WO2007045122A1

WO2007045122A1 - A soft demodulating method for 16qam in communication system

Info

Publication number: WO2007045122A1
Application number: PCT/CN2005/001708
Authority: WO
Inventors: Pingbao Lv; Jianwu Dou
Original assignee: Zte Corporation
Priority date: 2005-10-18
Filing date: 2005-10-18
Publication date: 2007-04-26
Also published as: CA2625111A1; CN101176325B; CN101176325A; CA2625111C

Abstract

A soft demodulating method for 16-ary QAM(quadrature amplitude modulation) in a communication system can obtain average power of each constellation in 16QAM by estimating receiving power in traffic channels; And then said received intermediate frequency signal can be made carrier wave delamination, and in-phase symbol sequence information I and quadrature symbol sequence information Q can be obtained; Based on the constellation mapping relation between said input binary bits sequence and I, Q branch, different decision segment and corresponding error probability decision curve can be determined; and in different decision segment, decision can be made for obtained in-phase symbol sequence information and quadrature symbol sequence information by using corresponding decision curve, in order to obtain real value soft information sequence; Finally, resulting real value sequence should be input to decoder and made error correction decoding, reception bits sequence corresponding to transmission bits can be decoded. Present invention can combine fully the advantage of hard decision and soft decision, the algorithm is simple, and it is can be realized easily.

Description

Soft demodulation method for hexadecimal quadrature amplitude modulation in communication system

Technical field

The present invention relates to a soft demodulation method, and more particularly to a soft demodulation method for 16 Quadrature Amplitude Modulation (16QAM) in a communication system. Background technique

Adaptive Modulation and Coding (AMC) is a widely used link adaptation technology in mobile communication systems. It adaptively selects the link modulation and coding method to adapt to the fading of the link, thereby increasing the system. Capacity and the purpose of improving communication quality.

The modulation methods often used in the AMC strategy are Quadrature Phase Shift Keying (QPSK) and 16QAM. 16QAM has higher bandwidth efficiency than QPSK (twice the QPSK), but its power efficiency is lower than QPSK, that is, to achieve the same bit error rate (Bit Error Rate, hereinafter referred to as BER), Eb/No required by 16QAM (The ratio of energy per bit to noise power spectral density) is higher than QPSK. In other words, 16QAM is less easy to demodulate because the constellation point of 16QAM is denser than QPSK. The demodulation process needs to estimate the phase and estimate the amplitude. .

There are two methods for demodulating at the receiving end: hard decision demodulation and soft decision demodulation. The main idea of the former is to hardly determine the bit information corresponding to the input of the modulator when demodulating, that is, the binary bit information after hard decision is input to the decoder, and the decoder uses the known code word structure to judge The code word at the input of the encoder. Hard decisions are not a good method because the demodulator loses some information that might be used for each hard decision. Using a combination of coding and modulation, the demodulator does not pass some errors to the decoder. The demodulator is only a temporary estimate of the various symbols, often referred to as soft decisions, so that some information useful to the decoder is not lost. In general, with soft decisions, the Eb/No of the signal has a 2dB advantage over the hard decision, so most of the actual systems use soft decision mode. In QPS modulation, one symbol carries two bits of information, which are mapped to the I (in-phase) branch and the Q (orthogonal) branch respectively. When the receiving end is in soft demodulation, the in-phase part of the symbol after stripping the received carrier is included. Mapped to the I branch, the orthogonal part is mapped to the Q branch, and soft demodulation can be realized, that is, the I branch and the Q branch respectively correspond to real value information of one binary bit, and the soft demodulated real value information string is Soft decision decoding can be implemented by transmitting the conversion to the decoder. For 16QAM, soft demodulation is more complicated. This is mainly because one symbol carries four bits of information when 16QAM modulation, and two bits of information are mapped to the I branch and the Q branch respectively, and the receiving end is soft demodulated. The in-phase portion of the received carrier stripped symbol corresponds to two bit information, and the same orthogonal portion also corresponds to two bit information, and the constellation amplitude corresponding to the symbol is also different.

Figure 1 shows a basic block diagram of a typical 16QAM coded modulation/demodulation decoding. After adding a Cyclic Redundancy Check (CRC) bit, the transport block is input to the Turbo coding module (step 101) for correction. Error coding, then performing physical layer hybrid automatic repeat request HARQ (step 102), 16QAM baseband modulation (step 103), followed by spreading processing (step 104), including channelization and scrambling operations, baseband signal modulation carrier signal, The modulated signal is transmitted through the channel (step 105). After receiving the signal, the UE receiving end performs carrier stripping, separates the in-phase orthogonal signals, and then performs despreading (step 106), and the despread in-phase and orthogonal symbols are sent to the 16QAM soft-decision demodulator (step 107). Performing soft demodulation to obtain a real-valued soft information sequence ₂ corresponding to the transmitted binary bit sequence, and then undergoing physical layer de- HARQ processing (step 108), and sending it to the Turbo decoder (step 109) for error correction decoding, translation A received bit sequence corresponding to the transmitted bit sequence.

At present, in order to solve this problem, a soft decision demodulation method for calculating an input to a Turbo decoder is generally used, and the idea is to calculate the logarithm of each bit corresponding to the in-phase component and the orthogonal component of each constellation point. The likelihood ratio (hereinafter referred to as LLR), that is, the information after soft demodulation is the LLR corresponding to the input bit of the modulator. When applying this method, for some LLRs, it may be necessary to estimate the carrier-to-interference ratio (C/I), and the error of C/I may affect the performance of soft demodulation. In addition, the calculation method of LLR More complicated, hardware implementation is more difficult.

Summary of the invention

The technical problem to be solved by the present invention is to provide a hexadecimal orthogonal amplitude in a communication system. A soft demodulation method for degree modulation to provide a simple and easy to implement 16QAM soft demodulation method, thereby facilitating adaptive modulation and coding strategies.

The present invention provides a solution to the above technical problem, and provides:

Estimating the received power of the traffic channel by the pilot power and the power deviation of the traffic channel from the pilot channel, thereby obtaining the average power P _{aV of the} hexadecimal quadrature amplitude modulation constellation _(;

Performing carrier stripping on the received intermediate frequency signal to obtain in-phase symbol sequence information I and orthogonal symbol sequence information Q;

a binary bit sequence input according to the hexadecimal quadrature amplitude modulation

The constellation mapping relationship of the I and Q branches determines different decision segments and their corresponding error probability decision curves, and accordingly uses the corresponding decision curves to obtain the in-phase symbol sequence information and orthogonal symbols in different decision segments. The sequence information is judged to obtain a real value soft information sequence.

Further, the present invention may further input the obtained real value sequence to a decoder, perform error correction decoding, and decode a received bit sequence corresponding to the transmitted bit.

The soft demodulation method of the hexadecimal quadrature amplitude modulation of the invention fully combines the advantages of hard decision and soft decision, and the algorithm is simple and easy to implement.

BRIEF abstract

Figure 1 is a basic block diagram of a typical 16QAM coded modulation/demodulation decoding;

Figure 2 is a constellation diagram of QPSK and 16QAM;

Figure 3 is a map of the 16QAM I branch and Q branch map

Figure 4 is a 16QAM segmentation soft decision section hatching;

Figure 5 shows the flow of the 16QAM soft decision demodulation algorithm.

BEST MODE FOR CARRYING OUT THE INVENTION

The basic idea of the present invention is to adopt a segmented soft decision method (CSD) which combines hard decision and soft decision, and fully utilizes soft decision for higher The decision merits of the uncertainty and the hard decision can prevent the advantages of the overestimation (ovei^estimations), so that the CSD soft decision algorithm is simple and easy to implement, and the performance is better.

The present invention will be further described in detail below by taking the 16QAM soft decision in the High Speed Downlink Packet Access (HSDPA) as an example.

HSDPA is a new technology proposed by 3GPP in the R5 protocol to meet the asymmetry of uplink/downlink data services. It solves the contradiction between system coverage and capacity, greatly improves system capacity and satisfies users. High-speed business needs. Compared with R99, HSDPA uses Adaptive Modulation and Coding (AMC) and Hybrid Automatic Repeat Request (HQQ) for link adaptation.

The core algorithm of 16QAM CSD soft decision is based on the characteristics of the four-bit corresponding constellation at the input of the 16QAM modulator and the error probability decision curves of the above four bits. The soft decision solution is implemented by using the proportional segmentation method similar to QPSK baseband mapping. Tuning, that is, by dividing the different decision sections corresponding to the above (corresponding to the idea of hard decision), respectively performing soft decisions to obtain four real-valued soft bit information sequences corresponding to four bits of the input of the 16QAM modulator

H. Table 1 shows the baseband modulation mapping of 16QAM in HSDPA. When 16QAM modulation is used, the four consecutive binary symbols are first stringed into the I branch and the Q branch ^, and then mapped according to the mapping rules of Table 1. It should be noted that the average constellation power of the constellation map mapped according to Table 1 is exactly equal to 1.

Table 1

i lql i2q2 I branch Q branch

0000 0. 3162 0. 3162

0001 0. 3162 0. 9487

0010 0. 9487 0. 3162

0011 0. 9487 0. 9487 0100 0. 3162 -0. 3162

0101 0. 3162 -0. 9487

0110 0. 9487 -0. 3162

0111 0. 9487 - 0. 9487

1000 -0. 3162 0. 3162

1001 -0. 3162 0. 9487

1010 -0. 9487 0. 3162

1011 -0. 9487 0. 9487

1100 -0. 3162 - 0. 3162

1101 -0. 3162 -0. 9487

1110 -0. 9487 -0. 3162

1111 - 0· 9487 -0. 9487

Figure 2 is a constellation diagram of QPSK and 16QAM. It can be clearly seen from the constellation diagram that the constellation amplitude of QPSK is the same, but the phase is different, and the phase and amplitude of the 16QAM constellation may be different, and the constellation is denser than QPSK, thus increasing understanding. Tuning is especially the complexity of soft decision demodulation.

Figure 3 summarizes the mapping rules of Table 1. When binary 0 is used, it must be mapped to a positive real-valued signal. If ^ζ or binary 1, it must be mapped to a negative real-valued signal. The mapping of and is more complicated.

FIG. 4 further visualizes the principle of the segmentation soft demodulation algorithm on the basis of FIG. Since the mapping law of ^ and is the same, and the mapping law of the sum is the same, the principle of 16QAM CSD soft decision demodulation will be described by taking '', M乍 as an example. It can be seen from Fig. 4 that when the in-phase symbol information I is positive, the corresponding ^ should tend to be 0, and the larger I is, the greater the probability of the decision; when I is negative, the corresponding ^ should tend to judge If 1, and the smaller I is, the greater the probability of correctness of the ι decision; the same orthogonal symbol information Q is positive, the corresponding should tend to be 0, and the larger the Q, the greater the probability of the decision; When Q is negative, the corresponding trend should be judged The smaller the Q, the greater the probability of the decision; therefore, the segmentation ratio algorithm is used for soft decision, which can reflect the above trend.

For 2, the in-phase symbol information 1>0.9487 or 1<-0.9487, the corresponding trend should be judged as 1; -0.3162<1<0.3162, the corresponding should tend to be 0; a 0.9487<1<-0.3162 or 0.3162 <1 <0.9487, corresponding to a judgment tends to 0 or 1 depending on the size of the I, the I approaches zero, shellfish ^ij; 2 correct decision probability is greater 0, 1 or more approaches I 1, the probability that the Bay decision is 1 is greater; the same rule is, when the orthogonal symbol information is 0>0.9487 or (3<-0.9487, the corresponding trend should be judged as 1; when 0.3162<Q<0.3162, Corresponding should tend to be 0; - 0.9487 <Q<-0.3162 or 0.3162<Q<0.9487, the corresponding trend tends to be 0 or 1 depending on the size of Q, and Q becomes closer to 0, then the decision is 0. The greater the probability of correctness, the closer Q is to 1 or a 1, the greater the probability of a decision being 1 is greater.

Therefore, the algorithm of the present invention performs a soft decision corresponding to a hard decision scale algorithm, which can reflect the above trend. In the soft decision, different soft segments are used to demodulate the corresponding soft information ₂ corresponding to different segments. However, it is found through analysis that the segment soft decision demodulation formula can be combined to obtain the corresponding FIG. Soft decision demodulation algorithm. Figure 5 corresponds to the in-phase or orthogonal division of QPSK baseband modulation with an average constellation power of 1 in the soft-decision demodulation formula of 0· ⁷⁰ .

The detailed flow of the 16QAM CSD soft decision demodulation algorithm of the present invention will be described below as shown in FIG.

Step 501: The receiving end of the UE estimates the received power of the traffic channel by using the pilot power and the power deviation of the traffic channel and the pilot channel, so as to obtain the average power of each point of the 16QAM constellation; Step 502: The I branch and the Q branch symbol information are obtained. Leave

Should, ^

Step 504: Determine the positive or negative of I, if 1 ^ 0, then go to step 505, if 1 < 0, then go to step 506; Step 505: According to the formula: Step 506: According to the formula:

Step 507: Determine the positive or negative of Q. If Q 0, go to step 508. If Q < 0, go to step 509; Step 508: Press the formula: Step 509: Press formula:

Calculating step 510: merging the soft demodulated Ί, ^, and sum into a real value sequence corresponding to the bit sequence W input to the 16QAM modulator;

Step 511: The real value sequence is input to the decoder, and error correction decoding is performed, and the received bit sequence corresponding to the transmission bit is decoded.

Claims

Claim

A soft demodulation method for hexadecimal quadrature amplitude modulation, which performs soft decision demodulation on a binary bit sequence modulated by a hexadecimal quadrature amplitude modulation at a transmitting end to obtain a corresponding real value soft information sequence ₂ The method includes the following steps: estimating the received power of the traffic channel by using pilot power and power deviation of the traffic channel and the pilot channel, thereby obtaining an average power Pave of the hexadecimal quadrature amplitude modulation constellation;

According to the hexadecimal quadrature amplitude modulation input binary bit sequence and I,

The constellation mapping relationship of the Q branch determines different decision segments and their corresponding error probability decision curves, and accordingly uses the corresponding decision curves to obtain the in-phase symbol sequence information and the orthogonal symbol sequence information in different decision segments. A decision is made to obtain a real value soft information sequence ^2.

2. The method of claim 1, further comprising:

The obtained real value sequence is input to a decoder, error correction decoding is performed, and a received bit sequence corresponding to the transmission bit is decoded.

3. The method according to claim 1, wherein said hexadecimal quadrature amplitude modulation is hexadecimal quadrature amplitude modulation in a high speed downlink packet access system; said determining according to a mapping relationship The different decision segments and their corresponding error probability decision curves are:

When the in-phase symbol information I is positive, the corresponding ^ should tend to be 0, and the larger the I, the greater the probability of the decision; when I is negative, the corresponding ^ should tend to be 1, and the smaller the I ,

^The greater the probability of a ι decision;

When the orthogonal symbol information Q is positive, the corresponding trend should tend to be 0, and the larger the Q, the greater the probability of the decision; when Q is negative, the corresponding trend should be judged as 1, and the smaller the Q, The probability of a correct judgment is greater;

When the in-phase symbol information 1>0. 9487 or 1<0. 9487, the corresponding trend should be judged as 1; -0. 3162<1<0. 3162, the corresponding ^ should tend to be 0; -0 9487 <1< -0. 3162 or 0. 3162<1<0. 9487, the corresponding trend is judged to be 0 or 1 Depending on the size of I, the closer I is to 0, the greater the probability that B is judged to be 0. The closer I is to 1 or 1 , the greater the probability that B will be 1;

When the orthogonal symbol information Q>0. 9487 or Q<—0. 9487, the corresponding trend should be judged as 1; -0. 3162<Q<0. 3162, the corresponding trend should be judged as 0; 9487 <Q< -0. 3162 or 0. 3162<Q<0. 9487, the corresponding ^ tends to judge whether it is 0 or 1. Depending on the size of Q, the closer Q is to 0, then the judgment is 0. The greater the probability, the closer Q is to 1 or a 1, the greater the probability of a decision being 1 is greater.

4. The method according to claim 1, wherein said hexadecimal quadrature amplitude modulation is hexadecimal quadrature amplitude modulation in a high speed downlink packet access system; said determining according to a mapping relationship The different decision segments and their corresponding error probability decision curves are:

Regardless of the value of I, the error probability decision curve of ^f i should satisfy the formula:

Regardless of the value of Q, the error probability decision curve should satisfy the formula:

^, =2*0 7071 /(^* 0 3162)

At 1 ^0, the error probability decision curve of ^ should satisfy the formula:

ί

_{1 2} d. 3162

h 0 7071

, 0 9487 - 0 3162 ₎

When 1<0, the error probability decision curve should satisfy the formula:

At Q 0, the error probability decision curve should satisfy the formula:

When Q<0, the error probability decision curve should satisfy the formula:

The method according to claim 4, wherein the value 0. 7071 corresponds to an in-phase or quadrature component of quadrature phase shift keying baseband modulation with an average constellation power of 1.