Method and apparatus for measuring and reporting a rank and a precoding matrix for multiple-input multiple-output communication

Multi-socket input and multi-output communication sequencing and precoding matrix measurement and report method and device

METHOD AND APPARATUS FOR MEASURING AND REPORTING A RANK AND A PRECODING MATRIX FOR MULTIPLE-INPUT MULTIPLE-OUTPUT COMMUNICATION

This application is related to wireless communication.

Spatial multiplexing of a wireless transmit/receive unit (WTRU) using multiple input multiple output (MIMO) communication devices may include determining multiple settings and parameters. The selection of these settings and parameters is expected to improve the quality and reliability of MIMO communication. For example, the WTRU is expected to determine a desired rank that indicates multiple useful transport layers.

For the open-loop spatial multiplexing mode, the reported rank equal to 1 indicates that transmit diversity should be used for MIMO communication, while the reported rank higher than 1 (for example, a value of 2, 3, or 4) indicates that a rank with 2 should be used. , 3, or 4 corresponding layers of large delay cyclic delay diversity (CDD).

For the closed-loop spatial multiplexing mode, the reported rank indicates that closed-loop precoding with the corresponding number of layers (for example, 1, 2, 3, or 4) should be used.

For both open-loop and closed-loop schemes, rank measurement and generation can be performed for MIMO spatial multiplexing.

In addition, determining the precoding matrix index (PMI) may be ideal for MIMO communication by the WTRU.

This application includes several example methods and devices for selecting and reporting rank and PMI for MIMO communications by a WTRU.

A method and device for measuring and reporting rank and/or precoding matrix for multiple input multiple output (MIMO) communication are disclosed. The measurement indicates the condition of the channel A metric, and the rank is selected based on the metric. The metric can be signal to interference and noise ratio (SINR), throughput, block error rate (BLER), system capacity, sum rate, and so on. Calculate the SINR of each resource block group (RBG) of each rank. The data rate of each RBG is calculated based on the SINR of each rank. The total rate of all RBGs is calculated for each rank. At least one rank is selected based on the total rate. At least one precoding matrix may be selected jointly with at least one rank or separately.

<p>100Wireless communication system</p><p>110WTRU</p><p>115, 125 processor</p><p>116,126Receiver</p><p>117, 127 transmitter</p><p>118,128antenna</p><p>120Node B</p><p>130CRNC</p><p>140SRNC</p><p>150Core network</p><p>SRNCService Radio Network Controller</p><p>CRNCControl radio network controller</p><p>WTRUWireless Transmit/Receive Unit</p><p>MIMOMultiple input and multiple output</p><p>PMIPrecoding Matrix Index</p>

The present invention can be understood in more detail from the following descriptions, which are given in the form of examples and can be understood in conjunction with the accompanying drawings, in which: Figure 1 shows the subsequent evolution of the long-term evolution according to the example of the present application (LTE-A) a schematic block diagram of a telecommunications system; Figure 2 is a schematic block diagram showing an example wireless transmit/receive unit (WTRU) and Node B according to the present application; Figure 3 is a schematic block diagram showing the application according to the present application A flowchart of an example method for selecting a rank for multiple-input multiple-output (MIMO) communication performed by a wireless transmit/receive unit (WTRU); A schematic block diagram of an example WTRU of the rank and/or precoding matrix index (PMI) of the WTRU; Figure 5 is a schematic block diagram illustrating an example method for selecting a joint PMI and rank for MIMO communication performed by the WTRU according to the present application Flow chart; and Figure 6 is a flow chart showing an example method of selecting a PMI for MIMO communication by a WTRU according to the present application.

The term "wireless transmit/receive unit (WTRU)" mentioned below includes but is not limited to user equipment (UE), mobile station, fixed or mobile subscriber unit, pager, cellular phone, personal digital assistant (PDA), computer or Any other type of user equipment capable of operating in a wireless environment. The term "base station" mentioned below includes but is not limited to Node B, site controller, access point (AP), or any other type of peripheral equipment capable of operating in a wireless environment.

Figure 1 shows a wireless communication system 100, which includes Multiple WTRUs 110, Node Bs 120, Control Radio Network Controller (CRNC) 130, Serving Radio Network Controller (SRNC) 140, and core network 150. Node B 120 and CRNC 130 may be collectively referred to as UTRAN.

As shown in Figure 1, the WTRU 110 is configured to perform multiple input multiple output (MIMO) communication with the Node B 120, and the Node B 120 communicates with the CRNC 130 and the SRNC 140. Although Figure 1 shows three WTRUs 110, one Node B 120, one CRNC 130, and one SRNC 140, it should be noted that the wireless communication system 100 may include any combination of wireless and wired devices.

Figure 2 is a functional block diagram 200 of the WTRU 110 and the Node B 120 of the wireless communication system 100 of Figure 1. As shown in Figure 2, the WTRU 110 communicates with the Node B 120, both of which may be configured to perform MIMO rank selection and/or PMI selection methods in the WTRU.

In addition to the elements that may be found in a typical WTRU, the WTRU 110 includes a processor 115, a receiver 116, a transmitter 117, and an antenna 118. The processor 115 is configured to perform rank and precoding matrix index (PMI) measurement and selection methods in the WTRU. The receiver 116 and the transmitter 117 communicate with the processor 115. The antenna 118 communicates with the receiver 116 and the transmitter 117 to facilitate the transmission and reception of wireless data.

In addition to the elements that can be found in a typical base station, the Node B 120 includes a processor 125, a receiver 126, a transmitter 127, and an antenna 128. The processor 125 is configured to perform the method of selecting MIMO rank selection and/or PMI selection in the WTRU. The receiver 126 and the transmitter 127 communicate with the processor 125. The antenna 128 communicates with the receiver 126 and the transmitter 127 to facilitate the transmission and reception of wireless data.

Example methods and devices for enhanced MIMO rank and/or PMI measurement and generation are disclosed. Also disclosed are methods and devices for reporting and reporting the selected PMI and indicating the rank information of the selected rank. Some example methods for enhanced MIMO rank and/or PMI measurement and generation are described in detail; however, these examples are not limiting.

During MIMO operation, it is desirable to measure and generate rank information at the receiver or WTRU, and the generated rank information is fed back to the transmitter or eNodeB. Rank measurement and generation can be performed for MIMO spatial multiplexing of open-loop and closed-loop schemes.

For spatial multiplexing, the WTRU determines the rank that indicates multiple useful transmission layers. For the open-loop spatial multiplexing mode, a reported rank equal to 1 indicates that transmission diversity should be used, while a reported rank higher than 1 (for example, 2, 3, or 4) indicates that a rank with 2, 3, or 4 should be used. Corresponding layer large delay cyclic delay diversity (CDD).

For the closed-loop spatial multiplexing mode, the reported rank indicates that closed-loop precoding with the corresponding number of layers (for example, 1, 2, 3, or 4) should be used.

Table 1 summarizes the rank values and the corresponding spatial multiplexing schemes for the open-loop and closed-loop modes.

<tables><img he="818" wi="1996" file="tw201351908a_d0001.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></tables>

The example method and device described below can determine the rank required by the current channel condition according to a predetermined standard. These standards may include metrics based on signal-to-interference and noise ratio (SINR). Other standards such as metrics based on liquidity, block error rate (BLER), system capacity, and rate can also be used.

The measurement and generation of a single rank and the feedback and reporting of a single rank are usually used for large bandwidths (for example, the entire bandwidth). Or, if needed, the same method can be extended to measure and generate multiple ranks for part of the bandwidth. For example, the rank for each subband or the rank of the subband group can be measured and generated. Therefore, the multiple ranks (ie, each rank used for a subband or subband group) can be fed back and reported as part of a multi-band rank or multi-rank measurement and reporting scheme.

For a 2×2 antenna configuration, 1 bit may be required to represent the rank (rank=1 or 2). For a 4×4 antenna configuration, 2 bits may be required to represent the rank (ie, rank=1, 2, 3, and 4).

<tables><img he="285" wi="1784" file="tw201351908a_d0002.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></tables><tables><img he="354" wi="1768" file="tw201351908a_d0003.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></tables>

Figure 3 is a flowchart showing an example method for rank measurement and generation in a WTRU using MIMO communication.

For each of the multiple possible ranks, in step 300, the rank is used to determine the value of the metric indicating the channel condition of the MIMO communication by the WTRU. The metric may be based on one or more measured quantities that indicate channel conditions, such as signal-to-interference and noise ratio (SINR), throughput, block error rate (BLER), system capacity, or sum rate.

Then in step 302, a rank may be selected from a plurality of ranks based on the value of the metric.

For example, if the metric is based on the sum rate of the SINR, it can be based on the channel measurement (such as channel estimation) for each rank, or each subband (ie, resource block (RB)), or its resource block group (RBG). Calculate SINR. The rate of each subband, or RBG, can then be calculated based on the measured SINR for the subband, or RBG. Then, the total sum rate of all subbands is calculated for each rank by adding the subband rates or RBG rates of each subband or RBG. The required rank is selected based on the total sum rate, that is, the rank with the largest sum rate is selected.

It is also possible to determine the value of the metric for each rank by determining the stream sub-value of the metric for each spatial stream used for MIMO communication. These stream sub-values of the metric can then be added to determine the value of the metric.

In the open-loop spatial multiplexing mode, rank 1 corresponds to the transmission diversity scheme, and rank 2 or higher corresponds to the large delay CDD scheme. The sum rate-based metric used here is taken as an example. These example methods include calculating the sum rate for each rank. When calculating the sum rate of rank 1, it is assumed that the transmission diversity scheme is used. Assuming that the transmission diversity scheme is used, the SINR is measured and calculated. When calculating the total rate as rank 2 or higher, a large delay CDD is assumed. Assuming a large delay CDD with a suitable number of layers, the SINR is measured and calculated. The total rate is then calculated based on the calculated SINR for each rank.

The SINR of the jth spatial stream and rank p, that is, the rank PJ RBG SINR, can be expressed as follows in the gth RBG:<maths><img he="280" wi="1863" file="TW201351908A_D0004.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

in<img file="TW201351908A_D0005.tif" wi="122" he="81" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0006.tif" wi="111" he="81" img-format="tif" img-content="character" orientation="portrait" inline="no" />Matrices<img file="TW201351908A_D0007.tif" wi="105" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0008.tif" wi="90" he="79" img-format="tif" img-content="character" orientation="portrait" inline="no" />The (j, k)th element of, and Ns is the number of data streams. P=2 is for 2×2 MIMO, and P=4 is for 4×4 MIMO.

matrix<img file="TW201351908A_D0009.tif" wi="96" he="73" img-format="tif" img-content="character" orientation="portrait" inline="no" />It can be obtained by the following formula:<maths><img he="161" wi="1638" file="TW201351908A_D0010.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where R_InIs the covariance matrix of interference and/or noise.

matrix<img file="TW201351908A_D0011.tif" wi="107" he="75" img-format="tif" img-content="character" orientation="portrait" inline="no" />It can be obtained by the following formula:<maths><img he="128" wi="1652" file="TW201351908A_D0012.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

aisle<img file="TW201351908A_D0013.tif" wi="107" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />Is the average effective channel matrix of rank p for the gth RBG. Large delay CDD can use several matrices. Large delay CDD can use two matrices for two layers, three matrices for three layers, and four matrices for four layers. In addition, the matrix can loop or skip different versions of the matrix to obtain diversity gain. SINR can be averaged across all matrices. The effective channel matrix can be averaged across all matrices to produce a single averaged matrix. The large delay matrix is expressed as follows:<i>T</i>=<i>W</i>(<i>i</i>)<i>D</i>(<i>i</i>)<i>U</i> Equation (4)

Where W(i) is a matrix selected in advance from the precoding codebook. W(i) can be a single matrix or selected from a subset of the precoding codebook. D(i) is a matrix that can depend on the index i of the data symbol. U is a predetermined fixed matrix. If there is a single matrix of W(i), for a 2-transmit antenna configuration, if it is a two-layer, there may be the following two matrices of D(i):<maths><img he="207" wi="1652" file="TW201351908A_D0014.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

If there are multiple matrices of W(i), for a 4-transmit antenna configuration, assume<i>W</i>(<i>i</i>)<img file="TW201351908A_D0015.tif" wi="40" he="49" img-format="tif" img-content="character" orientation="portrait" inline="no" />{<i>C</i>_1,<i>C</i>_2,...,<i>C</i>_<i>L</i>,}, each W(i) and D(i) are combined to generate a new matrix T. If it is three layers, there may be the following three matrices of D(i):<maths><img he="326" wi="1669" file="TW201351908A_D0016.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

If it is four layers, there may be the following four matrices of D(i):<maths><img he="377" wi="1604" file="TW201351908A_D0017.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

The SINR should be averaged across all matrices D(i) and/or W(i) according to the way the matrix W(i) is allocated. Once the matrix T is determined, the effective channel can be calculated accordingly. For each matrix T obtained, the corresponding SINR is calculated. Express the calculated SINR as SINR₁, SINR₂,..., SINR_N. The average SINR can be obtained by the following formula:<maths><img he="165" wi="1517" file="TW201351908A_D0018.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where α_<i>i</i>Is the weight coefficient used for averaging processing.

For two transmitting and one receiving antennas, the SINR for the transmission diversity of the Alamouti scheme using linear minimum mean square error (LMMSE) or maximum rate combining (MRC) can be calculated as follows:<maths><img he="170" wi="1600" file="TW201351908A_D0019.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Similarly, the SINR of transmission diversity using other schemes can be calculated appropriately. Once the SINR is calculated, the total rate can be calculated accordingly:<maths><img he="239" wi="1655" file="TW201351908A_D0020.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

in<img file="TW201351908A_D0021.tif" wi="102" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />It is the rank 1 for transmission diversity and the SINR of one stream. Nt and Nr are the number of antennas at the transmitter and receiver, respectively.

The sum rate of all spatial streams and all subbands or RBGs of rank p can be expressed as follows:<maths><img he="195" wi="1643" file="TW201351908A_D0022.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ng is the number of RBGs in the configured system bandwidth. Select a single rank. Choose the rank that maximizes the total rate of all RBGs,<maths><img he="142" wi="1663" file="TW201351908A_D0023.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ω_pIs the set of allowable ranks, namely Ω_p={1,2,...P}. Typically, for a 4×4 MIMO configuration, Ω_p={1,2,3,4}. The largest rank is equal to four. If p=1 is selected, it indicates that transmit diversity is better. If p=2, 3, or 4 is selected, it indicates that a large delay CDD with two, three, or four layers is better.

In the closed-loop spatial multiplexing mode, the rank x corresponds to the x-layer small delay CDD scheme. The value of x can be 1 to 4 for a 4×4 antenna configuration, and can be 1 or 2 for a 2×2 antenna configuration. The sum rate-based metric is used here as an example. These example methods include calculating the total rate for each rank. Assuming a small delay CDD with a suitable number of layers, the SINR is measured and calculated.

When using the sum rate criterion, it may be desirable to calculate the SINR for each given rank. The sum rate is then calculated based on the calculated SINR. An example process for measuring and generating rank information using and rate standards is described as follows:

Step 1: The bandwidth is divided into several subbands or RBGs of suitable size for measurement and channel averaging purposes. The size can be selected by execution for best performance.

Step 2: For each subband or RBG, calculate the SINR for each rank. The ranks are 1, 2,...P, where P is the maximum rank for a given antenna configuration of the Node B and the WTRU. For example, if the antenna configuration is 4×4, the maximum rank is P=4. SINR is calculated for each spatial stream of each rank.

Step 3: For each subband or RBG, calculate the data rate for each rank based on the calculated SINR in each subband or RBG. If the rank is 1, the data rate is calculated for one layer.

Step 4: Use the previously calculated sum rate of each subband or RBG to calculate the corresponding total sum rate of all subbands or RBGs of each rank 1, 2, ...P.

Step 5: Choose the rank with the highest total sum rate.

Denote the average effective channel matrix of the rank p of the g-th subband or RBG as<img file="TW201351908A_D0024.tif" wi="109" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />. By dividing the channel matrix of the gth subband or RBG<i>H</i>_<i>g</i>Precoding matrix with rank p<img file="TW201351908A_D0025.tif" wi="94" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />Multiply to obtain the average effective channel matrix of the rank p of the g-th subband or RBG<img file="TW201351908A_D0026.tif" wi="111" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />,as follows:<maths><img he="152" wi="1603" file="TW201351908A_D0027.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

The SINR of the j-th spatial stream of rank p in the g-th subband or RBG can be expressed as follows:<maths><img he="207" wi="1580" file="TW201351908A_D0028.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Alternatively, the measurement of the SINR of the j-th spatial stream of rank p in the g-th subband or RBG can be expressed as follows:<maths><img he="239" wi="1557" file="TW201351908A_D0029.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

in<img file="TW201351908A_D0030.tif" wi="124" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0031.tif" wi="116" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />Matrices<img file="TW201351908A_D0032.tif" wi="107" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0033.tif" wi="94" he="75" img-format="tif" img-content="character" orientation="portrait" inline="no" />The (j, k)th element of. matrix<img file="TW201351908A_D0034.tif" wi="96" he="75" img-format="tif" img-content="character" orientation="portrait" inline="no" />It can be obtained by the following formula:<maths><img he="110" wi="1557" file="TW201351908A_D0035.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where R_InIs the covariance matrix of interference and noise. matrix<img file="TW201351908A_D0036.tif" wi="107" he="75" img-format="tif" img-content="character" orientation="portrait" inline="no" />It can be obtained by the following formula:<maths><img he="124" wi="1584" file="TW201351908A_D0037.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

The sum rate of all spatial streams and all subbands or RBGs of rank p can be expressed as follows:<maths><img he="163" wi="1528" file="TW201351908A_D0038.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ng is the number of subbands or RBGs in the configured system bandwidth. A single value of rank is selected for large bandwidth. Choose the rank that maximizes the total rate of all subbands or RBGs, as shown below:<maths><img he="193" wi="1548" file="TW201351908A_D0039.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ω_pIs the set of allowable ranks (i.e. Ω_p={1,2,...,P}). For 4×4 MIMO configuration, Ω_p={1,2,3,4}, and the maximum rank is four.

It is expected that the second rank with the second highest overall rate or the second best metric can also be selected. In this example embodiment, the WTRU may signal the highest total rate or best metric (primary rank) and the second highest total rate or second best metric (secondary rank), respectively The first and second ranks. If the first rank or primary rank may not be desired from the network's point of view, this allows the network to have more flexibility to allocate ranks and schedule resources for data transmission. The network can signal to the WTRU whether it should report the second-best rank. The instruction can be done semi-statically or dynamically.

In another example embodiment, the best K ranks with K highest total rates or K best metrics can be selected. If certain ranks or ranks are not desired from the network's point of view, the WTRU may signal the best K ranks to allow the network the flexibility to allocate the number of data streams and schedule data transmission resource. The network can signal to the WTRU whether it should report the best K ranks. The instruction can be done semi-statically or dynamically.

In step 304, once the desired rank for MIMO communication by the WTRU has been selected, the WTRU sends rank information indicating the selected rank. When reporting rank information, the WTRU expects to report a single instance of the number of useful transport layers.

For each rank reporting interval of the closed-loop spatial multiplexing mode, the WTRU determines the rank from the set of supported rank values for the corresponding antenna configurations of the eNodeB and the WTRU, and reports the number in each rank report.

For each rank reporting interval of the open-loop spatial multiplexing mode, the WTRU determines the rank from the set of supported rank values for the corresponding antenna configurations of the eNodeB and the WTRU, and reports the number in each rank report.

Upon receiving the indication sent in the downlink scheduling grant, the WTRU can use the physical uplink shared channel (PUSCH) to perform aperiodic rank reporting. The minimum reporting interval for aperiodic reports of rank information can be one subframe. The WTRU may be semi-statically configured by higher layers to use one of the reporting modes to feed back rank information on the PUSCH.

The rank value determined by the WTRU may also be used to select a precoding matrix index (PMI) and calculate a channel quality index (CQI). The rank and PMI can be calculated and determined separately. Alternatively, the rank and PMI may be jointly calculated and determined by the WTRU.

The WTRU may be semi-statically configured by higher layers to use one of the reporting modes to periodically return rank information on the physical uplink control channel (PUCCH). It can be expected that the reporting interval of the rank information report is an integer multiple of the broadband CQI and/or PMI reporting period. The offset between the rank information report and the broadband CQI and/or PMI report event is configured, and the same or different offset values can be used for the rank and CQI/PMI report. Both the reporting interval and offset can be configured by higher layers. In the case where the rank information conflicts with the broadband CQI and/or PMI report, it is desirable to transmit the rank information and discard the broadband CQI and/or PMI report. In this case, rank information is considered to be more important than CQI and/or PMI.

In addition to selecting the desired rank for the MIMO communication performed by the WTRU, it is also desirable to select a PMI from among multiple possible PMIs. For MIMO communication using precoding, two example methods for PMI and rank selection-a joint method for rank and PMI selection and generation, and a separate measurement method for rank and PMI selection and generation are described in detail below. In the joint method, the rank and PMI are jointly measured and selected. In the separate measurement method, the rank and PMI are generated separately. Therefore, the combined method uses one-stage processing, while the separate measurement method uses two-stage processing.

In the separate measurement method, for each PMI of a plurality of PMIs, the PMI value using the selected rank metric can be determined. Then, one PMI is selected from a plurality of PMIs based on the measured PMI value, and then the selected PMI is transmitted.

As an example of this method, the above-described example is continued using SINR and sum rate as a metric. Assuming open-loop MIMO communication, the SINR of LMMSE can be calculated for each subband or RBG of the selected rank. The SINR calculated for the subband or RBG is then used to calculate the corresponding data rate for each subband or RBG of each rank. Add the total rates of all subbands or RBGs. Then select the rank that produces the highest total sum rate. Once the rank is selected, the corresponding precoding matrix is selected according to the following process. Given the selected rank, the SINR is calculated for each PMI of each RBG. The total sum rate is calculated by adding the data rates of all RBGs of each precoding matrix. Then the desired precoding matrix is selected for the selected rank in open-loop MIMO as the precoding matrix with the highest sum rate of the selected rank.

The SINR of LMMSE can be calculated for each RBG for a given rank and precoding matrix. The SINR calculated for each RBG can be used to calculate the corresponding data rate of each RBG. The total rate of all RBGs can be calculated for a given rank and precoding matrix. For a given precoding matrix F_iThe SINR of the j-th spatial stream with rank p can be expressed as follows:<maths><img he="289" wi="1530" file="TW201351908A_D0040.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

in<img file="TW201351908A_D0041.tif" wi="124" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0042.tif" wi="111" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />Matrices<img file="TW201351908A_D0043.tif" wi="113" he="75" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0044.tif" wi="98" he="75" img-format="tif" img-content="character" orientation="portrait" inline="no" />The (j, k)th element of. aisle<img file="TW201351908A_D0045.tif" wi="111" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />Is the average channel matrix of rank p for the gth RBG.

matrix<img file="TW201351908A_D0046.tif" wi="96" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />It can be obtained by the following formula:<maths><img he="142" wi="1571" file="TW201351908A_D0047.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where R_InIs the covariance matrix of interference and/or noise.

matrix<img file="TW201351908A_D0048.tif" wi="111" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />It can be obtained from the following:<maths><img he="151" wi="1536" file="TW201351908A_D0049.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

For a single PMI measurement, the sum rate of all spatial streams and all RBGs of rank p and PMI i can be expressed as follows:<maths><img he="170" wi="1449" file="TW201351908A_D0050.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ns is the maximum number of data streams, and Ng is the number of RBGs to be considered.

Choose the PMI that maximizes the total rate of all RBGs,<maths><img he="117" wi="1430" file="TW201351908A_D0051.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

It is expected that the second PMI with the second highest overall rate or the second best metric can also be selected. In this example embodiment, the WTRU may signal the first and second PMIs with the highest total rate or best metric (primary rank) and the second highest total rate or second best metric (secondary rank), respectively . If the primary rank is not desired from a network perspective, this allows the network to have more flexibility to allocate PMI and schedule resources for data transmission. The network can signal to the WTRU whether it should report the second best PMI. The instruction can be done semi-statically or dynamically.

In another example embodiment, the best K PMIs with K highest total rates or K best metrics can be selected. If certain PMIs are not expected from a network perspective, the WTRU can signal the best K PMIs to allow the network flexibility To allocate PMI and schedule resources for data transmission. The network can signal to the WTRU whether it should report the best K PMIs. The instruction can be done semi-statically or dynamically.

For multiple PMI measurements, given rank p, the total rate of all spatial streams of the g-th RB and PMI i can be expressed as follows:<maths><img he="163" wi="1600" file="TW201351908A_D0052.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Choose the PMI that maximizes the rate of each RBG,<maths><img he="117" wi="1539" file="TW201351908A_D0053.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

It is expected that the second PMI with the second highest overall rate or the second best metric can be selected for each offspring or RBG. In this example embodiment, the WTRU may signal the first and second with the highest total rate or best metric (primary rank) and the second highest total rate or second best metric (secondary rank), respectively. PMI. If the main rank for each RBG is not desired from the network's point of view, this allows the network to have more flexibility to allocate PMI and schedule resources for data transmission of each RBG. The network may signal to the WTRU whether it should report the second best PMI for a certain RBG or RBGs. The instruction can be done semi-statically or dynamically.

In another example embodiment, the best K PMIs for each subband or RBG with the K highest total rates or K best metrics for the subband or RBG may be selected. If some PMIs are not expected from a network perspective, the WTRU can signal the best K PMIs for each subband or RBG to allow the network the flexibility to allocate PMIs and schedule them for data transmission H. The network can signal to the WTRU whether it should report the best K PMIs for each subband or RBG. The instruction can be done semi-statically or dynamically.

For MIMO communication using precoding, rank measurement can be performed jointly with precoding matrix selection. Fig. 5 shows an example method of joint selection of selecting a PMI from multiple PMIs of MIMO communication performed by a WTRU and selecting a rank from multiple ranks. Only one stage of joint processing is used in this example method.

For each combination of PMI and rank, in step 500, the value of the metric indicating the channel condition of the MIMO communication performed by the WTRU using the PMI and rank is determined. The metric It can be based on one or more measured quantities that indicate channel conditions, such as signal-to-interference and noise ratio (SINR), throughput, block error rate (BLER), system capacity, or sum rate.

Then in step 502, a combination of PMI and rank may be selected from a plurality of combinations of PMI and rank based on the value of the metric.

For example, if the metric is based on the sum rate of the SINR, then for each combination of PMI and rank, the SINR is calculated based on the channel measurement (e.g., channel estimation). Alternatively, the SINR of each combination of PMI and rank on each subband (ie, resource block (RB)) or resource block group (RBG) can be measured. The rate of each subband or RBG can then be calculated based on the SINR measured for the subband or RBG. The subband rates or RBG rates of each subband or RBG are then added to calculate the total sum rate of all subbands for each combination of PMI and rank. The required combination of PMI and rank is selected based on the total sum rate, that is, the combination of PMI and rank with the largest sum rate is selected.

Using equations 19-21, the sum rate of all spatial streams and all RBGs of rank p and PMI i can be expressed as follows:<maths><img he="170" wi="1572" file="TW201351908A_D0054.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ns is the maximum number of data streams. Choose a single rank. Jointly select the rank and PMI that maximize the total rate of all RBGs,<maths><img he="159" wi="1526" file="TW201351908A_D0055.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

In step 504, once the required PMI and rank combination for MIMO communication by the WTRU has been selected, the WTRU sends the selected PMI and rank information indicating the selected rank.

It is expected that the second combination of the second PMI and rank with the second highest overall rate or the second best metric can also be selected. In this example embodiment, the WTRU may signal the PMI with the highest total rate or best metric (primary rank/PMI) and the second highest total rate or second best metric (secondary rank/PMI), respectively The first and second combination of rank and rank. If the combination of primary rank and PMI is not desired from a network perspective, this allows the network to have more flexibility to allocate rank and PMI and schedule resources for data transmission. For the rank sum of multiple RBGs For the measurement of the combination of PMI, the best combination of rank and PMI is selected for each RBG. In addition, the second best combination of rank and PMI can be selected for each RBG. The network may signal to the WTRU whether it should report the second best combination of rank and PMI. The instruction can be done semi-statically or dynamically.

In another example embodiment, the best K rank and PMI combination with the K highest total rates or K best metrics can be selected. If a certain rank and/or PMI is not desired from the network's point of view, the WTRU may signal the best combination of K ranks and PMI to allow the network the flexibility to allocate the number of data streams and PMI And schedule the resources used for data transmission. The network can signal to the WTRU whether it should report the best combination of K ranks and PMI. The instruction can be done semi-statically or dynamically.

Upon receiving the indication sent in the downlink scheduling grant, the WTRU can use PUSCH to perform aperiodic PMI and rank reporting. The minimum reporting interval for aperiodic reporting of PMI and rank information can be one subframe. The WTRU may be semi-statically configured by higher layers to use one of the reporting modes to feed back PMI and rank information on the PUSCH. The values of PMI and rank determined by the WTRU may also be used to calculate CQI.

The WTRU may be semi-statically configured by a higher layer to use one of the reporting modes to periodically return PMI and rank information on the PUCCH. It can be expected that the reporting interval of the rank information report is an integer multiple of the broadband CQI and/or PMI reporting period. The offset between the rank information report and the broadband CQI and/or PMI report event is configured, and the same or different offset values can be used for the rank and CQI/PMI report. Both the reporting interval and offset can be configured by higher layers. In the case where the rank information conflicts with the broadband CQI and/or PMI report, it is desirable to send the rank information and discard the broadband CQI and/or PMI report.

Figure 4 shows an example WTRU configured for MIMO communication, which may be used to perform different methods of this application. This example WTRU includes: a receiver 400; a channel condition processor 404, which is coupled to the receiver 400; a selection processor 406, which is coupled to the channel condition processor 404; and a transmitter 408, The transmitter 408 is coupled with the selection processor 406; and the antenna array 402 is coupled with the receiver 400 and the transmitter 408.

The receiver 400 is configured to receive the signal detected by the antenna array 402, and then The signal is then relayed to the channel status processor 404. The channel condition processor 404 is configured to determine the value of the metric indicating the channel condition based on the received signal. The channel status processor 404 may be configured such that the value of the metric is determined for each of a plurality of ranks, a plurality of PMIs, or a combination of a plurality of PMIs and ranks. The metric may be based on at least one of SINR, throughput, BLER, system capacity, or sum rate.

In addition, for each of the multiple PMIs and/or ranks, the channel status processor 404 may be configured to determine the value of the metric, and the value of the metric determines the metric of each spatial stream (or RBG) used for MIMO communication. The flow sub-value (or RBG sub-value), and then the measured flow sub-value (or RBG sub-value) is added to determine the value of the metric.

The selection processor 406 (or the joint selection processor in the case of the joint selection of PMI and rank) is configured to select the PMI based on the value of the metric determined by the channel condition processor 404 and/or select from multiple ranks rank.

The transmitter 408 is configured to transmit the selected PMI and/or rank information indicating the selected rank.

It should be noted that the channel status processor 404, the selection processor 406, and the transmitter 408 may be configured to execute the example method disclosed above in relation to FIG. 3 or FIG. 5, or the example method disclosed below in relation to FIG. 6 Any one of the method-related channel status determination, selection, and transmission process.

Figure 6 shows an example method for selecting a PMI from multiple PMIs for MIMO communication performed by the WTRU. This example method may include a process for PMI selection and generation for advanced WTRUs, such as advanced WTRUs using minimum mean square error (MMSE) continuous interference cancellation (SIC). An example process for PMI selection and generation for different MIMO precoding configurations (such as wideband precoding, M subband precoding, and per subband precoding) may also be included.

PMI selection can be based on different criteria. Several criteria selected by PMI include criteria based on minimum mean square error (MMSE), based on channel capacity, and based on correlation.

For each PMI in the plurality of PMIs, in step 600, the value of the metric indicating the channel condition of the MIMO communication performed by the WTRU using the PMI is determined. The metric may be based on one or more measured quantities that indicate channel conditions, such as signal-to-interference and noise ratio (SINR), throughput, block error rate (BLER), system capacity, or sum rate.

Then in step 602, a PMI can be selected from a plurality of PMIs based on the value of the metric.

For example, if the metric is based on the sum rate of the SINR, the SINR can be calculated for each PMI based on the channel measurement (e.g., channel estimation). Alternatively, the SINR of each PMI on each subband (ie, resource block (RB)) or resource block group (RBG) can be measured. The rate of each subband, or RBG, can then be calculated based on the measured SINR of the subband, or RBG. Then the sub-band rate or RBG rate of each sub-band or RBG is added to calculate the total sum rate of all sub-bands for each PMI. The required PMI is selected based on the total sum rate, that is, the PMI with the largest sum rate is selected.

In the following description of the example of the channel condition determination and selection step according to Figure 6, Ω is used to represent a codebook or a set of candidate precoding matrices F, such as Discrete Fourier Transform (DFT) or Haushold ( Householder) (HH) matrix. H and H_effThey are used to represent channel status information (CSI) and effective CSI, respectively. The effective CSI is generated from the specific precoding matrix F by the following equation:<i>H</i>_<i>eff</i>=<i>HF</i> Equation (28)

When using the MMSE linear detection receiver, according to the MMSE detection form used, the mean square error (MSE) can be expressed according to the following equation:<maths><img he="166" wi="1567" file="TW201351908A_D0056.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

or<maths><img he="174" wi="1510" file="TW201351908A_D0057.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where ρ is the signal-to-noise ratio.

In one method, if the corresponding MSE is minimized, PMI is selected, that is, if the trace of the MSE in equation (2) is minimized, then F is selected according to the following equation:<maths><img he="121" wi="1571" file="TW201351908A_D0058.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Given H and F (or effective channel H_effThe channel capacity of) can be expressed by the following formula for two different linear minimum mean square error (LMMSE) forms:<maths><img he="181" wi="1609" file="TW201351908A_D0059.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

or<maths><img he="170" wi="1565" file="TW201351908A_D0060.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

In another method, PMI is selected to maximize the channel capacity, that is, F is selected by the following rules:<maths><img he="128" wi="1539" file="TW201351908A_D0061.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Yet another method for selecting PMI includes estimating the channel response H and performing singular value decomposition (SVD) on the estimated H to obtain the precoding matrix V. For MIMO transmission of N streams, 1<img file="TW201351908A_D0062.tif" wi="40" he="53" img-format="tif" img-content="character" orientation="portrait" inline="no" /><i>N</i><img file="TW201351908A_D0063.tif" wi="47" he="51" img-format="tif" img-content="character" orientation="portrait" inline="no" /><i>N</i>_<i>t</i>, Suppose A is the V sub-matrix used to pre-encode N streams. Also suppose B_iIs the possible combination of N column vectors of matrix F. Search for all possible combinations of the column vectors of F, that is, for all possible B_iPerform a search and find the maximization A and B during the search_iThe sum of the norm of the inner product or the correlation of a B_i,thereby:<maths><img he="181" wi="1539" file="TW201351908A_D0064.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

The precoding matrix selection can also be performed based on the sum rate standard. The SINR for LMMSE can be calculated for each precoding matrix F. According to the type of LMMSE used, the precoding matrix F_iThe SINR of the j-th spatial stream can be expressed as follows:<maths><img he="223" wi="1622" file="TW201351908A_D0065.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

or<maths><img he="215" wi="1580" file="TW201351908A_D0066.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where H is the average channel matrix in the resource block group (RBG).

I-th precoding matrix F_iThe sum rate of the gth RBG can be expressed as follows:<maths><img he="177" wi="1558" file="TW201351908A_D0067.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ns is the number of data streams. Select the maximum and rate precoding matrix F_i： <maths><img he="151" wi="1562" file="TW201351908A_D0068.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Can be for each precoding matrix F_iCalculate the SINR of LMMSE with each spatial stream. Precoding matrix F_iThe SINR of the j-th spatial stream can be expressed as follows:<maths><img he="254" wi="1491" file="TW201351908A_D0069.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

in<i>w</i>_<i>g,j,k</i>and<i>z</i>_<i>g,j,k</i>Matrices<i>W</i>_<i>g</i>and<i>Z</i>_<i>g</i>The (j, k)th element of.

matrix<i>Z</i>_<i>g</i>It can be obtained by the following formula:<maths><img he="144" wi="1517" file="TW201351908A_D0070.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

in<i>R</i>_<i>In</i>Is the covariance matrix of interference and noise.

matrix<i>W</i>_<i>g</i>It can be obtained by the following formula:<i>W</i>_<i>g</i>=<i>Z</i>_<i>g</i><i>H</i>_<i>eff,g</i> Equation (39)

I-th precoding matrix F_iAnd the effective channel matrix of the gth RBG<i>H</i>_<i>eff,g</i>It can be obtained by the following formula:<i>H</i>_<i>eff,g</i>=<i>H</i>_<i>g</i><i>F</i>_<i>i</i> Equation (40)

I-th precoding matrix F_iThe sum rate of the gth RBG can be expressed as follows:<maths><img he="166" wi="1555" file="TW201351908A_D0071.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ns is the number of data streams. Select the precoding F that maximizes the sum rate of the g-th RBG_i： <maths><img he="129" wi="1567" file="TW201351908A_D0072.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Can be for each precoding moment F_iCalculate the SINR of MMSE-SIC with each spatial stream j. The SINR of the first detected codeword is the same as the SINR of the MMSE. The precoding matrix F is shown in equation (37)_iThe SINR of the j-th spatial stream.

The backward detection SINR depends on the detection order of the SIC. For MIMO 2×2, if the detection starts with codeword 1 (CW1 or the first layer), equation (37) is used to calculate the SINR of CW1, where j=1 is assumed.

For CW2 (or the second layer), take interference cancellation into account. Preferably, it is assumed that SIC ideally removes the interference generated from CW1. The equation is reduced to rank 1 transmission instead of rank 2 transmission. I-th precoding matrix F_iAnd the effective channel matrix of the gth RBG<i>H</i>_<i>eff,g</i>Becomes 2×1 matrix, and can be obtained by the following formula:<maths><img he="121" wi="1553" file="TW201351908A_D0073.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where F_i(:, 2) Representation matrix F_iThe second column vector.

Matrix with 1×2 dimensions<img file="TW201351908A_D0074.tif" wi="92" he="79" img-format="tif" img-content="character" orientation="portrait" inline="no" />It can be obtained by the following formula:<maths><img he="166" wi="1603" file="TW201351908A_D0075.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Scalar<img file="TW201351908A_D0076.tif" wi="107" he="79" img-format="tif" img-content="character" orientation="portrait" inline="no" />It can be obtained by the following formula:<maths><img he="144" wi="1519" file="TW201351908A_D0077.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Precoding matrix F_iThe SINR of the second spatial stream can be expressed as follows:<maths><img he="303" wi="1527" file="TW201351908A_D0078.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

in<img file="TW201351908A_D0079.tif" wi="120" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0080.tif" wi="111" he="81" img-format="tif" img-content="character" orientation="portrait" inline="no" />Matrices<img file="TW201351908A_D0081.tif" wi="107" he="77" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0082.tif" wi="92" he="79" img-format="tif" img-content="character" orientation="portrait" inline="no" />The (j, k)th element of.

I-th matrix F_iThe sum rate of the gth RBG can be expressed as follows:<maths><img he="166" wi="1553" file="TW201351908A_D0083.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ns is the number of data streams. Select the precoding matrix F that maximizes the sum rate of the g-th RBG_i： <maths><img he="125" wi="1565" file="TW201351908A_D0084.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

It can be seen from equations (37), (46), and (47) that the SIC detection sequence affects the calculated SINR of the first CW and the second CW. If the first CW is processed first, and then the second CW is detected, there are<img file="TW201351908A_D0085.tif" wi="79" he="75" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0086.tif" wi="85" he="75" img-format="tif" img-content="character" orientation="portrait" inline="no" />. If the second CW is processed first, and then the first CW is detected, there are<img file="TW201351908A_D0087.tif" wi="90" he="104" img-format="tif" img-content="character" orientation="portrait" inline="no" />and<img file="TW201351908A_D0088.tif" wi="100" he="104" img-format="tif" img-content="character" orientation="portrait" inline="no" />. This is because the CW processed first uses only MMSE detection. SINR depends on the CW signal strength and the interference generated by other CWs. Choosing a CW that is different from the CW processed first will result in a different SINR. In addition, the SINR of a CW takes SIC into consideration, and thus only depends on its own signal strength and noise level, and does not depend on the interference of other CWs. This is based on the backward detection of SINR.

calculate[<img file="TW201351908A_D0089.tif" wi="83" he="109" img-format="tif" img-content="character" orientation="portrait" inline="no" />,<img file="TW201351908A_D0090.tif" wi="83" he="113" img-format="tif" img-content="character" orientation="portrait" inline="no" />]and[<img file="TW201351908A_D0091.tif" wi="93" he="112" img-format="tif" img-content="character" orientation="portrait" inline="no" />,<img file="TW201351908A_D0092.tif" wi="95" he="109" img-format="tif" img-content="character" orientation="portrait" inline="no" />] Of each F_iThe sum rate is as follows:<maths><img he="211" wi="1559" file="TW201351908A_D0093.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

as well as<maths><img he="167" wi="1559" file="TW201351908A_D0094.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

The sum rate Π calculated in equations (49) and (50)_<i>g</i>(<i>F</i>_<i>i</i>)and<img file="TW201351908A_D0095.tif" wi="204" he="102" img-format="tif" img-content="character" orientation="portrait" inline="no" />Because SIC's detection order is different, they are different from each other. The SIC detection order can be determined based on the criterion that maximizes the sum rate in equations (49) and (50). If the total rate of a given detection sequence is the largest, then the detection sequence is selected. The previously described method can be used to generate rank, PMI, CQI, or a combination of rank, PMI, and/or CQI.

In wideband precoding, a single precoding matrix is generated for the entire bandwidth. Choose the size of the RBG so that the average channel response is good enough. Once the RBG size is determined, the entire bandwidth can be divided into multiple RBGs, called N_G. From this, the rate in each RBG can be calculated. I-th precoding matrix F_iThe sum rate of the gth RBG can be expressed as follows:<maths><img he="151" wi="1525" file="TW201351908A_D0096.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ns is the number of data streams. The total rate of the entire bandwidth is the sum of all respective rates of each RBG:<maths><img he="166" wi="1491" file="TW201351908A_D0097.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where N_GIs the number of RBGs for the entire bandwidth. Select the precoding matrix F that maximizes the sum rate of the entire bandwidth_i： <maths><img he="143" wi="1631" file="TW201351908A_D0098.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

In M-subband precoding, a single precoding matrix is generated for M better subbands in a given bandwidth. The given bandwidth is a set of subbands S, that is, M better subbands in a set of subbands S. M-subband precoding corresponds to the subband precoding and feedback selected by the WTRU. Choose the RBG size so that the average channel response is good enough. Once the RBG size is determined, each subband can be divided into one or more RBGs, called<i>G</i>_S. From this, the rate in each RBG can be calculated. Suppose there are<i>N</i>_<i>S</i>Sub-bands. When<i>N</i>_<i>S</i>When M subbands are selected among the subbands, there are a total of<i>Q</i>A combination of M-subbands, such as:<maths><img he="147" wi="1570" file="TW201351908A_D0099.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

The i-th precoding matrix F in the q-th combination_iThe sum rate of the gth RBG can be expressed as follows:<maths><img he="158" wi="1548" file="TW201351908A_D0100.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where N_SIs the number of data streams. The rate of each of the Q combinations (ie, each M-subband) is the sum of all individual rates of each RBG in each subband and the sum of all M subbands. For the qth combination, the rate is calculated as:<maths><img he="174" wi="1514" file="TW201351908A_D0101.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Denote the set of Q combinations as Ω_<i>Q</i>, Precoding matrix F_iAnd the qth combination that maximizes the total rate of the M subbands are jointly selected according to the following formula:<maths><img he="170" wi="1510" file="TW201351908A_D0102.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

In each subband precoding, a single precoding matrix is generated for each subband in a set of subbands S. Each subband precoding corresponds to multiple precoding feedback. Choose the size of the RBG so that the average channel response is good enough. Once the RBG size is determined, each subband can be divided into one or more RBGs, called<i>G</i>_<i>S</i>. Therefore, the rate in each RBG can be calculated. Suppose there are<i>N</i>_<i>S</i>Sub-bands. I-th precoding matrix F_iThe sum rate of the gth RBG can be expressed as follows:<maths><img he="181" wi="1574" file="TW201351908A_D0103.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Where Ns is the number of data streams. The rate of each subband is the sum of all the individual rates of the RBG in each subband:<maths><img he="174" wi="1559" file="TW201351908A_D0104.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

The precoding matrix that maximizes the rate of the subband is selected according to the following formula:<maths><img he="174" wi="1555" file="TW201351908A_D0105.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

An example codebook containing eight DFT matrices with different phase shifts is as follows:<maths><img he="1725" wi="1888" file="TW201351908A_D0106.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></maths>

Table 3 below shows an example codebook:<tables><img he="2761" wi="2000" file="tw201351908a_d0107.tif" img-content="drawing" img-format="tif" orientation="portrait" inline="no" /></tables>

In step 604, once the PMI required for the MIMO communication by the WTRU has been selected, the WTRU transmits the selected PMI. If M-subband precoding is used, sending the selected PMI includes sending the selected M subbands and the selected PMI combination. And if a separate PMI is selected for each subband, sending the selected PMI includes sending the subband PMI selected for each subband.

Although the features and elements of the present invention are described in specific combinations, each feature or element can be used alone without other features and elements, or used in various situations with or without other features and elements. . The method or flowchart provided here can be implemented in a computer program, software, or firmware executed by a general-purpose computer or processor, where the computer program, software, or firmware is included in a computer-readable storage medium in a tangible manner of. Examples of computer-readable storage media include read-only memory (ROM), random access memory (RAM), registers, cache memory, semiconductor memory devices, internal hard disks, and removable disks. Magnetic media, magneto-optical media, and optical media such as CD-ROM disks and digital versatile discs (DVD).

For example, suitable processors include: general-purpose processors, special-purpose processors, traditional processors, digital signal processors (DSP), multiple microprocessors, one or more microprocessors associated with DSP cores, Controller, microcontroller, dedicated integrated circuit (ASIC), field programmable gate array (FPGA) circuit, any kind of integrated circuit (IC) and/or state machine.

The processor associated with the software can be used to implement a radio frequency transceiver for use in a wireless transmit and receive unit (WTRU), user equipment (UE), terminal, base station, radio network controller (RNC), or any host computer To be used in. WTRU can be used in combination with modules implemented in hardware and/or software, such as cameras, camera modules, video phones, speaker phones, vibrating devices, speakers, microphones, TV transceivers, hands-free earphones, keyboards, bluetooth Bud® Module, Frequency Modulation (FM) Radio Unit, Liquid Crystal Display (LCD) Display Unit, Organic Light Emitting Diode (OLED) Display Unit, Digital Music Player, Media Player, Video Game Module, Internet Browser and/or any wireless local area network (WLAN) or ultra-wideband (UWB) module.

Example

Example 1. A method of selecting at least one rank from multiple ranks used for multiple-input multiple-output (MIMO) communication performed by a wireless transmit/receive unit (WTRU).

Example 2. As in the method described in embodiment 1, the method includes: for each rank of the plurality of ranks, using the rank to determine the channel state of the MIMO communication performed by the WTRU; Selecting the at least one rank from the plurality of ranks based on the value of the metric; and sending rank information indicating at least one selected rank for MIMO communication performed by the WTRU.

Example 3. The method according to embodiment 2, wherein determining the value of the metric of each rank of the plurality of ranks includes: for each spatial stream used for the MIMO communication, determining the stream sub-value of the metric; and the metric Add the flow sub-values to determine the value of the metric.

Example 4. The method according to embodiment 2, wherein determining the value of each rank metric in the plurality of ranks includes: for each resource block group (RBG) used for the MIMO communication, determining the RBG sub-value of the metric; and The RBG sub-values of the metric are added to determine the value of the metric.

Example 5. The method according to one of embodiments 2-4, wherein: the metric is a sum rate; the at least one rank is a rank; and selecting the at least one rank from the plurality of ranks includes selecting the one with the highest sum rate A rank.

Example 6. The method of embodiment 5, wherein determining each rank sum rate of the plurality of ranks includes: measuring the RBG signal-to-interference and noise ratio (SINR) of each RBG used by the WTRU; In the RBG SINR, the RBG rate of each RBG used by the WTRU is calculated; and the RBG rate is added to determine the sum rate of the rank.

Example 7. The method according to embodiment 6, wherein: MIMO communication uses open-loop spatial multiplexing; and determining the sum rate of each of the multiple ranks includes: for rank 1: measuring each of the WTRUs using transmit diversity The rank 1 RBG SINR of the RBG; calculate the rank 1 RBG rate of each RBG used by the WTRU from the measured rank 1 RBG SINR of the RBG; and add the rank 1 RBG rate to determine the rank 1 and rate; and Rank P, where P is an integer greater than 1: For each spatial stream J of rank P used for MIMO open-loop spatial multiplexing, where J is an integer from 1 to P, the measurement uses large delay cyclic delay diversity (CDD) The rank PJ RBG SINR of each RBG used by the WTRU; calculate the rank PJ RBG rate of each RBG used by the WTRU from the measured rank PJ RBG SINR of each RBG; and add the rank PJ RBG rate to determine The rank P and rate.

Example 8. The method according to embodiment 7, wherein for the J-th spatial stream of the P-th rank, measuring the rank PJ RBG SINR of each RBG used by the WTRU includes: For the MIMO open loop space of the rank P of the rank P For each precoding of the work, the measurement rank PJ RBG SINR And averaging the precoding sub-values of the rank PJ RBG SINR to determine the rank PJ RBG SINR.

Example 9. The method of embodiment 6, wherein: MIMO communication uses closed-loop spatial multiplexing, and determining the sum rate of each of the plurality of ranks includes: for each layer used by the WTRU for the rank: measurement use The layer RBG SINR of each RBG used by the WTRU with small delay CDD; calculate the layer RBG rate of the RBG used by the WTRU from the measured layer RBG SINR of each RBG; add the layer RBG rates to determine the layer rate ; And the layer rate of the layer used by the WTRU for the rank is added to determine the sum rate of the rank.

Example 10. The method of one of embodiments 5-9, wherein: the MIMO communication uses continuous interference cancellation (SIC) based on a set of codewords, the number of codewords is equal to the layer used by the WTRU for each rank Determining the sum rate of each of the multiple ranks includes: determining the SIC order and rate of the rank for each order of the codewords in the set of codewords by the following process: measuring the SIC used by the WTRU The signal-to-interference and noise ratio (SINR) of each layer; calculate the layer rate of each layer used by the WTRU from the measured SINR of the layer; and add the layer rates to determine the SIC sequential rate; select the highest of the rank SIC sequence rate to determine the sum rate of the rank and the selected sequence of codewords in a set of codewords for that rank; and send the selected order of codewords in a set of codewords of the selected rank order.

Example 11. The method of one of embodiments 2-4, wherein: the metric is the sum rate; the at least one rank is two ranks; and selecting at least one rank from the plurality of ranks includes selecting the two highest sums The two ranks of the rate.

Example 12. The method according to one of the embodiments 2-11, wherein the rank information is transmitted on the physical uplink shared channel (PUSCH).

Example 13. The method of embodiment 12, wherein the rank information is sent aperiodically.

Example 14. The method according to one of the embodiments 2-11, wherein: the rank information is transmitted on the physical uplink control channel (PUCCH); and the reporting interval of the rank information is an integer multiple of one of the following: the The channel quality indicator (CQI) reporting period of the WTRU; or the precoding matrix index (PMI) reporting period of the WTRU.

Example 15. As in the method described in one of the embodiments 2-14, the method The method further includes: for each PMI of the plurality of PMIs, using each of the at least one selected ranks to determine the PMI value of the metric; and based on the metric of each of the at least one selected ranks Select a PMI from the multiple PMIs; and send the selected PMI for each of the at least one selected ranks.

Example 16. A joint selection method for MIMO communication performed by the WTRU is to select at least one PMI from multiple PMIs and at least one rank from multiple ranks.

Example 17. As the method described in embodiment 16, the method includes: for each combination of PMI and rank, using the PMI and rank to determine the value of the metric, and the value of the metric indicates the channel status of the MIMO communication performed by the WTRU; The value of the metric selects at least one combination of PMI and rank; and sends information indicating the PMI and rank of each combination of at least one selected combination of rank and PMI for the MIMO communication performed by the WTRU.

Example 18. The method of embodiment 17, wherein determining the value of the metric for each combination of PMI and rank includes: for each RBG used for the MIMO communication, determining the RBG sub-value of the metric; and determining the RBG sub-value of the metric The values are added to determine the value of the metric.

Example 19. The method according to one of the embodiments 17 or 18, wherein: the metric is the sum rate; the combination of the at least one PMI and the rank is a combination of the PMI and the rank; and selecting the combination of the at least one PMI and the rank includes Choose a combination of PMI and rank with the highest sum rate.

Example 20. The method of embodiment 19, wherein determining the sum rate of each combination of PMI and rank includes: measuring the RBG SINR of each RBG used by the WTRU; calculating the RBG SINR used by the WTRU from the measured RBG SINR of the RBG The RBG rate of each RBG; and the RBG rate is added to determine the combined sum rate of the PMI and rank.

Example 21. The method according to one of embodiments 19 or 20, wherein: the MIMO communication uses SIC based on a set of codewords, the number of codewords is equal to the number of layers used by the WTRU for each rank; The sum rate of the combination of PMI and rank includes: the SIC order and rate of the combination of PMI and rank are determined by the following process for each ordering of the codewords in the set of codewords: each layer used by the WTRU using SIC is measured Calculate the layer rate of each layer used by the WTRU from the measured SINR of the layer; and add the layer rates to determine the SIC sequence rate; select the highest SIC sequence rate of the combination of PMI and rank to determine The sum rate of the combination of the PMI and the rank and the selected ordering of the codewords in the set of codewords of the combination of the PMI and the rank; and the code sent in the set of codewords of the combination of the selected PMI and rank The selected sort of words.

Example 22. The method according to one of the embodiments 17 or 18, wherein: the metric is the sum rate; the combination of the at least one PMI and the rank is a combination of two PMIs and the rank; and selecting at least one combination of the PMI and the rank includes The two combinations of PMI and rank with the two highest sum rates are selected.

Example 23. The method according to one of the embodiments 17-22, wherein the PMI and rank information are transmitted on the PUSCH.

Example 24. The method of embodiment 23, wherein the PMI and rank information are sent aperiodically.

Example 25. The method according to one of the embodiments 17-22, wherein: the PMI and rank information are transmitted on PUCCH; and the reporting interval of the rank information is an integer multiple of one of the following: CQI report of the WTRU Period; or the PMI reporting period of the WTRU.

Example 26. A method of selecting at least one PMI from a plurality of PMIs used for MIMO communication by the WTRU.

Example 27. As the method described in embodiment 26, the method includes: for each PMI in the plurality of PMIs, using the PMI to determine a value of a metric, the value of the metric indicating the channel status of the MIMO communication performed by the WTRU; based on the The value of the metric selects the at least one PMI from the plurality of PMIs; and transmits the at least one selected PMI for MIMO communication by the WTRU.

Example 28. The method of embodiment 27, wherein determining the value of the metric of each PMI of the plurality of PMIs includes: for each spatial stream used for the MIMO communication, determining the stream sub-value of the metric; and the metric Add the flow sub-values to determine the value of the metric.

Example 29. The method of embodiment 27, wherein determining the value of the metric of each PMI in the plurality of PMIs includes: for each RBG used for the MIMO communication, determining the RBG sub-value of the metric; and The RBG sub-values are added to determine the value of the metric.

Example 30. The method of one of embodiments 27-29, wherein: the metric is a sum rate; the at least one PMI is a PMI; and selecting the at least one PMI from the plurality of PMIs includes selecting the one with the highest sum rate A PMI.

Example 31. The method of embodiment 30, wherein determining the sum rate of each PMI in the plurality of PMIs includes: measuring the RBG SINR of each RBG used by the WTRU; calculating the WTRU from the measured RBG SINR of the RBG The RBG rate of each RBG used; and the RBG rate is added to determine the sum rate of the PMI.

Example 32. The method according to one of the embodiments 30 or 31, wherein: the MIMO communication uses SIC based on a set of codewords, and the number of codewords is equal to the number of layers used by the WTRU; determining the sum rate of each PMI includes: Determine the SIC order and rate of the PMI for each ordering of the codewords in the set of codewords by the following process: measure the SINR of each layer used by the WTRU using SIC; calculate the WTRU usage from the measured SINR of the layer The layer rate of each layer; and the layer rate is added to determine the SIC sequence rate; the highest SIC sequence rate of the PMI is selected to determine the sum rate of the PMI and the codeword in a set of codewords for the PMI And the selected order of codewords sent in a set of codewords of a selected PMI.

Example 33. The method of one of embodiments 27-29, wherein: the metric is a sum rate; the at least one PMI is two PMIs; and selecting at least one PMI from the plurality of PMIs includes selecting two highest sums Rate two PMI.

Example 34. As the method described in one of the embodiments 27-33, the method further includes: dividing the bandwidth of the MIMO communication performed by the WTRU into S subbands, where S is a predetermined integer greater than 1; Determine M better subbands from the S subbands of MIMO communication, where M is a predetermined positive integer less than S; where: the WTRU uses M-subband precoding; the metric is the sum rate; the multiple PMIs are determined The sum rate of each PMI in includes: measuring the RBG SINR of each RBG used by the WTRU; calculating the RBG rate of the RBG used by the WTRU from the measured RBG SINR of each RBG; and for the M subbands For each combination, the RBG rates of the RBGs in the combination of the M subbands are added to determine the sum rate of the combination of the M subbands using the PMI; selecting the at least one PMI from the plurality of PMIs includes selecting the PMI A combination of M subbands and at least one PMI with the highest sum rate; and sending at least one selected PMI includes sending the selected M subbands and to One less selected PMI combination.

Example 35. As the method described in one of the embodiments 27-33, the method further includes: dividing the bandwidth of the MIMO communication performed by the WTRU into S subbands, where S is a predetermined integer greater than 1, where: the metric is and Rate; determining the sum rate of each of the multiple PMIs includes: measuring the RBG SINR of each RBG used by the WTRU; calculating the RBG rate of the RBG used by the WTRU from the measured RBG SINR of each RBG ; And for each subband, add the RBG rate of the RBG in the subband to determine the subband and rate of the subband using the PMI; selecting the at least one PMI from the plurality of PMI includes, for each Selecting at least one subband PMI with the highest sum rate; and sending at least one selected PMI includes sending at least one selected subband PMI for each subband.

Example 36. The method of embodiments 2-15, 17-25, and 27-35, wherein the metric is at least one of SINR, throughput, block error rate (BLER), system capacity, or sum rate.

Example 37. A WTRU configured for MIMO communication using the method described in any one of embodiments 1-36.