Method and apparatus for controlling multi-antenna transmission in wireless communication network. RU patent 2474048.

FIELD: information technology.

SUBSTANCE: multi-antenna transmission control presented herein involves generating a set of virtual channel realisations in a transmitter (10) which shares the same second-order statistics as the actual channel realisations observed for a targeted receiver (12). By making the control-related quantities of interest at the transmitter (10) depend on the long-term statistics of the channel, the actual channel realisations are not needed for transmission control, e.g., for accurate multiple-input/multiple-output (MIMO) precoding and the related modulation and coding choice.

EFFECT: use of virtual channel realisations enables transmission control which approaches the closed-loop channel capacity which would be provided by full feedback of the channel status information without requiring the overhead signalling burden which accompanies full feedback.

17 cl, 4 dwg

The level of equipment

The technical field to which the invention relates

The present invention, in General, applies to wireless communication systems and, in particular, relates to the management of mnogoetajnoe transmitted in the wireless network connection, for example, managing the operation of pre-coding and selection of modulation and encoding speed of the transmission channel with multiple inputs and multiple output (MIMO).

The level of equipment

The availability of specific information (dissemination) the state of the channel to the transmitter plays a critical role in achieving the highest possible spectral efficiency for wireless communication system with numerous transmitting antennas. For example, E. Telatar, “Capacity of multi-antenna Gaussian channels”, Euro. Trans. Telecomm., vol. 10, no. 6, pp. 585-596, November 1999, shows that significant gains in bandwidth can be obtained with multiple antennas, when accurate information about the instantaneous state of the channel is available at the transmitter.

Feedback is instantaneous States of the channel from the target receiver transmitter is known mechanism for providing accurate information about the state of the channel, and it may be necessary, for example, in duplex systems with frequency division (FDD), in which the instantaneous States of the channel in the ascending line and downlink not directly connected. However, problematic potential number and complexity of distribution channels existing in (MIMO) systems, may require significant amounts of feedback channel, which may be feasible and desirable in many cases. In addition, even starting with the dubious assumption that the receivers can estimate the instantaneous States of the channel with the required accuracy, delayed feedback, which includes computational delays and delays in signal transmission, ensure that the feedback channel is received from the transmitter, leads to delay the actual conditions observed in the receiver. Essentially, adjust the transmission does not correspond to the actual instantaneous States of the channel in the target receivers.

As you exit the use of instantaneous States of the channel, as a basis for management mnogoetajnoe transmission, instead, some research has examined the optimal transmission schemes that use long-term statistical information on the channel (channels) of the distribution. Unlike information about the instantaneous state of the channel, which varies with speed fast fading, statistical information on the channel can be changed with a much slower speed (for example, with the speed of a slow fading (shadowing), or the rates of change of the corners of dispatch/receipt). Consequently, much more acceptable from the point of view of computational overhead and administrative costs for the alarm, just send feedback statistical information on the channel from the target receivers for the appropriate management mnogoetajnoe transmission.

Despite the fact that the Foundation of the transmission control feedback on statistical data channel reduce the volume and complexity alarm feedback channel from the target receivers, the implementation of such management has its own complications. In practice, the calculation of the various parameters that are required for optimal mnogoetajnoe transmission, on the basis of statistical data channel is often considerably more difficult than solving them based on instant information about the channel.

For example, a number of articles presents information related to the definition of the optimal linear matrix preliminary coding F

that maximizes ergodic bandwidth flat with MIMO n transmitting antennas and n a reception antenna. Such items include E. Visotsky and Madhow, “Space-Time Transmit Precoding with Imperfect Feedback, IEEE Trans. on Info. Thy., vol. 47, pp. 2632-2639, September 2001; S.H. Simon & A.L. Mavrobara, Optimizing MIMO Antenna Systems with Channel Covariance Feedback”, IEEE JSAC, vol. 21, pp. 406-417, April 2003; and A.M. Tulino, A. Lozano, S. Verdu, “Capacity-Achieving Input Covariance for Single-User, Multi-Antenna Channels”, IEEE Trans. on Wireless Comm., vol. 5, pp. 662-671, March 2006.

In accordance with these different materials F

can be calculated as

Equation (1)

More specifically, it is shown that optimal matrix pre-coding can be written as

Equation (2)

where U denotes the matrix columns which are vectors H E

H , D ( , , ..., ) denotes a diagonal matrix with { } as the diagonal elements, and where p denotes the fraction of the power, distributed in the j-th own transfer mode, which corresponds to the j-th column of the U .

In the context of the above-mentioned General scheme is shown that the relative power levels {p } must satisfy the following conditions:

Equation (3)

Equation (4)

Equation (5)

Equation (6)

} - column vectors converted channel = HU =[

]. It should be noted that as a member of the MMSE depends on {p } , the relative power levels {p } defined only implicitly.

To calculate {p } is suggested an iterative algorithm based on the joint probability distribution, given by p () (or alternatively, the joint probability distribution H ). As a first step, the algorithm sets {p } such that = 1 (for example, by setting p = 1/n for all j). Then the algorithm iterates equations with fixed decimal point up until the solution converges:

Equation (7)

for j=1, 2, ..., n

where MMSE calculated on the basis of equation (4) with p set to p .

At this point the algorithm stops if, for each j, such that p has converged to zero at the above stage,

Equation (8)

Otherwise, set p =0 for j, which corresponds to the smallest value E[SINR ].

Of the project implementation, which includes the equation (7) and equation (8), requires the calculation of several significant quantities, including

Equation (9)

. Equation (10)

Calculation of equations (9) and equation (10) requires the joint probability distribution p( a ) instantaneous States of the channel that is difficult, if not impossible, to determine even in the receiver, not to mention the transmitter. Despite the fact that the integrals that are included in these equations are of interest, can be approximated by averaging over many implementations observed in the receiver, this approach includes additional complexity. Because of the magnitude of interest, depend not only from , but also from distributed power levels {p } , these values should be estimated for different values of {p } in order to calculate the optimal power levels. In the result of multiple and/or large sets of implementations would need to keep working memory (for example, RAM) target receiver. In practice, however, it is undesirable to require enough memory and processing power in the target receivers in order to perform the above algorithm to calculate the optimal values of pre-transfer-encoding.

In addition to the calculation of weighting coefficients preliminary coding for mnogoetajnoe transfer, select the appropriate modulation and speeds of coding channel for each thread transfer on the basis of statistical data channel to the present time is considered enough. Such considerations will depend on the type of the detection algorithm, for example, serial maturity interference (SIC)used in the target receiver (receiver).

Summary of the invention

Management mnogoetajnoe transmission, provided in this application, involves the generation of a set of virtual channel implementations in the transmitter, which shares the same statistical data of the second order, and that the actual implementation of the canal, the observed for the target receiver. By forming the values associated with the management of interest, at the transmitter, are dependent on long-term statistical data, the actual channel implementations are not required to manage the transfer, for example, prior to encoding with multiple inputs and multiple output (MIMO) and bound by the choice of modulation and encoding. Essentially, the use of virtual channel implementations allows transmission control, which is approaching bandwidth closed loop, which would have been provided with full feedback information about the state of the channel, without the requirements of the load on the service alarms which accompanies the full feedback.

In one or more variants of implementation method, designed to manage the transfer of transmitter contains the generation of the initial set of “boilerplate” channel implementations, which is preferable distributed Gaussian, the definition of the statistical data of the second order for actual implementations of the channel in the target receiver and adaptation of the initial set of predefined channel implementations as statistical data of the second order of the channel to get a virtual channel implementations, which reflect the statistical data of the second order of the actual channel implementations. The method further includes the definition of one or more parameters of the transmission control functions as virtual channel implementations for transmission control in the target receiver.

The aforementioned method and its variants carried out in accordance with one or more choices of implementation presented in this application, transmitter, configured for use on the wireless network connection. Non-limiting examples include a cellular communication network of the 3G and 4G. For example, one or more schemas processing, for example, logic circuits based on microprocessors, or other programmable logic processing can be configured to run any version of the method presented in this application. These schemes can be implemented, for example, in a network base station or other transfer site.

Figure 1 - block diagram of option implementation of a wireless communication network, which includes a transmitter, a controller transfer, configured to define one or more parameters of the transmission control based on the statistical data of the second order of the channel.

Figure 2 - block diagram functional elements of the scheme for one or more options for the implementation of controller transfer, such as, for example, is depicted in figure 1.

Figure 3 - a logical block diagram of the sequence of stages one variant of the method of transmission control based on the statistical data of the second order of the channel that can be implemented in a logical scheme of handling the controller transfer.

Figure 4 - the following chart illustrates an example of the relative effectiveness of one or more options for the implementation of transmission control as provided in this application.

Detailed description of the invention

Figure 1 illustrates one way of implementation transmitter 10 to transfer in many target wireless devices 12 communication. Wireless devices 12 communication (abbreviated in the figure as “WCD”) may not all be the same, and you should understand that they are a lot of possible types of devices, such as cell phones, pagers, portable digital assistants, computers, network access map or combination of any of these devices. Essentially, for the balance of this discussion, they are simply referred to as “receivers 12”.

At least, in one embodiment, the transmitter 10 is the base station or another host transceiver in a wireless network 14 communication and supports wireless communication downlink (DL) and uplink connection (UL) in the 12 receivers and receiver 12. At least, in one embodiment, the transmitter 10 configured to work with multiple inputs and a single output (MISO) or multiple inputs and multiple output (MIMO)and receivers 12, respectively, with one or more of a reception antenna 18.

When working MIMO signal (signal)intended for one specific receiver receivers 12, passed from selected antennas antenna, 16, and certain of antenna used for transmission and distribution of relative power transmission can be (and usually is changed) modified dynamically depending on a number of considerations. Of particular interest in this application transmitter 10 includes controller 20-transfer”, which establishes and regulates or otherwise controls one or more parameters of transmission used radio frequency (RF) circuits 22 transceiver transmitter 10. As detailed in this application as an example, management is mainly based on the use of statistical data of the second order of the channel associated with the actual implementations of the channel in the target receivers 12.

Thus, at least in one embodiment, the transmitter 10 contains a transmitter MIMO configured for a wireless network connection, in which he identifies one or more parameters of the transmission control as a function virtual channel implementations for transmission control (MIMO) in the receiver 12 calculation of weighting coefficients pre-transfer-encoding and speeds of coding based virtual channel implementations for this, this receiver 12. Weights pre-transfer-encoding establish a distribution transmission power at the antenna, used for transmission to the receiver 12 of the corresponding antennas of two or more transmitting antennas 16.

In more detail, consider the following simple model of a received signal baseband frequency with the channel flat MIMO:

Equation (11)

where H denotes the channel response MIMO (matrix n

n ) with zero mean value, r denotes the received signal, s denotes the transmitted signal, and w denotes the component noise plus interference in wireless communication system with n transmitting antennas and n a reception antenna. Component noise w can be spatially painted by receiving antennas using the covariance matrix R

}, where E { * } denotes the expected value within the parentheses.

For the purposes of discussion, at least, this receiver receivers 12 may obtain accurate estimates of H channel relative to the transmitter 10 and the covariance of the noise R

(Also, when the noise has a non-zero mean value, the receiver 12 defines the covariance and the average value of noise). As a useful working definition of “given to white noise the channel response” to this receiver 12 can be determined from H and R

Equation (12)

At least in one aspect of the transmission control, as shown in this application, the receiver 12 sends feedback to the statistical data of the second order about his refer to white noise the channel response. In turn, the transmitter 10 uses statistical data of the second order to form a set of “virtual channel implementations. The transmitter uses 10 virtual channel implementations to define one or more parameters of the transmission control, such as the distribution of power transmitting antenna for MISO, or MIMO transmission, the choice of the scheme of preliminary coding and/or the type of modulation and coding (MCS)that maximize bandwidth communication line.

In accordance with one or more variants of the implementation of the transmitter 10 stores or otherwise holds a pre-computed values that contain, or otherwise represent a set of independent identically distributed (IID) sample matrix, distributed Gaussian marked with { H

} of dimension n on n . The variable N indicates the number previously stored samples that can be made longer as required, i.e. could be potentially larger the sample set.

From one perspective, consider the sample IID, distributed Gaussian, as “template” model or a default actual channel implementations in this receiver 12, which is not available in the transmitter 10. However, the scale matrix S of size n x n and identity matrix U of size n n n n n calculated and used to resize and convert a single sample matrix H

the virtual channel implementation H

. That is, in these cases the initial set of template-based implementations of channel used for a virtual channel implementation, contains a set (IID) samples, distributed Gaussian.

Mathematically each virtual channel implementation generate in accordance with

Equation (13)

where collectively account “ • ” denotes the product components two matrices a and b with the same size, vec( A ) denotes the vector generated by “to link all of the columns in A single vector, and mat ( X ) denotes the matrix m by n, formed by re-forming mn-dimensional vector X .

In particular, the above mentioned virtual channel implementations create from template data, i.e. IID samples matrix, distributed Gaussian, which can be pre-calculated and withheld in memory (or generated once on the run at startup or on demand or desire). In more detail, to create them, without any requirements to knowledge of anything about the actual state of the channel (although I admit that the model of the Gaussian distribution is accurate).

Although it is not a limiting example, figure 2 reveals an implementation option controller 20 transmission, presented in figure 1, which is beneficial, at least in some circumstances. For example, at least one variant of implementation of the transmitter 10 includes one or more schemas 21 on microprocessors, which may include the universal or specialized microprocessors, digital signal processors and other type (types) digital logic circuits processing. At least, in one embodiment, the controller 20 transmission contains one or more such digital processors, which are programmed with the ability to implement transmission control as a function of the statistical data of the second order of the channel. For example, the controller transfer can include memory or to access memory that stores software instructions, the implementation of which makes the controller 20-transfer to execute the method. It is also expected to conduct all or part of the required processing transmission control in FPGA (FPGA) or other programmable element (elements).

Having in mind the above, figure 2 illustrates the functional unit schema for one or more schemas processing controller 20 transfer, including calculate your 22 scaling/conversion, device 24 scaling Converter 26 and selectively includes devices 28 vectorization and generator 30 matrix. Controller 20 transfer additionally includes the device memory or associated with the device, for example, one or more devices 32 memory to save the initial set of predefined channel implementations, such as set the IID of samples distributed Gaussian in { H

} that is considered for one or more options for implementation.

When working calculate your 22 scaling/conversion calculates scale matrix S and transformation matrix U

(which can be based on f

described below). According to one embodiment of the full covariance matrix is given to white noise channel, specified by

Equation (14)

and make available for the transmitter 10 through feedback from the receiver 12. In this case, a single transformation matrix is chosen as U

denotes the matrix columns which are vectors f

denotes the identity matrix n, n , and

means the product is presented. Using this expression scale matrix S is removed from the square root of the components of the vectors f

Equation (17)

for all i {1, 2, ..., n } and j {1, 2, ..., n }, where

denotes the k-th eigenvalue of f

for k=1, 2, .... It should be noted that in this embodiment, the stages vectorization and forming a matrix can be skipped, i.e. items 28 and 30, depicted in figure 1, can be ignored. This pass is allowed because the virtual channel implementations can be generated just as

Equation (18)

can also be obtained from the full matrix of covariances of the channel f

. Namely, the element f

in the i-th row and j-th column is asked by delineating the relevant pieces and uses of n on n on f

Equation (19)

where [ A ] denotes A , which consists of the elements of the m-th line in the n-th row and the l-th column on the k-th column a, inclusive.

In one or more variants of implementation controller 20 transfer configured to use a set of virtual channel implementations { } in an iterative algorithm described in equations (7) to (10). That is at least one version of the exercise of the controller 20 transfer determines optimum distribution capacity for transmitting antennas for 16 different modes of transmission of MIMO in the receiver 12 on the basis of an appropriate set of virtual channel implementations { } , as revealed in the statistics of the second order of the actual channel implementations for this, this receiver 12.

In more detail, receiving virtual channel implementations from a set of initial channel implementations, distributed Gaussian default, provides a set of samples channel implementations in the transmitter 10, which reflect the statistical data of the second order of the actual channel implementations and, thus, can be used to assess the essential values {E[MMSE ]} and {E[SINR ]} as follows

Equation (20)

Equation (21)

for j=1, 2, ..., n , where

denotes the j-th column of the transformed matrix

It is clear that the use of the virtual channel implementations { } provides controller 20 transfer of a set of samples of sufficient size to exactly approximate integration, presented in equation (9) and equation (10), through summation presented in equation (20) and equation (21). This feature is particularly useful considering the fact that equation (9) and equation (10) depend on the fame of the probability distribution function, the actual channel implementations, p( ) in the transmitter 10, and they usually not known until then, still not used burdensome full feedback channel. Thus, when the controller 20 transfer configured to calculate power distribution for pre-transfer-encoding on the basis of one or more integrated, as to the probability density function of the actual channel implementations, it may be mostly configured with the possibility of approximation of this integration means of averaging over the set of samples of some or all of the virtual channel implementations.

Of course, the controller 20 transfer may base its determination of parameters of the transmission control other than the weights pre-transfer-encoding, virtual channel implementations. For example, in addition or in the alternative to the determination of the weights pre-transfer-encoding, controller 20 transmission can be configured to base their choice of modulation and coding (MCS) to this receiver 12 virtual channel implementations, for certain that this receiver 12. As an example, long-encoding speed on the thread {R } to use in the transfer of MISO/MIMO in this unit 12, working with serial maturity interference (SIC), can be computed from {

The equation (22)

for j=1, 2, ..., n .

More broadly, it should be understood that the definition of matrices pre-transfer-encoding and/or execution of elections MCS based virtual channel implementations remain in force as the primary consideration, but not limited examples transmission control, as shown in this application. Figure 3 illustrates an implementation option a broad manner in which the controller 20 transfer can be programmed or otherwise configured to run. Illustrated processing involves the given sequence of stages, but controlling the transmission, as provided in this application, is not necessarily limited illustrates the sequence. In addition, it should be understood that all or part of the illustrated by the treatment can be carried out on the continuous basis or recurring basis, and may be part of a larger set of processing operations control processing/communication in the transmitter 10.

Having in mind the above points, illustrated way transmission control from transmitter, for example, the transmitter 10, involves the generation of the initial set of “boilerplate” channel implementations, which preferably is distributed Gaussian (step 100), the definition of the statistical data of the second order for actual implementations of the channel in the target receiver (phase 102), for example, in this one receivers 12 and the adaptation of the initial set of predefined channel implementations as a function of the statistical data of the second order channel to get a virtual channel implementations, which reflect the statistical data of the second order of the actual channel channel implementations (phase 104). The method further includes the definition of one or more parameters of the transmission control functions as virtual channel implementations for transmission control in the target receiver (phase 106).

As mentioned, the generation of the initial set of predefined channel implementations may contain generation them out stored values, for example, forming a matrix with elements taken from a previously saved set of independent identically distributed (IID) sample Gauss. Set IID samples matrix, distributed Gaussian { H

} use in one or more variants of implementation, and { H

} can be generated from a pre-computed saved values. That is, sampling { H

} can be saved, and copies can be downloaded in working memory as needed.

It would not have been generated, the template implementation channel represented by { H

} , adapt by means of scaling and transformation to reflect the statistical data of the second order actual implementations of the channel in the target receiver 12 therefore, as explained in the context of equation (13). That is, in one or more variants of implementation controller 20 transmission uses a scale matrix S of size n x n and identity matrix U

n n n n n in order to resize and convert a single sample matrix H

in virtual channel implementations H

Statistical data of the second order of the actual channel implementations for this unit 12, for example, the covariance given to white noise of the channel response, given in equation (12)can be defined on the basis of receiving feedback from the receiver 12. At least, in one embodiment, themselves statistical data of the second order is passed through feedback. Thus, the receiver can determine its covariance given to white noise response of the channel and send feedback for this information transmitter 10. Alternatively, the transmitter 10 may determine the statistical data of the second order on the basis of observations known signal from the receiver 12. For example, the transmitter can retrieve statistical data of the second order of measurements made on the pilot signal uplink connection (or another known signal)transferred from the receiver 12.

It should also be stressed that these methods are directly applicable when responses channel MISO/MIMO interest, have a non-zero mean value. For example, in such cases, in addition to the covariance channel (statistical data of the second order of the channel), transferred through feedback using this receiver 12, or otherwise determined for him, this unit 12 can also send feedback long-term average or mean value of the response of the channel (statistical data of the first order). Set virtual channel implementations can be generated in the same way as in the case of zero mean value, except that the average response channel add in IID Gauss sampling matrix { H

} to scaling and unit conversion.

Controlling the transmission, as provided in this application, additional easily extends to the cases when the feedback channel MISO/MIMO of interest are the polling frequency. For example, if you want a single frequency independent matrix preliminary coding controller 20 transmission can be configured to use the processing described in this application on a system with orthogonal frequency multiplexing (OFDM) MIMO using the definition of F.

modified, respectively, as

Equation (23)

Equation (24)

where [k] is given to white noise the channel response in the frequency domain in the kth subcarrier, [n] indicates the corresponding n-th branch of the channel in the time domain, N denotes the number of subcarriers in the system, and L denotes the maximum number of branches of the channel in the time domain.

In any of the different options for the implementation of the management mnogoetajnoe transmission as a function of the statistical data of the second order of the channel provides numerous benefits. For example, using statistical data of the second order for actual implementations of the channel, in order to adapt the original set channel implementations, the default distributed Gaussian, so that adapted channel implementation reflects the statistics of the second order, potentially computationally intensive and requires memory tasks calculate the optimal long-term values of the matrix preliminary coding and appropriate long-term rate encoding on the thread for MIMO transmission can be performed more in the transmitter, than in the receiver. The approach also allows target receivers to convey through their feedback the covariance matrix is given to white noise channel (or covariance plus the average value of zero sinking average).

This type of feedback is relatively compact information with reduced non-productive costs alarm. In addition, the usual statistical information provided by the statistical data of the second order of the channel (and the first order), can also be used for other purposes, such as compression quality information channel (CQI), which usually passed through feedback through a fast line feedback. In addition, at least in some variants of implementation, particularly where the ascending and descending lines of communication are well correlated, the transmitter can determine necessary statistical data on the basis of observations of known signals, devotees from the target receiver (receivers).

Despite the effectiveness of the signalling and the computational efficiency, effectiveness transmission control based on the statistical data of the second order, as provided in this application, preferably compared with the ideal bandwidth closed loop proposed when the matrix pre-transfer-encoding MIMO calculate from full instant feedback channel status. For example, figure 4 illustrates the performance graph, which shows the efficiency of the transmission control as provided in this application, for MIMO-OFDM. Suppose that the entire bandwidth of the system is 5 MHz, while the size of the fast Fourier transform (FFT), 512. The number of employed subcarriers is 300, which are divided into 25 pieces of data, each of the 12 sub-carriers. In addition, the distance between the sub to 15 kHz, productivity simulated using spatial 3GPP model of a channel with the common profile of the channel in the environment of microwells.

Having in mind the above assumptions and models curve “+” is ergodic throughput achieved through optimal matrix preliminary coding, as calculated using actual (instant) channel implementations. Curve “x” represents the ergodic throughput that can be achieved using the matrix preliminary coding, calculated using virtual channel implementations, generated, as shown in this application, on the basis of statistical data of the second order of the actual channel channel implementations. As shown in the graph, virtual no loss of performance compared to the more onerous actual channel implementations.

Having in mind the above examples, or other variations or extensions of specialists in the given field of technology will understand that the previous description and the accompanying drawings represent non-limiting examples of ways and devices presented in this application to manage the transfer on the basis of statistical data of the second order of the channel. Essentially, this invention is not limited to the previous description and the accompanying drawings. Instead, the present invention is limited only by the following formula of the invention and its legal equivalent.

2. The method according to claim 1, additionally characterized by the fact that the generation of the initial set of predefined channel implementations contains the generation of the initial set of predefined channel implementations of precalculated saved values.

3. The method of claim 2, additionally characterized by the fact that the generation of the initial set of template-based implementations channel from the previously saved values calculated contains the formation of the initial set of predefined channel implementations from a previously saved set of independent identically distributed (IID) Gaussian samples matrix.

4. The method according to claim 3, additionally characterized in that adapting the initial set of predefined channel implementations as statistical data of the second order of the channel to obtain a set of virtual channel implementations, contains formation each virtual channel implementation in the set of virtual channel implementations using scaling and translation, the original template implementation channel in the set of the original template implementations of the channel as a function of the statistical data of the second order such that the set of virtual channel implementations reflects the statistics of the second order of the actual channel implementations.

5. The method according to claim 1, additionally characterized in that the definition of one or more parameters of the transmission control as a function of a virtual channel implementations contains the calculation assignments capacity for pre-transfer-encoding from the set of virtual channel implementations.

6. The method according to claim 5, additionally characterized by the fact that calculation of assignments capacity for pre-transfer-encoding from the set of virtual channel implementations contains iterative calculation of the optimal assignment of the power of the transmitting antenna, depending on the mentioned approximation of integration regarding the probability density function corresponding to the actual channel implementations, based on the aforementioned averaging set of samples of some or all of a set of virtual channel implementations.

7. The method according to claim 1, additionally characterized in that the definition of one or more parameters of the transmission control as a function of a virtual channel implementations contains the type of modulation and coding (MCS) for the target receiver (12) on the basis of a set of virtual channel implementations.

8. The method according to claim 1, additionally characterized in that the definition of the statistical data of the second order for actual implementations of the channel in the target receiver (12) contains the statistical data of the second order on the basis of observations known signal received from the transmitter (10) from the target receiver (12).

9. The method according to claim 1, additionally characterized in that the definition of the statistical data of the second order for actual implementations of the channel in the target receiver contains the definition of the statistical data of the second order on the basis of feedback from the target receiver (12).

10. The method of claim 9, additionally characterized in that the definition of the statistical data of the second order on the basis of feedback from the target receiver (12) contains the reception of the statistical data of the second order from the target receiver (12).

11. The method according to claim 1, additionally characterized in that the definition of the statistical data of the second order for actual implementations of the channel in the target receiver (12) contains the definition of the statistical data of the second order for this to white noise of the channel response to actual channel implementations.

12. The method according to claim 11, additionally characterized in that the definition of the statistical data of the second order for this to white noise of the channel response to actual implementations of the channel contains the definition of covariance given to white noise response of the channel.

13. The method indicated in paragraph 12 additionally characterized in that the definition of covariance given to white noise of the channel response contains the information is received covariance to refer to white noise response of the channel as feedback from the target receiver (12).

14. The method indicated in paragraph 12 additionally characterized in that phase, which will adapt the initial set of predefined channel implementations as a function of the statistical data of the second order of the channel that contains the scaling and transformation of the initial set of predefined channel implementations based on the covariance given to white noise response of the channel.

15. The method according to claim 1, advanced wherein the transmitter (10) is a radio transmitter with orthogonal frequency multiplexing (OFDM)configured for a wireless network connection, and that the definition of one or more parameters of the transmission control as a function of a virtual channel implementations for transmission control in the target receiver (12) contains the calculation of weighting coefficients pre-transfer-encoding for transmission portions OFDM data from adequate antennas of two or more transmitting antennas (16) on the basis of a set of virtual channel implementations.

16. The method according to claim 1, advanced wherein the transmitter (10) is a radio transmitter with many inputs and many output (MIMO)configured for a wireless network (14) connection, and that the definition of one or more parameters of the transmission control as a function of a virtual channel implementations for transmission control in the target receiver (12) contains the calculation of weighting coefficients pre-transfer-encoding for transmission in the target receiver (12) of the corresponding antennas of two or more transmitting antennas (16) based on the set virtual channel implementations.

17. Controller (20) transfer to transmitter (10), wherein the one or more schemes (21) processing, configured to: generate an initial set of predefined channel implementations mentioned initial set of predefined channel implementations contains many template channel implementations, definitions of statistical data of the second order for actual implementations of the channel in the target receiver (12), adaptation of the initial set of predefined channel implementations as statistical data of the second order of the channel to to obtain a set of virtual channel implementations, which reflect the statistical data of the second-order canal actual implementations of the channel, the set virtual channel implementations contains multiple virtual channel implementations, and define one or more parameters of the transmission control as a function of a virtual channel implementations for transmission control in the target receiver (12) on the basis of approximation of integration regarding the probability density function corresponding to the actual channel implementations, using averaging set of samples of some or all of the set virtual channel implementations.