Adaptive vector modulation method and system

[0001] In general, the present invention relates to the field of communication systems. More specifically, the present invention relates to multi-antenna transceivers and links within wireless systems.

[0002] Prior art wireless communication systems such as 2^nd and 3^rd generation cellular systems (e.g., ETSI GSM, TIA IS-95, 3GPP UMTS, etc.) and wireless local area network systems (e.g., IEEE 802.11b) are generally capable of sustaining maximum information bit rates of approximately 20 Mbps or less. Future applications for extremely high speed personal wireless communication links anticipate information bit rates well above 100 Mbps. Further, the personal (and possibly wearable) nature of the services anticipated to be supported by such links require extremely low-power operation, and therefore minimal complexity in baseband signal processing (including demodulation and error correction stages).

[0003] The present invention offers an improved capacity for communication systems employing a high speed link capacity with low associated signal processing complexity.

[0004] One form of the present invention is a method of operating a transceiver including one or more transmitter antennas, at least one receiver antenna, and one or more propagation channels between the one or more transmitter antennas and the one or more receiver antennas. First, a binary stream assembled into groups of bits forming symbol indices is received by the transceiver. Second, the transceiver generates at least one complex symbol value in response to a reception of the binary stream with each one complex symbol value being normalized over one or more channel coefficients associated with the propagation channel(s). Thereafter, one of the receiver antennas is selected to receiver the at least one complex symbol value from the transmitter antenna(s). In one aspect, the selection of the receiver antenna is a function of a metric proportional to an average injection power corresponding to the selected receiver antenna. In another aspect, the selection of the receiver antenna is a function of a vector norm corresponding to the selected receiver antenna.

[0005] The foregoing form as well as other forms, features and advantages of the invention will become further apparent from the following detailed description of the presently preferred embodiment, read in conjunction with the accompanying drawings. The detailed description and drawings are merely illustrative of the invention rather than limiting, the scope of the invention being defined by the appended claims and equivalents thereof.

[0006] FIG. 1 illustrates a block diagram of one embodiment of an adaptive vector modulation transceiver in accordance with the present invention;

[0007] FIG. 2 illustrates a block diagram of a first embodiment of a spatial mapping module and a first embodiment of a receiver of the FIG. 1 adaptive vector modulation transceiver;

[0008] FIG. 3 illustrates a flowchart representative of a first embodiment of a receiver antenna selection method in accordance with the present invention;

[0009] FIG. 4 illustrates a flowchart representative of a second embodiment of a receiver antenna selection method in accordance with the present invention;

[0010] FIG. 5 illustrates a block diagram of a second embodiment of a spatial mapping module of the FIG. 1 adaptive vector modulation transceiver; and

[0011] FIG. 6 illustrates a block diagram of a second embodiment of a receiver of the FIG. 1 adaptive vector modulation transceiver.

[0012] FIG. 1 illustrates one embodiment of an adaptive vector modulation (AVM) system transceiver 10 (hereinafter AVM transceiver 10) in accordance with the present invention. A transmitter of the AVM transceiver 10 includes a symbol modulator 20 and a spatial mapping module 30. The symbol modulator 20 receives as an input a binary information stream assembled into groups of K bits forming symbol indices J. In response thereto, the symbol modulator 20 conventionally provides a generally complex-valued Quadrature Amplitude Modulation (QAM) symbol s_J in the form of a member of any well-known constellation (e.g., 16-QAM, 64-QAM, etc.). The spatial mapping module 30 receives the modulation symbol s_J, and in response thereto, the spatial mapping module 30 generally directs a transmission of M_T complex symbol values (not shown) from each of M_T transmitter antennas where 1M_T. In the embodiment of FIG. 1, M_T3 whereby spatial mapping module 30 specifically directs a transmission of three (3) complex symbol values (not shown) from a transmitter antenna TX₀, a transmitter antenna TX₁ and a transmitter antenna TX₂, respectively.

[0013] The transmitted complex symbol values are received by one of M_R receiver antennas where 1M_R, and M_T and M_R may or may not have identical values. In the embodiment of FIG. 1, M_R3 whereby the transmitted complex symbol values are received by one of a receiver antenna RX₀, a receiver antenna RX₁, and a receiver antenna RX₂. The complex symbol values transmitted from each transmitter antenna TX₀-TX₂ are weighted versions of modulation symbol s_J where the weights are a function of the receiver antenna among the receiver antennas RX₀-RX₂ selected to receive the complex symbol values, and the propagation channels between the M_T transmitter antennas and the M_R candidate receiver antennas. The result is a normalization of the complex symbol values over the channel coefficients associated with the propagation channels.

[0014] Upon reception of the complex symbol values by the selected receiver antenna, the receiver 40 demodulates the complex symbol values to recover an estimate of the modulation symbol s_J to thereby generate an estimate of the transmit symbol index J to obtain the binary information sequence input to the symbol modulator 20.

[0015] The symbol modulator 20, the spatial mapping module 30, and the receiver 40 may be implemented in hardware (analog or digital), software, or any combination of hardware and software. From a subsequent description of various embodiments of the spatial mapping module 30 and the receiver 40, those having ordinary skill in the art will appreciate a sequential operation of the components of the symbol modulator 20, the spatial mapping module 30, and the receiver 40 (e.g., in a software implementation) and a concurrent operation of the symbol modulator 20, the spatial mapping module 30, and the receiver 40 (e.g., in a hardware implementation).

[0016] FIG. 2 illustrates a spatial mapping module 30a (for the particular case of M_T3) as one embodiment of the spatial mapping module 30 (FIG. 1) and a receiver 40a as one embodiment of the receiver 40 (FIG. 1). In response to the modulation symbol s_J, and selection of the i-th receiver antenna, the spatial mapping module 30a generates a complex symbol value x₀, a complex symbol value x₁, and a complex symbol value x₂ in accordance with the following general equation 1: $\begin{array}{cc}{x}_m=\frac{\sqrt{E_s}\left(h_{i,m}^{*}\right)}{\underset{j=0}{\overset{M_T-1}{}}|h_{i,j}{|}^2}{s}_{j}m=0,1,,M_T-1& \left[1\right]\end{array}$

[0017] where E_s is a desired mean received symbol energy, M_T is the number of transmitter antennas (M_T3 for the spatial mapping module 30a), and h_i,j is a complex-valued channel coefficient.

[0018] For this particular implementation, the complex-valued baseband-equivalent channels connecting the transmitter antennas TX₀-TX₂ with each of the receiver antennas RX₀-RX₂ is assumed to comprise a single coefficient (conventionally referred to as a flat channel). Accordingly, there are M_RM_T distinct channel coefficients. In one embodiment, the channel coefficients are organized into a matrix L₁ having M_R rows (or groups) of M_T coefficients (equivalently, M_R length-M_T row vectors) in accordance with the following equation 2: $\begin{array}{cc}{L}_1=\left[\begin{array}{cccc}{h}_{1,1}& h_{1,2}& & h_{1,M_T}\\ h_{2,1}& h_{2,2}& & \\ & & & \\ h_{M_R,1}& & & h_{M_R,M_T}\end{array}\right]=\left[\begin{array}{c}{h}_1\\ h_2\\ \\ h_{M_R}\end{array}\right]& \left[2\right]\end{array}$

[0019] where each group comprises the M_T channel coefficients connecting the transmitter antennas TX₀-TX₂ to one of the receiver antennas RX₀-RX_2.

[0020] A transmission of the complex symbol value x₀, the complex symbol value x₁, and the complex symbol value x₂ from the transmission antennas TX₀-TX₂, respectively, is directed along a subset of the channel coefficients of equation 2 in accordance with the antenna selection method of the present invention.

[0021] FIG. 3 illustrates a flowchart 50 representative of one embodiment of a receiver antenna selection method. During a stage S52 of the flowchart 50, a metric proportional to the average injection power (AIP) (where the averaging process is defined with respect to the transmitted symbol processwhose component symbols may not have constant energyand assumes the fading channel coefficients are approximately constant for a particular interval) corresponding to selection of the i-th (of M_R) candidate receiver antennas is computed in accordance with the following equation 3: $\begin{array}{cc}{\mathrm{AIP}}_i=\frac{1}{\underset{j=0}{\overset{M_T-1}{}}|h_{i,j}{|}^2}& \left[3\right]\end{array}$

[0022] During a stage S54 of the flowchart 50, the receiver antenna (selected over index i) having the smallest average injection power metric AIP_i is selected, and that receiver antenna is then used to receive the complex symbol value x₀, the complex symbol value x₁, and the complex symbol value x₂ after implicit processing by the corresponding selected channel coefficient vector h_i according to the definitions of equation 2.

[0023] FIG. 4 illustrates a flowchart 60 representative of another embodiment of a receiver antenna selection method. During a stage S62 of the flowchart 60, the L₂ vector norm (VN) of the row vector h_i corresponding to each of the M_R candidate receiver antennas is computed in accordance with the following equation 4: $\begin{array}{cc}{\mathrm{VN}}_i=\underset{j=0}{\overset{M_T-1}{}}|h_{i,j}{|}^2={h_i}_2^2& \left[4\right]\end{array}$

[0024] During a stage S64 of the flowchart 60, the channel having the highest vector norm VN_i is selected, and the corresponding i-th receiver antenna is used to receive the complex symbol value x₀, the complex symbol value x₁, and the complex symbol value x₂ after implicit processing by the corresponding selected channel coefficients h_i.

[0025] In one embodiment, the spatial mapping module 30a executes either the flowchart 50 or the flowchart 60, and communicates the receiver antenna selection to a switch 41 of the receiver 40a. In a second embodiment, the switch 41 executes either the flowchart 50 or the flowchart 60, and communicates the receiver antenna selection to the spatial mapping module 30a. In a third embodiment, both the spatial mapping module 30a and the switch 41 execute either the flowchart 50 and/or the flowchart 60, and handshake as to the receiver antenna selection.

[0026] In response to the receiver antenna selection, the switch 41 establishes a communication between the receiving antenna associated with the selected channel and a slicer 42 (which could equivalently be termed a hard-decision symbol estimator). For example, upon a selection of the channel 1, the switch 41 establishes a communication between the receiving antenna RX₁, and the slicer 42 as exemplary illustrated in FIG. 2. The slicer 42 receives a modulation signal _J that is an estimation of modulation signal s_J. In response thereto, the slicer 42 conventionally generates a group of bits as an estimate of the symbol index J (FIG. 1).

[0027] The description above was particularly intended for use in channels where each channel coefficient comprises a single complex-valued coefficient. The invention can be extended to include so-called frequency-selective channels (i.e., where the channel comprises a non-zero length time-series of complex-valued coefficients).

[0028] FIG. 5 illustrates a spatial mapping module 30b as another embodiment of spatial mapping module 30 (FIG. 1). The spatial mapping module 30b is an extension of the present invention to frequency-selective channels using the well-known technique of Orthogonal Frequency Division Modulation (OFDM). In response to a length-N sequence of modulation symbols s_J, a serial to parallel converter 31 conventionally concatenates the N modulation symbols s_J into a group of modulation symbols s_J0, . . . , s_JN1. A spatial mapper 32 provides a group of complex symbol values X₀0, . . . , X₀N1, a group of complex symbol values X₁0, . . . , X₁N1 and a group of complex symbol values X₂0, . . . , X₂N1 as a function of modulation symbols s_J0, . . . , s_JN1.

[0029] In one embodiment, given the selection of the i-th receiver antenna, the spatial mapper 32 generates complex symbol values X₀0, . . .,X₀N1, complex symbol values X₁0, . . .,X₁N1 and complex symbol values X₂0, . . . ,X₂N1 in accordance with the following equation 5: $\begin{array}{cc}{X}_j\left[k\right]=\frac{\sqrt{E_s}\left(H_{i,j}^{*}\left[k\right]\right)}{\underset{m=0}{\overset{M_T-1}{}}|H_{i,m}\left[k\right]{|}^2}{s}_j\left[k\right]k=0,1,,N-1;j=0,1,,M_T-1& \left[5\right]\end{array}$

[0030] where H_i,jk is generally the single frequency-domain complex-valued channel coefficient corresponding to the k-th (of N) OFDM sub-channel connecting the j-th transmitter antenna to the i-th receiver antenna.

[0031] An inverse Fast Fourier Transform 33a (IFFT 33a) conventionally transforms the group of frequency-domain complex symbol values X₀0, . . . ,X₀N1 into a group of time-domain complex symbol values x₀0, . . . ,x₀N1. A parallel/serial converter 34a conventionally converts the group of time-domain complex symbol values x₀0, . . . ,x₀N1 into a time-domain complex symbol sequence denoted x₀. A cyclic redundancy module 35a conventionally adds a cyclic prefix and/or cyclic postfix to the time-domain complex symbol sequence x₀ to yield a time-domain complex symbol sequence z₀.

[0032] An IFFT 33b conventionally transforms the group of frequency-domain complex symbol values X₁0, . . . ,X₁N1 into a group of time-domain complex symbol values x₁0, . . . ,x₁N1. A parallel/serial converter 34b conventionally converts the group of time-domain complex symbol values x₁0, . . . ,x₁N1 into a time-domain complex symbol sequence x₁. A cyclic redundancy module 35b conventionally adds a cyclic prefix and/or cyclic postfix to the time-domain complex symbol sequence x₁ to yield a time-domain complex symbol sequence z₁.

[0033] An IFFT 33c conventionally transforms the group of frequency-domain complex symbol values X₂0, . . . ,X₂N1 into a group of time-domain complex symbol values x₂0, . . . ,x₂N1. A parallel/serial converter 34c conventionally converts the group of time-domain complex symbol values x₂0, . . . ,x₂N1 into a time-domain complex symbol sequence x₂. A cyclic redundancy module 35c conventionally adds a cyclic prefix and/or cyclic postfix to the time-domain complex symbol sequence x₂ to yield a time-domain complex symbol sequence z₂.

[0034] The cyclic prefixes and/or cycle postfixes added to time-domain complex symbol sequences x₀-x₂ are determined by factors such as, for example, a maximum channel time-dispersion.

[0035] FIG. 6 illustrates an OFDM receiver 40b as another embodiment of the receiver 40 (FIG. 1). The switch 41 can operated as previously described herein in connection with the flowchart 50 (FIG. 3) with the metric proportional to the average injection power (AIP) (where averaging is again defined with respect to the transmitted symbol process, for a set of channel coefficients assumed constant over an interval) corresponding to each candidate (i -th) receiver antenna computed in accordance with the following equation 7: $\begin{array}{cc}{\mathrm{AIP}}_i=\underset{k=0}{\overset{N-1}{}}\left(\frac{1}{\underset{j=0}{\overset{M_T-1}{}}|H_{i,j}\left[k\right]{|}^2}\right)& \left[7\right]\end{array}$

[0036] Alternatively, the switch 41 can operated as previously described herein in connection with the flowchart 60 (FIG. 4) with the vector norm VN of the channel corresponding to the i-th receiver antenna is computed in accordance with the following equation 8: $\begin{array}{cc}{\mathrm{VN}}_i=\underset{k=0}{\overset{N-1}{}}\left(\underset{j=0}{\overset{M_T-1}{}}|H_{i,j}\left[k\right]{|}^2\right)=\underset{k=0}{\overset{N-1}{}}{H_i\left[k\right]}_2^2& \left[8\right]\end{array}$

[0037] wherein a manner similar to equation 2 the row vector H_ik is defined as the lengthM_T vector comprising

[0038] A cyclic redundancy module 42 receives a time-domain complex symbol sequence Z_r, and in response thereto, conventionally removes the cyclic prefix and/or the cyclic postfix to yield a time-domain complex symbol sequence x_r. A serial/parallel converter 43 conventionally converts the time-domain complex symbol sequence x_r into a group of time-domain complex symbol values x_r0, . . . ,x_rN1. A Fast Fourier Transform 44 (FFT 44) conventionally transforms the time-domain complex symbol values x_r0, . . .,x_rN1 into a group of frequency-domain complex symbol values _j0_jN1. A parallel/serial converter 45 conventionally converts the frequency-domain complex symbol values _J0_JN1 into a set of N estimated modulation symbols _J that is an estimate of modulation symbols s_J (FIG. 1). The slicer 46 conventionally generates N symbol index estimates each identifying K bits (FIG. 1).

[0039] The present invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description. All changes that come within the meaning and range of equivalency of the claims are to be embraced within their scope.