Mimo transmission with rank-dependent precoding

(54) Title: TRANSMISSÃO MIMO COM (57) Summary:

CLASSIFICATION PRECODIFICATION

DEPENDENT (30) Unionist Priority: 07/02/2008 us 12 / 027,921,

02/02/2007 US 60 / 889,255 (66) Internal Priority: 860446 (73) Holder (s): Qualcomm Incorporated (72) Inventor (s): Byoung-Hoon Kim, Hao Xu (74) Attorney (s): Montaury Pimenta, Machado & Lioce (86) International Order: pct US2008053512 of 08/02/2008 (87) International Publication: wo 2008 / 098225de 14/08/2008

500

<img file="BRPI0807108A2_D0001.tif" />

MIMO TRANSMISSION WITH DEPENDENT CLASSIFICATION PRECODING

This patent application claims the priority of the provisional application serial number

US60 / 889.255, entitled MULTIPLE INPUT MULTIPLE OUTPUT

ANTENNA METHODS AND DEVICES, filed on February 9, 2007, assigned to its assignee, and incorporated by reference in this document.

FUNDAMENTALS

I. Field of the Invention

The present disclosure refers in general to communication, and more specifically to techniques for transmitting data in a wireless communication system.

II. Foundations

Wireless communication systems are widely developed to provide various communication content such as voice, video, packet data, exchange of broadcast messages, etc. These wireless systems can be multiple-access systems capable of supporting multiple users by sharing available system resources. Examples of such multiple access systems include code division multiple access systems (CDMA), time division multiple access systems (TDMA), frequency division multiple access systems (FDMA), and multiple access systems by orthogonal frequency division (OFDMA), and single carrier FDMA systems (SC-FDMA).

A wireless communication system can support multiple input and multiple output (MIMO) transmission. For MIMO, a transmitter can use multiple transmit antennas (T) to transmit data to a receiver equipped with multiple receive antennas (R). The multiple transmit and receive antennas form a

2/33 MIMO channel that can be used to increase productivity and / or improve reliability. For example, the transmitter can transmit up to T data streams simultaneously from the T transmission antennas to improve productivity. Alternatively, the transmitter can transmit a single data stream from all transmission antennas to improve reliability. In any case, it is desirable to send a MIMO transmission in a way that achieves good performance.

SUMMARY

Techniques for performing dependent line precoding for a MIMO Transmission are described here. Precoding may include processing with a precoding array or matrix to send L data streams into L virtual antennas, formed by T physical antennas where, in general, 1 &lt;L &lt;T. L can also be considered as the line of a MIMO Channel. For dependent line precoding, each line can be associated with a set of at least one precoding vector or matrix that can provide good performance for that line. Different lines can be associated with different sets of vectors or precoding matrices.

In a project, a transmitter (for example, a Node B) can obtain a precoding vector for a line-1 transmission from a first set containing at least one column vector of a unitary matrix. The unitary matrix can be a Fourier matrix, an offset phase Fourier matrix, or some other matrix that has orthogonal columns. The transmitter can perform precoding for line-1 transmission based on the precoding vector. The transmitter can obtain a precoding matrix for a line-2 transmission to

3/33 from a second set containing an identity matrix that has numbers one along the diagonal and zeros elsewhere. The transmitter can perform precoding for line-2 transmission based on the precoding matrix.

In a project, the transmitter can determine whether the MIMO channel resembles a diagonal channel, which has a channel response matrix with small channel gains outside the diagonal. This determination may be based on the antenna configurations at the transmitter and receiver. The transmitter can select the identity matrix as the precoding matrix for the line-2 transmission if the MIMO channel looks like a diagonal channel. The second set may additionally include the unitary matrix. The transmitter can select the unitary matrix as the precoding matrix for the line-2 transmission if the MIMO channel does not look like a diagonal channel.

Several aspects and characteristics of the disclosure are described in further detail below.

BRIEF DESCRIPTION OF THE DRAWINGS

Ά Figure 1 shows a wireless multiple access communication system.

Figure 2 shows a block diagram of a Node B and a user equipment (UE).

Figure 3 shows a block diagram of a transmission data processor (TX) and a TX MIMO processor.

Figure 4 shows a block diagram of a receiving MIMO (RX) processor and an RX data processor.

Figure 5 shows a process for transmitting data with dependent line precoding.

4/33

Figure 6 shows an apparatus for transmitting data with dependent line precoding.

Figure 7 shows a process for receiving data with dependent line precoding.

Figure 8 shows a device for receiving data with dependent line precoding.

DETAILED DESCRIPTION

The techniques described here can be used by various wireless communication systems such as CDMA, TDMA, FDMA, OFDMA, SC-FDMA and other systems. The terms system and network are often used interchangeably. A CDMA system can implement radio technology such as Universal Terrestrial Radio Access (UTRA), cdma2000, etc. UTRA includes broadband-CDMA (W-CDMA) and other variants of CDMA. Cdma2000 covers the IS-2000, IS-95 and IS-856 standards. A TDMA system can implement radio technology such as the Global System for Mobile Communications (GSM). An OFDMA system can implement radio technology such as Evolved UTRA (E-UTRA), Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, FlashOFDM (R), etc. UTRA and E-UTRA are part of the Universal Mobile Telecommunication System (UMTS). The Long Term Evolution (LTE) 3GPP is a future launch of UMTS that uses E-UTRA. UTRA, E-UTRA, UMTS, LTE and GSM which are described in documents from an organization called the 3^The Generation (3GPP). Cdma2000 and UMB are described in documents from an organization called the 3^The Generation 2 (3GPP2). These various radio technologies and standards are known in the art.

Figure 1 shows a multi-access wireless communication system 100 with multiple B-Nodes 110 and

5/33 multiple UEs 120. A Node B can be a fixed station that communicates with the UEs and can also be designated as an evolved Node B (eNB), a base station, an access point, etc. Each Node B 110 provides communication coverage for a particular geographic area. UEs 120 can be dispersed throughout the system, and each UE can be stationary or mobile. A UE can also be designated as a mobile station, a terminal, an access terminal, a subscriber unit, a station, etc. A UE can be a cell phone, a personal digital assistant (PDA), a wireless modem, a wireless communication device, a portable device, a laptop computer, a cordless phone, etc. A UE can communicate with a Node B via downlink and uplink transmission. The downlink (or sequential link) refers to the communication link from Nodes B to the UEs, and the uplink (or reverse link) refers to the communication link from the UEs to the Nodes.

Figure 2 shows a block diagram of a design of Node B 110 and UE 120, which are one of Nodes B and one of the UEs in Figure 1. Node B 110 is equipped with multiple antennas (T) 234a a 234t. The UE 120 is equipped with multiple antennas (R) 252a to 252r. Each of the antennas 234 and 252 can be considered as a physical antenna.

At Node B 110, a TX 220 data processor can receive data from data source 212, process (for example, encode and map symbol) the data based on one or more modulation and encoding schemes, and provide symbols of data. As used here, a data symbol is a data symbol, a pilot symbol is a pilot symbol, and a symbol can be a real or complex value. The pilot data and symbol can be modulation symbols from a modulation scheme such as PSK or QAM. The pilot is a data that is known to

6/33 priori by both Node B and the EU. A TX MIMO 230 processor can process data and pilot symbols as

<td colspan="2">described below,</td><td>and</td><td colspan="2">can provide the</td><td>T flows</td><td>of symbol</td>

<td>output to</td><td>the</td><td>T</td><td>modulators</td><td>(MOD)</td><td>232a a</td><td>232t. Each</td>

<td>modulator 232</td><td>can</td><td colspan="2">process your</td><td>flow</td><td colspan="2">exit symbol</td>

<td>(for example,</td><td>for</td><td>O</td><td>OFDM) for</td><td>get</td><td>a stream</td><td>sample</td>

<td>about to leave.</td><td>Each</td><td></td><td>modulator</td><td> 232</td><td>can</td><td>condition</td>

additionally (for converting to analogue,

232t can be 234a to 234t, for example, filtering, amplifying, and rising converters) your outgoing sample stream and generating a downlink signal. The T downlink signals from modulators 232a to transmitted through the antennas respectively.

In UE 120, R antennas 252a to 252r can receive T downlink signals from Node B 110, and each antenna 252 can provide a received signal to an associated demodulator (DEMOD) 254. Each demodulator 254 can condition (for example , filter, amplify, perform a downward conversion, and digitize) your received signal to obtain samples, and can further process the samples (for example, for OFDM) to obtain the received symbols. Each demodulator 254 can provide received data symbols for an RX MIMO 260 processor and provide received symbol pilots for a 294 channel processor. The 294 channel processor can estimate the MIMO channel response from Node B 110 to UE 120 based on the received symbol pilots, and provide an estimated MIMO channel for the RX MIMO 2 60 processor. The RX MIMO 260 processor can perform MIMO detection on the received data symbols based on the estimated MIMO channel, and provide detected symbols, which are estimated from the transmitted data symbols. A RX 270 data processor can process (for example, demap and decode a symbol)

7/33 the detected symbols and provide decoded data to a collector.de, data 272 ,.

UE 120 can evaluate channel conditions and generate feedback information, which can comprise various types of information as described below. Feedback information and data from data source 278 can be processed (for example, encoded and mapped by symbol) by a TX 280 data processor, spatially processed by a TX MIMO 282 processor, and further processed by modulators 254a at 254r to generate uplink R signals, which can be transmitted via antennas 252a to 252r. At Node B 110, R uplink signals from UE 120 can be received by antennas 234a to 234t, processed by demodulators 232a to 232t, spatially processed by an RX MIMO 236 processor, and further processed (for example, demapped and decoded by symbol ) by an RX 238 data processor to retrieve the information and feedback data sent by the UE 120. The decoded data can be provided to a data collector 239. A controller / processor 240 can control data transmission to UE 120 based on feedback information.

Controllers / processors 240 and 290 can direct operation on Node B 110 and UE 120, respectively. Memories 242 and 292 can store data and program codes for Node B 110 and UE 120, respectively. A programmer 244 can select a UE 120 and / or other UEs for data transmission on the downlink and / or uplink based on the feedback information received from all UEs.

The techniques described here can be used for MIMO transmission in the downlink as well as in the uplink.

8/33

For clarity, certain aspects of the techniques are described below for MIMO transmission on the LTE downlink. LTE uses orthogonal frequency division multiplexing (OFDM) in the downlink and single carrier frequency division multiplexing (SC-FDM) in the uplink. OFDM and SC-FDM divide the system's broadband into multiple orthogonal (K) subcarriers, which are also commonly referred to as tones, bands, etc. Each subcarrier can be modulated with data. In general, modulation symbols are sent in the frequency domain with the OFDM and in the time domain with the SC-FDM. LTE uses localized frequency division multiplexing (LFDM), which is a variant of SC-FDM, for the uplink. With LFDM, modulation symbols are sent on a block of consecutive subcarriers.

Node B 110 can transmit L data symbols simultaneously through L layers in each subcarrier in each symbol period, where, in general, L&gt; 1. A layer can correspond to a spatial dimension for each subcarrier used for transmission. Node B 110 can transmit data using various MIMO transmission schemes.

In a project, Node B 110 can process data symbols for each subcarrier k, as follows:

x (k) = WUd / ÃJ, Eq (1) where d (k) is a vector L x 1 containing L data symbols to be sent through L layers in subcarrier k in a symbol period,

U is a permutation matrix L xL,

W is a Z xL precoding matrix, and

9/33 x (k) is a vector Lxl containing T output symbols for the JT transmission antennas on subcarrier k in a symbol period.

Equation (1) is for a subcarrier k. The same processing can be performed for each subcarrier used for the transmission. In the present description, a matrix can have one or multiple columns.

The previous matrix W can be used to form T virtual antennas with T physical antennas 234a to 234t at Node B 110. Each virtual antenna can be formed with a W column. A data symbol can be multiplied by a W column and can then be sent on one virtual antenna and on all physical T antennas. W can be determined as described below.

The permutation matrix U can be used to map the data symbols for the L layers to the L virtual antennas selected from the available virtual T antennas. OR can be defined based on a layer for mapping the virtual antenna selected for use. OR can also be an identity I matrix. The same or different permutation matrices can be used for the K subcarriers.

In general, Node B 110 can perform precoding based on one or more arrays. A precoding can include a virtual antenna, signaling, which is processed with the previous matrix W to obtain virtual antennas.

it can also include precoding processing with one or more cyclic delay matrices for each subcarrier to obtain a diversity of cyclic delays. For simplicity, many of the following descriptions assume that the previous ones include only one

10/33 virtual antenna signaling with W precoding matrix.

Figure 3 shows a block diagram of a design for a TX 220 data processor, TX MIMO 230 processor, and modulators 232a to 232t on Node B 110 in Figure 2. Within the TX 220 data processor, the S flows data can be provided for encoders S 320a to 320s, where, in general, S&gt; 1. Each encoder 320 can encode, merge, and scramble its data streams and provide encoded data to an associated symbol mapper 322. Each 322 symbol mapper can map its encoded data to the data symbols. Each data stream can carry a transport block or packet at each transmission time interval (TTI). Each encoder 320 can process its transport block to obtain a code word. The terms data flow, transport block, package, and codeword can be used interchangeably. Symbol mappers 322a through 322s can provide S data symbol streams.

Within the TX MIMO 230 processor, a layer 332 mapper can map the data symbols to the S data streams to the virtual L antennas selected for use. In a project, mapper 332 can map the data symbols for the S data streams to the L layers and can then map the data symbols for the L layers to the subcarriers and virtual antennas used for the transmission. A precoder / virtual antenna signaling unit 334 can multiply the symbols mapped from the layer mapper 332 for each subcarrier with the precoding matrix W to obtain output symbols for that subcarrier. The pilot symbols can be multiplexed at the input of the

11/33 precoder 334. The precoder 334 can provide T output symbol streams for the T modulators 232a to 232t.

Each modulator 232 can perform an OFDM modulation for a respective output symbol stream. Within each modulator 232, a discrete inverse Fourier transform unit (IDFT) 342 can perform an IDFT K-point on K output symbols to be sent on the total K subcarriers in an OFDM symbol period to obtain a profitable portion containing the K time domain samples. Each time domain sample is a complex value to be transmitted over a sample period. A cyclic prefix generator 344 can copy the last C samples from the profitable portion and attach the copied samples in front of the profitable portion to form an OFDM symbol containing K + C samples. The copied portion is designated as a cyclic prefix and is used to combat inter-symbol interference (ISI) caused by the disappearance of the selective frequency. Each 232 modulator can additionally condition its sample flow (not shown in Figure 3) to generate a downlink signal.

The controller / processor 240 can receive feedback information from the UE 120 and generate controls for the TX 220 data processor and for the TX MIMO 230 processor. The controller / processor 240 can also supply the W precoding matrix to the 334 precoder. .

The TX 280 data processor, TX MIMO 282 processor, and modulators 254 at UE 120 in Figure 2 can be implemented in a similar manner to the TX 220 data processor, TX MIMO 230 processor, and modulators 232, respectively, in Figure 3. For LFDM, a discrete Fourier transformed unit (DFT) can be inserted after

12/33 each symbol mapper 322 and can be used to transform data symbols from a time domain to a frequency domain. For OFDM, the DFT unit can be omitted as shown in Figure 3.

The symbols received at UE 120 for each subcarrier k can be expressed as: y (k) = H (k) x (k) + z (k) = H (k) WVd (k) + i (k) Eq (2 ) = H_ef(k) Ud (k) + i. (k) where H (k) is an R x T MIMO channel matrix for subcarrier k, n_ef(k) = H. (k) W is an R x L is an effective MIMO channel matrix for subcarrier k, y (k) is an R x 1 vector containing R symbols received in subcarrier k, and

J. (k) is a noise vector Rxl for subcarrier k.

The UE 120 can compute a spatial filter matrix Mfk / for each subcarrier k based on the MIMO channel matrix H / k / and the precoding matrix W and, according to a linear technique of minimum mean square error (MMSE), as follows:

M (k) = T&gt; (k) [H (k) H_ef (k) + crf I f H (k) Eq (3) where O (k) = [di_Theg {[H ^_f(k) H_ef(k) + a²2lJ-¹íl ^_f(k) ll_ef(k)} J-¹ is a diagonal matrix of graduation values ​​used to obtain normalized detected symbols, σ² is the noise variation, and

M / k / is an RxL spatial filter matrix for subcarrier k.

13/33

The UE 120 can perform MIMO detection as follows:

Ud (it) = M (it) y (ji :) = Ud (ifc) + z (i) Eq (4) where d (£) is a L χ 1 vector containing L symbols detected for the subcarrier k, ez ( fc) is a noise vector after MIMO detection.

Figure 4 shows a block diagram of a design of the RX MIMO 260 processor and the RX 27 0 data processor at UE 120 in Figure 2. A channel estimator 2 94 can derive an estimated MIMO U (k) channel, based on pilot symbols received from demodulators 254a to 254r. Within the RX MIMO 260 processor, a computing unit 410 can compute an M / jt spatial filter matrix; for each subcarrier k as shown in equation (3), or based on some other MIMO detection technique. A MIMO 412 detector can perform MIMO detection on the R data symbol streams received from the R demodulators 254a to -254r with the spatial filter matrix N1 (k) for each subcarrier k as shown in equation (4), and provide symbols detected for the selected virtual antennas L. A layer mapper 414 can map the detected symbols in a complementary manner to the mapping performed by layer mapper 332 in Figure 3 and can provide the S symbol streams detected for the S data streams.

In the RX 270 data processor, the S symbol mappers 420a to 420s can demap the symbol from the detected symbol streams and provide log likelihood ratios (LLRs). S decoders 422a through 422s can unscramble, deinterleave, and decode LLRs from symbol demappers 420a through 420s,

14/33 amplifiers used for the respectively, and provide S decoded data streams.

Figure 4 shows a linear MMSE receiver. For a linear MMSE with a receiver for successive interference cancellation (MMSE-SIC), a data stream can be detected and decoded, and the interference due to that stream can be estimated and canceled from the received data symbols. Another stream of data can then be detected and decoded after canceling the interference due to the decoded stream.

Various types of matrices can be used for the precoding matrix W. In one project, a unitary matrix VT xT is used for the precoding matrix W. The unitary matrix V is characterized by the properties V^//V = I and VV^w = I, which means that the columns of V are orthogonal to each other, the lines of V are also orthogonal to each other, and each column and each line has a unit energy. Unit matrix V can be defined so that all elements of the matrix are elements of unit of magnitude that have the same magnitude. The use of unitary matrix V with elements of unit of magnitude for the previous matrix W can (i) allow all T transmission antennas and their associated energy to be fully data transmission without taking into account the number of layers and (ii) and without affecting channel statistics, which may be unknown to the transmitter.

In a project, a Fourier matrix F, which is a unitary matrix with elements of unit of magnitude, can be used for the previous matrix W. The elements of a Fourier FT xT matrix can be expressed as:

15/33

-&gt; 2π— f = e ^τ

J u_tv for u = 0,, T-7 and v = 0, Tl

Fq (5) where f_uv is the element in the u-th row and in the v-th column of the Fourier matrix. A Fourier matrix is ​​also commonly known as a DFT matrix.

In another project, a Fourier phase-shifted matrix, which is also a unitary matrix with elements of unit of magnitude, can be used for the precoding matrix W. A phase-shifted Fourier matrix can be expressed as:

W = AF, ··· o

Eq (6) where it is a diagonal matrix of displaced phase, and θ_ν it is the phase for the seventh antenna.

A diagonal matrix is ​​a matrix with possible non-zero elements along the diagonal, and zero elements elsewhere. As shown in equation (6), the displaced phase Fourier matrix can be obtained by pre-multiplying the Fourier matrix with a diagonal matrix.

Pre-coding with a unitary matrix, as shown in equation (1), can allow each data symbol to be transmitted via all physical T antennas and can also allow associated power amplifiers to be used for data transmission even when only one data symbol is sent on a layer. In addition, each data symbol can be sent from all physical T antennas without regard to the number of layers, and can observe spatial diversity.

16/33

The response of the MIMO channel from the T transmitting antennas at Node B .110 to the R receiving antennas at

EU 120 can be expressed as:

H (*) =

M *) h_2l (k) h_l2(k) h₂₂(k) kyr (k) h_2T(k)

Eq (7) ^ R1 (£) ^ R2 (*) A_rt (*) where hfk) is a complex channel gain of the transmission antenna j to receive antenna i by subcarrier k.

The characteristics of complex channel gains in Wk) may depend on several factors such as the wireless environment, the type of antennas used in Node B 110, the type of antennas used in UE 120, etc. If an antenna configuration, such as uniform linear array (ULA), is used at Node B 110, then complex channel gains may be non-correlated, and precoding with a unitary matrix can achieve spatial diversity. However, if the cross polarization antenna configurations are used on Node B 110 and UE 120, then the cross polarization discrimination (XPD) of the MIMO Channel can be increased. When XPD is high, the MIMO H / £ / channel matrix can come close to a diagonal matrix, and precoding with a unitary matrix can make the matrix H effective_ef(k) of the MIMO channel, far from the diagonal matrix.

UE 120 may employ a linear MMSE receiver, an MMSE-SIC receiver, or some other MIMO receiver to process received symbols y (k). The linear MMSE receiver can perform linear MMSE detection on the received symbols to obtain detected symbols for all flows, which can be processed to recover the data sent in those flows. The MMSE-SIC receiver can perform linear MMSE detection and decoding for a flow in

17/33 time, estimate the interference due to each decoded stream, and cancel the previous estimated interference to perform MMSE detection and decode for the next stream. The MMSE-SIC receiver may be able to achieve good performance regardless of whether the effective MIMO channel is diagonal or not diagonal. However, the linear MMSE receiver may perform poorly for an effective non-diagonal MIMO channel. Thus, precoding with a unitary matrix in the presence of an almost diagonal H (k) matrix of the MIMO channel can degrade the performance of the linear MMSE receiver.

In one aspect, dependent line precoding can be performed in order to provide good performance for both the linear MMSE receiver and the MMSE-SIC receiver. For dependent line precoding, each classification can be associated with a set of at least one precoding vector or matrix that can provide one.

classification. Different associated with different good performance ratings for this set can be precoding vectors or matrices. Dependent line precoding can provide good performance even when Node B is equipped with cross-polarized antennas.

For clarity, dependent line precoding is described below for a 2x2 MIMO configuration with two transmit antennas and two receive antennas. For simplicity, the following description is for a subcarrier, and the subcarrier index k is omitted. Also for simplicity, the U is assumed to be an identity matrix and is omitted. For the 2x2 MIMO configuration, the symbols received in the UE can be expressed as:

18/33

An Α₁₂Λ₂ι A₂₂

Eq (8)

The MIMO channel matrix can look like a diagonal matrix if the XPD is too high. For an almost diagonal MIMO channel, the MIMO channel matrix can be expressed as:

H =

<td></td><td> ^12</td><td></td><td>to 0</td>

<td>Λΐ</td><td>THE₂₂_</td><td></td><td>.θ β.</td>

Eq (9) where a and β are complex channel gains. The matrix of the almost diagonal MIMO channel in equation (9) can be obtained, for example, when Node B and UE are both equipped with cross-polarized antennas.

For a 2-rating transmission using a 2x2 F = Fourier matrix

1 1 -1 as the

<td>^xi ~</td><td> 1</td><td> '7 /]</td><td></td><td></td><td></td>

<td> .^X2_</td><td> 42</td><td> _7 — 7_|</td><td>ά</td><td>r</td><td>and</td>

<td>'z'</td><td> 1</td><td>aa</td><td>Γ ^ ι</td><td></td><td>^z\~</td>

<td></td><td> 72</td><td>1 1 t</td><td></td><td> +</td><td></td>

precoding matrix W, the exit symbols at Node B and the symbols received at the UE can be expressed as:

Eq (10)

Eq (ll)

Equation (11) can be expressed as:

y_}= - ^ = (ad_l+ ad₂') + z_l and χ₂= - ^ == (/ 7 d_x- β · ά₂) + z₂ Eq (12)

A / 2 5/2

As shown in equation (12), precoding with the Fourier matrix can degrade the performance of the classification-2 transmission due to increased spatial interference if the UE employs the linear MMSE receiver, unless | oj = | /? | . If the UE employs an MMSESIC receiver, then precoding with the Fourier matrix may not degrade performance (ideally).

19/33

For a classification-2 transmission that uses the, identity matrix as the precoding matrix W, or W = I, the exit symbols on Node B and the symbols received on the UE can be expressed as:

<td>Xj</td><td> 1</td><td>Ί</td><td> 0'</td><td></td>

<td> -^X2.</td><td> 1'</td><td> 0</td><td> 1</td><td>d₂</td>

Eq (13)

<td>You</td><td> 1</td><td>to 0</td><td>~ df</td><td></td><td>^Z1</td>

<td>_T₂.</td><td>1/: II</td><td>1 O</td><td>d₂</td><td> +</td><td> _^Z2_</td>

Eq (14)

Equation (14) can be expressed - (X · d_x + ζ_λ ey₂ - β · d₂ + r as:

Eq (15) uses a column of the Fourier matrix F precoding W, the output symbols symbols received in the UE can be expressed

As shown in equation (15), precoding with the identity matrix can result in a small or non-spatial interference when the MIMO channel matrix is ​​almost diagonal. This can provide good classification-2 performance for both the linear MMSE receiver and the MMSE-SIC receiver. The identity matrix can therefore be preferred over the Fourier matrix for the classification-2 transmission with an almost diagonal MIMO channel matrix.

For a transmission of classification -1 that) mo the matrix of in Node B and the as:

Eq (16)

<td> 1_</td><td> 1</td><td>f</td>

<td>x₂</td><td>1 ^ II</td><td> ±1</td>

Tl

T₂

V2 a

± β.

d +

<img file="BRPI0807108A2_D0002.tif" />

Eq (17) ± sign in equations (16) and (17) depends on whether the first or second column of the Fourier matrix is ​​used as the precoding vector.

Equation (17) can be expressed as:

20/33 ii - ad + ζ_γ and _y₂^— ± β 'd + z₂

Eq (18)

As shown in equation (18), precoding with a column of the Fourier matrix for a rating-1 transmission can improve performance as long as the UE can obtain a combined channel energy of (| a |² + | β 1²) / 2, thereby making full use of the energy radiated from the two power amplifiers to the two transmission antennas at Node B.

For a classification-1 transmission that uses the left column of the identity matrix as the W precoding matrix, the symbols received in the UE can be expressed as:

¹ , jVi - cx 'dz_THE ej /₂ - z₂

Eq (19)

If the right column of the identity matrix is ​​used as the W precoding matrix, then the symbols received in the UE can be expressed as:

β -d + z₂

Eq (20)

As shown in equations (19) and (20), the UE can obtain channel energy from both | or |²/ 2 how much of j β j²/ 2 for classification-1 transmission depending on whether the left or right column of the identity matrix is ​​used for precoding. Thus, half of the total energy of the two power amplifiers can be lost by using a column of the identity matrix for the precoding of the classification-1 transmission. The Fourier matrix can therefore be preferred over the identity matrix for the classification-1 transmission.

21/33

In a first project, three hypotheses can be supported by classifications 1 and 2 in a 2x2 MIMO configuration, as follows:

. Use the identity matrix for classification

2, and

Use both the first and the second column of the Fourier matrix (or an offset phase Fourier matrix) for classification 1.

The first design can be used when the MIMO H channel matrix is ​​almost diagonal , for example, due to the high XPD for the cross-polarized antenna configurations. This design can provide good performance for both the linear MMSE receiver and the MMSE-SIÇ receiver for both classifications 1 and 2 for the almost diagonal matrix of the MIMO channel. This project can be used when the information in the precoding matrix is ​​not informed by the UE.

In a second project, four hypotheses can be supported for classifications 1 and 2 in a 2x2 MIMO configuration, as follows:

. Use both the identity matrix and the Fourier matrix (or an offset phase Fourier matrix) for classification 2, e. Use either the first or the second column of the Fourier matrix (or the shifted phase Fourier matrix) for classification 1.

The second project can support both the almost diagonal MIMO channel as well as the diagonal MIMO channel. Both the high XPD (MIMO channel almost diagonal) and the low XPD (far from the diagonal of the MIMO channel) can be dynamically observed even by the cross-polarized antenna configurations depending on the orientation of the

22/33 antenna, channel propagation, etc. In addition, different UEs can be equipped with different antenna configurations, for example, some UEs can be equipped with dipole antennas while other UEs can be equipped with cross-polarized antennas. Supporting both the identity matrix and the Fourier matrix for classification 2, good performance can be achieved for both the linear MMSE receiver and the MMSE-SIC receiver without regard to XPD or antenna configurations.

For the second project, a UE can select one of the four hypotheses based on the metric (for example, summability). The UE can inform the selected hypothesis using two bits for the feedback. Node B can apply the antecedent matrix corresponding to the selected hypothesis for data transmission to the UE.

For clarity, dependent line precoding has been described for a 2x2 MIMO configuration. In general, dependent line precoding can be used for any MIMO R x T configuration and can support any number of different classifications. Each classification can be associated with a set of at least one precoding vector or matrix. For classification 1, the set can include at least one column vector of a unitary matrix, which can be a Fourier matrix, a displaced Fourier matrix, or some other matrix. For classification 2, the set can include the identity matrix and possibly one or more unitary matrices. A set for a higher rating can include one or more matrices that can provide good performance for that rating. For example, a set for rating 4 may include a matrix that can provide good performance for

23/33 dual cross-polarized antennas. The set for each classification can also include other matrices. The sets of precoding vectors / matrices for different classifications can be defined to provide good performance for both the almost diagonal MIMO channel as well as distant from the diagonal MIMO channel.

In a project, Node B can select the precoding matrix and not have an information response from the precoding matrix by the UE. The UE can select the classification, and Node B can select the precoding matrix based on the selected classification. In another project, the UE can evaluate different possible precoding vectors / matrices for different classifications and can inform the precoding vector / matrix and classification. Node B can then apply the selected precoding matrix / vector.

As noted above, SC-FDM or OFDM can be used for transmission over a given link. The main motivation for choosing SC-FDM over OFDM for uplink is that a SC-FDM waveform has a lower peak-to-average energy ratio (PAR) than that of a waveform OFDM. The lowest PAR can allow a power amplifier to be operated closer to the peak energy level (or at medium high energy). SC-FDM can therefore have an advantage over OFDM in limited energy scenarios, such as for UEs that have a cell limit, due to their more efficient use of the energy amplifier.

However, UEs located close to Node B or in isolated cells can achieve sufficiently high geometry to justify MIMO transmission. For the MIMO (SU-MIMO) of a single 2x2 user, two streams can

24/33 be transmitted by an UE equipped with two antennas and two power amplifiers. For a SU-MIMO 4x4, four streams can be transmitted by an UE equipped with four antennas and four power amplifiers. In both cases, different streams can observe different channel conditions and can be reliably sent at different rates with different modulation and coding schemes (MCSs). The use of different modulation schemes for different flows can lead to different PARs for these flows. In addition, the processing of the MIMO transmitter such as permutation and

<td>precoding flows.</td><td>layer</td><td colspan="2">can also impact the PARs</td><td>From</td>

<td>An</td><td>simulation</td><td>computer</td><td>it was made</td><td>for</td>

<td>determine the</td><td>PARs of</td><td>waveforms</td><td>LFDM and</td><td>OFDM</td>

<td>for several</td><td>schemes</td><td>MIMO and</td><td colspan="2">modulation for</td>

<td colspan="2">2x2 MIMO configuration.</td><td>The simulation</td><td>computer</td><td>was</td>

performed for the following MIMO schemes:

. Through the antenna rate control (PARC) each stream is sent from a physical antenna without precoding or layer exchange.

. Layer permutation - each stream is sent over all antennas used for MIMO transmission, and

Precoding (or mapping the virtual antenna) - each stream is sent on a virtual antenna formed with a column of a precoding matrix.

PARC can be achieved by omitting layer permutation and performing precoding with the identity matrix. The layer permutation can be achieved through the cycle by the antennas in different subcarriers and / or in different symbol periods. Layer permutation can allow a flow to observe a

25/33 signal to noise ratio and interference (SINR) for all antennas.

Two streams can be sent using two antennas based on one of the MIMO schemes. The computer simulation indicates that the PAR of an LFDM waveform is lower than the PAR of an OFDM waveform for all MIMO schemes and modulation schemes. For LFDM with PARC, PAR for QPSK is lower than PAR for 16-QAM, which is lower than PAR for p 64-QAM. For LFDM, the PAR of each output stream with a layer permutation is between the PARs of the two output streams with the PARC. Also for LFDM, the PARs of the outgoing streams with precoding are higher than (i) the PARs of the outgoing streams with a layer permutation and (ii) the PARs of the outgoing streams with the PARC.

The following observations can be made:

For a classification-1 transmission, it may be advantageous from a PAR perspective to perform precoding with a unitary matrix in order to use all available power amplifiers.

. If the number of streams is equal to the number of antennas, or L = T, then precoding with a unitary matrix can degrade performance due to the increase in PAR. Pre-coding with the identity matrix can provide a lower PAR.

. If the number of streams is less than the number of antennas, or L &lt;T, then it may be advantageous to perform precoding with a unitary matrix in order to use all available energy amplifiers.

Figure 5 shows a design of a process 500 for transmitting data with dependent line precoding. Process 500 can be performed by a

26/33 transmitter, which can be a Node B for the transmission of the downlink or a UE for the transmission of the uplink.

The transmitter can obtain a precoding vector for a rating-1 transmission from a first set comprising at least one column vector of a unitary matrix (block 512). The unitary matrix can be a Fourier matrix, an offset phase Fourier matrix, or some other type of unitary matrix. The transmitter can perform precoding for the classification -1 transmission based on the precoding vector (block 514). The transmitter can obtain a precoding matrix for a classification-2 transmission from a second set comprising an identity matrix (block 516). The transmitter can perform precoding for the classification-2 transmission based on the precoding matrix (block 518).

For block 514, the transmitter can perform precoding for a data stream with precoding vector to obtain multiple output streams for multiple transmission antennas. For block 518, the transmitter can perform precoding for two data streams with the precoding matrix to obtain multiple output streams for the multiple transmission antennas.

In a 516 block design, the transmitter can determine if a MIMO Channel looks like a diagonal channel that has an almost diagonal channel response matrix with small off-diagonal channel gains. This determination can be based on antenna configurations (i) at Node B and UE, (ii) an estimated MIMO channel obtained by the receiver, and / or (iii) some other information. Ό transmitter can select the identity matrix as the

27/33 precoding matrix for rating-2 transmission if the MIMO channel looks like a diagonal channel. For example, the transmitter can select the identity matrix if Node B and / or the UE is equipped with cross-polarized antennas. The second set may additionally comprise the unitary matrix. The transmitter can select the unitary matrix as the precoding matrix for the classification-2 transmission if the MIMO channel does not look like a diagonal channel.

In a project, the transmitter can select the identity matrix as a precoding matrix for the L-rating transmission if L is equal to the number of transmission antennas. The transmitter can then perform precoding for the L-rating transmission based on the identity matrix. The transmitter can select a unitary matrix as the precoding matrix for the L-rated transmission if L is less than the number of transmission antennas. The transmitter can then perform precoding for the L-rating transmission based on the unit matrix.

The transmitter can select the precoding vector for the rating-1 transmission and the precoding matrix for the rating-2 transmission. Alternatively, the transmitter can receive the precoding vector and / or the precoding matrix from the receiver. The receiver can evaluate different possible precoding vectors in the first set and different possible matrices in the second set. The receiver can then send the best performing precoding vector and matrix to the transmitter.

Figure 6 shows a design of a device 600 for data transmission with line precoding

28/33 dependent. The apparatus 600 includes mechanisms for obtaining a precoding rating-1 vector from a transmission of a first set comprising at least one column vector of a unitary matrix (module

612), mechanisms for performing precoding for the rating-1 transmission based on the precoding vector (module 614), mechanisms for obtaining a precoding matrix for a rating-2 transmission from a second set comprising an identity matrix (module 616), and mechanisms to perform precoding for the classification-2 transmission based on the precoding matrix (module 618).

Figure 7 shows a design of a 700 process for receiving data with dependent line precoding. Process 700 can be performed by a receiver, which can be a UE for the transmission of the downlink or a Node B for the transmission of the uplink.

The receiver can receive a classification-1 transmission sent from multiple transmission antennas with a precoding vector selected from a first set comprising at least one column vector of a unitary matrix (block 712). The receiver can process the rating-1 transmission to retrieve data streams sent in the rating-1 transmission (block 714). The receiver can receive a classification-2 transmission sent from multiple transmission antennas with a precoding matrix selected from a second set comprising an identity matrix (block 716). The receiver can process the classification-2 transmission to retrieve two data streams sent in the classification-2 transmission (block

718) .

29/33 [0114] For block 714, the receiver can derive a spatial filter vector for rating-1 transmission based on the precoding vector. The receiver can then perform detection for the classification-1 transmission based on the spatial filter vector. For block 718, the receiver can derive a spatial filter matrix for the classification-2 transmission based on the precoding matrix. The receiver can then perform MIMO detection for rating-2 transmission based on the spatial filter matrix. The receiver can perform MMSE detection or an MMSE-SIC detection for classification-2 transmission.

In a project, the receiver can evaluate at least one vector in the first set and at least one matrix in the second set based on a metric, for example, the transmission capacity of the sum. The receiver can select a vector or matrix with the best metric, for example, the transmission capacity of the highest sum. The receiver can send feedback information comprising the selected matrix or vector to the transmitter.

Figure 8 shows a design of a device 800 to receive data with dependent line precoding. The apparatus 800 includes mechanisms for receiving a classification-1 transmission sent from multiple transmission antennas with a precoding vector selected from a first set comprising at least one column vector of a unitary matrix (module 812), mechanisms for processing the classification-1 transmission to retrieve a data stream sent in the classification-1 transmission (module 814), mechanisms for receiving a classification-2 broadcast sent from multiple antennas of

30/33 transmission with a precoding matrix selected from a second set comprising an identity matrix (module 816), and mechanisms for processing the classification-2 transmission to retrieve two data streams sent in the classification-2 transmission (module 818).

The modules in Figures 6 and 8 can comprise processors, electronic devices, hardware devices, electronic components, logic circuits, memories, etc., or any combination of them.

Those skilled in the art would understand that information and signals can be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that can be taken as a reference throughout the above description can be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, fields or optical particles , or any combination of them.

Those skilled in the art could additionally consider that the various illustrative logic blocks, modules, circuits, and algorithm steps described in connection with this disclosure can be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, several illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or as software depends on the application and design restrictions in particular imposed on the general system. Those skilled in the art can implement the functionality described in different ways for each

31/33 application in particular, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.

The various illustrative logic blocks, modules, and circuits described in connection with this disclosure can be implemented or carried out with a processor that has a general purpose, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate arrangement (FPGA) or other programmable logic devices, discrete or transistor logic gate, discrete hardware components, or any combination of them designed to perform the functions described here. A general purpose processor can be a microprocessor, but in an alternative, the processor can be any conventional processor, controller, micro controller, or state machine. A processor can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other similar configuration.

The steps of a method or algorithm described in connection with this disclosure can be incorporated directly into hardware, a software module run by a processor, or a combination of the two. A software module can reside in RAM, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from, and write information to,

32/33 the storage medium. Alternatively, the storage medium can be incorporated into the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside on a user terminal. Alternatively, the processor and the storage medium can reside as discrete components in a user terminal.

In one or more example projects, the functions described can be implemented in hardware, software, firmware, or any combination of them. If implemented in software, the functions can be stored in, or transmitted through, as one or more instructions or code in a computer-readable medium. A computer-readable media includes both storage media and computer communication media including any means that facilitates the transfer of a computer program from one place to another. A storage medium can be any available medium that can be accessed by a general purpose or special purpose computer. By way of example, and not by way of limitation, such computer-readable media may comprise RAM, ROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage, or other magnetic storage devices, or any other means that can be used to transmit or store desired program code mechanisms in the form of instructions or data structures and that can be accessed through a general purpose or special purpose computer, or a general purpose processor or special purpose. In addition, any connection is properly designated as a computer-readable medium. For example, if the software is transmitted through a web site, server, or other

33/33 remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, and then coaxial cable, fiber optic cable , twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave, are included in the definition of medium. Disc or disc, as used here, includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disc and blu-ray disc where discs usually reproduce data magnetically, while discs reproduce data optically with lasers . Combinations of the above should also be included within the scope of computer-readable media.

The foregoing description of the disclosure has been provided to allow anyone skilled in the art to make or use that disclosure. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined here can be applied to other variations without departing from the spirit or scope of the disclosure. Therefore, the disclosure is not intended to be limited to the examples and projects described here, but the broader scope consistent with the principles and new features revealed here should be recognized.

wireless