# Method and apparatus for using factorised precoding

FIELD: radio engineering, communication.

SUBSTANCE: invention relates to wireless communication systems and is intended to improve user equipment (UE) channel state information (CSI) feedback due to that a precoder part of a CSI feedback report contains factorised precoder feedback. In one or more such embodiments, the factorised precoder feedback corresponds to at least two precoder matrices, including a recommended "conversion" precoder matrix and a recommended "tuning" precoder matrix. The recommended conversion precoder matrix restricts the number of channel dimensions considered by the recommended tuning precoder matrix and, in turn, the recommended tuning precoder matrix matches the recommended precoder matrix to an effective channel that is defined in part by said recommended conversion precoder matrix.

EFFECT: improving user equipment channel state information feedback.

42 cl, 7 dwg

RELATED APPLICATIONS

This application claims the priority of provisional patent application U.S. No. 61/264495, filed November 25, 2009.

The technical FIELD

This invention in General relates to pre-coding signal transmission and, in particular, refers to the use of well pre-encoding.

The LEVEL of TECHNOLOGY

Mnogoaktnye methods can significantly increase the data transmission speed and reliability of wireless communication systems. Performance, in particular, is improved if both the transmitter and receiver are equipped with multiple antennas, which leads to a communication channel with multiple inputs and multiple outputs (MIMO). Such systems and/or related methods are usually referred to as MIMO.

The LTE standard is currently being developed with enhanced support MIMO. The basic component in LTE is to support deployments of MIMO antennas and associated with MIMO methods. The current working assumption in LTE advanced is mode support 8-layer spatial multiplexing with potential-dependent channel pre-coding. The spatial multiplexing is designed for high data rates in favorable channel conditions. According to this multiplexing carrying information symb the local vector s is multiplied by the matrix W_{
NT×r}N_{T}×r pre-coder, which is used for distribution of energy transfer in the subspace of N_{T}-d (which corresponds to N_{T}antenna ports) of a vector space.

This matrix is pre-coder is usually chosen from a codebook of possible matrices pre-coder and generally indicated by the indicator matrix pre-coder (PMI), which defines a unique matrix pre-coder in the code book. If the matrix is pre-coder is limited orthonormal columns, then the design of the codebook matrices pre-coder corresponds to the task of unifying the subspaces of Grassman. Every r symbols in s correspond to a certain level, and r is called the rank of the transmission. Thus, spatial multiplexing, as many symbols can be transmitted simultaneously on the same resource element (RE). The number of characters r usually adjusted to reflect the current properties of the channel.

LTE uses OFDM in the downlink (and DFT pre-coded OFDM in the uplink communication) and, therefore, adopted by the NR×1 vector y_{n}for some element resource on subcarrier n (or alternative RE data number n), and assuming no mistovich interference t is thus,
simulated by

where e_{n}is the noise vector that is obtained as a realization of a random process. Advanced encoder, W_{NT×r}may be a broadband pre-coder, which is a constant frequency or polling frequency. Matrix pre-coder is often chosen to match the characteristics of the N_{R}×N_{T}MIMO channel H, which leads to the so-called dependent channel pre-coding. It is also commonly called pre-encoded with closed circuit and essentially expresses the desire to focus the energy transfer in the subspace, which is strong in the sense of transferring most of the transmitted energy to the UE. In addition, the matrix pre-coder may also be selected from a desire to orthogonalization of the channel, which means that right after linear correction to UE inter-level interference is reduced.

In the preliminary encoding of the closed loop, the UE transmits, on the basis of the measurement channel in a direct line of communication (downlink), recommendations for eNodeB by using appropriate pre-coder. The only pre-coder, it is believed, covers a large bandwidth (broadband pre is kodirovanie), can be brought back. It may also be useful to agree on the frequency channel change and implementation instead of feeding back the message frequency-selective pre-coding, for example, several pre-coders, one per sub-band. This is an example of a more General case of feedback information about the state of the channels (CSI), which also covers the feed back other than pre-coders objects to support the eNodeB in subsequent transmissions to the UE. Such other information may include indicators of channel quality (CQI), and the indicator grade transfer (RI).

One problem with prior coding with closed circuit are the service load feedback caused by the alarm indicator matrix pre-coder (PMI) and indicator rank pre-coder (i.e., RI) - especially in systems with large antenna configurations, where it is necessary to characterize the many dimensions of the channel. With modern design feedback, costs feedback for systems with many transmitting antennas in many cases will lead to unacceptable costs feedback. Complexity can also be a problem if you are using standard feedback loops, as the size of the antenna array increases. This is the compared, search "best" pre-coder among matrices candidate pre-coder in a large code book requires a large amount of computation, as it essentially involves an exhaustive search over a large number of elements in the code book.

DISCLOSURE of INVENTIONS

According to one or more aspects of the solutions presented here improve feedback information about the state of the channels (CSI) user equipment (UE) through the provision of content feedback well advanced encoder, part of the pre-coder messages CSI feedback. In one or more such embodiments, implementation, feedback is well advanced encoder corresponds to at least two matrices pre-coder, which includes the recommended matrix "conversion" pre-coder and featured the matrix "settings" pre-coder. Featured transformation matrix pre-encoder limits the number of dimensions of the channel in question featured a matrix configuration pre-coder, and, in turn, recommended a matrix configuration advanced encoder meets the recommended matrix pre-coder for the effective channel, which is set partly KJV is anotai recommended by the transformation matrix pre-coder.

Next, the recommended transformation matrix pre-coder has dimension rows-columns N_{T}×k, where the number of rows N_{T}equal to the number of ports of the transmission antennas in the first device, and the number of columns k is equal to the dimension of the transformation, which is less than the value of N_{T}in order to restrict the number of dimensions of the channel in question featured a matrix configuration pre-coder. Pre-encoder conversion usually, but not necessarily, connected with a coarser granularity in time and/or frequency than the pre-coder settings for the savings of the alarm and/or complexity.

A specific aspect is the fact that the dimension k of the transformation is not necessarily equal to the number of antenna ports N_{T}and either configured by the first device, which, for example, may be an LTE eNodeB, or is configured with a second device, which, for example, may be LTE mobile terminal or another type of UE. For a given N_{T}the transmission rank r and dimension k conversion are related as N_{T}≥k≥r, hence, available over the possible values of k and r. A specific aspect is the fact that there is at least one combination of N_{T}and r for which k can take at least two different values. Specifically, k can b shall be strictly less
than N_{T}that gives the possibility of reducing the dimensionality. Another aspect is that there is at least one combination of N_{T}and k for which r can take at least two different values.

The recommendations of the transformation matrices and configure advanced encoder, including the choice of dimension k conversion and transmission rank r, usually performed by the second device so that the second device uses feedback well pre-coder for supplying the first device recommended matrix pre-coder. Accordingly, the first device receives the recommended matrix pre-coder through such feedback, it necessarily follows such recommendations, and can extract information about the state of the channels from this feedback, which is used to determine the entry to be applied prior to encoding. The alternative considered here is that some portion or all of these settings are configured by the first device. These definitions are signalled from the first device to the second device, which uses this information to determine the remaining parameters, representing the recommendation of the pre-encoding.

BRIEF DESCRIPTION of DRAWINGS

p> Fig.1 is a block diagram of exemplary embodiments of the first device and a second device, where the second device is configured to send the recommendations of the pre-coding to the first device through feedback well advanced coder.Fig.2 is a block diagram of additional exemplary details for the device, introduced in Fig.1.

Fig.3A and 3B illustrate an exemplary codebook for maintaining information pre-coder conversion and adjustment in accordance with the decisions presented here.

Fig.4 is a block diagram of a variant of implementation of the scheme pre-coder configured for transmission pre-coding in accordance with the decisions presented here.

Fig.5 is a logical block diagram of one variant of the method of forming and sending feedback well pre-coder in the second device, to provide recommendations preliminary coding for the first device.

Fig.6 is a logical block diagram of one variant of the method of receiving and evaluating feedback well pre-coder in the first device, where this feedback is well pre-coder which provides recommendations preliminary coding from the second device.

DETAILED DESCRIPTION

Fig.1 illustrates the first device 10 ("device 1"), which transmits the pre-coded signal 12 to the second device 14 ("device 2") using a number of antennas 16 transmission. In turn, the second device 14 includes a number of antenna 18 for receiving the pre-encoded signal 12 and data return and signaling to the first device 10, which includes feedback 20 well pre-coder. Feedback 20 well advanced encoder contains recommendations pre-coder for the first device 10. The first device 10 shall consider, but not necessarily follows the recommendations of the preliminary coding included in the feedback 20 well advanced encoder, when determining the operation of the preliminary coding, which it uses to form the pre-encoded signal 12. However, one advantageous aspect of the solutions presented here, is that feedback 20 well pre-coder offers greatly improved efficiency in terms of processing required to determine feedback 20 well pre-coder, and/or in terms of service load (ed is rgec) signalling required to send feedback 20 well advanced coder.

In at least one embodiment, the second device 14 recommends that the matrix pre-coder for the first device 10 by specifying recommended the transformation matrix pre-coder for the first device 10 and/or by specifying the recommended matrix settings pre-coder for the first device 10. At least one such embodiment, feedback 20 well advanced encoder contains the alarm, providing such instructions for the first device 10. For example, in at least one embodiment, the second device 14 supports (maintains) one or more code books 22, which include a number of possible matrices 24 conversion pre-coder and a number of possible matrices 26 configuring advanced coder. The first device 10 supports the same one or more code books 22 (or equivalent, stores the information code books, which is extracted or depends on elements of the code books, supported by the second device 14).

In one or more such embodiments, the implementation of the second device 14 sends the values of the index matrix tentative is the main coder (PMI),
where these values identify the elements of the codebook representing the recommendations of the matrix pre-coder to be considered first device 10 in the operation definition preliminary coding for use in the formation of the pre-encoded signal 12. For example, presenting the recommended matrix pre-coder by W, feedback 20 well pre-coder in at least one embodiment, contains an index value that identifies a specific matrix of possible matrices 24 conversion pre-coder as recommended transformation matrix pre-encoder, denoted as W_{1}and additionally contains an index value that identifies a specific matrix of possible matrices 26 conversion pre-coder as recommended matrix settings advanced encoder, denoted as W_{2}. The device 10 configured accordingly with the possibility of the formation of the recommended matrix W pre-encoder as the product (matrix multiplication) recommended matrix W_{1}convert pre-coder and recommended matrix W_{2}settings pre-coder. Namely, W=W_{1}×W_{2}. The device 10 examines recom is huemul matrix W pre-coder in the operation definition pre-encoding,
which it applies. For example, it formulates a matrix pre-encoder used to generate a pre-encoded signal 12 is at least partially based on the recommended pre-coder W.

Thus, by receiving feedback 20 well advanced encoder, the first device 10 are informed about the recommended matrices W_{1}and W_{2}transform and customize pre-coder and considers CSI specified by such feedback in determining their operations pre-coding. The first device 10 evaluates W, for example, to determine whether to coordinate its operation prior to encoding with the recommended matrix W pre-coder. Namely, the first device 10 accepts and understands feedback 20 well advanced encoder, but the operation of pre-coding really used the first device 10 may follow or not to follow the recommendations of the pre-coder from the second device 14. Valid pre-coding in the first device 10 depends on a number of factors outside of the recommendations received from the second device 10.

As a non-limiting example, Fig.2 illustrates one Varian is the implementation of the first and second devices 10 and 14.
According to the illustrated example, the first device 10 includes a receiver 34, configured to receive feedback 20 well pre-coder from the second device 14. As discussed, feedback 20 well advanced encoder specifies at least one of the recommended transformation matrix pre-coder (W_{1}and featured matrix settings pre-coder (W_{2}), which together represent the recommended matrix pre-coder (W), which is a matrix multiplication of the recommended transformation matrices and settings pre-coder. As will be further detailed here later featured the transformation matrix pre-encoder limits the number of dimensions of the channel in question featured a matrix configuration pre-coder, and featured the matrix settings pre-coder will agree featured matrix pre-coder with an effective channel, which is partially set featured the transformation matrix pre-coder.

The first device 10 further comprises a transmitter 36, which includes a scheme 38 pre-coder. The transmitter 36 configured to determine the operation of the seat reservation coding for the formation of the pre-encoded signal 12, at least partially based on the assessment referred featured matrix pre-coder. Here, the operation of pre-coding" will be understood as a preliminary encoding, which is actually used by the first device 10 for forming a pre-encoded signal 12, and it can follow or not to follow the recommended matrix pre-coder corresponding to the recommended transformation matrices and settings pre-coder. The transmitter 36 configured to transmit the pre-coded signal 12 to the second device 14, where the pre-coded signal 12 is encoded according to a pre-operations pre-coding applied by the device 10.

In determining the actual operation preliminary coding to use the transmitter 36 is configured, for example, determine whether to use the recommended matrix pre-coder as a matrix pre-coder, actually used in the schema 38 pre-coder for the formation of the pre-encoded signal 12. Namely, the operation of pre-coding carried out by the first device 10 may follow or not to follow the recommended operation p is adveritising encoding, depending on a number of conditions. However, it will be clear that the first device 10 can follow these recommendations and, in any case, configured to understanding and consideration of feedback 20 well advanced encoder as the basis for identification of those recommendations.

Further, in at least one embodiment, the first device 10 configured to maintain one or more code books 22 as a two-dimensional table 28 possible matrices pre-coder. Cm. Fig.3A for a sample table 28, where table 28 will be understood, for example, as a data structure stored in the memory device 10. Table 28 includes a number of elements that individually represented by "W" in this illustration. Each W is a possible matrix pre-coder, formed as a matrix multiplication of a specific combination of possible matrices 24 and 26 convert and configure the pre-coder. Namely, each of some or all of W_{s}in table 28 represents the work of various pairwise connections possible matrix 24 conversion pre-coder and possible matrix 26 configuring advanced coder. Thus, each row (or column) of the table 28 are the same as the concrete matrix in the set of possible matrices 24 conversion pre-coder,
and each column (or row) of the table 28 corresponding to a particular matrix in the set of possible matrices 26 configuring advanced coder.

In this embodiment, feedback 20 well advanced encoder contains at least one of the value line index and the index of the column to identify the specific matrix of these possible matrices pre-coder in table 28 as recommended matrix pre-coder. Each value of the index of the row (or column) can be understood as representing a specific recommendation pre-encoder conversion, and each value of the index of the column (or row) can be understood as representing a specific recommendation pre-coder settings.

Note that the values of the indices of row and column can be fed back with different levels of detail, and note that such options making matrix 24 conversion pre-coder and possible matrix 26 configuring advanced encoder do not explicitly defined in a separate code books; instead, the product of specific possible matrix 24 conversion pre-coder and specific possible matrix 26 configuring advanced encoder is stored in a particular cell of the table 28.

B the children clear in such scenarios, the implementation of the second device 14 may also be configured to support such table 28 in memory of the second device 14. Thus, then, the second device 14 determines the value (s) of the index table corresponding to its recommendations, the preliminary encoder, and sends specify these values back to the first device 10 through feedback 20 well pre-coder. Namely, the second device 14 sends back the value line index and/or index value of the column as feedback 20 well pre-coder. To the extent that the first device 10 selects the advanced encoder settings, for example, the second device 14 does not need to be sent back to index values as row and column).

In another embodiment, such as suggested in Fig.1, the first device 10 configured to maintain one or more code books 22 possible matrices 24 conversion pre-coder and the possible matrices 26 configuring advanced coder. Accordingly, the receiver 34 of the first device 10 configured to receive feedback 20 well advanced encoder as at least one index value, ukazyvaya what about the at least one of the specific matrix of possible matrices 24 conversion pre-coder as recommended pre-encoder conversion, and the concrete matrix of the possible matrices 26 configuring advanced encoder as recommended matrix settings pre-coder.

Fig.3B illustrates an example of such a variant implementation, where the first device 10 configured to maintain one or more code books 22 by maintaining the first code book 30 possible matrices 24 conversion pre-coder and the second codebook 32 possible matrices 26 configuring advanced coder. In such scenarios, the implementation of feedback 20 well advanced encoder contains at least one of the first index value for the first code book 30 and the second index value to the second code book 32. It will be clear that the second device 14 maintains copies of one or both of the code books 30 and 32.

Regardless of the specific organization code books, in at least one embodiment, the first device 10 supports one or more code books 22 possible matrices 24 conversion pre-coder and the possible matrices 26 configuring advanced encoder, where each matrix 24 conversion pre-coder has some specific configuration. In particular, each of the possible transformation matrices pre-coder has a dimension string-Tolstov N_{
T}×k, where the number of rows N_{T}equal to the number of ports of the transmission antennas in the first device 10, and the number of columns k is equal to the dimension of the transformation, which is less than the value of N_{T}in order to restrict the number of dimensions of the channel in question featured a matrix configuration pre-coder. It will be clear that the second device 14 can support a similarly structured code book code book) 22.

In Fig.4 shows an exemplary implementation of circuit 38 pre-coder, which includes pre-encoder 50, which pre-encodes the signals for transmission to the first device 10 according to some pre-coding, which, as noted, is determined at least partially based on the evaluation recommended pre-coder determined from feedback 20 well pre-coder. In more detail, the circuit 38 pre-coder includes a tiered scheme 52 processing, which process the input symbols in the vector s of symbols used for each level (spatial multiplexing) (e.g., level 1, level 2 and so on).

These vectors characters are encoded according to the pre-active matrix pre-coding prior the first encoder 50,
and pre-coded vectors are ignored in scheme 54 processing inverse fast Fourier transform (IFFT), and the output signals of this circuit then routed to the appropriate ports of the N_{T}antenna ports 56. It will be clear that the number of N_{T}antenna ports that are available for use by the first device 10 in conducting pre-coded transmission, specifies the maximum number of dimensions of the channel under consideration by the operations of pre-coding the first device 10. As will be explained here in more detail later, the size and/or complexity of one or more code books 22 (and the size and/or complexity of feedback 20 pre-coding) can be advantageously reduced by limiting the number of dimensions of the channel considered less than N_{T}.

In the above embodiment, at least one of the possible matrices 24 conversion pre-coder contains a block-diagonal matrix. Further, at least one of the possible matrices 26 configuring advanced encoder has a row of the matrix, which alter the phasing of the blocks in this block-diagonal matrix. Here each block in a block-diagonal matrix can be understood in terms of beam forming for forming a set of rays emitted from the corresponding the corresponding subset of N_{
T}antenna ports 56, and consider "phasing" here represents the phase shift between the beams on both blocks of the block-diagonal matrix.

Further, in at least one embodiment, the dimension k of the transformation is a configurable parameter. In the case when the dimension k of the transform is configured by the first device 10, the first device 10 configured to alarm specify the dimension k of the transformation from the first device 10 to the second device 14. Accordingly, the second device 14, in this case, configured to receive signaled values of dimension k conversion and consider this value in the implementation of his recommendations, the preliminary coding - i.e., it limits his choice recommended the transformation matrix pre-coder because signaled k values.

Still further, in at least one embodiment, the recommended transformation matrix pre-coder selects the first device 10, but not the second device 14. In this case, the first device 10 configured to alarm indications recommended the transformation matrix pre-coder for the second device 14. Accordingly, the second device 14 konfigurera is but with the ability to receive guidance recommended the transformation matrix pre-coder from the first device 10 and use this signaled guidance in selecting the recommended matrix settings pre-encoder - i.e., the second device 14 restricts its consideration of possible matrices 26 settings pre-coder those matrices which are suitable (in terms of dimensions) for use with the recommended transformation matrix pre-coder.

As an additional benefit solutions presented here, in one or more embodiments implement one or more code books 22 includes a set of possible matrices 24 conversion pre-coder in such a way that the number of unique vectors that form a particular column from a set of possible transformation matrices pre-coder is greater than the number of unique vectors forming another column from a set of possible transformation matrices pre-coder.

Further, in at least one embodiment, the first device 10 configured to receive feedback 20 well pre-coder from the second device 14 as the first alarm, adopted by the first device 10 with the first granularity in time or frequency, which indicates the recommended transformation matrix pre-coder, the second alarm, adopted by the first device 10 with the second granularity on time or cha is the Thoth, indicates the recommended matrix settings pre-coder. In particular, the first granularity is larger than the second granularity. Accordingly, the second device 14 configured to alarm recommended the transformation matrix pre-coder with a first granularity, and featured alarm matrix settings pre-coder with the second level of detail.

More broadly and with reference to Fig.2 it will be clear that the second device 14 configured to specify a recommended matrix pre-coder for the first device 10. In support of this configuration, a sample implementation of the second device 14 includes a receiver 40, which is configured to estimate channel conditions relative to the first device 10. In this regard, the second device 14 receives, for example, specific to the antenna reference signals for N_{T}antenna ports 56. These signals allow the receiver 40 to perform channel estimation for each antenna that allows the second device 14 to determine, for example, the number of levels of spatial multiplexing that it can support, and thus to use this definition in the implementation of the recommendations of the pre-coder is La the first device 10.

Accordingly, the receiver 40 is additionally configured to determine feedback 20 well advanced encoder at least partially based on the channel conditions. As noted above, feedback 20 well advanced encoder specifies at least one of the recommended transformation matrix pre-coder and recommended matrix settings advanced encoder, where recommended the transformation matrix and configure pre-coder together represent the recommended matrix pre-coder, which is a matrix multiplication of the recommended transformation matrices and settings pre-coder.

As previously recommended by the transformation matrix pre-encoder limits the number of dimensions of the channel in question featured a matrix configuration pre-coder, and featured the matrix settings pre-coder will agree featured matrix pre-coder with an effective channel between the first and second devices 10 and 14, which is partially set featured the transformation matrix pre-coder. The second device 14 additionally includes a transmitter 42, configured to send feedback 20 factorize the preliminary data of the encoder to the first device 10, to specify the recommended matrix pre-coder for the first device 10.

Referring to the above examples, the first and second devices, Fig.5 illustrates one variant of the method, implemented in a first device 10 according to the solutions presented here. Illustrated method 500 provides a transmission pre-coding from the first device 10 to the second device 14. The method 500 includes receiving feedback 20 well pre-coder from the second device 14 (block 502), where this feedback indicates at least one of the recommended transformation matrix pre-coder and recommended matrix settings pre-coder (with the previously detailed structure/nature). The method 500 additionally includes an operation definition pre-coding (for pre-coding for the second device 14) at least partially based on the assessment referred featured matrix pre-coder (block 504). Additionally, the method includes transmitting the pre-coded signal 12 to the second device 14, which is encoded according to a pre-defined operations preliminary coding (block 506).

Fig.6 illustrates an example method 600, R is Lisovenko in the second device 14, where this method involves the assessment of the conditions of the channel relative to the first device 10 (block 602) and the definition of feedback 20 well advanced encoder at least partially based on the conditions of the channel (block 604). As before, feedback 20 well advanced encoder specifies at least one of the recommended transformation matrix pre-coder and recommended matrix settings pre-coder. The method 600 optionally includes sending feedback 20 well advanced encoder to the first device 10 (block 606) to specify the recommended matrix pre-coder for the first device 10.

As an additional example, in one or more embodiments, the implementation presented here, at least some aspects of the recommendations of the pre-coder based on the definition of the square root of the channel covariance. This treatment, therefore, is concerned with the assessment of channel conditions between the first and second devices 10 and 14. At least one such embodiment, the first device 10 is an eNodeB, for example, based on the LTE wireless communication network. Accordingly, the second device 14 is a mobile terminal or other item of user equipment (UE), konfigurera is hydrated to work in based on the LTE wireless network.

eNodeB determines a matrix pre-coder for pre-coding transmission for the UE, where this determination is made, at least partially based on consideration of the recommendations of the pre-coder from the UE provided in the feedback form 20 well advanced encoder, as discussed earlier. In particular, one method of determining the UE recommendations pre-coder for eNodeB based on the following:

1. UE estimates the N_{R}×N_{T}channel matrix H_{n}for a given set of resource elements (RE) orthogonal multiplexing frequency division (OFDM), where these estimates are based on specific antenna reference signal from the eNodeB.

2. UE generates an estimate of the covariance matrix of the transmission channel R_{tr}=E[H*H], for example, by forming a sample estimatewhere the summation is over the set of RE. Such averaging is taken over the set of RE in time, uses the fact that the correlation properties of the channel can often slow to change over time, whereas a similar averaging in terms of frequency uses the fact that the correlation properties can be rather constant with frequency. Thus, a typical operation is that the averaging is performed across the frequency band of the system (OAD is emer,
across the bandwidth of the considered OFDM carrier) and includes many podkatov time. A weighted average can also be formed to account for the fact that the correlation properties will eventually become obsolete over time or frequency.

3. UE matrix then takes the square root of R_{tr}for example, R_{tr}^{1/2}=V Λ^{1/2}where V is the eigenvectors of the covariance matrix of the transmission channel, and a diagonal matrix Λ^{1/2}contains the square root of the corresponding eigenvalues, sorted in descending order. (Note that there are other forms of matrix square roots, and here it is assumed that other forms may be used).

4. UE now hypothetically assumes a value of dimension k conversion (which implicitly restricts the rank of transmission through k). This implies that these are only the first k columns of R_{tr}. These columns are scaled to some fixed norm Frobenius and then elementwise quanthouse.

5. Featured matrix (W_{1}) convert a pre-coder now fixed for hypothetical values of k to match the truncated columns, quantized and scaled to the square root of the covariance matrix of the transmission channel.

6. UE now the hypothesis is automatic admits a rank value transfer r provided this hypothetical k.

7. UE now refers to a new effective channel H_{n}W_{1}for which it tries to choose a consistent hypothetical pre-coder settings (agreed on a set of RE, for example, the subrange in LTE) to optimize some measure of performance. For example, this selection may be optimized, for example, predicted performance, or may be aimed at higher transport format, giving a BLER not exceeding 10%. The pre-coder settings can be selected from a codebook W_{m}={W_{2,1}, W_{2,2}, ...}. Namely, the set of possible pre-coders 26 settings shown for codebook 32 in Fig.3B, may include a finite set of choices of different pre-coders settings W_{2,1}, W_{2,2}and so, for a hypothetical values of dimension k conversion and rank r transmission. Various choices of such sets can be supported for different values of k and r. Code book (books) pre-coder settings could, for example, to correspond to the respective rank of the transmission codebook 2 or 4 antenna ports available in LTE Rel-8.

8. The UE then performs a search on some or all of the different possible combinations of k and r through repetition of the above stages 4-7 and finally selects the final best combination of matrices change the Oia and configure advanced encoder, including the choice of k and r. Here, "best" combination may be a combination of possible pre-coder 24 conversion and possible pre-coder 26 settings from the code book (books) 22, which gives the highest or otherwise best value for the selected performance measures. Alternatively, the best dimension k conversion was selected and reported in the previous moment, but still used, and only the rank r is determined on the basis of previously defined dimension transformation, through repetition of stages 4-7.

9. In continuation, the UE converts scalar quantized elements recommended pre-coder convert the bit sequence that is encoded and sent to the eNodeB. Similarly, it also reported the index pointing to the code book pre-coder settings. This last index would directly correspond to the PMI reported in LTE. Note also that instead of scalar quantization recommended pre-encoder conversion can also be selected from the code book, for example, by selecting a possible pre-coder 24 conversion, which corresponds to the covariance of the transfer in the sense of maximizing the received signal-to-noise ratio (SNR) or ergodic measures throughput ka is Ala. Further, even if the alarm feedback well pre-coder is performed using scalar quantization, the UE may still have an internal code book pre-encoder conversion as a way requirements desired properties to the matrix prior encoder conversion, which is selected as the recommended transformation matrix pre-coder, before rounding to the nearest scalar quantization.

Further, as noted earlier, the actual feedback message can be done a number of ways. For example, in LTE, the feedback could be carried out on the control channel uplink communication PUCCH, to periodically transmit status information channels (CSI) to the eNodeB, where the CSI may include feedback is well advanced encoder, which here is of interest. CSI can also be transmitted through explicit CSI request message on the PUSCH. In one or more embodiments, the implementation of the UE reports only the recommended transformation matrix pre-coder on PUSCH, together with the message featured multiple matrices settings advanced encoder, each of the pre-coder settings aimed at specific sub-band of the entire strip is of astot system. It is also expected to change the content based on the PUSCH of the message so that sometimes passed the recommended transformation matrix pre-coder, and for other podkatov transmitted featured matrix or matrix configuration pre-coder.

What recommendations should be transmitted from the UE to the eNodeB, in one or more embodiments of implementation is indicated as part of the access permissions ascending line on the PDCCH. For example, this permission includes some bits or some of the available bit pattern that UE interprets as an indicator of what recommendations should be sent. In support of this method, the ratio of the strict synchronization can be established between the various recommendations of messages pre-coder from the UE, so as to UE and eNodeB clear which time/frequency resources correspond to the specific recommendations matrix pre-coder from the UE. As an alternative, the UE configured to transmit its recommendations matrix pre-coder in the higher point of the Protocol stack, as the access control to the data transfer medium (WT) or via a signaling Protocol radio resource control (RRC).

Further, the eNodeB does not necessarily know how UE selects the pre is sustained fashion coders it recommends. In fact, the typical case is that the UE does not know, but rather only knows that the UE somehow prefers pre-coders, which it reports. In particular, the eNodeB may not be aware of the basis on which the UE recommends specific transformation matrix pre-coder. The alternative proposed here for one or more embodiments is that the transformation matrix is pre-encoder should be selected on the basis of the square root of the covariance of the transmission channel, or even the fact that the covariance matrix of the transmission as a whole is fed back from the UE to the eNodeB. This approach, however, presents some challenges in terms of testing and guarantee such behavior UE on multiple UE merchants.

These challenges occur because the channel properties, such as the covariance of the transmission channel, visible only inside the UE, it is not easy to observe from the outside and, therefore, no easy way to ensure that the reported covariance has the correct value, especially because part of the input stages of the receiver in the UE may contribute to this covariance. Clearly inform the pre-coder, in contrast, allows a hypothetical transfer and, as a consequence, in terms of the transport format, which contain about 0% BLER for a hypothetical transmission, reported by CQI. This can be observed by inspection of the ACK/NACK of the UE and evaluation BLER. These aspects of feedback is not limited to any particular embodiment described herein, and applies to the following for more details.

In at least one embodiment, the dimension k of the transformation adapted to suit different correlation properties of the channel. In this respect, the choice of dimension k conversion serves as a simple way of limiting the energy transferred by the reduced subspace of dimension N_{T}-dimensional vector space. Roughly speaking, this focuses the energy in certain preferred "directions" and, thus, eliminates the need for pre-coder settings to cover more than the required subspace. For example, the code book (book) 22 include a number of possible matrices 24 conversion pre-coder, which is limited by the dimension of k) in some subspace of N_{T}-dimensional vector space, and a set (sets) of possible matrices 26 configuring advanced encoder thereby simplified.

Otherwise, the requirement of accounting matrix settings pre-coder full N_{T}-dimensional vector about transto would require more code books and
thus, higher costs of signaling between the UE and eNodeB and/or requires a higher complexity in the search for pre-coder in the UE and/or eNodeB. To understand why the adaptation values of dimension k conversion is best, consider a scenario with four jointly polarized and closely spaced (about half the wavelength) transmitting antennas at the eNodeB. For the purposes of this example, the first device 10 may be understood as eNodeB, and the antenna 16, therefore, contain four together polarized and closely spaced antennas. If the angular divergence in the eNodeB is small enough, then the channels, corresponding to different transmission antennas, will be highly correlated and the covariance of the transmission channel will therefore have a very strong own value, and the remaining eigenvalues will be weak. For such a channel forming a beam with a single level is appropriate.

This may be realized through a well pre-coding as follows:

that gives effective pre-coder

Here the dimension k of the transformation is equal to 1, and the transmission rank r equals to one, while w_{BF}is the beam shaper with a single level, which focuses all the energy transfer in the strongest direction of the channel, thereby improving SINR at the receiving side. In this case, the UE would report information that describes or otherwise indicates the recommended transformation matrix pre-coder, while the corresponding recommended matrix settings pre-coder is a constant, and, thus, do not need to spend bits for a message about it.

Beam shaper could be taken from a codebook based on the columns of the matrices of the discrete Fourier transform (DFT), forming a lattice of beams for selection. Alternatively, the beam shaper can be based on the covariance matrix of the transmission channel. However, as the angular divergence (beam) increases, the eigenvalues of the covariance matrix of the transmission channel become more similar. Therefore, the most powerful of its own is no longer dominant as before. Then it may be advantageous distribution of some energy in more than one direction. Therefore, has the meaning in order to make the dimension of the k conversion more than 1. At the same time, the transmission rank r can be equal to 1, for example, because the SNR is not high enough to guarantee mnogolampovoy transmission. In this case, k>1 and r=1. For k=2 the recommended matrix settings pre-coder could be selected from a pre-coder with two antenna ports in LTE Rel-8, i.e., as

The latter case is the dimension of the transformation is equal to two, it also makes sense, if the antenna array in the eNodeB consists of several closely spaced cross-poles. Each polarization then forms the group is co-polarized and closely-spaced antennas, for which the channel correlation is high, if the angular divergence is low enough. Beamforming on each polarization is then reasonable and is followed by the pre-coder settings, which tries to adjust the relative phase between the two polarizations. Featured matrix W pre-coder can be prepared for this operation by defining W as an effective matrix pre-coder W_{eff}=W_{1}W_{2}where the recommended matrix W_{1}conversion prewar the positive encoder and featured the matrix W_{
2}settings pre-coder could take the form

The above details show that the choice between different values of dimension k conversion is beneficial. The actual choice of k can be performed in a manner similar to the search conducted in an exemplary embodiment relating to determine which of the possible matrices 24 conversion pre-coder should be recommended on the basis of matrix square roots.

Fixture rank transmission is an additional aspect in one or more embodiments, the implementation presented here. Namely, the rank r of the transmission also varies. Here recognized that it is important to allow r to vary, even though the dimension k of the transformation and the number of N_{T}ports of transmission antennas remain fixed. Consider again the case of the antenna array transmission with multiple closely spaced cross-poles. As shown above, the pre-coder W_{1}conversion can take a block-diagonal form

The dimension k of the transformation here is two, which corresponds to two orthogonal polarizations, thus, also implies that the corresponding pre-coder settings has two lines. The pre-coder settings could, however, have one or two columns, depending on the rank r of the transmission, which is supported by the channel. For example, if the SINR is low, you may transfer with a single level is preferred. Still maintaining k=2 is advantageous because it allows pre-coder settings to adjust the relative phase between the two polarizations and, thus, to achieve a coherent combination of the transmission signals at the receiving side. However, if the SINR is high, you may use two levels is better than using only a single level, and therefore, the pre-coder settings would have two columns.

Matrix settings pre-coder, which is recommended for use, therefore, could be selected from a codebook of unitary 2×2 matrices which tend to orthogonalizing effective 2×2 channel formed by the works of the channel matrix and the pre-coder conversion. Similar arguments apply what I for cluster antenna arrays, where the band antennas have channels with a high correlation, but the correlation between the groups is low, therefore, there is a need for a pre-coder settings for the phase settings. It will be clear that one or more code books 22 can be filled with a large set of possible matrices 26 configuring advanced encoder, where one or more defined subsets of them have the above properties. (In General, the data subset of possible matrices 26 settings pre-coder in the code book (books) 22 will correspond to the data values of dimension k conversion and transmission rank r, so that the selected matrix settings pre-coder that is appropriate for the selected transform matrix pre-coder).

Another aspect of one or more embodiments presented here is supported eNodeB selection of pre-coder. Even though the recommendations of the pre-coder is usually performed by a UE as the UE usually has the best channel measurement downlink, the design is well advanced coder presented here, advantageously allows the design, where the eNodeB supports the choice of the pre-coder. This support is based, for example, on the channel measure the x in the return line connection (uplink communication), where reciprocity can be used to acquire information about the channel for downlink. Supported eNodeB selection of pre-coder is particularly suitable for systems duplex time division (TDD), where reciprocity can be properly used, but also system duplex frequency division (FDD) can benefit from such support through the use of parameters for large scale of the channel, which are also mutual duplex on large distances.

One such approximate variant of the implementation is the implementation of choice eNodeB (not UE) of dimension k conversion and signaling the selected dimension conversion to UE via direct signaling, and in this case, the UE determines k by decoding the messages sent from the eNodeB. In this configuration, the UE will be limited to the message prior encoder conversion, satisfying customized dimension conversion. This configuration has the advantage that the eNodeB could take into account those factors in the selection that is not available in the UE, such as the presence of a jointly planned UE in case of multi-user systems with multiple input and multiple output (MU-MIMO). This decision, therefore, may be beneficial, even though the channel measured what I usually are more accurate in the UE.

In an additional exemplary embodiment, eNodeB, in addition, carries out the recommendation of the transformation matrix pre-coder and signals this selection for the UE via direct signaling. In such cases, the UE determines k and the choice of the pre-coder transform-based decoding of this alarm. In this configuration, the UE is limited by the choice of matrix pre-coder implemented by the eNodeB when determining its recommendation, the pre-coder settings.

Further, as for MU-MIMO, when planning many UE on the same amount of time-frequency is significant that the eNodeB is able to spatially separate streams for simultaneous transmission. For such applications configuring dimension k conversion should be less restrictive, so are not only more dominant own fashion channel, but also moderately strong own fashion, in which the UE is still sensitive to noise. This approach can be achieved through the implementation choices eNodeB dimension k conversion, consistent with the above variants of implementation, or, if the UE selects the dimension of the transformation, through the following options: eNodeB may configure ogranichennoi criterion E applies to the choice of dimension k conversion. To be effective, this configuration must be set for the selection of rank r.

As a further consideration here, we note that the dimension k of the transform determines the number of dimensions of the channel (N_{T}-k) are exactly the truncated (aquantance) regarding feedback 20 well pre-coder. Columns recommended the transformation matrix pre-coder determines the effective dimension of the channel, which must be quantized. Could, however, be useful to have a smoother transition between quantized dimensions and truncated dimensions. Such a smooth transition is achieved in one or more embodiments, the implementation presented here, through such elements codebook for possible matrix 26 configuring advanced encoder that row of the matrix settings pre-coder quanthouse with different resolution. As an example, for any given matrix of possible matrices 26 configuring advanced encoder in one or more codebooks 22, the first row of this matrix has the highest resolution, and the resolution decreases with the increase of the index of the row (the last row has the most coarse resolution quantization).

With this design code book pre-coder mood and sorting the columns of the transformation matrix pre-coder becomes relevant, as each column is associated with a corresponding row of the matrix settings pre-coder. Therefore, if you are implementing a reduced resolution quantization of strings preliminary encoding settings, then more bits of feedback pre-coder settings will be spent on the choice of rotations of the first column of the transformation matrix pre-coder than in the last column of the transformation matrix pre-coder. These columns of the transformation matrix pre-coder, thus, must be ordered so that the first column represents the most important dimension of the channel, and the last column is the least important of the k most important) dimension of the channel ("directions"). In a broad sense, then one or more embodiments presented here use the elements of a codebook, which quantuum line of possible matrices 26 configuring advanced encoder with different permissions.

Referring to the above variations, the approaches described herein provide a solution for operation with spatial multiplexing with closed circuit, as well as with MU-MIMO, and this is done using a managed service load feedback. Increased efficiency and simplicity provided by the use of the feedback 20 well advanced coder (and associated processing), provide special advantages for large antenna configurations.

As a further non-limiting examples of the various advantages of the disclosed solution provide a reduced service load feedback for this performance downlink; improved performance downlink for the service load feedback; reduced computational complexity by reducing the dimensionality of the estimates used for dynamic message pre-coder provided feedback 20 well pre-coder; good suitability for MU-MIMO transmission, as the recommendations of the preliminary encoder conversion are reported with high resolution phase quantization.

In addition, although terminology from 3GPP LTE has been used in various sections of this document to provide meaningful configuration and working examples, such use LTE examples should not be construed as limiting the volume of the solutions presented here. It is assumed that these solutions can be extended, for example, WCDMA, WiMax, UMB and GSM. A more General way, it should be understood that the foregoing details and accompanying illustrations provide a non-limiting exemplary embodiments of the solution is, disclosed here.

1. The method (500) pre-coding transmission from the first device (10) to the second device (14), and the above-mentioned method comprises:

receiving (502) feedback well pre-coder from the second device (14) that indicates at least one of the recommended transformation matrix pre-coder and recommended matrix settings pre-coder, which together represent the recommended matrix pre-coder, which is a matrix multiplication of the recommended transformation matrices and settings pre-coder, and referred to the recommended transformation matrix pre-encoder limits the number of dimensions of the channel in question referred to the recommended matrix settings pre-coder, and the above-mentioned recommended matrix settings advanced encoder coordinates referred to the recommended matrix pre-coder with some effective channel which is defined in part by the aforementioned recommended the transformation matrix pre-coder;

determining (504) operations pre-coding at least partially on the basis of the assessment referred featured matrix pre-coder; and

transmission (506) prior is part of the encoded signal (12) to the second device (14),
pre-coded in accordance with the above operation prior encoding.

2. The method (500) under item 1, in which the assessment referred featured matrix pre-coder includes determining whether to utilize the recommended matrix pre-coder in the above-mentioned operations pre-coding for the formation of the pre-encoded signal (12).

3. The method (500) under item 1 or 2, additionally containing maintaining one or more code books (22) as two-dimensional tables (28) possible matrices pre-coder, in which each row or column of the referenced table (28) corresponding to a particular matrix in the set of possible matrices (24) the transformation of the pre-coder and each column or row of the referenced table (28) corresponding to a particular matrix in the set of possible matrices (26) setting pre-coder, and where mentioned feedback (20) well pre-coder contains at least one of the value line index and the index of the column to identify the specific matrix of these possible matrices pre-coder in the above table (28) as mentioned recommended matrix pre-coder.

4. The method (500) under item 1 or 2, additionally containing supported by the E. one or more code books (22) possible matrices (24) the transformation of the pre-coder and the possible matrices (26) configure advanced encoder, where mentioned feedback (20) well advanced encoder contains at least one index value indicating at least one of the specific matrix of these possible matrices (24) the transformation of the pre-coder as mentioned recommended pre-encoder conversion and the specific matrix of these possible matrices (26) setting pre-coder as mentioned recommended matrix settings pre-coder.

5. The method (500) for p. 4, which referred to the maintenance of one or more code books (22) includes maintaining the first codebook (30) mentioned the possible matrices (24) the transformation of the pre-coder and the second codebook (32) mentioned the possible matrices (26) configure advanced encoder and in which the mentioned feedback (20) well advanced encoder contains at least one of the first index value for the first code book (30) and the second index value to the second code book (32).

6. The method (500) for p. 1, additionally containing maintaining one or more code books (22) possible matrices (24) the transformation of the pre-coder and the possible matrices (26) configure advanced encoder and in which each possible matrix (24) the transformation of the pre-coder had the t dimension rows-columns N_{
T}×k, where the number of rows N_{T}equal to the number of ports (56) of the transmission antennas in the first device (10), and the number of columns k is equal to the dimension of the transformation, which is less than the value of N_{T}in order to limit the number of dimensions of the channel in question referred to the recommended matrix settings pre-coder.

7. The method (500) for p. 6, in which at least one of the mentioned possible matrices (24) the transformation of the pre-coder contains a block-diagonal matrix.

8. The method (500) for p. 7, in which at least one of the mentioned possible matrices (26) setting pre-coder has a row of the matrix, which alter the phasing of the blocks in the above-mentioned block-diagonal matrix.

9. The method (500) for p. 6 in which the said dimension k conversion is configured referred to the first device (10) or referred to the second device (14) and which, if referred to the dimension k of the transform is configured referred to the first device (10), the above-mentioned method (500) further comprises an alarm indication mentioned dimension k conversion from the first device (10) of the said second device (14).

10. The method (500) for p. 9, which referred to the recommended transformation matrix pre-coder, proc is guerrette first-mentioned device (10) and in which the said means (500) is provided for alarm indications mentioned recommended the transformation matrix pre-coder of said the first device (10) of the said second device (14).

11. The method (500) for p. 3, in which one or more code books (22) include the set of possible matrices (26) setting pre-coder in such a way that the number of unique vectors that form a specific row of the above set of possible matrices settings pre-coder is greater than the number of unique vectors forming another row of the above set of possible matrices settings pre-coder.

12. The method (500) under item 1, in which the said receiving feedback (20) well pre-coder from the second device (14) includes receiving signaling from the first granularity in time or frequency, which specifies mentioned featured a transformation matrix pre-coder, and receiving signaling from the second level of detail in time or frequency, indicating mentioned recommended matrix settings pre-coder, and in which the mentioned first granularity is coarser than said second degree of detail.

13. The first device (10) configured for transmission pre-coding to the second device (14), and referred to the first device (10) contains

the receiver (34), configured to receive the feedback (20) well pre-coder from the second device (14),
indicating at least one of the recommended transformation matrix pre-coder and recommended matrix settings pre-coder, which together represent the recommended matrix pre-coder, which is a matrix multiplication of the recommended transformation matrices and settings pre-coder, and referred to the recommended transformation matrix pre-encoder limits the number of dimensions of the channel in question referred to the recommended matrix settings pre-coder, and referred to the recommended matrix settings advanced encoder coordinates referred to the recommended matrix pre-coder with an effective channel, which is partially set mentioned recommended by the transformation matrix pre-coder; and

transmitter (36), which includes a circuit (38) pre-encoder and configured to

the operation definition pre-coding at least partially on the basis of the assessment referred featured matrix pre-coder and

transmitting the pre-coded signal (12) to the second device (14), which is pre-coded in accordance with the above operation prior encoding.

14. The first device (10) p is p. 13, in which the said transmitter (36) configured to assessment referred featured matrix pre-encoder by determining whether to utilize the recommended matrix pre-coder in the above-mentioned operations pre-coding used to generate pre-coded signal (12).

15. The first device (10) under item 13 or 14, in which is mentioned the first device (10) configured to maintain one or more code books (22) as two-dimensional tables (28) possible matrices pre-coder, and each row or column of the referenced table (28) corresponding to a particular matrix in the set of possible matrices (24) the transformation of the pre-coder and each column or row of the referenced table (28) corresponding to a particular matrix in the set of possible matrices (26) setting pre-coder and mentioned feedback (20) well pre-coder contains at least one of the value line index and the index of the column to identify the specific matrix of these possible matrices pre-coder as mentioned recommended matrix pre-coder.

16. The first device (10) under item 13 or 14, in which the first mentioned is a device (10) configured to maintain one or more code books (22) possible matrices (24) the transformation of the pre-coder and the possible matrices (26) configure advanced encoder, moreover, the above-mentioned receiver (34) configured to receive the mentioned feedback (20) well advanced encoder as at least one index value indicating at least one of: a specific matrix of these possible matrices (24) the transformation of the pre-coder as mentioned recommended pre-encoder conversion and the specific matrix of these possible matrices (26) setting pre-coder as mentioned recommended matrix settings pre-coder.

17. The first device (10) according to p. 16, in which is mentioned the first device (10) configured to maintain one or more code books (22) by maintaining the first codebook (30) mentioned the possible matrices (24) the transformation of the pre-coder and the second codebook (32) mentioned the possible matrices (26) configure advanced encoder and in which the mentioned feedback (20) well advanced encoder contains at least one of the first index value for the first code book (30) and the second index value to the second code book (32).

18. The first device (10) p. 13, in which is mentioned the first device (10) configured to maintain one or more code books (22) the possibility of the s matrix (24) the transformation of the pre-coder and the possible matrices (26) configure advanced encoder,
each possible matrix (24) the transformation of the pre-coder is the dimension of the row-column N_{T}×k, where the number of rows N_{T}equal to the number of ports (56) of the transmission antennas in the first device (10), and the number of columns k is equal to the dimension of the transformation, which is less than the value of N_{T}in order to limit the number of dimensions of the channel in question referred to the recommended matrix settings pre-coder.

19. The first device (10) p. 18, in which at least one of the mentioned possible matrices (24) the transformation of the pre-coder contains a block-diagonal matrix.

20. The first device (10) according to p. 19, in which at least one of the mentioned possible matrices (26) setting pre-coder has a row of the matrix, which alter the phasing of the blocks in the above-mentioned block-diagonal matrix.

21. The first device (10) under item 18 in which the said dimension k conversion is configured referred to the first device (10) or referred to the second device (14) and which, if referred to the dimension k of the transform is configured referred to the first device (10) referred to the first device (10) configured to alarm indications mentioned dimension k conversion from the PE the first device (10) of the said second device (14).

22. The first device (10) according to p. 21, in which is mentioned the recommended transformation matrix pre-coder selects the first-mentioned device (10), and referred to the first device (10) configured to alarm indications mentioned recommended the transformation matrix pre-coder of said first device (10) of the said second device (14).

23. The first device (10) according to p. 15, in which one or more code books (22) include the set of possible matrices (26) setting pre-coder in such a way that the number of unique vectors that form a specific row of the above set of possible matrices settings pre-coder is greater than the number of unique vectors forming another row of the above set of possible matrices settings pre-coder.

24. The first device (10) p. 13, and referred to the first device (10) configured to receive feedback (20) well pre-coder from the second device (14) as the first alarm, adopted by the first device (10) with a first degree of detail in time or frequency, which specifies mentioned featured a transformation matrix pre-coder, and the second alarm, adopted the first elimination of the ETS (10) with the second granularity in time or frequency, pointing mentioned recommended matrix settings pre-coder, and mentioned the first granularity is coarser than said second degree of detail.

25. The method (600) of the second device (14) specify the recommended matrix pre-coder for the first device (10) with the above-mentioned method contains

assessment of channel conditions on the said first device (10);

the definition of feedback (20) well advanced encoder at least partially on the basis of the conditions of the channel, and mentioned feedback (20) indicates at least one of the recommended transformation matrix pre-coder and recommended matrix settings pre-coder, and mentioned recommended the transformation matrix and configure pre-coder together represent the recommended matrix pre-coder, which is a matrix multiplication of the recommended transformation matrices and settings pre-coder, and said transformation matrix pre-encoder limits the number of dimensions of the channel in question referred to the recommended matrix settings advanced encoder, and referred to the recommended matrix configuration predvaritelnogo the encoder coordinates referred to the recommended matrix pre-coder with an effective channel between the first and second devices (10,
14), which is defined partly mentioned recommended by the transformation matrix pre-coder; and

the reference mentioned feedback (20) well pre-coder of the said first device (10) for instructions mentioned recommended matrix pre-coder for the said first device (10).

26. The method (600) according to p. 25, optionally containing maintaining a two-dimensional table (28) possible matrices pre-coder, and each matrix can be selected as mentioned recommended matrix pre-coder, and each row or column of the referenced table (28) corresponding to a particular matrix in the set of possible matrices (24) the transformation of the pre-coder and each column or row of the referenced table (28) corresponding to a particular matrix in the set of possible matrices (26) setting pre-coder, and mentioned feedback (20) well pre-coder contains at least one of the value line index and the index of the column to identify the specific matrix of these possible matrices pre-coder as mentioned recommended matrix pre-coder.

27. The method (600) according to p. 25 or 26, in which the above mentioned reference feedback (20) well the CSO pre-coder contains the alarm mentioned recommended pre-encoder conversion with the first degree of detail in the time or frequency and alarming mentioned recommended pre encoder configuration with the second granularity in time or frequency, and mentioned the first granularity is coarser than said second degree of detail.

28. The method (600) according to p. 25, optionally containing maintaining in said second device (14) one or more code books (22) containing the set of possible matrices (24) the transformation of the pre-coder, and each has dimension rows-columns N_{T}×k, where the number of rows N_{T}equal to the number of ports (56) of the transmission antennas in the first device (10), and the number of columns k is equal to the dimension of the transformation, which is less than the value of N_{T}in order to limit the number of dimensions of the channel in question referred to the recommended matrix settings pre-coder.

29. The method (600) for p. 28, in which at least one of the mentioned possible matrices (24) the transformation of the pre-coder contains a block-diagonal matrix.

30. The method (600) according to p. 29, in which one or more code books (22) additionally contain the set of possible matrices (26) configure advanced encoder, and at least one of the mentioned possible matrices (26) setting pre-coder has a row of the matrix, which alter the phasing of the blocks in the above-mentioned block-diagonal matrix.

31. JV the property (600) p. 28, in which the said dimension k conversion is configured referred to the first device (10) or referred to the second device (14) and which, if referred to the dimension k of the transform is configured referred to the first device (10), the above-mentioned method (600) further includes receiving instructions mentioned dimension k conversion from the first device (10).

32. The method (600) according to p. 31, in which is mentioned the recommended transformation matrix pre-coder selects the first-mentioned device (10) and in which the said means (600) further includes receiving instructions mentioned recommended the transformation matrix pre-coder of said first device (10).

33. The method (600) according to p. 25, optionally containing maintaining one or more code books (22) in the above-mentioned second device (14), and one or more code books (22) include the set of possible matrices (26) setting pre-coder in such a way that the number of unique vectors that form a specific row of the above set of possible matrices settings pre-coder is greater than the number of unique vectors forming another row of the above set of possible matrices settings pre-coder.

34. The second from trojstvo (14),
configured to specify a recommended matrix pre-coder for the first device (10), and referred to a second device (14) includes

the receiver (40) configured with the ability

assessment of channel conditions on the said first device (10) and

determine feedback (20) well advanced encoder at least partially on the basis of the conditions of the channel, and mentioned feedback (20) well advanced encoder specifies at least one of the recommended transformation matrix pre-coder and recommended matrix settings pre-coder, and mentioned recommended the transformation matrix and configure pre-coder together represent the recommended matrix pre-coder, which is a matrix multiplication of the recommended transformation matrices and settings pre-coder, and said transformation matrix pre-encoder limits the number of dimensions of the channel in question referred to the recommended matrix settings advanced encoder, and referred to the recommended matrix settings advanced encoder coordinates referred to the recommended matrix pre-coder with effective the channel between the said first and second devices (10,
14), which is partially set mentioned recommended by the transformation matrix pre-coder; and

transmitter (42), configured to send the mentioned feedback (20) well pre-coder of the said first device (10) for instructions mentioned recommended matrix pre-coder for the said first device (10).

35. The second device (14) p. 34, which referred to a second device (14) configured to maintain a two-dimensional table (28) possible matrices pre-coder, and each matrix can be selected as mentioned recommended matrix pre-coder, and each row or column of the referenced table (28) corresponding to a particular matrix in the set of possible matrices (24) the transformation of the pre-coder and each column or row of the referenced table (28) corresponding to a particular matrix in the set of possible matrices (26) setting pre-coder, and mentioned feedback (20) well pre-coder contains the value of the index table identifying the specific matrix of these possible matrices pre-coder as mentioned recommended matrix pre-coder.

36. The second device (14) p. 34, and cited Otoe second device (14) configured to send the mentioned feedback (20) well pre-coder through the alarm mentioned recommended pre-coder conversion from the first granularity in time or frequency and alarm systems mentioned recommended pre-coder settings with the second granularity in time or frequency, and mentioned the first granularity is coarser than said second degree of detail.

37. The second device (14) p. 34 or 36, in which is mentioned a second device (14) configured to maintain one or more code books (22) containing the set of possible matrices (24) the transformation of the pre-coder, and each has dimension rows-columns N_{T}×k, where the number of rows N_{T}equal to the number of ports (56) of the transmission antennas in the first device (10), and the number of columns k is equal to the dimension of the transformation, which is less than the value of N_{T}in order to limit the number of dimensions of the channel in question referred to the recommended matrix settings pre-coder.

38. The second device (14) p. 37, in which at least one of the mentioned possible matrices (24) the transformation of the pre-coder contains a block-diagonal matrix.

39. The second device (14) p. 38, in which one or more code books (22) additionally contain the set of possible matrices (26) configure advanced encoder, and at least on the on mentioned possible matrices (26) setting pre-coder has a row of the matrix, which alter the phasing of the blocks in the above-mentioned block-diagonal matrix.

40. The second device (14) p. 37, in which the said dimension k conversion is configured referred to the first device (10) or referred to the second device (14) and which, if referred to the dimension k of the transform is configured referred to the first device (10) referred to the second device (14) configured to receive instructions mentioned dimension k conversion from the first device (10).

41. The second device (14) p. 40, which referred to the recommended transformation matrix pre-coder selects the first-mentioned device (10), and referred to a second device (14) configured to receive instructions mentioned recommended the transformation matrix pre-coder of said first device (10).

42. The second device (14) p. 34 or 36, which referred to a second device configured to maintain one or more code books (22), and one or more code books (22) include the set of possible matrices (26) setting pre-coder in such a way that the number of unique vectors that form a specific row of the above set of possible matrices has preliminarily setting the th encoder, is greater than the number of unique vectors forming another row of the above set of possible matrices settings pre-coder.

**Same patents:**

FIELD: radio engineering, communication.

SUBSTANCE: invention relates to wireless communication systems, more specifically to communication between a primary station and one or more secondary stations in a multiple input and, multiple output mode. The method comprises a step where the primary station transmits to the first secondary station an indication of an integration matrix during reception, which the first secondary station must use when integrating signals received at the said plurality of antennae thereof from the first subsequent transmission from the primary station.

EFFECT: improved communication.

13 cl, 2 dwg

FIELD: radio engineering, communication.

SUBSTANCE: invention relates to wireless communication. A method and a device for data transmission based on the interrelation between the first and the second channel are shown. The method might involve measuring the first channel, which corresponds to the first antenna of a wireless terminal, as well as measuring the second channel, which corresponds to the second antenna of the said terminal. The method might also involve the determination of the interrelation between the first and the second channels based on measuring both of them. The method may also involve transmitting the data related to the upperlink transmission. And the said data might be based upon the said interrelation.

EFFECT: demand of UE in order to promote the decision making based upon some measurements of the signals received by multiple antennas at a UE side.

20 cl, 5 dwg

FIELD: radio engineering, communication.

SUBSTANCE: methods and devices are provided wherein user equipment transmits using at least two uplink transmit antennae and receives a set of control signals in the downlink direction from a cellular network. The user equipment estimates a received signal quality for each control signal in said set of control signals and determines, based on said received signal quality, which control signals that have been reliably received. The user equipment derives one or more parameters related to the uplink transmit diversity operation using a subset of control signals from the set of control signals. Said subset only includes control signals determined as reliably received and transmits in the uplink direction applying the derived one or more parameters to control the uplink transmit diversity operation.

EFFECT: invention improves the accuracy of the transmit diversity parameter values derived/set by the UE, which enhances the performance of the uplink transmit diversity and also reduces interference in neighbour cells.

32 cl, 4 dwg

FIELD: radio engineering, communication.

SUBSTANCE: invention relates to wireless communication systems. A re-entry method includes steps of receiving, by a base station of a wireless communication network, a message from a mobile station which includes an indication that the mobile station is in coverage loss recovery mode, and a mobile station identifier to identify the mobile station. The method further includes a step of determining whether a static context and/or a dynamic context associated with the mobile station identifier is stored at a previous serving base station of the mobile station and transmitting a message to the mobile station to indicate which re-entry actions are to be performed to facilitate re-entry of the mobile station into the wireless communication network.

EFFECT: simple procedure of re-entry into a network.

26 cl, 7 dwg

FIELD: radio engineering, communication.

SUBSTANCE: invention relates to wireless communication. The invention discloses, in particular, a method of transmitting control information in a WLAN, which comprises transmitting first control information by beam formation with diversity with cyclic shift delay and transmitting second control information. The first control information comprises information needed for each of a plurality of target stations from the second control information in order to receive the second control information. The second control information is transmitted to the plurality of target stations by beam formation.

EFFECT: invention is intended to provide control information and transmit frames in a wireless local area network (WLAN) which supports a multiple-antenna technology at the transmitting side and the receiving side for multiple users.

29 cl, 38 dwg

FIELD: radio engineering, communication.

SUBSTANCE: invention relates to multi-user multiple input multiple output (MU-MIMO) wireless communication systems. An aspect of the invention consists in improving an apparatus and a method for providing and using control information in a mobile communication system. The invention particularly discloses a method for outputting control information in MU-MIMO wireless communication systems, which includes receiving a plurality of resource elements (RE) including downlink control information (DCI), determining, using the DCI, a set of RE to which a plurality of downlink reference signals (DRS) may be mapped, determining remaining RE as RE to which data are mapped, and demodulating the data using a precoding vector of a DRS corresponding to the MS.

EFFECT: improved communication.

12 cl, 39 dwg

FIELD: radio engineering, communication.

SUBSTANCE: invention relates to wireless communication using a multiple-user multiple input, multiple output (MU-MIMO) system and discloses a method for communication in a network including a primary station and at least a first secondary station, wherein the first secondary station transmits to the primary station an indication of a first set of precoding vectors, and the number of first precoding vectors is greater than a preferred rank of transmission from the primary station to the first secondary station.

EFFECT: improved communication.

17 cl, 2 dwg

FIELD: radio engineering, communication.

SUBSTANCE: invention relates to a wireless mobile communication system and is intended to improve system performance by reducing signalling overhead. The method comprises steps of: setting a rank of uplink control information to a rank of uplink data; multiplexing a first control information item output from the control information with the data; channel interleaving the multiplexed output with control information other than the first control information item from said control information; and transmitting the interleaved signal.

EFFECT: invention discloses a method of transmitting uplink data and control information in a wireless mobile communication system that supports multiple transmitting antennae and multiple receiving antennae (MIMO).

14 cl, 18 dwg

FIELD: radio engineering, communication.

SUBSTANCE: during precoding, channel coherence and system capacity are taken into account. A base station adjusts phase and/or amplitude of a precoding matrix corresponding to each precoded unit to maintain coherence of associated information of the entire precoding channel. The associated information of the precoding channel includes, for example, channel status information (CSI) or eigenvalue matrix of the precoding channel. Further, a mobile terminal performs channel estimation based on reference signals of multiple precoded units, thereby eliminating the limitation in prior art that a mobile terminal can perform channel estimation only within one or more resource blocks limited by a precoding granularity.

EFFECT: high accuracy of precoding.

15 cl, 11 dwg

FIELD: radio engineering, communication.

SUBSTANCE: radio communication method includes periodic calibration in each calibration interval to obtain a Node B calibration vector, wherein periodic calibration involves selecting a group of user equipment UE for calibration, and groups of UE are selected based on channel quality indicators (CQI) received from said UE; and forming a beam pattern for at least one UE in each calibration interval and applying said calibration vector, obtained for said calibration interval in which periodic calibration involves, in each calibration interval, calculation of at least one initial calibration vector for each UE in the selected group, and calculation of a Node B calibration vector based on the initial calibration vectors for all UE in the selected group.

EFFECT: high quality of radio communication.

11 cl, 13 dwg

FIELD: mobile communications.

SUBSTANCE: proposed distributed system of intelligent antennas has N antennas, N radio-frequency transceivers, main frequency band digital signal processor in base station of wireless communication system, feeders, and data bus; N antennas and N radio-frequency transceivers are grouped to obtain a number of radio-frequency transceiver groups disposed at different consistent-reception locations of base station, including different buildings or different floors of one building.

EFFECT: improved consistency of reception.

12 cl, 4 dwg

FIELD: communication systems with distributed transmission, in particular, method and device for non-zero complex weighing and space-time encoding of signals for transmission by multiple antennas.

SUBSTANCE: method and device provide for expansion of space-time block code N×N' to space-time block code M×M', where M>N, with utilization of leap-like alternation of symbols phase in space-time block code N×N', to make it possible to transfer space-time block code from distributed antennas in amount, exceeding N'.

EFFECT: distribution of transmission from more than two antennas.

2 cl, 16 dwg

FIELD: automatic adaptive high frequency packet radio communications.

SUBSTANCE: each high frequency ground station contains at least one additional high frequency receiver for "surface to surface" communication and at least one additional "surface to surface" demodulator of one-tone multi-positional phase-manipulated signal, output of which is connected to additional information input of high frequency controller of ground station, and input is connected to output of additional high frequency "surface to surface" receiver, information input of which is connected to common high frequency receiving antenna, while control input is connected to additional control output of high frequency controller of ground station.

EFFECT: prevented disconnection from "air to surface" data exchange system of technically operable high frequency ground stations which became inaccessible for ground communications sub-system for due to various reasons, and also provision of possible connection to high frequency "air to surface" data exchange system of high frequency ground stations, having no access to ground communication network due to absence of ground communication infrastructure at remote locations, where these high frequency ground stations are positioned.

2 cl, 12 dwg, 2 tbl

FIELD: radio engineering, in particular, signal transfer method (variants) and device for realization thereof (variants), possible use, for example, in cellular radio communication systems during transmission of information signal in direct communication channel from backbone station to mobile station.

SUBSTANCE: technical result is achieved due to correction of spectrum of copies of information signal being transferred, transferring copies of information signal from each adaptive antenna array in each effective transmission direction, estimating transfer functions of direction transmission channels on basis of pilot signals transferred from each antenna element, on basis of pilot signals for distributed transmission, sent from each adaptive antenna array in each one of effective transmission directions, and also combining two given estimates.

EFFECT: increased efficiency of transmission of information signal in direct communication channel, and, therefore, maximized quality of receipt of information signal at mobile station.

5 cl, 10 dwg

FIELD: wireless communication receivers-transmitters and, in particular, wireless communication receivers-transmitters which use a multi-beam antenna system.

SUBSTANCE: when controlling a multi-beam antenna system for a downstream line of wireless communications, generation of polar pattern and signaling of distribution during transmission in closed contour are combined, each beam signal is adjusted at transmitter on basis of check connection from wireless communication mobile station in such a way, that signals received by wireless communication mobile station may be coherently combined.

EFFECT: increased traffic capacity and productivity of the system, improved power consumption, cell coverage and communication line quality characteristics.

2 cl, 5 dwg

FIELD: method for estimating a channel in straight direction in radio communication systems.

SUBSTANCE: in accordance to the method, in straight direction, beam is created by a set of antennas, and at least one vector is created for beam generation, subject to application for connection of at least one base station to at least one mobile station, is defined by at least one client station and from at least one client station to at least one base station information is transmitted, which contains information about aforementioned at least one beam generation vector. In accordance to the invention, at least one base station transmits information about beam generation vector used for connection of at least one base station to at least one client station, to at least one client station, on basis of which at least one client station estimates the channel.

EFFECT: transmission of information about beam generation vector due to selection and transmission of pilot-signal series.

2 cl, 2 dwg

FIELD: physics, communications.

SUBSTANCE: invention concerns data transfer, particularly frequency-time-space block coding in a transmitter with three transmitting Tx antennae. Input symbol sequence is transferred by three Tx antennae according to permutation method via selected transmission matrix.

EFFECT: increased data transfer speed.

28 cl, 10 dwg

FIELD: physics; communications.

SUBSTANCE: present invention pertains to communication techniques. The transmitting object carries out spatial processing using control matrices, so that data transmission is held in a set of "effective" channels, formed on the real channel used for transmitting data, and control matrices, used for PRTS. Control matrices can be formed by sampling a base matrix, which can be a Walsh or Fourier matrix. Different combinations of scalars are then chosen, each combination of which consists of at least one scalar, of at least row of the base matrix. Each scalar can be a real or complex value. Different control matrices are formed by multiplying the base matrix by each of the different combinations of scalars. Control matrices are different transpositions of the base matrix.

EFFECT: generation and use of control matrices for pseudorandom transmission control (PRTS).

55 cl, 3 dwg, 1 tbl

FIELD: physics.

SUBSTANCE: invention is related to device and method for beams shaping in telecommunication system of mobile communication CDMA with application of intellectual antennas technology, using specified device and method, multiple fixed beams are shaped in sector, and multiple fixed beams are used to shape traffic channel with narrow beams and common channel with sector beams in one and the same intellectual antenna system, and problem of phases discrepancy is solved in appropriate channels due to differences in time and temperature oscillations without application of complicated correcting technology. Since fixed beams in some area correlate and interact with each other, or considerably weaken due to correlative summation of space vectors of every fixed beam in process of common channels transfer in CDMA system with multiple antennas, then appropriate ratio is established between power of pilot channel and traffic channel in coverage area, and signal-noise ratio is increased for signals received by mobile communication station. As a result of addition of optical transceivers system between system of the main frequency band and system of radio frequency transceivers (TRX), the main frequency band system services more sectors. Radio frequency unit is located in close proximity to antennas, and consumed power is reduced accordingly.

EFFECT: increased throughput capacity and efficiency of CDMA system with multiple antennas.

15 cl, 6 dwg

FIELD: information technologies.

SUBSTANCE: separation of transmitting antennas with feedback is applied to special channel of downstream communications line, and separation of transmitting antennas without feedback is applied to control channel of downstream communications line in accordance with high-efficiency method of transmission over upstream communications line. The objective of present invention is to determine how the station with separation during transmission which implements advanced upstream communications line (EUL) should apply separation of transmitting antennas to level 1 confirmation information transmission channels (E-HICH), relative transmission rate channels (E-RGCH) and absolute transmission rate channels (E-AGCH).

EFFECT: providing communications system stability and reliability.

9 cl, 15 dwg