BRPI0706477A2 - spherical detection and rate selection for minimal transmission - Google Patents

Abstract

SPHERE DETECTION AND RATE SELECTION FOR MIMO TRANSMISSION Techniques for performing spherical detection are described in order to recover the data symbols sent in a MIMO transmission. In one respect, spherical detection is performed for data symbols generated with at least two modulation schemes. In another aspect, spherical detection is performed for data symbols in a given order based on at least one attribute of the data symbols, which can be error probabilities, modulation schemes and / or link margins for the symbols of Dice. In yet another aspect, the rates for the various data streams detected by spherical detection are selected based on the channel state information. The qualities of the data streams can be estimated based on channel status information, such as (1) the upper triangular matrix used for spherical detection and / or (2) the assumption that the interference of the data streams already detected is canceled. Rates for data streams can be selected based on estimated signal qualities.

Description

SPHERIC DETECTION AND RATE SELECTION FOR MIMO TRANSMISSION

Priority Claim under 35 U.S.C.§119

The present patent application takes precedence for provisional application No. 60 / 758,344, entitled "SPHERIC DETECTION AND RATE SELECTION FOR A MINIMUM TRANSMISSION", filed on January 11, 2006 and is hereby incorporated by reference hereby.

BACKGROUND

I. Field

The present disclosure relates generally to communications and, more specifically, to techniques for performing rate detection and selection for a multiple input and multiple output (MIMO) transmission.

II. Background

A MIMO transmission is a transmission sent from various transmission antennas (T) to multiple receiving antennas (R). A MIMO channel formed by the transmitting T antennas and the receiving R antennas can be composed into M space channels, where M <min {T, R}. Space McAns can be used to transmit data to achieve higher total transmission capacity and / or greater security.

A transmitter can encode and transmit data streams in parallel via the transmission antennas. A receiver obtains received symbol streams through the receiving antennae, detects received symbol streams and decodes the detected symbol streams in order to retrieve the transmitted data streams. For optimal detection performance, the receiver would need to evaluate many hypotheses for all possible bit sequences that might have been transmitted based on all information available at the receiver. Such exhaustive search is computationally intensive and is prohibitive for many applications.

There is therefore a technical need in the art to perform detection with reduced complexity while achieving good performance.

SUMMARY

Techniques for performing spherical detection to retrieve data symbols sent in a MIMO transmission are described herein. In one aspect, spherical detection is performed for data symbols generated with at least two modulation schemes. In another aspect, spherical detection is performed for data symbols in a determined order based on at least one attribute of the data symbols, which may be error probabilities for data symbols, modulation schemes used for data symbols, link margins. for the finger symbols and so on. In yet another aspect, rates for spherically sensed sensed data streams are selected based on channel state information. Channel state information may include channel estimates, noise estimates, interference estimates, power measurements, signal quality estimates, and so on. In one or more embodiments, the signal qualities of the data streams may be estimated based on (1) the upper triangular matrix used for spherical detection and / or (2) on the assumption that interference from already detected data streams is canceled. The rates for data streams can then be selected based on the estimated design qualities. In other embodiments, rates may be selected based on channel state information in other ways. Spherical detection and rate selection are described in detail below. Various aspects and features of the invention are also described below.

BRIEF DESCRIPTION OF DRAWINGS

The characteristics and nature of the present invention will become more apparent from the following detailed description, when considered in conjunction with the drawings, in which the same references identify the same elements everywhere.

Figure 1 shows a perspective block diagram of a transmitter and a receiver.

Figure 2 shows a perspective block diagram of a transmit data (TX) processor and a TX spatial processor in the transmitter.

Figure 3 shows aspects of an exemplary spindle tree for spherical detection.

Figure 4 shows aspects of a process for performing spherical detection in a selected order.

Figure 5 shows aspects of an equipment for performing spherical detection in a selected order.

Figure 6 shows aspects of a process for performing spherical detection on data symbols generated with various modulation schemes.

Figure 7 shows aspects of an apparatus for performing spherical detection for data symbols generated with various modulation schemes.

Figure 8 shows aspects of a process for selecting rates for data streams.

Figure 9 shows aspects of equipment to select rates for data streams.

Figure 10 shows a block diagram of a receiving space processor (RX) and a receiving RX data processor block. DETAILED DESCRIPTION

The word "exemplary" is used herein to mean "serving as an example, occurrence or illustration". Any embodiment or design described herein as "exemplary" should not necessarily be construed as preferred or advantageous compared to other embodiments or designs.

Acid-rate detection and selection techniques can be used in various communication systems, where multiple data streams are transmitted in parallel through a communication channel. For example, these techniques can be used in a single subcarrier MIMO system, a multiple subcarrier MIMO system, a Code Division Multiple Access (CDMA) system, a Frequency Division Multiple Access (FDMA) system , a Time Division Multiple Access (TDMA) system, and so on. Multiple subcarriers may be obtained with orthogonal frequency division multiplexing (OFDM), single behavioral frequency division multiple access (SC-FDMA) or some other demodulation technique. OFDM and SC-FDMA partition the total bandwidth of the system into various orthogonal subcarriers, also called tones, binaries, and so on, each of which can be modulated with independent data. In general, modulation symbols are sent in the frequency domain with OFDM and in the time domain with SC-FDMA. For the sake of clarity, much of the following description is for a single sub-carrier MIMO system.

Figure 1 shows a block diagram of a transmitter 110 and a receiver 150 in a MIMO 100 system. The transmitter 110 is equipped with several antennas (T), and the receiver 150 is equipped with several antennas (R). For downlink (or direct link) transmission, the transmitter 110 may be part of, and may contain some or all of the functionality of, a base station, an access point, a Node B, and so on. The receiver 150 may be part of, and may contain some or all of the functionality of, a mobile station, a user terminal, a user equipment and so on. For uplink (or reverse link) transmission, the transmitter 110 may be part of a mobile station, a user terminal, a godsend and so on, and the receiver 150 may be part of a base station, an access point, a Node B and so on.

At transmitter 110, a TX120 data processor receives traffic data from a data source 112 and processes (formats, encodes, interleaves, and maps symbols, for example) the traffic data to generate data symbols, which are modulation symbols for traffic data. A TX 130 spatial processor multiplexes data symbols with pilot symbols, which are demodulation pilot symbols. A pilot is a transmission which is known a priori by both the transmitter and receiver and may also be referred to as a training signal, a reference, a preamble and so on. The TX 130 space processor performs transmitter spatial processing and provides T transmitting symbol streams to T transmitter units (TMTR) 132a to 132t.

Each transmitter unit 132 processes (modulates with OFDM, converts to analog, filters, amplifies, and converts upwards, for example) its transmission symbol stream and generates a modulated signal. Modulated signals from transmitter units 132a to 132t are transmitted from antennas 134a to 134t, respectively. At receiver 150, antennas 152a to 152r receive the modulated signals, and each antenna 152 provides a received signal to a respective receiver unit (RCVR) 154.

Each receiver unit 154 processes its received signal in a manner complementary to the processing performed by transmitter units 132 in order to obtain received symbols. Each receiver unit 154 provides symbols received for traffic data to a spatial processor RX 160 and provides symbols received for pilot to a channel processor 194. Channel processor 194 estimates MIMO channel response from transmitter 110 to receiver 150 based on symbols received for pilot and provides channel estimates to the RX 160 space processor. The RX 160 space processor performs spherical detection of our received symbols with the channel estimates and detected symbols, which are estimates of the transmitted finger symbols. An RX 170 data processor additionally processes (deinterleaves and decodes, for example) the detected symbols and provides decoded data to a data store 172.

Receiver 150 may send feedback information to help transmitter 110 control data transmission to receiver 150. Feedback information may indicate a specific transmission mode to be used for transmission, a specific packet rate or format to be used for each data stream. , acknowledgments (ACKs) and / or negative acknowledgments (NAKs) for packets decoded by receiver 150, channel state information and so forth, or any combination thereof. Feedback information is processed (encoded, interleaved, and symbol mapped, for example) by a TX 180 signaling processor, multiplexed with pilot symbols, and spatially processed by a TX182 space processor, and further processed by transmitter units 154a through 154r to generate R modulated signals which are transmitted via antennas 152a to 152r.

At the transmitter 110, the modulated R signals are received by antennas 134a and 134t, processed by receiver units 132a to 132t, spatially processed by an RX 136 space processor and further processed (e.g., de-interleaved and decoded) by an RX 138 signaling processor so as to retrieve the feedback information. A controller / processor 140 controls the data transmission to receiver 150 based on the received feedback information. A channel processor 144 can estimate the MIMO channel response from receiver 150 to transmitter 110 and can derive spatial mapping matrices used by the TX 130 space processor.

Controllers / processors 140 and 190 control operation on transmitter 110 and receiver 150, respectively. Memories 142 and 192 store data and program codes for transmitter 110 and receiver 150, respectively.

Figure 2 shows a block diagram of the TX 120 data processor and TX 130 processors in transmitter 110. In this embodiment, a common coding scheme is used for all data streams, and a separate code rate and a separate modulation scheme. can be used for each data stream. For clarity, the following description assumes that M data streams are sent over M spatial channels.

However, this need not be the case, and a data stream can spread across multiple spatial channels.

Within the TX data processor 120, an encoder 220 encodes traffic data according to an encoding scheme and generates code bits. The decoding scheme may include a convolutional code, a Turbo code, a low density parity check (LPDC) code, a cyclic redundancy check code (CRC), a block code, and so on or a combination thereof. A Demultiplexer (Demux) 222 multiplies or parses the code bits in M stream and provides the M code bit streams in M sets of processing units. Each set includes a punching unit 224, a channel interleaver 226, and a symbol mapper 228. For each set, the punching unit 224 punctures or erases the code bits as needed to obtain a code rate selected for its flow. and sends the codeword bits to an associated channel interleaver 226. Channel interleaver 226 interleaves or reorders the code bits based on an interleaving scheme and provides interleaved bits to an associated symbol mapper 228.

Interleaving can be performed separately for each data stream (as shown in Figure 2) or through some or all of the data streams (not shown in Figure 2).

In one or more modalities, each data stream may be sent with a modulation scheme selected for that stream. In general, the same or different modulation schemes may be used for the data flows, depending on system operation, channel conditions and / or other factors. Each symbol mapper 228 maps its interleaved bits according to the selected modulation scheme for its stream and provides a data symbol stream {sm}. Mapping to symbols for flow m can be obtained by (1) grouping together Qm bits to form Qmbits values, where Qm> 1 and (2) mapping each Qm value to one of 2 ° points in a constellation of signals for the selected modulation scheme. Each mapped signal point is a complex value for a data symbol. Symbol mapping can be based on DeGray mapping or non-Gray mappings. With Gray mapping, the neighboring points in the desinal constellation (in both horizontal and vertical directions) differ by only one of Qm bit positions. Gray mapping reduces the number of bit errors to most likely error events, which correspond to a detected symbol that is mapped to a correct location near the correct location, in which case only one encoded bit would be incorrectly detected. With non-Gray mapping, neighboring points may differ by more than one bitposition.

Within the TX 130 space processor, a multiplexer (Mux) 230 receives the M symbol data streams from M mapper symbols 228a to 228m and multiplexes data symbols with pilot symbols. A matrix multiplier 232 multiplies the data and / or pilot symbols by spatial mapping matrices P and provides transmission symbols. In one or more embodiments, spatial mapping arrays are an identity matrix I, which results in the absence of non-transmitter spatial processing. In other embodiments, different spatial mapping arrays are used for different symbol periods and / or different subcarriers to achieve similar performance for M data streams. In still other embodiments, spatial mapping arrays are self-vector arrays. .

Figure 2 shows aspects of a common decoding scheme, and separate code rates and demodulation schemes can be used for M data streams. Different code rates can be obtained for M data streams by using different operating patterns for these streams. In other embodiments, a common coding scheme and common coding rate are used for all data streams, and separate modulation schemes may be used for data streams. In still other embodiments, a common coding scheme, a common data rate, and a common modulation scheme are used for all data streams. In still other embodiments, each data stream is independently processed based on a selected coding and modulation scheme for that data stream. In general, the same or different decoding schemes, the same or different code rates and the same or different modulation schemes may be used through the subcarriers. If several subcarriers are available, then the same or different coding schemes, the same or different coding schemes, and the same or different demodulation schemes may be used across the subcarriers.

Transmitter 110 typically encodes each data packet separately. A package can be split into several blocks, with each block containing Kbits of code. The K bits of code in each block can be mapped to M data symbols as follows:

<formula> formula see original document page 11 </formula> where s = [si sm2. .. Sm] 1 "is a vector with M data symbols;

<formula> formula see original document page 12 </formula>

is a vector with K bits of code in a block;

bm is a vector with Qjn code bits used to form the data symbol sm;

bm, q, for m = 1, ..., Meg = I,. . ., Qm is gth bit of code in the vector bm;

bk, for k = 1, .., K, is the Jcth noveltor code bit b; and

"r" denotes a transposition.

Equation (1) indicates that there is a one-to-one mapping between a given bit vector b and a data vector s.

In general, the same or different modulation schemes may be used for the M data symbols sent in a given data vector s. Therefore, Qi to Qm can be the same or different for the M data symbols in the vector s.

In one or more embodiments, the M data streams are coded together so that a single packet may be sent on several (all M, for example) space channels. In other embodiments, the data streams are independently encoded so that each packet is sent on a spatial channel. In still other embodiments, some data streams are coded together, while other data streams are independently coded.

For clarity, the following description assumes that a data stream is sent on each space channel. The terms "data stream" and "space channel" are therefore interchangeable for much of the description that follows. The number of data streams can be configurable and can be selected based on channel conditions and / or other factors. For clarity, the following description assumes that M data streams are sent on M spatial channels.

1. Detection

The symbols received at the receiver can be expressed as follows:

2 = Hch · P * s + η = H * s + η, Eq (2)

where P is a TxM spatial mapping matrix used by the transmitter;

Hch is a real MIMO channel response matrix RxT, H = Hch · P is an effective MIMO channel response matrix RxM;

Y is an R χ 1 vector with R symbols received noreceptor; and

η is an R χ 1 noise vector.

It can be assumed that the noise is an additive white Gaussian noise (AWGN) with a zero mean vector and a variance matrix of, where is the noise variance.

Response to the effective MIMO channel H includes response to the actual MIMO channel Hch and the spatial mapping matrix P used by the transmitter. The effective MIMO channel response matrix can be given as:

<formula> formula see original document page 13 </formula>

where the input hr, mr for r = 1, ..., R and m = 1, ..., M denotes the complex channel gain observed by the data stream at the receiving antenna r. For simplicity, the MIMO channel is assumed to have flat fading without frequency selectivity. The receiver typically derives, which is an estimate of H, and uses for detection. For simplicity, in the present description it is assumed that there is no channel estimation error, so that H is also referred to as the MIMO channel response matrix.

For a MIMO transmission on a single sub-carrier, the receiver obtains a received symbol vector γ at each symbol period used in the transmission. For a MIMO transmission on multiple subcarriers, the receiver has a received symbol vector jr for each subcarrier in each symbol period used for transmission. In one or more embodiments, the receiver detects separately for each received symbol vector.

In other embodiments, the receiver performs joint detection for various received symbol vectors. The receiver can perform detection in many ways.

The receiver can perform maximum probability detection (ML) on received symbols to obtain detected symbols. For ML detection, the receiver evaluates each of 2K symbol vectors presented as hypotheses that may have been transmitted to the symbol vector data s. For each symbol vector presented as a hypothesis, the receiver computes a distance metric that can be given as:

<formula> formula see original document page 48 </formula>

where is a vector of symbols presented as hypotheses as having been transmitted to the vector s; and

is the distance to the symbol vector presented as hypothesis. Equation (4) can also be referred to as the decust function, error function, and so on. can also be referred to as cost value or error value for S vector.

For ML detection, the receiver obtains 2K distances for the 2K symbol vectors presented as hypotheses. The receiver can then derive the detected symbols as follows:

<formula> formula see original document page 15 </formula>

where ~ is an M χ 1 vector with M detected symbols. In equation (5), the minimum distance between 2K distances for the 2K symbol vectors presented as hypotheses is identified. The symbol vector presented as hypothesis with the minimum distance is presented as the detected symbol vector which is an estimate of the transmitted data symbol vector s.

For ML detection, an exhaustive search is performed through all combinations of data symbols that may have been transmitted to the data symbol vector s. The exhaustive search considers all 2K hypotheses possible for the data symbol vector s. Therefore, the ML detection complexity is exponential in the number of bits (K) used to form the data symbol vector s. ML detection can provide good performance. However, exhaustive abuse is computationally intensive and can be prohibitive for many applications. For example, if four data streams are sent via QPSK for all streams, then K = 8 and 256 hypotheses are evaluated for each received symbol vector. However, if 16-QAM is used for all four streams, then K = 16 and 65,536. Hypotheses are evaluated for each received symbol vector that is much more complex. If the four data streams are sent using QPSK, 16-QAM, 64-QAM, and 256-QAM, then K = 20 and more than one million hypotheses are evaluated for each received symbol vector χ, which may be impractical. This example illustrates the rapid increase in the number of hypotheses for larger signal constellation sizes.

The number of hypotheses to consider can be reduced by performing spherical detection (SD), which is also referred to as spherical list detection, spherical decoding, and so on. Spherical detection seeks to reduce the ML detection search space by conducting a search for candidate hypotheses and discarding less likely hypotheses.

For spherical detection, the distance metric in equation (4) can be simplified by decomposing QR from the MIMO H channel response matrix as follows:

<formula> formula see original document page 16 </formula>

where Q is an orthonormal matrix RxM and R is an upper triangular matrix MxM. The orthonormal matrix Q has normal columns and unit power for each column, or Qh · Q = I, where "H" denotes a conjugate transposition. The upper triangular matrix R contains zeros below the diagonal. The structure of the upper triangular matrix R can be explored to reduce the number of hypotheses to be evaluated.

Equation (2) can be rewritten as follows: <formula> formula see original document page 17 </formula>

where - is a rotated version of y and n '= QH. no

The distance metric in equation (4) can be described as follows:

<formula> formula see original document page 17 </formula>

Equation (9) can be expressed as follows:

<formula> formula see original document page 17 </formula>

For M = 4, equation (10) can be further expanded as follows:

<formula> formula see original document page 17 </formula>

The set of equations (11) can be generalized to any value of M as follows:

<formula> formula see original document page 17 </formula> where Dm + ι = 0. For equation (12), the index i moves backwards from M down to 1.

As shown in equations (10) to (12) the metric distance can be computed in an incremental way with M terms Di to Dm. The term Dbι only depends on the symbol presented as a hypothesis and represents the distance to this symbol. The term Aj-ι depends on the symbols>: "e and represents the aggregate distance for these two symbols. Each subsequent term depends on one symbol presented as an additional hypothesis. The term Ddepends on all M symbols presented as hypotheses a and represents the total distance for all these symbols. The distance metric can be computed incrementally at M levels, a term Diem each level, beginning with the last term Dll of the first level.For each level, Di is computed for all applicable hypotheses for that level.

Spherical detection can be performed in several ways. Several embodiments of spherical detection are described below.

For total spherical detection, all hypotheses with distances equal to or less than a Dth limit are retained, and all other hypotheses are discarded. The limit Dth is also called sphere radius. Full spherical detection can be performed as follows. For the first level with i = M, a Pm list is formed with 2nd symbols presented as hypotheses that may have been transmitted to the data symbol sM, which is generated based on a constellation of signals with 2nd "signal points such as, for example, 20 "1 QAM. Dm is computed for the 2M symbols presented as hypotheses in the PM list, as shown in equation (12), to obtain 20 1 distances. All symbols presented as hypotheses with distances less than or equal to the limit are stored in a list of candidates. the deisissues presented as hypotheses are discarded, which has the effect of pruning all symbol vectors presented as hypotheses containing the discarded symbols.

For the second level with i = M - 1, a PM-i list is formed with 2Qm_1 symbols that may have been transmitted to the sM-i data symbol, which is generated by combining a signal constellation with 2Qm ~ 1 signal points.

Ai-i is computed for all symbol pairs presented as valid hypotheses in order to obtain the distances for these symbol pairs presented as hypotheses. The symbol pairs presented as valid hypotheses include all possible combinations of each symbol in the list of Cmd candidates with each symbol in the list ΡΜ-. All symbol pairs presented as hypotheses with distances less than or equal to the limit are stored in a candidate list CM_i, and all other symbol pairs presented as hypotheses are discarded.

Each of the remaining levels can be assessed in a similar way. A P ± list is formed with 2Ql symbols presented as hypotheses that may have been transmitted to the s ± f data symbol which is generated by combining a signal constellation with 2Ql signal points.

Di is computed for all symbol sets presented as valid hypotheses' so as to obtain distances for these symbol sets presented as hypotheses. Symbol sets presented as valid hypotheses include all possible combinations of each hypothesis in the list of candidates Ci + i with each symbol All of the symbol sets presented as hypotheses with distances less than or equal to the limit are stored in a candidate list Ci, and all other symbol sets are discarded After all Levels have been evaluated, the detected symbols can be determined based on the hypotheses. For more clarity, the above description uses different candidate lists for different levels.A single C candidate list can also be used for all M levels and can be updated at each level.

Figure 3 shows an exemplary search tree for spherical detection of data symbols that can be generated with different modulation schemes. In this example, M = 4 and four terms Di to D4. For the first level with i = 4, D4 is computed for 2 ° hypotheses for 2 ° 4 possible data symbols that may have been transmitted to data symbol S4. The 2 ° 4 symbols presented as hypotheses are denoted 4V 'a * in Figure 3. Two hypotheses have distances less than or equal to the limit and are shown with black filled nodes. For the second level with i = 3, D3 is computed for 2 · 2 ° 3 hypotheses for 2 · 203 possible symbol pairs that may have been passed to data symbols S3 and s4. Again, two hypotheses have distances less than or equal to the limit and are shown with black filled nodes. For the third level with i = 2, D2 is computed for data symbols s2, S3 and S4. Three hypotheses have distances less than or equal to the limit and are shown with black filled nodes. For the last level with i = 1, D1 is computed for 3 · 201 hypotheses for 3 · 2Q1 possible sets of symbols that can be transmitted to the data symbols s \, S2, S3 and S4.

Four hypotheses have distances less than or equal to the limit and are shown with black filled nodes. The set of symbols with the shortest distance is shown by the thickened line.

For partial spherical detection, the best hypothesis Nbs are retained for each level and used to form hypotheses for the next level. As shown in the set of equations (11), the QR decomposition allows data symbol S4 to be detected separately by removing interference from other data symbols. The detection of the following data symbolS3 relieves interference from the data symbolS4. This interference is given as in equation (Ilb).

The accuracy of interference estimation and the effectiveness of interference jamming both depend on symbol correction *. If there are no errors in the channel estimates, then interference from data symbol s4 can be completely canceled from detection of data symbol S3.

In one or more embodiments, the number of best hypothesis symbols (Nbs) to be retained for each level is a fixed value, such as 2, 3, 4, and so on. In other embodiments, Nbs is a configurable value which may depend on the size of the constellation for the detected Si data symbol, signal quality and other status information (s) for the Si data symbol and / or other criteria. Signal quality can be quantified by signal-to-noise ratio (SNR), signal-to-noise-to-interference ratio (SINR), total energy-to-noise-to-noise ratio (Eo / N0), and so on. For example, Nbs can be set to a quarter of the size of the constellation, Nbs = 2Ql / 4 or some other percentage. Nbs may be selected such that the probability that the transmitted data symbol is among the best Nbs presented as hypotheses reaches or exceeds a predetermined percentage, such as 95% or some other percentage.

For restricted spherical detection, the best hypothesized symbol is retained for each level and is referred to as a "hard" decision for that level. Restricted spherical detection is a special case of partial spherical detection with Nbs = 1. The hard decision for level can be given as:

Sj =. arg {min Di>, Eq (13)

where * is a hard decision for the transmitted data symbol s ±. In equation (13), the symbol presented as hypothesis * that produces the minimum distance for Di is presented as the hard decision for the data symbol s ±. The hard decision for each level can be taken to the next level and used to compute the distances to the next level.

The distance metric in equation (12) can then be expressed as:

<formula> formula see original document page 22 </formula> The summation term in equation (14) can be considered as the interference of earlier detected symbols. In this case, the modified received symbol 1 obtained after interference interference can be expressed as:

<formula> formula see original document page 23 </formula>

for i = M,. . . .1 Eq (15)

The distance metric in equation (14) can then be rewritten as:

for i = M, ..., 1 Eq (16)

As an example, if two data streams are sent and then the distance to the Sz data symbol can be expressed as:

<formula> formula see original document page 23 </formula>

The distance to the data symbol Si can then be expressed as:

<formula> formula see original document page 23 </formula>

Where

For partial and restricted spherical detection, the number of hypotheses to be evaluated at each level after the first level can be substantially reduced by retaining the Nbs best symbols presented as hypotheses for each level, and the number of hypotheses to be evaluated at each level is 2Ql. For the second level, different symbols presented as hypotheses and the hard decision can be evaluated, instead of different pairs of symbols presented as hypotheses ^ 5.1- '. To the third level, different symbols presented as hypotheses * and hard and ^ decisions can be evaluated instead of different sets of symbols.

The restricted spherical detection scheme has computed distances using hard decisions to already detected data symbols, as shown in equation (14). The performance of spherical detection then depends on the reliability of hard decisions. The reliability of the hard decisions for a given data stream depends on the modulation scheme and the desinal quality of that flow. In one or more embodiments, the detection order is selected based on the symbol probabilities or expected symbol error rates (SERs) for the data streams. The SER for each data stream can be estimated or verified based on information about channel state (s). In this mode, detection is first performed for the data stream with the lowest SER. This data stream should have the highest link margin with respect to unencoded signal quality, such as SNR, which is a requirement for the modulation scheme used for this data stream. Detection is then performed for the lowest SER data stream, and so on. This detection ordering reduces the timing of earlier detected stream symbol (s) errors in distance computation for later detected streams. The rate for each data stream can be selected to obtain a desired SER.

A rate or packet format may be selected for each data stream based on its signal quality, as described below. The rate may be associated with a specific spectral efficiency, which may be given in units of bits per second by Hertz (bps / Hz). A given rate can be obtained with different modulation scheme and code rate combinations. For example, a rate of 3 bps / Hz can be obtained with (1) 3/4 code rate and 16-QAM, (2) 1/2 code rate and 64-QAM or (3) 3/8 code rate. and 256-QAM. These different code rate and modulation scheme combinations may require slightly different signal qualities in order to obtain the target packet error rate (PER).

However, these different combinations of code rate and modulation scheme may have substantially different SERs. For a given signal quality, SER for 16-QAM is lower than SER for 64-QAM, which is lower than SER for 256-QAM. The progressively higher SERs for 16-QAM, 64-QAM, and 256-QAM are due to the progressively higher codes (or lower code rates) for 16-QAM, 64-QAM, and 256-QAM.

The performance of restricted spherical detection is affected by the SERs for the previous data streams (the first data streams, for example) to be detected.

The rate for each of the previous detected streams can be selected to obtain the target SER or a lower SER. This target SER can be 5%, 10% or some other value. In one or other embodiments, if the SER for a previous detected stream (the first, for example) exceeds the target SER, then the rate for the stream is reduced to a higher rate with a lower order modulation scheme, which then reduces the spread of errors for interference cancellation. Selecting a lower order modulation scheme may reduce the transmission capacity of the previous detected stream, but may improve the transmission capabilities of the subsequent detected streams. Computer simulation indicates that this rate selection strategy can improve overall transmission capacity for certain channel conditions, such as high SNRs.

For the spherical detection schemes described above, the number of candidate hypotheses to be stored in list C can be trimmed in several ways. In one or more embodiments, all hypotheses with distances equal to or less than the Dth limit are retained. For this mode, the number of candidate hypotheses to be stored at each level is not necessarily constant.

In other embodiments, the number of candidate assumptions to be retained at each level may be a function of the expected SER, which depends on the modulation scheme and signal quality of the data stream that is detected. In still other embodiments, even better hypotheses with distances equal to or less than the Dth limit are retained at each level. In still other embodiments, better hypotheses with distances equal to or less than the Dth limit are retained for each node. Nbi and Nbn may be selected based on a trade-off between detection performance, complexity and / or other considerations. For example, Nble Nbn can be selected based on the size of the signal constellation, so that more chances are stored for larger signal constellations. Nbi and Nbn can also be limited to being equal to or greater than some minimum value (Nmin = 2, for example), which ensures that at least Nltlin candidate hypotheses are stored for each level or node. In general, any number of hypotheses can be stored in the C-candidate list. Upon completion of the spherical detection, Yog-Likelihood ratios (LLRs) can be computed for the C-code candidate bits, as follows:

<formula> formula see original document page 27 </formula>

where is a bit vector that corresponds to the symbolic vector presented as hypothesis;

is a vector with all code bits of vector S h except the code bit;

is a vector with a priori LLRs for all code bits and

is a subset of the C candidate list and contains assumptions for which bk = + 1;

is a subset of candidate list C and contains hypotheses for which bk = -1; and

is the extrinsic LLR for code bit bk.

In general, detected symbols may be presented as LLRs or in some other form.

Equation (18) can be evaluated for each decode bit in the transmitted bit vector b. For each decode bit bk, all symbol vectors presented asS.

assumptions from candidate list C may be considered. Each symbol vector presented as a - · hypothesis has a bit vector presented as a corresponding hypothesis. For equation (18), the expression within the max operation is computed for each byte vector presented as a hypothesis & in order to obtain a result for this bit vector. The highest result for all bit vectors presented as hypotheses - · combk = +1 is identified. The highest result for all bit vectors presented as hypotheses - · with bk = -1 is also identified. The LLR for code bit bk equals the difference between the highest result for bk = +1 and the highest result for bk = -1.

Figure 4 shows aspects of a process 400 for performing spherical detection. A data symbol detection order in a MIMO transmission is selected based on at least one data symbol attribute (block412). In one or more modalities, the order is selected based on error probabilities for the data symbols, starting with the data symbol with the lowest error probability. In other embodiments, the order is selected based on modulation schemes for the data symbols, starting with the data symbol with the lowest order modulation scheme. In still other embodiments, the order is selected based on link margins for data symbols, starting with the data symbol with the largest link margin.

Spherical detection is performed for data symbols in the selected order (block 414). For spherical detection, a channel response matrix can be decomposed to obtain a superior triangular matrix. Data symbols can be detected one at a time in the selected order. For each data symbol, the distances for various hypotheses of the data symbol can be computed based on the received symbols, the upper triangular matrix, and the candidate assumptions and / or hard decisions for the already detected data symbols. data are determined based on the computed distances. The LLRs for the code bits of the data symbols are computed based on the candidate assumptions (block 416).

Figure 5 shows aspects of an equipment 500 for performing spherical detection. The apparatus 500 includes devices for selecting a detection order of data symbols sent in a MIMO transmission based on at least one attribute of the data symbols, such as error probabilities, modulation schemes and / or link margins (block 512), spherical detection devices for data symbols in the selected order (block 514) and devices for computing LLRs for data symbol code bits based on the spherical detection candidate hypotheses (block 516).

Figure 6 shows aspects of a process 600 for performing spherical detection. A detection order of data symbols sent in a MIMO transmission is selected, for example, based on deer probabilities, modulation schemes, link margins, and so on (block 612). Spherical detection is then performed on the received symbols in order to detect the generated data symbols with at least two modulation schemes (block 614). Spherical detection can be carried out using basic modulation schemes used for finger symbols. In one or more embodiments, the number of assumptions to be evaluated for each data symbol is determined based on the modulation scheme for the data symbol. In other embodiments, the number of assumptions to be retained for each data symbol is determined based on the schema. modulation for the data symbol. The LLRs for the data symbols data bits are computed based on candidate assumptions for the data symbols (block 616).

Figure 7 shows aspects of an equipment 700 for performing spherical detection. The apparatus 700 includes devices for selecting a detection order of data symbols sent in a MIMO transmission (block 712), devices for spherical detection of received symbols to detect finger symbols generated with at least two modulation schemes (block 714), and devices for computing LLRs for data symbol code bits based on hypothesized data symbol (block 716).

2. Rate Selection

A rate or packet format can be selected for each data stream to obtain a target performance level, which can be quantified by a target PER, such as a 1% PER. The rate for each data stream may be selected based on information about channel state (s), such as data flow signal quality, which can be estimated as described below.

For spherical detection with QR decomposition, signal quality, such as SNR, of each data stream may depend on the order in which the flow is detected. For the simple case of two data streams, with stream 2 first detected, followed by stream 1, the SNR of each data stream can be expressed as follows: <formula> formula see original document page 31 </formula>

Where

are the SNRs of flows 1 and 2 respectively, with spherical detection.

The receiver may also implement a successive interference cancellation (SIC) scheme and may perform spatial matching filtering and successive interference cancellation using hard decisions. For the SIC scheme, the receiver retrieves the M-Stages data streams, a data stream at each stage, and estimates and balances the interference caused by each retrieved data stream. For the first stage, the receiver performs spatially matched filtering on received symbols γ and obtains detected symbols for a data stream. Spatial-space filtering can be based on the force-to-zero (ZF), minimum mean square error (MMSE), maximal combination (MRC), or some other technique. In order to encode interference coding, the receiver processes (unclips symbols, decalculates and decodes, for example) the detected symbols to obtain encoded data and also processes (encodes, interleaves, and modulates, for example) to obtain decoded symbols, which are estimates. for data symbols that have just been decoded. Oreceptor also processes remodulated symbols with channel estimates and obtains the iM interference components due to the recovered data stream. The interference components iM are then subtracted from the received symbols j; in order to obtain modified received symbols which have the interference components removed. The modified received symbols m-I are then processed by the next stage.

The restricted spherical detection scheme is equivalent to the SIC scheme with uncoded interference cancellation. For spherical strict detection, hard decisions are obtained for the Sj + i to sM finger symbols and used for interference cancellation. The modified received symbol is based on hard decisions for the already detected data symbols, as shown in equation (15). Ideally, it is desirable to perform interference cancellation using the remodeling symbols generated from the decoder output, as these symbols tend to be more reliable than hard decisions. In many cases, however, demodulated symbols are not available due to processing complexity and / or latency.

A parallel can be drawn between spherical detection and the SIC scheme. SNRs of spherically detected detected data streams can be estimated by the SNRs of data streams retrieved with the SIC scheme.

For the SIC scheme, a spatial filter vector can be derived to data stream m based on the stress-to-zero (ZF) or MMSE technique, as follows:

<formula> formula see original document page 32 </formula> where Hm is an Rxm reduced channel response matrix for data stream m;

hm is an R χ 1 channel response vector for data stream m; and

MmmrfW S ~ Q R χ 1 vectors ^ and filter for data flow m for forcing to zero eMMSE techniques, respectively.

Hm contains m H columns for m undetected data streams, with M - m H columns for data streams already detected at earlier stages removed.

The detected symbol for data stream m can be expressed as follows:

<formula> formula see original document page 33 </formula>

where is an R χ 1 received symbol vector modified for stage m; and

mm can be equal to

The data stream SNR for the SIC schema can be expressed as follows:

<formula> formula see original document page 33 </formula>

where e are the data stream SNRs with techniques of forcing to zero and MMSE, respectively. AsSNRs computed for SIC schema-based data streams with the force-to-zero or MMSE technique can be used to select the rates for spherically detected detected data streams.

It can be shown that, for the case of two data streams and perfect interference cancellation, for the data stream detected second with the SIC scheme is equal to for the data stream detected second with spherical detection. '* n "Si'i ~ ssc- ~ for the first detected data flow with the SCI scheme is almost identical to Λ' Λ for the first detected data flow with spherical detection at high SNRs.

Thus, rates for spherically sensed data streams can be selected based on SNRs computed for data streams with the MMSE-SIC scheme.

The signal qualities of spherically sensing data streams can also be estimated in other ways. Estimated signal qualities for data streams can be used to select appropriate rates for data streams.

In one or more embodiments, the system supports a set of rates or packet formats. Each supported rate may be associated with specific spectral efficiency, a specific code rate, a specific modulation scheme, and a specific minimum SNR required to obtain the target PER for a non-fading AWGN channel. Supported rates and required SNRs can be stored in a lookup table. The rate for each data stream can be selected depending on the SNR computed for that stream.

In other embodiments, the system supports a vector-quantized rate set, which may also be called a modulation coding scheme (MCS) set. The vector-quantized rate set contains only certain rate combinations. The rates for the M data streams can be selected together from the rate set combinations of rates.

For both modalities, different rates may be selected for different data streams based on their signal qualities and / or other factors. The ability to use different rates for data streams can improve overall transmission capacity.

In one or more embodiments, rates for detected data streams with spherical detection may be iteratively selected. Initial rates can be selected for data streams based on their signal quality. If the SER for an earlier detected flow is higher than the target SER, then another rate with a lower order modulation scheme may be selected for the flow, and the transmission capacity may be determined for all flows. The combination of rates with the highest transmission capacity for all data streams can be selected for use.

Figure 8 shows aspects of a process 800 for selecting rates for data streams. Spherical detection is performed for multiple data streams (block812). Channel state (s) information is obtained for the spherically sensed detected data streams (block814). Channel status information (s) may comprise channel estimates (channel response matrix, for example), noise estimates, interference estimates, power measurements, signal quality estimates, and / or other information. Rates are selected for data streams based on channel state (s) information (block 816). In one or more modalities, rates for data streams are selected based on the signal qualities (SNRs, for example) of the data streams, which can be estimated based on channel state (s) information. Signal qualities for data streams can be estimates based on a superior triangular matrix used for spherical detection. The upper triangular matrix may be derived from a channel response matrix, which may be part of the state information (s). The signal quality of each data stream may also be estimated based on the SIC scheme with the assumption that the interference of the flows already detected data is canceled. The rate for each data stream can be independently selected. Rates for all data streams can also be selected together. The rate for a data stream (the first data stream to be detected, for example) can be selected to obtain a target SER or better for that data stream. An initial rate may be selected for the data stream based on its estimated signal quality, and a revised rate with a lower order modulation scheme may be selected if the initial rate results in exceeding the target SER by the data stream.

Figure 9 shows aspects of 900 equipment for selecting rates for data streams. Equipment900 includes devices for performing spherical detection for multiple data streams (block 912), devices for obtaining channel state information (s) for data streams (block 914), and devices for selecting rates for data streams based on state information. channel (s) (block 916). The rate for one or more finger streams can be selected to obtain the target SER or better.

Figure 10 shows a block diagram of the space processor RX 160 and data processor RX 170 on receiver 150. Within the special processors 160, a computing unit 1010 receives H-channel estimates from channel processor 194 and derives the orthonormal matrix Q and matrix. upper triangular R. A spherical detector 1020 performs spherical detection of our received symbols from the receiver units 154a to 154r with the matrices QeRe generates detected symbols or candidate hypotheses. Spherical detector 1020 can detect in an order determined by controller 190. Spherical detector 1020 can make hard decisions on our detected symbols and can compute distances based on hard decisions. Spherical detector 1020 may consider all assumptions for each data symbol or, if spatial matched filtering is performed, may only consider a subset of hypotheses that are close to a detected symbol provided by spatial matched filtering.

An LLR 1030 computing unit computes the LLRs for the code bits based on the detected symbol or candidate hypotheses of the spherical detector 1020.

Within the RX 170 data processor, 1040m to 1040m channel deinterleavers receive the LLRs for the LLRs1030 compute unit data streams. Each channel deinterleaver 1040 deinterleaves asLLRs for its flow in a manner complementary to the interleaving performed by channel interleaver 226 for that flow.

A 1050 multiplexer multiplexes or serializes the deinterleaved LLRs of the 1040a to 1040m channel deinterleaver. A 1060 decoder decodes the de-interleaved LLRs and generates decoded data. The techniques described herein can be implemented by various devices. For example, these techniques may be implemented in hardware, firmware, software, or a combination of them. For a hardware implementation, processing units used for sensing, rate selection, and so forth may be implemented within one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing apparatus (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, other electronic units designed to perform the functions described herein or a combination thereof.

For a firmware and / or software implementation, techniques may be implemented with modules (such as procedures, functions, and so forth) that perform the functions described herein. Firmware and / or software codes may be stored in memory (memory 192 of Figure 1, for example) and executed by a processor (processor 190, for example). Memory can be deployed inside the processor and outside the processor.

The foregoing description of the disclosed embodiments is presented to enable any person skilled in the art to manufacture or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the defined generic principles may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown, but should be given the widest scope consistent with the principles and novel aspects disclosed herein.

Claims (46)

Equipment comprising: at least one processor configured to select an order of detection of data symbols sent in a MIMO transmission based on at least one attribute of the data symbols and to perform spherical detection for data symbols in the selected order; a memory coupled to at least one processor. The apparatus of claim 1, wherein at least one attribute comprises error probabilities for data symbols, and wherein at least one processor is configured to select the order based on error probabilities for data symbols starting with a data symbol with the lowest probability of error. The apparatus of claim 1, wherein at least one attribute comprises modulation schemes for data symbols, and wherein at least one processor is configured to select order based on demodulation schemes for data symbols, starting with a data symbol with the modulation scheme of the lowest order. The apparatus of claim 1, wherein at least one attribute comprises link margins for data symbols, and wherein at least one processor is configured to select order based on delink margins for data symbols, starting with the largest symbol. Link margin. The apparatus of claim 1, wherein at least one processor is configured to decompose a channel response matrix to obtain a superior triangular matrix for the selected order, to compute distances for a series of hypotheses for each combase data symbol in the upper triangular matrix. and to determine candidate hypotheses for data symbols based on computed distances. The apparatus of claim 1, wherein at least one processor is configured to identify at least one hypothetical symbol for data symbol and to determine candidate hypotheses for data symbols based on at least one hypothetical symbol for each data symbol. The apparatus of claim 1, wherein at least one processor is configured to obtain at least one hard decision for at least one data symbol and to determine candidate hypotheses for the data symbols with at least one hard decision. The apparatus of claim 1, wherein the data symbols comprise first and second data symbols, and wherein at least one processor is configured to compute a first set of at least one distance to at least one hypothesis for the first data symbol to obtain a first hard decision for the first data symbol based on the first set of at least one distance, and for computing a second set of at least one distance for at least one hypothesis for the second data symbol with the first hard decision. The apparatus of claim 8, wherein the data symbols further comprise a third data symbol, and wherein at least one processor is configured to obtain a second hard decision for the second data symbol based on the second set at least one distance, and for compute a third set of at least one distance to at least one hypothesis for the third data symbol with the first and second hard decisions. A method comprising: selecting an order of detection of data symbols sent in a MIMO transmission based on at least one attribute of the data symbols; and perform spherical detection for data symbols in the selected order. The method of claim 10, wherein at least one attribute comprises error probabilities, demodulation schemes, link margins or a combination thereof for data symbols. The method of claim 10, wherein performing spherical detection comprises comprising a channel response matrix so as to obtain an upper triangular matrix for the selected order, computing the distances for a series of hypotheses for each data symbol based on the upper triangular matrix; and determine candidate hypotheses for data symbols based on computed distances. The method of claim 10, wherein the execution of spherical detection comprises obtaining at least one hard decision for at least one data symbol, and determining candidate hypotheses for data symbols with at least one hard decision. The method of claim 10, which further comprises: identifying at least one hypothetical symbol for each data symbol; and determine candidate hypotheses for data symbols based on at least one symbol presented as hypothesis for each data symbol. Equipment comprising: devices for selecting an order of detection of data symbols sent in a MIMO transmission based on at least one attribute of the data symbols; devices to perform spherical detection for data symbols in the order selected. The apparatus of claim 15, wherein at least one attribute comprises error probabilities, modulation schemes, link margins or a combination thereof for the data symbols. The apparatus of claim 15, wherein the device for performing spherical detection comprising devices for decomposing a channel response matrix to obtain an upper triangular matrix for the selected order, devices for computing distances for a series of hypotheses for each data symbol with based on the upper triangular matrix, and devices for determining candidate hypotheses for data symbols based on computed distances. The apparatus of claim 15, wherein the device for performing spherical detection comprises devices for obtaining at least one hard decision for at least one data symbol, and device for determining candidate hypotheses for the data symbols with at least one hard decision. 19. Processor readable medium for storing actionable instructions for: selecting an order of detection of finger symbols sent in a MIMO transmission based on at least one attribute of the data symbols; and perform spherical detection for the finger symbols in the selected order. Equipment comprising: at least one processor configured to perform spherical detection on received symbols to detect the generated data symbols with at least two modulation schemes; a memory coupled to at least one processor. The apparatus of claim 20, wherein at least one processor is configured to determine a data symbol detection order based on at least two modulation schemes, and to perform spherical detection for the data symbols in the given order. The apparatus of claim 20, wherein at least one processor is configured to determine a specific number of assumptions to be evaluated for the data symbol based on the modulation scheme used for the data symbol. The apparatus of claim 20, wherein at least one processor is configured to identify a subset of all hypotheses for each symbol data based on the modulation scheme used for the data symbol and to evaluate the hypothesis subset for each data symbol. The apparatus of claim 20, wherein at least one processor is configured to hold a specific number of assumptions for each data symbol based on the modulation scheme used for the data symbol. The apparatus of claim 20, wherein at least one processor is configured to hold a specific number of assumptions for each data symbol with sufficient link margin to the data symbol. A method comprising: obtaining received symbols for a MIMO transmission; and performing spherical detection on received symbols so as to detect data symbols generated with at least two modulation schemes. 27. Equipment comprising: devices for obtaining received symbols for a MIMO transmission; devices for spherical detection of our received symbols to detect the generated data symbols with at least two modulation schemes. The apparatus of claim 27 further comprising: devices for determining an order of detecting data symbols based on at least two modulation schemes, and wherein spherical detection for data symbols is performed in the given order. The apparatus of claim 27, further comprising: devices for determining a specific number of hypotheses to be evaluated for each data symbol based on the modulation scheme used for the data symbol. 30. Computer readable medium for storing actionable instructions for: obtaining symbols received for a MIMO transmission; and performing spherical detection on received symbols so as to detect data symbols generated with at least two modulation schemes. 31. Equipment comprising: at least one processor configured to perform spherical detection for multiple data streams, to obtain channel state information for various data streams, and to select rates for various data streams based on channel state information; a memory coupled to at least one processor. The apparatus of claim 31, wherein the channel state information comprises a channel response matrix, and wherein at least one processor is configured to decompose the channel response matrix so as to obtain an upper triangular matrix used for spherical detection, to estimate the signal qualities of the various data streams based on the upper triangular matrix, and to select the rates for the various data streams based on the channel qualities. The apparatus of claim 32, wherein at least one processor is configured to perform spherical detection for the various data streams in a sequential order, one data stream at a time, and to estimate the signal quality of each data stream assuming that interference from already detected data streams is The apparatus of claim 31, wherein at least one processor is configured to estimate the signal quality of the various data streams based on channel state information and successive interference cancellation. The apparatus of claim 31, wherein at least one processor is configured to perform spherical detection for the various data streams in a sequential order, starting with a first data stream, and to select a rate for the first data stream to obtain a rate. target symbol errors or better for the first data stream. The apparatus of claim 31, wherein at least one processor is configured to perform spherical detection for the various data streams in a sequential order starting with a first data stream to estimate the signal quality of the first data stream based on the state information of the data stream. channel, to select an initial rate for the first data stream based on the signal quality of the first data stream, and to select a revised rate with a lower order modulation scheme for the first data stream if the initial rate results in a rate overrun. error target symbol. The apparatus of claim 31, wherein at least one processor is configured to estimate the signal quality of each of the various data streams based on channel state information and independently select a rate for each data stream based on the quality of the data. data flow signal. The apparatus of claim 31, wherein at least one processor is configured to select rates for various data streams from a set of rate combinations. The apparatus of claim 31, wherein at least one processor is configured to cause rate indications for the various data streams to be sent to a transmitter. 40. A method comprising: performing spherical detection for multiple data streams, obtaining channel state information for the various data streams; and select rates for the various data streams based on channel state information. The method of claim 40, wherein performing spherical detection for the various data streams comprises performing spherical detection for the various data streams in a sequential order, a one-time data stream, and wherein selecting the rates for the various streams. Data comprises estimating the signal quality of each data stream based on the channel state information and assuming interference from already detected data streams is canceled. The method of claim 40, wherein performing spherical detection for the various data streams comprises performing spherical detection for the various data streams in a sequential order starting with a first data stream, and wherein selecting the rates for the various data streams. comprises selecting a rate for the first data stream to obtain a target symbol rate or better for the first data stream. 43. Equipment comprising: devices for performing spherical detection for various data streams; devices for obtaining channel state information for the various data streams; devices to select rates for the various data streams based on channel status information. The apparatus of claim 43, wherein the device for performing spherical detection for the various data streams comprises devices for performing spherical detection for the various data streams in a sequential order, one data stream at a time, and wherein the device for selecting the rates. for the various data streams comprises a device for estimating the signal quality of each data stream based on channel state information and assuming interference from already detected data streams is canceled. The apparatus of claim 43, wherein the device for performing spherical detection for the various data streams comprises a device for performing spherical detection for the various data streams in a sequential order starting with a first data stream, and wherein the device for selecting the data streams. Rates for various data streams comprises devices for selecting a rate for the first data stream to obtain a target symbol error rate or better for the first data stream. 46. Processor readable medium for storing actionable instructions to: perform spherical detection for multiple data streams, obtain channel state information for the various data streams; and select rates for the various data streams based on channel state information.