WO2011160566A1

WO2011160566A1 - Method and system for dynamically feeding back long term channel state information

Info

Publication number: WO2011160566A1
Application number: PCT/CN2011/075903
Authority: WO
Inventors: 孙云锋; 朱常青; 郭森宝
Original assignee: 中兴通讯股份有限公司
Priority date: 2010-06-21
Filing date: 2011-06-17
Publication date: 2011-12-29
Also published as: CN102291217B; CN102291217A

Abstract

The present invention discloses a method for dynamically feeding back long term Channel State Information (CSI). The method comprises: the User Equipment (UE) determines whether the long term CSI needs to be fed back to the network side, and notifies the network side of the determination result. The present invention simultaneously discloses a system for dynamically feeding back long term CSI. The system comprises a determining unit and a notifying unit; the determining unit and the notifying unit are placed in the UE, wherein the determining unit is used to determine whether the long term CSI needs to be fed back to the network side, and the notifying unit is used to notify the network side of the determination result of the determining unit. In the present invention, by comparing the current long time CSI and the former long time CSI in use, the long time CSI is fed back to the network side when the difference exceeds a preset threshold; and determining to feed back long term CSI to the network side enables flexible feedback, thus economizing feedback overhead.

Description

Dynamic feedback method and system for long-term channel state information

The present invention relates to the technology of Long Term Channel State Information, and in particular to a dynamic feedback method and system for long-term channel state information. Background technique

In a Long Term Evolution (LTE) system, a base codebook is shared between a base station (eNB) and a user equipment (UE, User Equipment), and the UE passes a certain criterion, for example, based on maximum signal dry noise. Select the appropriate Precoding Matrix Index (PMI, Precoding Matrix Index) for the SINR (Signal to Interference plus Noise Ratio) criterion. The LTE system implements precoding on the eNB side by means of UE feedback PMI, thereby improving system performance. A base codebook can be viewed as a single rank or multiple rank approximate effective channel option for a set of non-correlated channels. However, these codes are specifically designed for a particular transmitter, in the sense that they are fixed codes and cannot adaptively change as the channel changes.

In the Advanced Long Term Evolution (LTE-A or LTE-Advanced, Long Term Evolution Advanced) system, the eNB can improve system performance by feeding back long-term channel state information (such as spatial and temporal correlation matrices). By adapting the characteristics of the actual relevant channel, the base code can be transformed according to statistical knowledge. A common method for the eNB to obtain long-term channel state information is to receive a feedback report sent upstream from the UE. Although this approach is attractive from a performance perspective, its additional signaling overhead will be inevitable.

With the popularity of wireless communication technologies and the continuous improvement of service bandwidth, there are fewer and fewer wireless spectrum resources. Therefore, how to improve the performance of wireless communication and reduce the overhead of signaling is the principle that communication systems adhere to. Uplink feedback from the UE is necessary, but its signaling overhead is none. If it can avoid the signaling overhead, it will undoubtedly save valuable spectrum resources. Summary of the invention

In view of this, the main object of the present invention is to provide a dynamic feedback method and system for long-term channel state information, which can dynamically implement long-term CSI feedback and save signaling overhead.

In order to achieve the above object, the technical solution of the present invention is achieved as follows:

A dynamic feedback method for long-term channel state information, including:

The user equipment UE determines whether it is necessary to feed back the long-term channel state information CSI to the network side, and notifies the network side of the determination result.

Preferably, the UE determines whether it is necessary to feed back the long-term CSI to the network side, which is:

The current long-term CSI is compared with the previous long-term CSI. When the difference exceeds the set threshold, the long-term CSI is fed back to the network side. Otherwise, the long-term CSI is not fed back.

Preferably, the method further includes:

When the UE determines to feed back the long-term CSI to the network side, feedback or feedback the long-term CSI according to the network side indication.

Preferably, the notifying the network side of the determination result is:

The UE notifies the network side whether to feed back the long-term CSI to the network side by using a notification message;

Or the setting, by the UE, in the message carrying the long-term CSI, whether to carry the long-term CSI indication information, indicating whether to carry the long-term CSI.

Preferably, when the UE determines to feed back the long-term CSI to the network side, the long-term CSI is fed back as:

The UE notifies the network side to feed back the long-term CSI to the network side by using the notification message, and carries the long-term CSI in the k-th subframe after the subframe that carries the notification message, and feeds back to the network side; where k is an advance The determined value of the setting;

Or, when the UE determines to feed back the long-term CSI to the network side, the long-term CSI is carried in the k1th subframe after the current frame, and is fed back to the network side; wherein, kl is preset Value.

Preferably, when the UE determines to feed back the long-term CSI to the network side, the long-term CSI is fed back according to the network side indication, which is:

The UE notifies the network side to feed back the long-term CSI to the network side through the notification message;

The network side carries information indicating whether the UE feeds back the long-term CSI in the k2th subframe after the subframe that carries the notification message;

When the network side indication information indicates that the UE feeds back the long-term CSI, the UE carries the long-term CSI in the k3th subframe after the subframe that carries the network side indication information, and feeds back to the network side; The network side; wherein, k2 and k3 are all predetermined determination values.

Preferably, the k, k1, k2, and k3 are constant values in a frequency division duplex (FDD) system frame, and in a time division duplex (TDDD) system frame, according to the system Determine the frame uplink and downlink configuration mode.

Preferably, when the UE feeds back the long-term CSI, it also feeds back the short-term PMI.

A dynamic feedback system for long-term channel state information, including a determining unit and a notification unit; the determining unit and the notification unit are disposed in the UE;

a determining unit, configured to determine whether the long-term channel state information CSI needs to be fed back to the network side; and the notifying unit is configured to notify the network side of the determination result of the determining unit.

Preferably, the determining unit further compares the current long-term CSI with the previous long-term CSI, and feeds back the long-term CSI to the network side when the difference exceeds the set threshold, otherwise the long-term CSI is not fed back.

Preferably, the system further comprises:

The feedback unit is configured to: when the determining unit determines to feed back the long-term CSI to the network side, feedback or feed back the long-term CSI according to the network side indication.

Preferably, the notification unit notifies the network side whether to feed back long-term CSI to the network side by using a notification message; Or the notification unit sends the long-term CSI message to the network side, where the long-term CSI is carried in the message carrying the long-term CSI, and indicates whether the long-term CSI is carried.

Preferably, the notification unit notifies the network side to feed back the long-term network side through the notification message

The CSI, the feedback unit carries the long-term CSI in the kth subframe after the subframe that carries the notification message, and feeds back to the network side; where k is a preset determination value;

Or, the determining unit determines to feed the long-term CSI to the network side, where the feedback unit carries the long-term CSI in the k1th subframe after the current frame, and feeds back to the network side; where, k1 is a preset determination. value.

CSI;

The system further includes a sending unit, configured to be configured on the network side, configured to carry information indicating whether the UE feeds back the long-term CSI in a k2th subframe after the subframe that carries the notification message; When the UE feeds back the long-term CSI, the feedback unit carries the long-term CSI in the k3th subframe after the subframe that carries the network side indication information, and feeds back to the network side; and feeds back to the network side; wherein, k2 And k3 are all predetermined values determined in advance.

In the present invention, by comparing the current long-term CSI with the previous long-term CSI, the long-term CSI is fed back to the network side when the difference exceeds the set threshold; and, in determining the long-term feedback to the network side

When CSI is used, feedback can be flexibly and feedback costs can be saved. DRAWINGS

1 is a flowchart of a dynamic feedback method for long-term channel state information according to the present invention;

2 is a schematic diagram showing a first composition structure of a dynamic feedback system for long-term channel state information according to the present invention;

3 is a schematic diagram of a second component structure of a dynamic feedback system for long-term channel state information according to the present invention; 4 is a schematic diagram of a third component structure of a dynamic feedback system for long-term channel state information according to the present invention. detailed description

The present invention will be further described in detail below with reference to the accompanying drawings.

1 is a flowchart of a dynamic feedback method for long-term channel state information according to the present invention. As shown in FIG. 1, the dynamic feedback method for long-term channel state information of the present invention includes the following steps:

Step 101: The UE determines whether the long-term channel state information CSI needs to be fed back to the network side, and notifies the network side of the determination result.

The UE compares the current long-term CSI with the previous long-term CSI, and feeds back the long-term CSI to the network side when the difference exceeds the set threshold, otherwise the long-term CSI is not fed back. For example, you can set the threshold of the difference. If the threshold is exceeded, the uplink feedback will be performed. Otherwise, the uplink feedback will not be performed.

The UE notifies the network side whether to feed back the long-term CSI to the network side through the notification message; the notification message may be carried in the PUCCH in the subframe or in other uplink channels. The notification message can be a new notification message, or an existing message can be used as a notification message.

Of course, the UE may also set whether to carry the long-term CSI indication information in the message carrying the long-term CSI, and indicate whether to carry the long-term CSI. That is, when transmitting long-term CSI, the setting indicates whether the transmitted information contains long-term CSI. If a special indicator bit is set therein, it is determined whether the long-term CSI is included by parsing the indication bit.

When the UE feeds back the long-term CSI, the long-term CSI is carried in the subframe after the set interval after the subframe that carries the notification message; or is carried in the subframe after the set interval after receiving the subframe where the indication information of the network is located Long-term CSI.

The above long-term CSI, including a correlation matrix of channel information, or a correlation matrix including PMI

PML

Step 102: The network side receives the long-term CSI fed back by the UE, or re-instructs the UE to perform long-term Feedback from the CSI, the UE performs long-term CSI feedback according to the network side indication.

The essence of the technical solution of the present invention will be further clarified by the following examples.

Embodiment 1

This example specifically includes the following steps.

Step 1. Simulate and test the difference value R0 according to the actual situation. Here, the threshold R0 is obtained by simulation of the actual communication system and related tests, and is mainly used to measure the current long-term CSI and the previous use. The degree of difference between long-term CSI. This value is an empirical value. Those skilled in the art will appreciate that the above setting of R0 is easy to implement.

Step 2: The user equipment (UE) first calculates the chord distance L. There are many specific calculation methods, which can be realized by the prior art familiar to those skilled in the art. The chord distance L is the actual difference between the current long-term CSI and the long-term CSI previously used.

Step 3: It is assumed that the correlation matrix corresponding to the subframe time is the correlation matrix corresponding to the time of the subframe w, and the original codebook at the eNB and the UE is PM/. At the time of the subframe w, if the UE measurement does not require feedback, that is, 1^<1 0, the UE still uses the previous long-term channel information? («.) to perform codebook conversion, that is, R( _n .) X PMI, that is, only The PMI is not required to be fed back, and the Long Term CSI is not fed back. Step 4: The UE sets the lbit indication information on the Physical Uplink Control Channel (PUCCH) of the subframe to indicate whether the network side needs to feed back the long-term. CSI, for example, when the indication information is 0, the eNodeB is notified that it does not need to receive long-term CSI feedback information in the "+ uplink subframe, and the eNodeB saves the last received long-term CSI information? («.) until the next update. Then use the long-term CSI information? («.) to correct the short-term PM/ of the feedback. The value of k can be designed according to FDD and TDD respectively. For example, it is always equal to 4 in FDD, and k is selected according to different configurations and subframes in TDD as shown in Table 1.

TDD UL DL UL subframe index n

Configuration 0 1 2 3 4 5 6 7 8 9

0 4 7 6 4 7 6 2 6 6

3 6 6 6

4 6 6

5 6

6 4 6 6 4 7

Table 1

In Table 1, there are 7 ways to configure TDD frames, and each frame has 10 subframes. Taking configuration mode “0” as an example, when n is the second subframe, k is 4, when n is the third subframe, k is 7; when n is the fourth subframe in the frame, k is 6 When n is the 7th subframe, k is 4; when n is the 8th subframe, k is 7; when n is the ninth subframe, k is 6. The frame and k values of other configurations are the same as those described above, and are not described here.

Step 5: If the long-term CSI needs to be fed back, that is, L>R0, the UE uses the current Long-Term CSI to perform codebook conversion, that is, XPM/generates the updated codebook, and the UE can use the maximum SINR criterion to update the codebook. Select the corresponding PM/. The corresponding PM/ and step 6 need to be fed back. At this time, the UE has a lbit information of 1 on the PUCCH on the subframe w, and the eNodeB is notified to receive the long-term channel information R( _n in the uplink subframe of the tt+k, the eNodeB saves the UE. R(n) fed back in the n + kth uplink subframe until the next update, and then using the received R(n) to correct the short-term PMI of the feedback. The value of k can be designed according to FDD and TDD, for example, It is always equal to 4 in FDD, and is selected according to different configurations and subframes in TDD as shown in Table 1.

Step 7. The receiving end determines whether to update the Long Term CSI according to the received control information, so as to perform the next codebook conversion. On the eNodeB side, depending on the feedback method, the currently used Long-Term CSI may be ?(«.) or codebook conversion, and find the codeword corresponding to the PMI for precoding;

Step 8. Repeat steps 3 through 6.

Embodiment 2

This example specifically includes the following steps.

Step 1, before the simulation and testing according to the actual situation, the threshold R0 is obtained; this step is the same as in the first embodiment. Step 1 is the same;

Step 2: First, calculate the chord distance L at the receiving end (UE), and there are many specific calculation methods, which are not described here; this step is the same as step 2 in the first embodiment;

Step 3, assuming a sub-frame. The correlation matrix corresponding to the time is ?(«.), and the correlation matrix corresponding to the time of the subframe w is that the original codebook at the eNodeB and the UE is PMI. At the time of subframe n, if it is not necessary to feed back the long-term CSI, that is, L < R0, the UE still uses the previous long-term channel information? («.) #丈码本转换, ie? («.) PMI, that is, only feedback is needed. Corresponding Μ/, does not feedback Long Term CSI;

Step 4: At this time, the information of the lbit identifier bit of the UE on the PUCCH of the subframe w is 0, and the eNodeB is notified that the eNodeB does not need to receive the feedback information of the long-term CSI and the short-term PMI in the ++ uplink subframe, and the eNodeB saves The long-term CSI information received recently («.) until the next update, and then use the long-term CSI information («.) to correct the short-term PM/ of the feedback. The value can be designed according to FDD and TDD respectively. For example, it is always equal to 4 in FDD, and it is selected according to different configurations and subframes in TDD.

Step 5: If the long-term CSI needs to be fed back, that is, L>R0, the UE uses the current Long Term.

For CSI, code-based conversion, that is, x PM/ generation of updated codebook, the UE can use the maximum SINR criterion and select the corresponding PMI in the update codebook. Need to feedback the corresponding PMI and .

Step 6: At this time, the UE has a lbit information of 1 on the PUCCH of the subframe w, and informs the eNodeB that the PM/sum needs to be received in the w+ uplink subframe, and the eNodeB saves the feedback of the UE in the w+ uplink subframe until the next time. Update and then use the short-term PMI received to correct the feedback. The values can be designed according to FDD and TDD respectively. For example, in FDD, it is equal to 4, and in TDD, it is selected according to different configurations and subframes as shown in Table 1.

In step 7, the receiving end determines whether to update the Long-Term CSI according to the received control information, so as to perform the next codebook conversion. On the eNodeB side, depending on the feedback method, the currently used Long-Term CSI may be ? («.) or R(n) for codebook conversion, and find the corresponding PMI. Codewords are precoded;

Step 8. Repeat steps 3 through 6.

Embodiment 3

This example specifically includes the following steps.

Step 1 : Simulate and test according to the actual situation to obtain a threshold value R0; this step is the same as step 1 in the first embodiment;

Step 3, assuming that the correlation matrix corresponding to the time in the subframe "?" is ?(«.), the correlation matrix corresponding to the time of the subframe w is that the original codebook at the eNodeB and the UE side is the PMI. At the time of the subframe n, if not Need to feedback long-term CSI, that is, L < R0, the UE still uses the previous long-term channel information? («.) to do the codebook conversion, ie? («.) PMI, that is, only the corresponding M/, the feedback is not required. Term CSI;

Step 4: At this time, the information of the lbit identifier bit added by the UE on the PUCCH of the subframe is 0, and the eNodeB is notified that the CSI indicator bit of the DCI Format 0 of the subframe is not required to be updated. 0. The CQI indication bit value of the DCI Format 0 of the eNodeB in the first subframe is not completely controlled by the lbit information fed back by the UE, and may also be determined by the eNodeB to be set to 1. If the UE receives the CQI indication of the eNodeB as 1, judge The lbit information of the PUCCH fed back in the subframe is 0 or 1, if it is 0, the short-term PMI is fed back, and if it is 1, the long-term channel information is fed back. In this example, it is determined by the network side whether to perform feedback of long-term CSI, but it is determined by the UE and directly notified to the network side. . The value is optional, and can also be designed according to FDD and TDD, for example, a value equal to 1 to 10 in FDD, and selected according to different configurations and subframes in TDD. Then the UE does not feed back the long-term CSI information in the ++ uplink subframe. The eNode saves the CSI information of the last feedback? («.) until the next update, and then the long-term CSI information? («.) to correct the short-term feedback. PMI. The value of k can be designed according to FDD and TDD respectively. For example, k is always equal to 4 in FDD, and is selected according to different configurations and subframes in TDD as shown in Table 2.

Table 2

In Table 2, there are 7 ways to configure TDD frames, and each frame has 10 subframes. Taking configuration mode “0” as an example, when n is the 0th subframe, k is 4, when n is the first subframe, k is 6; when n is the 5th subframe in the frame, k is 4 , when n is the sixth subframe, k is 6. The frame and k values of other configurations are the same as those described above, and are not described here.

Step 5: If the long-term CSI needs to be fed back, that is, L>R0, the UE uses the current Long-Term CSI to perform codebook conversion, that is, the X PMI generates an updated codebook, and the UE can use the maximum SINR criterion in the updated codebook. Select the corresponding PMI. Need to feedback the corresponding PMI and ;

Step 6: At this time, the lbit information added by the UE on the PUCCH of the subframe is 1 to notify the eNodeB that the long-term CSI needs to be updated, and then the eNodeB sets the CQI indication bit of the DCI Format 0 of the first subframe to be set to 1. The eNodeB is in the first sub- The CQI indication bit value of the DCI Format 0 of the frame is not completely controlled by the lbit information fed back by the UE, and may also be determined by the eNodeB. If the UE receives the CQI indication of the eNodeB as 1, it determines the feedback on the "subframe". Whether the PUCCH lbit information is 0 or 1, if it is 0, it feeds back the short-term PMI, and if it is 1, it feeds back the long-term channel information. The value is optional, and can also be designed according to FDD and TDD, for example, ^ is a value equal to 1 to 10 in FDD, and is selected according to different configurations and subframes in TDD. Then the UE feeds back the long-term CSI feedback information in the ++ uplink subframe. The eNodeB saves the long-term CSI and saves it until the next update, and then uses the received long-term CSI to correct the short-term PMI of the feedback. The value of k can be designed according to FDD and TDD respectively, for example, it is always equal to 4 in FDD, and is selected according to different configurations and subframes in TDD as shown in Table 2.

Step 7. The receiving end determines whether to update the Long Term CSI according to the received control information, so as to perform the next codebook conversion. On the eNodeB side, depending on the feedback method, the currently used Long Term CSI may be ?(«.) or codebook conversion, and find the codeword corresponding to the PMI for precoding;

Step 8. Repeat steps 3 through 6.

Embodiment 4

This example specifically includes the following steps.

Step 3, assuming a sub-frame. The correlation matrix corresponding to the time is ?(«.), and the correlation matrix corresponding to the time of the subframe is that the original codebook at the eNodeB and the UE is the PMI. At the time of the subframe n, if the long-term CSI is not required to be fed back, that is, L<R0 , UE still uses the previous long-term channel information? («.) #丈码本转换, ie? («.) PMI, that is, only need to feedback the corresponding Μ /, do not feedback Long Term CSI;

Step 4: At this time, the information of the identifier of the lbit added by the UE on the PUCCH of the subframe w is 0, and the eNodeB is notified that the CSI indicator bit of the DCI Format 0 of the first subframe is not set to 0. . The eNodeB is at the first. The CQI indication bit value of the DCI Format 0 of the subframe is not completely controlled by the lbit information fed back by the UE, and may also be determined by the eNodeB to be set. If the UE receives the CQI indication of the eNodeB as 1, it determines the "subframe" Whether the feedback PUCCH lbit information is 0 or 1, if it is 0, it feeds back the short-term PMI, and if it is 1, it feeds back the long-term channel information and the short-term PMI information. The values are selectable, and can also be designed according to FDD and TDD, for example, a value equal to 1 to 10 in FDD, and selected in TDD according to different configurations and subframes. Then the UE does not feed back the long-term CSI information and the short-term PMI information in the ++ uplink subframe, and the eNodeB saves the CSI information of the last feedback? (w.) until the next update, and then the long-term CSI information? (w.) Correct the short-term PMI of the feedback. The value of k can be designed according to FDD and TDD respectively, for example, it is always equal to 4 in FDD, and is selected according to different configurations and subframes in TDD as shown in Table 2.

Step 6: At this time, the lbit information added by the UE on the PUCCH of the subframe is 1 to notify the eNodeB that the long-term CSI needs to be updated, and then the eNodeB sets the CQI indication bit of the DCI Format 0 of the first subframe to 1. The eNodeB is in the first sub- The CQI indication bit value of the DCI Format 0 of the frame is not completely controlled by the lbit information fed back by the UE, and may also be determined by the eNodeB. If the UE receives the CQI indication of the eNodeB as 1, it determines the feedback on the "subframe". Whether the PUCCH lbit information is 0 or 1, if it is 0, it feeds back the short-term PMI, and if it is 1, it feeds back the long-term channel information and the short-term channel information. The values are optional and can be designed separately for FDD and TDD, for example. A value equal to 1 to 10 in FDD, . The selection is made in TDD according to different configurations and subframes. Then, the UE feeds back the long-term CSI feedback information and the short-term PM/information in the “+ uplink subframe. The eNodeB saves the long-term CSI and saves it until the next update, and then uses the received long-term CSI to correct the short-term feedback. PMI. The value of k can be designed according to FDD and TDD respectively. For example, it is always equal to 4 in FDD. The selection in TDD according to different configurations and sub-frames is shown in Table 2.

Step 7. The receiving end determines whether to update the Long Term CSI according to the received control information. So do the next codebook conversion. On the eNodeB side, depending on the feedback method, the currently used Long Term CSI may be ? («.) or codebook conversion, and find the codeword corresponding to the PMI for precoding;

Step 8. Repeat steps 3 through 6.

Embodiment 5

This example specifically includes the following steps.

Step 4: The value of the UE in the subframe "+ feedback PMI information is selectable, and may also be designed according to FDD and TDD respectively, for example, a value equal to 1 to 10 in FDD, according to different in TDD The configuration and the subframe are selected. The eNodeB does not need to know in advance whether the UE feeds back long-term CSI or short-term PMI information through signaling. The eNodeB can know whether the channel information is long-term CSI information or short-term in an explicit or implicit manner. PMI information, for example, an explicit indication: a new UCI format needs to be defined, that is, the number of bits of the PMI is increased by 1 bit on the basis of the original UCI. The eNodeB receives the channel information fed back by the UE. Perform a series of demapping, demodulation, detection, decoding and other processes to determine that the first bit (MSB) of the channel information is 0 to indicate the PMI1 fed back by the UE. Save the last updated long-term CSI to correct the feedback. Short-term PMI to get the next pass The precoding weight of the input.

Step 5: If the long-term CSI needs to be fed back, ^ L> R0, the UE uses the current Long Term CSI to perform codebook conversion, that is, x PM/generates the updated codebook, and the UE can use the maximum SINR criterion in the updated codebook. Select the corresponding PMI. Need to feedback the corresponding PMI and ;

Step 6, the value of the UE in the subframe "+ feedback PMI2 information is selectable, and may also be designed according to FDD and TDD respectively, for example, a value equal to 1 to 10 in FDD, according to different in TDD Configuration and sub-frame selection. The eNodeB does not need to know in advance through signaling whether the UE feeds back long-term CSI or short-term PMI information. The eNodeB can learn whether the channel information is long-term CSI information or short-term PMI information in an explicit or implicit manner. eNodeB only needs to indicate a bit in the DCI Format CQI set of K subframes ₀ 0. For example, an explicit indication is: a new UCI format needs to be defined, that is, the number of bits of the PMI is increased based on the original UCI, and the lbit feedback information is performed after the L eNodeB receives the channel information fed back by the UE. A series of processes, such as demapping, demodulation, detection, and decoding, determine that the first bit (MSB) of the channel information is 0, indicating the PMI1 (short-term channel information) fed back by the UE, and if 1 is the PMI2 (long-term channel) that the UE feeds back Information). The UE updates the long-term CSI with PMI2, and then uses the updated long-term CSI to correct the short-term PMI of the feedback to obtain the precoding weight for the next transmission.

Step 7. The receiving end determines whether to update the Long-Term CSI according to the received control information, so as to perform the next codebook conversion. On the eNodeB side, depending on the feedback method, the currently used Long-Term CSI may be ?(«.) or codebook conversion, and find the codeword corresponding to the PMI for precoding;

Step 8. Repeat steps 3 through 6.

Embodiment 6

This example specifically includes the following steps.

Step 3, assuming a sub-frame. The correlation matrix corresponding to the time is ?(«.), and the correlation matrix corresponding to the time of the subframe w is that the original codebook at the eNodeB and the UE side is PM/. At the time of the subframe w, if it is not necessary to feed back the long-term CSI, L<R0, the UE still uses the previous long-term channel information R(n ₀ )#-codebook conversion, that is, R(n ₀ ) X PMI , that is, only feedback corresponding is needed. ΡΜΙ, do not feedback Long-Term CSI;

Step 4: At this time, the information of the lbit identifier bit added by the UE on the PUCCH of the subframe w is 0, and the eNodeB is notified. In the PMI for the next feedback, the broadband PMI part is a short-term broadband PMI, and the eNodeB receives the UE feedback. After the channel information, the short-term PMI of the last updated long-term CSI de-correction feedback is saved to obtain the pre-encoding weight of the next transmission.

Step 5: If the long-term CSI needs to be fed back, ^ L> R0, the UE uses the current Long Term CSI to perform codebook conversion, that is, x PM/generates the updated codebook, and the UE can use the maximum SINR criterion in the updated codebook. Select the corresponding PM/. Need to feedback the corresponding PM / and

Step 6: At this time, the lbit information added by the UE on the PUCCH of the subframe is 1 to notify the eNodeB. In the PMI for the next feedback, the broadband PMI part is the long-term broadband PMI, and the eNodeB uses the channel information fed back by the UE. The updated long-term CSI is used to correct the short-term PMI of the feedback to obtain the precoding weight for the next transmission.

Step 8. Repeat steps 3 through 6.

2 is a first structural diagram showing a dynamic feedback system for long-term channel state information according to the present invention; In an embodiment, as shown in FIG. 2, the dynamic feedback system of the present example of the long-term channel state information includes a determining unit 20 and a notification unit 21; wherein the determining unit 20 and the notifying unit 21 are disposed in the UE;

The determining unit 20 is configured to determine whether the long-term channel state information CSI needs to be fed back to the network side; the notifying unit 21 is configured to notify the network side of the determination result of the determining unit.

The determining unit 20 further compares the current long-term CSI with the previous long-term CSI, and feeds back the long-term CSI to the network side when the difference exceeds the set threshold, otherwise the long-term CSI is not fed back.

FIG. 3 is a schematic diagram of a second component structure of a dynamic feedback system for long-term channel state information according to the present invention. As shown in FIG. 3, based on the system shown in FIG. 2, the dynamic feedback system of the present long-term channel state information further includes:

The feedback unit 22 is configured to feed back the long-term CSI according to the network side indication when the determining unit determines to feed back the long-term CSI to the network side.

The notification unit 21 notifies the network side whether the network side feedbacks the long-term to the network side through the notification message.

CSI;

Alternatively, the notification unit 21 sends the long-term CSI message to the network side. The message carrying the long-term CSI is set in the message carrying the long-term CSI to indicate whether to carry the long-term CSI.

The notification unit 21 notifies the network side to feed back the long-term CSI to the network side by using the notification message, and the feedback unit carries the long-term CSI in the kth subframe after the subframe that carries the notification message, and feeds back to the network side; , k is a predetermined value determined in advance;

Alternatively, when the determining unit 20 determines to feed back the long-term CSI to the network side, the feedback unit 22 carries the long-term CSI in the k1th subframe after the current frame, and feeds back to the network side; where k1 is a preset determined value. .

The notification unit 21 notifies the network side to feed back the long-term CSI to the network side through the notification message; FIG. 4 is a schematic diagram of the third component structure of the dynamic feedback system for long-term channel state information according to the present invention, such as As shown in FIG. 4, based on the system shown in FIG. 3, the dynamic feedback system of the present long-term channel state information further includes:

The sending unit 23 is configured to be configured, on the network side, to carry information indicating whether the UE feeds back the long-term CSI in the k2th subframe after the subframe that carries the notification message;

When the indication information indicates that the UE feeds back the long-term CSI, the feedback unit carries the long-term CSI in the k3th subframe after the subframe that carries the network side indication information, and feeds back to the network side; and feeds back to the network. Side; wherein k2 and k3 are all predetermined determination values.

It should be understood by those skilled in the art that the dynamic feedback system of the long-term channel state information of the present invention is designed to implement the foregoing dynamic feedback method for long-term channel state information, and the implementation functions of the foregoing processing units can be understood by referring to the related description of the foregoing method. . The functions of the various processing units in the figure can be implemented by a program running on a processor or by a specific logic circuit.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention.

Claims

Claim

A dynamic feedback method for long-term channel state information, characterized in that the method comprises:

2. The method according to claim 1, wherein the UE determines whether it is necessary to feed back the long-term CSI to the network side, which is:

The method according to claim 1 or 2, wherein the method further comprises: when the UE determines to feed back long-term CSI to the network side, feedback or feedback the long-term CSI according to the network side indication.

4. The method according to claim 3, wherein the notifying the network side of the determination result is:

The method according to claim 4, wherein when the UE determines to feed back long-term CSI to the network side, the long-term CSI is fed back as:

Alternatively, when the UE determines to feed back the long-term CSI to the network side, the long-term CSI is carried in the k1th subframe after the current frame, and is fed back to the network side; where k1 is a preset determination value.

The method according to claim 4, wherein, when the UE determines to feed back long-term CSI to the network side, the long-term CSI is fed back according to the network side indication, as follows:

7. The method according to claim 6, wherein the k, kl, k2, and k3 are constant values in a frequency division multiplexed FDD system frame, and in a time division multiplexed TDD system frame, according to the system Determine the frame uplink and downlink configuration mode.

The method according to claim 3, wherein when the UE feeds back the long-term CSI, the short-term precoding matrix indication PMI is also fed back.

A dynamic feedback system for long-term channel state information, wherein the system includes a determining unit and a notification unit; the determining unit and the notification unit are disposed in the UE; wherein the determining unit is configured to determine whether The network side feeds back long-term channel state information CSI;

a notification unit, configured to notify the network side of the determination result of the determining unit.

The system according to claim 9, wherein the determining unit further compares the current long-term CSI with the previous long-term CSI, and feeds back the long-term CSI to the network side when the difference exceeds the set threshold, otherwise Feedback on this long-term CSI.

11. The system of claim 10 or 11, wherein the system further comprises:

And a feedback unit, configured to: when the determining unit determines to feed back the long-term CSI to the network side, feedback or feed back the long-term CSI according to the network side indication.

The system according to claim 11, wherein the notification unit is specifically The notification message notifies the network side whether to feed back the long-term CSI to the network side;

Or the notification unit sends the long-term CSI message to the network side. The message indicating whether the long-term CSI is carried in the message carrying the long-term CSI is set to indicate whether the long-term CSI is carried.

The system according to claim 12, wherein the notification unit notifies the network side to feed back the long-term CSI to the network side by using a notification message, and the feedback unit is the kth child after the subframe carrying the notification message. The long-term CSI is carried in the frame, and is fed back to the network side; where k is a predetermined value determined in advance;

The system according to claim 12, wherein the notification unit notifies the network side to feed back the long-term CSI to the network side by using a notification message;