HK1080661B - Rate control for multi-channel communication systems - Google Patents

Description

Rate control for multi-channel communication systems

FIELD

The present invention relates generally to data communications, and more specifically to techniques for controlling data transmission rates for multi-channel communication systems.

Background

OrthogonalFrequency Division Multiplexing (OFDM) communication systems effectively divide the overall system bandwidth into multiple (N)_F) A sub-bandwidth, which may also be referred to as a frequency sub-channel or frequency bin (bin). Each frequency bin is associated with a different subcarrier (or tone) upon which data may be modulated. For an OFDM system, the data to be transmitted (i.e., the information bits) is first encoded using a particular coding scheme to produce coded bits, which are further grouped into multi-bit symbols and then mapped to modulation symbols. Each modulation symbol corresponds to a point in a signal constellation (constellation) defined by a particular modulation scheme (e.g., M-PSK or M-QAM) for data transmission. In each time interval according to the bandwidth of each frequency subchannel, a modulation symbol may be in N_FIs transmitted on each of the frequency subchannels. OFDM may be used to combat inter-symbol interference (ISI) caused by frequency selective fading, which is characterized by different amounts of bandwidth fading over the system bandwidth.

A multiple-input-multiple-output (MIMO) communication system using multiple (N)_T) Transmitting antenna and a plurality of (N)_R) The receive antennas are used for data transmission. From N_TA transmitting antenna and N_RThe MIMO channel formed by the receiving antennas can be decomposed into N_SA separate channel of which N_S≤min{N_T，N_R}。N_SEach of the individual channels may also be referred to as a spatial subchannel of the MIMO channel and corresponds to a dimension. MIMO systems may provide improved performance (e.g., increased transmission capacity) if the multiple transmit and receive antennas being used create additional dimensionalities.

For a MIMO system using OFDM (i.e., a MIMO-OFDM system), there is N on each spatial subchannel_FOne frequency subchannel is used for data transmission. The frequency subchannels of each spatial subchannel may be referred to as transmission channels, N_F·N_SA transport channel can thus be used at N_FA transmitting antenna and N_SAnd data transmission is carried out among the receiving antennas.

For MIMO-OFDM systems, N for each spatial subchannel_FThe frequency subchannels may experience different channel conditions (e.g., different fading and multipath effects) and may achieve different signal-to-noise-and-interference ratios (SNRs). Each transmitted modulation symbol is affected by the channel response of the transmission channel used to transmit the symbol. Depending on the multipath profile (profile) of the communication channel between the transmitter and the receiver, the frequency response may vary widely in the system bandwidth for each spatial subchannel, and may further vary widely between spatial subchannels.

For multipath channels with a frequency response that is not flat (flat), the information rate that can be reliably transmitted on each transport channel (i.e., the number of information bits in each modulation symbol) differs from transport channel to transport channel. If the modulation symbols for a particular data packet are transmitted on multiple transmission channels and if the responses of these transmission channels vary widely, these modulation symbols may be received with a wide range of SNRs. The SNR may then vary accordingly throughout the received packet, which may make it difficult to determine the correct rate for the data packet.

Since different receivers may experience different (and possibly widely varying) channel conditions, it is not practical to use the same transmit power and/or data rate for all receivers. Fixing these transmission parameters results in wasted transmit power, sub-optimal data rates for some receivers and unreliable communication for other receivers, which can result in an undesirable reduction in system capacity. Furthermore, the channel conditions may vary over time. As a result, the data rate supportable by the transmission channel may also vary over time. The different transmission capabilities of the communication channels of the different receivers, coupled with the multipath and time-varying nature of these communication channels, make it challenging to efficiently transmit data in a MIMO-OFDM system.

Accordingly, there is a need in the art for techniques to control the rate of data transmission in a multi-channel communication system, such as a MIMO-OFDM system.

Summary of The Invention

Techniques are provided herein for controlling the rate of data transmission in a multi-channel communication system having multiple transmission channels. In an aspect, the rate for each data stream is determined based on a metric related to the data stream. The metric may be derived based on an equivalent system for modeling a set of transmission channels to be used for the data stream. The equivalent system is defined as having an AWGN channel (i.e., flat frequency response) and an average spectral efficiency S equal to the group of transmission channels_avgSpectral efficiency S of_equiv(i.e., the equivalent system has a total capacity equal to the total capacity of the transport channel group).

A particular embodiment provides a method for determining a set of rates for a set of data streams to be transmitted over a wireless communication channel in a multi-channel communication system (e.g., a MIMO-OFDM system). In the method, a set of transport channels to be used for each data stream is initially identified.

An equivalent system for each group of transmission channels is then defined based on one or more estimated characteristics of the transmission channels in the group. In one embodiment, the equivalent system for each transport channel group may be defined by: (1) obtaining an estimate of the SNR for each transmission channel, (2) estimating the spectral efficiency of each transmission channel based on the estimated SNR and a spectral efficiency function f (x), and (3) determining the average spectral efficiency of the transmission channels in the group based on the estimated spectral efficiencies of the individual transmission channels. An equivalent system is defined as having an AWGN channel and a spectral efficiency equal to the average spectral efficiency of the group of transmission channels.

The metric for each transport channel group is then derived based on the associated equivalent system. In one embodiment, the metric is set to the SNR required for the equivalent system to support the average spectral efficiency. The SNR is called the equivalent SNR and may be based on the inverse function f^-1(x) And is determined.

The rate for each data stream is then determined based on metrics associated with the data stream. This may be accomplished by estimating one or more available rates. For each estimated rate, the required SNR for the data rate supported by the communication system is determined, and if the required SNR is less than or equal to the metric, the rate is considered to be supported by the communication system.

Various aspects and embodiments of the invention are described in further detail below. The present invention also provides methods, receiver units, transmitter units, receiver systems, transmitter systems, and other apparatuses and elements that implement various aspects, embodiments, and features of the invention, as described in further detail below.

Brief Description of Drawings

The features, nature, and advantages of the present invention will become more apparent from the detailed description set forth below when taken in conjunction with the drawings in which like reference characters identify correspondingly throughout and wherein:

FIG. 1A is a block diagram of a model of a multi-channel communication system;

FIG. 1B is a graphical illustration of rate selection for a multi-channel communication system with multi-path channels based on an equivalent system;

FIG. 2 is a flow diagram of one embodiment of a process for determining the maximum data rate supported by a SISO-OFDM system based on an equivalent system;

fig. 3 is a diagram illustrating the spectral efficiency of a SISO-OFDM system with multipath channels.

FIG. 4A is a graph showing the required SNR versus data rate for a system supporting a discrete set of data rates;

FIG. 4B illustrates determining the number of back-offs (back-off) to use when estimating whether a particular data rate is supported;

fig. 5A is a diagram illustrating spectral efficiency of spatial subchannels in a MIMO system with a multipath channel.

Fig. 5B is a graph illustrating the spectral efficiency of an equivalent SISO system used to model the MIMO-OFDM system shown in fig. 5A.

FIG. 6 is a flow diagram of one embodiment of a process for controlling one or more independently processed data streams in a multi-channel system;

FIG. 7 is a block diagram of one embodiment of a transmitter system and a receiver system in a multi-channel system;

fig. 8 is a block diagram of a transmitter unit in a transmitter system.

Fig. 9 and 10 are block diagrams of two embodiments of a receiver processor in a receiver system.

Detailed Description

An Orthogonal Frequency Division Multiplexing (OFDM) communication system effectively partitions the bandwidth of the overall system into a plurality of N_F) A sub-bandwidth, which may also be referred to as a frequency sub-channel or frequency bin (bin). Each frequency subchannel is associated with a different subcarrier (or tone) upon which data may be modulated.

A multiple-input-multiple-output (MIMO) communication system using multiple (N)_T) Transmitting antenna and a plurality of (N)_R) The receiving antennas are used for data transmission and are marked as (N)_T，N_R) Provided is a system. From N_TA transmitting antenna and N_RThe MIMO channel formed by the receiving antennas can be decomposed into N_SA separate channel of which N_S≤min{N_T，N_R}。N_SEach of the individual channels may also be referred to as a spatial subchannel of the MIMO channel. The number of spatial subchannels is determined by the number of eigenmodes (eigenmodes) of the MIMO system, which in turn depends on a channel response matrix,H(k) which describes N_TA transmitting antenna and N_RThe response between the receiving antennas. For the sake of brevity, the following description is providedIn the above, the channel response matrixH(k) Is assumed to be full rank and the number of spatial subchannels is given as NS-N_T≤N_R。

The rate control techniques described herein may be used in a variety of multi-channel communication systems having multiple transport channels that may be used for data transmission. These multi-channel systems include MIMO systems, OFDM systems, MIMO-OFDM systems, and the like. The transmission channel may be (1) a spatial subchannel in a MIMO system, (2) a frequency subchannel in an OFDM system, or (3) a frequency subchannel of a spatial subchannel in a MIMO-OFDM system.

Fig. 1A is a block diagram of a model of a multi-channel communication system 100. At the transmitter 110, traffic data is provided from a data source 112 to a Transmit (TX) data processor 114. TX data processor 114 may demultiplex (demultiplex) the traffic data into N_DA data stream, N_DIs any integer greater than or equal to 1. Each data stream may be independently processed and then transmitted over a different set of transmission channels. For each data stream, the TX data processor encodes data in a particular coding scheme, interleaves the coded data in a particular interleaving scheme, and modulates the interleaved data in a particular modulation scheme. The modulation (i.e., symbol mapping) is accomplished as follows: the coded and interleaved bits are grouped into multi-bit symbols and each multi-bit symbol is mapped to a point in the signal constellation corresponding to a selected modulation scheme (e.g., QPSK, M-PSK, or M-QAM). Each mapped signal point corresponds to a modulation symbol.

In one embodiment, for each data stream, the data rate is determined by a data rate control, the coding scheme is determined by a coding control, and the modulation scheme is determined by a modulation control. These controls are provided by a controller 130 based on feedback information received from receiver 150.

A pilot may also be transmitted to the receiver to assist it in performing a number of functions, such as channel estimation, acquisition, frequency and timing synchronization, coherent data demodulation, and so forth. In this case, the pilot data is provided to a TX data processor 114, which then processes and multiplexes the pilot data with the traffic data.

For OFDM, in a transmitter (TMTR)116, modulated data (i.e., modulation symbols) to be transmitted from each transmit antenna is converted to the time domain by an Inverse Fast Fourier Transform (IFFT) unit to provide OFDM symbols. Each OFDM symbol is N to be transmitted on one transmit antenna in one transmission symbol period_FN transmitted on frequency sub-channels_FA time domain representation of a vector of modulation symbols. With respect to a "time-coded" system of a single carrier, an OFDM system can efficiently transmit modulation symbols "in the frequency domain" through an IFFT that transmits the modulation symbols of traffic data in the time domain.

Transmitter 116 provides a stream of OFDM symbols for each transmit antenna used for data transmission. Each OFDM symbol stream is further processed (not shown in fig. 1A for simplicity) to generate a corresponding modulated signal. Each modulated signal is then transmitted over a wireless channel through a different transmit antenna to a receiver. The communication channel will distort the modulated signal due to the specific channel response and will have an additional variance N₀White Gaussian Noise (AWGN) to further degrade the modulated signal.

At receiver 150, the transmitted modulated signals are received by each receive antenna, and the received signals from all the receivers are provided to a receiver (RCVR) 160. In receiver 160, each received signal is conditioned and digitized to provide a corresponding sample stream. For each sample stream, a Fast Fourier Transformer (FFT) receives and transforms the samples to the frequency domain to provide a corresponding received symbol stream. The received symbol streams are then provided to a Receive (RX) data processor 162.

An RX data processor processes the stream of received symbols to provide decoded data for the transmitted data stream. The processing by RX data processor 162 may include spatial or space-time processing, demodulation (i.e., symbol demapping), deinterleaving, and decoding. RX data processor 162 may further provide each receivedThe status of the data packet. The channel estimator 164 processes the "detected" symbols from the demodulator/decoder 162 to provide an estimate of one or more characteristics of the communication channel, e.g., channel frequency response, channel noise variance N₀Signal-to-noise-and-interference ratio (SNR) of the detected symbol, and the like. However, it is within the scope of the present invention that the SNR may also be estimated based on the data symbols, or a combination of pilot and data symbols.

Rate selector 166 receives the channel estimates, as well as possibly other parameters, from channel estimator 164 and determines an "appropriate" rate for each data stream. The data indicates a set of parameter values ready for subsequent transmission of the data stream. For example, the rate may indicate (or may be associated with) a particular data rate, a particular coding scheme and/or code rate, a particular modulation scheme, and/or the like, that is to be prepared for the data stream.

Controller 170 receives the rate from rate selector 166 and the packet status from RX data processor 162 and provides appropriate feedback information to transmitter 110. The feedback information may include a rate, a channel estimate, some other information, or a combination thereof. The feedback information may be used to increase the efficiency of the system by adjusting the processing at the transmitter so that data may be transmitted at the best known setting of power and rate supported by the communication channel. The feedback information is then sent back to the transmitter 110 and used to adjust the processing of the data transmitted to the receiver 150. For example, transmitter 110 may adjust a data rate, a coding scheme, a modulation scheme, or any combination thereof (based on the feedback information) for each data stream to be transmitted to receiver 150.

In the embodiment shown in fig. 1A, the rate selection is made by receiver 150 and the rate selected for each data stream is provided to transmitter 110. In other embodiments, rate selection is performed by the transmitter based on feedback information provided by the receiver, or may be performed by both the transmitter and the receiver in combination.

In a single carrier communication system, transmitted symbols may be received all at a receiver with a similar SNR. The relationship between data packets of "constant SNR" and the error Probability (PE) of the packets is well known in the art. Approximately, the maximum rate supported by a single carrier system with a particular SNR can be estimated as the maximum data rate supported by an AWGN channel with the same SNR. The main characteristic of an AWGN channel is that its frequency response is flat or constant across the system bandwidth.

However, in a multi-channel communication system, the modulation symbols that make up a data packet may be transmitted over multiple frequency subchannels and/or multiple spatial subchannels. In general, the communication channel between the transmitter and receiver is not flat, but rather is dispersive or frequency selective, with different amounts of attenuation across different sub-bands of the system bandwidth. Furthermore, for a MIMO channel, the frequency response of each spatial subchannel may be different from one another. Thus, the SNR may vary throughout the packet depending on the characteristics of the transmission channel used to transmit the packet. This problem of "varying SNR" of packets is exacerbated for wider system bandwidths and multipath channels. For a multipath channel, the data rate for each data stream may be selected to account for multipath or frequency selective characteristics of the communication channel.

The main challenge for multi-channel communication systems is next to determine the maximum data rate that can be used for each data stream while achieving a particular level of performance, which is quantified by a particular Packet Error Rate (PER), Frame Error Rate (FER), block error rate (BLER), Bit Error Rate (BER), or any other criteria that can be used to quantify performance. For example, a desired level of performance may be achieved by maintaining the PER within a small range (e.g., P) around a particular nominal value_e1%) was performed.

Techniques are provided herein for controlling data transmission in a multi-channel communication system having multi-path channels. In an aspect, the rate for each data stream is determined based on a metric associated with the data stream. The metric may be derived based on an equivalent system that models the set of transmission channels for the data stream, as will be described in detail below.

Fig. 1B is an illustration of rate selection for a multi-channel communication system with a multi-path channel based on an equivalent system. For a given channel response h (k) and noise variance N₀Defining multipath channels, theoretical multichannel systems being able to support S using a modulation scheme M_avgWhere M may be different for different frequency subchannels. As used herein, spectral efficiency refers to the general concept of "capacity in each dimension," where a dimension may be frequency and/or space. Spectral efficiency is typically measured in bits per second per hertz (bps/Hz). As used herein, a theoretical system is one without any penalty, while an actual system is one with (1) implementation penalty, e.g., due to hardware deficiency, and (2) coding penalty, since the coding actually implemented does not operate at that capacity. The S is_avgAnd providing channel conditions h (k) and N₀The average effectiveness of the theoretical system of (a). The average spectral efficiency S_avgMay be determined based on a spectral efficiency function f (x), where x represents a set of input parameters for the function f (·), as will be described below.

Equivalent systems with AWGN channels can operate at SNR_equivSNR of (S) supports spectral efficiency S_avg. This equivalent system is also a theoretical system. Equivalent SNR, SNR_equivModulation scheme M may be used and is based on function g (x) f^-1(x) Derived spectral efficiency S_avgWherein f is^-1(x) Is the inverse function of f (x).

Implementing a multi-channel system with an AWGN channel can use a modulation scheme M and a coding scheme C with P_ePER and SNR of_equivThe SNR of (b) supports the data rate R. The data rate R is normalized to bits/sec/hz, which is the same unit used for spectral efficiency. The required SNR may be determined based on computer simulations, empirical measurements, or some other means, and may be stored in a table. The required SNR as a function of data rate is based onThe particular modulation scheme M and coding scheme C used are selected. If the SNR required for the data rate is less than the equivalent SNR, then the data rate is considered to be supported by the implemented multi-channel system. As the data rate R increases, the SNR required by the implemented system also increases, since the equivalent SNR is determined by the channel conditions h (k) and N₀By definition, the equivalent SNR is substantially constant (except for deviations due to modulation scheme M). Thus, the maximum rate supported by an implemented multi-channel system with multi-path channels is limited by the channel conditions.

For clarity of explanation, rate control will first be described for a single-input single-output (SISO) system, then extended to a single-input multiple-output (SIMO) system, and finally extended to a MIMO system. In the following description, SISO, SIMO and MIMO systems all use OFDM.

SISO system

For SISO-OFDM system, there is only one spatial subchannel and the channel response is defined as { h (k) }, where k is 0, 1_F-1). For a channel response of { h (k) } and a noise variance of N₀For each frequency subchannel k, these parameters may be mapped to an snr (k). If total transmission power P of SISO-OFDM system_totalIs fixed and allocates transmit power to N_FThe frequency subchannels are uniform and fixed, and the SNR for each frequency subchannel k may be expressed as:

the spectral efficiency of each frequency subchannel k having snr (k) may be estimated based on a function f (x), which may be a constrained or unconstrained spectral efficiency function. The absolute or unconstrained frequency efficiency of a system is generally given as the theoretical maximum data rate that can be reliably transmitted over a channel with a given channel response and noise variance. The constrained spectral efficiency of the system is further based on the particular modulation scheme or signal constellation used for the data transmission. The constrained spectral efficiency (due to the fact that the modulation symbols are limited to a particular point on the signal constellation) is lower than the absolute spectral efficiency (which is not limited by any signal constellation).

In one embodiment, the function f (x) may be based on a constrained spectral efficiency function f_const(k) And is defined, it can be expressed as:

wherein M is_kIn relation to modulation schemes M (k), i.e. modulation schemes M (k) correspondThe constellation of dimensions (e.g.,QAM of dimensions) among others in the constellationEach of the points may be defined by M_kBit identification;

α_iand alpha_jIs thatPoints in a dimensional constellation;

β is a complex (complex) Gaussian random variable with a mean value of 0 and variance bits of 1/SNR (k); and

e [. cndot. ] is the expected value operation, which is for the variable β in equation (2).

Equation 2 shows that each frequency subchannel may use a different modulation scheme m (k). For simplicity, one modulation scheme M may be used for all N_FOne frequency subchannel for data rate R (i.e., M (k) ═ M, for all k).

Constrained spectral efficiency function f shown in equation (2)_const(k) There is no closed form solution. Thus, the function may numerically derive a variety of modulation schemes and SNR values, and the results may be stored in one or more tables. Thereafter, the function f_const(k) Can be estimated by accessing the appropriate table with the particular modulation scheme and SNR.

In another embodiment, the function f (x) is based on a Shannon (Shannon) (or theoretical) spectral efficiency function f_unconst(k) But is defined, it can be expressed as:

f_unconst(k)＝log₂[1+SNR(k)] (3)

as shown in equation (3), shannon spectral efficiency is not limited by any given modulation scheme (i.e., m (k) is not a parameter in equation (3)).

The spectral efficiency function provides the spectral efficiency of the system based on a set of input parameters. These spectral efficiency functions are related to the channel capacity function that provides the (constrained or unconstrained) capacity of the channel. Spectral efficiency (typically in bps/Hz) is related to capacity (typically in bps) and can be considered equal to normalized capacity.

The particular choice of function for f (x) may depend on different factors. For a typical system using one or more particular modulation schemes, it has been found that a constrained spectral efficiency function f is used_const(k) The function f (x) can yield an accurate estimate of the maximum data rate supported by a SISO-OFDM system with multipath channels.

In a typical communication system, a discrete set of data rates may be defined, R ═ R (R), R ═ 1, 2 … P, and only these data rates may be used. Each data rate R (R) in the set R may be associated with a particular modulation scheme or signal constellation m (R) and a particular coding rate c (R). Each data rate may also require an SNR_req(r) or better SNR to achieve P as desired_ePER of (4). The SNR_req(r) is determined for a practical SISO-OFDM system with AWGN channel.

Each data rate r (r) may thus be associated with a set of parameters characterizing it. These parameters may include the modulation scheme M (r), the code rate C (r), and the required SNR_req(r) represented by the following formula:

where r is an index of the data rate, i.e. r ═ r1. 2 … P, and P is the total number of data rates that can be used. Equation (4) indicates that the data rate R (r) can be transmitted using the modulation scheme M (r) and the coding rate C (r) and further requires SNR_req(r) to achieve the desired P_ePER of (4).

Fig. 2 is a flow diagram of one embodiment of a process 200 for determining a maximum data rate supported by a SISO-OFDM system based on an equivalent system. For this embodiment, the constrained spectral efficiency function shown in equation (2) is used as f (x) to determine the average spectral efficiency of the transmission channel for data transmission. Since each data rate r (r) may be associated with a different modulation scheme m (r), and since the constrained spectral efficiency function is dependent on m (r), the average spectral efficiency of the transmission channel may be different for different data rates. The equivalent system is based on the average spectral efficiency and thus determines each data rate in fig. 2.

Initially, the P data rates supported by the SISO-OFDM system may be ordered in the following order R (1) < R (2) < … < R (P). The highest available rate r (p) is then selected (e.g., by setting the variable r equal to the index of the highest data rate, i.e., r ═ p) (step 212). Parameter values and (1) transmission channels for data transmission, e.g. channel response h (k) and noise variance N₀And (2) the selected data rate r (r), e.g., modulation scheme m (r), is correlated and then determined (step 214). Each data rate may be associated with one or more modulation schemes, depending on the design of the SISO-OFDM system. For simplicity, it will be assumed in the following that only one modulation scheme is associated with each data rate.

Average spectral efficiency S of a transmission channel_avgAnd then determined (step 216). This may be achieved by first determining the snr (k) of each transmission channel, as shown in equation (1) above. The spectral efficiency of each transmission channel is then estimated for snr (k) and modulation scheme m (r) using the constrained spectral efficiency function, as shown in equation (2). N is a radical of_FThe spectral efficiencies of the sum frequency subchannels are then averaged to obtain an average spectral efficiency S_avgThe following formula:

fig. 3 is a graphical illustration of the spectral efficiency of a SISO-OFDM system with multipath channels. For multipath channels with variable SNR across the system bandwidth, the SISO-OFDM system has different spectral efficiencies for different frequency subchannels, as illustrated by curve 310. All N for data transmission_FThe spectral efficiency of the individual frequency subchannels may be averaged to obtain an average spectral efficiency S_avgWhich is shown by curve 312. Average spectral efficiency S if the communication channel is an AWGN channel rather than a multipath channel_avgCan be regarded as N in SISO-OFDM system_FSpectral efficiency of each of the frequency subchannels. A constrained or unconstrained spectral efficiency function may thus be used to map the multipath channel to an equivalent AWGN channel.

Referring back to fig. 2, the metric Ψ is then determined based on the equivalent system (step 218). The equivalence system 218 is defined as having an AWGN channel and an average spectral efficiency S_equivThe average spectral efficiency S_equivEqual to the average spectral efficiency (i.e., S) of a SISO-OFDM system with multipath channels_equiv＝S_avg). Etc. ofEffective system support S_equivThe required SNR for the data rate of can be based on the SNR used to derive S_avgIs determined, in this case the constrained spectral efficiency function. The metric Ψ may then be set equal to the equivalent SNR as follows:

Ψ＝g(x)＝f^-1(x) (6)

wherein f is^-1(x) Denotes the inverse function of f (x). Both metric Ψ and the equivalent SNR indicate N_F"goodness" (goodness) of each frequency subchannel.

The constrained spectral efficiency function f (x) has two inputs, snr (k) and m (r), and maps them to a spectral efficiency value. Inverse constrained spectral efficiency function f^-1(x) Having two inputs, S_avgAnd m (r) and maps them to an SNR value. Function g (S)_avgM (r) thus determines that the equivalent system support is equal to the average spectral efficiency S using the constellation M (r)_avgSNR required for spectral efficiency. The metric Ψ may thus be determined once for each modulation scheme (i.e., each signal constellation). The functions g (x) for the different modulation schemes may also be determined and stored in a table.

By the actual SISO-OFDM system with the desired P_ePER of (b) transmits the SNR, required for the selected data rate r (r)_req(r) is then determined (step 220). The required SNR is a function of the modulation scheme m (r) and the code rate c (r) associated with the selected data rate r (r). The required SNR for each possible data rate may be determined by computer simulation, empirical measurements, or by some other means and stored in a table for later use.

It is next determined whether the selected data rate r (r) is supported by the SISO-OFDM system (step 222). This may be achieved by comparing the metric Ψ to the SNR required to determine the data rate to be used for the selection. If the metric Ψ is greater than or equal to the required SNR (i.e., Ψ ≧ SNR)_req(r)) indicating that the SNR achieved by the SISO-OFDM system for the multipath channel is sufficientFor desired P, holding data rate R (r)_eThen the data rate is selected for use (step 226). Otherwise, the next lower available data rate is selected for estimation (e.g., by decreasing the variable r, or r-1) (step 224). The next lower data rate is then estimated by returning to step 214. Steps 214 through 222 may be repeated until (1) the maximum data rate supported is identified and provided in step 226, or (2) all available data rates have been estimated.

If a constrained spectral efficiency function is used, the metric Ψ is dependent on the channel conditions (e.g., h (k) and N₀) And a modulation scheme m (r). The required SNR is a monotonic function that increases as the data rate increases. The embodiment shown in fig. 2 estimates the available data rates from the maximum available data rate to the minimum available data rate, one at a time. A metric Ψ equal to or less than the highest data rate associated with the required SNR is selected for use.

The metric Ψ may be determined based on equations (2), (5), and (6). In equation (5), f (x) is summed to accumulate the spectral efficiency of the individual frequency subchannels to provide N_FSpectral efficiency of the individual frequency subchannels. Then by adding N_FDividing the spectral efficiency of the individual frequency subchannels by the number of frequency subchannels to obtain an average spectral efficiency S_avg. Function g (S)_avgM (r)) next determines to use the modulation scheme M (r) to average the spectral efficiency S_avgEqual spectral efficiency reliably transmits the equivalent SNR required by an equivalent system of data.

Equation (5) assumes that the same modulation scheme M (r) is used for all N in a SISO-OFDM system_FA frequency subchannel. This limitation may simplify processing at the transmitter and receiver in the system, but may sacrifice performance.

The metric Ψ may also be defined for the case where different modulation schemes are used for different frequency subchannels. The use of different modulation schemes and/or coding rates for different frequency subchannels is commonly referred to as "bit loading".

In fig. 2, an equivalent system is determined for each data rate to be estimated. This implementation covers a scheme where different data rates are associated with different modulation schemes. However, if different data rates are associated with the same modulation scheme, only an equivalent system for each different modulation scheme that may be used for the data rate to be estimated need be determined. This may simplify the calculation.

Further simplification, if the average spectral efficiency S of the frequency sub-channels_avgRelying only on snr (k) and not on the modulation scheme, which would occur if the unconstrained spectral efficiency function were used as f (x), the equivalent system would only need to be estimated once, rather than for each data rate. The equivalent SNR for an equivalent system may be determined once using the manner described above. Thereafter, the required SNR for each data rate (starting with the highest data rate) may be compared to the equivalent SNR.

In another embodiment, the metric Ψ is defined as the post-detection (SNR) used by the single carrier communication system for the multipath channel after equalization. The post-detection SNR represents the ratio of the total signal power and the noise plus interference at the receiver after equalization. The theoretical value of the post-detection SNR in an equalized single carrier system may represent the performance of a SISO-OFDM system and, thus, may be used to determine the maximum data rate supported in a SISO-OFDM system. Different types of equalizers may be used to process signals received in a single carrier system to compensate for distortion introduced in the received signal due to multipath channels. Such equalizers include, for example, minimum mean square error linear equalizers (MMSE-LE), Decision Feedback Equalizer (DFE), and the like.

The post-detection SNR of the MMSE-LE (of infinite length) can be expressed as:

wherein J_minIs given by

Wherein X (e)^jωT) Is the folded spectrum of the channel transfer function h (f).

The post-detection SNR of a DFE (of infinite length) can be expressed as:

the post-detection SNRs for MMSE-LE and DFE shown in equations (7) and (8), respectively, are representative of theoretical values. Post-detection SNRs for MMSE-LE and DFE are further elaborated by j.g. proakis, in chapters 10-2-2 and 10-3-2, entitled "digital communications" 3 rd edition, published by McGraw, Hill in 1995, both of which are incorporated by reference.

The post-detection SNR of the MMSE-LE and DFE can also be estimated at the receiver based on the received signal, as described in U.S. patent application Ser. Nos.09/826,481 and 09/956,449, both entitled "Method and Apparatus for adapting Channel State Information Wireless Communication System", filed on 23/2001 and 9/18/2001, respectively, and U.S. patent application Ser. No.09/854,235, entitled "Method and Apparatus for Processing Data in Multi-input Multi-MIMO-output (MIMO) Communication System adapting Channel State Information", filed on 11/2001. All assigned to the assignee of the present application and incorporated herein by reference.

Post-detection SNRs such as those described in the analytical representations shown in equations (7) and (8) are determined for the multipath channel and may be used as estimates of the metric Ψ (i.e., Ψ ≈ SNR)_mmse-1eOr Ψ ≈ SNR_dfe). Post-detection SNR (i.e., SNR) for equivalent AWGN channels_mmse-1eOr SNR_dfe) Can meet the required SNR_req(r) are compared to derive a set of parameter values, R (r), M (r), C (r), and P_eTo determine the data rate that can be used in a SISO-OFDM system with multipath channels.

An equivalent system for modeling the transmission channel of a user data stream may be defined as having an AWGN channel and a spectral efficiency equal to the average spectral efficiency of the transmission channel. An equivalent system may also be defined based on the post-detection SNR for a multipath channel by a single carrier communication system. Equivalent systems may also be defined using other forms and are within the scope of the present invention. It is within the scope of the invention that the metric Ψ may be defined based on other functions and/or in other manners.

The data rate selected for use in a SISO-OFDM system uses the metric Ψ to represent the predicted supportable by the multipath channel for the desired P_eThe data rate of the PER. For any data rate prediction scheme, there is inevitably a prediction error. To ensure that the desired PER can be achieved, the prediction error may be estimated and a back-off factor may be used in determining the data rate supported by the multipath channel. This backoff reduces the throughput of the system. Therefore, it is desirable to keep the back-off as small as possible while still achieving the desired PER. An accurate prediction scheme as described herein will minimize the backoff to be applied and thus maximizeThe system capacity is changed.

Fig. 4A shows a plot of required SNR versus data rate for a system supporting a discrete set of data rates. In fig. 4A, the discrete data rate is denoted on the horizontal axis as r (r), where r ═ 1, 2,. P_oEach data rate R (r) is implemented as P with a system having an AWGN channel_eAre correlated with different SNRs required by the PER. The required SNR is denoted SNR on the vertical axis_req(r) of (A). Discrete operating points at (r), SNR (r)), where r ═ 1, 2,. P correspond to the minimum SNR required to support the corresponding data rate, represented by solid dots 412. The spectral efficiency function of the system is represented by curve 410 (thick solid line).

For a given multipath channel, the average spectral efficiency S_avgMay be determined as shown in equation (5) and the metric Ψ for the average spectral efficiency may be determined as shown in equation (6). As shown, Ψ and S_avgWhich may be represented by point 414 in fig. 4A, is labeled "x". If the point is the shaded portion above curve 410, then the sum of Ψ and S is selected_avgThe associated data rates are considered to be supported by the system.

Since the selected data is based on theoretical values, it is necessary to back-off the selected data. For example, coding losses and implementation losses may result in a higher SNR being required to achieve a desired SNR. Implementation losses due to imperfections in the pre-decoding steps of the receiver cause an increase in SNR, while losses due to imperfections in the decoder and transmitter are generally negligible. The coding loss versus capacity value may be estimated and calculated along with the backoff. The back-off amount for considering the coding loss may be determined as described below

Fig. 4B illustrates a determination of the amount of backoff to use in estimating whether to support a particular data rate. As described above, set { SNR_req(r), wherein r 1, 2,. P indicates the desired P in a practical system_ePER ofAnd obtaining the SNR. The ideal SNR can be determined for each data rate based on a spectral efficiency function (constrained or unconstrained) and displayed on the vertical axis. Set { SNR_cap(r) }, where r 1, 2,. P denotes the desired P in an ideal system (i.e. no penalty is realized) in order to obtain the desired P_ePER of the signal to noise ratio. Note that for all r, due to SNR_cap(r) is the SNR required for an ideal system and SNR_req(r) is the SNR required for the actual system, SNR_cap(r)＜SNR_req(r) of (A). The set { Δ SNR (r) }, where r ═ 1, 2,. P can be defined to represent the additional SNR needed by the actual system to account for losses in the actual system (mainly including coding losses).

Average spectral efficiency S determined in equation (5)_avgWill lie between two successive data rates, e.g., R (R) and R (R +1), which are normalized to bits/sec/hz. The corresponding back-offs in SNR at these two data rates are Δ SNR (r) and Δ SNR (r +1), respectively. In one embodiment, the amount of back-off used to measure Ψ may be determined by linear interpolation between Δ SNR (r) and Δ SNR (r +1), as follows:

backoff metric Ψ_BOCan be labeled as:

Ψ_BO＝Ψ-ΔΨ (10)

referring back to FIG. 2, the backoff metric Ψ_BO(instead of the metric Ψ) may be compared to the required SNR in step 222 to determine whether the selected data rate r (r) is supported by a SISO-OFDM system.

SIMO system

For SIMO systems, N_RThe plurality of receiving antennas are used to receive data transmitted from a single transmitting antenna. A single transmitting antenna and N_RThe channel response between the receive antennas is represented ash(k) Or { h_i(k) N, wherein i ═ 1, 2_RAnd k is 0, 1, · (N)_F-1), wherein h_i(k) Is the coupling (i.e., complex gain) between the transmit antenna and the ith receive antenna on the kth frequency subchannel.

(1，N_R) The spectral efficiency function of the SIMO system is the same as for the SISO system, except that the SNR received by the SIMO system is over all N_RThe sum of the received SNRs for the individual receive antennas. Therefore, the received SNR for the k-th frequency subchannel in the SIMO-OFDM system can be expressed as:

where the transmit power for each frequency subchannel is normalized to 1. For simplicity, equation (11) assumes that all N are present_RReceiving the same noise variance N on several receiving antennas₀. Equation (11) may be modified to calculate different noise variances N received by different receive antennas₀. In comparison, the received SNR for the k-th frequency subchannel in the SISO-OFDM system can be represented by equation (1). For SIMO-OFDM systems, the received SNR determined in equation (11) may be used for the spectral efficiency function f (x). In addition to calculating the change in SNR, rate control for the SIMO-OFDM system may be performed in a manner similar to that described above for the SISO-OFDM system.

MIMO system

For MIMO-OFDM systems, in N_TA transmitting antenna and N_RThe channel response between the receive antennas may use N_R×N_TChannel impulse response matrixHAnd (4) showing. Matrix arrayHThe element in (1) is a vector containing channel pulseh _i，jComposition, wherein i ═ 1, 2_RAnd j is 0, 1,. M_T，h _i，jRepresenting the coupling between the jth transmit antenna and the ith receive antenna. Each vectorh _i，jConsists of L taps (tap), which can be expressed as:

h _i，j＝[h_i，j(1)h_i，j(2)...h_i，j(L)]^T (12)

where L taps may be modeled as complex gaussian coefficients for a rayleigh (rayleigh) fading channel. For a given pair of (i, j) transmit and receive antennas, a signal transmitted from the jth transmit antenna may be received by the ith receive antenna over several propagation paths, and multipath components associated with these propagation paths are assumed to be uncorrelated, which may be expressed as:

where p and q represent two multipath components, h^*Is the complex conjugate of h, and δ_p-qIs a Delta-Dirac function that is only 1 when p ═ q and 0 at other times. Furthermore, it is assumed that the channel responses for different transmit-receive antenna pairs are uncorrelated, i.e.For different m, n, i, j, where h^HRepresenting the conjugate transpose of h.

Channel impulse response matrixH(n) is a time domain representation of the MIMO channel response. Corresponding channel frequency response matrixH(k) Can be passed through atH(n) obtained by performing a Fast Fourier Transform (FFT), which can be expressed as:

wherein k ═ O, 1, · (N)_F-1) and N_FIs more than or equal to L. In particular, N_FFFT of points may be performed by N_FA sequence of sample values toHThe elements given inh _i，jTo derive correspondingHOf (2) element(s)h _i，jN of (A)_FA sequence of coefficients.HEach element of (a) is thusHFFT of the corresponding element in (a).HEach of which isThe element is a compound having N_FA plurality of complex-valued vectors (i.e. ah _i，j＝[h_i，j(0)，h_i，j(1)...h_i，j(N_F-1)]^T) Which represents the frequency response of the transmission path for a particular (i, j) transmit-receive antenna pair. The matrix can be viewed as comprising N_FAn arrayH(k) Wherein k is 0, 1_F-1) each having N_R×N_TAnd (5) maintaining.

For a MIMO-OFDM system, data may be processed and transmitted using several processing schemes. Each processing scheme may specify (1) the manner in which the data is processed (i.e., encoded, interleaved, and modulated) prior to transmission, and (2) the transport channels used to transmit each independently processed data stream.

In an All Antenna Processing (AAP) scheme, data streams are transmitted on all transmit antennas and frequency subchannels. For this scheme, data to be transmitted may be encoded, interleaved, modulated, and then multiplexed into N_TOne symbol stream for N_TA transmitting antenna. For the AAP scheme, the encoded data packet may be interleaved in both the frequency domain and the spatial domain.

In a Per Antenna Processing (PAP) scheme, a data stream is transmitted on all frequency subchannels of each transmit antenna. For this scheme, data to be transmitted is first multiplexed into N_TUsing data streams for N_TA transmitting antenna. Each data stream is independently encoded, interleaved, modulated and at N_TIs transmitted on one of the transmit antennas. N is a radical of_TThe data rate and coding and modulation scheme for the individual data streams may be the same or different. For the PAP scheme, each data stream is interleaved only in the frequency domain.

Each independently processed data stream may include one or more encoded data packets or codewords. Each such codeword is generated at the transmitter by encoding a data packet based on a particular coding scheme and may be decoded at the receiver based on a complementary decoding scheme. Decoding for each code word may be accomplished by first recovering the transmitted modulation symbols for that code word. The processing scheme selected at the transmitter affects the processing scheme available at the receiver.

The model of the MIMO-OFDM system can be expressed as:

y(k)＝H(k)x(k)+nfor k ═ 0, 1. (N)_F-1)， (15)

Whereiny(k) Is N for the k-th frequency subchannel_RA vector of received symbols (i.e., a "received" vector for tones (tones) k), which may be represented asWhereiny _i(k) Input of tone k received by the i-th receiving antenna, where i ═ 1, 2,. N_R；

x(k) Is N of tone k_TA vector of modulation symbols (i.e., a vector that is "transmitted"), which may be represented asWherein x is_j(k) Is a modulation symbol for tone k transmitted from the jth transmit antenna, where j 1, 2_T；

H(k) A channel frequency response matrix for the MIMO channel that is tone k; and

nis a vector having an average value of0And covariance (covariance) matrixΛ _n＝N₀ IAdditive White Gaussian Noise (AWGN), wherein0Is a vector of zero values and is,Iis a feature matrix of 1 on the diagonal and 0 elsewhere, N₀Is the noise variance.

For simplicity, the impact of OFDM processing at both the transmitter and receiver, which can be ignored, is not shown in equation (5).

From N due to scattering in the propagation environment_TThe NT symbol streams transmitted by the transmit antennas may interfere with each other at the receiver. In particular, a given symbol stream transmitted from one transmit antenna may be differentiated by all NR receive antennasAmplitude and phase reception. Each received symbol stream may include N_TA component of each of the transmitted symbol streams. N is a radical of_RThe received symbol stream may include all N in total_TA stream of transmitted symbols. However, these N_TThe data stream is dispersed in N_RAmong the received symbol streams.

At the receiver, N may be processed using different processing techniques_RA received symbol stream to detect N_TA stream of transmitted symbols. These receiver processing techniques can be divided into two main categories:

spatial and space-time receiver processing techniques (which may also be referred to as equalization techniques) and

receiver processing techniques (also referred to as "successive interference cancellation" (SIC) processing techniques) "successive nulling/equalization and interference cancellation".

Spatial and space-time receiver processing techniques may provide better performance for the AAP scheme, while SIC processing techniques may provide better performance for the PAP scheme. These receiver processing techniques are described in detail below.

For the sake of brevity, the following terms will be used herein:

"transmitted" symbol stream — the modulated symbol stream transmitted from the transmit antenna;

"received" symbol stream-the input to the spatial and space-time processors (if used, at the first stage of the SIC receiver, as shown in fig. 10);

"modified" symbol stream-the input to a spatial or space-time processor in a subsequent stage of the SIC receiver.

"detected" symbol stream-output from spatial or space-time processor (in stage I of SIC receiver, N can be detected_T-l +1 symbol streams); and

"recovered" symbol stream-a symbol stream recovered at the receiver to obtain a decoded data stream (only one symbol stream detected in each stage of the SIC receiver).

Spatial and space-time receiver processing techniques attempt to separate the transmitted symbol streams at the receiver. Each transmitted symbol stream may be "detected" (1) by the following steps based on an estimate of the channel response that will be at N_RDifferent components of a transmitted symbol stream in the received symbol streams are combined, and (2) interference due to other transmitted symbol streams is removed (or cancelled). Each receiver processing technique attempts to one of (1) decorrelate the individual transmitted symbol streams so that no interference is from the other transmitted symbol streams, and (2) maximize the SNR of each detected symbol stream in the presence of noise from the other transmitted symbol streams. Each detected symbol stream is then further processed (e.g., demodulated, deinterleaved, and decoded) to obtain a corresponding data stream.

For simplicity, assume a linear Zero Forcing (ZF) equalizer performs spatial processing by projecting (projecting) the received symbol stream into an interference-free word space to obtain the transmitted symbol stream. Linear ZF equalizer with responseW _ZF(k) It can be expressed as:

W _ZF(k)＝H(k)(H ^H(k)H(k))^-1 (16)

the stream of detected symbols is then decoded to obtain a stream of symbols,for the transmitted symbol streamxThe estimate may be expressed as:

as shown on the right side of equation (17), the detected symbol stream,comprising a transmitted symbol streamxPlus filtered noiseW _ZF ^H(k)nWhich is usually associated with a covariance matrixAnd (4) correlating. The correlation occurs between the same frequency subchannels of different transmit antennas. The correlation may be used in systems that use full antenna processing (AAP).

The analysis may also be based on other linear receivers, as known to those skilled in the art.

Successive interference cancellation receiver processing techniques attempt to recover the transmitted symbol streams, one at each step, using spatial or space-time receiver processing. As each symbol stream is recovered, the interference caused by the recovered symbol stream on the remaining, yet unrecovered symbol streams is estimated and canceled from the received symbol stream, and the modified symbol stream is similarly processed in a next step to recover the next transmitted symbol stream.

For a SIC receiver, the first step is first at N_RSpatial or time-space processing on the modified symbol streams to attempt separation (N)_T-l +1) transmitted symbol streams that have not been recovered. If the SIC receiver uses a linear ZF equalizer, each transmitted symbol stream may be filtered by using a filter matched to the transmitted symbol stream to filter the N_RThe modified symbol streams are separated. For simplicity, the following description assumes that the transmitted symbol streams are recovered in ascending order (i.e., the symbol streams from transmit antennas 1 are recovered first, the symbol streams from transmit antennas 2 are recovered next, and so on, from transmit antennas N_TWill be finally recovered). However, this is not required and the transmitted symbols may be recovered in other orders.

The matched filter used to recover the l symbol stream in the l stage has a unity norm vectorW _t(k) K for each tone with N_RFilter coefficients, where k is 0, 1_F-1). In order to minimize the noise from other (N)_T-l) interference caused by symbol streams not yet recovered on the l symbol stream, vectorW _t(k) Is defined as being directly connected withh _j(k) Is orthogonal, where j ═ l +1, l +2,. N_T. The condition may be expressed as a condition that,W _t(k)h _j(k) 0, wherein j is l +1, l +2_TAnd also k 0, 1 for each tone k_F-l). Since the transmitted symbol streams from the other (l-1) transmit antennas have been recovered in the previous stages and are recovered from the modified data stream for the l-th stagey ^t(k) Vector of medium cancellationW _t(k) Not necessarily withh _j(k) Quadrature, where j is 1, 2_F-1)。

Channel response of matched filterW _t(k) May be derived based on different spatial or space-time processing techniques. For example, matched filter responseW _t(k) Can be derived using a linear ZF equalizer. For SIC receivers, the channel response matrixH(k) Since the transmitted symbol stream is recovered, one column is reduced in each stage. Reduced channel response signal for stage IH ¹(k) Is one (N)_R×(N_TL +1)) matrix for (l-1) columns of transmit antennas of (l-1) symbol streams before recovery from the original matrixH(k) Is removed. ZF equalizer response matrix for l stageW _ZF ^l(k) May be based on a reduced channel response matrixH ¹(k) Is derived as shown in equation (16). However, due to each stageH ¹(k) Is different, each stageW _ZF ^l(k) As well as different. Matched filter response for ith symbol stream in ith stagew _l(k) Can be expressed asw _l(k)＝w _ZF ^l(k) Whereinw _ZF ^l(k) Corresponding to the l-th transmit antenna and is the ZF equalizer response matrixW _ZF ^l(k) The first column in (1), which is derived for the l-th stage.

Detected symbol stream for the l-th transmit antennaIt can be estimated as follows:

spatial or space-time processing for the l-th step of the SIC receiver can be provided (N)_TL +1) detected symbol streams,wherein j ═ l, l +1_T. Each detected symbol stream includes all N at different transmit antennas_FModulation of transmissions on frequency sub-channelsAnd (4) code elements. Spatial processing thus effectively maps a MIMO system to several parallel SISO systems. For (N) detected in stage l_T-l +1) symbol streams, the symbol stream corresponding to the l transmit antenna being selected for further processing to obtain data for that symbol stream.

Cancellation of interference due to the recovered symbol stream is effective if the symbol stream can be recovered without error (or with minimal error) and if the channel response estimate is reasonably accurate. The later recovered symbol stream will experience less interference and may achieve a higher SNR. In this way, higher performance may be achieved for all recovered symbol streams (possibly except for the first recovered symbol stream). The SIC processing technique may be better than the space/space-time receiver processing technique if the interference due to each recovered stream can be accurately estimated and canceled. This requires error-free or low-error recovery of the transmitted symbol stream, which can be achieved in part by applying an error correction code to the symbol stream.

In general, one important consideration for SIC receivers is the order of detection of the transmitted symbol streams. If the same data rate is used for all transmit antennas, the highest SNR in the detected symbol stream can be selected for recovery. However, using the rate control described herein, the rates for the transmit antennas may be selected such that all detected symbol streams have similar reliability. With rate control, it is not an important consideration in which order to use to detect the symbol stream.

In an aspect, in a multi-channel system that uses multiple transport channels for data transmission, each independently processed data stream may be modeled using an equivalent SISO system. Rate control may then be performed for each data stream in a manner similar to that described above for SISO systems.

MIMO-OFDM system using AAP

If AAP is used in a transmitter of a MIMO-OFDM system, it is in each transmission symbol periodSpatial or space-time processing at a receiver provides from N_TN transmitted by transmitting antenna_TA detected OFDM symbol. Each detected OFDM symbol includes symbols for N_FN of one frequency subchannel_FAnd a modulation symbol. N is a radical of_TThe detected OFDM symbols are typically attenuated independently, and each OFDM symbol is distorted by the response of the spatial subchannel used to receive the OFDM symbol.

For the AAP scheme, interleaving is performed in both the frequency and spatial domains. Thus, the codeword may be in all N_TInterleaving is performed on the detected OFDM symbols. MIMO-OFDM system using AAP (using all N)_TN_FOne transmission channel to transmit a codeword) may then be mapped to use N_TN_FEquivalent SISO system of sub-carriers and occupying a single spatial sub-channel N_TMultiple bandwidth (and thus, meeting with N_TL multipath channels). If the mapping is valid, the equivalent SNR for an equivalent SISO system with AWGN can then be used to select a suitable data rate for a MIMO-OFDM system with a multipath channel.

Fig. 5A is a diagram illustrating spectral efficiency of spatial subchannels in a MIMO system with a multipath channel. For MIMO-OFDM systems, if the channel response matrixH(k) Is of full rank (i.e. N)_S＝N_T≤N_R) And then has N_TA spatial subchannel. In this case, each spatial subchannel is associated with a different transmit antenna and has a bandwidth of W. The channel response of each spatial subchannel (or each transmit antenna) is determined byh _j(k) Wherein j is 0, 1_TAnd k is 0, 1. (N)_F-1) definition, whereinh _j(k) Is a matrixH(k) And includes for N_RN of one receiving antenna_RAnd (4) each element.

For each with a channel responseh _j(k) Sum noise variance N₀For N, of a transmitting antenna_FThe curve 510 of the spectral efficiency of the individual frequency subchannels may be based on the equation (2) or (3)The constrained or unconstrained spectral efficiency function shown in (a). Average spectral efficiency S for each transmit antenna_avgCan be derived from equation (5). As shown in fig. 5A, due to the independent attenuation of these spatial subchannels for N_TA transmitting antenna (or N_TSpatial subchannels) the spectral efficiency curves 510a through 510t may be different.

Fig. 5B is a graph illustrating the spectral efficiency of an equivalent SISO system used to model the MIMO-OFDM system shown in fig. 5A. An equivalent SISO system is defined as having an AWGN channel and a spectral efficiency equal to the average spectral efficiency of the MIMO-OFDM system being modeled. For having N_TMIMO-OFDM system with parallel colored-noise (colored-noise) channels, each channel occupying W bandwidth and overall capacity C_mimoCan be expressed as:

where | Σ | is the determinant of Σ, and Σ_SIs a diagonal matrix with post-equalizer signal power. Diagonal matrix sigma_SCan be derived based on equation (18) and can be expressed as:

capacity C of MIMO-OFDM system_mimoCan be expressed as:

wherein S_jIs the spectral efficiency in bits/hz corresponding to the jth transmit antenna. For simplicity, the lower limit in equation (21), i.e.And may be used in the following description. However, the true capacity of a MIMO-OFDM system may also be used and this is within the scope of the invention.

Occupied N_TW-Bandwidth equivalent SISO System Capacity C_SISOCan be expressed as:

C_siso＝N_TWS_equiv (22)

wherein S_equivIs the spectral efficiency in bits/hertz of an equivalent SISO system with an AWGN channel.

Set up C_sisoIs equal to C_mimoAnd combining equations (22) and (23), the spectral efficiency S of the equivalent SISO system_equivCan be expressed as:

spectral efficiency S of each transmit antenna in a MIMO-OFDM system_jCan be expressed as:

wherein W_j(k) Is the ZF equalizer response for the jth transmit antenna, e.g., the matrix determined in equation (16)W _ZF(k) Column j in (d).

The function f (x) in equation (24) is a function of snr (k) and modulation scheme m (r). The SNR of the kth frequency subchannel for the jth transmit antenna may be expressed as:

average spectral efficiency S of MIMO-ODFM system using AAP_avg，AAPCan be expressed as:

average spectral efficiency S of MIMO-OFDM system using AAP_avg，AAPWhich can then be used as the spectral efficiency S of an equivalent SISO system_equiv(i.e., S)_equiv＝S_avg，AAP)。

Spectral efficiency S in equivalent SISO systems_equivMay then be determined for the MIMO-OFDM system using AAP, as shown in equation (6):

Ψ＝SNR_equiv＝g(S_equiv，M(r)) (27)

as shown in equation (27), the equivalent SNR is the spectral efficiency S obtained for the equivalent system_equivAs shown in equations (24) and (26), by dividing all N_TSpectral efficiency S of a transmitting antenna_jObtained after averaging, wherein j is 0, 1_T. Spectral efficiency S of each transmitting antenna_jSequentially for all N_FThe spectral efficiency of the individual frequency subchannels is averaged. Therefore, the equivalent SNR is determined by the average spectral efficiency of all frequency subchannels and spatial subchannels, as shown in fig. 5B. The equivalent SNR can then be used as the metric Ψ to determine the rate of data transmission on all transmit antennas, in a manner similar to that described above for the SISO system.

As shown in fig. 5B, due to N_THair setting deviceSpectral efficiency function f of a radiating antenna_j(x) Wherein j is 0, 1_TThe spectral efficiency profile 520 of an equivalent SISO system may have discontinuities. However, the effect of discontinuities can be mitigated by interleaving the data between the frequency and spatial domains prior to transmission by an interleaver used at the transmitter.

MIMO-OFDM system using PAP

If PAP is used at the transmitter of the MIMO-OFDM system, the secondary N may be selected_TN transmitted by transmitting antenna_TEach of the data streams is rate controlled. At the receiver, space/space-time processing or SIC processing may be used to recover N_TA stream of transmitted symbols. Since SIC processing may provide better performance than space/space-time processing for PAP, the following description will be for a SIC receiver.

For a SIC receiver, to recover the symbol stream from the l-th transmit antenna in the l-th stage, the interference hypothesis from (l-1) symbol streams before recovery is cancelled, and from the others (N)_T-l) interference of the symbol streams not yet recovered can be detected by selecting a filter response that is correctly matched to the symbol stream to be recovered in that stagew _l(k) To minimize (or eliminate (null out)). Matched filter responsew _l(k) Including for N_RN of one receiving antenna_RAn element, wherein each element is a logical element having a value for N_FN of one frequency subchannel_FA vector of coefficients. Thus, each stage of the SIC receiver resembles a single (1, N)_R) SIMO system.

Average spectral efficiency S of each transmitting antenna in a MIMO-OFDM system using PAP_{avg，PAP，l}Can be expressed as:

whereinh _l(k) Andw _l(k) representing the channel response and the filter response associated with the ith transmit antenna, respectively. Average spectral efficiency S per transmit antenna in a MIMO-OFDM system using PAP_{avg，PAP，l}Spectral efficiency S that can be used as an equivalent SISO system_equiv(i.e., S)_eqyiv＝S_{avg，PAP，l}) To determine the velocity of the transmit antenna.

The function f (x) in equation (28) is a function of SNR and modulation scheme m (r). The SNR of the kth frequency subchannel for the ith transmit antenna may be expressed as:

as noted above, the matched filter response for the symbol stream recovered in stage Iw _l(k) Is ZF equalizer response matrixW _ZF ^l(k) One column of (a). Matrix for l-th stageW _ZF ^l(k) Reduced based channel response matrixH ¹(k) And (l-1) columns for the symbol stream before (l-1) recovery are removed.

Average spectral efficiency S of equivalent SISO system for each transmit antenna in MIMO-OFDM system using PAP_equivCan be determined as shown in equation (28) for spectral efficiency S_equivMay then be as followsIs determined as shown in equation (27). The equivalent SNR for each transmit antenna is determined by averaging all frequency subchannels for the transmit antenna, as shown in fig. 5A. The equivalent SNR for each transmit antenna may be used as a metric Ψ to determine the rate for that transmit antenna, in a manner similar to that described previously for SISO systems.

Multi-channel system using MCP

For a multipath processing (MCP) scheme, one or more data streams are independently processed (e.g., encoded, interleaved, and modulated) at a transmitter to provide one or more corresponding symbol streams, each of which may be transmitted on a different set of transmission channels. Each transport channel group may include (1) some or all of the frequency subchannels of a spatial subchannel, (2) some or all of the frequency subchannels of a plurality of spatial subchannels, (3) some or all of the spatial subchannels of a frequency subchannel, (4) some or all of the spatial subchannels of a plurality of frequency subchannels, (5) a combination of any transport channels, or (6) all of the transport channels. The rate of each independently processed data stream can be controlled so that improved performance (i.e., high throughput) can be achieved. AAP and PAP can be considered as variations of MCP schemes.

Fig. 6 is a flow diagram of one embodiment of a process 600 for controlling one or more independently processed data streams in a multi-channel system, each of which is transmitted on a different set of transmission channels.

Initially, a first data stream to be rate controlled is selected, for example, by setting a variable m for indicating the data stream to 1 (i.e., m-1) (step 612). For data stream d_mThe set of transmission channels of (a) is then determined (step 614). For the AAP scheme, a data stream is transmitted on all frequency subchannels of all spatial subchannels, and a transport channel group may include all transport channels. For the PAP scheme, a data stream is transmitted on all frequency subchannels of each spatial subchannel, and the set of transmission channels will include a data stream d for transmission_mAll of the transmitting antennas ofA frequency subchannel. For the MCP scheme, a group of transmission channels may include any combination of frequency and spatial subchannels.

Can be used for data stream d_mHighest available rate R_m(r) is then selected for evaluation (step 616). If the available rates are included in an ascending set of orders, the highest available rate may be selected by setting a variable r equal to P (i.e., r-P), which is the highest index number in the set. The same rate set may be used for all data streams, or each data stream may be associated with a different rate set.

And d_mAnd R_m(r) the relevant parameters are next determined (step 618). Some parameters may be associated with the processing data stream d_mCorrelation, e.g. modulation scheme M for the data stream_m(r) of (A). Some other parameters may be associated with the communication channel, such as the channel response h for each transport channel in the group_i，j(k) And the variance N of the noise₀。

The metric Ψ is then determined for data stream d_m(block 620). In one embodiment, metric Ψ and for modeling are used to transmit data stream d_mThe SNR of the equivalent SISO system of the transport channel group of (a). The metric Ψ may be determined by first determining the metric for stream d_mOf all transmission channels S_{avg，MCP，m}To obtain (step 622). It can be expressed as:

wherein h is_nAnd w_nRespectively, a channel response and a filter response associated with the nth transmission channel, where n is an index comprising (i, j, k), M_m(r) is for data stream d_mModulation scheme of (1), and N_mIs for data stream d_mThe number of transmission channels. For data stream d_mThe same modulation scheme may be used for all transport channels, as shown in equation (30), or different modulation schemes may be used for different transport channels.

The spectral efficiency of the equivalent SISO system is then set equal to that for transmitting data stream d_mOf the transmission channel (i.e., S)_equiv，m＝S_{avg，MCP，m}) (step 624). Supporting rate S in equivalent SISO system_equiv，mThe required equivalent SNR is then determined based on equation (27) (step 626). The equivalent SNR may be adjusted by considering the amount of backoff to achieve the loss, as described above for the SISO system (step 628). This step is optional and step 628 is represented using a dashed box. The metric Ψ is then set equal to the unadjusted or adjusted equivalent SNR (step 630). Multi-channel system with AWGN channel at rate R_m(r) to reliably transmit data stream d_mAs requiredThe SNR is then determined, for example from a table (step 632).

The rate R is then determined_m(r) whether or not it is for data stream d_mSupported by the transport channel group (step 636). If the metric Ψ is greater than or equal to the required SNR (i.e., Ψ ≧ SNR)_req) Then rate R_m(r) is considered to be supported for data stream d_mThen the process proceeds to step 640. Otherwise, the next lower available rate is selected for data stream d by decreasing the index r (i.e., r-1)_m(step 638). The process then returns to estimating the new rate at step 618.

At step 640, it is determined whether rate control has been performed for all data streams. If the answer is no, then the next data stream is rate controlled by incrementing the variable m (i.e., m +1) (step 642). The process then returns to step 614 to determine a new data stream d for_mThe rate of (c). Otherwise, if all data streams have been rate controlled, then provision is made for N_DSet of rates of independently processed data streams R_m(r) }, wherein m ═ 1, 2,. N_D(step 644). The process terminates.

Computer simulations have shown that the rate control techniques described herein can achieve the performance of a desired rate selection scheme. The ideal rate selection scheme is a non-realistic scheme that tests each available rate and selects a PER that meets the desired P_eThe highest data rate of the PER. The rate control techniques described herein may thus be used to implement a viable rate control scheme with high performance.

Fig. 7 is a block diagram of one embodiment of a transmitter system 110a and a receiver system 150a in the multipath communication system 100.

At transmitter system 110a, traffic data is provided from a data source 708 to a TX data processor 710. TX data processor 710 can multiplex the data into a number of data streams and further format, code, and interleave the data streams based on a coding scheme to provide corresponding coded data streams. The data rate and coding for each data stream may be determined by data rate control and coding control, respectively, provided by controller 730.

The encoded data is then provided to a modulator 720, which also receives pilot data (e.g., data used for channel estimation and other functions). The pilot data may be multiplexed with the encoded data, e.g., using time division multiplexing or code division multiplexing on all or a subset of the transmission channels used to transmit traffic data. For OFDM, the processing by modulator 720 may include (1) modulating the received data using one or more modulation schemes, (2) transforming the modulated data to form OFDM symbols, and (3) adding a cyclic prefix to each OFDM symbol to form a corresponding transmission symbol. The modulation is based on modulation control provided by a controller 730, and the stream of transmission symbols is then provided to each transmitter (TMTR) 722.

Each transmitter 722 converts a received stream of transmission symbols into one or more analog signals and further conditions (e.g., amplifies, filters, and frequency upconverts) the analog signals to generate a modulated signal suitable for transmission over the communication channel. The modulated signals from each transmit antenna 722 are then transmitted through an associated antenna 724 to a receiver system.

At receiver system 150a, the transmitted modulated signals are received by each of antennas 752a through 752r and the received signal from each antenna is provided to an associated receiver (RCVR) 754. Each receiver 754 conditions (e.g., filters, amplifies, and downconverts) its received signal and digitizes the conditioned signal to provide data samples. The sample streams from receivers 754a through 754r are then provided to a receiver processor 760, which includes a demodulator 762 and a RX data processor 764.

For OFDM, the processing by demodulator 762 may include (1) removing the cyclic prefix previously appended to each OFDM symbol, (2) converting each recovered OFDM symbol, and (3) demodulating the recovered modulated data in accordance with one or more demodulation schemes that are complementary to the one or more modulation schemes used at the transmitter system. RX data processor 764 then decodes the demodulated data to recover the transmitted traffic data. The processing by demodulator 762 and RX data processor 764 is complementary to the processing by modulator 720 and TX data processor 710, respectively, at transmitter system 110 a.

As shown in fig. 7, modulator 762 may derive estimates of the channel characteristics (e.g., channel response and noise variance) and provide these channel estimates to a controller 770. RX data processor 764 may also derive and provide the status of each received packet and may further provide one or more other performance metrics indicative of the decoding results. Based on the different types of information received from demodulator 762 and RX data processor 764, controller 770 may determine or select a particular rate for each independently processed data stream based on the techniques described above. Feedback information in the form of a set of selected rates for the data streams, channel response estimates, ACKs/NACKs for received packets, etc., or any combination thereof, may be provided by controller 770, processed by TX data processor 778, modulated by modulator 780, conditioned by transmitter 754, and transmitted by antenna 752 back to transmitter system 110 a.

At transmitter system 110a, the modulated signals from receiver system 150a are received by antennas 724, conditioned by receivers 722, demodulated by a demodulator 740, and processed by a RX data processor 742 to recover the feedback information transmitted by the receiver system. The feedback information is then provided to the controller 730 and used to control the processing of the data stream. For example, the rate for each data stream may be determined based on a selected rate provided by the receiver system or may be determined based on a channel estimate from the receiver system. The particular coding and modulation schemes associated with the selected rates are determined and reflected in the coding and modulation controls provided to TX data processor 710 and modulator 720. The received ACK/NACK may be used to initiate an incremental (elementary) transmission in which a small portion of the packet that was received in error is retransmitted to allow the receiver to correctly recover the packet.

Controllers 730 and 770 direct the operation at the transmitter and receiver systems, respectively. Memories 732 and 772 provide storage of program code and data used by controllers 730 and 770, respectively.

Fig. 8 is a block diagram of a transmitter unit 800, which is an embodiment of the transmitter portion of transmitter system 110a shown in fig. 7. Transmitter unit 800 includes (1) a TX data processor 710a that encodes each data stream in accordance with a particular coding scheme to provide a corresponding coded data stream, and (2) a modulator 720a that modulates and OFDM processes the coded data to provide a stream of transmission symbols.

In one embodiment, each data stream may be associated with its own data rate and coding and modulation scheme identified by controls provided by controller 730. The rate control for each data stream may be as described above.

In the embodiment shown in FIG. 8, TX data processor 710a includes a demultiplexer (demultiplexer)810, N_DEncoders 812a to 812s, and N_DChannel interleavers 814a through 814s (i.e., one set of encoder and channel interleaver for each data stream). Demultiplexer 810 demultiplexes traffic data (i.e., information bits) into N_DA data stream of which N_DAny integer greater than or equal to 1 may be used. N is a radical of_DA data stream to be determined as being composed of N_DThe data rates supported by a set of transport channels for these data streams are provided. Each data stream is provided to a respective encoder 812.

Each encoder 812 encodes a respective data stream based on a particular coding scheme selected for that data stream to provide encoded bits. The encoding increases the reliability of the data transmission. The coding scheme may include any combination of Cyclic Redundancy Check (CRC) codes, convolutional codes, Turbo codes, block coding, and the like. The encoded bits from each encoder 812 are then provided to a respective channel interleaver 814, which interleaves the encoded bits based on a particular interleaving scheme. The interleaving provides time diversity of the coded bits,allowing data to be transmitted based on the average SNR of the transmission channel for the data channel, countering fading, and further removing correlation between the coded bits used to form each modulation symbol. N is a radical of_DThe encoded data streams are then provided to a modulator 720 a.

In the embodiment shown in FIG. 8, modulator 720a includes N_DA symbol mapping element 822a through 822s (one for each data stream), a multiplexer/demultiplexer 824, and N_TOFDM modulators (one for each transmit antenna), each OFDM modulator comprising an inverse fourier transform (IFFT) unit 826 and a cyclic prefix generator 828.

Each symbol mapping element 822 receives a respective encoded data stream and maps the coded and interleaved bits based on a modulation scheme selected for that data stream to form modulation symbols. Each mapping element 822 maps a set q of_mThe encoded and interleaved bits are grouped to form a non-binary symbol, and the non-binary symbol is further mapped to a particular point on a signal constellation corresponding to the selected modulation scheme (e.g., QPSK, M-PSK, or M-QAM). Each mapped signal point corresponds to an M_mDimension modulation symbol, where M_mCorresponding to d as data stream_mA particular modulation scheme selected, and. The pilot data may also be symbol mapped to provide pilot symbols, which are then multiplexed (e.g., using TDM or CDM) with the modulation symbols for traffic data. Symbol mapping elements 822a through 822s are then provided for N_DThe modulation symbols for the individual data streams are passed to a multiplexer/demultiplexer 824.

Each data stream is transmitted over a respective set of transmission channels, and each set of transmission channels may include any number or combination of spatial and frequency subchannels.A multiplexer/demultiplexer 824 provides the modulation symbols for each data stream to a transmission channel ready for the data stream. Multiplexer/demultiplexer 824 then provides N_TOne modulation symbol stream to N_TAn OFDM modulator.

For the AAP scheme, a data stream is transmitted on all transport channels, and only one set of encoder 812, channel interleaver 814, and symbol mapping element 822 is needed. Multiplexer/demultiplexer 824 then demultiplexes the modulation symbols into N_TOne for N_TA stream of modulation symbols for each transmit antenna.

For the PAP scheme, a data stream is transmitted on all frequency subchannels of each transmit antenna, and N needs to be provided_TGroup encoder 812, channel interleaver 814, and symbol mapping element 822 (i.e., N)_D＝N_S). Multiplexer/demultiplexer 824 then simply passes the modulation symbols from each symbol mapping element 822 to an associated IFFT 826.

For the MCP scheme, each data stream is transmitted on a respective set of transmission channels. A multiplexer/demultiplexer 824 appropriately multiplexes/demultiplexes the modulation symbols onto the correct transport channels.

In each OFDM modulator, IFFT 826 receives a stream of modulation symbols, N for each group_FThe individual modulation symbols are grouped to form a corresponding vector of modulation symbols, and the vector is converted to its time-domain representation (referred to as an-OFDM symbol) using an inverse fast fourier transform. For each OFDM symbol, cyclic prefix generator 828 repeats a portion of the OFDM symbol to form a corresponding transmission symbol. The cyclic prefix ensures that the transmission symbol maintains its orthogonal properties in the presence of multipath delay spread, thus improving performance against deleterious path effects such as channel dispersion caused by frequency selective fading. Cyclic prefix generator 828 then provides a stream of transmission symbols to an associated transmitter 722.

Each transmitter 722 receives and processes a respective transmission symbol stream to generate a modulated signal, which is then transmitted from an associated antenna 724.

The coding and modulation of MIMO systems with or without OFDM will be described in further detail in the following U.S. patent applications:

U.S. patent application Ser. No.09/993,087 entitled "Multiple-Access Multiple-Input Multiple-output (MIMO) Communication System", filed on 11/6/2001;

U.S. patent application Ser. No.09/854,235, entitled "Method and Apparatus for processing Data in Multiple-Input Multiple-output (MIMO) communication System Utilizing Channel State Information", filed on 11/3/2001;

U.S. patent application Ser. Nos.09/826,481 and 09/956,449, both entitled "method and Apparatus for Utilizing Channel State Information in a Wireless communication System", filed on 3/23 and 9/18 of 2001, respectively;

U.S. patent application Ser. No.09/776,075, entitled "Coding Scheme for aWireless Communication System", filed on 2.1.2001; and

U.S. patent application Ser. No.09/532,492, entitled "High Efficiency, High Performance Communication System Employing Multi-Carrier Modulation", filed on 3/30/2000.

All of which are assigned to the assignee of the present application and incorporated herein by reference. Other designs for the transmitter unit may be implemented and are within the scope of the invention.

Fig. 9 is a block diagram of one embodiment of a receiver processor 760a, which is one embodiment of receiver processor 760 in fig. 7. The transmitted modulated signals are received by antennas 752 and processed by receivers 754 to provide NR sample streams, which are then provided to an RX OFDM processor 910 in a demodulator 762 a.

In demodulator 762a, each sample stream is provided to a respective OFDM demodulator, which includes a cyclic prefix removal element 912 and an FFT unit 914. Element 912 removes the cyclic prefix included in each transmission symbol to provide a corresponding recovered OFDM symbol. FFT 914 then transforms each recovered OFDM symbol using a fast Fourier transform to provide a symbol for N in each transmission symbol period_FN of one frequency subchannel_FThe recovered modulation symbols. FFT units 914a through 914r provide N_REach received symbol stream is then passed to a spatial processor 920.

Spatial processor 920 is at each received N_RSpatial or space-time processing on the symbol streams to provide N_TA detected symbol stream, which is for N_TAn estimate of the transmitted symbol streams. Spatial processor 920 may implement a linear ZF equalizer, a Channel Correlation Matrix Inverse (CCMI) equalizer, a Minimum Mean Square Error (MMSE) equalizer, an MMSE linear equalizer (MMSE-LE), a Decision Feedback Equalizer (DFE), or some other equalizer, which are described in detail in the aforementioned U.S. patent application serial nos. 09/993,087, 09/854,235, 09/826,481, 09/956,449.

A multiplexer/demultiplexer 922 then multiplexes/demultiplexes the detected symbols and will be used for N_DN of one data stream_DThe total (aggregated) detected symbols are provided to N_DSymbol demapping element 924. Each symbol demapping element 924 then demodulates the detected symbols in accordance with a demodulation scheme that is complementary to the modulation scheme used for the data stream. From N_DN of symbol mapping element 924_DThe demodulated data streams are then provided to a RX data processor 764 a.

In RX data processor 764a, each demodulated data stream is deinterleaved by a channel deinterleaver 932 in a manner complementary to the interleaving performed on the data stream at the transmitter system, and the deinterleaved data is further decoded by a decoder 934 in a manner complementary to the encoding performed at the transmitter system. For example, a Turbo decoder or a Viterbi (Viterbi) decoder may be used for decoder 934 if Turbo or convolutional coding, respectively, is performed at the transmitter unit. Decoder 934 may also provide the status of each received packet (e.g., indicating whether it was received correctly or in error). Decoder 934 may further hold demodulated data for packets that are not decoded correctly so that the data can be combined with data from subsequent delta transmissions and decoded.

In the embodiment shown in fig. 9, channel estimator 940 estimates the channel response and noise variance and provides these estimates to controller 770. The channel response and noise variance may be estimated based on the detected symbols for the pilot.

The controller 770 may be designed to perform various functions related to rate. For example, controller 770 may determine a maximum data rate for each data stream based on the channel estimates and other parameters such as modulation scheme.

Fig. 10 is a block diagram of one embodiment of a receiver processor 760b, which is another embodiment of receiver processor 760 in fig. 7. If the PAP or MCP scheme is used at the transmitter system, the receiver processor 760b performs SIC processing and may be used. For simplicity, the following description of the receiver processor 760b assumes that the PAP scheme is used.

In the illustrated embodiment of FIG. 10, receiver processor 760b may include (1) an RX OFDM processor 910 that processes N_RA sample stream to provide N_RA stream of received symbols, as described above, and (2) a space/data processor 1000. Spatial/data processor 1000 includes several successive (i.e., cascaded) receiver processing stages 1010a through 1010t that each recover a symbol stream. Each receiver processing stage 1010 (except for the final stage 1010t) includes a spatial processor 1020, an RX data processor 1030, and an interference canceller 1040. The final stage 1010t may include only the spatial processor 1020t and the RX data processor 1030 t.

For the first stage 1010a, the spatial processor 1020a is based onA particular spatial or space-time equalizer (e.g., a linear ZF equalizer, a CCMI equalizer, an MMSE-LE, or a DFE) receives and processes N from RX OFDM processor 910_RA received symbol stream (marked as a vector)y ¹) To provide N_TA stream of detected symbols (labeled as vectors)). A data stream is selected for recovery and spatial processor 1020a provides a stream of detection symbols for the data streamTo RX data processor 1030 a. Processor 1030a further processes (e.g., demodulates, deinterleaves, and decodes) the selected detected symbolsTo provide a mutual decoded data stream. Spatial processor 1020a may further provide an estimate of the channel response, which may be used in all stages for spatial or space-time processing.

For the first stage 1010a, interference canceller 1040a receives N from the receiver_RA stream of received symbols (i.e., vectors)y ¹). Interference canceller 1040a also receives and processes (e.g., encodes, interleaves, and symbol maps) the decoded data from RX data processor 1030a to provide a stream of remodulated symbolsWhich is an estimate of the symbol stream just recovered. Remodulated symbol streamIs further processed in the time or frequency domain to derive interference components (denoted as interference vectors) to the just recovered symbol streami ¹) Is estimated. For time domain implementation, symbol streams are remodulatedIs OFDM processed to obtain a stream of transmission symbols, which is further encoded by an impulse response vector in a channelh _lN in (1)_REach of the elements is convolved to derive N_RAn interference component due to the just recovered symbol stream. Vector quantityh _lIs a channel impulse response matrixHOne column in (c), corresponding to the transmit antenna/used for the symbol stream just recovered. Vector quantityh _lComprising N_RA defined transmitting antenna l and N_RN of channel responses between receive antennas_RAnd (4) each element. For frequency domain implementation, symbol streams are remodulatedAnd channel frequency response vectorh _l(it is a matrix)HOne column of (1) of_RMultiplying each of the elements to derive N_RAn interference component. Interference componenti ¹Is then derived from the input symbol stream of the first stepy ₁Is subtracted to derive N_RA modified symbol stream (labeledy ²) Which includes all but the subtracted (i.e., cancelled) interference component. N is a radical of_RThe modified symbol streams are then provided to the next step.

From each of the second to last stages 1010b to 1010t, the spatial processor for that stage receives and processes N from the interference canceller in the previous stage_RA modified symbol stream to derive a detected symbol stream for the stage. For each stage, a detected symbol is selected and processed by the RX data processor to provide a corresponding decoded data stream. For each of the second to last stages, the interference canceller in that stage receives N from the interference canceller in the preceding stage_RThe modified symbol streams and the decoded data stream from the RX data processor in the same stage, deriving N due to the symbol streams recovered in that stage_RAn interference component and provides N_RThe modified symbol stream goes to the next stage.

Successive interference cancellation receiver processing techniques are described in further detail in the aforementioned U.S. patent application serial nos. 09/993,087 and 09/854,235.

Fig. 7 and 9 show a simple design in which the receiver sends back the rate for the data stream. Other designs may also be implemented and are within the scope of the invention. For example, the channel estimates may be transmitted to the transmitter (instead of the rate), which then determines the rate of the data stream based on these channel estimates.

The rate control techniques described herein may be implemented using other designs. For example, channel estimator 940 of fig. 9, which is used to derive and provide channel estimates, may be implemented in a receiver system using different elements. All or some of the processing to determine the rate may be performed by the controller 770 (e.g., using one or more look-up tables stored in the memory 772). Other designs for rate control are also contemplated and within the scope of the present invention.

The rate control techniques described herein may be implemented in different ways. For example, these techniques may be implemented in hardware, software, or a combination thereof. For a hardware implementation, some of the elements used to implement rate control may be implemented within one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described herein, or any combination thereof.

For a software implementation, some portions of the rate control may be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. The software codes may be stored in a memory unit (e.g., memory 732 or 772 in fig. 7) and executed by a processor (e.g., controller 730 or 770). The memory unit may be implemented within the processor or external to the processor, and may be communicatively coupled to the processor via various means as is known in the art.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art. The general principles defined herein may be applied to other embodiments without departing from the spirit and scope of the invention. Thus, the present invention should not be limited to the embodiments shown herein but should be accorded the widest scope consistent with the principles and inventive features disclosed herein.

Claims (41)

1. A method of determining a rate for data transmission over a wireless communication channel in a multi-channel communication system, comprising:

identifying a plurality of transmission channels to be used for data transmission;

defining an equivalent system of transmission channels based on one or more estimated channel characteristics of the transmission channels;

deriving a metric for a transmission channel based on the equivalent system;

a particular rate for data transmission is determined based on the metric.

2. The method of claim 1, further comprising:

determining an average spectral efficiency of the transmission channel based on the one or more estimated channel characteristics,

wherein said defining an equivalent system of transmission channels comprises defining said equivalent system as having an additive white gaussian noise AWGN channel and a spectral efficiency equal to an average spectral efficiency of said transmission channels.

3. The method of claim 2, further comprising:

estimating the spectral efficiency of each transmission channel based on one or more estimated channel characteristics, an

Wherein the determining of the average spectral efficiency of the transmission channel is based on an estimated spectral efficiency of the transmission channel.

4. The method of claim 3, wherein the estimating the spectral efficiency of each transmission channel is based on a constrained spectral efficiency function.

5. The method of claim 4, wherein estimating the spectral efficiency of each transmission channel is further based on a modulation scheme used for the data transmission.

6. The method of claim 3, wherein the estimating the spectral efficiency of each transmission channel is based on an unconstrained spectral efficiency function.

7. The method of claim 2, wherein the deriving the metric for the transmission channel comprises:

an equivalent signal-to-noise-and-interference ratio, SNR, for an equivalent system is determined, and wherein the metric is related to the equivalent SNR.

8. The method of claim 7, wherein the determining the equivalent SNR is based on an inverse function of a spectral efficiency function used to estimate a spectral efficiency of each transmission channel.

9. The method of claim 7, wherein the deriving the metric for the transmission channel further comprises:

the equivalent SNR is adjusted to account for losses in the communication system, and wherein the metric is related to the adjusted equivalent SNR.

10. The method of claim 1, further comprising:

determining a particular modulation scheme for data transmission, and wherein the defining an equivalent system is further based on the modulation scheme.

11. The method of claim 7, further comprising:

determining a required SNR for supporting a particular data rate by the communication system, wherein the particular data rate is determined to be supported by the transmission channel if the required SNR is less than or equal to the metric.

12. The method of claim 1, wherein the one or more estimated channel characteristics comprise an SNR for each transmission channel.

13. The method of claim 1, wherein the one or more estimated channel characteristics comprise an estimated frequency response and an estimated noise variance for the transmission channel.

14. The method of claim 1, wherein the transmission channel is at least one of a frequency subchannel and a spatial subchannel in a multipath wireless communication channel with frequency selective fading.

15. The method of claim 1, wherein the multi-channel communication system is a multiple-input multiple-output (MIMO) communication system and the transmission channels correspond to spatial subchannels.

16. The method of claim 1, wherein the multi-channel communication system is an orthogonal frequency division multiplexing, OFDM, communication system and the transmission channels correspond to frequency subchannels.

17. The method of claim 1, wherein the multi-channel communication system is a multiple-input multiple-output, MIMO, communication system using orthogonal frequency division multiplexing, OFDM, and wherein a transmission channel corresponds to a frequency subchannel of a spatial subchannel.

18. The method of claim 1, wherein a set of rates can be used for data transmission, the method further comprising:

each of the one or more available rates is estimated to determine a highest rate supported by the transmission channel.

19. A method of determining a rate for data transmission over a wireless communication channel in a multi-channel communication system, comprising:

identifying a set of transmission channels to be used for data transmission;

obtaining an estimated signal-to-noise-and-interference ratio, SNR, for each transmission channel;

estimating a spectral efficiency of each transmission channel based on the estimated SNR for the transmission channel;

determining an average spectral efficiency for each transmission channel based on the estimated spectral efficiency of the transmission channel;

determining an equivalent SNR for an equivalent system having a spectral efficiency equal to an average spectral efficiency of the transmission channel;

determining a required SNR for supporting a particular data rate by the communication system; and

it is determined whether a particular rate is supported by a transmission channel for data transmission based on the equivalent SNR and the required SNR.

20. The method of claim 19, wherein the estimating the spectral efficiency of each transmission channel is based on an unconstrained spectral efficiency function.

21. The method of claim 19, wherein estimating the spectral efficiency of each transmission channel is further based on a modulation scheme used for the data transmission.

22. The method of claim 19, wherein the multi-channel communication system is a MIMO communication system using OFDM.

23. A method of determining a set of rates for a set of data streams to be transmitted over a wireless communication channel in a multi-channel communication system, comprising:

identifying a set of transport channel groups to be used for each data stream;

defining an equivalent system for each transport channel group based on one or more estimated channel characteristics of the transport channels in the transport channel group;

deriving a metric for each transport channel group based on the associated equivalent system;

a rate for each data stream is determined based on a metric associated with the data stream.

24. The method of claim 23, further comprising:

estimating a spectral efficiency of each transmission channel based on the one or more estimated channel characteristics;

determining an average spectral efficiency of the transmission channels in each group based on the estimated spectral efficiency of the transmission channels,

wherein said defining an equivalent system comprises defining an equivalent system for each group of transmission channels to have an additive white gaussian noise AWGN channel and a spectral efficiency equal to the average spectral efficiency of the transmission channels in the group.

25. The method of claim 24, wherein the estimating the spectral efficiency of each transmission channel is based on an unconstrained or constrained spectral efficiency function.

26. The method of claim 24, wherein the deriving the metric comprises:

an equivalent signal-to-noise-and-interference ratio, SNR, for an equivalent system is determined, and wherein the metric is related to the equivalent SNR.

27. The method of claim 26, further comprising:

for each data stream, a required SNR for supporting a particular rate by the communication system is determined, wherein the particular rate is determined to be supported by the set of transmission channels for the data stream if the required SNR is less than or equal to a metric associated with the data stream.

28. The method of claim 23, wherein the multi-channel communication system is a MIMO communication system using OFDM, and wherein a transmission channel corresponds to a frequency subchannel of a spatial subchannel.

29. The method of claim 28, wherein each data stream is transmitted on a respective transmit antenna, and each group of transmission channels includes all frequency subchannels for one transmit antenna.

30. A receiver unit in a multi-channel communication system, comprising:

a channel estimator configured to derive estimates of one or more channel characteristics of a plurality of transmission channels; and

a rate selector configured to

An equivalent system is defined based on one or more estimated channel characteristics of the transmission channel,

deriving metrics for the transmission channel based on the equivalent system, an

A particular rate for data transmission is determined based on the metric.

31. The receiver unit of claim 30, wherein the rate selector is further configured to

Estimating a spectral efficiency of each transmission channel based on the one or more estimated channel characteristics;

determining an average spectral efficiency of the plurality of transmission channels based on the estimated spectral efficiency of the transmission channels, an

Wherein the rate selector is configured to define the equivalent system as having an additive white gaussian noise AWGN channel and a spectral efficiency equal to an average spectral efficiency of the transmission channel.

32. The receiver unit of claim 31, wherein the rate selector is configured to estimate the spectral efficiency of each of the transmission channels based on a constrained or unconstrained channel spectral efficiency function.

33. The receiver unit of claim 31, further comprising:

a memory configured to store one or more tables of functions for estimating spectral efficiency of each transmission channel.

34. The receiver unit of claim 30, further comprising:

a controller configured to provide feedback information including a particular rate.

35. An apparatus in a multi-channel communication system, comprising:

means for identifying a plurality of transmission channels to be used for data transmission;

means for defining an equivalent system based on one or more estimated channel characteristics of the transmission channel;

means for deriving a metric for a transmission channel based on the equivalent system;

means for determining a particular rate for data transmission based on the metric.

36. The apparatus of claim 35, further comprising:

means for estimating a spectral efficiency of each transmission channel based on the one or more estimated channel characteristics; and

means for determining an average spectral efficiency of the plurality of transmission channels based on the estimated spectral efficiency of the transmission channels, an

Wherein the means for defining defines the equivalent system as having an additive white gaussian noise AWGN channel and a spectral efficiency equal to an average spectral efficiency of the transmission channel.

37. The apparatus of claim 36, further comprising:

means for storing one or more tables of functions for estimating spectral efficiency for each transmission channel.

38. A transmitter unit in a multi-channel communication system, comprising:

a controller configured to identify rates in a wireless communication channel for data transmission on a plurality of transmission channels, wherein the rates are determined based on metrics of the transmission channels derived according to an equivalent system defined for the transmission channels based on one or more estimated channel characteristics of the transmission channels;

a transmit data processor configured to encode the data provided at the identified rate in accordance with a particular coding scheme to provide encoded data; and

a modulator configured to modulate the encoded data according to a particular modulation scheme to provide modulated data.

39. The transmitter unit of claim 38, further comprising:

a transmitter configured to generate at least one modulated signal of modulated data.

40. The transmitter unit of claim 38, wherein the multi-channel communication system is a MIMO communication system using OFDM, and the transmission channels correspond to frequency subchannels of the spatial subchannels.

41. An apparatus in a wireless communication system, comprising:

means for identifying a rate in a wireless communication channel for data transmission on a plurality of transmission channels, wherein the rate is determined based on a metric for the transmission channel derived from an equivalent system defined for the transmission channel based on one or more estimated channel characteristics for the transmission channel;

means for encoding the data provided at the identified rate in accordance with a particular encoding scheme to provide encoded data; and

means for modulating the encoded data in accordance with a particular modulation scheme to provide modulated data.