RU2339186C1 - Method and device for channel feedback - Google Patents

Abstract

FIELD: physics, communication.

SUBSTANCE: invention is related to communication systems. Method of channel scores transmission on multiple subcarriers between transmitting device and receiving device is based on the fact that transmitting device determines channel scores on multiple subcarriers, and then codes these multiple scores into at least one coded channel signal, then transmitting device sends at least one coded signal to receiving device.

EFFECT: provides efficient presentation of channel data to transmitter with the purpose of their application in closed transmission.

10 cl, 11 dwg, 1 tbl

Description

FIELD OF THE INVENTION

The present invention generally relates to communication systems and, in particular, relates to a method and apparatus for providing channel information to a transmitter in a transmission system with many inputs and many outputs (MIMO).

State of the art

A system with many inputs and many outputs (MIMO) is a transmission method that includes many transmit antennas and many receive antennas, which in the future promise a significant increase in the capacity of communication lines in wireless communication systems. Different transmission strategies require that the transmitting antenna array has a certain amount of information regarding the channel characteristics between each element of the transmitting antenna and each element of the receiving antenna, and the implementation of such strategies is often called a closed MIMO system. Obtaining full information about the broadband channel on the transmitter side is possible by using technologies such as uplink sounding in time division duplex (TDD), if the antenna array is calibrated in the transmitter. However, the uplink sensing method in frequency division duplex (FDD) systems does not work when a repeater is used in the TDD system or when the antenna array is not calibrated at the transmitter. Thus, there is a need for a method for efficiently providing complete channel information in a transmitter when uplink sounding is ineffective for this purpose.

Brief Description of the Drawings

Figure 1 is a timing chart for feedback in a TDD system;

2 is a timing chart for feedback in an FDD system;

figure 3 - one option for the effective transmission of channel information feedback;

figure 4 is a second variant of the effective transmission of channel information feedback;

5 is an example of using phase shift to efficiently encode channel information;

6 is an example of orthogonal-frequency coding of channel information in the frequency domain;

7 is an example of orthogonal-frequency coding of channel information in the time domain;

Fig - a device for transmitting channel information;

Fig.9 is a block diagram showing the operation of the device of Fig.8;

figure 10 - device for restoring channel estimates;

11 is a block diagram showing the operation of the device of figure 10.

Detailed Description of Drawings

For simplicity, the invention is presented in terms of providing the base station (BS) with channel information when performing closed-loop transmission to a subscriber station (SS). It should be clear that this invention is also applicable to scenarios where the roles of the BS and SS stations are reversed with respect to those described here. For example, the invention can be applied to a scenario where the SS station needs to be provided with channel information in order to be able to transmit from the SS station to the BS station. Therefore, although this description focuses mainly on the case of transmission from a BS station to an SS station, the term “source communication unit” refers to a communication unit (eg, a BS station, an SS station, or another transceiver) that can perform closed-loop transmission to a “communication unit destination. "

In this description, some terms are also used interchangeably. Terms such as “channel response”, “frequency selective channel profile”, “spatial frequency response of a channel” refer to channel response information that a base station needs to use closed-loop transmission technology. This information about channel characteristics can also be called channel information (channel knowledge). The terms waveform and waveform are also used interchangeably. The term “subscriber unit” or “subscriber station” (SS) is sometimes referred to as a mobile station (MS) or simply a mobile phone, the invention being equally applicable to cases where the subscriber unit is stationary or mobile (i.e. non-stationary). The receiving device may be a "base station (BS)", a "subscriber station (SS)", or any combination thereof. The transmitting device may also be a BS station, an SS station, an MS station, or any combination thereof. In addition, if the system has repeaters, repeaters, or other similar devices, then the receiver or transmitter can be a repeater, relay, or other similar device. A repeater or repeater can be considered equivalent to an SS if the BS is transmitting feedback to the repeater / repeater. A repeater or repeater can be considered equivalent to a BSBS station if the repeater is transmitting feedback to the SS station. The relay may also be a unidirectional relay, for example, an uplink relay or a downlink relay. For example, the uplink repeater receives the uplink signal from the SS station and relays the signal or information represented by this signal to the BS station. The term “fast Fourier transform (FFT)” and the term “inverse fast Fourier transform (IFFT)” refer to a discrete Fourier transform (or similar transform) and an inverse discrete Fourier transform (or similar transform), respectively.

Obtaining channel information in a transmitter or source communication unit is very important to obtain the benefits that technologies such as transmitting antenna arrays promise such as maximum signal-to-noise ratio transmission, spatial division multiple access (SDMA), and multi-input closed-loop transmission technology and many outputs (MIMO). Two ways to obtain channel information are feedback and channel sensing. Channel sounding is only suitable for time division duplex (TDD) systems and is based on the fact that the radio frequency (RF) channels of the uplink and downlink are mutually inverse, so that the BS can receive information about the downlink channel communication based on sounding the uplink channel by the mobile station. It is known that in order to probe a channel for subsequent operation, the BS must calibrate its antenna arrays of the uplink and downlink. Unlike channel sensing, channel feedback is suitable for working with frequency division duplex (FDD) systems and TDD systems without calibrating antenna arrays.

The present invention provides a method for efficiently transmitting multiple channels of feedback (e.g., multiple channel estimates for multiple subcarriers, for example, for one or more antennas in an OFDM system (orthogonal frequency division multiplexing)) from a mobile station (s) to a station BS (this method is also suitable for the case when the base and mobile stations switch roles). This method is implemented by sending a training station (for example, pilot symbols) from each of its transmitting antennas to the BS, so that the mobile stations can measure (or estimate) the complex channel response for each of the BS antennas (channel estimates can also be obtained in other ways, which do not require training data, for example, channel estimation with decision management or channel estimation "blindly"). Then, each mobile station that is planned by the base station for such an estimate (or who knows / wishes to perform the estimation in some other way) encodes channel estimates for each antenna of the BS (generates a channel signal), after which many of these mobile stations transmit simultaneously encoded channel estimates back to the BS station to help the BS station in determining channel downlink estimates.

The channel coding process according to the present invention at a particular mobile station does not require subsequent channel quantization, which is already done in accordance with the intrinsic accuracy of the mobile station receiver (for example, standard feedback transmissions carry binary information per character using standard combinations such as QPSK ( phase shift keying with Quaternary phase signals), while the present invention can produce substantially non-quantized channel estimates (e.g., t necessarily limited to the four values of QPSK combinations and confined within its own processing by receiving and transmitting to the mobile station)); rather, this process provides efficient (from the point of view of using the frequency band) combination of a number of measured channels (for example, channels from each antenna of a BS station to one or more antennas of a mobile station) into a single channel signal, which can also be called an encoded channel signal, for transmission to the BS station. Many mobile stations can transmit their encoded channel signals using the same time / frequency resource in the SDMA method. The BS then uses its many receiving antennas to extract the coded channel signals from each mobile station, and then determines the channel estimates by canceling the encoding. Note that in a MIMO system for mobile stations with multiple antennas, each antenna on it can transmit an encoded channel signal different from the others (for example, a signal as a result of encoding channels from all BS antennas to the antenna of the mobile station from which the encoded channel signal is transmitted). Of course, the MIMO transmission of the encoded channel signal from one mobile station according to the SDMA method may be blocked by the transmission of the encoded channel signal from another mobile station. By efficiently coding channel estimates and using SDMA and / or MIMO technology for uplink transmission, very high bandwidth utilization for channel feedback can be achieved. The present invention provides efficient reverse transmission of broadband channels in the frequency domain (for example, on multiple subcarriers of the OFDM system) or in the time domain (for example, by replacing data symbols of the time domain with channel samples) by selective combination of SDMA, MIMO, channel estimation of multiple sources, scrambling to eliminate excess values of the ratio of peak power to average power, as well as the selection of the shape of the pilot signal.

Figure 1 shows an exemplary timing diagram for feedback in a TDD communication system (DL - downlink, UL - uplink), where the base station requests channel information to perform a closed transmission (loop) of a certain type using a transmitting adaptive antenna array (TxAA) according to the present invention. For the uplink, SDMA reception technologies can be used so that many mobile stations can send their feedback data using the same time-frequency resources.

First, the BS sends training data (e.g., pilot symbols) closer to the end of the downlink frame (in fact, the training data can be anywhere in the downlink frame, but the feedback delay is reduced if training data is sent at the end downlink frame). In the case where the BS station has multiple transmit antennas, it can perform multiple transmissions, for example, a first transmission from a first transmitting antenna, a second transmission from a second transmitting antenna, and so on (multiple transmissions), where this multiple transmissions preferably includes pilot symbols that enable the SS to estimate the channel over a plurality of subcarriers for each of the transmit antennas of the BS. Multiple transmissions can be performed simultaneously, preferably with orthogonal / distinguishable pilot sequences between antennas (for example, using different sets or sets of subcarriers with different antennas) or can be performed separately, but it is preferable that these transmissions be close in time. Then, each mobile station requested by the BS to send channel information estimates the channels for each transmitting antenna of the BS (for example, by defining multiple channel estimates for multiple subcarriers), encodes the channel estimates, creating an encoded channel signal (creates or generates a channel signal), and then reports the channel information by sending (transmitting) the encoded channel signal to the base station during transmission on the uplink. In one embodiment, the channel estimates are encoded to allow channel estimates to be sent from the multiple antennas of the BS station and / or the multiple receive antennas of the SS station in a single symbol time interval. Note that in one embodiment, channel estimates are sent along with pilot symbols that are used to estimate the uplink channel at the BS station (uplink channel estimates are used by the BS station to allow its receiving antenna array to separate the uplink channel feedback for many mobile stations, and also determines the encoded channel estimates from each mobile station). Pilot symbols may generate a channel sounding signal (eg, pilot symbols on a set of OFDM subcarriers) to enable estimation of uplink channels at a BS station. The BS uses the feedback channel information to perform a closed transmission (e.g., TxAA transmission) on the downlink, after the uplink (closed transmission (closed loop), it is preferable to perform on a very close interval of the downlink frame in order to minimize the change in the channel between the moment of measurement and the use of this measurement, but the invention is also applicable to cases where the delay is greater)). Since feedback channel information does not need to be quantized to a standard character combination (e.g., QPSK), the resolution and dynamic range of the channel information can be improved (e.g., substantially without quantization) compared to standard schemes. Thus, in one embodiment, channel estimates transmitted in feedback are not substantially quantized. In other embodiments, the channel estimate may be converted to one of a predetermined set of values (for example, the nearest predetermined set of complex values), may be limited, or may be set to zero if its value is lower than a specific value, etc.

Figure 2 shows an exemplary timing diagram for feedback in an FDD communication system (DL - downlink, UL - uplink), where the BS again requests channel information to perform a closed transmission (e.g., TxAA). For the uplink, SDMA reception technologies can be used so that many mobile stations can send their feedback data using the same time-frequency resources.

The operation in this case is similar to the TDD mode, but in the FDD mode, the delay between channel measurement and the use by the BS of downlink channel estimates can be significantly reduced compared to the TDD mode due to simultaneous transmission in the downlink and the uplink. For simplicity, suppose that in the feedback mechanism the downlink and uplink of the FDD system use the same modulation method (eg, OFDM) with the same frequency bandwidth. However, the proposed feedback mechanism is easy to expand for the case when different modulation methods and / or different frequency bands are used in the uplink and downlink. In addition, the invention contemplates the use of multiple uplink transmission modes, for example, IFDMA, CDMA, TDMA with one carrier, or a combination thereof for regular uplink data transmission (any of the modes providing a low peak power to average ratio), and OFDM / OFDMA for feedback transmissions (providing feedback transmission of the integrated channel response in the frequency domain). In this case, feedback transmissions are preferably multiplexed in time or frequency along with regular data transmissions. Thus, the present invention can transmit a channel signal using a different modulation method than that used to transmit other data from the destination communication unit.

Note that the feedback portion of the uplink frame shown in FIG. 1 and FIG. 2 consists of two OFDM symbol time slots. The first symbol transmission period is for sending pilot symbols to each mobile station (for example, a channel sounding signal) that are used to estimate the channel, and the second symbol transmission period is used to send feedback data to help the source communication unit in determining channel estimates for a plurality subcarriers.

This possible feedback option is shown in FIG. 3, in which each rectangle corresponds to one subcarrier of one OFDM symbol. Another feedback option, as shown in FIG. 4, in which each rectangle corresponds to one subcarrier of one OFDM symbol, requires only one OFDM symbol by mixing pilot symbols (e.g., channel sounding signal) with feedback data. The mix shown in FIG. 4 contains frequency multiplexing, but other types of mix can also be used. Obviously, an arbitrary mixture of pilot signals and channel signal is possible, for example, the dispersion of pilot symbols and channel signal samples over a plurality of symbol intervals. In addition, pilot symbols may not be explicitly sent with the channel signal, for example, when the source communication unit uses channel estimates from normal data traffic to detect the channel signal transmitted from the destination communication unit.

Compared with the channel reuse (or reversibility) strategy (that is, in TDD mode, the mobile station probes the uplink, and the BS relies on the reversibility between the uplink and downlink channels) and other known technologies, the proposed feedback method has following advantages:

1. The proposed feedback method can be used both in TDD mode and in FDD mode.

2. The proposed method is implemented in a mobile unit much easier than methods based on code dictionaries and vector quantization.

3. Using the proposed feedback method in TDD mode does not require antenna calibration at a BS station or a mobile station.

4. The proposed feedback method provides a channel feedback mechanism in TDD systems with uplink relays (uplink channel sounding is not possible since the mobile station does not transmit directly to the BS station).

5. The proposed feedback method can be very useful during calibration of the antenna array. For example, the proposed feedback method can be used to send a channel inversion from an SS station to a BS station.

6. The proposed feedback method will work even if the antenna arrays of the uplink and downlink at the BS are different from each other.

7. The mobile station may use downlink broadcast pilots to determine the composite channel (that is, the effective channel perceived by the mobile station, which is a combination of transmit weights and the actual RF channel) if the mobile station knows how the BS calculates these transfer weights. The reason for this is that the mobile station knows the channel that the BS uses to calculate the transmission weights, which channel is not suitable for uplink channel sounding.

Some aspects of the proposed feedback method are described in detail below. Let the BS station have M _b antennas (for both downlink and uplink); and then two cases for mobile stations are investigated: 1) the case when the mobile station u has only one transmitting antenna and M _{m, u of the} receiving antennas, and 2) the case when the mobile station u has M _{m, u of} transmitting and receiving antennas. (Note that the feedback method is applicable to any number of transmit antennas in a mobile station. Also, the feedback method is easy to expand for the case where the base station has a number of receive antennas other than the number of transmit antennas). If we assume that the OFDM downlink uses K subcarriers, then the received signal for mobile station u at its receiving antenna m for subcarrier k (0kK-1) and time point b for the symbol is defined as:

. Заметим, что временной индекс b используется для указания символьного времени пилот-символов (может быть множество значений b, если в целях оценки канала необходимо более одного символа OFDM). where H _{u, m, l} (k, b) is the channel in the frequency domain from the antenna l of the BS station (for example, the source communication unit) to the mth receiving antenna of the mobile station u (for example, the destination communication unit) on the subcarrier k at the time symbol time b; X _l (k, b) is the training symbol transmitted from the l-th antenna of the BS station, and N _{u, m} (k, b) is the additive noise with power

. Note that the time index b is used to indicate the symbol time of the pilot symbols (there can be many b values if more than one OFDM symbol is needed for channel estimation).

в момент b, предпочтительно для двух или более значений k (определяя множество канальных оценок для множества поднесущих). Заметим, что канальные оценки обычно имеют комплексные значения (представляющие как амплитуду, так и фазу). Станция BS может передавать ортогональные пилот-сигналы от множества своих антенн, чтобы дать возможность выполнить простую канальную оценку для множества антенн на мобильной станции. Также можно использовать другие способы канальной оценки (например, с поддержкой принятия решений, итеративные и т.д.). Затем мобильные станции кодируют канальную информацию для создания канального сигнала на основе одного из следующих способов. Mobile stations can trigger a channel estimator from multiple sources from a received signal to determine channel estimates

at time b, preferably for two or more k values (defining a plurality of channel estimates for a plurality of subcarriers). Note that channel estimates usually have complex values (representing both amplitude and phase). A BS may transmit orthogonal pilots from multiple of its antennas to enable simple channel estimation for multiple antennas at a mobile station. You can also use other methods of channel estimation (for example, with decision support, iterative, etc.). Then, the mobile stations encode channel information to create a channel signal based on one of the following methods.

Method 1: Time-sharing feedback for channels in the frequency domain

This feedback method uses a ramp / ramp phase shift (also known as Steiner phase shift) to encode measured channels (channel estimates) from multiple transmit antennas, so that the time domain channels are orthogonal (i.e., channel estimates are shared in the time domain). An example of the resulting phase-shift coding of the frequency domain in the time domain is shown in FIG. 5 for M _b transmit antennas at the base station for a single antenna mobile station. Such coding makes the channels for each transmit antenna orthogonal (distinguishable) in the time domain. As you can see, the corresponding phase shift leads to the fact that the channels in the time domain for each transmitting antenna of the base station occupy different parts of the time domain. According to this feedback method, mobile stations encode channel estimates from each BS antenna for efficient transmission by creating a subsequent uplink signal (i.e., an encoded channel signal or channel signal) for mobile station u with one transmit antenna (note that the time index d is used to indicate the time position of the uplink feedback that occurs at some point after the time index b, where the pilot symbols are transmitted from the BS to et opportunity to channel estimation on the SS):

where S _{u, m, l} (k, d) is the coding sequence for channel estimates of the user u, and β _u is the scaling (scale factor) necessary to make the average transmit power (averaged over all frequencies and, if applicable , then in time) Z _u (k, d) equal to one (or some other desired power level). It is important to have a single scaling for all channel estimates for one mobile station, so that relative power levels between any channel estimates are maintained. If scaling is not necessary, it can be eliminated by setting β _u = 1. Note that if M _b is two or more, then the BS has multiple antennas, and determining a plurality of channel estimates for a plurality of subcarriers may comprise determining a plurality of channel estimates for each of the plurality of subcarriers based on the plurality of transmissions (e.g., each transmission from another antennas) received from the source communication unit.

One example of a coding sequence is the following signal, which has two aspects. The first aspect is the Steiner phase shift, which allows you to separate the channels in the time domain, and the second aspect is the scrambling sequence:

where q _u (k, d) is any known / scrambling sequence (for example, a random sequence with a constant modulus, such as random BPSK characters), and α _u is the shift coefficient for user u (for example, α _u = M _{m, u} M _b ) Note that the use of a scrambling sequence is an important aspect of the invention, since multiplying the channel estimates by a scrambling sequence prevents the appearance of excessively large peak power to average (PAPR) ratios in the transmitted uplink signal (i.e., the IFFT transform for Z _u (k, d) ), especially for some conditions in the channel (for example, channels with uniform fading).

When the mobile station u has many transmitting antennas, it is possible to transmit feedback data of the MIMO type, and then the encoded channel signal (for the mth transmitting antenna at the mobile station u) is specified as:

where β _u is the scaling necessary to make the average transmit power (averaged over all frequencies, all mobile antennas and, if applicable, over time) Z _{u, m} (k, d) equal to unity (or some other arbitrary power level). The coding sequence S _{u, m, l} (k, d) is given in the form:

where q _{u, m} (k, d) is any known / scrambling sequence (for example, a random sequence with a constant modulus, such as random BPSK characters), and α _u is the shift coefficient for user u (for example, α _u = M _b ) .

Note that when using equation (2) or (4), the channel of the frequency domain on subcarrier k represents feedback on subcarrier k. In the TDD system, this means that if the subcarrier k freezes, then the feedback channel will be sent on the low amplitude subcarrier. To avoid this problem, an additional aspect of the invention includes interleaving or reordering Z _u (k, d) or Z _{u, m} (k, d) in frequency to efficiently scramble the frequency feedback. Note that interleaving should be performed with the encoded channel signal Z _u (k, d) or Z _{u, m} (k, d), and not with channel estimates, and before the phase shift is applied. If multiple OFDM symbols are used to send feedback data on the uplink, then time interleaving can be beneficial in addition to frequency interleaving, especially when different channels have different average powers (for example, for two channel estimates, each of which has K values to be transmitted back through two OFDM symbols, half of each channel estimate (K / 2 values from each estimate) should be transmitted in each OFDM symbol, so that the total transmit power will be the same for each OFDM symbol).

Also note that equations (2) and (4) can be modified for propagation in the case where the pilot signals are mixed with the feedback signal, as shown in FIG. 4. These equations can also be modified to encode channels when the frequency band and / or the downlink modulation method (eg, one carrier) is different from the uplink.

Method 2: Orthogonal-frequency feedback for a channel in the frequency domain.

In this method, the measured downlink channels are separated in the frequency domain so that they do not interfere. An example of this encoding method is shown in FIG. 6 for M _b transmit antennas at a BS station and one receive antenna at a mobile station, where each rectangle in the figure corresponds to one subcarrier of one OFDM symbol. As you can see, the frequency domain channels for each antenna are carried on different subcarriers (for example, mapping or coding of channel estimates for a particular antenna into a predetermined set of subcarriers) and therefore do not cause mutual interference. To recover channel estimates on all subcarriers of a BS, it is necessary to interpolate channel estimates. 6 can be modified for distribution in the case when the pilot signals are mixed with feedback data (for example, by analogy with figure 4). In the form of an equation, the encoded channel signal for this type of channel coding of the frequency domain (when the mobile station u has only one transmitting antenna) is defined as:

where (n) _N means n modulo N; δ (n) is equal to one if n is equal to zero, and equal to zero otherwise; q _u (k, d) is any known / scrambling sequence (for example, a random sequence with a constant modulus, such as random BPSK symbols), and β _u is the scaling necessary to make the average transmit power (averaged over all frequencies and if applicable, in time) Z _u (k, d) equal to one (or some other required power level). As with time separability, q _u (k, d) helps to avoid large ratios of peak power to average in the transmitted uplink signal. Also, as in the case of time separability, interleaving Z _u (k, d) can be used to scramble the encoded signal in frequency before transmission. When the mobile station u has M _{m, u} transmit antennas, the encoded channel signal for the transmit antenna m of the mobile station is defined as:

where q _{u, m} (k, d) is any known / scrambling sequence (for example, a random sequence with a constant modulus, such as random BPSK symbols), and β _{u is the} scaling necessary to make the average transmit power (averaged over all frequencies, all mobile antennas and, if applicable, in time) Z _{u, m} (k, d) equal to one (or some arbitrary power level.

Note that for simplicity, equations (6) and (7) can be written explicitly without a time index, namely

Note that if M _b is 2 or more, then determining the set of channel estimates for the set of subcarriers may include determining channel estimates for the first set of subcarriers corresponding to the first transmission of the source communication unit (for example, from the first antenna) and determining channel estimates for the second set subcarriers corresponding to the second transmission of the source communication unit (for example, from a second antenna).

Method 2 can also be categorized as decimation associated with subcarrier thinning, since the delta function results in the placement of zeros on some subcarriers, so channel estimates are sent back for the decimated set of subcarriers. Another set of channel estimates (for example, for a different value of l or for another mobile station) can be assigned subcarriers set to zero, as can be seen from Fig.6). Method 1 can also be attributed to a method where cyclic-shift separability is used, since the sawtooth phase change in equation 5 based on α _u introduces a cyclic shift when representing the channel in the time domain (as can be seen from FIG. 5).

From the mathematical representations of methods 1 and 2, it can be seen that both of these methods create a channel signal based on a variety of channel estimates. Since the structure of channel signals includes many channel estimates, it can be considered that channel signals contain a combination of many channel estimates. Also q _{u, m} (k, d) and S _{u, m, l} (k, d) are known sequences, and channel signals can be considered as containing channel estimates modulated by a known sequence. For example, q _{u, m} (k, d) is preferably a binary sequence (e.g., BPSK), but may be any known sequence.

After creating the channel signal, it can be transmitted to provide information about the channel or to help the source communication unit in determining channel estimates for multiple subcarriers.

Method 3: Orthogonal-frequency feedback for a channel in the time domain

An additional feedback method is for sending an estimated time-domain channel back to the frequency domain or in the time domain. Essentially, this method is similar to the method described in the previous section for sending a channel thinned in the frequency domain, except that channel estimates in the frequency domain are replaced by channel estimates or characteristics in the time domain. Thus, FIG. 6 remains valid, but the channel in this figure now denotes a channel in the time domain. Another possible embodiment is shown in FIG. 7, where each channel response in the time domain for one transmit antenna and a mobile station antenna is sent in a continuous block of subcarriers. Each rectangle in the figure corresponds to one subcarrier of one OFDM symbol. In the form of an equation, the feedback version as a continuous block (when the mobile station u has one transmitting antenna) is expressed as:

for 0nL-1, 1lM _b , 1mM _{m, u}

where L is the estimated channel length of the time domain, q _u (k, d) is any sequence (for example, a random sequence with a constant modulus, such as random BPSK symbols), and β _u is the scaling necessary to make the average transmit power (averaged over all frequencies and, if applicable, over time) Z _u (k, d) equal to one (or some desired value). When the mobile station u has a plurality of transmit antennas, the encoded signal for the transmit antenna m of the mobile station is defined as:

for 0nL-1 and 1lM _b

where q _{u, m} (k, d) is any sequence (for example, a random sequence with a constant modulus, such as random BPSK symbols), and β _u is the scaling necessary to make the average transmit power (averaged over all frequencies, all mobile antennas and, if applicable, in time) Z _{u, m} (k, d) equal to one (or some other arbitrary value).

ниже порогового значения, то его можно обнулить, или, если его величина выше порогового значения, его можно ограничить, либо возможно понадобиться квантование его амплитуды и/или фазы). Другой способ, который можно использовать либо для управления динамическим диапазоном, либо для возможного улучшения характеристик в условиях шума, состоит в выполнении компандирования по значениям

. Необязательно, на стороне приемника станции BS операция компандирования может быть обращена для восстановления некоторой части динамического диапазона.Note that all the described encoding schemes do not necessarily limit the dynamic range of the encoded channel. Channels are expected to have a limited dynamic range by nature, but in some implementations, mobile stations need the encoded channels to be in a narrower dynamic range suitable for the design and implementation of the transmitter. Thus, the encoded channel signal can be further processed so that the resulting signal matches the dynamic range transmitter requirements of the mobile station. Various technologies are possible to ensure that the encoded channel signal corresponds to the dynamic range, such as (but not limited to): limiting the encoded channel signal so that it does not exceed a certain value or voltage; mapping the encoded channel signal into some combination of symbols; and zeroing the encoded signal for samples in which the amplitude is below a threshold value. Also, these processing methods can be applied directly to channel estimates before creating a composite (channel) feedback signal (for example, if

below the threshold value, it can be reset, or if its value is higher than the threshold value, it can be limited, or it may be necessary to quantize its amplitude and / or phase). Another method that can be used either to control the dynamic range or to possibly improve performance under noise conditions is to perform a value companding

. Optionally, on the receiver side of the BS, a companding operation may be inverted to recover some part of the dynamic range.

For ease of explanation, some aspects of the invention have been described above based on the prospect of providing channel information for all K subcarriers of the OFDM system or for a thinned subset of K subcarriers of the OFDM system. This situation may occur because pilot symbols from the BS station typically allow all subcarriers in the MS station to be estimated. But even under these conditions, the MS may be requested to feed back channel information for only one subset of the subcarriers (e.g., selected subcarriers or subcarriers selected in a particular range of subcarriers). In this case, it is not required that the subcarriers used by the MS to probe the uplink channel and transmit the encoded channel signal coincide with the subcarriers whose channel information is requested by the BS until it is clear to the BS which subcarriers the feedback corresponds. In one example, when channel information of an adjacent block of subcarriers (i.e., a subband) is fed back, all three of the above methods can be used to create an encoded channel signal. But in this example, the third method (i.e., orthogonal-frequency feedback for a channel in the time domain) preferably models the frequency response of a subband with a channel in the time domain that has a larger channel spacing than in the case of a full frequency band. In another example, when channel information for a set of disjoint subcarriers is fed back, the MS can, one at a time, forward the channel corresponding to each antenna of the BS for each subcarrier, or at the same time, as in the case of MIMO transmission.

на станции SS (предпочтительно с ограничением по выходной величине) и обеспечив на базовой станции BS обратную операцию после приема данных по обратной связи. Можно рассчитать характеристику вход-выход расширителя таким образом, чтобы улучшить отношение SNR или другой показатель качества на выходе обратной операции на станции BS (например, SNR или MSE (среднеквадратическую ошибку), усредненный по частоте, коэффициент направленного действия при формировании пучка на основе обработанных данных обратной связи и т.д.).Some environments are also possible where the encoded channel signal is received under conditions of low SNR (signal-to-noise ratio) or C / I (ratio of carrier power to interference level). To improve the quality of feedback, the mobile station can retransmit the encoded channel signal (in time and / or frequency). Alternatively, instead of repeating the entire encoded channel signal, channel estimates may be repeated before creating the encoded channel signal. For example, in the orthogonal-frequency method (FIG. 6) on a subcarrier, where the feedback “channel for antenna 1 on subcarrier-M _b ” is assumed, “channel for antenna 1 on subcarrier-0” can be sent again. The compromise of improving quality on subcarrier-0 is an increased decimation factor in the frequency domain, since the channel on subcarrier-M _b will not be transmitted. In addition to these feedback enhancement technologies, you can reduce the effect of noise on the feedback channel by applying a reverse companding unit (or expander) to the values

at the SS station (preferably with a limited output value) and providing the reverse operation at the BS base station after receiving feedback data. You can calculate the input-output characteristic of the expander in such a way as to improve the SNR ratio or another quality indicator at the output of the reverse operation at the BS station (for example, SNR or MSE (standard error), frequency-averaged, directional coefficient during beam formation based on the processed data feedback, etc.).

для любого из вышеописанных способов осуществления обратной связи могут посылаться другие сигналы. Заметим, что для способов 1 и 3 может потребоваться, чтобы сигнал, передаваемый по обратной связи в частотной области, имел эквивалентное представление во временной области с конечным временным диапазоном. Например, для облегчения реализации указанных технологий, таких как калибровка антенной решетки, можно посылать обратные аналоги (инверсии) каналов. Другим примером является передача частотно-избирательного уровня SNR на каждой поднесущей для его использования на станции BS согласно технологиям мощностной нагрузки. Note that in alternative embodiments of the invention, instead of channel estimation

for any of the above feedback methods, other signals may be sent. Note that for methods 1 and 3, it may be required that the feedback signal in the frequency domain have an equivalent representation in the time domain with a finite time range. For example, to facilitate the implementation of these technologies, such as calibrating the antenna array, you can send inverse analogs (inversions) of the channels. Another example is the transmission of a frequency-selective SNR on each subcarrier for use at a BS station according to power load technologies.

Reception and decoding of feedback data when all mobile stations have one transmitting antenna

The BS receives a channel signal (Z), the channel signal having a plurality of channel estimates, as described above, where the plurality of channel estimates contains at least one channel estimate for each of the plurality of subcarriers. Note that the set of subcarriers does not have to have each subcarrier in the OFDM system - it can be a subset of subcarriers, for example, a set of neighboring subcarriers or a set of non-neighboring subcarriers (e.g., thinned subcarriers), and the set of subcarriers can cover any portion of the channel band. The BS estimates the receive channel for the received signal, preferably based on the pilot signal that is received with the channel signal, and restores the estimates for the plurality of channel estimates based on the received signal and the estimated receive channel. The following is a detailed description of an example disclosing some aspects of the above operations for the case where the BS receives N _u channel signals.

N _u mobile stations transmit their encoded sequences at the same time, and the M _b × 1 received signal vector representing the channel signals received at the BS station is defined as:

(I_n - матрица идентичности n×n). where G _u (k, d) is the M _b × 1 (receiving) channel vector of the uplink on the subcarrier k at time d for the mobile station u, and N (k, d) is the additive Gaussian noise with a correlation matrix defined as

(I _n is the identity matrix n × n).

восходящей линии связи каждой мобильной станции станция BS может использовать следующий составной вес с учетом MMSE (минимальная среднеквадратическая ошибка) для мобильной станции u:To restore signal estimates

the uplink of each mobile station, the BS may use the following composite weight taking into account MMSE (minimum mean square error) for the mobile station u:

- (приемная) канальная оценка восходящей линии связи, полученная в результате обработки пилот-символов, посланных от каждой мобильной станции (и принятых станцией BS), для оценки приемного канала. Заметим, что канальные оценки

обычно являются комплексными и их можно назвать оцененными комплексными канальными коэффициентами усиления. Оценка сигнала восходящей линии связи для мобильной станции u задается в виде:Where

- (receiving) uplink channel estimate obtained by processing the pilot symbols sent from each mobile station (and received by the BS) to estimate the receiving channel. Note that channel estimates

are usually complex and can be called estimated complex channel gains. The uplink signal estimate for the mobile station u is defined as:

Using this estimate together with the coding strategy for the mobile station u (i.e., whether channel estimates are sent in the frequency domain or in the time domain and whether time division coding or orthogonal frequency coding is used, which will be known to the BS if the BS indicates station MS in the message about which specific method is used to provide feedback), the base station can receive (restore) channel estimates that were obtained as a result of measurements on each mobile station stations and used to create the channel signal transmitted by the mobile station. Note that in this step, standard channel estimation procedures can be used to help gain gain that produces noise excess plus interference from channel coding for multiple BS station antennas together. The BS now has downlink channel estimates for use in TxAA operations at time t (for total feedback delay for t-b OFDM symbols).

An example of receiving and decoding feedback data when all mobile stations have multiple transmit antennas

N _u mobile stations transmit their encoded sequences from all their transmitting antennas at the same time, and M _b × 1 the received signal vector at the BS station is set in the form:

(I_n - матрица идентичности n×n). where G _{u, m} (k, d) is the M _b × 1 channel vector of the uplink on subcarrier k at time d for the mth transmitting antenna of the mobile station u, and N (k, d) is the additive Gaussian noise with a correlation matrix given as

(I _n is the identity matrix n × n).

восходящей линии связи каждой мобильной станции станция BS использует следующий составной вес с учетом MMSE для передающей антенны m на мобильной станции u:To restore signal estimates

The uplink of each mobile station, the BS uses the following composite weight taking into account the MMSE for the transmit antenna m at the mobile station u:

- канальная оценка восходящей линии связи, полученная из пилот-символов, посланных от каждой мобильной станции.Where

- uplink channel estimate obtained from pilot symbols sent from each mobile station.

The estimate of the uplink signal of the m-th transmitting antenna for the mobile station u is given in the form:

Using this estimate together with the coding strategy for the mobile station u (that is, whether channel estimates are sent in the frequency domain or in the time domain, and whether time division coding or orthogonal frequency coding is used), the base station BS can obtain channel estimates that were obtained as a result of measurements at each mobile station. Note that in this step, standard channel estimation procedures can be used to help gain gain that produces noise excess plus interference from channel coding for multiple BS station antennas together. The BS now has downlink channel estimates for use in TxAA operations at time t (for total feedback delay for t-b OFDM symbols).

Reception and decoding of feedback data in the repeater

As described previously, the block receiving the feedback signal may be a repeater / repeater. In this case, the operation / processing of the received channel signal performed in the repeater is basically the same as that of the BS.

Another aspect of the present invention is a signaling methodology that supports an effective channel feedback methodology. An example is the draft IEEE 802.16e / D5 specification, where this methodology allows subscriber stations (SS) to transmit channel sounding signals on the uplink to allow the base station to estimate the channel response of the uplink between each SS antenna and each BS antenna. This uplink channel sounding methodology from section 8.4.6.2.7.1 of the draft IEEE 802.16e / D5 specification may be modified to include and support an efficient feedback methodology according to the present invention. The idea is to use the same uplink channel sounding methodology from section 8.4.6.2.7.1 of the draft IEEE 802.16e / D5 specification, which allows a BS to estimate the uplink channel response. However, each symbol in the sensing zone (the portion of the uplink (UL) frame where the SS stations transmit sounding signals) and each symbol interval in which the sounding signals of the UL line are sent may be followed by a symbol interval in which the feedback signal (channel signal) can be sent by SS stations that have transmitted sounding signals. The sounding signals for the UL line allow the BS to evaluate the UL channel, which is then used to evaluate the transmitted feedback signals (channel signals) embodying the effective feedback methodology of the present invention.

The following is a first example of incorporating an effective feedback methodology according to the present invention into a communication system defined in the draft IEEE 802.16e / D5 specification. The uplink channel sounding methodology in Section 8.4.6.2.7 of the IEEE 802.16e specification is modified to include, in addition to the sounding signal, commands enabling direct transmission of channel DL (downlink) coefficients (channel estimates) for multiple subcarriers. Such a modification extends the transmission of channel sounding signals for the UL line, allowing the implementation of closed loop transmission in FDD and TDD systems in which the BS antenna transceiver is not calibrated. This modification consists of introducing an additional field in the UL_Sounding_Command_IE () message (UL line sensing command), which the BS station can configure and transmit to the SS station to indicate or issue a command about whether channel coefficients should be transmitted together with the probing signal in the sensing zone. Using this functionality to directly transmit channel coefficients, the probe signal given by the sensing command allows the BS to estimate the UL channel, which the BS then uses to estimate the channel DL coefficients sent by the SS in the feedback section of the probe signal. Then, these estimated channel DL coefficients can be used by the BS to perform a closed (closed loop) transmission. One example of a message that can be configured and transmitted by the BS to indicate or issue a command to the SS to send an effective feedback signal is as follows:

Table 1
UL_Sounding_Command_IE () Syntax The size Notes UL_Sounding_Command_IE () { Advanced UIUC 4 bits 0x09 Length 4 bits Variable Sounding_type
(sounding type) 1 bit 0 = Type A
1 = Type B Sending a sensing message flag 1 bit If (Sounding_Type == 0) { Enabling Feedback Symbols 0 = No feedback characters
1 = Enable feedback characters
(See section 8.4.6.2.7.3) Num_Sounding_symbols 3 bits The total number of sounding characters allocated in the sounding command is from ("000") to 2 ³ = 8 ("111") Separability type 1 bit 0: occupies all subcarriers in the allocated bands;
1: occupies thinned subcarriers If (type of separability == 0) { (use of cyclic shift separability) Index P of maximum cyclic shift 2 bits 00: P = 4;
01: P = 8;
10: P = 16;
11: P = 32 } otherwise { (use decimation based separability) Thinning value D 3 bits Sounding of each Dth subcarrier in the sounding distribution. The value of the decimation parameter D is 2 to the power (2 plus this value), that is, 4.8, ... up to a maximum of 64. Thinning shear randomization 1 bit 0 = no decimation shear randomization
1 = decimation shift defined in a pseudo-random manner } For (i = 0; i <Num_Sounding_symbols; i ++) { Character Sensing Index 3 bits The character index in the sensing zone, from 1 (bits "000") to 2 ³ = 8 (bits "111") Number of CID (Customer Identifier) 4 bits The number of CIDs sharing this sounding distribution For (j = 0; j <number of CID; j ++) { Shortened Base CID 12 bit 12 LSBs of the base CID for MSS (Mobile SS) Start frequency band 7 bit From a maximum of 96 lanes (depending on FFT size) Number of frequency bands 7 bit Adjacent bands used for sounding Power Assignment Method 2 bits 0b00 = equal power;
0b01 = reserved;
0b10 = depending on mutual interference. Power limitation for each subcarrier;
0b11 = depending on mutual interference. General power limitation Power boost 1 bit 0 = no power increase
1 = power increase Multiple Antenna Flag 1 bit 0 = MSS probes only the first antenna
1 = MSS probes all antennas If (type of separability == 0) { Time Cycle Index m 5 bit Cyclically shifts a character in the time domain by a number (from 0 to P-1) that is a multiple of N / P, where N = FFT size, and P = maximum cyclic shift index } otherwise { Thinning shift d 6 bit Relative initial shift position for the first subcarrier occupied for sounding in the sounding distribution } Periodicity 2 bits 00 = single command, not periodic, or completion of periodicity
01 = repeat sounding once per frame until completion
10 = repetition of commands one by 2 frames
11 = repeating commands one by 4 frames } } } otherwise { Rearrangement 2 bits 0b00 = PUSC swap
0b01 = permutation FUSC
0b10 = optional FUSC permutation
0b11 = permutation of adjacent subcarriers IDcell 6 bit Num_Sounding_Symbols 3 bits for
(i = 0; i <Num_Sounding_Symbols; i ++) { CID Number 7 bit For (j = 0; j <number of CID; j ++) { Shortened Base CID 12 bit 12 low-order bits of the base CID for the MSS Subchannel shift 7 bit The subchannel with the minimum index used to carry the packet, starting from subchannel 0 Number of subchannels 3 bits The number of subchannels with sequential indices used to carry the packet. Periodicity 2 bits 00 = single command, not periodic, or completion of periodicity
01 = repeat sounding once per frame until completion
10 = repetition of commands one by 2 frames
11 = repeating commands one by 4 frames Assignment Method
power 2 bits 0b00 = equal power;
0b01 = reserved;
0b10 = depending on mutual interference. Power limitation for each subcarrier;
0b11 = depending on mutual interference. General power limitation Power boost 1 bit 0 = no power increase
1 = power increase } } } Filling variable Zero-padding IE to the octet boundary. Bits must be set to 0 }

Note that commands in IE can specify a special character index and a special type or separability method from a variety of methods for constructing a channel signal. If the "Enable feedback symbols" field is set to 1, then the UL_Sounding_Command_IE () command enables or forces the SS (MSS) station to transmit channel DL coefficients directly to the BS station along with the UL line sounding signal. This functionality provides the base station with information about the state of the downlink channels in both FDD systems and TDD systems that do not calibrate the BS transceiver antenna transceiver. Given the specified functionality or enabled command, the channel DL coefficients are encoded as described below and transmitted in one or more feedback symbols that immediately follow each symbol used to transmit the sounding signals of the UL line. In this case, the UL line sounding signal is used by the BS to estimate the UL channel channel, so that feedback symbols transmitted by the SSs can be estimated by the BS. The encoded feedback symbols may then be used to initiate a closed downlink transmission.

Two cases are distinguished depending on the value of the field / command of the separability type. In the first case, if the separability type is 0 (separability with cyclic shift in the probing signal), then each probing symbol distributed using the UL_Sounding_Command_IE () command is followed by one feedback symbol. In this feedback symbol, the antenna of the SS station, which transmits a sounding symbol, will transmit an encoded feedback signal that occupies the same sounding bands allocated to the sounding signal. The encoded feedback signal for the u-th station SS (where u is the cyclic shift index in the UL line sensing command) is determined for two cases. The first case relates to the situation when the SS station has one transmitting antenna, but many receiving antennas, and it is informed by a sensing command IE to probe all antennas (the multiple antenna flag is set to 1). In this case, one transmitting antenna transmits a probing signal suitable for one transmitting antenna in the probing symbol and transmits the following feedback signal in the following symbol interval:

- оцененный канальный коэффициент (канальная оценка) линии DL между l-й передающей антенной станции BS и m-й приемной антенной u-й станции SS для поднесущей k; β_u - масштабирование, необходимое для того, чтобы сделать среднюю мощность передачи сигнала обратной связи (усредненную по всем частотам) Z_u(k), равной единице; s_u(k) - зондирующая последовательность (известная последовательность) из раздела 8.4.6.2.7.1; M_m,u - количество приемных антенн на u-й станции SS; α_u - M_m,uM_b, а M_b - количество передающих антенн станции BS. Where

- estimated channel coefficient (channel estimate) of the DL line between the lth transmitting antenna of the BS station and the mth receiving antenna of the uth station SS for the subcarrier k; β _u is the scaling necessary to make the average transmit power of the feedback signal (averaged over all frequencies) Z _u (k) equal to unity; s _u (k) is the probe sequence (known sequence) from section 8.4.6.2.7.1; M _{m, u} is the number of receiving antennas at the u-th station SS; α _u - M _{m, u} M _b , and M _b - the number of transmitting antennas of the BS station.

A second case for type 0 separability refers to a situation where an SS station has a number of transmit antennas equal to the number of receive antennas. In this case, the encoded feedback signal to be transmitted to the antenna station SS, to which the cyclic shift index u is assigned in the UL line sensing command, is represented as:

- оцененный канальный коэффициент линии DL между l-й передающей антенной станции BS и антенной станции SS, которой присвоен индекс u циклического сдвига в команде зондирования линии UL для поднесущей k; β_u - масштабирование, необходимое для того, чтобы сделать среднюю мощность передачи сигнала обратной связи (усредненную по всем частотам) Z_u(k), равной единице; s_u(k) - зондирующая последовательность (известная последовательность) из раздела 8.4.6.2.7.1; α_u - M_b и M_b - количество передающих антенн станции BS. Where

- the estimated channel coefficient of the DL line between the lth transmitting antenna station BS and the antenna station SS, which is assigned the cyclic shift index u in the UL line sensing command for the subcarrier k; β _u is the scaling necessary to make the average transmit power of the feedback signal (averaged over all frequencies) Z _u (k) equal to unity; s _u (k) is the probe sequence (known sequence) from section 8.4.6.2.7.1; α _u - M _b and M _b - the number of transmitting antennas of the BS station.

в l-м символе вслед за зондирующим символом, где

- канальный коэффициент линии DL от l-й антенны станции BS до указанной антенны станции SS.When using a UL type 1 separability command in the probing command (decimation separability in the probing signal), each distributed probing symbol is followed by several feedback symbols, the number of which is equal to the number of BSBS station antennas. In this case, the antenna of the SS station, which transmits the probe signal on the subcarrier k, will transmit the channel DL channel coefficient for the ith antenna of the base station to the antenna of the SS station for the kth subcarrier on the subcarrier k of the i-th feedback symbol following distributed sounding symbol. According to the equation, the SS station, which transmits the probe signal, on the subcarrier k of the probe symbol will transmit

in the l-th symbol after the probing symbol, where

- channel coefficient of the DL line from the l-th antenna of the BS station to the specified antenna of the SS station.

Note that the UL line sensing command is a message that the BS configures for the destination communication unit (SS), this message among other things indicating a specific method used to obtain channel estimates from the destination communication unit for multiple subcarriers. For example, one particular method is to introduce feedback symbols with decimation type separability. As another example, another specific method can be cited, which consists in using a probing signal with decimation type separability. After the BS station transmits this message to the destination communication unit, the destination communication unit receives it, generates a channel signal according to a specific method, and transmits this channel signal to the BS station, providing it with channel information for multiple subcarriers. The BS receives a channel signal from a destination communication unit generated in accordance with a particular method, and can then determine channel estimates for a plurality of subcarriers based on the received channel signal, as described above.

Also note that a message with a UL line sensing command can be configured for a plurality of destination communication units, this message instructing each of a plurality of destination communication units to simultaneously transmit a channel signal. When a BS receives a plurality of channel signals, it can then determine channel estimates for each of the plurality of destination communication units, as described previously.

The above example of incorporating an effective feedback methodology according to the present invention into a communication system defined by the draft IEEE 802.16e / D5 specification can be modified for the following additional features or combinations thereof:

- You can add an indicator to this table that determines whether the feedback signals transmitted by the multiple antennas of the SS stations should occupy the same symbol feedback interval (and be distinguished by the BS station using the general symbol estimation / processing by an SDMA receiver) or occupy different feedback symbols (where the BS can decode them without causing interference). If the feedback signals transmitted by the SS antennas must occupy different feedback symbols, then the specific feedback symbol that the antenna of the SS station will occupy will be determined by the value of the cyclic time shift index for this antenna of the SS station in the probe IE command.

- An indicator can be added to the table that determines whether the channel coefficients should be transmitted in the feedback symbol (s) with or without phase shift coding in the previous equations. If the channel coefficients are to be transmitted in the feedback symbol (s) without phase shift coding, then the channel feedback coefficients for each BS station, each antenna of the SS station and the given subcarrier are transmitted individually to combinations of subcarriers and symbol feedback intervals. Another indicator may be added to the table, indicating that Mb channel coefficients for a given antenna of an SS station are transmitted on a specific symbol feedback interval with Mb coefficients for each of a Mb station BS antennas for a specified SS station antenna over Mb sequential subcarriers in this symbol feedback interval. In this case, the indicator from the previous paragraph may indicate whether the feedback symbols for different SS stations occupy the same ones (using processing associated with receiving SDMA type at the BS station) or different symbol feedback intervals.

- It is important to note that the type of separability (cyclic or decimation) used in the probe symbol should not be used in one or more subsequent feedback symbols. To implement the effective feedback strategy of the present invention, the probe symbol is used simply to allow the BS to estimate the channel response of the UL line so that feedback signals transmitted at consecutive feedback symbol intervals can be decoded. Any combination of coding methodology for sounding signals with a coding methodology for feedback signals may be used.

8, an apparatus 800 for transmitting channel information is shown. As shown, apparatus 800 includes a receiver 801, channel estimation circuits 803, channel signal circuits 805, and a transmitter 807. Receiver circuits 801 are provided for receiving a signal from a source communication unit containing a plurality of subcarriers. Channel estimation schemes determine a plurality of channel estimates for a plurality of subcarriers received from a source communication unit, and channel signal circuits 805 create a channel signal based on a plurality of channel estimates. Then, the channel signal is transmitted to the source communication unit to provide channel information to the source communication unit.

Fig. 9 is a flowchart illustrating the operation of the device 800. In particular, Fig. 9 shows the steps necessary for a destination communication unit to transmit a channel information (for example, a downlink channel) to a communication unit. As described above, the channel contains many subcarriers. The logical flowchart begins at step 901, at which a signal is received (for example, a downlink signal). At 903, a plurality of channel estimates for this signal are determined. In particular, at least one channel estimate is determined for each subcarrier of the received signal. In step 905, a channel signal is generated. As described above, a channel signal is based on a plurality of channel estimates. Finally, in step 907, the channel signal is transmitted to the source communication unit for sending downlink channel information.

10, an apparatus 1000 for receiving channel estimates (eg, channel downlink estimates) is shown. As shown, device 1000 comprises a transmitter / receiver 1001, channel estimation circuits 1003, and a channel signal recovery circuit 1005. Transmitter / receiver circuits 1001 are used to receive a signal containing a channel signal, and are additionally used to transmit a message defining one of the many possible methods that will need to be used to generate the channel signal.

As described above, a channel signal comprises a plurality of channel estimates, wherein the plurality of channel estimates includes at least one channel estimate for each of the plurality of subcarriers. Evaluation circuits 1003 are provided for estimating a receiving channel (eg, an uplink channel) for a received signal. Finally, channel signal recovery schemes 1005 are provided for recovering estimates from a plurality of channel estimates based on an uplink signal based on a received signal and an estimated receiving channel.

11 is a flowchart illustrating the operation of the device 1000. The logical flow begins with step 1101, where an uplink signal (eg, an uplink signal) is received. As described above, the uplink signal contains a channel signal, and the channel signal has many channel estimates (e.g., downlink channel estimates), and the many channel estimates contain at least one channel estimate for each of the multiple subcarriers. In step 1103, the estimation schemes 1003 create a channel estimate for the uplink channel. In particular, in order to recover information from an uplink signal, the channel needs to be appropriately estimated. Once the channel is appropriately estimated, information can be retrieved from the uplink signal. Thus, in step 1105, the circuit 1005 reconstructs the downlink channel estimates in the uplink signal by reconstructing the channel signal embedded in the uplink signal and evaluates (e.g., performs the process inverse to the channel signal encoding process, if necessary) channel estimates from a channel signal. As obvious to those skilled in the art, channel estimates can be used to help the source communications unit adapt the characteristics of the subsequent transmission to the destination communications unit. Relevant examples include, but are not limited to, determining and applying transmission weights to form a transmission beam or transmission with multiple inputs and multiple outputs (MIMO), performing frequency selectivity scheduling, selecting bands, selecting a modulation method and coding rate, etc. .

It should be noted that the above discussion provides a method for efficiently providing channel information to a transmitter for use in a closed transmission. A mobile station encodes multiple broadband channels for transmission in one or more OFDM symbols. Many mobile stations transmit their feedback data simultaneously, and the base station uses its antenna array to separate feedback data from multiple users. It should be noted that the equations used above are given as examples of various embodiments of the invention. It will be apparent to those skilled in the art that they can be replaced by other equations if they do not go beyond the essence and scope of the invention. In addition, it should be clear to those skilled in the art that many different modifications, changes, and many combinations can be made to the above described options without departing from the spirit and scope of the invention, and that these modifications, changes and combinations should be considered within the concept inventions. It is assumed that these modifications, changes and combinations are included in the scope of the following claims.

Claims (10)

1. A method for transmitting, by a communication unit, a destination to a communication unit of a source of channel information containing a plurality of subcarriers, the method comprising: determining a plurality of channel estimates for a plurality of subcarriers received from a source communication unit; creating a channel signal based on a plurality of channel estimates; and transmitting channel information by transmitting a channel signal to a source communication unit, wherein the channel signal is based on at least one of:

;

;

; and

; where H _{u, m, l} (k) is the channel estimate between the transmitting antenna with index 1 of the source communication unit and the receiving antenna with index m in the destination communication unit for the subcarrier with index k; H _{u, l} (k) is the channel estimate between the transmitting antenna with index 1 of the source communication unit and the receiving antenna in the destination communication unit for the subcarrier with index k; δ (n) is equal to one if n is equal to zero, and equal to zero otherwise; β _u - scale factor; q _u (k), q _{u, m} (k,) and s _u (k) are known sequences; M _{m, u} is the number of receiving antennas in the communication unit of the addressee; α _u is the shear coefficient; and M _b is the number of transmitting antennas in the source communication unit. 2. The method of claim 1, wherein the channel signal comprises a combination of a plurality of channel estimates. 3. The method according to claim 1, further comprising transmitting the channel sounding signal to the communication unit of the destination together with the channel signal. 4. A method of obtaining channel estimates, comprising the steps of: receiving a signal comprising a channel signal having a plurality of channel estimates, the plurality of channel estimates comprising at least one channel estimate for each of the plurality of subcarriers; estimating a reception channel for a received signal; and reconstructing estimates from a plurality of channel estimates based on a received signal and an estimated receiving channel. 5. The method according to claim 4, further comprising: pilot reception; and wherein estimating the reception channel comprises estimating the reception channel based on the received pilot. 6. A method for transmitting channel estimates, comprising: receiving a message, the message indicating at least a specific method to be used to provide channel estimates from a destination communication unit for a plurality of subcarriers; generating a channel signal in accordance with a particular method for providing channel estimates for multiple subcarriers; and channel signal transmission to assist the source communication unit in determining channel estimates for multiple subcarriers. 7. The method according to claim 6, in which the channel signal contains at least one of: sounding signal; and multiple channel estimates for multiple subcarriers. 8. A method of obtaining channel estimates in the source communication unit, the method comprising: configuring a message for the destination communication unit, the message indicating at least a specific method to be used to obtain channel estimates from the destination communication unit for the plurality of subcarriers; sending a message to the communication unit of the addressee; receiving a channel signal from the communication unit of the destination, where the channel signal is generated in accordance with a specific method; and determining channel estimates for multiple subcarriers based on the received channel signal. 9. The method of claim 8, in which the channel signal comprises at least one of: sounding signal; and multiple channel estimates for multiple subcarriers. 10. A device for transmitting channel information, the device comprising: receiver circuits for receiving a signal from a source communication unit comprising a plurality of subcarriers; channel estimation schemes for determining a plurality of channel estimates for a plurality of subcarriers received from a source communication unit; channel signal schemes for generating a channel signal based on a plurality of channel estimates; and transmission schemes for transmitting a channel signal to a source communication unit to provide channel information to a source communication unit, wherein the channel signal is based on at least one of:

;

;

; and

; where H _{u, m, l} is the channel estimate between the transmitting antenna with index 1 of the source communication unit and the receiving antenna with index m in the destination communication unit for the subcarrier with index k; H _{u, l} (k) is the channel estimate between the transmitting antenna with index 1 of the source communication unit and the receiving antenna in the destination communication unit for the subcarrier with index k; δ (n) is equal to one if n is equal to zero, and equal to zero otherwise; β _u - scale factor; q _u (k), q _{u, m} (k,) and s _u (k) are known sequences; M _{m, u} is the number of receiving antennas in the communication unit of the addressee; α _u is the shear coefficient; and M _b is the number of transmitting antennas in the source communication unit.

Images