WO2012031384A1

WO2012031384A1 - Method for demodulating dcm signals and apparatus thereof

Info

Publication number: WO2012031384A1
Application number: PCT/CN2010/076661
Authority: WO
Inventors: Yongfu Huang; Jun Ma; Junfeng Wang
Original assignee: Panovel Technology Corporation
Priority date: 2010-09-07
Filing date: 2010-09-07
Publication date: 2012-03-15
Also published as: CN104040983A

Abstract

A method for demodulating Dual Carrier Modulated (DCM) signals is disclosed. The method includes correcting phase rotation of received DCM signals; decoupling real components and imaginary components of the phase-corrected DCM signals; and performing log-likelihood ratio (LLR) computation separately on the real components and the imaginary components.

Description

METHOD FOR DEMODULATING DCM SIGNALS AND APPARATUS

THEREOF

FIELD OF THE INVENTION

[0001] The present disclosure generally relates to receiver technology in high-speed wireless communication systems, and, more particularly, to method and apparatus for optimal demodulations for Dual Carrier Modulation in a Multi-Band Orthogonal Frequency Division Multiplexing (MB-OFDM) based Ultra Wide-Band (UWB) system.

BACKGROUND

[0002] Increasing demand for more powerful and convenient data and information communication has resulted in a number of advancements, particularly in wireless communication technologies. Among recently emerging communication

technologies—especially those needing high data transfer rates—various ultra- wideband (UWB) technologies are gaining support and acceptance. UWB technologies are utilized for wireless transmission of video, audio or other high bandwidth data between various devices.

[0003] WiMedia MultiBand Orthogonal Frequency-Division Multiplexing (MB-OFDM) technique is an implementation of next generation networks utilizing ultra- wideband (UWB) wireless technology. The MB-OFDM modulation based ultra-wideband (UWB) system specifies eight data rates in terms of Mb/s, among which 320,400,480 date rates employ 16QAM Dual Carrier Modulation (DCM) to provide diversity gain to improve error effectiveness for a high speed data rate by broadening in frequency domain.

[0004] FIG. 1 A and FIG. IB illustrate a mapping rule for dual carrier modulation.

According to the mapping rule, on a transmitter side, every four data bits b₀ bi,b₂ and b₃ are mapped onto two complex numbers, namely, d(j),j=0, l, as shown in two 16QAM

constellations in FIG. 1A and FIG. IB, respectively. Suppose there are 100 subcarriers in total. The mapping rule may thus be formulated as follows:

where n is the symbol number and k is the OFDM subcarrier index, code bits have been mapped as

[0005] Referring to FIG. 1, bO and bl among the 4 bits are mapped to an in-phase component of the two subcarriers, and the b2 and b3 are mapped to a quadrature-phase component of the two subcarriers.

[0006] On the receiver side, taken the channel interference and noise into account, the received symbols may be expressed as follows:

where

h_cm{j) are the equivalent baseband channel coefficients, N_cm are the i.i.d Gaussian noise with variance σ² .

[0007] To demodulate the received symbols, several conventional solutions are proposed.

[0008] One of the well-known solutions is based on the observation that the real parts of do and dj are the nonsingular linear transformation (see equation 1) of b₀, bj, so that estimations of b₀ and bj can be computed through solving a 2 by 2 linear equation. The same way can be applied for b₂, b₃ and the imaginary part of d₀ and dj as well. The estimations of b₀ bi,b₂ and b₃ can be sent to the Viterbi decoder directly. However, in order to compute b₀ bi,b₂ and b₃, d₀ and dj have to be derived through channel equalization by dividing r_cm by the channel factor h_cm . For instance, in equation (2), suppose N_cm{j) is small, then, d₀ and dj can be expressed as follows:

[0009] Considering both equations (4) and (5), the following equation can be derived:

[0010] As can be seen from above, in the case of a small h_cm{j) , such as in a fading indoor channel, this approach often amplifies noise, and results in performance degradation. [0011] Patent Application Number US 7,512, 185 B2 also proposes a method for DCM demodulation. The method includes first solving a linear equation system, then computing a Log-Likelihood Ratio (LLR) of b₀ bi,b₂ and b₃.

[0012] However, such method also requires channel equalization which is likely to amplify noise, and result in performance degradation. Another drawback is that this method involves cumbersome computational complexity in LLR implementation.

SUMMARY

[0013] Detailed herein is a new DCM demodulation solution which, among other things, decouples the real part and the imaginary part of the received DCM symbols, eliminates the needs for channel equalization, and uses LLR as the sole method of soft metrics computation instead of solving linear equations, which significantly reduces LLR computational complexity, while still ensuring a well performance. In addition, the present invention is VLSI implementation friendly and the loss is negligible.

[0014] In one aspect of the invention, a method for demodulating Dual Carrier Modulated (DCM) signals is provided according to one embodiment of the present invention. The method includes correcting phase rotation of received DCM signals; decoupling real components and imaginary components of the phase-corrected DCM signals; and performing log-likelihood ratio (LLR) computation separately on the real components and the imaginary components.

[0015] In another aspect of the invention, an apparatus for demodulating Dual Carrier Modulated (DCM) signals is provided according to one embodiment of the present invention. The apparatus includes a correcting module for correcting phase rotation of received DCM signals; a decoupling module for decoupling real components and imaginary components of the phase-corrected DCM signals; and a log-likelihood ratio (LLR) processing module for performing LLR computation separately on the real components and the imaginary components. Such apparatus for demodulating DCM signals may be integrated in a receiver or may be implemented as a separate device.

[0016] It is to be noted that this Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In the drawings, each identical or nearly identical component that is illustrated in various figures is represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. In the drawings:

[0018] FIG. 1 A illustrates a constellation for DCM modulation;

[0019] FIG. IB illustrates a constellation for DCM modulation;

[0020] FIG. 2 illustrates an optimized ASIC implementation architecture according to one embodiment of the present invention;

[0021] FIG. 3 A illustrates a simulation under CM1 channel model;

[0022] FIG. 3B illustrates a simulation under CM2 channel model; and

[0023] FIG. 4 illustrates a flowchart of a method for DCM demodulation according to one embodiment of the present invention.

DETAILED DESCRIPTION

[0024] The purposes, technical solutions and advantages concerning the embodiments of the present invention will become more readily appreciated by reference to the following description of the embodiments, when taken in conjunction with the accompanying drawings.

[0025] FIG. 1 illustrates two constellations for DCM modulation where the X-axis refers to the real component and the Y-axis refers to the imaginary component. As previously described, dual carrier modulation (DCM) can be expressed by two 16QAM constellations, see FIG. 1 A and FIG. IB. However, at the receiver side, the constellations are generally phase-rotated due to the fading problem. If the phase-rotation can be eliminated first by the receiver, that is, the constellations can be rotated back to the shape as shown in FIG. 1 A and FIG. IB at the receiver, the subsequent computational complexity in demodulation will be greatly reduced.

[0026] In view of this, the present invention provides a method for optimal DCM demodulation. According to the present invention, the phase rotation is corrected before LLR computation is performed. Consequently, there is no need for channel equalization and the LLR computational complexity is thus reduced significantly.

[0027] In one embodiment, the phase rotation is corrected by multiplying exp(-z^'#) to both sides of equation (2). Subsequently, b₀ bi are decoupled from b₂ and b₃ by taking the real part and imaginary part of the received symbol, separately. Hence, the following equations can be obtained:

where

[0028] As can be seen, (b₀ bi) and (b₂ b₃) have been decoupled. It is to be noted that the multiplication of exp(-z^'#) is used in algorithm derivation only, not needed in algorithm implementation. Because we can prove that r and h can be replaced by the original r_cm and h_cm as shown in the following and the channel equalization is not needed.

[0100] The followings is an example illustrating one of the benefits of the present invention. Assume FIG. 1 A is the constellation whose phase rotation has been rectified by the receiver according to the present invention. Then, if the real component is known (e.g., -3), the value of b₀ bi can be easily determined, which are 0,0. Consequently, the task of calculating b₀ bi,b₂ and b₃ is thus reduced to calculating b₂ and b₃, which greatly reduces the computational complexity_. In other words, the computation of b₀ bi is decoupled from that of b₂ and b₃. In addition, as can be seen, no channel equalization is involved during this process. Thus, the problem of noise amplification, which is caused by channel equalization, can be addressed.

[0029] After the decoupling process, LLR computation may be performed. In one embodiment, LLR computation may be applied to equation (7) to compute soft metrics of b_0, b],b₂ and b₃ (equation (9)). Since the computation of b₀ and bj is separate from that of b₂ and b₃, the LLR computational complexity is reduced by 75% compared with the method without the decoupling process.

[0030] The above LLR algorithm is referred as low-complexity-LLR (LC-LLR).

Function "In" can be implemented as lookup table. In one embodiment, A may be 2 σ² .

[0031] Since the DCM is employed in a relatively high S R condition and the logarithmic term in equations (9), (10),(11), (12) can be neglected in the case of a high (e.g. snr>6dB)S R, the above LC-LLR can be further simplified through a process called MAX-LOG approximation (MAX-LOG-LC-LLR), which is illustrated below:

[0032] In one embodiment, the present invention may be implemented with an optimized ASIC architecture. The optimized ASIC implementation architecture is shown in FIG. 2, where c ~ c_w are constants. As can be seen from the architecture, some of the intermediate variables are repetitively used and invoked for a plurality of times. Advantageously, this optimized ASIC architecture is VLSI implementation friendly and the loss is negligible. Block Q in FIG. 2 is a quantization block. In order to produce soft metrics with reasonable bit width for the Viterbi decoder, the LLR result should be scaled and truncated to a smaller bit width.

[0033] FIG. 3 illustrates simulations under CM1 and CM2 channel models, where X-axis denotes SNR and Y-axis denotes Bit Error Ratio (BER). The BER of three algorithms are compared: (a) EQL, equalization before solving equations for b₀ bi,b₂ and b₃; (b) LC-LLR; (c) MAX-LOG-LC-LLR. It can be seen that LC-LLR algorithm and MAX-LOG-LC-LLR algorithm are both obviously better than EQL algorithm by up to 3 dB.

MAX-LOG-LC-LLR algorithm is slightly worse than LC-LLR algorithm (less than 0.2 dB), but MAX-LOG-LC-LLR algorithm is more preferable due to its low implementation complexity.

[0034] FIG. 4 illustrates a flowchart of a method for DCM demodulation according to one embodiment of the present invention. The method includes correcting phase rotation of received DCM signals (401); decoupling real components and imaginary components of the phase-corrected DCM signals (403); and performing log-likelihood ratio (LLR) computation separately on the real components and the imaginary components (405).

[0035] In one embodiment, an apparatus for demodulating DCM signals is provided. The apparatus may include a correcting module for correcting phase rotation of received DCM signals; a decoupling module for decoupling real components and imaginary components of the phase-corrected DCM signals; and a log-likelihood ratio (LLR) processing module for performing LLR computation separately on the real components and the imaginary components. The apparatus may further include a scaling apparatus configured to scale the result of LLR computation for process by a Viterbi decoder.

[0036] Advantageously, the present invention enjoys the below benefits: 1) no channel equalization is involved. 2) LLR is the sole method of soft metrics computation, instead of solving linear equations. 3) The LLR method decouples computation of b₀ b] from that of b₂ b₃, significantly reducing computational complexity with only 1/4 complexity of

conventional LLR, but without loss in performance.

[0037] It is to be noted that the embodiments described herein may be provided as desired, for example as software, hardware, firmware or embedded logic or otherwise. In one embodiment, depending upon the particular requirements of the application in the areas of speed, expense, tooling costs, and the like, the present invention may be implemented by constructing an application- specific integrated circuit (ASIC) having thousands of tiny integrated transistors. Such an ASIC may be implemented with CMOS (complementary metal oxide semiconductor), TTL (transistor-transistor logic), VLSI (very large systems integration), or another suitable construction. Other alternatives include a digital signal processing chip (DSP), discrete circuitry (such as resistors, capacitors, diodes, inductors, and transistors), field programmable gate array (FPGA), programmable logic array (PLA), programmable logic device (PLD), and the like.

[0038] Certain terms are used throughout the description and following claims to refer to particular system components. As one skilled in the art will appreciate, different companies may refer to a component by different names. This document does not intend to distinguish between components that differ in name but not function. In the specification and in the claims, the terms "including" and "comprising" are used in an open-ended fashion, and thus should be interpreted to mean "including, but not limited to . . . ."

[0039] Apparently, various modifications and variations may be made by those skilled in the art without departing from the spirit and scope of the present invention. Accordingly, the present invention is intended to cover these modifications and variations if these modification and variation according to the present invention fall within the scope of the claims of the present invention and equivalent thereof.

Claims

CLAIMS What is claimed is:

1. A method for demodulating Dual Carrier Modulated (DCM) signals comprising: correcting phase rotation of received DCM signals; decoupling real components and imaginary components of the phase-corrected DCM signals; and performing log-likelihood ratio (LLR) computation separately on the real components and the imaginary components.

2. The method of claim 1, wherein the step of correcting phase rotation of received DCM signals further comprises: multiplying exp(-z^'#) to the received DCM signals.

3. The method of claim 1, wherein the step of performing LLR computation further comprising: neglecting a logarithmic term in the result of LLR computation in the case of a high Signal-to-Noise Ratio (S R).

4. The method of claim 1, further comprising reusing intermediate results during LLR computation.

5. The method of claim 1, further comprising scaling the result of LLR computation for process by a decoder.

6. The method of claim 5, wherein the decoder is a Viterbi decoder.

7. An apparatus for demodulating Dual Carrier Modulated (DCM) signals, comprising: a correcting module for correcting phase rotation of received DCM signals; a decoupling module for decoupling real components and imaginary components of the phase-corrected DCM signals; and a log-likelihood ratio (LLR) processing module for performing LLR computation separately on the real components and the imaginary components.

8. The apparatus of claim 7, wherein the correcting module is further configured to multiply exp(-z^'#) to the received DCM signals.

9. The apparatus of claim 7, wherein the LLR processing module is configured to neglect a logarithmic term in the result of LLR computation in the case of a high

Signal-to-Noise Ratio (S R).

10. The apparatus of claim 7, further comprising a scaling apparatus configured to scale the result of LLR computation for process by a decoder.

11. The apparatus of claim 10, wherein the decoder is a Viterbi decoder.