WO2012000277A1

WO2012000277A1 - Decision-feedback device and method for receiver

Info

Publication number: WO2012000277A1
Application number: PCT/CN2010/078886
Authority: WO
Inventors: 张骏凌
Original assignee: 中兴通讯股份有限公司
Priority date: 2010-06-29
Filing date: 2010-11-18
Publication date: 2012-01-05
Also published as: CN102316057A; CN102316057B

Abstract

A decision-feedback device and method for a receiver are disclosed by the present invention, wherein the device includes: a reception module (102), which is used for receiving the demodulation output of a front-end demodulation element; a weighting function decision module (104), which is used for making decision for a weighting function of which constellation point weightings are the probabilities of equalized equivalent complex additive white Gaussian noises (AWGNs) according to the demodulation output and based on a hard-decision method; and a feedback module (106), which is used for feeding the decision result of the weighting function back to the front-end demodulation element. With the present invention, the defect that faults fed back to a channel estimation device or an interference elimination device are enlarged and spread due to hard-decision faults is overcome, thus a balance between the performance and efficiency of the communication system is realized.

Description

TECHNICAL FIELD The present invention relates to the field of communications, and in particular to a device and method for receiver decision-feedback. BACKGROUND OF THE INVENTION In the field of communications, receivers typically perform channel estimation. At present, receiver decision-response channel estimation can generally be divided into three categories: pilot-based decision-feedback; joint channel decoder iterative decision-feedback; hard decision-feedback. In the "pilot-based decision-feedback" method, the receiver is fully aware of the pilot sequence. A more common application is that the receiver uses the pilot sequence to adaptively estimate the channel, and obtains the channel estimate before demodulating. Or interference cancellation, etc., such as the receiver of the US ATSC standard. The advantage of this scheme is that the performance of the algorithm is better due to the existence of the pilot sequence. The weak point is that the insertion of the pilot sequence will reduce the system efficiency. In addition, the lower pilot insertion rate will result in the system not being able to move in the high-speed mobile channel environment. Work effectively. The second "joint channel decoder iterative decision-feedback, the method can be regarded as an improvement of the first method. Since the current channel decoders, such as viterbi, turbo, RS and other decoders have larger decoding gains, This scheme utilizes this decoding gain to re-transmit the decoded data through the transmitting device as the pilot in the first method to the front-end channel estimation or interference cancellation unit to enhance the performance of the channel estimation or interference cancellation algorithm. The advantage of this algorithm is that the pilot sequence in the first method can be completely eliminated, and the performance equivalent to the "pilot-based decision-feedback" method is achieved, thereby improving system efficiency, and its disadvantages are channel decoder delays. And the transmitting device requires higher hardware complexity for feedback. The third "hard decision-feedback" method is a simplification of the "joint channel decoder iterative decision-feedback" method. This method uses the hard decision method instead of "joint channel decoding." The iterative decision-feedback "channel decoder output in the method, thereby eliminating the higher hardware complexity required for demodulation-feedback. But the performance of this method is three The worst of the methods, in practical applications, generally requires a greater number of iterations to achieve performance improvements. In the related art, the receiver decision-feedback method has the following drawbacks: The feedback to the channel estimation or the error expansion and propagation in the interference canceling device due to the hard decision error is poor. In response to the above problems, there is currently no solution. SUMMARY OF THE INVENTION A primary object of the present invention is to provide a receiver decision-feedback method and apparatus for solving feedback to channel estimation or interference cancellation due to a hard decision error in the above-described conventional receiver decision-feedback mode. Errors in the device that expand and spread, and poor performance. According to an aspect of the present invention, a receiver decision-feedback apparatus is provided, including: a receiving module, configured to receive a demodulated output of a front end demodulation device; a weight function decision module, configured to The decision method determines the probability of the equal-additional additive white Gaussian noise AWGN as the weight function of the constellation point weight; the feedback module is configured to feed back the decision result of the weight function to the front-end demodulation device. Preferably, the weight function decision module comprises: a second receiving point module, configured to confirm that the receiving point after the equalization at time t is ^) = () + «. ), where jc(t)e{C0,Cl, ......,C«} is the emission constellation point,

2

«. (t) is the equivalent complex AWGN after equalization, both the real part and the imaginary part satisfy the N(0, ) distribution, where n is the constellation point number; the second distance metric module is used to obtain the distance metric di = i = 0,1, 2, ......, η; second decision module, used to obtain the weight function decision result

x _p (t) = ∑Ci-p(r _eq (t)\Ci), Cie{C0, Cl, ..., C"} , where Cie{C0, Cl, ..., Cw}.

Preferably, the apparatus further comprises: a hard decision module, configured to perform a hard decision of the constellation transmission point that minimizes the distance metric according to the demodulation output; an adaptive module, configured to perform an adaptive selection according to the preset decision threshold The function decision module or the hard decision module is executed; the feedback module is further configured to feed back the result of the weight function decision or the hard decision to the front end demodulation device according to the selection result of the adaptive module. Preferably, the adaptive module is configured to: when the distance metric is less than the decision threshold, the selection weight function decision module performs; otherwise, the hard decision module is selected to perform. Preferably, the hard decision module includes: a first receiving point module, configured to confirm that the receiving point after the equalization at time t is 0 e o, ci, ..., c"} is a transmitting constellation point, ". (0 is the equalization complex AWGN after equalization; the first distance metric module is used to obtain the distance metric di = r _eq (t) - Ci , i = 0, l, 2, ... '"; the first decision module, For obtaining the hard decision result x _H (t) = Ci I min(i//'), Ci e {CO, CI, , Cn) Preferably, the device is applied to a quadrature phase shift keying QPSK modulation system, 16 In a quadrature amplitude modulation Q AM system, or a 64Q AM modulation system. Preferably, the apparatus, when applied to the QPSK modulation system, the n=3; or when applied to the 16QAM modulation system, The n=15; or when applied to the 64QAM modulation system, the n=63. Preferably, the front end demodulation device is a channel estimation device or a interference cancellation device. According to another aspect of the present invention A method for receiver decision-feedback is provided, comprising: receiving a demodulation output of a front-end demodulation device; and based on a demodulation output, based on a hard decision method, the probability of equalizing the equivalent complex AWGN as a weight of a constellation point The function performs a decision; the decision result of the weight function is fed back to the front-end demodulation device. Preferably, the method can Applied to QPSK modulation systems, 16 quadrature amplitude modulation QAM systems, or 64QAM modulation systems. When applied to QPSK modulation systems, n=3; or when applied to 16QAM modulation systems, n=15; or when applied In the 64QAM modulation system, n=63. By the present invention, the "hard decision-feedback" method combining the equivalent AWGN probability after equalization is used to solve the problem that the existing receiver decision-feedback mode is caused by a hard decision error. Feedback to the channel estimation or interference cancellation device in the problem of error expansion and propagation, poor performance, and thus achieve a balance between communication system performance and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are set to illustrate,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram of a receiver decision-feedback apparatus according to an embodiment of the apparatus of the present invention; FIG. 2 is a schematic diagram of a receiving point in a receiver-response apparatus of a receiver according to an embodiment of the present invention; EMBODIMENT OF THE INVENTION Embodiment 4 is a schematic diagram of a device for receiver decision-feedback in an Orthogonal Frequency Division Multiplexing (OFDM) receiver; FIG. 4 is a device for receiving a receiver decision-feedback according to Embodiment 5 of the present invention. FIG. 5 is a flowchart of a receiver decision-response method according to an embodiment of a method of the present invention; FIG. 6 is a flowchart of a receiver decision-feedback method according to an embodiment of the method of the present invention; . BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict. The invention provides a device and method for receiver decision-feedback. Based on the traditional hard decision method, the 4 rate of the additive additive white Gaussian noise (AWGN) is constructed. The weight function decision of the constellation point weight. After equalization, the equivalent complex AWGN is the equivalent complex AWGN after equalization, wherein AWGN includes real and imaginary parts that are uncorrelated. The present invention, in combination with a front end demodulation device, can prevent feedback to channel estimation or error propagation and propagation in the interference cancellation device due to hard decision errors. Apparatus Embodiment 1: FIG. 1 is a schematic diagram of a receiver decision-feedback apparatus according to an embodiment of the apparatus of the present invention. The connection relationship in this embodiment is indicated by a solid line in FIG. As shown in FIG. 1, the embodiment includes: a receiving module 102, configured to receive a demodulation output of a front end demodulation device; a weight function decision module 104, The receiving module 102 is connected to determine, according to the demodulation output, the probability of the equalized additive additive white Gaussian noise AWGN as a weight function of the constellation point weight based on the hard decision method; the feedback module 106, and the weight function decision module 104 is connected, used to feed back the decision result of the weight function to the front-end demodulation device. This embodiment can be applied to a Quadrature Phase Shift Keying (QPSK) modulation system, a 16 Quadrature Amplitude Modulation (QAM) system, and a 64QAM modulation system. This embodiment is based on the traditional hard decision-feedback method, and uses the probability of the equalized complex AWGN as the weight function decision of the constellation point weight, and overcomes the feedback to the channel estimation or the interference cancellation device due to the hard decision error. The defects of error expansion and propagation, thus achieving a balance between communication system performance and efficiency. Apparatus Embodiment 2: This embodiment will further explain the receiver decision-feedback apparatus based on the apparatus embodiment 1. As shown in Fig. 1, the connection relationship of this embodiment is indicated by a broken line. The receiver decision-response device may further include: a hard decision module 108 connected to the receiving module 102 for performing a demodulation output, performing a hard decision of a constellation transmission point that minimizes the distance metric; the adaptive module 110, and the weight function decision The module 104 is connected to the hard decision module 108 for use in a preset decision threshold, an adaptive selection function decision module 104 or a hard decision module 108. The feedback module 106 is connected to the adaptive module and is also used according to the adaptive module. As a result of the selection, the result of the weight function decision or the hard decision is fed back to the front end demodulation device. In this embodiment, the combination of the weight function decision and the hard decision combines the higher execution efficiency of the conventional receiver decision-response device, and avoids the error hard feedback feedback to the channel estimation or the interference cancellation module. The error is expanded and spread, thereby improving system performance. Device Embodiment 3: This embodiment will specifically describe the hard decision method and the weight function decision method in the device embodiment 1 and the device embodiment 2. 2 is a schematic diagram of a receiving point in a receiver-response device of a receiver according to an embodiment of the present invention; Figure. The weight function decision module specifically includes: a second receiving point module, a second distance measuring module, and a second determining module. The hard decision module specifically includes: a first receiving point module, a first distance measuring module, and a first determining module. The second receiving point module in the weight function decision module, the second distance measuring module and the first receiving point module in the hard decision module, the first distance measuring module has the same calculation process, the same function, and preferably takes a set Just fine. In this embodiment, the QPSK modulation system will be mainly described as an example. As shown in Figure 2: In the first (b) receiving point module, the normalized QPSK constellation points used by the transmitter are C0, Cl, C2, and C3. At the receiving end, the receiving point after equalization at a certain time t is: r _eq {t) = x{t) + n ₀ {t) (Formula 1) jt) e{C0, Cl, C2, C3 in the above formula } for launching constellation points, ". (t) is the equivalent complex AWGN after equalization, both the real part and the imaginary part satisfy the distribution Ν(0, ^-). In the first (b) distance metric module, the distance metric is calculated using the traditional hard decision method: di = i = 0, 1, 2, 3

(Formula 2) In the first decision module, the hard decision that α is the minimum distance α is x(t): x _H (t) = Ci\mm(di\ C/e {C0, C1, C2, C3} (Formula 3) In Equation 3, 3⁄4y(t) is the smallest Ci. As shown in Figure 2, d0 is the minimum distance, then x _H (t) = CO. In the second embodiment, the hard decision module 108 can Use the hard decision function as shown in Equation 3. In the second decision module, consider "in the formula 1." (When 0 is large, then x(t) is more likely to be non-CO, as assumed x(t) = C3 , then the hard decision = CO feedback to the channel estimation or the interference cancellation algorithm is error expansion and propagation, resulting in faster performance degradation, which is the main reason for the poor performance of the "hard decision-feedback" method. Consider the conditional probability: Cie{C0, Cl, C2, C3} (Equation 4)

Constructing a weight function decision based on the conditional probability - Equation 4: x _p (t) = ^α·ρ (r _eq (t) I Ci), Ci e {CO, I, 2, 3} (Equation 5) Weight function module 104 A weight function decision as shown in Equation 6 can be used. For traditional hard decisions such as those shown in Equation 3, the probability-based weight function decision avoids the error expansion and propagation caused by the error hard decision feedback to the channel estimation or interference cancellation module, thereby improving system performance. On the other hand, if the hard decision method as shown in Equation 3 is correct, unnecessary noise introduced by the weight function decision as shown in Equation 5 can be avoided. From the perspective of further improving system performance, the adaptive module 110 can adaptively use hard decisions and weight function decisions.

Cie {C0, Cl, C2, C3} (Equation 6)

In addition, the present invention is also applicable to a 16QAM or 64QAM modulation system, and only needs to increase the number of constellation points in the above steps. For the 16QAM modulation system, the constellation points are taken as 15. For the 64QAM modulation system, the number of constellation points is 63, and the other descriptions are unchanged, and should also be included in the protection scope of the present invention. In this embodiment, the hard decision method and the weight function decision method in the device embodiment 1 and the device embodiment 2 are described in detail, and all the beneficial effects of the foregoing embodiment are obtained, and the implementation is stronger, and the jt匕 is no longer heavy. Said. Apparatus Embodiment 4: FIG. 3 is a schematic diagram of an apparatus for receiving a receiver decision-feedback according to Embodiment 4 of the present invention in an OFDM receiver. As shown in Figure 3, the receiver without decision feedback uses only "pilot-assisted channel estimation" and "first equalizer". As shown in Figure 3, the receiver using hard decision-feedback channel estimation uses the "first equalizer" The output performs a hard decision output f _ff (t) as shown in Equation 3 as a feedback symbol to the second equalizer. In this case, the receiver uses "pilot + feedback symbol assisted channel estimation". If the hard decision f _ff (t) is correct, "pilot + feedback symbol assisted channel estimation" relative to "pilot-assisted channel estimation" can be regarded as pilot due to all hard decision feedback symbols, and generally by data The number of hard decision symbols formed is much larger than the number of pilots and thus the quality of the channel estimate is better than that of "pilot-assisted channel estimation". But once the hard decision is wrong, as above, the misjudgment of the hard decision f _ff (t) will cause the channel estimation error formed by its feedback to propagate and amplify through the "second equalizer", when the signal to noise ratio is low. The phenomenon is particularly serious. In order to avoid this error propagation and expansion caused by hard decision errors, as shown in FIG. 3, the output of the "first equalizer" is further subjected to the weight function decision as shown in Equation 5, completely avoiding the error of the hard decision, and then using Using the adaptive module of Equation 6, the threshold selection is performed to form a feedback symbol, and the gain of the hard decision and the error of the weight function decision are simultaneously obtained according to different signal-to-noise ratios on different symbols, thereby being used in the "pilot + feedback symbol-assisted channel" Estimate "obtain error-free propagation and gain channel estimation. Device Embodiment 5: FIG. 4 is a schematic diagram of an apparatus for receiving a receiver decision-feedback according to Embodiment 5 of the present invention in an OFDM interference cancellation receiver. As shown in Figure 4, "weight function decision-feedback channel estimation" and "first equalizer" form a receive useful signal path. The "pilot assisted channel estimation", the second equalizer, the weight function decision module, the hard decision module, the adaptive module, and the interference symbol reconstruction module form a interference signal reconstruction path. On the interference reconstruction path, if the hard decision is correct, the output quality of "equalization 1" is improved by "reconstructing the interference symbol, and eliminating the influence of the interference signal in the received signal; if the hard decision is wrong, then Similarly, the interference symbol reconstruction module propagates to the "weight function decision-feedback channel estimation" and the first equalizer and amplifies the error in the first equalizer. As before, the weight function decision is introduced on the interference signal reconstruction path. The "and" threshold selection forms a feedback symbol "to simultaneously obtain gains obtained from accurate hard decisions of the interference symbols and to eliminate error propagation and expansion caused by hard decisions. Method Embodiment 1: FIG. 5 is a receiver according to an embodiment of the method of the present invention. A flow chart of the decision-response method. As shown in FIG. 5, this embodiment includes: Step S502, receiving a demodulation output of the front-end demodulation device. Step S504, performing a demodulation output, and determining, according to a hard decision method, a weight ratio function of the equal-respective complex AWGN as a weight function of the constellation point weight; step S506, The decision result of the weight function is fed back to the front end demodulation device. This embodiment is based on the traditional hard decision-feedback method, and uses the probability of equalizing the complex complex AWGN as the weight function decision of the constellation point weight, and overcomes the feedback to the channel estimation or the interference canceling device due to the hard decision error. The defects of error expansion and propagation, thus achieving a balance between communication system performance and efficiency. Method Embodiment 2: FIG. 6 is a flowchart of a method for receiver decision-feedback according to Embodiment 2 of the method of the present invention. As shown in FIG. 6, the embodiment includes: Step S602: Receive a demodulation output of a front-end demodulation device; Step S604, perform a demodulation output, and based on a hard decision method, use 4 rates of the equal-respective complex AWGN as a constellation The weight function of the point weight is determined; step S606, according to the demodulation output, performing a hard decision of the constellation transmission point that minimizes the distance metric; step S608, performing pre-calculation, according to the preset decision threshold, the adaptive selection weight function decision Or a hard decision; Step S610, feeding back the result of the weight function decision or the hard decision to the front end demodulation device according to the selection result. This embodiment can be applied to a QPSK modulation system, a 16QAM modulation system, and a 64QAM modulation system. When applied to a QPSK modulation system, the number of constellation points is 3; or when applied to a 16QAM modulation system, the number of constellation points is 15; or when applied to a 64QAM modulation system, the number of constellation points is 63. For a detailed description of the embodiment, reference may be made to the second embodiment of the device, and the specific method of the hard decision and the weight function decision may be referred to the third embodiment. For the application in the OFDM receiver and the OFDM processor, reference may be made to the fourth embodiment. And the fifth embodiment, and have all the beneficial effects of the above embodiments, It is no longer repeated. Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device, which can be concentrated on a single computing device or distributed over a network composed of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device, such that they may be stored in the storage device by the computing device and, in some cases, may be different from the order herein. The steps shown or described are performed, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps are fabricated as a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software. The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the scope of the present invention are intended to be included within the scope of the present invention.

Claims

A device for receiver decision-feedback, characterized in that it comprises:

a receiving module, configured to receive a demodulated output of the front end demodulation device;

a weight function decision module, configured to determine, according to the demodulation output, a probability of the equal-additional additive white Gaussian noise AWGN as a weight function of the constellation point weight based on the hard decision method;

And a feedback module, configured to feed back a decision result of the weight function to the front end demodulation device. The apparatus according to claim 1, wherein the weight function decision module comprises: a second receiving point module, configured to confirm that the receiving point after the equalization at time t is r _eq (t) = x(t) + n ₀ (t), where x(t)e{C0,Cl, ......,C«} is the emission constellation point, ". (t) is the equivalent complex AWGN after equalization, both the real part and the imaginary part satisfy the N(0, ) distribution, where n is the number of constellation points;

a second distance metric module, configured to obtain a distance metric di=r _eq (t)-Ci , = 0,1,2, . . . , rr; a second decision module, configured to obtain a weight function decision result x _p (t) =∑Ci-p(r _eq (t)\Ci), Cie{C0,Cl,...,C"} , where Cie{C0,Cl,..., Cw}.

The device according to claim 1, further comprising:

a hard decision module, configured to perform a hard decision on a constellation transmission point that minimizes a distance metric according to the demodulated output;

An adaptive module, configured to adaptively select the weight function decision module or the hard decision module to execute according to a preset decision threshold;

The feedback module is further configured to feed back the result of the weight function decision or the hard decision to the front end demodulation device according to the selection result of the adaptive module. 4. Apparatus according to claim 3 wherein:

The adaptive module is configured to: when the distance metric is less than the decision threshold, select the weight function decision module to execute; otherwise, select the hard decision module to execute. The apparatus according to claim 4, wherein the hard decision module comprises: a first receiving point module, configured to confirm that the receiving point after equalization at time t is r _eq (t) = x(t) + n ₀ (t) , where x(t) e {C0, Cl, ...... , C«} is the emission constellation point, ". (t) is an equivalent complex AWGN after equalization;

a first distance metric module for obtaining a distance metric di = r _eq (t) - Ci , = 0, 1, 2, ..., _{n a} first decision module for obtaining a hard decision result x _H (t) = Ci I min(i//'), Ci e {CO, CI, , Cn).

6. Apparatus according to any one of claims 2-5, characterized by: being applied to a quadrature phase shift keying QPSK modulation system, a 16 quadrature amplitude modulation Q AM system, or a 64Q AM modulation system. The device of claim 6 wherein:

When applied to the QPSK modulation system, the n=3; or

When applied to the 16QAM modulation system, the n=15; or

When applied to the 64QAM modulation system, the n = 63. The apparatus according to any one of claims 1 to 5, wherein: said front end demodulator is a channel estimation device or a interference canceling device.

A receiver decision-feedback method, comprising:

Receiving a demodulated output of the front end demodulation device;

Determining, according to the demodulation output, a weighted function of the equalized additive additive white Gaussian noise AWGN as a weight function of the constellation point weight based on the hard decision method;

The decision result of the weight function is fed back to the front end demodulation device.

10. The method according to claim 9, wherein the constructing the weighted function of the equalized complex AWGN as the weight function of the constellation point weight based on the hard decision method comprises: Confirm that the reception point after the equalization at time t is r _eg (t) = x{t) + n ₀ {t), where x(t)e{C0,Cl, ......,C«} is Launch the constellation point, ". (t) is the equivalent complex AWGN after equalization, in fact

2

Both the imaginary and imaginary parts satisfy the N(0, ) distribution, where n is the number of constellation points; the acquisition distance metric· = / = 0,1,2, ..., _n

Get the weight function decision result) = Ο··ρ(Γ )|Ο· C/e{C0,Cl, ......,Cn) , where Cie{C0,Cl,...,Cw}.

The method according to claim 9, wherein the receiving the demodulated output further comprises:

Determining, according to the demodulation output, a hard decision of a constellation transmission point that minimizes a distance metric; adaptively selecting the weight function decision or the hard decision according to a preset decision threshold; The result of the weighting function decision or the hard decision is fed back to the front end demodulation device.

12. Method according to claim 10 or 11, characterized in that:

When applied to the quadrature phase shift keying QPSK modulation system, the η=3; or when applied to the 16 quadrature amplitude modulation QAM modulation system, the η=15; or

When applied to the 64QAM modulation system, the η = 63.