiving the first packet is in the The response information is sent after the first time period.
  • the obtaining unit 410 is specifically configured to obtain the second time period of the second moment, wherein:
  • the first node monitors the response information including the second indication information at the second moment, it determines that the second time period at the second moment is 0, and the second indication information is used to indicate that the second time period includes the second indication information. 2.
  • the sending result of the message corresponding to the response information of the indication information is successful;
  • the first node monitors the response information including the third indication information at the second moment, it determines that the second time period of the second moment is the time period preceding the second moment in the target period.
  • the sending result of the message corresponding to the response information is failed.
  • the obtaining unit 410 is specifically configured to obtain the total number of interfering nodes at the second moment, wherein:
  • the first node receives the first broadcast frame broadcast by the second node
  • the first node determines the total number of interfering nodes parsed from the first broadcast frame as the total number of interfering nodes at the second moment.
  • the obtaining unit 410 is specifically configured to obtain the second time period of the second moment, wherein:
  • the first node When the first node receives the second broadcast frame at the second moment, it determines that the second time period of the second moment is 0, and the second broadcast frame is the second broadcast frame that is monitored by the second node. It is broadcast to the first node when the interfering node successfully sends the message through the shared channel.
  • the second broadcast frame includes a first time point, and the first time point is the time when the second node monitors that the interfering node successfully sends a packet through the shared channel.
  • the obtaining unit 410 is specifically configured to execute:
  • the first node determines the time interval between the first time point and the second time as the first time interval
  • the channel access apparatus further includes a training unit 440, and the training unit 440 is configured to perform:
  • the first node takes the network state information at each moment of the target period as an input and updates the target neural network with the sending result of the to-be-sent message as a label.
  • the channel access apparatus further includes a training unit 440, and the training unit 440 is specifically configured to perform: updating the target neural network according to a loss, where the loss includes comparing the prediction result with the to-be-to-be The error between the transmission results of the transmitted message.
  • the loss further includes a reward value at the first moment, where the reward value is based on the prediction result and the availability of the shared channel monitored by the first node at the first moment status determined.
  • the channel access apparatus further includes a training unit 440, and the training unit 440 is further configured to perform:
  • the sending result of the message to be sent is a failure, it is determined that the reward value is a negative value of the total number of interfering nodes at the first moment.
  • the channel access apparatus further includes a training unit 440, and the training unit 440 is further configured to perform:
  • the first node monitors the shared channel at the first moment
  • the first node monitors at the first moment that the shared channel has response information indicating that the message is sent successfully, it determines that the reward value is the second time period of the first moment and the first moment. the ratio of the first time period;
  • the first node monitors at the first moment that the shared channel has response information indicating that the message transmission fails, it determines that the reward value is M, and M is a positive number;
  • the first node determines that the reward value is 0 when monitoring at the first moment that no response information exists on the shared channel.
  • the apparatus 400 may be a device on the side of the access point AP, or a chip in the access point AP, so as to support the access point AP to implement the corresponding functions in the method; the apparatus 400 may also be on the side of the station STA The device, or a chip in the station STA, is used to support the STA to implement the corresponding functions in the method.
  • FIG. 11 shows another channel access apparatus 500 provided by an embodiment of the present application.
  • the channel access apparatus 500 includes at least a processor 510 , a memory 520 and a transceiver 530 , and the processor 510 , the memory 520 and the transceiver 530 are connected to each other through a bus 540 .
  • the memory 520 includes, but is not limited to, a random access memory (Random Access Memory, RAM), a read-only memory (Read-Only Memory, ROM), or an erasable programmable read-only memory (Erasable Programmable Read-Only Memory, EPROM or fast memory). Flash memory), the memory 520 is used to store related instructions and data.
  • RAM Random Access Memory
  • ROM read-only memory
  • EPROM Erasable Programmable Read-Only Memory
  • Flash memory Flash memory
  • the transceiver 530 may include a receiver and a transmitter, for example, a radio frequency module.
  • the processor 510 described below receives or sends a certain message. Specifically, it can be understood that the processor 510 receives or sends a message through the transceiver 530. .
  • the processor 510 may be one or more central processing units (Central Processing Units, CPUs). When the processor 510 is a CPU, the CPU may be a single-core CPU or a multi-core CPU.
  • CPUs Central Processing Units
  • the processor 510 in the channel access device 500 is configured to read the program codes stored in the memory 520, and perform the following operations:
  • the first node acquires network status information at each moment in a target period, where the target period is a time period of a preset duration before the first moment including the first moment; the network status information includes the first moment
  • the first time period at the second moment is used to indicate the difference between the time when the first node successfully sent a packet through the shared channel last time before the second moment and the second moment.
  • time interval, the second time period of the second time is used to indicate the time interval between the time when the interfering node successfully sends a message through the shared channel last time before the second time and the second time;
  • the interfering node is a node that is monitored by the second node and sends a message through the shared channel except the first node; the second moment is any moment in the target period;
  • the first node inputs the network state information at each moment into the target neural network to obtain a prediction result, where the prediction result includes a first prediction value and a second prediction value, and the first prediction value is used to indicate the the probability that the first node successfully sends a packet to the second node through the shared channel at the first moment, and the second predicted value is used to indicate that the first node passes the the probability that the shared channel fails to send the message to the second node;
  • the first node When the first predicted value is greater than the second predicted value, the first node sends the message to be sent to the second node through the shared channel.
  • the network status information further includes the busy and idle status of the shared channel monitored by the first node and the total number of the interfering nodes.
  • acquiring, by the first node, network status information at each moment in the target period includes: acquiring a second time period of the second moment;
  • the acquiring the second time period of the second moment includes:
  • the first node listens to the response information at the second time according to the first indication information, it determines that the second time period of the second time is 0, and the second time is when the first node listens to the first time.
  • the packet header of the first packet includes the first indication information, and the first indication information is used to indicate that the node receiving the first packet is in the The response information is sent after the first time period.
  • acquiring, by the first node, network status information at each moment in the target period includes: acquiring a second time period of the second moment;
  • the acquiring the second time period of the second moment includes:
  • the first node monitors the response information including the second indication information at the second moment, it determines that the second time period at the second moment is 0, and the second indication information is used to indicate that the second time period includes the second indication information. 2.
  • the sending result of the message corresponding to the response information of the indication information is successful;
  • the first node monitors the response information including the third indication information at the second moment, it determines that the second time period of the second moment is the time period preceding the second moment in the target period.
  • the sending result of the message corresponding to the response information is failed.
  • acquiring, by the first node, network status information at each moment in the target period includes: acquiring the total number of interfering nodes at the second moment;
  • the first node receives the first broadcast frame broadcast by the second node
  • the first node determines the total number of interfering nodes parsed from the first broadcast frame as the total number of interfering nodes at the second moment.
  • acquiring, by the first node, network status information at each moment in the target period includes: acquiring a second time period of the second moment;
  • the acquiring the second time period of the second moment includes:
  • the first node When the first node receives the second broadcast frame at the second moment, it determines that the second time period of the second moment is 0, and the second broadcast frame is the second broadcast frame that is monitored by the second node. It is broadcast to the first node when the interfering node successfully sends the message through the shared channel.
  • the second broadcast frame includes a first time point, and the first time point is the time when the second node monitors that the interfering node successfully sends a packet through the shared channel .
  • determining that the second time period of the second moment is 0, including:
  • the first node determines the time interval between the first time point and the second time as the first time interval
  • the method further includes:
  • the first node takes the network state information at each moment of the target period as an input and updates the target neural network with the sending result of the to-be-sent message as a label.
  • the first node takes the network state information of the target period as input and uses the sending result of the to-be-sent message as a label to update the target neural network, including:
  • the target neural network is updated according to a loss, the loss including the error between the prediction result and the transmission result of the to-be-sent message.
  • the loss further includes a reward value at the first moment, where the reward value is based on the prediction result and the shared channel monitored by the first node at the first moment is determined by the availability status.
  • the sending result of the message to be sent is a failure, it is determined that the reward value is a negative value of the total number of interfering nodes at the first moment.
  • the first node monitors the shared channel at the first moment
  • the first node monitors, at the first moment, that the shared channel has response information indicating that the message is successfully sent, determine that the reward value is the second time period of the first moment and the first moment The ratio of the first time period;
  • the first node monitors at the first moment that the shared channel has response information indicating that the message transmission fails, it determines that the reward value is M, and M is a positive number;
  • the first node determines that the reward value is 0 when monitoring at the first moment that no response information exists on the shared channel.
  • the apparatus 500 may be an apparatus on the side of the access point AP, or a chip in the access point AP; the apparatus 500 may also be an apparatus on the side of the station STA, or a chip in the station STA.
  • Embodiments of the present application also provide a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, when the computer or processor runs on the computer or processor, the computer or processor executes any one of the foregoing physical layer protocol data One or more steps in a channel access method for a unit. If each component module of the above device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in the computer-readable storage medium.
  • the above-mentioned computer-readable storage medium may be an internal storage unit of the channel access device described in any of the foregoing embodiments, such as a hard disk or a memory of the channel access device.
  • the above-mentioned computer-readable storage medium can also be an external storage device of the above-mentioned channel access device, such as a plug-in hard disk equipped on the above-mentioned channel access device, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, flash memory card (Flash Card), etc.
  • the above-mentioned computer-readable storage medium may also include both an internal storage unit of the above-mentioned channel access apparatus and an external storage device.
  • the above-mentioned computer-readable storage medium is used for storing the above-mentioned computer program and other programs and data required by the above-mentioned channel access device.
  • the above-mentioned computer-readable storage medium can also be used to temporarily store data that has been output or is to be output.
  • the program can be stored in a computer-readable storage medium, and when the program is executed , which may include the processes of the above-mentioned method embodiments.
  • the aforementioned storage medium includes various media that can store program codes, such as ROM, RAM, magnetic disk, or optical disk.
  • the modules in the apparatus of the embodiment of the present application may be combined, divided and deleted according to actual needs.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请实施例公开了一种信道接入方法和装置。在该方法中，第一节点获取目标周期内每一个时刻的网络状态信息，其中，网络状态信息包括第一节点监听到的共享信道的忙闲状态、第一时间段、第二时间段和干扰节点的总数，干扰节点为第二节点的监听到的除第一节点外通过共享信道发送报文的节点，第一节点将上述每一个时刻的网络状态信息输入目标神经网络，得到第一预测值和第二预测值，在第一预测值大于第二预测值时，第一节点通过共享信道向第二节点发送待发送报文。实施本技术方案，节点可以根据当前的网络状态决策是否接入共享信道，该方法能够提高节点接入信道的成功率，增加信道的吞吐量。

Description

一种信道接入方法和装置
本申请要求于2020年12月29日提交中国国家知识产权局、申请号为202011603244.0、申请名称为“一种信道接入方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本方案涉及通信领域，尤其涉及一种信道接入方法和装置。
背景技术
在短距离传输网络系统和无线保真(wireless fidelity，WIFI)系统等无线网络中，信道是共享的，也即是，在该无线网络中的多个节点会使用同一共享信道进行通信。
在该网络中，如果两个或两个以上的节点同时发送报文，会产生通信冲突，出现节点的报文传输失败的情况，导致该信道的吞吐量降低和网络中的节点通信的时延增加，极大影响了用户体验。可见，节点信道接入决策对通信起到重要的影响，对用户体验起到决定性作用。
在分布式场景中，没有中心节点对网络中的所有节点进行统一的调度和协调，直接由节点决策是否接入共享信道。目前，节点通常可以通过随机接入共享信道，或者获取大量数据来预测接入共享信道的结果而决策是否接入信道，该方法成功接入信道的概率较低，信道的吞吐量少。
发明内容
本申请实施例提供了一种信道接入方法和装置。该方法中，第一节点可以在通过共享信道向第二节点发送报文前，将获取到的目标周期内每一个时刻的网络状态信息输入目标神经网络中，得到第一时刻向第二节点发送报文的预测结果，最后，根据该预测结果确定是否通过该共享信道向第二节点发送报文。
其中，目标周期为包括第一时刻在内的第一时刻之前的预设时长的时间段；网络状态信息可以包括第一时间段、第二时间段、第一节点监听到的共享信道的忙闲状态和干扰节点的总数，第二时刻的第一时间段用于指示第一节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔，第二时刻的第二时间段用于指示干扰节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔；干扰节点为第二节点的监听到的除第一节点外通过共享信道发送报文的节点；第二时刻为目标周期中的任意一个时刻。
实施本技术方案，节点可以根据过去一段时间的网络状态决策是否接入共享信道，该方法能够提高节点接入信道的成功率，增加信道的吞吐量。
第一方面，本申请实施例提供了一种信道接入方法，其特征在于，该方法包括：
第一节点获取目标周期内每一个时刻的网络状态信息，上述目标周期为包括第一时刻在内的上述第一时刻之前的预设时长的时间段；网络状态信息包括第一时间段和第二时间段，第二时刻的第一时间段用于指示第一节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔，第二时刻的第二时间段用于指示干扰节点在第二时刻 之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔；干扰节点为第二节点的监听到的除第一节点外通过共享信道发送报文的节点；第二时刻为目标周期中的任意一个时刻；
第一节点将每一个时刻的网络状态信息输入目标神经网络，得到预测结果，预测结果包括第一预测值和第二预测值，第一预测值用于指示第一节点在第一时刻通过共享信道向第二节点发送报文成功的概率，第二预测值用于指示第一节点在第一时刻通过共享信道向第二节点发送报文失败的概率；
在第一预测值大于第二预测值时，第一节点通过共享信道向第二节点发送待发送报文。
结合第一方面，在一种可能的实现方式中，网络状态信息还包括第一节点监听到的共享信道的忙闲状态和干扰节点的总数。
结合第一方面，在一种可能的实现方式中，第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取第二时刻的第二时间段；
获取第二时刻的第二时间段，包括：
第一节点根据第一指示信息在第二时刻监听到响应信息时，确定第二时刻的第二时间段为0，第二时刻为第一节点监听到第一报文的第一时长后的时刻，第一报文的包头包括第一指示信息，第一指示信息用于指示在接收到第一报文的节点在第一时长后发送响应信息。
结合第一方面，在一种可能的实现方式中，第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取第二时刻的第二时间段；
获取第二时刻的第二时间段，包括：
第一节点在第二时刻监听到包括第二指示信息的响应信息时，确定第二时刻的第二时间段为0，第二指示信息用于指示包括第二指示信息的响应信息对应的报文的发送结果为成功；
第一节点在第二时刻监听到包括第三指示信息的响应信息时，确定第二时刻的第二时间段为目标周期内第二时刻的上一个时刻的第二时间段与第一时间间隔之和，第一时间间隔为上一个时刻与第二时刻的时间间隔，第三指示信息用于指示包括第二指示信息的响应信息对应的报文的发送结果为失败。
结合第一方面，在一种可能的实现方式中，第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取第二时刻的干扰节点的总数；
获取第二时刻的干扰节点的总数，包括：
在第二时刻时，第一节点接收第二节点广播的第一广播帧；
第一节点将从第一广播帧中解析到的干扰节点的总数确定为在第二时刻的干扰节点的总数。
结合第一方面，在一种可能的实现方式中，第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取第二时刻的第二时间段；
获取第二时刻的第二时间段，包括：
第一节点在第二时刻接收到第二广播帧时，确定第二时刻的第二时间段为0，第二广播帧是由第二节点在监听到干扰节点通过共享信道成功发送报文时广播至第一节点的。
结合第一方面，在一种可能的实现方式中，第二广播帧包括第一时间点，第一时间点为第二节点监听到干扰节点通过共享信道成功发送报文的时刻。
结合第一方面，在一种可能的实现方式中，第一节点在第二时刻接收到第二广播帧时，确定第二时刻的第二时间段为0，包括：
第一节点将第一时间点与第二时刻的时间间隔确定为第一时间间隔；
将第二时刻的上一个第二时间段清零的时刻与第二时刻的时间间隔确定为第二时间间隔；
在第一时间间隔小于第二时间间隔时，确定第二时刻的第二时间段为0。
结合第一方面，在一种可能的实现方式中，第一节点通过共享信道向第二节点发送待发送报文后，方法还包括：
第一节点获取待发送报文的发送结果；
第一节点以目标周期每一个时刻的网络状态信息为输入且以待发送报文的发送结果为标签更新目标神经网络。
结合第一方面，在一种可能的实现方式中，第一节点以目标周期的网络状态信息为输入且以待发送报文的发送结果为标签更新目标神经网络，包括：
根据损失更新目标神经网络，损失包括将预测结果与待发送报文的发送结果之间的误差。
结合第一方面，在一种可能的实现方式中，损失还包括第一时刻的回报值，回报值是根据预测结果和第一节点在第一时刻监听到的共享信道的忙闲状态决定的。
结合第一方面，在一种可能的实现方式中，在待发送报文的发送结果为成功时，确定回报值为第一时间段与第二时间段的比值；
在待发送报文的发送结果为失败时，确定回报值为第一时刻的干扰节点的总数的负值。
结合第一方面，在一种可能的实现方式中，在第一预测值小于第二预测值时，第一节点在第一时刻监听共享信道；
第一节点在第一时刻监听到共享信道有指示报文发送成功的响应信息时，确定回报值为第一时刻的第二时间段与第一时刻的第一时间段的比值；
第一节点在第一时刻监听到共享信道有指示报文发送失败的响应信息时，确定回报值为M，M为正数；
第一节点在第一时刻监听到共享信道不存在响应信息时，确定回报值为0。
第二方面公开一种信道接入装置，该装置包括：
获取单元，用于获取目标周期内每一个时刻的网络状态信息，目标周期为包括第一时刻在内的第一时刻之前的预设时长的时间段；网络状态信息包括第一时间段和第二时间段，第二时刻的第一时间段用于指示第一节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔，第二时刻的第二时间段用于指示干扰节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔；干扰节点为第二节点的监听到的除第一节点外通过共享信道发送报文的节点；第二时刻为目标周期中的任意一个时刻；
预测单元，用于将每一个时刻的网络状态信息输入目标神经网络，得到预测结果，预测结果包括第一预测值和第二预测值，第一预测值用于指示第一节点在第一时刻通过共享信道向第二节点发送报文成功的概率，第二预测值用于指示第一节点在第一时刻通过共享信道向第二节点发送报文失败的概率；
决策单元，用于在第一预测值大于第二预测值时，第一节点通过共享信道向第二节点发送待发送报文。
作为一种可能的实施方式，获取单元，还用于获取第一节点监听到的共享信道的忙闲状态和干扰节点的总数。
作为一种可能的实施方式，获取单元，具体用于获取第二时刻的第二时间段，其中：
第一节点根据第一指示信息在第二时刻监听到响应信息时，确定第二时刻的第二时间段 为0，第二时刻为第一节点监听到第一报文的第一时长后的时刻，第一报文的包头包括第一指示信息，第一指示信息用于指示在接收到第一报文的节点在第一时长后发送响应信息。
作为一种可能的实施方式，获取单元，具体用于获取第二时刻的第二时间段，其中：
第一节点在第二时刻监听到包括第二指示信息的响应信息时，确定第二时刻的第二时间段为0，第二指示信息用于指示包括第二指示信息的响应信息对应的报文的发送结果为成功；
第一节点在第二时刻监听到包括第三指示信息的响应信息时，确定第二时刻的第二时间段为目标周期内第二时刻的上一个时刻的第二时间段与第一时间间隔之和，第一时间间隔为上一个时刻与第二时刻的时间间隔，第三指示信息用于指示包括第二指示信息的响应信息对应的报文的发送结果为失败。
作为一种可能的实施方式，获取单元，具体用于获取第二时刻的干扰节点的总数，其中：
第二时刻时，第一节点接收第二节点广播的第一广播帧；
第一节点将从第一广播帧中解析到的干扰节点的总数确定为在第二时刻的干扰节点的总数。
作为一种可能的实施方式，获取单元，具体用于获取第二时刻的第二时间段，其中：
第一节点在第二时刻接收到第二广播帧时，确定第二时刻的第二时间段为0，第二广播帧是由第二节点在监听到干扰节点通过共享信道成功发送报文时广播至第一节点的。
作为一种可能的实施方式，第二广播帧包括第一时间点，第一时间点为第二节点监听到干扰节点通过共享信道成功发送报文的时刻。
作为一种可能的实施方式，获取单元具体用于执行：
第一节点将第一时间点与第二时刻的时间间隔确定为第一时间间隔；
将第二时刻的上一个第二时间段清零的时刻与第二时刻的时间间隔确定为第二时间间隔；
在第一时间间隔小于第二时间间隔时，确定第二时刻的第二时间段为0。
作为一种可能的实施方式，信道接入装置还包括训练单元，训练单元用于执行：
第一节点获取待发送报文的发送结果；
第一节点以目标周期每一个时刻的网络状态信息为输入且以待发送报文的发送结果为标签更新目标神经网络。
作为一种可能的实施方式，信道接入装置还包括训练单元，训练单元具体用于执行：根据损失更新目标神经网络，损失包括将预测结果与待发送报文的发送结果之间的误差。
作为一种可能的实施方式，损失还包括第一时刻的回报值，回报值是根据预测结果和第一节点在第一时刻监听到的共享信道的忙闲状态决定的。
作为一种可能的实施方式，信道接入装置还包括训练单元，训练单元还用于执行：
在待发送报文的发送结果为成功时，确定回报值为第一时间段与第二时间段的比值；
在待发送报文的发送结果为失败时，确定回报值为第一时刻的干扰节点的总数的负值。
作为一种可能的实施方式，信道接入装置还包括训练单元，训练单元还用于执行：
在第一预测值小于第二预测值时，第一节点在第一时刻监听共享信道；
第一节点在第一时刻监听到共享信道有指示报文发送成功的响应信息时，确定回报值为第一时刻的第二时间段与第一时刻的第一时间段的比值；
第一节点在第一时刻监听到共享信道有指示报文发送失败的响应信息时，确定回报值为M，M为正数；
第一节点在第一时刻监听到共享信道不存在响应信息时，确定回报值为0。
第三方面公开一种信道接入装置，该信道接入装置可以包括处理器、存储器、输入接口和输出接口，所述输入接口用于接收来自所述信道接入装置之外的其它信道接入装置的信息，所述输出接口用于向所述信道接入装置之外的其它信道接入装置输出信息，当所述处理器执行所述存储器存储的计算机程序时，使得所述处理器执行第一方面或第一方面的任一实施方式公开的信道接入方法。
第四方面公开一种信道接入网络，该信道接入网络包括第二方面的信道接入装置和第三方面的信道接入装置。
第五方面公开一种计算机可读存储介质，该计算机可读存储介质上存储有计算机程序或计算机指令，当该计算机程序或计算机指令运行时，实现如各方面公开的信道接入方法。
附图说明
下面对本申请实施例用到的附图进行介绍。
图1为本申请实施例提供的一种效果对比示意图；
图2为本申请实施例提供的一种信道接入网络架构示意图；
图3为本申请实施例提供的另一种信道接入网络架构示意图；
图4为本申请实施例提供的一种信道接入方法的流程图；
图5为本申请实施例提供的一种第一时间段和第二时间段的示意图；
图6A为本申请实施例提供的一种获取第二时间段的示意图；
图6B为本申请实施例提供的一种响应报文的示意图；
图7为本申请实施例提供的一种目标神经网络的示意图；
图8为本申请实施例提供的另一种信道接入方法的流程图；
图9A为本申请实施例提供的一种广播帧的信息单元的示意图；
图9B为本申请实施例提供的另一种广播帧的信息单元的示意图；
图10为本申请实施例提供的一种信道接入装置400的结构示意图；
图11为本申请实施例提供的另一种信道接入装置500的结构示意图。
具体实施方式
本申请以下实施例中所使用的术语只是为了描述特定实施例的目的，而并非旨在作为对本申请实施例的限制。如在本申请实施例的说明书和所附权利要求书中所使用的那样，单数表达形式“一个”、“一种”、“所述”、“”、“该”和“这一”旨在也包括复数表达形式，除非其上下文中明确地有相反指示。还应当理解，本申请实施例中使用的术语“和/或”是指并包含一个或多个所列出项目的任何或所有可能组合。
在本申请实施例中，第一节点为向第二节点发送报文的节点，第二节点为接收来自第一节点的报文的节点。由于第一节点、第二节点和干扰节点均通过一个共享信道进行通信，若干扰节点在第一节点接入该共享信道的同时通过该上述共享信道发送报文，则会影响第一节点向第二节发送报文。因此，在第一节点在向第二节点发送报文之前，第一节点可以执行本申请实施例提供的信道接入方法，确定是否接入上述共享信道。
本申请实施例提供了一种信道接入方法和装置。该方法中，第一节点可以在通过共享信道向第二节点发送报文前，将获取到的目标周期内每一个时刻的网络状态信息输入目标神经网络中，得到第一时刻向第二节点发送报文的预测结果，最后，根据该预测结果确定是否通 过该共享信道向第二节点发送报文。
其中，目标周期为包括第一时刻在内的第一时刻之前的预设时长的时间段；网络状态信息可以包括第一时间段、第二时间段、第一节点监听到的共享信道的忙闲状态和干扰节点的总数，第二时刻的第一时间段用于指示第一节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔，第二时刻的第二时间段用于指示干扰节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔；干扰节点为第二节点的监听到的除第一节点外通过共享信道发送报文的节点；第二时刻为目标周期中的任意一个时刻。
实施本技术方案，节点可以根据当前的网络状态决策是否接入共享信道，该方法能够提高节点接入信道的成功率，增加信道的吞吐量。
具体的，请参见图1，图1为本申请实施例提供的一种效果对比示意图。如图1中的(A)所示，本技术方案增加了信道的吞吐量；如图1中的(B)所示，本技术方案降低了节点之间通信的平均时延和最大时延。
为了更好地理解本申请实施例提供的一种信道接入方法和装置，下面先对本申请实施例使用的网络架构进行描述。
请参见图2，图2为本申请实施例提供的一种信道接入网络架构示意图。如图2所示，该网络架构包括第一节点10、第二节点20和干扰节点30，其中，干扰节点30可以包括多个节点，上述第一节点10、第二节点20和干扰节点30均通过同一个共享信道进行通信。其中：
具体的，第一节点10可以在通过共享信道向第二节点20发送报文前，将获取到的目标周期内每一个时刻的网络状态信息输入目标神经网络中，得到第一时刻向第二节点20发送报文的预测结果，最后，第一节点10根据该预测结果确定是否通过该共享信道向第二节点20发送报文。
第一节点10可以获取目标周期内每一个时刻的网络状态信息，其中，该目标周期为包括第一时刻在内的第一时刻之前的预设时长的时间段；该网络状态信息可以包括第一节点10监听到的共享信道的忙闲状态、第一时间段、第二时间段和干扰节点的总数，第二时刻的第一时间段用于指示第一节点10在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔，第二时刻的第二时间段用于指示干扰节点30在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔，第二时刻为目标周期中的任意一个时刻；干扰节点30为第二节点20的监听到的除第一节点10外通过共享信道发送报文的节点。进而，第一节点10将每一个时刻的网络状态信息输入目标神经网络，得到预测结果，预测结果包括第一预测值和第二预测值，第一预测值用于指示第一节点10在第一时刻通过共享信道向第二节点20发送报文成功的概率，第二预测值用于指示第一节点10在第一时刻通过共享信道向第二节点20发送报文失败的概率；在第一预测值大于第二预测值时，第一节点10通过共享信道向第二节点20发送待发送报文。
其中，本申请实施例中的第一节点10、第二节点20和干扰节点30中的任意一个节点可以为接入点AP，也可以为站点STA。
其中，站点STA可以指用户设备、接入终端、用户单元、用户站、移动站、移动台、远方站、远程终端、移动设备、用户终端、终端、无线通信设备、用户代理或用户装置。站点还可以是蜂窝电话、无绳电话、会话启动协议(session initiation protocol，SIP)电话、无线本地环路(wireless local loop，WLL)站、个人数字处理(personal digital assistant，PDA)、 具有无线通信功能的手持设备、计算设备或连接到无线调制解调器的其它处理设备、车载设备、可穿戴设备，未来5G网络中的终端设备或者未来演进的公用陆地移动通信网络(public land mobile network，PLMN)中的终端设备等，本申请实施例对此并不限定。
接入点AP可以是用于与站点通信的设备。该接入点可以是任意一种具有无线收发功能的设备或可设置于该设备的芯片，该设备包括但不限于：演进型节点B(evolved Node B，eNB)、无线网络控制器(radio network controller，RNC)、节点B(Node B，NB)、基站控制器(base station controller，BSC)、基站收发台(base transceiver station，BTS)、家庭基站(例如，home evolved NodeB，或home Node B，HNB)、基带单元(base band Unit，BBU)，无线保真(wireless fidelity，WIFI)系统中的接入点(access point，AP)、无线中继节点、无线回传节点、传输点(transmission point，TP)或者发送接收点(transmission and reception point，TRP)等，还可以为5G，如，NR，系统中的gNB，或，传输点(TRP或TP)，5G系统中的基站的一个或一组(包括多个天线面板)天线面板，或者，还可以为构成gNB或传输点的网络节点，如基带单元(BBU)，或，分布式单元(distributed unit，DU)等。
请参见图3，图3为本申请实施例提供的另一种信道接入网络架构示意图。如图3所示，信道接入网络架构包括第一节点、第二节点和干扰节点，其中，第二节点为AP，第一节点和干扰节点均为与第二节点连接的STA。可以理解的，第二节点可以获取同一网络下所有STA发送报文的响应。
在一种实现中，第二节点可以在获取同一网络下所有STA发送报文的响应时，根据获取到该响应的时刻生成广播帧，将上述广播帧广播至同一网络下的所有STA。可以理解的，接收到该广播帧的STA在执行本申请实施例的信道执行方法时，可以根据该广播帧更新第二时间段，详细内容可以参见以下本申请实施例的信道接入方法的相关内容。
需要说明的是，本申请的技术方案既可以应用于AP与AP之间的通信，还可以应用于AP与STA之间的通信，也可以应用于STA与STA之间的通信。下面以AP与STA之间的通信为例进行说明，并不构成对本申请实施例的限定。
可以理解的是，图2和图3中的信道接入网络架构只是本申请实施例的示例性的实施方式，本申请实施例中的信道接入网络架构包括但不仅限于以上信道接入网络架构。
请参考图4，图4为本申请实施例提供的一种信道接入方法的流程图，该方法可应用于上述图2或图3的信道接入网络中，该方法可以包括以下部分或全部步骤。
S101、第一节点获取目标周期内每一个时刻的网络状态信息，该目标周期为包括第一时刻在内的第一时刻之前的预设时长的时间段，该网络状态信息包括第一时间段和第二时间段。
具体的，第一节点可以获取目标周期内每一个时刻的网络状态信息。其中，目标周期为包括第一时刻在内的第一时刻之前的预设时长的时间段；网络状态信息可以包括第一时间段和第二时间段，第二时刻的第一时间段用于指示第一节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔，第二时刻的第二时间段用于指示干扰节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔；干扰节点为第二节点的监听到的除第一节点外通过共享信道发送报文的节点；第二时刻为目标周期中的任意一个时刻。
请参见图5，图5为本申请实施例提供的一种第一时间段和第二时间段的示意图。其中，Node0(节点0)表示本申请实施例中的第一节点，Nodei≠0表示本申请实施例中的干扰节点，响应信息(Acknowledge character，ACK)表示节点成功发送报文的响应时间，如图5所示， 节点0距离上次成功发送的时间
Figure PCTCN2021122982-appb-000001
即为本申请实施例中的第一时间段，任意干扰接收段的节点距离上次成功发送的时间
Figure PCTCN2021122982-appb-000002
即为本申请实施例中的第二时间段。
具体的，步骤S101包括如下部分或全部步骤。
S1011、第一节点获取目标周期内每一时刻的第一时间段。
具体的，第一节点可以在发送报文之后，接收该报文的响应，根据上述响应可以确定接收到上述响应的时刻的第一时间段。
在接收到报文发送成功的响应时，第一节点可以确定接收到上述响应的时刻的第一时间段为0；在接收到报文发送失败时，第一节点可以确定接收到上述响应的时刻的第一时间段为第一节点上一次清零到接收到上述响应的时刻的时间间隔。例如，第一节点在第一时刻接收到上述报文成功发送的响应，则第一时刻的第一时间段为0；第一节点在第一时刻接收到上述报文发送失败响应，第一时刻距离上次第二时间段清零的时间间隔为t，则第一节点确定第一时刻的第一时间段为t。
需要说明的是，第一节点没有发送报文的时刻的第二时间段为该时刻之前第一节点成功发送报文的时刻距离该时刻的时间间隔。
S1012、第一节点获取目标周期内每一时刻的第二时间段。
具体的，第一节点可以监听干扰节点发送的报文，根据上述报文确定干扰节点上一次成功发送报文的时间，根据上述时间，获取当前时刻的第二时间段。
在一些实施例中，第一节点可以解调全部接收的报文，根据报文的ID(MAC地址)和类型，判断该报文是否来自第二节点的干扰节点以及该报文是否为响应信息，该方法复杂度高，功耗较大。
以下介绍本申请实施例提供的三种获取目标周期内每一时刻的第二时间段的实现方式，上述实现方式为立即响应的方法，能够降低复杂度和功耗。
在一种实现中，干扰节点发送的数据报文中包括第一指示信息，该第一指示信息用于指示接受到该数据报文的节点在接收到该数据报文的第一时长后发送响应信息。
具体的，在目标周期内的每一个时刻，第一节点可以监听第一目标范围的数据报文，其中，第一目标范围可以为第一节点能够监听到报文的最大范围。第一节点在监听到包含第一指示信息的数据报文时，可以在第一时长后监听是否有响应信息。在监听到响应信息时，第一节点可以确定该数据报文发送成功，进而，将当前时刻的第二时间段确定为0。
请参见图6A，图6A为本申请实施例提供的一种获取第二时间段的示意图。其中，该第一指示信息可以在报文的物理层包头上，第一节点可以在短帧间间隔(Short interframe space，SIFS)之后监听该响应。可以理解的，第一时长越短，则第一节点更新第二时间段的时延就越短，第二时间段的准确度就越高。
在另一种实现中，第一节点可以根据监听到的响应报文中的指示信息，确定监听到响应报文的时刻的第二时间段。其中，该响应报文中的指示信息可以为指示该响应报文对应的报文发送失败的指示信息，也可以为指示该响应报文对应的报文发送成功的指示信息。
具体的，第一节点在第二时刻监听到包括第二指示信息的响应信息时，确定第二时刻的第二时间段为0，第二指示信息用于指示包括第二指示信息的响应信息对应的报文的发送结果为成功；第一节点在第二时刻监听到包括第三指示信息的响应信息时，确定第二时刻采集到的第二时间段为第二时刻的上一个采集时刻的第二时间段与第一时间间隔之和，第一时间间隔为目标周期内第二时刻的上一个时刻与第二时刻的时间间隔，第三指示信息用于指示包括第二指示信息的响应信息对应的报文的发送结果为失败。
请参见图6B，图6B为本申请实施例提供的一种响应报文的示意图。其中，如图6B所示，上述响应报文中的指示信息可以是该响应报文的包头中携带特殊的比特、相位和域值等，以包头中携带特殊的比特、相位和域值等作为区分该响应报文对应的报文的发送结构的规则。
在又一种实现中，第一节点是根据第二节点发送的广播帧来确定每一个时刻的第二时间段的。
在一些实施例中，第一节点可以在第二时刻接收到第二广播帧时，确定第二时刻的第二时间段为0，第二广播帧是由第二节点在监听到干扰节点通过共享信道成功发送报文时广播至第一节点的。
例如，在图3所示的网络架构中，第二节点为AP，其他节点均为STA，则第二节点可以在获取到干扰节点发送报文成功的响应时，生成第二广播帧，将该第二广播帧广播至该网络中的每一个STA。可以理解的，该网络中的每一个STA均可以执行本申请实施例提供的信道接入方法，故每一个STA均可以利用第二节点广播的第二广播帧确定每一个时刻的第二时间段，该方法能够有效利用第二节点获取的数据，同时，提高了该网络中每一个节点获取网络状态信息的速度，从而缩短了该网络中每一个节点决策接入信道的时间，提高了该网络的通信速度。
优选的，第二广播帧包括第一时间点，第一时间点为第二节点监听到干扰节点通过共享信道成功发送报文的时刻。
在一些实施例中，第一节点可以根据第二广播帧中的第二时间点，确定第二时间段。具体的，第一节点可以将第二时刻的上一个第二时间段清零的时刻与第二时刻的时间间隔确定为第二时间间隔；在第一时间间隔小于第二时间间隔时，第一节点确定第二时刻的第二时间段为0。可以理解的，该方法可以解决在第二广播帧丢包或延时的情况下，第一节点接收到比上一次清零第二时间段的时刻早的第二广播帧而再次清零第二时间段的问题。
需要说明的，第一节点还可以通过其他方式获取任一时刻的第二时间段，此处不作限定。
S102、第一节点将每一个时刻的网络状态信息输入目标神经网络，得到预测结果，预测结果包括第一预测值和第二预测值。
具体的，第一节点可以将上述目标周期内的每一个时刻的第一时间段和第二时间段输入目标神经网络模型中，得到第一预测值和第二预测值。其中，上述第一预测值用于指示上述第一节点在上述第一时刻通过上述共享信道向上述第二节点发送报文成功的概率，上述第二预测值用于指示上述第一节点在上述第一时刻通过上述共享信道向上述第二节点发送报文失败的概率。
在一些实施例中，上述目标神经网络可以是以样本周期中每一个时刻的网络状态信息为输入且以样本时刻发送报文的结果为标签训练得到的，样本周期为包括样本时刻在内的样本时刻之前预设时长的时间段。其中，样本周期中每一个时刻的网络状态信息包括每一个时刻的第一时间段和第二时间段。
需要说明的是，目标神经网络可以是深度神经网络(Deep Neural Networks，DNN)，也可以是门控循环单元(Gate Recurrent Unit，GRU)，还可以是长短期记忆网络(Long Short-Term Memory，LSTM)等神经网络。
请参见图7，图7为本申请实施例提供的一种目标神经网络的示意图。其中，x表示每一个时刻的网络状态，如图7所示，目标周期有T个时刻，第一节点将T个时刻的网络状态输入目标神经网络中，x t表示第一时刻的网络状态信息，则第一时刻的输出为y t。
第一节点可以采用如下所示的损失函数对目标神经网络模型进行训练：
损失函数：
Figure PCTCN2021122982-appb-000003
其中，r t表示第一时刻的回报值，γ表示折扣因子(discount factor)，e t＝(s t,a t,r t,s t+1)表示经验(experience)，E表示经验池，N E表示经验池中经验e t的数量，Q(s t+1,a′；θ -)表示目标神经网络的输出，θ -为目标神经网络的参数，Q(s t,a t；θ)表示目标神经网络的输出，θ为目标神经网络的参数。
可以理解的，训练的目标为最小化损失函数，训练的过程可以利用小批量梯度下降法(mini-batch gradient descent)的方法更新目标神经网络的参数。每完成C次训练，将上述参数赋值给目标神经网络，例如，C可以取100。其中，每一次训练过程中会生成一个优化参数h，该如图7所示，在输入为x t-T+1的训练中会生成h t-1，在下一次训练中将h t-1和x t-T+1作为输入进行训练，得到h t。
需要说明的是，本申请实施例还可以使用其他损失函数，此处不作限定。
S103、在第一预测值大于第二预测值时，第一节点通过共享信道向第二节点发送待发送报文。
具体的，第一节点在得到第一预测值和第二预测值后，可以比较第一预测值和第二预测值的大小，在第一预测值大于第二预测值时，第一节点通过共享信道向第二节点发送待发送报文；在第一预测值小于第二预测值时，第一节点不通过共享信道向第二节点发送待发送报文。
在一些实施例中，第一节点可以根据S202得到的发送和不发送的状态-动作值函数的函数值确定是否向第二节点发送待发送报文。具体的，若Q(s t,0)>Q(s t,1)，则第一节点向第二节点发送待发送报文，反之，第一节点不向第二节点发送待发送报文。
请参考图8，图8为本申请实施例提供的另一种信道接入方法的流程图。该方法可应用于上述图2或图3的信道接入网络中，该方法可以包括以下部分或全部步骤。
S201、第一节点获取目标周期内每一个时刻的网络状态信息，该目标周期为包括第一时刻在内的第一时刻之前的预设时长的时间段，网络状态信息包括第一节点监听到的共享信道的忙闲状态、第一时间段、第二时间段和干扰节点的总数。
具体的，第一节点考研获取目标周期内每一个时刻的网络状态信息。其中，目标周期为包括第一时刻在内的第一时刻之前的预设时长的时间段；网络状态信息包括第一节点监听到的共享信道的忙闲状态、第一时间段、第二时间段和干扰节点的总数，第二时刻的第一时间段用于指示第一节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔，第二时刻的第二时间段用于指示干扰节点在第二时刻之前通过共享信道最近一次成功发送报文的时间与第二时刻之间的时间间隔；干扰节点为第二节点的监听到的除第一节点外通过共享信道发送报文的节点；第二时刻为目标周期中的任意一个时刻。
具体的，步骤S201包括如下部分或全部步骤：
S2011、第一节点获取目标周期内每一时刻的共享信道的忙闲状态。
具体的，第一节点可以通过载波侦听机制来获取目标周期每一个时刻的共享信道的忙闲状态。其中，第一节点在侦听到有其他节点在使用共享信道传输报文时，可以得到该时刻的共享信道处于忙状态；第一节点在侦听到没有其他节点在使用共享信道传输报文时，可以得到该时刻的共享信道处于闲状态。进而，第一节点可以根据预设规定记录共享信道的忙闲状态。
其中，第一节点获取共享信道的忙闲状态的方法，此处不做限定。
S2012、第一节点获取目标周期内每一时刻的干扰节点的总数。
具体的，第二节点可以在每一个时刻获取干扰节点的总数，在获取到干扰节点的总数时， 生成包含上述干扰节点的总数的广播帧，并将该广播帧广播至第一节点，相应的，第一节点可以从该广播帧解析到每一个时刻的干扰节点数。
例如，在第二时刻时，第一节点接收第二节点广播的第一广播帧，该第一广播帧包括干扰节点的总数；第一节点将从第一广播帧中解析到的干扰节点的总数确定为在第二时刻的干扰节点的总数。
请参见图9A，图9A为本申请实施例提供的一种广播帧的信息单元的示意图。如图9A所示，该信息单元可以包括元素标识符(Element ID)、长度(Length)、元素标识符扩展(Element ID Extension)和干扰节点的总数(Number of interfering STAs)。
在一些实施例中，网络架构可以为图3所示，其中，第二节点是AP，第一节点和干扰节点为与AP相关的STA。具体的，第二节点可以在获取到干扰节点的总数时，通过信标(beacon)将该干扰节点的总数广播至同一网络架构下的每一个STA，相应的，上述每一个STA(包括第一节点)可以接收到该广播帧，从该广播帧中解析得到上述干扰节点的总数。
在另一些实施例中，第二节点为STA，则第二节点可以通过探测请求(probe request)帧将干扰节点的总数广播至第一节点。
请参见图9B，图9B为本申请实施例提供的另一种广播帧的信息单元的示意图。如图9B所示，干扰节点的总数可以存在于HT控制字段(HT control field)中的字段(A-control subfield)中。具体的，图中有控制列表(Control List)和补齐(Padding)，一个Control List包括多个控制(Control)，其中一个控制可以包括控制标识符(Control ID)和控制信息(Control information)，Number of interfering STAs可以存在与上述Control information中。
其中，第一节点获取每一时刻的第一时间段和第二时间段的详细内容可以参见步骤S101中的相关内容，此处不再赘述。
S202、第一节点将每一个时刻的网络状态信息输入目标神经网络，得到预测结果，预测结果包括第一预测值和第二预测值。
具体的，第一节点可以将上述目标周期内的每一个时刻的第一节点监听到的共享信道的忙闲状态、第一时间段、第二时间段和干扰节点的总数输入目标神经网络模型中，得到第一预测值和第二预测值。其中，第一预测值用于指示第一节点在第一时刻通过共享信道向第二节点发送报文成功的概率，第二预测值用于指示第一节点在第一时刻通过上述共享信道向第二节点发送报文失败的概率。
在一些实施例中，上述目标神经网络可以是以样本周期中每一个时刻的网络状态信息为输入且以样本时刻发送报文的结果为标签训练得到的，样本周期为包括样本时刻在内的样本时刻之前预设时长的时间段。其中，样本周期中每一个时刻的网络状态信息包括每一个时刻的第一节点监听到的共享信道的忙闲状态、第一时间段、第二时间段和干扰节点的总数。
请参见图7，图7是本申请实施例提供的一种目标神经网络的示意图。其中，x表示每一个时刻的网络状态，如图7所示，目标周期有T个时刻，第一节点将T个时刻的网络状态信息输入目标神经网络中，可以得到第一时刻的输出y t。
其中，第一时刻的网络状态信息为：
Figure PCTCN2021122982-appb-000004
其中，o t-1表示上一时刻载波侦听结果，
Figure PCTCN2021122982-appb-000005
表示第一时刻的第一时间段，
Figure PCTCN2021122982-appb-000006
表示第一时刻的第二时间段，N表示第一时刻的干扰节点数。
第一节点将目标周期的每一个时刻的网络状态信息输入目标神经网络后，可以得到预测结果，该输出为：y t＝[Q(s t,0),Q(s t,1)]，其中，s t＝(x t-T+1,x t-T+2,…,x t)。Q(s t,0)和Q(s t,1)分别表示发送和不发送的状态-动作值函数(state-action value function)的函数值，其中，Q(s t,0)为第一时刻的第一预测值，Q(s t,1)为第一时刻的第二预测值。
S203、在第一预测值大于第二预测值时，第一节点通过共享信道向第二节点发送待发送报文。
具体的，第一节点在得到第一预测值和第二预测值后，可以比较第一预测值和第二预测值的大小，在第一预测值大于第二预测值时，第一节点通过共享信道向第二节点发送待发送报文；在第一预测值小于第二预测值时，第一节点不通过共享信道向第二节点发送待发送报文。
在一些实施例中，第一节点可以根据S202得到的发送和不发送的状态-动作值函数的函数值确定是否向第二节点发送待发送报文。具体的，若Q(s t,0)>Q(s t,1)，则第一节点向第二节点发送待发送报文，反之，第一节点不向第二节点发送待发送报文。
在一些实施例中，在步骤S103和步骤S203之后，该信道接入方法还包括以下部分或全部步骤：
S301、第一节点通过共享信道向第二节点发送待发送报文后，获取待发送报文的发送结果。
具体的，第一节点可以在通过共享信道向第二节点发送待发送报文后，等待响应，进而，根据该响应，确定待发送报文的发送结果。例如，在接收到带有指示成功的指示信息的响应时，确定待发送报文的发送结果为成功：在接受到带有指示失败的指示信息的响应时，确定待发送报文的发送结果为失败。
S302、第一节点以目标周期的网络状态信息为输入且以待发送报文的发送结果为标签更新目标神经网络。
具体的，第一节点可以根据损失更新目标神经网络，损失包括将预测结果与待发送报文的发送结果之间的误差。第一节点可以根据待发送报文的发送结果和目标神经网络的预测结果，得到损失，再根据该损失更新目标神经网络，
其中，损失还可以包括第一时刻的回报值，回报值是根据待发送报文的发送结果和第一节点在第一时刻监听到的共享信道的忙闲状态决定的。具体的，回报值的计算过程可以参见以下计算回报值的实施例。
第一节点可以采用如下所示的损失函数对目标神经网络模型进行训练：
损失函数：
Figure PCTCN2021122982-appb-000007
其中，r t表示第一时刻的回报值，γ表示折扣因子(discount factor)，e t＝(s t,a t,r t,s t+1)表示经验(experience)，E表示经验池，N E表示经验池中经验e t的数量，Q(s t+1,a′；θ -)表示目标神经网络的输出，θ -为目标神经网络的参数，Q(s t,a t；θ)表示目标神经网络的输出，θ为目标神经网络的参数。
需要说明的是，本申请实施例还可以使用其他损失函数，此处不作限定。
以下介绍本申请实施例提供的一种计算回报值的方法。
在一些实施例中，决策结果是第一节点在第一时刻向第二节点发送报文。进而，第一节点可以根据该报文的发送结果，确定第一时刻的回报值。
具体的，第一节点可以在待发送报文的发送结果为成功时，确定上述回报值为第一时间段与第二时间段的比值；在待发送报文的发送结果为失败时，确定上述回报值为第一时刻的干扰节点的总数的负值。
在另一些实施例中，决策结果是第一节点在第一时刻不向第二节点发送报文。则第一节点可以监听第一时刻的共享信道的忙闲状态，确定第一时刻的回报值。
具体的，第一节点可以在第一时刻监听到共享信道有指示报文发送成功的响应信息时， 确定上述回报值为第一时刻的第二时间段与第一时刻的第一时间段的比值；第一节点在第一时刻监听到共享信道有指示报文发送失败的响应信息时，确定上述回报值为M，其中，M为正数，例如，N可以为1；第一节点在第一时刻监听到共享信道不存在响应信息时，确定上述回报值为0。
为了便于更好地实施本申请实施例的上述方案，相应地，下面还提供用于配合实施上述方案的相关装置。
参见图10，图10为本申请实施例提供的一种信道接入装置400的结构示意图。其中：
获取单元410，用于获取目标周期内每一个时刻的网络状态信息，所述目标周期为包括第一时刻在内的所述第一时刻之前的预设时长的时间段；所述网络状态信息包括第一时间段和第二时间段，第二时刻的第一时间段用于指示所述第一节点在所述第二时刻之前通过共享信道最近一次成功发送报文的时间与所述第二时刻之间的时间间隔，所述第二时刻的第二时间段用于指示干扰节点在所述第二时刻之前通过所述共享信道最近一次成功发送报文的时间与所述第二时刻之间的时间间隔；所述干扰节点为所述第二节点的监听到的除所述第一节点外通过所述共享信道发送报文的节点；所述第二时刻为所述目标周期中的任意一个时刻；
预测单元420，用于将所述每一个时刻的网络状态信息输入目标神经网络，得到预测结果，所述预测结果包括第一预测值和第二预测值，所述第一预测值用于指示所述第一节点在所述第一时刻通过所述共享信道向所述第二节点发送报文成功的概率，所述第二预测值用于指示所述第一节点在所述第一时刻通过所述共享信道向所述第二节点发送报文失败的概率；
决策单元430，用于在所述第一预测值大于所述第二预测值时，所述第一节点通过所述共享信道向所述第二节点发送待发送报文。
在一个实施例中，所述获取单元410，还用于获取所述第一节点监听到的所述共享信道的忙闲状态和所述干扰节点的总数。
在一个实施例中，所述获取单元410，具体用于获取所述第二时刻的第二时间段，其中：
所述第一节点根据第一指示信息在所述第二时刻监听到响应信息时，确定所述第二时刻的第二时间段为0，所述第二时刻为所述第一节点监听到第一报文的第一时长后的时刻，所述第一报文的包头包括所述第一指示信息，所述第一指示信息用于指示在接收到所述第一报文的节点在所述第一时长后发送所述响应信息。
在一个实施例中，所述获取单元410，具体用于获取所述第二时刻的第二时间段，其中：
所述第一节点在所述第二时刻监听到包括第二指示信息的响应信息时，确定所述第二时刻的第二时间段为0，所述第二指示信息用于指示所述包括第二指示信息的响应信息对应的报文的发送结果为成功；
所述第一节点在所述第二时刻监听到包括第三指示信息的响应信息时，确定所述第二时刻的第二时间段为所述目标周期内所述第二时刻的上一个时刻的第二时间段与第一时间间隔之和，所述第一时间间隔为所述上一个时刻与所述第二时刻的时间间隔，所述第三指示信息用于指示所述包括第二指示信息的响应信息对应的报文的发送结果为失败。
在一个实施例中，所述获取单元410，具体用于获取所述第二时刻的干扰节点的总数，其中：
所述第二时刻时，所述第一节点接收所述第二节点广播的第一广播帧；
所述第一节点将从所述第一广播帧中解析到的干扰节点的总数确定为在所述第二时刻的干扰节点的总数。
在一个实施例中，所述获取单元410，具体用于获取所述第二时刻的第二时间段，其中：
所述第一节点在所述第二时刻接收到第二广播帧时，确定所述第二时刻的第二时间段为0，所述第二广播帧是由所述第二节点在监听到所述干扰节点通过所述共享信道成功发送报文时广播至所述第一节点的。
在一个实施例中，所述第二广播帧包括第一时间点，所述第一时间点为所述第二节点监听到所述干扰节点通过所述共享信道成功发送报文的时刻。
在一个实施例中，所述获取单元410具体用于执行：
所述第一节点将所述第一时间点与所述第二时刻的时间间隔确定为第一时间间隔；
将所述第二时刻的上一个第二时间段清零的时刻与所述第二时刻的时间间隔确定为第二时间间隔；
在所述第一时间间隔小于所述第二时间间隔时，确定所述第二时刻的第二时间段为0。
在一个实施例中，所述信道接入装置还包括训练单元440，所述训练单元440用于执行：
所述第一节点获取所述待发送报文的发送结果；
所述第一节点以所述目标周期每一个时刻的网络状态信息为输入且以所述待发送报文的发送结果为标签更新所述目标神经网络。
在一个实施例中，所述信道接入装置还包括训练单元440，所述训练单元440具体用于执行：根据损失更新所述目标神经网络，所述损失包括将所述预测结果与所述待发送报文的发送结果之间的误差。
在一个实施例中，所述损失还包括所述第一时刻的回报值，所述回报值是根据所述预测结果和所述第一节点在所述第一时刻监听到的共享信道的忙闲状态决定的。
在一个实施例中，所述信道接入装置还包括训练单元440，所述训练单元440还用于执行：
在所述待发送报文的发送结果为成功时，确定所述回报值为所述第一时间段与所述第二时间段的比值；
在所述待发送报文的发送结果为失败时，确定所述回报值为所述第一时刻的干扰节点的总数的负值。
在一个实施例中，所述信道接入装置还包括训练单元440，所述训练单元440还用于执行：
在所述第一预测值小于所述第二预测值时，所述第一节点在所述第一时刻监听所述共享信道；
所述第一节点在所述第一时刻监听到所述共享信道有指示报文发送成功的响应信息时，确定所述回报值为所述第一时刻的第二时间段与所述第一时刻的第一时间段的比值；
所述第一节点在所述第一时刻监听到所述共享信道有指示报文发送失败的响应信息时，确定所述回报值为M，M为正数；
所述第一节点在所述第一时刻监听到所述共享信道不存在响应信息时，确定所述回报值为0。
可以理解的，装置400可以是接入点AP侧的装置，或，接入点AP内的芯片，以支持接入点AP实现所述方法中相应的功能；装置400也可以是站点STA侧的装置，或者是站点STA内的芯片，以支持STA实现所述方法中相应的功能。
参见图11，图11为本申请实施例提供的另一种信道接入装置500。该信道接入装置500 至少包括：处理器510、存储器520和收发器530，该处理器510、存储器520和收发器530通过总线540相互连接。
存储器520包括但不限于是随机存取存储器(Random Access Memory，RAM)、只读存储器(Read-Only Memory，ROM)或可擦除可编程只读存储器(Erasable Programmable Read-Only Memory，EPROM或者快闪存储器)，该存储器520用于存储相关指令及数据。
该收发器530可以包括一个接收器和一个发送器，例如，无线射频模块，以下描述的处理器510接收或者发送某个消息，具体可以理解为该处理器510通过该收发器530来接收或者发送。
处理器510可以是一个或多个中央处理器(Central Processing Unit，CPU)，在处理器510是一个CPU的情况下，该CPU可以是单核CPU，也可以是多核CPU。
该信道接入装置500中的处理器510用于读取该存储器520中存储的程序代码，执行以下操作：
第一节点获取目标周期内每一个时刻的网络状态信息，所述目标周期为包括第一时刻在内的所述第一时刻之前的预设时长的时间段；所述网络状态信息包括第一时间段和第二时间段，第二时刻的第一时间段用于指示所述第一节点在所述第二时刻之前通过共享信道最近一次成功发送报文的时间与所述第二时刻之间的时间间隔，所述第二时刻的第二时间段用于指示干扰节点在所述第二时刻之前通过所述共享信道最近一次成功发送报文的时间与所述第二时刻之间的时间间隔；所述干扰节点为所述第二节点的监听到的除所述第一节点外通过所述共享信道发送报文的节点；所述第二时刻为所述目标周期中的任意一个时刻；
所述第一节点将所述每一个时刻的网络状态信息输入目标神经网络，得到预测结果，所述预测结果包括第一预测值和第二预测值，所述第一预测值用于指示所述第一节点在所述第一时刻通过所述共享信道向所述第二节点发送报文成功的概率，所述第二预测值用于指示所述第一节点在所述第一时刻通过所述共享信道向所述第二节点发送报文失败的概率；
在所述第一预测值大于所述第二预测值时，所述第一节点通过所述共享信道向所述第二节点发送待发送报文。
在一种可能的实现方式中，所述网络状态信息还包括所述第一节点监听到的所述共享信道的忙闲状态和所述干扰节点的总数。
在一种可能的实现方式中，所述第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取所述第二时刻的第二时间段；
所述获取所述第二时刻的第二时间段，包括：
所述第一节点根据第一指示信息在所述第二时刻监听到响应信息时，确定所述第二时刻的第二时间段为0，所述第二时刻为所述第一节点监听到第一报文的第一时长后的时刻，所述第一报文的包头包括所述第一指示信息，所述第一指示信息用于指示在接收到所述第一报文的节点在所述第一时长后发送所述响应信息。
在一种可能的实现方式中，所述第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取所述第二时刻的第二时间段；
所述获取所述第二时刻的第二时间段，包括：
所述第一节点在所述第二时刻监听到包括第二指示信息的响应信息时，确定所述第二时刻的第二时间段为0，所述第二指示信息用于指示所述包括第二指示信息的响应信息对应的报文的发送结果为成功；
所述第一节点在所述第二时刻监听到包括第三指示信息的响应信息时，确定所述第二时 刻的第二时间段为所述目标周期内所述第二时刻的上一个时刻的第二时间段与第一时间间隔之和，所述第一时间间隔为所述上一个时刻与所述第二时刻的时间间隔，所述第三指示信息用于指示所述包括第二指示信息的响应信息对应的报文的发送结果为失败。
在一种可能的实现方式中，所述第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取所述第二时刻的干扰节点的总数；
获取所述第二时刻的干扰节点的总数，包括：
在所述第二时刻时，所述第一节点接收所述第二节点广播的第一广播帧；
所述第一节点将从所述第一广播帧中解析到的干扰节点的总数确定为在所述第二时刻的干扰节点的总数。
在一种可能的实现方式中，所述第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取所述第二时刻的第二时间段；
所述获取所述第二时刻的第二时间段，包括：
所述第一节点在所述第二时刻接收到第二广播帧时，确定所述第二时刻的第二时间段为0，所述第二广播帧是由所述第二节点在监听到所述干扰节点通过所述共享信道成功发送报文时广播至所述第一节点的。
在一种可能的实现方式中，所述第二广播帧包括第一时间点，所述第一时间点为所述第二节点监听到所述干扰节点通过所述共享信道成功发送报文的时刻。
在一种可能的实现方式中，所述第一节点在所述第二时刻接收到第二广播帧时，确定所述第二时刻的第二时间段为0，包括：
所述第一节点将所述第一时间点与所述第二时刻的时间间隔确定为第一时间间隔；
将所述第二时刻的上一个第二时间段清零的时刻与所述第二时刻的时间间隔确定为第二时间间隔；
在所述第一时间间隔小于所述第二时间间隔时，确定所述第二时刻的第二时间段为0。
在一种可能的实现方式中，所述第一节点通过所述共享信道向所述第二节点发送待发送报文后，所述方法还包括：
所述第一节点获取所述待发送报文的发送结果；
所述第一节点以所述目标周期每一个时刻的网络状态信息为输入且以所述待发送报文的发送结果为标签更新所述目标神经网络。
在一种可能的实现方式中，所述第一节点以所述目标周期的网络状态信息为输入且以所述待发送报文的发送结果为标签更新所述目标神经网络，包括：
根据损失更新所述目标神经网络，所述损失包括将所述预测结果与所述待发送报文的发送结果之间的误差。
在一种可能的实现方式中，所述损失还包括所述第一时刻的回报值，所述回报值是根据所述预测结果和所述第一节点在所述第一时刻监听到的共享信道的忙闲状态决定的。
在一种可能的实现方式中，在所述待发送报文的发送结果为成功时，确定所述回报值为所述第一时间段与所述第二时间段的比值；
在所述待发送报文的发送结果为失败时，确定所述回报值为所述第一时刻的干扰节点的总数的负值。
在一种可能的实现方式中，在所述第一预测值小于所述第二预测值时，所述第一节点在所述第一时刻监听所述共享信道；
所述第一节点在所述第一时刻监听到所述共享信道有指示报文发送成功的响应信息时， 确定所述回报值为所述第一时刻的第二时间段与所述第一时刻的第一时间段的比值；
所述第一节点在所述第一时刻监听到所述共享信道有指示报文发送失败的响应信息时，确定所述回报值为M，M为正数；
所述第一节点在所述第一时刻监听到所述共享信道不存在响应信息时，确定所述回报值为0。
可以理解的，装置500可以是接入点AP侧的装置，或者是接入点AP内的芯片；装置500也可以是站点STA侧的装置，或者是站点STA内的芯片。
本申请实施例还提供了一种计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在计算机或处理器上运行时，使得计算机或处理器执行上述任一个物理层协议数据单元的信道接入方法中的一个或多个步骤。上述装置的各组成模块如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在所述计算机可读取存储介质中。
上述计算机可读存储介质可以是前述任一实施例所述的信道接入装置的内部存储单元，例如信道接入装置的硬盘或内存。上述计算机可读存储介质也可以是上述信道接入装置的外部存储设备，例如上述信道接入装置上配备的插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)等。进一步地，上述计算机可读存储介质还可以既包括上述信道接入装置的内部存储单元也包括外部存储设备。上述计算机可读存储介质用于存储上述计算机程序以及上述信道接入装置所需的其他程序和数据。上述计算机可读存储介质还可以用于暂时地存储已经输出或者将要输出的数据。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，可通过计算机程序来指令相关的硬件来完成，该的程序可存储于计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。而前述的存储介质包括：ROM、RAM、磁碟或者光盘等各种可存储程序代码的介质。
本申请实施例方法中的步骤可以根据实际需要进行顺序调整、合并和删减。
本申请实施例装置中的模块可以根据实际需要进行合并、划分和删减。
以上所述，以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (16)

  1. 一种信道接入方法，其特征在于，所述方法包括：
    第一节点获取目标周期内每一个时刻的网络状态信息，所述目标周期为包括第一时刻在内的所述第一时刻之前的预设时长的时间段；所述网络状态信息包括第一时间段和第二时间段，第二时刻的第一时间段用于指示所述第一节点在所述第二时刻之前通过共享信道最近一次成功发送报文的时间与所述第二时刻之间的时间间隔，所述第二时刻的第二时间段用于指示干扰节点在所述第二时刻之前通过所述共享信道最近一次成功发送报文的时间与所述第二时刻之间的时间间隔；所述干扰节点为所述第二节点的监听到的除所述第一节点外通过所述共享信道发送报文的节点；所述第二时刻为所述目标周期中的任意一个时刻；
    所述第一节点将所述每一个时刻的网络状态信息输入目标神经网络，得到预测结果，所述预测结果包括第一预测值和第二预测值，所述第一预测值用于指示所述第一节点在所述第一时刻通过所述共享信道向所述第二节点发送报文成功的概率，所述第二预测值用于指示所述第一节点在所述第一时刻通过所述共享信道向所述第二节点发送报文失败的概率；
    在所述第一预测值大于所述第二预测值时，所述第一节点通过所述共享信道向所述第二节点发送待发送报文。
  2. 根据权利要求1所述的方法，其特征在于，所述网络状态信息还包括所述第一节点监听到的所述共享信道的忙闲状态和所述干扰节点的总数。
  3. 根据权利要求1或2所述的方法，其特征在于，所述第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取所述第二时刻的第二时间段；
    所述获取所述第二时刻的第二时间段，包括：
    所述第一节点根据第一指示信息在所述第二时刻监听到响应信息时，确定所述第二时刻的第二时间段为0，所述第二时刻为所述第一节点监听到第一报文的第一时长后的时刻，所述第一报文的包头包括所述第一指示信息，所述第一指示信息用于指示在接收到所述第一报文的节点在所述第一时长后发送所述响应信息。
  4. 根据权利要求1或2所述的方法，其特征在于，所述第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取所述第二时刻的第二时间段；
    所述获取所述第二时刻的第二时间段，包括：
    所述第一节点在所述第二时刻监听到包括第二指示信息的响应信息时，确定所述第二时刻的第二时间段为0，所述第二指示信息用于指示所述包括第二指示信息的响应信息对应的报文的发送结果为成功；
    所述第一节点在所述第二时刻监听到包括第三指示信息的响应信息时，确定所述第二时刻的第二时间段为所述目标周期内所述第二时刻的上一个时刻的第二时间段与第一时间间隔之和，所述第一时间间隔为所述上一个时刻与所述第二时刻的时间间隔，所述第三指示信息用于指示所述包括第二指示信息的响应信息对应的报文的发送结果为失败。
  5. 根据权利要求1或2所述的方法，其特征在于，所述第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取所述第二时刻的干扰节点的总数；
    获取所述第二时刻的干扰节点的总数，包括：
    在所述第二时刻时，所述第一节点接收所述第二节点广播的第一广播帧；
    所述第一节点将从所述第一广播帧中解析到的干扰节点的总数确定为在所述第二时刻的干扰节点的总数。
  6. 根据权利要求1或2所述的方法，其特征在于，所述第一节点获取目标周期内每一个时刻的网络状态信息，包括：获取所述第二时刻的第二时间段；
    所述获取所述第二时刻的第二时间段，包括：
    所述第一节点在所述第二时刻接收到第二广播帧时，确定所述第二时刻的第二时间段为0，所述第二广播帧是由所述第二节点在监听到所述干扰节点通过所述共享信道成功发送报文时广播至所述第一节点的。
  7. 根据权利要求6所述的方法，其特征在于，所述第二广播帧包括第一时间点，所述第一时间点为所述第二节点监听到所述干扰节点通过所述共享信道成功发送报文的时刻。
  8. 根据权利要求7所述的方法，其特征在于，所述第一节点在所述第二时刻接收到第二广播帧时，确定所述第二时刻的第二时间段为0，包括：
    所述第一节点将所述第一时间点与所述第二时刻的时间间隔确定为第一时间间隔；
    将所述第二时刻的上一个第二时间段清零的时刻与所述第二时刻的时间间隔确定为第二时间间隔；
    在所述第一时间间隔小于所述第二时间间隔时，确定所述第二时刻的第二时间段为0。
  9. 根据权利要求1或2所述的方法，其特征在于，所述第一节点通过所述共享信道向所述第二节点发送待发送报文后，所述方法还包括：
    所述第一节点获取所述待发送报文的发送结果；
    所述第一节点以所述目标周期每一个时刻的网络状态信息为输入且以所述待发送报文的发送结果为标签更新所述目标神经网络。
  10. 根据权利要求9所述的方法，其特征在于，所述第一节点以所述目标周期的网络状态信息为输入且以所述待发送报文的发送结果为标签更新所述目标神经网络，包括：
    根据损失更新所述目标神经网络，所述损失包括将所述预测结果与所述待发送报文的发送结果之间的误差。
  11. 根据权利要求10所述的方法，其特征在于，所述损失还包括所述第一时刻的回报值，所述回报值是根据所述预测结果和所述第一节点在所述第一时刻监听到的共享信道的忙闲状态决定的。
  12. 根据权利要求11所述的方法，其特征在于，所述方法还包括：
    在所述待发送报文的发送结果为成功时，确定所述回报值为所述第一时间段与所述第二时间段的比值；
    在所述待发送报文的发送结果为失败时，确定所述回报值为所述第一时刻的干扰节点的总数的负值。
  13. 根据权利要求1或2所述的方法，其特征在于，所述方法还包括：
    在所述第一预测值小于所述第二预测值时，所述第一节点在所述第一时刻监听所述共享信道；
    所述第一节点在所述第一时刻监听到所述共享信道有指示报文发送成功的响应信息时，确定所述回报值为所述第一时刻的第二时间段与所述第一时刻的第一时间段的比值；
    所述第一节点在所述第一时刻监听到所述共享信道有指示报文发送失败的响应信息时，确定所述回报值为M，M为正数；
    所述第一节点在所述第一时刻监听到所述共享信道不存在响应信息时，确定所述回报值为0。
  14. 一种信道接入装置，其特征在于，包括：
    获取单元，用于获取目标周期内每一个时刻的网络状态信息，所述目标周期为包括第一时刻在内的所述第一时刻之前的预设时长的时间段；所述网络状态信息包括第一时间段和第二时间段，第二时刻的第一时间段用于指示所述第一节点在所述第二时刻之前通过共享信道最近一次成功发送报文的时间与所述第二时刻之间的时间间隔，所述第二时刻的第二时间段用于指示干扰节点在所述第二时刻之前通过所述共享信道最近一次成功发送报文的时间与所述第二时刻之间的时间间隔；所述干扰节点为所述第二节点的监听到的除所述第一节点外通过所述共享信道发送报文的节点；所述第二时刻为所述目标周期中的任意一个时刻；
    预测单元，用于将所述每一个时刻的网络状态信息输入目标神经网络，得到预测结果，所述预测结果包括第一预测值和第二预测值，所述第一预测值用于指示所述第一节点在所述第一时刻通过所述共享信道向所述第二节点发送报文成功的概率，所述第二预测值用于指示所述第一节点在所述第一时刻通过所述共享信道向所述第二节点发送报文失败的概率；
    决策单元，用于在所述第一预测值大于所述第二预测值时，所述第一节点通过所述共享信道向所述第二节点发送待发送报文。
  15. 一种信道接入装置，其特征在于，包括处理器、存储器、输入接口和输出接口，所述输入接口用于接收来自所述信道接入装置之外的其它信道接入装置的信息，所述输出接口用于向所述信道接入装置之外的其它信道接入装置输出信息，所述处理器调用所述存储器中存储的计算机程序实现如权利要求1-13任一项所述的方法。
  16. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中存储有计算机程序或计算机指令，当所述计算机程序或计算机指令被运行时，实现如权利要求1-13任一项所述的方法。
PCT/CN2021/122982 2020-12-29 2021-10-11 一种信道接入方法和装置 WO2022142573A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP21913318.8A EP4258793A1 (en) 2020-12-29 2021-10-11 Channel access method and apparatus
US18/342,468 US20230345536A1 (en) 2020-12-29 2023-06-27 Channel access method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011603244.0A CN114698138A (zh) 2020-12-29 2020-12-29 一种信道接入方法和装置
CN202011603244.0 2020-12-29

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/342,468 Continuation US20230345536A1 (en) 2020-12-29 2023-06-27 Channel access method and apparatus

Publications (1)

Publication Number Publication Date
WO2022142573A1 true WO2022142573A1 (zh) 2022-07-07

Family

ID=82131472

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/122982 WO2022142573A1 (zh) 2020-12-29 2021-10-11 一种信道接入方法和装置

Country Status (4)

Country Link
US (1) US20230345536A1 (zh)
EP (1) EP4258793A1 (zh)
CN (1) CN114698138A (zh)
WO (1) WO2022142573A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024027511A1 (zh) * 2022-08-05 2024-02-08 华为技术有限公司 信道接入方法及相关装置

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117693035A (zh) * 2022-08-31 2024-03-12 华为技术有限公司 一种信道聚合方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106686699A (zh) * 2017-01-09 2017-05-17 电子科技大学 一种基于可用带宽的多ap选择方法
CN106888501A (zh) * 2015-12-16 2017-06-23 上海贝尔股份有限公司 预测网络拥塞程度并将链路接入信道的方法及设备
US20170273101A1 (en) * 2015-11-02 2017-09-21 Telefonaktiebolaget Lm Ericsson (Publ) Advanced laa scheduling for multi-point subset transmission in shared cells
CN109245840A (zh) * 2018-10-15 2019-01-18 哈尔滨工业大学 认知无线电系统中基于卷积神经网络的频谱预测方法
CN111919423A (zh) * 2018-04-06 2020-11-10 华为技术有限公司 网络通信中的拥塞控制

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170273101A1 (en) * 2015-11-02 2017-09-21 Telefonaktiebolaget Lm Ericsson (Publ) Advanced laa scheduling for multi-point subset transmission in shared cells
CN106888501A (zh) * 2015-12-16 2017-06-23 上海贝尔股份有限公司 预测网络拥塞程度并将链路接入信道的方法及设备
CN106686699A (zh) * 2017-01-09 2017-05-17 电子科技大学 一种基于可用带宽的多ap选择方法
CN111919423A (zh) * 2018-04-06 2020-11-10 华为技术有限公司 网络通信中的拥塞控制
CN109245840A (zh) * 2018-10-15 2019-01-18 哈尔滨工业大学 认知无线电系统中基于卷积神经网络的频谱预测方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024027511A1 (zh) * 2022-08-05 2024-02-08 华为技术有限公司 信道接入方法及相关装置

Also Published As

Publication number Publication date
CN114698138A (zh) 2022-07-01
EP4258793A1 (en) 2023-10-11
US20230345536A1 (en) 2023-10-26

Similar Documents

Publication Publication Date Title
US10721786B2 (en) Radio resource allocation in Wi-Fi aware neighborhood area network data links
US9419752B2 (en) Transmission opportunity operation of uplink multi-user multiple-input-multiple-output communication in wireless networks
CN107113782B (zh) 用于数字通信中避免干扰的系统和方法
US11387965B2 (en) Method for replying with acknowledgement frame, apparatus, and data transmission system
WO2022142573A1 (zh) 一种信道接入方法和装置
WO2021022408A1 (zh) 随机接入消息发送方法、装置及存储介质
KR20050115253A (ko) 무선 네트워크에서의 서비스 품질 차별화 방법 및 무선장치
EP3621360A1 (en) System information transmission method and related device
US8953564B2 (en) Systems and methods to achieve fairness in wireless LANs for cellular offloading
JP2009117891A (ja) 無線通信装置
CN110392994B (zh) Harq反馈方法、装置及可读存储介质
WO2020063266A1 (zh) 一种harq进程id的确定方法、装置、终端及介质
US20160337974A1 (en) Power save trigger
US20160135183A1 (en) Multimode user equipment accessing wireless sensor network
WO2022077227A1 (zh) 直连通信方法、装置及存储介质
KR102607771B1 (ko) 다중 사용자 직교 주파수 분할 다중 액세스를 옵트인 및 옵트아웃하기 위한 스위칭 방식
WO2022262734A1 (zh) 一种信道接入方法及相关装置
KR20230116722A (ko) Emlsr 동작을 지원하는 무선랜에서 저지연 통신을 위한 방법 및 장치
US20160337899A1 (en) Power save trigger
CN109906646A (zh) 信息传输方法、基站和终端设备
WO2020010991A1 (zh) 一种信道接入的方法及装置
KR102666491B1 (ko) 다중 사용자 직교 주파수 분할 다중 액세스를 옵트인 및 옵트아웃하기 위한 스위칭 방식
He et al. The analysis of IEEE 802.11 PCF protocol based on LEO satellite Wi-Fi
WO2024032228A1 (zh) 一种强化学习的训练方法及相关装置
JP2008227737A (ja) 無線装置およびそれを備えた無線通信ネットワーク

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21913318

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2021913318

Country of ref document: EP

Effective date: 20230731