WO2024109596A1 - 信息传输方法、装置、相关设备及存储介质 - Google Patents

信息传输方法、装置、相关设备及存储介质 Download PDF

Info

Publication number
WO2024109596A1
WO2024109596A1 PCT/CN2023/131633 CN2023131633W WO2024109596A1 WO 2024109596 A1 WO2024109596 A1 WO 2024109596A1 CN 2023131633 W CN2023131633 W CN 2023131633W WO 2024109596 A1 WO2024109596 A1 WO 2024109596A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
channel
downlink channel
threshold
terminal
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2023/131633
Other languages
English (en)
French (fr)
Inventor
石子烨
韩双锋
李刚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Mobile Communications Group Co Ltd
Research Institute of China Mobile Communication Co Ltd
Original Assignee
China Mobile Communications Group Co Ltd
Research Institute of China Mobile Communication Co Ltd
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 China Mobile Communications Group Co Ltd, Research Institute of China Mobile Communication Co Ltd filed Critical China Mobile Communications Group Co Ltd
Publication of WO2024109596A1 publication Critical patent/WO2024109596A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/0231—Traffic management, e.g. flow control or congestion control based on communication conditions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/02—Arrangements for detecting or preventing errors in the information received by diversity reception
    • H04L1/06—Arrangements for detecting or preventing errors in the information received by diversity reception using space diversity
    • 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
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information

Definitions

  • the present application relates to the field of wireless communications, and in particular to an information transmission method, apparatus, related equipment and storage medium.
  • CSI channel state information
  • FDD frequency division duplex
  • the terminal can use a data-driven deep learning method to perform channel compression and feed back the channel compression result to the base station so that the base station can obtain accurate downlink CSI, that is, the terminal can use a deep neural network (i.e., the encoding network in Figure 1) to extract low-dimensional features from a large amount of channel data for compression and feed back the channel compression result to the base station, and the base station can use the corresponding deep neural network (i.e., the decoding network in Figure 1) to restore the channel compression result to obtain the state of the original channel.
  • a deep neural network i.e., the encoding network in Figure 1
  • the base station can use the corresponding deep neural network (i.e., the decoding network in Figure 1) to restore the channel compression result to obtain the state of the original channel.
  • the terminal when adopting the downlink channel compression and feedback scheme based on deep learning, the terminal periodically feeds back the channel compression results to the base station according to a specific period, which may degrade the system performance.
  • the embodiments of the present application provide an information transmission method, apparatus, related equipment and storage medium.
  • the present application provides an information transmission method, which is applied to a terminal and includes:
  • the second information represents the stability of the downlink channel
  • the third information indicates a channel compression mode of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network side; the third information is determined by at least using the second information;
  • At least the third information is used to perform channel compression on the downlink channel to obtain a channel compression result, and the third information is used to send fourth information to the network side, where the fourth information at least includes the channel compression result.
  • the first information includes fifth information and sixth information
  • the fifth information includes the number of eigenvectors K in the channel matrix of the downlink channel, K is an integer greater than 0
  • the sixth information includes at least one threshold, and the at least one threshold corresponds to at least two stability levels of the downlink channel
  • the at least using the first information to determine the second information includes:
  • the first parameter represents a similarity between K eigenvectors in two channel matrices corresponding to the downlink channel at two adjacent moments;
  • the first parameter is compared with the at least one threshold to obtain a first comparison result, and the second information includes the first comparison result.
  • the sixth information includes a first threshold and a second threshold, the first threshold is greater than 0 and less than the second threshold; the second threshold is less than 1; the interval formed by the second threshold and 1 corresponds to the first state of the downlink channel, the interval formed by the first threshold and the second threshold corresponds to the second state of the downlink channel, and the interval formed by the first threshold and 0 corresponds to the third state of the downlink channel; the stability of the downlink channel in the first state is higher than the stability in the second state, and the stability of the downlink channel in the second state is higher than the stability in the third state.
  • the first comparison result when the first parameter is greater than or equal to 0 and less than the first threshold, the first comparison result includes a first identifier, and the first identifier represents the third state;
  • the first comparison result includes a second identifier, and the second identifier represents the first state.
  • the sixth information further includes a third threshold value, and the third threshold value is greater than the first threshold value and less than the second threshold value; and the first parameter is compared with the at least one threshold value to obtain a first comparison result, including:
  • the first parameter is greater than or equal to the first threshold and less than the second threshold, for two channel matrices corresponding to the downlink channel at two adjacent moments, determine a similarity between each eigenvector of K eigenvectors of one channel matrix and a corresponding eigenvector in the other channel matrix, and compare the determined similarity with the third threshold to obtain K second comparison results;
  • the first comparison result is obtained by using the K second comparison results.
  • the first comparison result represents the K second comparison results in the form of a bitmap sequence; wherein, in the case where the bit position in the bitmap sequence is the first value In this case, the corresponding second comparison result represents that the determined similarity is less than or equal to the third threshold and greater than or equal to 0; when the bit in the Bitmap sequence is the second value, the corresponding second comparison result represents that the determined similarity is greater than the third threshold and less than or equal to 1.
  • the feedback mode includes one of the following:
  • the third information includes an identifier of at least one model, and the at least one model is used to perform channel compression on the downlink channel.
  • the third information includes a third identifier, and the third identifier represents a model for performing channel compression on O eigenvectors in the channel matrix of the downlink channel, where O is an integer greater than 0;
  • the third information includes a third identifier and a fourth identifier, the third identifier represents a model for performing channel compression on O eigenvectors in the channel matrix of the downlink channel, and the fourth identifier represents a model for performing channel compression on P eigenvectors whose changes satisfy a first condition among the O eigenvectors in the channel matrix of the downlink channel, where O and P are integers greater than 0.
  • the sending the second information to the network side includes:
  • the obtaining of the first information includes:
  • the first information is obtained from locally stored information, or the first information sent by the network side is received.
  • the receiving the first information sent by the network side includes one of the following:
  • the embodiment of the present application also provides an information transmission method, which is applied to a network device, including:
  • the third information indicates a channel compression method of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network device;
  • the channel compression result is obtained by performing channel compression on the downlink channel using at least the third information.
  • the method further comprises:
  • First information is sent to the terminal, where the first information is used to determine the stability of the downlink channel, and the second information is determined using at least the first information.
  • the method further comprises:
  • the first information includes fifth information and sixth information, where the fifth information includes the number K of eigenvectors in the channel matrix of the downlink channel, where K is an integer greater than 0; and the sixth information includes at least one threshold value, and the at least one threshold value corresponds to at least two stability levels of the downlink channel.
  • the second information includes a first comparison result, which is obtained by comparing the first parameter with the at least one threshold value, and the first parameter is determined at least using the fifth information.
  • the first parameter represents the similarity between K eigenvectors in the two channel matrices corresponding to the downlink channel at two adjacent moments.
  • the sixth information includes a first threshold and a second threshold, the first threshold is greater than 0 and less than the second threshold; the second threshold is less than 1; the interval formed by the second threshold and 1 corresponds to the first state of the downlink channel, the interval formed by the first threshold and the second threshold corresponds to the second state of the downlink channel, and the interval formed by the first threshold and 0 corresponds to the third state of the downlink channel; the stability of the downlink channel in the first state is higher than the stability in the second state, and the stability of the downlink channel in the second state is higher than the stability in the third state.
  • the first comparison result when the first parameter is greater than or equal to 0 and less than the first threshold, the first comparison result includes a first identifier, and the first identifier represents the third state;
  • the first comparison result includes a second identifier, and the second identifier represents the first state.
  • the sixth information also includes a third threshold, and the third threshold is greater than the first threshold and less than the second threshold; when the first parameter is greater than or equal to the first threshold and less than the second threshold, the first comparison result is obtained using K second comparison results; the K second comparison results are obtained by determining the similarity between each eigenvector of the K eigenvectors of a channel matrix and the corresponding eigenvector in another channel matrix for two channel matrices corresponding to the downlink channel at two adjacent moments, and comparing the determined similarity with the third threshold.
  • the first comparison result represents the K second comparison results in the form of a Bitmap sequence; wherein, when the bit in the Bitmap sequence is the first value, the corresponding second comparison result represents that the determined similarity is less than or equal to the third threshold and greater than or equal to 0; when the bit in the Bitmap sequence is the second value, the corresponding second comparison result represents that the determined similarity is greater than the third threshold and less than or equal to 1.
  • the feedback mode includes one of the following:
  • the terminal feeds back to the network device a channel compression result of O eigenvectors in the channel matrix of the downlink channel, where O is an integer greater than 0;
  • the terminal periodically feeds back to the network device channel compression results of O eigenvectors in the channel matrix of the downlink channel according to a first period, where O is an integer greater than 0;
  • the terminal After the terminal feeds back the channel compression results of the O eigenvectors in the channel matrix of the downlink channel to the network device once, the terminal periodically feeds back the channel compression results of the P eigenvectors among the O eigenvectors in the channel matrix of the downlink channel whose changes meet the first condition to the network device according to a second period, where O and P are integers greater than 0.
  • the third information includes an identifier of at least one model, and the at least one model is used to perform channel compression on the downlink channel.
  • the third information includes a third identifier, and the third identifier represents a model for performing channel compression on O eigenvectors in the channel matrix of the downlink channel, where O is an integer greater than 0;
  • the third information includes a third identifier and a fourth identifier, the third identifier represents a model for performing channel compression on O eigenvectors in the channel matrix of the downlink channel, and the fourth identifier represents a model for performing channel compression on P eigenvectors whose changes satisfy a first condition among the O eigenvectors in the channel matrix of the downlink channel, where O and P are integers greater than 0.
  • the receiving the second information sent by the terminal includes:
  • a CSI report sent by the terminal is received, where the CSI report includes the second information.
  • the sending the first information to the terminal includes one of the following:
  • the first information is sent to the terminal in a unicast manner.
  • the embodiment of the present application further provides an information transmission device, which is arranged on a terminal and includes:
  • An acquiring unit configured to acquire first information, where the first information is used to determine the stability of a downlink channel
  • a first processing unit configured to determine second information using at least the first information, wherein the second information represents stability of the downlink channel
  • a first sending unit configured to send the second information to a network side
  • a first receiving unit configured to receive third information sent by the network side, wherein the third information indicates a channel compression mode of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network side; the third information is determined by at least using the second information;
  • a second processing unit is configured to perform channel compression on the downlink channel using at least the third information to obtain a channel compression result
  • the second sending unit is configured to use the third information to send fourth information to the network side, where the fourth information at least includes the channel compression result.
  • the embodiment of the present application further provides an information transmission device, which is arranged on a network device and includes:
  • a second receiving unit configured to receive second information sent by the terminal, where the second information represents the stability of the downlink channel
  • a third processing unit configured to determine third information using at least the second information, the third information indicating a channel compression method of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network device;
  • a third sending unit configured to send the third information to the terminal
  • the third receiving unit is configured to receive fourth information sent by the terminal, where the fourth information at least includes a channel compression result, and the channel compression result is obtained by performing channel compression on the downlink channel by at least using the third information.
  • the present application also provides a terminal, including:
  • a first processor is configured to obtain first information, where the first information is used to determine the stability of a downlink channel; and determine second information using at least the first information, where the second information represents the stability of the downlink channel;
  • the first communication interface is configured to send the second information to the network side; receive the third information sent by the network side, the third information indicating the channel compression mode of the downlink channel and the feedback mode for feeding back the channel compression result of the downlink channel to the network side; the third information is determined by at least using the second information;
  • the first processor is further configured to perform channel compression on the downlink channel using at least the third information to obtain a channel compression result
  • the first communication interface is further configured to use the third information to send fourth information to the network side, and the fourth information at least includes the channel compression result.
  • the present application also provides a network device, including:
  • a second communication interface configured to receive second information sent by the terminal, where the second information represents the stability of the downlink channel
  • the second processor is configured to use at least the second information to determine third information, wherein the third information indicates a channel compression method of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network device;
  • the second communication interface is further configured to send the third information to the terminal; receive fourth information sent by the terminal, the fourth information at least including a channel compression result, and the channel compression result is obtained by performing channel compression on the downlink channel using at least the third information.
  • the embodiment of the present application further provides a terminal, comprising: a first processor and a first memory configured to store a computer program that can be run on the processor,
  • the first processor is configured to execute the steps of any of the above-mentioned terminal-side methods when running the computer program.
  • the embodiment of the present application also provides a network device, including: a second processor and a a second memory capable of storing a computer program executed on the processor,
  • the second processor is configured to execute the steps of any one of the above-mentioned methods on the network device side when running the computer program.
  • An embodiment of the present application also provides a storage medium on which a computer program is stored.
  • the computer program is executed by a processor, the steps of any of the above-mentioned methods on the terminal side are implemented, or the steps of any of the above-mentioned methods on the network device side are implemented.
  • the terminal obtains first information, the first information is used to determine the stability of the downlink channel; after obtaining the first information, the terminal at least uses the first information to determine second information, and sends the second information to the network device, the second information characterizing the stability of the downlink channel; after receiving the second information, the network device at least uses the second information to determine third information, and sends the third information to the terminal, the third information indicating the channel compression method of the downlink channel and the feedback mode for feeding back the channel compression result of the downlink channel to the network device; after receiving the third information, the terminal at least uses the third information to perform channel compression on the downlink channel to obtain the channel compression result, and uses the third information to send fourth information to the network device, the fourth information at least including the channel compression result.
  • the solution provided by the embodiment of the present application is that since the channel compression method for the terminal to perform downlink channel compression and the feedback mode for the terminal to feed back the channel compression result of the downlink channel to the network side are determined by the network side according to the stability of the downlink channel, the network side can flexibly configure the channel compression method for the terminal to perform downlink channel compression and the feedback mode for the terminal to feed back the channel compression result of the downlink channel to the network side according to the stability of the downlink channel, thereby improving system performance.
  • FIG1 is a schematic diagram of a channel compression and feedback process based on deep learning in the related art
  • FIG2 is a schematic diagram of a flow chart of an information transmission method according to an embodiment of the present application.
  • FIG3 is a flow chart of another information transmission method according to an embodiment of the present application.
  • FIG4 is a schematic diagram of a flow chart of a third information transmission method according to an embodiment of the present application.
  • FIG5 is a schematic diagram of the compression and feedback process of the downlink channel feature vector in the application example of the present application.
  • FIG6 is a schematic diagram of the structure of an information transmission device according to an embodiment of the present application.
  • FIG7 is a schematic diagram of the structure of another information transmission device according to an embodiment of the present application.
  • FIG8 is a schematic diagram of the terminal structure of an embodiment of the present application.
  • FIG9 is a schematic diagram of the structure of a network device according to an embodiment of the present application.
  • FIG. 10 is a schematic diagram of the structure of the information transmission system according to an embodiment of the present application.
  • the terminal can feedback CSI to the base station based on the codebook.
  • the terminal can usually support CSI type (type) I, type II, type II enhanced (enhanced) and other codebook types for feedback of CSI such as rank indication (RI), precoding matrix indication (PMI), and channel quality indication (CQI).
  • CSI type type I
  • type II type II enhanced (enhanced)
  • other codebook types for feedback of CSI such as rank indication (RI), precoding matrix indication (PMI), and channel quality indication (CQI).
  • the codebook used in related technologies is mainly based on the discrete Fourier transform (DFT) matrix, which is characterized by treating the channel as an isotropic random distribution, and then using the DFT vector to evenly divide the beam space.
  • DFT discrete Fourier transform
  • the channels in the actual system do not strictly follow the isotropic distribution law, but are concentrated in a certain subspace.
  • the DFT codebook is limited in performance because it cannot effectively utilize the subspace information where the channel is located; in other words, the CSI fed back by the terminal to the base station based on the codebook may not accurately reflect the actual channel status.
  • the deep learning-based downlink channel compression and feedback scheme shown in Figure 1 can be optimized for specific scenarios, that is, it can be optimized for the actual distribution pattern of the channel, so as to improve the channel feedback accuracy under the same feedback overhead; in other words, compared with the codebook-based CSI feedback scheme, the terminal adopts the deep learning-based downlink channel compression and feedback scheme to enable the base station to obtain accurate downlink CSI.
  • the deep learning-based downlink channel compression and feedback scheme shown in Figure 1 does not take into account the stability of the current channel environment. That is, regardless of whether the downlink channel changes drastically, the terminal continuously feeds back the channel compression results of the downlink channel to the network side at a specific period. The feedback overhead of the terminal is large, which may reduce system performance.
  • the network side flexibly configures the channel compression method of the terminal for downlink channel compression and the feedback mode of the terminal for feeding back the channel compression result of the downlink channel to the network side according to the stability of the downlink channel, thereby improving system performance.
  • the present application provides an information transmission method, which is applied to a terminal. As shown in FIG2 , the method includes:
  • Step 201 Acquire first information, where the first information is used to determine the stability of a downlink channel
  • Step 202 Determine second information using at least the first information, and send the second information to a network side, where the second information represents the stability of the downlink channel;
  • Step 203 receiving third information sent by the network side, wherein the third information indicates a channel compression mode of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network side; the third information is determined by at least using the second information;
  • Step 204 perform channel compression on the downlink channel using at least the third information to obtain a channel compression result, and use the third information to send fourth information to the network side, wherein the fourth information at least includes the channel compression result.
  • the terminal may also be called user equipment (UE, User Equipment) or user.
  • UE User Equipment
  • user User Equipment
  • the stability of the downlink channel can also be understood as the degree of change of the downlink channel or the severity of the change of the downlink channel.
  • the expression form can be set according to requirements, such as the degree of change in the correlation of the characteristic vectors of the downlink channel at adjacent moments (ie, the similarity of the characteristic vectors of the downlink channel at adjacent moments), etc., which is not limited in the embodiments of the present application.
  • the first information may indicate the number K of eigenvectors in the channel matrix of the downlink channel, where K is an integer greater than 0; accordingly, the terminal may determine the similarity between the K eigenvectors in the two channel matrices corresponding to the downlink channel at two adjacent moments based on the channel matrix and K of the downlink channel obtained by measuring the CSI reference signal (CSI-RS), that is, determine the stability of the downlink channel.
  • CSI-RS CSI reference signal
  • the first information may include fifth information, where the fifth information includes the number of eigenvectors K in the channel matrix of the downlink channel, where K is an integer greater than 0; accordingly, a specific implementation of step 202 may include:
  • At least the fifth information is used to determine a first parameter, where the first parameter represents the similarity between K eigenvectors in two channel matrices corresponding to the downlink channel at two adjacent moments; and the second information includes the first parameter.
  • an eigenvector may refer to a singular vector; accordingly, an eigenvalue corresponding to the eigenvector may refer to a singular value corresponding to the singular vector.
  • the obtaining of the first information may include: obtaining the fifth information.
  • the terminal may obtain the fifth information from locally stored information; or, the terminal may obtain the fifth information from the network side, that is, the network side may send the fifth information to the terminal, and the terminal may receive the fifth information sent by the network side.
  • the specific manner in which the terminal obtains the fifth information may be set as required, and this embodiment of the present application does not limit this.
  • the network device on the network side when the fifth information is sent from the network side to the terminal, the network device on the network side (specifically including a base station) can decide the size of K according to one or more factors such as the current user's business requirements (such as CSI accuracy requirements, etc.), channel characteristics, feedback overhead, etc., that is, determine the fifth information. For example, the higher the user's demand for CSI accuracy, the larger the value of K; the specific way in which the network device determines the fifth information can be set according to demand, and this embodiment of the application does not limit this.
  • K needs to be less than or equal to the total number of eigenvectors in the channel matrix of the downlink channel.
  • the terminal can select K left singular vectors or K right singular vectors in the two channel matrices corresponding to the downlink channel at two adjacent moments to calculate the first parameter, and the terminal can calculate the first parameter based on the calculation of cosine similarity (which can be expressed as Cosine Similarity in English).
  • H(t) U(t) ⁇ (t)V T (t).
  • the K eigenvectors in the channel matrix H(t) of the downlink channel at time t may include U 1 (t), U 2 (t)..., U K (t) or Since the cosine similarity is the cosine of the angle between two j-dimensional vectors in a j-dimensional space (j is an integer greater than 0), it is equal to the dot product (i.e., vector product) of the two vectors divided by the product of the lengths (or sizes) of the two vectors.
  • the cosine similarity (i is an integer greater than 0) between the i-th eigenvector in H(t 1 ) and the i-th eigenvector in H(t 2 ) can be calculated by left singular vectors (i.e., formula (1)) or right singular vectors (i.e., formula (2)):
  • the first parameter can be calculated by the following formula:
  • the network device when the second information includes the first parameter, after the network device receives the second information, it can at least use the first parameter to flexibly configure the channel compression method of the downlink channel for the terminal and the feedback mode for feeding back the channel compression result of the downlink channel to the network device.
  • the at least one threshold may include a first threshold and a second threshold, wherein the first threshold is greater than 0 and less than the second threshold, and the second threshold is less than 1; the interval formed by the second threshold and 1 (i.e., (second threshold, 1] or [second threshold, 1] or [second threshold, 1) or (second threshold, 1)) may correspond to the first state of the downlink channel, the interval formed by the first threshold and the second threshold (i.e., (first threshold, second threshold] or [first threshold, second threshold] or [first threshold, second threshold) or (first threshold, second threshold)) may correspond to the second state of the downlink channel, and the interval formed by the first threshold and 0 (i.e., (0, first threshold] or [0, first threshold] or [0, first threshold) or (0, first threshold)) may correspond to the third state of the downlink channel; the stability of the downlink channel in the first state is higher than that in the second state, and the stability of the downlink channel in the second state is higher than that in the third state.
  • the way to divide the stability level of the downlink channel, the channel compression method corresponding to each stability level, and the feedback mode of the channel compression result can be set according to needs, and the embodiments of the present application do not limit this.
  • the network device can configure the terminal to no longer periodically feedback the channel compression result to the network side, such as feeding back the channel compression result once or not feeding back the channel compression result; that is, the feedback mode corresponding to the first state may include: the terminal feeds back the channel compression result of O eigenvectors in the channel matrix of the downlink channel to the network side once, O is an integer greater than 0, and O is less than or equal to the total number of eigenvectors in the channel matrix of the downlink channel.
  • the network device may configure the terminal to feed back the channel compression result to the network side according to a specific period; that is, the feedback mode corresponding to the third state may include: the terminal periodically feeds back the channel compression results of O eigenvectors in the channel matrix of the downlink channel to the network side according to the first period.
  • the network device can configure the terminal to incrementally feedback the channel compression result, that is, the feedback mode corresponding to the second state may include: after the terminal feeds back the channel compression results of O eigenvectors in the channel matrix of the downlink channel to the network side once, according to the second period, periodically feeds back to the network side the channel compression results of P eigenvectors in the O eigenvectors of the channel matrix of the downlink channel whose changes meet the first condition, where P is an integer greater than 0, and P is less than or equal to O.
  • the network device can flexibly configure the feedback mode of the terminal to feed back the channel compression result to the network side according to the stability of the downlink channel, such as the feedback modes corresponding to the first state, the second state and the third state; in this way, the terminal no longer needs to continuously feed back the channel compression result to the network side according to a specific period, but can adopt a feedback mode that flexibly changes with the stability of the downlink channel to feed back the channel compression result to the network side, thereby reducing the feedback overhead of the terminal, that is, effectively reducing the wireless transmission overhead, and thus improving the system performance.
  • the stability of the downlink channel such as the feedback modes corresponding to the first state, the second state and the third state
  • first period and the second period may be the same or different in size.
  • the sizes of the first cycle and the second cycle can be set according to specific needs, and the embodiments of the present application do not limit this.
  • the size of O can be set according to needs, and the embodiment of the present application does not limit the size of O and its decision-making method.
  • the sizes of O and K can be the same or different, that is, O can be equal to, greater than, or less than K, and can be set according to needs.
  • the embodiment of the present application does not limit the size relationship between O and K.
  • the first condition may include a third threshold value, which is greater than the first threshold value and less than the second threshold value.
  • the similarity between the i-th eigenvector in the K eigenvectors of one channel matrix and the corresponding eigenvector in the other channel matrix can be calculated using formula (1) or formula (2).
  • the first threshold, the second threshold and the third threshold can be set according to the requirements, and the embodiment of the present application does not limit this.
  • the first threshold can be equal to 0.2
  • the second threshold can be equal to 0.8
  • the third threshold can be equal to 0.7.
  • the first threshold and the second threshold may be pre-set on the terminal, or the network device may send the first threshold and the second threshold to the terminal. After the terminal determines the first parameter, the comparison result of the first parameter with the first threshold and the second threshold may be directly fed back to the network device as the second information.
  • the first information may further include sixth information (that is, the first information includes fifth information and sixth information), and the sixth information includes at least one threshold value, and the at least one threshold value corresponds to at least two stability levels of the downlink channel; accordingly, another specific implementation of step 202 may include:
  • the first parameter represents a similarity between K eigenvectors in two channel matrices corresponding to the downlink channel at two adjacent moments;
  • the first parameter is compared with the at least one threshold to obtain a first comparison result, and the second information includes the first comparison result.
  • the obtaining of the first information may further include: obtaining the sixth information.
  • the terminal may obtain the sixth information from locally stored information; or, the terminal may obtain the sixth information from the network side, that is, the network side may send the sixth information to the terminal, and the terminal may receive the sixth information sent by the network side.
  • the specific manner in which the terminal obtains the sixth information may be set according to requirements, and this embodiment of the application does not make any provision for this. limited.
  • the sixth information may include the first threshold and the second threshold.
  • the first comparison result may include a first identifier, and the first identifier represents the third state;
  • the first comparison result may include a second identifier, and the second identifier represents the first state.
  • the network device may configure the terminal to periodically feed back the channel compression results of the O eigenvectors in the channel matrix of the downlink channel to the network side according to the first period according to the first identifier; or, according to the second identifier, configure the terminal to feed back the channel compression results of the O eigenvectors in the channel matrix of the downlink channel to the network side once.
  • the feedback mode indicated by the third information may include: periodically feeding back the channel compression results of the O eigenvectors in the channel matrix of the downlink channel to the network side according to the first period; when the first comparison result includes the second identifier, the feedback mode indicated by the third information may include: feeding back the channel compression results of the O eigenvectors in the channel matrix of the downlink channel to the network side once.
  • the expression forms of the first identifier and the second identifier can be set according to requirements, and the embodiment of the present application does not limit this.
  • the first identifier can be recorded as -1 and the second identifier can be recorded as -2.
  • the network device can configure the terminal to incrementally feedback the channel compression result; at this time, the third threshold can be pre-set on the terminal, or the network device can send the third threshold to the terminal, so that the terminal can determine whether the change of the characteristic vector meets the first condition.
  • the sixth information may further include the third threshold; and the comparing the first parameter with the at least one threshold to obtain the first comparison result may include:
  • the first parameter is greater than or equal to the first threshold and less than the second threshold, for two channel matrices corresponding to the downlink channel at two adjacent moments, determine a similarity between each eigenvector of K eigenvectors of one channel matrix and a corresponding eigenvector in the other channel matrix, and compare the determined similarity with the third threshold to obtain K second comparison results;
  • the first comparison result is obtained by using the K second comparison results.
  • the similarity between each eigenvector of the K eigenvectors of one channel matrix and the corresponding eigenvector in the other channel matrix can be calculated using formula (1) or formula (2).
  • the first comparison result may represent the K second comparison results in the form of a Bitmap sequence (that is, the first comparison result may include a K-dimensional Bitmap sequence);
  • the corresponding second comparison result represents that the determined similarity is less than or equal to the third threshold and is greater than or equal to 0, that is, the change of the corresponding feature vector meets the first condition;
  • the bit in the Bitmap sequence is the second value
  • the corresponding second comparison result represents that the determined similarity is greater than the third threshold and is less than or equal to 1, that is, the change of the corresponding feature vector does not meet the first condition.
  • the first value can be equal to 0, and the second value can be equal to 1; or, the first value can be equal to 1, and the second value can be equal to 0; the size of the first value and the second value can be set according to specific needs, and the embodiment of the present application is not limited to this.
  • the network device can configure the terminal to feed back the channel compression results of the O eigenvectors in the channel matrix of the downlink channel to the network side once according to the Bitmap sequence, and then, according to the second period, periodically feed back to the network side the channel compression results of the P eigenvectors whose changes in the O eigenvectors of the channel matrix of the downlink channel meet the first condition, that is, periodically feed back to the network side the channel compression results of the P eigenvectors whose corresponding bits in the Bitmap sequence are the first value.
  • the feedback mode indicated by the third information may include: after feeding back the channel compression results of the O eigenvectors in the channel matrix of the downlink channel to the network side once, according to the second period, periodically feeding back to the network side the channel compression results of P eigenvectors whose changes in the O eigenvectors of the channel matrix of the downlink channel meet the first condition (i.e., periodically feeding back to the network side the channel compression results of the P eigenvectors whose corresponding bits in the Bitmap sequence are the first values).
  • the feedback mode indicated by the third information may include one of the following:
  • the mode of single-transfer compression of the channel feature vector corresponding to the first state (which may be referred to as the first feedback mode in the subsequent description) is to feed back to the network side the channel compression result of O feature vectors in the channel matrix of the downlink channel once;
  • a mode of updating the eigenvector corresponding to the first value in the Bitmap sequence periodically corresponding to the second state (this mode can be understood as an incremental feedback model, which can be recorded as the second feedback mode in the subsequent description), that is, after feeding back the channel compression results of the O eigenvectors in the channel matrix of the downlink channel to the network side once, according to the second period, periodically feeding back the channel compression results of the P eigenvectors (that is, the eigenvectors whose corresponding bits in the Bitmap sequence are the first value) among the O eigenvectors of the channel matrix of the downlink channel whose changes meet the first condition to the network side;
  • the mode of periodically transmitting compressed channel eigenvectors corresponding to the third state (which may be referred to as the third feedback mode in the subsequent description) is, that is, according to the first period, periodically feeding back to the network side the channel compression results of the O eigenvectors in the channel matrix of the downlink channel.
  • the third information may include an identifier of at least one model.
  • a model is used to perform channel compression on the downlink channel; in other words, the channel compression method for the downlink channel indicated by the third information includes: performing channel compression on the downlink channel using at least one model corresponding to an identifier of at least one model contained in the third information.
  • the network side may collect downlink channel estimation data in advance, preprocess the downlink channel estimation data through singular value decomposition, and use the preprocessed singular vectors and other relevant information (such as CSI accuracy requirements, the number of eigenvectors, etc.) as inputs to offline train a channel compression model (the number (i.e., the identifier) of the compression model can be recorded as E1 , E2 ..., EN ) and a channel decompression model (the number (i.e., the identifier) of the decompression model can be recorded as D1 , D2 ..., DN ) for channel eigenvectors based on an artificial intelligence (AI) autoencoder.
  • a channel compression model the number (i.e., the identifier) of the compression model can be recorded as E1 , E2 ..., EN
  • a channel decompression model the number (i.e., the identifier) of the decompression model can be recorded as D1 , D2 ...,
  • the channel compression model and the channel decompression model are one-to-one corresponding, that is, the channel compression result obtained by using a channel compression model needs to be restored by using a corresponding channel decompression model.
  • N can be an integer greater than 0, and the size of N can be set according to requirements (such as CSI accuracy requirements, number of eigenvectors, etc.), that is, different channel compression models/channel decompression models can correspond to different CSI accuracy requirements, different numbers of eigenvectors, etc.; in other words, different channel compression models/channel decompression models have different functions.
  • the channel compression model and the corresponding number can be synchronized to the terminal (for example, configured locally in the terminal before leaving the factory), and the channel decompression model and the corresponding number (i.e., the identifier) can be synchronized to the network device (for example, pre-configured locally in the base station).
  • the network side can set the association between the model number and the model function (the function of the model can be reflected by parameters such as CSI accuracy and the number of eigenvectors) on the terminal and the network device.
  • the network side may deploy an AI model training system, which collects downlink channel estimation data in advance, pre-processes the downlink channel estimation data through singular value decomposition, and uses the pre-processed singular vectors and other related information (such as CSI accuracy requirements, number of eigenvectors, etc.) as input to perform offline training on the channel compression model and channel decompression model for channel eigenvectors based on the AI autoencoder.
  • an AI model training system which collects downlink channel estimation data in advance, pre-processes the downlink channel estimation data through singular value decomposition, and uses the pre-processed singular vectors and other related information (such as CSI accuracy requirements, number of eigenvectors, etc.) as input to perform offline training on the channel compression model and channel decompression model for channel eigenvectors based on the AI autoencoder.
  • the specific deployment method of the AI model training system can be set according to demand, and the embodiments of the present application are not limited to this.
  • the AI model training system can be deployed on a centralized unit (CU) and/or distributed unit (DU) of a base station, that is, the channel compression model and channel decompression model are trained by one base station; or, the AI model training system can be deployed across CUs and/or DUs, that is, deployed on CUs and/or DUs of multiple base stations. In other words, at least two base stations jointly train the channel compression model and channel decompression model.
  • CU centralized unit
  • DU distributed unit
  • the network device that sends the first information to the terminal may participate in the training of the channel compression model and the channel decompression model, or may not participate in the training of the channel compression model and the channel decompression model.
  • the specific setting can be based on the requirements (such as network deployment requirements, etc.), and the embodiment of the present application does not limit this.
  • At least one threshold value i.e., the sixth information
  • the network device may set the first information based on the sixth information.
  • K i.e., the fifth information
  • K is determined based on one or more factors such as the terminal's business requirements (such as CSI accuracy requirements, etc.), channel characteristics, feedback overhead, etc., and K and at least one pre-set threshold are sent to the terminal, i.e., the first information is sent to the terminal.
  • the network device in the process of the network device at least using the second information to determine the third information, can first use the first comparison result to determine the feedback mode (the first feedback mode or the second feedback mode or the third feedback mode) in which the terminal feeds back the channel compression result of the downlink channel to the network side, and then determine the channel compression method of the downlink channel based on the determined feedback mode, the characteristic vector number O and the association between the pre-set model number (i.e., the identifier) and the model function (the function of the model can be reflected by parameters such as the characteristic vector number), that is, determine at least one model for performing channel compression on the downlink channel.
  • the feedback mode the first feedback mode or the second feedback mode or the third feedback mode
  • the channel compression method of the downlink channel based on the determined feedback mode, the characteristic vector number O and the association between the pre-set model number (i.e., the identifier) and the model function (the function of the model can be reflected by parameters such as the characteristic vector number), that is, determine at least one model for
  • the third information may include a third identifier, the third identifier representing a model for performing channel compression on O eigenvectors in the channel matrix of the downlink channel; in a case where the feedback mode indicated by the third information includes the second feedback mode, the third information may include a third identifier, a fourth identifier and a second period, the fourth identifier representing a model for performing channel compression on P eigenvectors whose changes satisfy the first condition among the O eigenvectors in the channel matrix of the downlink channel; in a case where the feedback mode indicated by the third information includes the third feedback mode, the third information may include the third identifier and the first period.
  • the terminal after receiving the third information, when the third information includes the third identifier, the terminal can use the model corresponding to the third identifier to perform channel compression on the O eigenvectors in the channel matrix of the downlink channel, obtain a first channel compression result including O eigenvalues, and send the fourth information including at least the first channel compression result to the network side once.
  • the terminal can use the model corresponding to the third identifier to perform channel compression on the O eigenvectors in the channel matrix of the downlink channel, obtain a second channel compression result including O eigenvalues, and send the fourth information including at least the second channel compression result to the network side once; after completing the first feedback of the channel compression result, the terminal can use the model corresponding to the fourth identifier to perform channel compression on the P eigenvectors whose changes meet the first condition among the O eigenvectors of the channel matrix of the downlink channel according to the second period, obtain a third channel compression result including P eigenvalues, and periodically send the fourth information including at least the third channel compression result to the network side.
  • the terminal can periodically use the model corresponding to the third identifier to perform channel compression on O eigenvectors in the channel matrix of the downlink channel according to the first period, obtain a fourth channel compression result including O eigenvalues, and periodically send fourth information including at least the fourth channel compression result to the network side.
  • the first information may include the fifth information and the sixth information; the terminal may obtain the fifth information from the locally stored information, or Alternatively, the terminal may obtain the fifth information from the network side; at the same time, the terminal may obtain the sixth information from the locally stored information, or the terminal may obtain the sixth information from the network side.
  • obtaining the first information may include:
  • the first information is obtained from locally stored information, or the first information sent by the network side is received.
  • the obtaining of the first information may include one of the following:
  • the sixth information is obtained from the locally stored information, and the fifth information sent by the network side is received.
  • the network device may broadcast the first information, or may send the first information to the terminal in a multicast or unicast manner.
  • the receiving of the first information sent by the network side may include one of the following:
  • the terminal may send the second information to the network device via a CSI report (which may be expressed as CSI-ReportConfig in English).
  • CSI report (which may be expressed as CSI-ReportConfig in English).
  • sending the second information to the network side may include:
  • the network device may send the third information to the terminal via RRC signaling.
  • the receiving the third information sent by the network side may include: receiving RRC signaling sent by the network side, the RRC signaling including the third information.
  • the form in which the terminal sends the fourth information to the network side can be set according to needs, such as sending the fourth information to the network side through a CSI report; it can be understood that in addition to the channel compression result, the fourth information may also include other information that the terminal needs to feedback to the network side (such as any information that the CSI report can carry).
  • the terminal may periodically determine the second information according to the third period, and periodically send the second information to the network side; after the network side determines and sends the third information to the terminal for the first time, it may periodically update the third information using the second information sent by the terminal, and periodically send the updated third information to the terminal; the terminal may obtain and feedback the channel compression result using the updated third information.
  • the size of the period can be set according to demand, for example, the third period can be set to be consistent with the CSI-RS transmission period, etc., which is not limited in the embodiment of the present application.
  • the third period can be pre-set on the terminal, or the network device can configure the third period to the terminal in a specific manner (such as RRC signaling, etc., the embodiment of the present application does not limit the configuration method of the third period).
  • an embodiment of the present application further provides an information transmission method, which is applied to a network device (specifically, may include a base station). As shown in FIG3 , the method includes:
  • Step 301 receiving second information sent by a terminal, where the second information represents stability of a downlink channel;
  • Step 302 using at least the second information, determining third information, and sending the third information to the terminal, the third information indicating the channel compression mode of the downlink channel and a feedback mode for feeding back the channel compression result of the downlink channel to the network device;
  • Step 303 Receive fourth information sent by the terminal, wherein the fourth information at least includes a channel compression result, and the channel compression result is obtained by performing channel compression on the downlink channel by at least using the third information.
  • the method may further include:
  • First information is sent to the terminal, where the first information is used to determine the stability of the downlink channel, and the second information is determined using at least the first information.
  • the method may further include:
  • the first information includes fifth information and sixth information, where the fifth information includes the number K of eigenvectors in the channel matrix of the downlink channel, where K is an integer greater than 0; and the sixth information includes at least one threshold value, and the at least one threshold value corresponds to at least two stability levels of the downlink channel.
  • the receiving the second information sent by the terminal may include:
  • a CSI report sent by the terminal is received, where the CSI report includes the second information.
  • sending the first information to the terminal may include one of the following:
  • the first information is sent to the terminal in a unicast manner.
  • the embodiment of the present application further provides an information transmission method, as shown in FIG4 , the method comprising:
  • Step 401 The terminal obtains first information, determines second information by at least using the first information, and sends the second information to a network device, wherein the first information is used to determine the stability of a downlink channel, and the second information represents the stability of the downlink channel;
  • Step 402 The network device receives the second information sent by the terminal, determines third information using at least the second information, and sends the third information to the terminal, wherein the third information indicates a channel compression method of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network device;
  • Step 403 the terminal receives the third information sent by the network device, performs channel compression on the downlink channel by at least using the third information to obtain a channel compression result, and sends fourth information to the network device by using the third information, wherein the fourth information at least includes the channel compression result;
  • Step 404 The network device receives the fourth information sent by the terminal.
  • a terminal obtains first information, and the first information is used to determine the stability of a downlink channel; after obtaining the first information, the terminal at least uses the first information to determine second information, and sends the second information to a network device, wherein the second information represents the stability of the downlink channel; after receiving the second information, the network device at least uses the second information to determine third information, and sends the third information to the terminal, wherein the third information indicates a channel compression method of the downlink channel and a feedback mode for feeding back the channel compression result of the downlink channel to the network device; after receiving the third information, the terminal at least uses the third information to perform channel compression on the downlink channel to obtain a channel compression result, and uses the third information to send fourth information to the network device, wherein the fourth information at least includes the channel compression result.
  • the network side can flexibly configure the channel compression mode for the terminal to perform downlink channel compression and the feedback mode for the terminal to feed back the channel compression result of the downlink channel to the network side according to the stability of the downlink channel, such as the feedback modes corresponding to the first state, the second state and the third state; in this way, the terminal no longer needs to continuously feed back the channel compression result to the network side according to a specific period, but can adopt a feedback mode that flexibly changes with the stability of the downlink channel to feed back the channel compression result to the network side, thereby reducing the feedback overhead of the terminal, that is, effectively reducing the wireless transmission overhead, and thus improving the system performance.
  • the network device includes a base station; the first threshold, the second threshold and the third threshold are called correlation thresholds or change degree thresholds, the first threshold is represented by ⁇ 1,K , the second threshold is represented by ⁇ 2,K , and the third threshold is represented by ⁇ 3,K ; O is equal to K.
  • the base station decides the number of channel feature vectors K and the correlation thresholds ⁇ 1,K , ⁇ 2,K and ⁇ 3,K , the terminal determines the downlink channel correlation ⁇ (i.e. the above-mentioned first parameter) based on K, ⁇ 1,K , ⁇ 2,K and ⁇ 3,K (i.e.
  • the base station decides the downlink channel compression model and feedback mode (i.e. the above-mentioned third information) according to the comparison result of the channel correlation fed back by the terminal and the correlation threshold, and the terminal performs channel compression based on the decision of the base station and feeds back the channel compression result.
  • the compression and feedback process for the downlink channel feature vector may include the following steps:
  • Step 501 The network side collects data, trains an AI-based channel compression/decompression model offline, synchronizes the data to the base station and the terminal, and presets a threshold parameter (i.e., the correlation threshold), and then executes step 502;
  • a threshold parameter i.e., the correlation threshold
  • Step 502 The terminal accesses the network and enters the RRC-CONNECTED state, and then executes step 503;
  • Step 503 The base station determines the number of channel compression vectors, and then executes step 504;
  • Step 504 the base station sends a channel correlation threshold (i.e., the sixth information) and a number of eigenvectors (i.e., the fifth information) to the terminal, and sends a CSI-RS according to the configuration, and then executes step 505;
  • a channel correlation threshold i.e., the sixth information
  • a number of eigenvectors i.e., the fifth information
  • Step 505 The terminal calculates the downlink channel correlation (ie, the first parameter mentioned above), and then executes step 506;
  • Step 506 The terminal feeds back the comparison result of the downlink channel correlation and the threshold (ie, the second information) through CSI-ReportConfig, and then executes step 507;
  • Step 507 The base station determines the compression model number and other parameters (ie, the third information), and then executes step 508;
  • Step 508 The base station sends the model number and other parameters to the terminal, and then executes step 509;
  • Step 509 The terminal compresses the channel information according to the model number in the configuration parameter, and then executes step 510;
  • Step 510 the terminal feeds back the compressed channel information (i.e. the fourth information at least including the channel compression result) according to the configuration, and then executes step 511;
  • the compressed channel information i.e. the fourth information at least including the channel compression result
  • Step 511 The base station decompresses the channel information, and then executes step 512;
  • Step 512 According to the CSI-RS channel correlation of the terminal, the base station updates the sending model number and other parameters, and the terminal adjusts the model and reports the compressed channel information.
  • an AI model training system can be deployed in the wireless access network (i.e., the above-mentioned network side), and the AI model training system can be deployed in the CU and/or DU of the base station, or on a logical entity across CU and/or DU.
  • the AI model training system can pre-collect downlink channel estimation data, pre-process the downlink channel estimation data by singular value decomposition, and use the pre-processed singular vectors and other related information as input to offline train the channel compression model (the number (i.e., identifier) of the compression model can be recorded as E 1 , E 2 ..., E N ) and the channel decompression model (the number (i.e., identifier) of the decompression model can be recorded as D 1 , D 2 ..., D N ) for the channel feature vector based on the AI autoencoder.
  • the AI model training system can synchronize the trained channel compression model and the corresponding number to the terminal, and synchronize the trained channel decompression model and number to the base station.
  • the base station may record the comparison result T d between the downlink channel correlation fed back by the terminal and the correlation threshold; in step 501, the base station may initially set (ie, initialize) T d to -3.
  • the base station may decide the size of the number K of channel feature vectors that the terminal needs to compress based on one or more factors such as the current user's service requirements (such as CSI accuracy requirements, etc.), channel characteristics, feedback overhead, etc.
  • the base station may also pre-set the same or different correlation thresholds corresponding to different numbers K of feature vectors in step 501 according to the requirements for dividing the channel variation degree.
  • the base station may send K, ⁇ 1,K , ⁇ 2,K and ⁇ 3,K to the terminal by broadcast, unicast or multicast, and send CSI-RS according to the configuration.
  • the terminal may calculate the downlink channel correlation based on the measurement of the CSI-RS.
  • the channel correlation may be obtained by calculating the correlation (i.e., similarity) of the channel eigenvectors, and the calculation of the channel eigenvector correlation may select the left singular vector or the right singular vector, and the method for calculating the channel correlation may include but is not limited to the cosine similarity method.
  • the terminal may feed back the comparison result Td to the base station via CSI-ReportConfig.
  • the base station can select a model number (i.e., the third identifier) that can compress K channel feature vectors and decide the feature vector feedback period T1 (i.e., the first period); if T d is -2, the base station can select a model number (i.e., the third identifier) that can compress K channel feature vectors, and the compression model corresponding to the number is used for the first compression feedback of the channel feature vector, that is, only the compressed channel information (i.e., the fourth information that at least includes the channel compression result) is fed back once; when T d is a K-dimensional Bitmap sequence, the base station can decide the feedback period T2 (i.e., the second period) and select two compression models, the first compression model (i.e., the model corresponding to the third identifier) is used to compress K channel feature vectors for the first time and feedback them, and the second compression model (i.e., the model corresponding to the fourth identifier) is
  • the base station may not execute step 507 (ie, the base station does not actually receive T d fed back by the terminal, and T d is still in the initialization state), and ends the current process, or goes to step 505 or step 512.
  • the base station may send the model number (ie, the third identifier, or the third identifier and the fourth identifier) and other parameters (eg, periods T1 and T2) to the terminal.
  • the model number ie, the third identifier, or the third identifier and the fourth identifier
  • other parameters eg, periods T1 and T2
  • the terminal can select a corresponding channel compression model according to the channel compression model number (ie, the third identifier, or the third identifier and the fourth identifier) determined by the base station, and complete Channel compression for the first K eigenvectors of the channel.
  • the channel compression model number ie, the third identifier, or the third identifier and the fourth identifier
  • the terminal may feed back the channel compression information (ie, the fourth information at least including the channel compression result) to the base station according to the feedback mode agreed with the base station (ie, the feedback mode indicated by the base station in step 508).
  • the channel compression information ie, the fourth information at least including the channel compression result
  • the base station may restore (ie, decompress) corresponding downlink channel information using a corresponding decompression model (ie, a channel decompression model paired with a compression model used by the terminal) according to the feedback from the terminal.
  • a corresponding decompression model ie, a channel decompression model paired with a compression model used by the terminal
  • the terminal can periodically calculate the channel correlation according to a preset period (i.e., the third period mentioned above, which can be consistent with the CSI-RS transmission period) and feed back the comparison result between the correlation and the threshold to the base station; the base station can update (i.e., reconfigure) the compression model number and channel feedback mode of the terminal according to the feedback of the terminal, and the terminal can report the compressed channel information according to the compression model and feedback mode reconfigured by the base station.
  • the terminal can periodically calculate the channel correlation according to the period of sending CSI-RS by the base station and record the comparison result between the correlation and the threshold. like is equal to T d , then steps 509 to 510 can be executed; if is not equal to T d , then you need to use Update T d , and execute steps 506 to 510 .
  • the solution provided by this application example adopts a feedback mode that can flexibly change with the stability of the downlink channel to feed back the channel compression result of the downlink channel to the network side, which can reduce the feedback overhead of the terminal, thereby effectively reducing the wireless transmission overhead, and further improving the system performance.
  • the embodiment of the present application further provides an information transmission device, which is arranged on the terminal, as shown in FIG6 , and includes:
  • An acquiring unit 601 is configured to acquire first information, where the first information is used to determine the stability of a downlink channel;
  • a first processing unit 602 is configured to determine second information using at least the first information, where the second information represents the stability of the downlink channel;
  • a first sending unit 603 is configured to send the second information to the network side
  • the first receiving unit 604 is configured to receive third information sent by the network side, wherein the third information indicates a channel compression mode of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network side; the third information is determined using at least the second information;
  • the second processing unit 605 is configured to perform channel compression on the downlink channel by at least using the third information to obtain a channel compression result
  • the second sending unit 606 is configured to send fourth information to the network side using the third information, where the fourth information at least includes the channel compression result.
  • the first information includes fifth information and sixth information
  • the fifth information includes the number of eigenvectors K in the channel matrix of the downlink channel, K is an integer greater than 0
  • the sixth information includes at least one threshold, and the at least one threshold corresponds to at least two stability levels of the downlink channel
  • the first processing unit 602 is further configured to:
  • the first parameter represents a similarity between K eigenvectors in two channel matrices corresponding to the downlink channel at two adjacent moments;
  • the first parameter is compared with the at least one threshold to obtain a first comparison result, and the second information includes the first comparison result.
  • the sixth information further includes a third threshold, and the third threshold is greater than the first threshold and less than the second threshold; the first processing unit 602 is further configured to:
  • the first parameter is greater than or equal to the first threshold and less than the second threshold, for two channel matrices corresponding to the downlink channel at two adjacent moments, determine a similarity between each eigenvector of K eigenvectors of one channel matrix and a corresponding eigenvector in the other channel matrix, and compare the determined similarity with the third threshold to obtain K second comparison results;
  • the first comparison result is obtained by using the K second comparison results.
  • the acquisition unit 601 is further configured to acquire the first information from locally stored information, or receive the first information sent by the network side.
  • the first sending unit 603 is further configured to send a CSI report to the network side, where the CSI report includes the second information.
  • the obtaining unit 601 is further configured to perform one of the following operations:
  • the acquisition unit 601 can be implemented by the processor in the terminal in combination with the communication interface; the first sending unit 603, the first receiving unit 604 and the second sending unit 606 can be implemented by the communication interface in the terminal; the first processing unit 602 and the second processing unit 605 can be implemented by the processor in the terminal.
  • the embodiment of the present application further provides an information transmission device, which is arranged on the network device, as shown in FIG7 , and includes:
  • the second receiving unit 701 is configured to receive second information sent by the terminal, where the second information represents the stability of the downlink channel;
  • the third processing unit 702 is configured to determine third information using at least the second information, wherein the third information indicates a channel compression method of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network device;
  • a third sending unit 703, configured to send the third information to the terminal
  • the third receiving unit 704 is configured to receive fourth information sent by the terminal, where the fourth information at least includes a channel compression result, and the channel compression result is obtained by performing channel compression on the downlink channel by at least using the third information.
  • the apparatus may further include a fourth sending unit 705 configured to send first information to the terminal, wherein the first information is used to determine the stability of the downlink channel, and the second information is determined at least using the first information.
  • a fourth sending unit 705 configured to send first information to the terminal, wherein the first information is used to determine the stability of the downlink channel, and the second information is determined at least using the first information.
  • the device may also include a fourth processing unit 706, configured to determine the first information, the first information includes fifth information and sixth information, the fifth information includes the number of eigenvectors K in the channel matrix of the downlink channel, K is an integer greater than 0; the sixth information includes at least one threshold, and the at least one threshold corresponds to at least two stability levels of the downlink channel.
  • a fourth processing unit 706 configured to determine the first information, the first information includes fifth information and sixth information, the fifth information includes the number of eigenvectors K in the channel matrix of the downlink channel, K is an integer greater than 0; the sixth information includes at least one threshold, and the at least one threshold corresponds to at least two stability levels of the downlink channel.
  • the second receiving unit 701 is further configured to receive a CSI report sent by the terminal, where the CSI report includes the second information.
  • the fourth sending unit 705 is further configured to perform one of the following operations:
  • the first information is sent to the terminal in a unicast manner.
  • the second receiving unit 701, the third sending unit 703, the third receiving unit 704 and the fourth sending unit 705 can be implemented by the communication interface in the network device;
  • the third processing unit 702 and the fourth processing unit 706 can be implemented by the processor in the network device.
  • the information transmission device provided in the above embodiment only uses the division of the above program modules as an example when performing information transmission.
  • the above processing can be assigned to different program modules as needed, that is, the internal structure of the device is divided into different program modules to complete all or part of the processing described above.
  • the information transmission device provided in the above embodiment and the information transmission method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.
  • the embodiment of the present application further provides a terminal, as shown in FIG8 , the terminal 800 includes:
  • the first communication interface 801 is capable of exchanging information with other terminals and/or the network side;
  • a first processor 802 is connected to the first communication interface 801 to implement information interaction with other terminals and/or the network side, and is configured to execute the method provided by one or more technical solutions of the above terminal side when running a computer program;
  • a first memory 803 on which the computer program is stored.
  • the first processor 802 is configured to obtain first information, where the first information is used to determine the stability of the downlink channel; determine second information using at least the first information, where the second information represents the stability of the downlink channel;
  • the first communication interface 801 is configured to send the second information to the network side; receive the third information sent by the network side, the third information indicating the channel compression mode of the downlink channel and the feedback mode for feeding back the channel compression result of the downlink channel to the network side; the third information is determined by at least using the second information; wherein,
  • the first processor 802 is further configured to perform channel compression on the downlink channel by at least using the third information to obtain a channel compression result;
  • the first communication interface 801 is further configured to use the third information to send fourth information to the network side, where the fourth information at least includes the channel compression result.
  • the first information includes fifth information and sixth information
  • the fifth information includes the number of eigenvectors K in the channel matrix of the downlink channel, K is an integer greater than 0
  • the sixth information includes at least one threshold, and the at least one threshold corresponds to at least two stability levels of the downlink channel
  • the first processor 802 is further configured to:
  • the first parameter represents a similarity between K eigenvectors in two channel matrices corresponding to the downlink channel at two adjacent moments;
  • the first parameter is compared with the at least one threshold to obtain a first comparison result, and the second information includes the first comparison result.
  • the sixth information further includes a third threshold, and the third threshold is greater than the first threshold and less than the second threshold; the first processor 802 is further configured to:
  • the first parameter is greater than or equal to the first threshold and less than the second threshold, for two channel matrices corresponding to the downlink channel at two adjacent moments, determine a similarity between each eigenvector of K eigenvectors of one channel matrix and a corresponding eigenvector in the other channel matrix, and compare the determined similarity with the third threshold to obtain K second comparison results;
  • the first comparison result is obtained by using the K second comparison results.
  • the first communication interface 801 is further configured to send a CSI report to the network side, where the CSI report includes the second information.
  • the first processor 802 is further configured to obtain the first information from locally stored information, or to receive the first information sent by the network side through the first communication interface 801.
  • the first communication interface 801 is further configured to perform one of the following operations:
  • bus system 804. the various components in the terminal 800 are coupled together through the bus system 804. It can be understood that the bus system 804 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 804 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, various buses are marked as the bus system 804 in FIG. 8.
  • the first memory 803 in the embodiment of the present application is configured to store various types of data to support the operation of the terminal 800. Examples of such data include: any computer program for operating on the terminal 800.
  • the method disclosed in the above embodiment of the present application can be applied to the first processor 802, or implemented by the first processor 802.
  • the first processor 802 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the first processor 802 or an instruction in the form of software.
  • the first processor 802 may be a general-purpose processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • DSP Digital Signal Processor
  • the first processor 802 may implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present application.
  • a general-purpose processor may be a microprocessor or any conventional processor, etc.
  • the software module may be located in a storage medium, which is located in the first memory 803, and the first processor 802 reads the information in the first memory 803, and completes the steps of the aforementioned method in conjunction with its hardware.
  • terminal 800 can be implemented by one or more application specific integrated circuits (ASIC), DSP, programmable logic device (PLD), complex programmable logic device (CPLD), field programmable gate array (FPGA), general processor, controller, microcontroller (MCU), microprocessor, or other electronic components to execute the aforementioned method.
  • ASIC application specific integrated circuits
  • DSP digital signal processor
  • PLD programmable logic device
  • CPLD complex programmable logic device
  • FPGA field programmable gate array
  • MCU microcontroller
  • microprocessor or other electronic components to execute the aforementioned method.
  • the embodiment of the present application further provides a network device, as shown in FIG. 9 , the network device 900 includes:
  • the second communication interface 901 is capable of exchanging information with other network devices and/or terminals;
  • a second processor 902 is connected to the second communication interface 901 to implement information interaction with other network devices and/or terminals, and is configured to execute the method provided by one or more technical solutions on the network device side when running a computer program;
  • a second memory 903 on which the computer program is stored.
  • the second communication interface 901 is configured to receive second information sent by the terminal, where the second information represents the stability of the downlink channel;
  • the second processor 902 is configured to use at least the second information to determine third information, wherein the third information indicates a channel compression method of the downlink channel and a feedback mode for feeding back a channel compression result of the downlink channel to the network device;
  • the second communication interface 901 is further configured to send the third information to the terminal; receive fourth information sent by the terminal, the fourth information at least including a channel compression result, and the channel compression result is obtained by performing channel compression on the downlink channel using at least the third information.
  • the second communication interface 901 is further configured to send first information to the terminal, where the first information is used to determine the stability of the downlink channel, and the second information is determined at least using the first information.
  • the second processor 902 is further configured to determine the first information, wherein the first information includes fifth information and sixth information, the fifth information includes the number K of eigenvectors in the channel matrix of the downlink channel, K is an integer greater than 0; the sixth information includes at least one threshold, and the at least one threshold corresponds to at least two levels of stability of the downlink channel.
  • the second communication interface 901 is further configured to receive a CSI report sent by the terminal, where the CSI report includes the second information.
  • the second communication interface 901 is further configured to perform one of the following operations:
  • the first information is sent to the terminal in a unicast manner.
  • bus system 904. the various components in the network device 900 are coupled together through the bus system 904. It can be understood that the bus system 904 is used to realize the connection and communication between these components.
  • the bus system 904 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, various buses are marked as the bus system 904 in Figure 9.
  • the second memory 903 in the embodiment of the present application is configured to store various types of data to support the operation of the network device 900.
  • Examples of such data include: any computer program for operating on the network device 900.
  • the method disclosed in the above embodiment of the present application can be applied to the second processor 902, or implemented by the second processor 902.
  • the second processor 902 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware or software instructions in the second processor 902.
  • the second processor 902 may be a general-purpose processor, a DSP, or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.
  • the second processor 902 can implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present application.
  • a general-purpose processor may be a microprocessor or any conventional processor, etc.
  • the software module may be located in a storage medium, which is located in the second memory 903, and the second processor 902 reads the information in the second memory 903 and completes the steps of the above method in combination with its hardware.
  • the network device 900 may be implemented by one or more ASICs, DSPs, PLDs, CPLDs, FPGAs, general purpose processors, controllers, MCUs, Microprocessors, or other electronic components to perform the aforementioned methods.
  • the memory (first memory 803, second memory 903) of the embodiment of the present application can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories.
  • the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a ferromagnetic random access memory (FRAM), a flash memory, a magnetic table
  • the volatile memory may be a random access memory (RAM), which is used as an external cache.
  • RAM random access memory
  • SRAM static random access memory
  • SSRAM synchronous static random access memory
  • DRAM dynamic random access memory
  • SDRAM synchronous dynamic random access memory
  • DDRSDRAM double data rate synchronous dynamic random access memory
  • ESDRAM enhanced synchronous dynamic random access memory
  • SLDRAM synchronous link dynamic random access memory
  • DRRAM direct memory bus random access memory
  • the embodiment of the present application also provides an information transmission system, as shown in FIG. 10 , the system includes: a terminal 1001 and a network device 1002 .
  • the embodiment of the present application further provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, for example, including a first memory 803 storing a computer program, and the above-mentioned computer program can be executed by the first processor 802 of the terminal 800 to complete the steps described in the aforementioned terminal-side method.
  • a storage medium namely a computer storage medium, specifically a computer-readable storage medium, for example, including a first memory 803 storing a computer program, and the above-mentioned computer program can be executed by the first processor 802 of the terminal 800 to complete the steps described in the aforementioned terminal-side method.
  • the above-mentioned computer program can be executed by the second processor 902 of the network device 900 to complete the steps described in the aforementioned network device-side method.
  • the computer-readable storage medium can be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface storage, optical disk, or CD-ROM.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种信息传输方法、装置、终端、网络设备及存储介质。其中，方法包括：终端获取第一信息，所述第一信息用于确定下行信道的稳定性；至少利用所述第一信息，确定第二信息，并向网络侧发送所述第二信息，所述第二信息表征所述下行信道的稳定性；接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果，并利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。

Description

信息传输方法、装置、相关设备及存储介质
相关申请的交叉引用
本申请基于申请号为202211456381.5、申请日为2022年11月21日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请涉及无线通信领域，尤其涉及一种信息传输方法、装置、相关设备及存储介质。
背景技术
在多天线(MIMO，Multiple Input Multiple Output)系统中，获取信道状态信息(CSI，Channel State Information)是基站进行波束赋形来提升传输性能的关键条件。对于频分双工(FDD，Frequency Division Duplexing)系统，因不存在完整的上行和下行信道的互易性，基站需要借助终端反馈的方式才能获取完整的下行CSI。
相关技术中，如图1所示，终端可以使用数据驱动的深度学习方法进行信道压缩并向基站反馈信道压缩结果以使基站获得精确的下行CSI，即终端可以使用深度神经网络(即图1中的编码网络)从大量信道数据中提取低维特征进行压缩并向基站反馈信道压缩结果，基站可以使用对应的深度神经网络(即图1中的解码网络)对信道压缩结果进行恢复来得到原始信道的状态。
然而，采用基于深度学习的下行信道压缩和反馈方案时，终端按照特定周期周期性地向基站反馈信道压缩结果，可能会降低系统性能。
发明内容
为解决相关技术问题，本申请实施例提供一种信息传输方法、装置、相关设备及存储介质。
本申请实施例的技术方案是这样实现的：
本申请实施例提供一种信息传输方法，应用于终端，包括：
获取第一信息，所述第一信息用于确定下行信道的稳定性；
至少利用所述第一信息，确定第二信息，并向网络侧发送所述第二信息， 所述第二信息表征所述下行信道的稳定性；
接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；
至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果，并利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。
上述方案中，所述第一信息包含第五信息和第六信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度；所述至少利用所述第一信息，确定第二信息，包括：
至少利用所述第五信息确定第一参数，所述第一参数表征所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个特征向量之间的相似度；
将所述第一参数与所述至少一个阈值进行比较，得到第一比较结果，所述第二信息包含所述第一比较结果。
上述方案中，所述第六信息包含第一阈值和第二阈值，所述第一阈值大于0，且小于所述第二阈值；所述第二阈值小于1；所述第二阈值和1形成的区间对应所述下行信道的第一状态，所述第一阈值和第二阈值形成的区间对应所述下行信道的第二状态，所述第一阈值和0形成的区间对应所述下行信道的第三状态；所述下行信道在所述第一状态下的稳定程度高于在所述第二状态下的稳定程度，且所述下行信道在所述第二状态下的稳定程度高于在所述第三状态下的稳定程度。
上述方案中，在所述第一参数大于或等于0，且小于所述第一阈值的情况下，所述第一比较结果包括第一标识，所述第一标识表征所述第三状态；
或者，
在所述第一参数小于或等于1，且大于或等于所述第二阈值的情况下，所述第一比较结果包括第二标识，所述第二标识表征所述第一状态。
上述方案中，所述第六信息还包含第三阈值，所述第三阈值大于所述第一阈值，且小于所述第二阈值；所述将所述第一参数与所述至少一个阈值进行比较，得到第一比较结果，包括：
在所述第一参数大于或等于所述第一阈值，且小于所述第二阈值的情况下，针对所述下行信道在相邻两个时刻对应的两个信道矩阵，确定一个信道矩阵的K个特征向量中的每个特征向量与另一个信道矩阵中的对应特征向量之间的相似度，并将确定的相似度与所述第三阈值进行比较，得到K个第二比较结果；
利用所述K个第二比较结果，得到所述第一比较结果。
上述方案中，所述第一比较结果通过位图(Bitmap)序列的方式表征所述K个第二比较结果；其中，在所述Bitmap序列中的比特位是第一值的情 况下，对应的第二比较结果表征确定的相似度小于或等于所述第三阈值，且大于或等于0；在所述Bitmap序列中的比特位是第二值的情况下，对应的第二比较结果表征确定的相似度大于所述第三阈值，且小于或等于1。
上述方案中，所述反馈模式包括以下之一：
向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果，O为大于0的整数；
按照第一周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果，O为大于0的整数；
向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果后，按照第二周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩结果，O和P为大于0的整数。
上述方案中，所述第三信息包含至少一个模型的标识，所述至少一个模型用于进行所述下行信道的信道压缩。
上述方案中，所述第三信息包含第三标识，所述第三标识表征用于进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩的模型，O为大于0的整数；
或者，
所述第三信息包含第三标识和第四标识，所述第三标识表征用于进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩的模型，所述第四标识表征用于进行针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩的模型，O和P为大于0的整数。
上述方案中，所述向所述网络侧发送所述第二信息，包括：
向所述网络侧发送CSI报告，所述CSI报告包含所述第二信息。
上述方案中，所述获取第一信息，包括：
从本地存储的信息中获取所述第一信息，或者，接收所述网络侧发送的所述第一信息。
上述方案中，所述接收所述网络侧发送的所述第一信息，包括以下之一：
接收所述网络侧广播的所述第一信息；
接收所述网络侧通过多播方式发送的所述第一信息；
接收所述网络侧通过单播方式发送的所述第一信息。
本申请实施例还提供一种信息传输方法，应用于网络设备，包括：
接收终端发送的第二信息，所述第二信息表征下行信道的稳定性；
至少利用所述第二信息，确定第三信息，并向所述终端发送所述第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；
接收所述终端发送的第四信息，所述第四信息至少包含信道压缩结果， 所述信道压缩结果是至少利用所述第三信息进行所述下行信道的信道压缩得到的。
上述方案中，所述方法还包括：
向所述终端发送第一信息，所述第一信息用于确定所述下行信道的稳定性，所述第二信息是至少利用所述第一信息确定的。
上述方案中，所述方法还包括：
确定所述第一信息，所述第一信息包含第五信息和第六信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度。
上述方案中，所述第二信息包含第一比较结果，所述第一比较结果是将第一参数与所述至少一个阈值进行比较得到的，所述第一参数是至少利用所述第五信息确定的，所述第一参数表征所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个特征向量之间的相似度。
上述方案中，所述第六信息包含第一阈值和第二阈值，所述第一阈值大于0，且小于所述第二阈值；所述第二阈值小于1；所述第二阈值和1形成的区间对应所述下行信道的第一状态，所述第一阈值和第二阈值形成的区间对应所述下行信道的第二状态，所述第一阈值和0形成的区间对应所述下行信道的第三状态；所述下行信道在所述第一状态下的稳定程度高于在所述第二状态下的稳定程度，且所述下行信道在所述第二状态下的稳定程度高于在所述第三状态下的稳定程度。
上述方案中，在所述第一参数大于或等于0，且小于所述第一阈值的情况下，所述第一比较结果包括第一标识，所述第一标识表征所述第三状态；
或者，
在所述第一参数小于或等于1，且大于或等于所述第二阈值的情况下，所述第一比较结果包括第二标识，所述第二标识表征所述第一状态。
上述方案中，所述第六信息还包含第三阈值，所述第三阈值大于所述第一阈值，且小于所述第二阈值；在所述第一参数大于或等于所述第一阈值，且小于所述第二阈值的情况下，所述第一比较结果是利用K个第二比较结果得到的；所述K个第二比较结果是针对所述下行信道在相邻两个时刻对应的两个信道矩阵，确定一个信道矩阵的K个特征向量中的每个特征向量与另一个信道矩阵中的对应特征向量之间的相似度，并将确定的相似度与所述第三阈值进行比较得到的。
上述方案中，所述第一比较结果通过Bitmap序列的方式表征所述K个第二比较结果；其中，在所述Bitmap序列中的比特位是第一值的情况下，对应的第二比较结果表征确定的相似度小于或等于所述第三阈值，且大于或等于0；在所述Bitmap序列中的比特位是第二值的情况下，对应的第二比较结果表征确定的相似度大于所述第三阈值，且小于或等于1。
上述方案中，所述反馈模式包括以下之一：
所述终端向所述网络设备反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果，O为大于0的整数；
所述终端按照第一周期，周期性地向所述网络设备反馈针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果，O为大于0的整数；
所述终端向所述网络设备反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果后，按照第二周期，周期性地向所述网络设备反馈针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩结果，O和P为大于0的整数。
上述方案中，所述第三信息包含至少一个模型的标识，所述至少一个模型用于进行所述下行信道的信道压缩。
上述方案中，所述第三信息包含第三标识，所述第三标识表征用于进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩的模型，O为大于0的整数；
或者，
所述第三信息包含第三标识和第四标识，所述第三标识表征用于进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩的模型，所述第四标识表征用于进行针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩的模型，O和P为大于0的整数。
上述方案中，所述接收所述终端发送的第二信息，包括：
接收所述终端发送的CSI报告，所述CSI报告包含所述第二信息。
上述方案中，所述向所述终端发送第一信息，包括以下之一：
广播所述第一信息；
通过多播方式向所述终端发送所述第一信息；
通过单播方式向所述终端发送所述第一信息。
本申请实施例还提供一种信息传输装置，设置在终端上，包括：
获取单元，配置为获取第一信息，所述第一信息用于确定下行信道的稳定性；
第一处理单元，配置为至少利用所述第一信息，确定第二信息，所述第二信息表征所述下行信道的稳定性；
第一发送单元，配置为向网络侧发送所述第二信息；
第一接收单元，配置为接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；
第二处理单元，配置为至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果；
第二发送单元，配置为利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。
本申请实施例还提供一种信息传输装置，设置在网络设备上，包括：
第二接收单元，配置为接收终端发送的第二信息，所述第二信息表征下行信道的稳定性；
第三处理单元，配置为至少利用所述第二信息，确定第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；
第三发送单元，配置为向所述终端发送所述第三信息；
第三接收单元，配置为接收所述终端发送的第四信息，所述第四信息至少包含信道压缩结果，所述信道压缩结果是至少利用所述第三信息进行所述下行信道的信道压缩得到的。
本申请实施例还提供一种终端，包括：
第一处理器，配置为获取第一信息，所述第一信息用于确定下行信道的稳定性；至少利用所述第一信息，确定第二信息，所述第二信息表征所述下行信道的稳定性；
第一通信接口，配置为向网络侧发送所述第二信息；接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；其中，
所述第一处理器，还配置为至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果；
所述第一通信接口，还配置为利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。
本申请实施例还提供一种网络设备，包括：
第二通信接口，配置为接收终端发送的第二信息，所述第二信息表征下行信道的稳定性；
第二处理器，配置为至少利用所述第二信息，确定第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；其中，
所述第二通信接口，还配置为向所述终端发送所述第三信息；接收所述终端发送的第四信息，所述第四信息至少包含信道压缩结果，所述信道压缩结果是至少利用所述第三信息进行所述下行信道的信道压缩得到的。
本申请实施例还提供一种终端，包括：第一处理器和配置为存储能够在处理器上运行的计算机程序的第一存储器，
其中，所述第一处理器配置为运行所述计算机程序时，执行上述终端侧任一方法的步骤。
本申请实施例还提供一种网络设备，包括：第二处理器和配置为存储能 够在处理器上运行的计算机程序的第二存储器，
其中，所述第二处理器配置为运行所述计算机程序时，执行上述网络设备侧任一方法的步骤。
本申请实施例还提供一种存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现上述终端侧任一方法的步骤，或者实现上述网络设备侧任一方法的步骤。
本申请实施例提供的信息传输方法、装置、相关设备及存储介质，终端获取第一信息，所述第一信息用于确定下行信道的稳定性；获取到所述第一信息后，所述终端至少利用所述第一信息，确定第二信息，并向网络设备发送所述第二信息，所述第二信息表征所述下行信道的稳定性；接收到所述第二信息后，所述网络设备至少利用所述第二信息，确定第三信息，并向所述终端发送所述第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；接收到所述第三信息后，所述终端至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果，并利用所述第三信息向所述网络设备发送第四信息，所述第四信息至少包含所述信道压缩结果。本申请实施例提供的方案，由于终端进行下行信道压缩的信道压缩方式及终端向网络侧反馈下行信道的信道压缩结果的反馈模式是网络侧根据下行信道的稳定性确定的，所以网络侧能够根据下行信道的稳定性，灵活地配置终端进行下行信道压缩的信道压缩方式及终端向网络侧反馈下行信道的信道压缩结果的反馈模式，从而能够提高系统性能。
附图说明
图1为相关技术中基于深度学习的信道压缩和反馈流程示意图；
图2为本申请实施例一种信息传输方法的流程示意图；
图3为本申请实施例另一种信息传输方法的流程示意图；
图4为本申请实施例第三种信息传输方法的流程示意图；
图5为本申请应用示例针对下行信道特征向量的压缩与反馈流程示意图；
图6为本申请实施例一种信息传输装置结构示意图；
图7为本申请实施例另一种信息传输装置结构示意图；
图8为本申请实施例终端结构示意图；
图9为本申请实施例网络设备结构示意图；
图10为本申请实施例信息传输系统结构示意图。
具体实施方式
下面结合附图及实施例对本申请再作进一步详细的描述。
在新空口(NR，New Radio)系统中，终端可以基于码本向基站反馈CSI，终端通常可以支持CSI类型(type)I、type II、type II增强(enhanced)等码本种类，用于反馈秩指示(RI，Rank Indicator)、预编码矩阵指示(PMI，Precoding matrix indicator)、信道质量指示(CQI，Channel Quality Indicator)等CSI。
然而，相关技术中使用的码本主要基于离散傅里叶变换(DFT，Discrete Fourier Transform)矩阵实现，其特点是视信道为各向同向性的随机分布，进而使用DFT矢量对波束空间进行均匀划分。但实际系统中的信道不会严格遵循各向同向性的分布规律，而是会集中于某个子空间中，DFT码本由于无法有效利用信道所在的子空间信息而性能受限；换句话说，终端基于码本向基站反馈的CSI可能无法精确地反映实际的信道状态。
图1所示的基于深度学习的下行信道压缩和反馈方案可以针对具体场景进行优化，即针对信道实际的分布规律进行优化，从而能够在同样反馈开销下提高信道反馈精度；换句话说，与基于码本的CSI反馈方案相比，终端采用基于深度学习的下行信道压缩和反馈方案能够使基站获得精确的下行CSI。
然而，图1所示的基于深度学习的下行信道压缩和反馈方案未考虑当前信道环境的稳定性，即无论下行信道是否剧烈变化，终端均持续地按照特定周期向网络侧反馈下行信道的信道压缩结果，终端的反馈开销较大，可能会降低系统性能。
基于此，在本申请的各种实施例中，网络侧根据下行信道的稳定性，灵活地配置终端进行下行信道压缩的信道压缩方式及终端向网络侧反馈下行信道的信道压缩结果的反馈模式，从而能够提高系统性能。
本申请实施例提供一种信息传输方法，应用于终端，如图2所示，该方法包括：
步骤201：获取第一信息，所述第一信息用于确定下行信道的稳定性；
步骤202：至少利用所述第一信息，确定第二信息，并向网络侧发送所述第二信息，所述第二信息表征所述下行信道的稳定性；
步骤203：接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；
步骤204：至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果，并利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。
实际应用时，所述终端也可以称为用户设备(UE，User Equipment)，还可以称为用户。
实际应用时，所述下行信道的稳定性，也可以理解为所述下行信道的变化程度或者所述下行信道变化的剧烈程度。所述下行信道的稳定性的具体 表现形式可以根据需求来设置，比如所述下行信道的特征向量在相邻时刻的相关性变化程度(即所述下行信道的特征向量在相邻时刻的相似度)等，本申请实施例对此不作限定。
实际应用时，在所述下行信道的稳定性的表现形式包括所述下行信道的特征向量在相邻时刻的相关性变化程度的情况下，所述第一信息可以指示所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；相应地，所述终端可以基于对CSI参考信号(CSI-RS)的测量所得到的所述下行信道的信道矩阵和K，确定所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个特征向量之间的相似度，即确定所述下行信道的稳定性。
基于此，在一实施例中，所述第一信息可以包含第五信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；相应地，步骤202的一种具体实现可以包括：
至少利用所述第五信息确定第一参数，所述第一参数表征所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个特征向量之间的相似度；所述第二信息包含所述第一参数。
这里，需要说明的是，在本申请的各种实施例中，特征向量可以指奇异向量；相应地，特征向量对应的特征值，可以指奇异向量对应的奇异值。
实际应用时，所述获取第一信息，可以包括：获取所述第五信息。具体地，所述终端可以从本地存储的信息中获取所述第五信息；或者，所述终端可以从所述网络侧获取所述第五信息，即所述网络侧可以向所述终端发送所述第五信息，所述终端可以接收所述网络侧发送的所述第五信息。所述终端获取所述第五信息的具体方式可以根据需求来设置，本申请实施例对此不作限定。
实际应用时，在由所述网络侧向所述终端发送所述第五信息的情况下，所述网络侧的网络设备(具体可以包括基站)可以根据当前用户的业务需求(比如CSI精度需求等)、信道特征、反馈开销等一种或多种因素决策K的大小，即确定所述第五信息，比如用户对CSI精度需求越高，K的值越大；所述网络设备确定所述第五信息的具体方式可以根据需求来设置，本申请实施例对此不作限定。另外，可以理解，K需要小于或等于所述下行信道的信道矩阵中的特征向量总数。
实际应用时，可以理解，除所述第五信息外，还可以存在其他数据参与所述第一参数的计算，比如对CSI-RS进行测量所得到的所述下行信道的信道矩阵等，本申请实施例对计算所述第一参数的具体方式不作限定，只要实现其功能即可。示例性地，所述终端可以选择所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个左奇异向量或者K个右奇异向量计算所述第一参数，并且，所述终端可以基于计算余弦相似度(英文可以表达为Cosine Similarity)的方式计算所述第一参数。
其中，假设所述终端在t时刻对CSI-RS进行测量所得到的所述下行信 道的二维信道矩阵为H(t)，对H(t)进行奇异值分解可以表示为H(t)＝U(t)∑(t)VT(t)，∑(t)中包含的奇异值按照绝对值从大到小排序，可以记作需要从∑(t)中选取最大K个奇异值(0≤K≤min(LM-1,LM))时，其对应的左奇异向量可以表示为U1(t),U2(t)…,UK(t)，其对应的右奇异向量可以表示为换句话说，t时刻所述下行信道的信道矩阵H(t)中的K个特征向量可以包括U1(t),U2(t)…,UK(t)或者由于余弦相似度是j维空间中两个j维向量之间角度的余弦(j为大于0的整数)，它等于两个向量的点积(即向量积)除以两个向量长度(或大小)的乘积，所以针对所述下行信道在相邻两个时刻(记作t1和t2)对应的两个信道矩阵H(t1)和H(t2)，H(t1)中第i个特征向量与H(t2)中第i个特征向量之间的余弦相似度(i为大于0的整数)可以选择左奇异向量计算(即公式(1))或者选择右奇异向量计算(即公式(2))：

基于公式(1)或公式(2)，所述第一参数可以通过以下公式计算：
实际应用时，在所述第二信息包含所述第一参数的情况下，所述网络设备接收到所述第二信息后，至少可以利用所述第一参数灵活地为所述终端配置所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式。示例性地，可以在所述网络设备上预先设置至少一个阈值(阈值的具体数量可以根据需求(比如所述下行信道的稳定程度的划分需求)来设置，本申请实施例对此不作限定)，所述至少一个阈值可以形成至少两个区间，每个区间可以对应所述下行信道的一种稳定程度，即所述至少一个阈值可以对应所述下行信道的至少两种稳定程度；每种稳定程度可以对应一种信道压缩方式以及信道压缩结果的反馈模式，所述网络设备可以通过将所述第一参数与预先设置的至少一个阈值进行比较来确定所述下行信道的稳定程度，从而根据确定的稳定程度确定需要为所述终端配置的信道压缩方式以及信道压缩结果的反馈模式。
其中，所述至少一个阈值可以包含第一阈值和第二阈值，所述第一阈值 大于0，且小于所述第二阈值，所述第二阈值小于1；所述第二阈值和1形成的区间(即(第二阈值，1]或[第二阈值，1]或[第二阈值，1)或(第二阈值，1))可以对应所述下行信道的第一状态，所述第一阈值和第二阈值形成的区间(即(第一阈值，第二阈值]或[第一阈值，第二阈值]或[第一阈值，第二阈值)或(第一阈值，第二阈值))可以对应所述下行信道的第二状态，所述第一阈值和0形成的区间(即(0，第一阈值]或[0，第一阈值]或[0，第一阈值)或(0，第一阈值))可以对应所述下行信道的第三状态；所述下行信道在所述第一状态下的稳定程度高于在所述第二状态下的稳定程度，且所述下行信道在所述第二状态下的稳定程度高于在所述第三状态下的稳定程度。
这里，所述下行信道的稳定程度划分方式、每种稳定程度所对应的信道压缩方式以及信道压缩结果的反馈模式可以根据需求来设置，本申请实施例对此不作限定。示例性地，在所述下行信道处于所述第一状态的情况下，说明所述下行信道较为稳定，所述网络设备可以配置所述终端不再周期性地向所述网络侧反馈信道压缩结果，比如反馈一次信道压缩结果或者不反馈信道压缩结果；即所述第一状态对应的反馈模式可以包括：所述终端向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果，O为大于0的整数，且O小于或等于所述下行信道的信道矩阵中的特征向量总数。
在所述下行信道处于所述第三状态的情况下，说明所述下行信道剧烈变化，所述网络设备可以配置所述终端按照特定周期向所述网络侧反馈信道压缩结果；即所述第三状态对应的反馈模式可以包括：所述终端按照第一周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果。
在所述下行信道处于所述第二状态的情况下，说明所述下行信道存在变化但变化较小，所述网络设备可以配置所述终端增量反馈信道压缩结果，即所述第二状态对应的反馈模式可以包括：所述终端向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果后，按照第二周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩结果，P为大于0的整数，且P小于或等于O。
从上面的描述可以看出，所述网络设备能够根据所述下行信道的稳定性，灵活地配置所述终端向所述网络侧反馈信道压缩结果的反馈模式，比如所述第一状态、第二状态和第三状态对应的反馈模式；如此，所述终端不需要再持续地按照特定周期向所述网络侧反馈信道压缩结果，而是可以采用随下行信道的稳定性变化而灵活变化的反馈模式向所述网络侧反馈信道压缩结果，从而能够降低终端的反馈开销，即有效降低无线传输开销，进而能够提高系统性能。
实际应用时，所述第一周期和第二周期的大小可以相同或不同，所述第 一周期和第二周期的大小具体可以根据需求来设置，本申请实施例对此不作限定。
实际应用时，O的大小可以根据需求来设置，本申请实施例对O的大小及其决策方式不作限定。另外，O与K的大小可以相同或不同，即O可以等于或大于或小于K，具体可以根据需求来设置，本申请实施例对O与K的大小关系也不作限定。
实际应用时，所述第一条件可以包括第三阈值，所述第三阈值大于所述第一阈值，且小于所述第二阈值。针对所述下行信道在相邻两个时刻对应的两个信道矩阵，在一个信道矩阵的K个特征向量中的第i个特征向量与另一个信道矩阵中的对应特征向量(即另一个信道矩阵的K个特征向量中的第i个特征向量)之间的相似度小于或等于所述第三阈值，且大于或等于0的情况下，可以确定该特征向量(即第i个特征向量)的变化情况满足所述第一条件；在一个信道矩阵的K个特征向量中的第i个特征向量与另一个信道矩阵中的对应特征向量之间的相似度大于所述第三阈值，且小于或等于1的情况下，可以确定该特征向量的变化情况不满足所述第一条件。其中，针对所述下行信道在相邻两个时刻对应的两个信道矩阵，一个信道矩阵的K个特征向量中第i个特征向量与另一个信道矩阵中的对应特征向量之间的相似度可以利用公式(1)或公式(2)计算。
实际应用时，所述第一阈值、第二阈值和第三阈值的大小具体可以根据需求来设置，本申请实施例对此不作限定。示例性地，所述第一阈值可以等于0.2，所述第二阈值可以等于0.8，所述第三阈值可以等于0.7。
实际应用时，可以将所述第一阈值和第二阈值预先设置在所述终端上，或者，所述网络设备还可以将所述第一阈值和第二阈值发送给所述终端。所述终端确定所述第一参数后，可以直接将所述第一参数与所述第一阈值和第二阈值进行比较的比较结果作为所述第二信息反馈给所述网络设备。
基于此，在一实施例中，所述第一信息还可以包含第六信息(即所述第一信息包含第五信息和第六信息)，所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度；相应地，步骤202的另一种具体实现可以包括：
至少利用所述第五信息确定第一参数，所述第一参数表征所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个特征向量之间的相似度；
将所述第一参数与所述至少一个阈值进行比较，得到第一比较结果，所述第二信息包含所述第一比较结果。
其中，所述获取第一信息，还可以包括：获取所述第六信息。具体地，所述终端可以从本地存储的信息中获取所述第六信息；或者，所述终端可以从所述网络侧获取所述第六信息，即所述网络侧可以向所述终端发送所述第六信息，所述终端可以接收所述网络侧发送的所述第六信息。所述终端获取所述第六信息的具体方式可以根据需求来设置，本申请实施例对此不作 限定。
从上面的描述可以看出，实际应用时，所述第六信息可以包含所述第一阈值和第二阈值。在所述第一参数大于或等于0，且小于所述第一阈值的情况下，所述第一比较结果可以包括第一标识，所述第一标识表征所述第三状态；在所述第一参数小于或等于1，且大于或等于所述第二阈值的情况下，所述第一比较结果可以包括第二标识，所述第二标识表征所述第一状态。相应地，所述网络设备接收到所述第二信息后，可以根据所述第一标识，配置所述终端按照所述第一周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果；或者，可以根据所述第二标识，配置所述终端向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果。换句话说，在所述第一比较结果包括所述第一标识的情况下，所述第三信息指示的反馈模式可以包括：按照所述第一周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果；在所述第一比较结果包括所述第二标识的情况下，所述第三信息指示的反馈模式可以包括：向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果。
实际应用时，所述第一标识和第二标识的表现形式可以根据需求来设置，本申请实施例对此不作限定，示例性地，所述第一标识可以记作-1，所述第二标识可以记作-2。
实际应用时，在所述第一参数大于或等于所述第一阈值，且小于所述第二阈值的情况下，所述下行信道处于所述第二状态，所述网络设备可以配置所述终端增量反馈信道压缩结果；此时，可以将所述第三阈值预先设置在所述终端上，或者，所述网络设备可以将所述第三阈值发送给所述终端，以供所述终端判断特征向量的变化情况是否满足所述第一条件。
基于此，在一实施例中，所述第六信息还可以包含所述第三阈值；所述将所述第一参数与所述至少一个阈值进行比较，得到第一比较结果，可以包括：
在所述第一参数大于或等于所述第一阈值，且小于所述第二阈值的情况下，针对所述下行信道在相邻两个时刻对应的两个信道矩阵，确定一个信道矩阵的K个特征向量中的每个特征向量与另一个信道矩阵中的对应特征向量之间的相似度，并将确定的相似度与所述第三阈值进行比较，得到K个第二比较结果；
利用所述K个第二比较结果，得到所述第一比较结果。
实际应用时，针对所述下行信道在相邻两个时刻对应的两个信道矩阵，可以利用公式(1)或公式(2)计算一个信道矩阵的K个特征向量中的每个特征向量与另一个信道矩阵中的对应特征向量之间的相似度。
实际应用时，所述第一比较结果可以通过Bitmap序列的方式(即所述第一比较结果可以包括K维的Bitmap序列)表征所述K个第二比较结果； 其中，在所述Bitmap序列中的比特位是第一值的情况下，对应的第二比较结果表征确定的相似度小于或等于所述第三阈值，且大于或等于0，即对应的特征向量的变化情况满足所述第一条件；在所述Bitmap序列中的比特位是第二值的情况下，对应的第二比较结果表征确定的相似度大于所述第三阈值，且小于或等于1，即对应的特征向量的变化情况不满足所述第一条件。
实际应用时，所述第一值可以等于0，所述第二值可以等于1；或者，所述第一值可以等于1，所述第二值可以等于0；所述第一值和所述第二值的大小具体可以根据需求来设置，本申请实施例对此不作限定。
实际应用时，可以理解，在所述第一比较结果包括K维的Bitmap序列的情况下，所述下行信道处于所述第二状态。所述网络设备接收到所述第二信息后，可以根据所述Bitmap序列，配置所述终端向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果后，按照所述第二周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵的O个特征向量中变化情况满足所述第一条件的P个特征向量的信道压缩结果，即周期性地向所述网络侧反馈在所述Bitmap序列中对应的比特位是所述第一值的P个特征向量的信道压缩结果。换句话说，在所述第一比较结果包括K维的Bitmap序列的情况下，所述第三信息指示的反馈模式可以包括：向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果后，按照所述第二周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩结果(即周期性地向所述网络侧反馈在所述Bitmap序列中对应的比特位是所述第一值的P个特征向量的信道压缩结果)。
从上面的描述可以看出，所述第三信息指示的反馈模式可以包括以下之一：
所述第一状态所对应的单次传递压缩信道特征向量的模式(后续描述中可以记作第一反馈模式)，即向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果；
所述第二状态所对应的周期更新所述Bitmap序列中所述第一值对应的特征向量的模式(该模式可以理解为一种增量反馈模型，后续描述中可以记作第二反馈模式)，即向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果后，按照所述第二周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量(即在所述Bitmap序列中对应的比特位是所述第一值的特征向量)的信道压缩结果；
所述第三状态所对应的周期传递压缩信道特征向量的模式(后续描述中可以记作第三反馈模式)，即按照所述第一周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果。
在一实施例中，所述第三信息可以包含至少一个模型的标识，所述至少 一个模型用于进行所述下行信道的信道压缩；换句话说，所述第三信息所指示的所述下行信道的信道压缩方式包括：利用所述第三信息包含的至少一个模型的标识所对应的至少一个模型，进行所述下行信道的信道压缩。
实际应用时，所述网络侧可以预先采集下行信道估计数据，通过奇异值分解对下行信道估计数据进行预处理，以预处理后的奇异向量及其他相关信息(比如CSI精度需求、特征向量数等)为输入，离线训练基于人工智能(AI)自编码器(autoencoder)的针对信道特征向量的信道压缩模型(压缩模型的编号(即标识)可以记作E1,E2…,EN)和信道解压模型(解压模型的编号(即标识)可以记作D1,D2…,DN)。
其中，所述信道压缩模型和信道解压模型是一一对应的，即利用一个信道压缩模型得到的信道压缩结果需要利用对应的一个信道解压模型进行信道恢复。另外，N可以为大于0的整数，N的大小可以根据需求(比如CSI精度需求、特征向量数等)来设置，即不同的信道压缩模型/信道解压模型可以对应不同的CSI精度需求、不同的特征向量数等；换句话说，不同的信道压缩模型/信道解压模型的功能不同。
实际应用时，所述网络侧训练好所述信道压缩模型和信道解压模型后，可以将所述信道压缩模型及对应编号(即标识)同步到终端(比如在出厂前配置到终端本地)，并将所述信道解压模型及对应编号(即标识)同步到网络设备(比如预先配置到基站本地)。同时，所述网络侧可以在所述终端和所述网络设备上设置模型编号与模型功能(可以通过CSI精度、特征向量数等参数体现模型的功能)之间的关联关系。
实际应用时，所述网络侧可以部署AI模型训练系统，由所述AI模型训练系统预先采集下行信道估计数据，通过奇异值分解对下行信道估计数据进行预处理，以预处理后的奇异向量及其他相关信息(比如CSI精度需求、特征向量数等)为输入，离线训练基于AI autoencoder的针对信道特征向量的信道压缩模型和信道解压模型。其中，所述AI模型训练系统的具体部署方式可以根据需求来设置，本申请实施例对此不作限定。示例性地，所述AI模型训练系统可以部署在一个基站的集中单元(CU)和/或分布单元(DU)上，即由一个基站进行所述信道压缩模型和信道解压模型的训练；或者，所述AI模型训练系统可以跨CU和/或DU部署，即部署在多个基站的CU和/或DU上，换句话说，由至少两个基站协同训练所述信道压缩模型和信道解压模型。
实际应用时，在所述终端从所述网络侧获取所述第一信息的情况下，向所述终端发送所述第一信息的网络设备可以参与所述信道压缩模型和信道解压模型的训练，也可以不参与所述信道压缩模型和信道解压模型的训练，具体可以根据需求(比如网络部署需求等)来设置，本申请实施例对此不作限定。所述网络设备上可以预先设置有至少一个阈值(即所述第六信息)，所述终端通过所述网络设备接入所述网络侧后，该网络设备可以根据所述 终端的业务需求(比如CSI精度需求等)、信道特征、反馈开销等一种或多种因素决策K的大小(即所述第五信息)，并将K和预先设置的至少一个阈值发送给所述终端，即向所述终端发送所述第一信息。
实际应用时，在所述网络设备至少利用所述第二信息确定所述第三信息的过程中，所述网络设备可以先利用所述第一比较结果，确定所述终端向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式(所述第一反馈模式或第二反馈模式或第三反馈模式)，再基于确定的反馈模式、特征向量数O以及预先设置的模型编号(即标识)与模型功能(可以通过特征向量数等参数体现模型的功能)之间的关联关系，确定所述下行信道的信道压缩方式，即确定用于进行所述下行信道的信道压缩的至少一个模型。
其中，在所述第三信息指示的反馈模式包括所述第一反馈模式的情况下，所述第三信息可以包含第三标识，所述第三标识表征用于进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩的模型；在所述第三信息指示的反馈模式包括所述第二反馈模式的情况下，所述第三信息可以包含第三标识、第四标识和第二周期，所述第四标识表征用于进行针对所述下行信道的信道矩阵的O个特征向量中变化情况满足所述第一条件的P个特征向量的信道压缩的模型；在所述第三信息指示的反馈模式包括所述第三反馈模式的情况下，所述第三信息可以包含所述第三标识和第一周期。
实际应用时，所述终端在接收到所述第三信息之后，在所述第三信息包含所述第三标识的情况下，所述终端可以利用所述第三标识对应的模型，进行一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩，得到一个包含O个特征值的第一信道压缩结果，并向所述网络侧发送一次至少包含所述第一信道压缩结果的第四信息。在所述第三信息包含所述第三标识、第四标识和第二周期的情况下，首先，所述终端可以利用所述第三标识对应的模型，进行一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩，得到一个包含O个特征值的第二信道压缩结果，并向所述网络侧发送一次至少包含所述第二信道压缩结果的第四信息；完成信道压缩结果的首次反馈之后，所述终端可以按照所述第二周期，周期性地利用所述第四标识对应的模型进行针对所述下行信道的信道矩阵的O个特征向量中变化情况满足所述第一条件的P个特征向量的信道压缩，得到包含P个特征值的第三信道压缩结果，并周期性地向所述网络侧发送至少包含所述第三信道压缩结果的第四信息。在所述第三信息包含所述第三标识和第一周期的情况下，所述终端可以按照所述第一周期，周期性地利用所述第三标识对应的模型进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩，得到包含O个特征值的第四信道压缩结果，并周期性地向所述网络侧发送至少包含所述第四信道压缩结果的第四信息。
从上面的描述可以看出，实际应用时，所述第一信息可以包含所述第五信息和第六信息；所述终端可以从本地存储的信息中获取所述第五信息，或 者，所述终端可以从所述网络侧获取所述第五信息；同时，所述终端可以从本地存储的信息中获取所述第六信息，或者，所述终端可以从所述网络侧获取所述第六信息。
基于此，在一实施例中，所述获取第一信息，可以包括：
从本地存储的信息中获取所述第一信息，或者，接收所述网络侧发送的所述第一信息。
具体地，在一实施例中，所述获取第一信息，可以包括以下之一：
从本地存储的信息中获取所述第五信息和第六信息；
接收所述网络侧发送的所述第五信息和第六信息；
从本地存储的信息中获取所述第五信息，并接收所述网络侧发送的所述第六信息；
从本地存储的信息中获取所述第六信息，并接收所述网络侧发送的所述第五信息。
实际应用时，在所述终端从所述网络侧获取所述第一信息的情况下，所述网络设备可以广播所述第一信息，或者可以通过多播或单播方式向所述终端发送所述第一信息。
基于此，在一实施例中，所述接收网络侧发送的第一信息，可以包括以下之一：
接收所述网络侧广播的所述第一信息；
接收所述网络侧通过多播方式发送的所述第一信息；
接收所述网络侧通过单播方式发送的所述第一信息。
实际应用时，所述终端可以通过CSI报告(英文可以表达为CSI-ReportConfig)向所述网络设备发送所述第二信息。
基于此，在一实施例中，所述向所述网络侧发送所述第二信息，可以包括：
向所述网络侧发送CSI报告，所述CSI报告包含所述第二信息。
实际应用时，所述网络设备可以通过RRC信令向所述终端发送所述第三信息。相应地，所述接收所述网络侧发送的第三信息，可以包括：接收所述网络侧发送的RRC信令，所述RRC信令包含所述第三信息。
实际应用时，所述终端向所述网络侧发送所述第四信息的形式可以根据需求来设置，比如通过CSI报告向所述网络侧发送所述第四信息；可以理解，所述第四信息除所述信道压缩结果外，还可以包含所述终端需要向所述网络侧反馈的其他信息(比如CSI报告可以携带的任一信息)。
实际应用时，所述终端可以按照第三周期，周期性地确定所述第二信息，并周期性地向所述网络侧发送所述第二信息；所述网络侧在第一次确定并向所述终端发送所述第三信息之后，可以周期性地利用所述终端发送的第二信息更新第三信息，并周期性地向所述终端发送更新后的第三信息；所述终端可以利用更新后的第三信息得到和反馈信道压缩结果。其中，所述第三 周期的大小可以根据需求来设置，比如可以设置所述第三周期与CSI-RS的发送周期一致等，本申请实施例对此不作限定。另外，所述第三周期可以预先设置在所述终端上，或者，所述网络设备可以按照特定方式(比如RRC信令等，本申请实施例对第三周期的配置方式不作限定)将所述第三周期配置给所述终端。
相应地，本申请实施例还提供了一种信息传输方法，应用于网络设备(具体可以包括基站)，如图3所示，该方法包括：
步骤301：接收终端发送的第二信息，所述第二信息表征下行信道的稳定性；
步骤302：至少利用所述第二信息，确定第三信息，并向所述终端发送所述第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；
步骤303：接收所述终端发送的第四信息，所述第四信息至少包含信道压缩结果，所述信道压缩结果是至少利用所述第三信息进行所述下行信道的信道压缩得到的。
其中，在一实施例中，该方法还可以包括：
向所述终端发送第一信息，所述第一信息用于确定所述下行信道的稳定性，所述第二信息是至少利用所述第一信息确定的。
在一实施例中，该方法还可以包括：
确定所述第一信息，所述第一信息包含第五信息和第六信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度。
在一实施例中，所述接收所述终端发送的第二信息，可以包括：
接收所述终端发送的CSI报告，所述CSI报告包含所述第二信息。
在一实施例中，所述向终端发送第一信息，可以包括以下之一：
广播所述第一信息；
通过多播方式向所述终端发送所述第一信息；
通过单播方式向所述终端发送所述第一信息。
相应地，本申请实施例还提供了一种信息传输方法，如图4所示，该方法包括：
步骤401：终端获取第一信息，至少利用所述第一信息，确定第二信息，并向网络设备发送所述第二信息，所述第一信息用于确定下行信道的稳定性，所述第二信息表征所述下行信道的稳定性；
步骤402：所述网络设备接收所述终端发送的所述第二信息，至少利用所述第二信息，确定第三信息，并向所述终端发送所述第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；
步骤403：所述终端接收所述网络设备发送的所述第三信息，至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果，并利用所述第三信息向所述网络设备发送第四信息，所述第四信息至少包含所述信道压缩结果；
步骤404：所述网络设备接收所述终端发送的所述第四信息。
本申请实施例提供的信息传输方法，终端获取第一信息，所述第一信息用于确定下行信道的稳定性；获取到所述第一信息后，所述终端至少利用所述第一信息，确定第二信息，并向网络设备发送所述第二信息，所述第二信息表征所述下行信道的稳定性；接收到所述第二信息后，所述网络设备至少利用所述第二信息，确定第三信息，并向所述终端发送所述第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；接收到所述第三信息后，所述终端至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果，并利用所述第三信息向所述网络设备发送第四信息，所述第四信息至少包含所述信道压缩结果。本申请实施例提供的方案，由于终端进行下行信道压缩的信道压缩方式及终端向网络侧反馈下行信道的信道压缩结果的反馈模式是网络侧根据下行信道的稳定性确定的，所以网络侧能够根据下行信道的稳定性，灵活地配置终端进行下行信道压缩的信道压缩方式及终端向网络侧反馈下行信道的信道压缩结果的反馈模式，比如所述第一状态、第二状态和第三状态对应的反馈模式；如此，所述终端不需要再持续地按照特定周期向所述网络侧反馈信道压缩结果，而是可以采用随下行信道的稳定性变化而灵活变化的反馈模式向所述网络侧反馈信道压缩结果，从而能够降低终端的反馈开销，即有效降低无线传输开销，进而能够提高系统性能。
下面结合应用示例对本申请再作进一步详细的描述。
在本应用示例中，所述网络设备包括基站；所述第一阈值、第二阈值和第三阈值称为相关性阈值或变化程度阈值，所述第一阈值表示为δ1,K，所述第二阈值表示为δ2,K，所述第三阈值表示为δ3,K；O等于K。基站决策信道特征向量数K和相关性阈值δ1,K、δ2,K和δ3,K，终端基于K、δ1,K、δ2,K和δ3,K(即上述第一信息)确定下行信道相关性δ(即上述第一参数)并向基站反馈下行信道相关性与相关性阈值的比较结果Td(即上述第二信息)；基站根据终端反馈的信道相关性与相关性阈值的比较结果决策下行信道压缩模型及反馈模式(即上述第三信息)，终端基于基站的决策进行信道压缩并反馈信道压缩结果。
具体地，在本应用示例中，如图5所示，针对下行信道特征向量的压缩与反馈流程可以包括以下步骤：
步骤501：网络侧采集数据，离线训练基于AI的信道压缩/解压模型，同步至基站和终端，并预置阈值参数(即所述相关性阈值)，之后执行步骤502；
步骤502：终端接入网络，进入RRC连接态(RRC-CONNECTED)，之后执行步骤503；
步骤503：基站决策信道压缩向量数，之后执行步骤504；
步骤504：基站向终端发送信道相关性阈值(即上述第六信息)及特征向量数(即上述第五信息)，并按照配置发送CSI-RS，之后执行步骤505；
步骤505：终端计算下行信道相关性(即上述第一参数)，之后执行步骤506；
步骤506：终端通过CSI-ReportConfig反馈下行信道相关性与阈值的比较结果(即上述第二信息)，之后执行步骤507；
步骤507：基站决策压缩模型编号及其他参数(即上述第三信息)，之后执行步骤508；
步骤508：基站向终端发送模型编号及其他参数，之后执行步骤509；
步骤509：终端按照配置参数中的模型编号压缩信道信息，之后执行步骤510；
步骤510：终端按照配置反馈压缩后的信道信息(即上述至少包含信道压缩结果的第四信息)，之后执行步骤511；
步骤511：基站解压信道信息，之后执行步骤512；
步骤512：根据终端CSI-RS信道相关性，基站更新发送模型编号及其他参数，终端调整模型并上报压缩信道信息。
其中，步骤501中，无线接入网(即上述网络侧)中可以部署AI模型训练系统，所述AI模型训练系统可以部署在基站的CU和/或DU，或者跨CU和/或DU的逻辑实体上。所述AI模型训练系统可以预先采集下行信道估计数据，通过奇异值分解对下行信道估计数据进行预处理，以预处理后的奇异向量及其他相关信息为输入，离线训练基于AI autoencoder的针对信道特征向量的信道压缩模型(压缩模型的编号(即标识)可以记作E1,E2…,EN)和信道解压模型(解压模型的编号(即标识)可以记作D1,D2…,DN)。所述AI模型训练系统可以将训练好的信道压缩模型及对应编号同步至终端，并将训练好的信道解压模型及编号同步至基站。基站可以根据信道变化程度的划分需求预先设置能够体现信道变化程度的相关性阈值δ1,K、δ2,K和δ3,K，其中0<δ1,K<δ3,K<δ2,K<1；示例性地，基站可以设置δ1,K＝0.2,δ2,K＝0.8,δ3,K＝0.7。
实际应用时，基站可以记录终端反馈的下行信道相关性与相关性阈值的比较结果Td；步骤501中，基站可以将Td初始置(即初始化)为-3。
步骤503中，基站可以根据当前用户的业务需求(比如CSI精度需求等)、信道特征、反馈开销等一种或多种因素决策该终端需要压缩的信道特征向量数量K的大小。
实际应用时，基站也可以在步骤501中根据信道变化程度的划分需求预先设置对应不同特征向量数K的相同或不同的相关性阈值。
步骤504中，基站可以通过广播或单播或多播的方式将K、δ1,K、δ2,K和δ3,K发送给终端，并按照配置发送CSI-RS。
步骤505中，终端可以根据对CSI-RS的测量计算下行信道相关性。这里，所述信道相关性可以通过计算信道特征向量的相关性(即相似度)得到，信道特征向量相关性的计算可以选择左奇异向量或右奇异向量，并且，计算信道相关性的方法可以包括但不限于余弦相似度的方法。具体地，采用计算余弦相似度的方式计算信道相关性时，终端可以利用公式(3)计算下行信道的变化程度δ(即所述下行信道相关性)，将下行信道的变化程度δ与相关性阈值δ1,K、δ2,K和δ3,K进行比较，并记录比较结果Td；若δ∈[0,δ1,K)，则将Td记为-1(即上述第一标识)；若δ∈[δ2,K,1]，则将Td记为-2(即上述第二标识)；若δ∈[δ1,K,δ2,K)，则利用公式(1)或公式(2)计算K个特征向量中每个特征向量的相似度，得到δi(i＝1,2,…K)，并将δi与δ3,K进行比较，若δi∈[0,δ3,K]，则记为0(即上述第一值)，否则记为1(即上述第二值)，将δi顺次排列，得到一个K维的Bitmap序列，即将Td记为一个K维的Bitmap序列。示例性地，在δ1,K＝0.2,δ2,K＝0.8,δ3,K＝0.7的情况下，若δ∈[0,0.2)，则将Td记为-1；若δ∈[0.8,1]，则将Td记为-2；若δ∈[0.2,0.8)，则利用公式(1)或公式(2)计算K个特征向量中每个特征向量的相似度，得到δi(i＝1,2,…K)，并将δi与δ3,K＝0.7进行比较，若δi∈[0,0.7]，则记为0，否则记为1，将δi顺次排列，将Td记为一个K维的Bitmap序列。
步骤506中，终端可以将比较结果Td通过CSI-ReportConfig反馈给基站。
步骤507中，若Td为-1，基站可以选择一个可压缩K个信道特征向量的模型编号(即上述第三标识)并决策特征向量反馈周期T1(即上述第一周期)；若Td为-2，基站可以选择一个可压缩K个信道特征向量的模型编号(即上述第三标识)，该编号对应的压缩模型用于信道特征向量的首次压缩反馈，即仅反馈一次压缩后的信道信息(即上述至少包含信道压缩结果的第四信息)；当Td为K维Bitmap序列时，基站可以决策反馈周期T2(即上述第二周期)并选择两个压缩模型，第一个压缩模型(即上述第三标识对应的模型)用于首次压缩K个信道特征向量并反馈，第二个压缩模型(即上述第四标识对应的模型)用于完成首次压缩反馈后以周期T2压缩Bitmap序列中数字0对应的特征向量并反馈。
实际应用时，若Td为-3，则基站可以不执行步骤507(即基站实际未接收到终端反馈的Td，Td仍为初始化状态)，并结束当前流程，或者转到步骤505或步骤512。
步骤508中，基站可以将模型编号(即上述第三标识，或者上述第三标识和第四标识)及其他参数(比如周期T1、T2)发送给终端。
步骤509中，终端可以根据基站决策的信道压缩模型编号(即上述第三标识，或者上述第三标识和第四标识)选择相应的信道压缩模型，并完成 针对信道前K个特征向量的信道压缩。
步骤510中，终端可以按照与基站约定的反馈模式(即基站在步骤508中指示的反馈模式)将信道压缩信息(即上述至少包含信道压缩结果的第四信息)反馈给基站。
步骤511中，基站可以根据终端的反馈，使用对应的解压模型(即与终端使用的压缩模型配对的信道解压模型)恢复(即解压)相应的下行信道信息。
步骤512中，终端可以按照预设的周期(即上述第三周期，该周期可以与CSI-RS的发送周期一致)，周期性地计算信道相关性并将相关性与阈值的比较结果反馈给基站；基站可以根据终端的反馈更新(即重新配置)终端的压缩模型编号及信道反馈模式，终端可以按照基站重新配置的压缩模型及反馈模式上报压缩后的信道信息。示例性地，终端可以根据基站发送CSI-RS的周期，周期性地计算信道相关性并记录相关性与阈值的比较结果若与Td相等，则可以执行步骤509至510；若不等于Td，则需要利用更新Td，并执行步骤506至步骤510。
本应用示例提供的方案，与终端持续地按照特定周期向网络侧反馈信道压缩结果的反馈模式相比，终端采用可以随下行信道的稳定性变化而灵活变化的反馈模式向网络侧反馈下行信道的信道压缩结果，能够降低终端的反馈开销，从而有效降低无线传输开销，进而能够提高系统性能。
为了实现本申请实施例终端侧的方法，本申请实施例还提供了一种信息传输装置，设置在终端上，如图6所示，该装置包括：
获取单元601，配置为获取第一信息，所述第一信息用于确定下行信道的稳定性；
第一处理单元602，配置为至少利用所述第一信息，确定第二信息，所述第二信息表征所述下行信道的稳定性；
第一发送单元603，配置为向所述网络侧发送所述第二信息；
第一接收单元604，配置为接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；
第二处理单元605，配置为至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果；
第二发送单元606，配置为利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。
其中，在一实施例中，所述第一信息包含第五信息和第六信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度；所述第一处理单元602，还配置为：
至少利用所述第五信息确定第一参数，所述第一参数表征所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个特征向量之间的相似度；
将所述第一参数与所述至少一个阈值进行比较，得到第一比较结果，所述第二信息包含所述第一比较结果。
在一实施例中，所述第六信息还包含第三阈值，所述第三阈值大于所述第一阈值，且小于所述第二阈值；所述第一处理单元602，还配置为：
在所述第一参数大于或等于所述第一阈值，且小于所述第二阈值的情况下，针对所述下行信道在相邻两个时刻对应的两个信道矩阵，确定一个信道矩阵的K个特征向量中的每个特征向量与另一个信道矩阵中的对应特征向量之间的相似度，并将确定的相似度与所述第三阈值进行比较，得到K个第二比较结果；
利用所述K个第二比较结果，得到所述第一比较结果。
在一实施例中，所述获取单元601，还配置为从本地存储的信息中获取所述第一信息，或者，接收所述网络侧发送的所述第一信息。
在一实施例中，所述第一发送单元603，还配置为向所述网络侧发送CSI报告，所述CSI报告包含所述第二信息。
在一实施例中，所述获取单元601，还配置为执行以下操作之一：
接收所述网络侧广播的所述第一信息；
接收所述网络侧通过多播方式发送的所述第一信息；
接收所述网络侧通过单播方式发送的所述第一信息。
实际应用时，所述获取单元601可由所述终端中的处理器结合通信接口实现；所述第一发送单元603、第一接收单元604和第二发送单元606可由所述终端中的通信接口实现；所述第一处理单元602和第二处理单元605可由所述终端中的处理器实现。
为了实现本申请实施例网络设备侧的方法，本申请实施例还提供了一种信息传输装置，设置在网络设备上，如图7所示，该装置包括：
第二接收单元701，配置为接收终端发送的第二信息，所述第二信息表征下行信道的稳定性；
第三处理单元702，配置为至少利用所述第二信息，确定第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；
第三发送单元703，配置为向所述终端发送所述第三信息；
第三接收单元704，配置为接收所述终端发送的第四信息，所述第四信息至少包含信道压缩结果，所述信道压缩结果是至少利用所述第三信息进行所述下行信道的信道压缩得到的。
其中，在一实施例中，如图7所示，该装置还可以包括第四发送单元705，配置为向所述终端发送第一信息，所述第一信息用于确定所述下行信道的稳定性，所述第二信息是至少利用所述第一信息确定的。
在一实施例中，如图7所示，该装置还可以包括第四处理单元706，配置为确定所述第一信息，所述第一信息包含第五信息和第六信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度。
在一实施例中，所述第二接收单元701，还配置为接收所述终端发送的CSI报告，所述CSI报告包含所述第二信息。
在一实施例中，所述第四发送单元705，还配置为执行以下操作之一：
广播所述第一信息；
通过多播方式向所述终端发送所述第一信息；
通过单播方式向所述终端发送所述第一信息。
实际应用时，所述第二接收单元701、第三发送单元703、第三接收单元704和第四发送单元705可由所述网络设备中的通信接口实现；所述第三处理单元702和第四处理单元706可由所述网络设备中的处理器实现。
需要说明的是：上述实施例提供的信息传输装置在进行信息传输时，仅以上述各程序模块的划分进行举例说明，实际应用中，可以根据需要而将上述处理分配由不同的程序模块完成，即将装置的内部结构划分成不同的程序模块，以完成以上描述的全部或者部分处理。另外，上述实施例提供的信息传输装置与信息传输方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
基于上述程序模块的硬件实现，且为了实现本申请实施例终端侧的方法，本申请实施例还提供了一种终端，如图8所示，该终端800包括：
第一通信接口801，能够与其他终端和/或网络侧进行信息交互；
第一处理器802，与所述第一通信接口801连接，以实现与其他终端和/或网络侧进行信息交互，配置为运行计算机程序时，执行上述终端侧一个或多个技术方案提供的方法；
第一存储器803，所述计算机程序存储在所述第一存储器803上。
具体地，所述第一处理器802，配置为获取第一信息，所述第一信息用于确定下行信道的稳定性；至少利用所述第一信息，确定第二信息，所述第二信息表征所述下行信道的稳定性；
所述第一通信接口801，配置为向所述网络侧发送所述第二信息；接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；其中，
所述第一处理器802，还配置为至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果；
所述第一通信接口801，还配置为利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。
在一实施例中，所述第一信息包含第五信息和第六信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度；所述第一处理器802，还配置为：
至少利用所述第五信息确定第一参数，所述第一参数表征所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个特征向量之间的相似度；
将所述第一参数与所述至少一个阈值进行比较，得到第一比较结果，所述第二信息包含所述第一比较结果。
在一实施例中，所述第六信息还包含第三阈值，所述第三阈值大于所述第一阈值，且小于所述第二阈值；所述第一处理器802，还配置为：
在所述第一参数大于或等于所述第一阈值，且小于所述第二阈值的情况下，针对所述下行信道在相邻两个时刻对应的两个信道矩阵，确定一个信道矩阵的K个特征向量中的每个特征向量与另一个信道矩阵中的对应特征向量之间的相似度，并将确定的相似度与所述第三阈值进行比较，得到K个第二比较结果；
利用所述K个第二比较结果，得到所述第一比较结果。
在一实施例中，所述第一通信接口801，还配置为向所述网络侧发送CSI报告，所述CSI报告包含所述第二信息。
在一实施例中，所述第一处理器802，还配置为从本地存储的信息中获取所述第一信息，或者，通过所述第一通信接口801接收所述网络侧发送的所述第一信息。
在一实施例中，所述第一通信接口801，还配置为执行以下操作之一：
接收所述网络侧广播的所述第一信息；
接收所述网络侧通过多播方式发送的所述第一信息；
接收所述网络侧通过单播方式发送的所述第一信息。
需要说明的是：所述第一通信接口801和所述第一处理器802的具体处理过程可参照上述方法理解，这里不再赘述。
当然，实际应用时，终端800中的各个组件通过总线系统804耦合在一起。可理解，总线系统804用于实现这些组件之间的连接通信。总线系统804除包括数据总线之外，还包括电源总线、控制总线和状态信号总线。但是为了清楚说明起见，在图8中将各种总线都标为总线系统804。
本申请实施例中的第一存储器803配置为存储各种类型的数据以支持终端800的操作。这些数据的示例包括：用于在终端800上操作的任何计算机程序。
上述本申请实施例揭示的方法可以应用于所述第一处理器802中，或者由所述第一处理器802实现。所述第一处理器802可能是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法的各步骤可以通过所述第一处理器802中的硬件的集成逻辑电路或者软件形式的指令完成。所 述第一处理器802可以是通用处理器、数字信号处理器(DSP，Digital Signal Processor)，或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。所述第一处理器802可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者任何常规的处理器等。结合本申请实施例所公开的方法的步骤，可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于存储介质中，该存储介质位于第一存储器803，所述第一处理器802读取第一存储器803中的信息，结合其硬件完成前述方法的步骤。
在示例性实施例中，终端800可以被一个或多个应用专用集成电路(ASIC，Application Specific Integrated Circuit)、DSP、可编程逻辑器件(PLD，Programmable Logic Device)、复杂可编程逻辑器件(CPLD，Complex Programmable Logic Device)、现场可编程门阵列(FPGA，Field-Programmable Gate Array)、通用处理器、控制器、微控制器(MCU，Micro Controller Unit)、微处理器(Microprocessor)、或者其他电子元件实现，用于执行前述方法。
基于上述程序模块的硬件实现，且为了实现本申请实施例网络设备侧的方法，本申请实施例还提供了一种网络设备，如图9所示，该网络设备900包括：
第二通信接口901，能够与其他网络设备和/或终端进行信息交互；
第二处理器902，与所述第二通信接口901连接，以实现与其他网络设备和/或终端进行信息交互，配置为运行计算机程序时，执行上述网络设备侧一个或多个技术方案提供的方法；
第二存储器903，所述计算机程序存储在所述第二存储器903上。
具体地，所述第二通信接口901，配置为接收终端发送的第二信息，所述第二信息表征下行信道的稳定性；
所述第二处理器902，配置为至少利用所述第二信息，确定第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；其中，
所述第二通信接口901，还配置为向所述终端发送所述第三信息；接收所述终端发送的第四信息，所述第四信息至少包含信道压缩结果，所述信道压缩结果是至少利用所述第三信息进行所述下行信道的信道压缩得到的。
在一实施例中，所述第二通信接口901，还配置为向所述终端发送第一信息，所述第一信息用于确定所述下行信道的稳定性，所述第二信息是至少利用所述第一信息确定的。
在一实施例中，所述第二处理器902，还配置为确定所述第一信息，所述第一信息包含第五信息和第六信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度。
在一实施例中，所述第二通信接口901，还配置为接收所述终端发送的CSI报告，所述CSI报告包含所述第二信息。
在一实施例中，所述第二通信接口901，还配置为执行以下操作之一：
广播所述第一信息；
通过多播方式向所述终端发送所述第一信息；
通过单播方式向所述终端发送所述第一信息。
需要说明的是：所述第二通信接口901和所述第二处理器902的具体处理过程可参照上述方法理解，这里不再赘述。
当然，实际应用时，网络设备900中的各个组件通过总线系统904耦合在一起。可理解，总线系统904用于实现这些组件之间的连接通信。总线系统904除包括数据总线之外，还包括电源总线、控制总线和状态信号总线。但是为了清楚说明起见，在图9中将各种总线都标为总线系统904。
本申请实施例中的第二存储器903配置为存储各种类型的数据以支持网络设备900的操作。这些数据的示例包括：用于在网络设备900上操作的任何计算机程序。
上述本申请实施例揭示的方法可以应用于所述第二处理器902中，或者由所述第二处理器902实现。所述第二处理器902可能是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法的各步骤可以通过所述第二处理器902中的硬件的集成逻辑电路或者软件形式的指令完成。所述第二处理器902可以是通用处理器、DSP，或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。所述第二处理器902可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者任何常规的处理器等。结合本申请实施例所公开的方法的步骤，可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于存储介质中，该存储介质位于第二存储器903，所述第二处理器902读取第二存储器903中的信息，结合其硬件完成前述方法的步骤。
在示例性实施例中，网络设备900可以被一个或多个ASIC、DSP、PLD、CPLD、FPGA、通用处理器、控制器、MCU、Microprocessor、或其他电子元件实现，用于执行前述方法。
可以理解，本申请实施例的存储器(第一存储器803、第二存储器903)可以是易失性存储器或者非易失性存储器，也可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(ROM，Read Only Memory)、可编程只读存储器(PROM，Programmable Read-Only Memory)、可擦除可编程只读存储器(EPROM，Erasable Programmable Read-Only Memory)、电可擦除可编程只读存储器(EEPROM，Electrically Erasable Programmable Read-Only Memory)、磁性随机存取存储器(FRAM，ferromagnetic random access memory)、快闪存储器(Flash Memory)、磁表 面存储器、光盘、或只读光盘(CD-ROM，Compact Disc Read-Only Memory)；磁表面存储器可以是磁盘存储器或磁带存储器。易失性存储器可以是随机存取存储器(RAM，Random Access Memory)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用，例如静态随机存取存储器(SRAM，Static Random Access Memory)、同步静态随机存取存储器(SSRAM，Synchronous Static Random Access Memory)、动态随机存取存储器(DRAM，Dynamic Random Access Memory)、同步动态随机存取存储器(SDRAM，Synchronous Dynamic Random Access Memory)、双倍数据速率同步动态随机存取存储器(DDRSDRAM，Double Data Rate Synchronous Dynamic Random Access Memory)、增强型同步动态随机存取存储器(ESDRAM，Enhanced Synchronous Dynamic Random Access Memory)、同步连接动态随机存取存储器(SLDRAM，SyncLink Dynamic Random Access Memory)、直接内存总线随机存取存储器(DRRAM，Direct Rambus Random Access Memory)。本申请实施例描述的存储器旨在包括但不限于这些和任意其他适合类型的存储器。
为了实现本申请实施例提供的方法，本申请实施例还提供了一种信息传输系统，如图10所示，该系统包括：终端1001及网络设备1002。
这里，需要说明的是：所述终端1001及网络设备1002的具体处理过程已在上文详述，这里不再赘述。
在示例性实施例中，本申请实施例还提供了一种存储介质，即计算机存储介质，具体为计算机可读存储介质，例如包括存储计算机程序的第一存储器803，上述计算机程序可由终端800的第一处理器802执行，以完成前述终端侧方法所述步骤。再比如包括存储计算机程序的第二存储器903，上述计算机程序可由网络设备900的第二处理器902执行，以完成前述网络设备侧方法所述步骤。计算机可读存储介质可以是FRAM、ROM、PROM、EPROM、EEPROM、Flash Memory、磁表面存储器、光盘、或CD-ROM等存储器。
需要说明的是：“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。
另外，本申请实施例所记载的技术方案之间，在不冲突的情况下，可以任意组合。
以上所述，仅为本申请的较佳实施例而已，并非用于限定本申请的保护范围。

Claims (32)

  1. 一种信息传输方法，应用于终端，包括：
    获取第一信息，所述第一信息用于确定下行信道的稳定性；
    至少利用所述第一信息，确定第二信息，并向网络侧发送所述第二信息，所述第二信息表征所述下行信道的稳定性；
    接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；
    至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果，并利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。
  2. 根据权利要求1所述的方法，其中，所述第一信息包含第五信息和第六信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度；所述至少利用所述第一信息，确定第二信息，包括：
    至少利用所述第五信息确定第一参数，所述第一参数表征所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个特征向量之间的相似度；
    将所述第一参数与所述至少一个阈值进行比较，得到第一比较结果，所述第二信息包含所述第一比较结果。
  3. 根据权利要求2所述的方法，其中，所述第六信息包含第一阈值和第二阈值，所述第一阈值大于0，且小于所述第二阈值；所述第二阈值小于1；所述第二阈值和1形成的区间对应所述下行信道的第一状态，所述第一阈值和第二阈值形成的区间对应所述下行信道的第二状态，所述第一阈值和0形成的区间对应所述下行信道的第三状态；所述下行信道在所述第一状态下的稳定程度高于在所述第二状态下的稳定程度，且所述下行信道在所述第二状态下的稳定程度高于在所述第三状态下的稳定程度。
  4. 根据权利要求3所述的方法，其中，在所述第一参数大于或等于0，且小于所述第一阈值的情况下，所述第一比较结果包括第一标识，所述第一标识表征所述第三状态；
    或者，
    在所述第一参数小于或等于1，且大于或等于所述第二阈值的情况下，所述第一比较结果包括第二标识，所述第二标识表征所述第一状态。
  5. 根据权利要求3所述的方法，其中，所述第六信息还包含第三阈值，所述第三阈值大于所述第一阈值，且小于所述第二阈值；所述将所述 第一参数与所述至少一个阈值进行比较，得到第一比较结果，包括：
    在所述第一参数大于或等于所述第一阈值，且小于所述第二阈值的情况下，针对所述下行信道在相邻两个时刻对应的两个信道矩阵，确定一个信道矩阵的K个特征向量中的每个特征向量与另一个信道矩阵中的对应特征向量之间的相似度，并将确定的相似度与所述第三阈值进行比较，得到K个第二比较结果；
    利用所述K个第二比较结果，得到所述第一比较结果。
  6. 根据权利要求5所述的方法，其中，所述第一比较结果通过位图Bitmap序列的方式表征所述K个第二比较结果；其中，在所述Bitmap序列中的比特位是第一值的情况下，对应的第二比较结果表征确定的相似度小于或等于所述第三阈值，且大于或等于0；在所述Bitmap序列中的比特位是第二值的情况下，对应的第二比较结果表征确定的相似度大于所述第三阈值，且小于或等于1。
  7. 根据权利要求1至6任一项所述的方法，其中，所述反馈模式包括以下之一：
    向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果，O为大于0的整数；
    按照第一周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果，O为大于0的整数；
    向所述网络侧反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果后，按照第二周期，周期性地向所述网络侧反馈针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩结果，O和P为大于0的整数。
  8. 根据权利要求1至6任一项所述的方法，其中，所述第三信息包含至少一个模型的标识，所述至少一个模型用于进行所述下行信道的信道压缩。
  9. 根据权利要求8所述的方法，其中，所述第三信息包含第三标识，所述第三标识表征用于进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩的模型，O为大于0的整数；
    或者，
    所述第三信息包含第三标识和第四标识，所述第三标识表征用于进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩的模型，所述第四标识表征用于进行针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩的模型，O和P为大于0的整数。
  10. 根据权利要求1至6任一项所述的方法，其中，所述向所述网络侧发送所述第二信息，包括：
    向所述网络侧发送信道状态信息CSI报告，所述CSI报告包含所述 第二信息。
  11. 根据权利要求1至6任一项所述的方法，其中，所述获取第一信息，包括：
    从本地存储的信息中获取所述第一信息，或者，接收所述网络侧发送的所述第一信息。
  12. 根据权利要求11所述的方法，其中，所述接收所述网络侧发送的所述第一信息，包括以下之一：
    接收所述网络侧广播的所述第一信息；
    接收所述网络侧通过多播方式发送的所述第一信息；
    接收所述网络侧通过单播方式发送的所述第一信息。
  13. 一种信息传输方法，应用于网络设备，包括：
    接收终端发送的第二信息，所述第二信息表征下行信道的稳定性；
    至少利用所述第二信息，确定第三信息，并向所述终端发送所述第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；
    接收所述终端发送的第四信息，所述第四信息至少包含信道压缩结果，所述信道压缩结果是至少利用所述第三信息进行所述下行信道的信道压缩得到的。
  14. 根据权利要求13所述的方法，其中，所述方法还包括：
    向所述终端发送第一信息，所述第一信息用于确定所述下行信道的稳定性，所述第二信息是至少利用所述第一信息确定的。
  15. 根据权利要求14所述的方法，其中，所述方法还包括：
    确定所述第一信息，所述第一信息包含第五信息和第六信息，所述第五信息包含所述下行信道的信道矩阵中的特征向量数K，K为大于0的整数；所述第六信息包含至少一个阈值，所述至少一个阈值对应所述下行信道的至少两种稳定程度。
  16. 根据权利要求15所述的方法，其中，所述第二信息包含第一比较结果，所述第一比较结果是将第一参数与所述至少一个阈值进行比较得到的，所述第一参数是至少利用所述第五信息确定的，所述第一参数表征所述下行信道在相邻两个时刻对应的两个信道矩阵中的K个特征向量之间的相似度。
  17. 根据权利要求15所述的方法，其中，所述第六信息包含第一阈值和第二阈值，所述第一阈值大于0，且小于所述第二阈值；所述第二阈值小于1；所述第二阈值和1形成的区间对应所述下行信道的第一状态，所述第一阈值和第二阈值形成的区间对应所述下行信道的第二状态，所述第一阈值和0形成的区间对应所述下行信道的第三状态；所述下行信道在所述第一状态下的稳定程度高于在所述第二状态下的稳定程度，且所述下行信道在所述第二状态下的稳定程度高于在所述第三状态下的稳 定程度。
  18. 根据权利要求17所述的方法，其中，在所述第一参数大于或等于0，且小于所述第一阈值的情况下，所述第一比较结果包括第一标识，所述第一标识表征所述第三状态；
    或者，
    在所述第一参数小于或等于1，且大于或等于所述第二阈值的情况下，所述第一比较结果包括第二标识，所述第二标识表征所述第一状态。
  19. 根据权利要求17所述的方法，其中，所述第六信息还包含第三阈值，所述第三阈值大于所述第一阈值，且小于所述第二阈值；在所述第一参数大于或等于所述第一阈值，且小于所述第二阈值的情况下，所述第一比较结果是利用K个第二比较结果得到的；所述K个第二比较结果是针对所述下行信道在相邻两个时刻对应的两个信道矩阵，确定一个信道矩阵的K个特征向量中的每个特征向量与另一个信道矩阵中的对应特征向量之间的相似度，并将确定的相似度与所述第三阈值进行比较得到的。
  20. 根据权利要求19所述的方法，其中，所述第一比较结果通过Bitmap序列的方式表征所述K个第二比较结果；其中，在所述Bitmap序列中的比特位是第一值的情况下，对应的第二比较结果表征确定的相似度小于或等于所述第三阈值，且大于或等于0；在所述Bitmap序列中的比特位是第二值的情况下，对应的第二比较结果表征确定的相似度大于所述第三阈值，且小于或等于1。
  21. 根据权利要求13至20任一项所述的方法，其中，所述反馈模式包括以下之一：
    所述终端向所述网络设备反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果，O为大于0的整数；
    所述终端按照第一周期，周期性地向所述网络设备反馈针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果，O为大于0的整数；
    所述终端向所述网络设备反馈一次针对所述下行信道的信道矩阵中的O个特征向量的信道压缩结果后，按照第二周期，周期性地向所述网络设备反馈针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩结果，O和P为大于0的整数。
  22. 根据权利要求13至20任一项所述的方法，其中，所述第三信息包含至少一个模型的标识，所述至少一个模型用于进行所述下行信道的信道压缩。
  23. 根据权利要求22所述的方法，其中，所述第三信息包含第三标识，所述第三标识表征用于进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩的模型，O为大于0的整数；
    或者，
    所述第三信息包含第三标识和第四标识，所述第三标识表征用于进行针对所述下行信道的信道矩阵中的O个特征向量的信道压缩的模型，所述第四标识表征用于进行针对所述下行信道的信道矩阵的O个特征向量中变化情况满足第一条件的P个特征向量的信道压缩的模型，O和P为大于0的整数。
  24. 根据权利要求13至20任一项所述的方法，其中，所述接收所述终端发送的第二信息，包括：
    接收所述终端发送的CSI报告，所述CSI报告包含所述第二信息。
  25. 根据权利要求14至20任一项所述的方法，其中，所述向所述终端发送第一信息，包括以下之一：
    广播所述第一信息；
    通过多播方式向所述终端发送所述第一信息；
    通过单播方式向所述终端发送所述第一信息。
  26. 一种信息传输装置，包括：
    获取单元，配置为获取第一信息，所述第一信息用于确定下行信道的稳定性；
    第一处理单元，配置为至少利用所述第一信息，确定第二信息，所述第二信息表征所述下行信道的稳定性；
    第一发送单元，配置为向网络侧发送所述第二信息；
    第一接收单元，配置为接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；
    第二处理单元，配置为至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果；
    第二发送单元，配置为利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。
  27. 一种信息传输装置，包括：
    第二接收单元，配置为接收终端发送的第二信息，所述第二信息表征下行信道的稳定性；
    第三处理单元，配置为至少利用所述第二信息，确定第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向网络设备反馈所述下行信道的信道压缩结果的反馈模式；
    第三发送单元，配置为向所述终端发送所述第三信息；
    第三接收单元，配置为接收所述终端发送的第四信息，所述第四信息至少包含信道压缩结果，所述信道压缩结果是至少利用所述第三信息进行所述下行信道的信道压缩得到的。
  28. 一种终端，包括：
    第一处理器，配置为获取第一信息，所述第一信息用于确定下行信道的稳定性；至少利用所述第一信息，确定第二信息，所述第二信息表征所述下行信道的稳定性；
    第一通信接口，配置为向网络侧发送所述第二信息；接收所述网络侧发送的第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络侧反馈所述下行信道的信道压缩结果的反馈模式；所述第三信息是至少利用所述第二信息确定的；其中，
    所述第一处理器，还配置为至少利用所述第三信息进行所述下行信道的信道压缩，得到信道压缩结果；
    所述第一通信接口，还配置为利用所述第三信息向所述网络侧发送第四信息，所述第四信息至少包含所述信道压缩结果。
  29. 一种网络设备，包括：
    第二通信接口，配置为接收终端发送的第二信息，所述第二信息表征下行信道的稳定性；
    第二处理器，配置为至少利用所述第二信息，确定第三信息，所述第三信息指示所述下行信道的信道压缩方式及用于向所述网络设备反馈所述下行信道的信道压缩结果的反馈模式；其中，
    所述第二通信接口，还配置为向所述终端发送所述第三信息；接收所述终端发送的第四信息，所述第四信息至少包含信道压缩结果，所述信道压缩结果是至少利用所述第三信息进行所述下行信道的信道压缩得到的。
  30. 一种终端，包括：第一处理器和配置为存储能够在处理器上运行的计算机程序的第一存储器，
    其中，所述第一处理器配置为运行所述计算机程序时，执行权利要求1至12任一项所述方法的步骤。
  31. 一种网络设备，包括：第二处理器和配置为存储能够在处理器上运行的计算机程序的第二存储器，
    其中，所述第二处理器配置为运行所述计算机程序时，执行权利要求13至25任一项所述方法的步骤。
  32. 一种存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现权利要求1至12任一项所述方法的步骤，或者实现权利要求13至25任一项所述方法的步骤。
PCT/CN2023/131633 2022-11-21 2023-11-14 信息传输方法、装置、相关设备及存储介质 Ceased WO2024109596A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211456381.5 2022-11-21
CN202211456381.5A CN118057874A (zh) 2022-11-21 2022-11-21 信息传输方法、装置、相关设备及存储介质

Publications (1)

Publication Number Publication Date
WO2024109596A1 true WO2024109596A1 (zh) 2024-05-30

Family

ID=91069188

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/131633 Ceased WO2024109596A1 (zh) 2022-11-21 2023-11-14 信息传输方法、装置、相关设备及存储介质

Country Status (2)

Country Link
CN (1) CN118057874A (zh)
WO (1) WO2024109596A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130100922A1 (en) * 2010-06-22 2013-04-25 Lg Electronics Inc. Method and apparatus for transmitting channel state information
CN112468203A (zh) * 2020-11-19 2021-03-09 杭州勒贝格智能系统股份有限公司 深度迭代神经网络用低秩csi反馈方法、存储介质及设备
CN113381950A (zh) * 2021-04-25 2021-09-10 清华大学 基于网络聚合策略的高效mimo信道反馈方法及装置
CN113726375A (zh) * 2021-08-17 2021-11-30 北京理工大学重庆创新中心 一种基于深度学习的信道信息压缩反馈重建方法
CN115021787A (zh) * 2022-05-30 2022-09-06 大连大学 一种基于复数卷积神经网络的信道状态信息反馈方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130100922A1 (en) * 2010-06-22 2013-04-25 Lg Electronics Inc. Method and apparatus for transmitting channel state information
CN112468203A (zh) * 2020-11-19 2021-03-09 杭州勒贝格智能系统股份有限公司 深度迭代神经网络用低秩csi反馈方法、存储介质及设备
CN113381950A (zh) * 2021-04-25 2021-09-10 清华大学 基于网络聚合策略的高效mimo信道反馈方法及装置
CN113726375A (zh) * 2021-08-17 2021-11-30 北京理工大学重庆创新中心 一种基于深度学习的信道信息压缩反馈重建方法
CN115021787A (zh) * 2022-05-30 2022-09-06 大连大学 一种基于复数卷积神经网络的信道状态信息反馈方法

Also Published As

Publication number Publication date
CN118057874A (zh) 2024-05-21

Similar Documents

Publication Publication Date Title
JP7693025B2 (ja) 通信方法および装置
CN116032419B (zh) 用于反馈信道状态的方法和装置
CN115996160A (zh) 通信系统中的方法和设备
US20230412230A1 (en) Systems, methods, and apparatus for artificial intelligence and machine learning based reporting of communication channel information
CN116260552B (zh) 一种csi发送和接收方法及装置
WO2022199142A1 (zh) 网络参数集合信息传输方法、装置、终端、基站和介质
WO2023030538A1 (zh) 信道状态信息的处理方法、终端、基站、计算机可读存储介质
WO2023179476A1 (zh) 信道特征信息上报及恢复方法、终端和网络侧设备
CN115001629A (zh) 信道量化反馈方法、装置、电子设备及存储介质
US20250183962A1 (en) Information transmission method and device, and storage medium
WO2023246618A1 (zh) 信道矩阵处理方法、装置、终端及网络侧设备
WO2025035776A1 (zh) 信道状态信息的发送和接收方法、通信装置、及存储介质
WO2025148327A1 (zh) 信道状态信息的发送和接收方法、通信装置及存储介质
CN114692774A (zh) 编码器压缩率的调整方法、数据的反馈方法及装置
CN117459104A (zh) 一种传输方法、通信节点及存储介质
WO2025129950A1 (zh) 波束赋形方法、装置、智能超表面中继系统和基站
WO2022199090A1 (zh) 信道状态信息传输方法、装置、终端、基站和存储介质
CN112994832A (zh) 参数确定方法、装置、相关设备及存储介质
CN115694722A (zh) 一种通信方法及装置
WO2024149157A1 (zh) Csi传输方法、装置、终端及网络侧设备
WO2022151064A1 (zh) 信息发送方法、信息接收方法、装置、设备及介质
WO2023179570A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
WO2023185980A1 (zh) 信道特征信息传输方法、装置、终端及网络侧设备
KR20250157987A (ko) 통신 채널 정보의 머신 러닝 기반 보고를 위한 모델 상호 운용성 및 협업
CN118057874A (zh) 信息传输方法、装置、相关设备及存储介质

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: 23893686

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 02/09/2025)

122 Ep: pct application non-entry in european phase

Ref document number: 23893686

Country of ref document: EP

Kind code of ref document: A1