CN103888399A

CN103888399A - Quadrature amplitude modulation soft de-mapping method

Info

Publication number: CN103888399A
Application number: CN201210560341.5A
Authority: CN
Inventors: 汤武惊; 郭江波; 罗笑南; 杨艾琳; 刘海亮; 吴超如; 李圳龙
Original assignee: Shenzhen Research Institute of Sun Yat Sen University
Current assignee: Shenzhen Research Institute of Sun Yat Sen University
Priority date: 2012-12-20
Filing date: 2012-12-20
Publication date: 2014-06-25

Abstract

The invention provides a quadrature amplitude modulation soft de-mapping method which is a de-mapping method used for a receiver and used for deducing constellation interval by utilizing M middle soft bit values yj of an I component and a Q component of an input signal. For signals yj supporting N soft bit outputs to the maximum degree, look-up table index containing 2<(N+1)> items can be established through M values within the limitation range and by utilizing derivation index; and an I component and a Q component of an output signal can be deduced through the look-up table index and M soft bit data yj of each signal.

Description

The soft de-mapping method of quadrature amplitude modulation

Technical field

The present invention relates to general significance Shangdi communication sink, and the soft bit further relating in orthogonal frequency division multiplexi (OFDM) receiver goes mapping.

Below statement will be described the digital video broadcast ing (DVB) for receiving terrestrial DTV (DTV).Its concept is transmission and other receivers that are equally applicable to any DTV receiver

Or orthogonal frequency division multiplexi (OFDM) use standard.

Include but are not limited to and comprise global radio standard.As WLAN (wireless local area network) 802.lla and g, HiperLAN/2, digital audio broadcasting (DAB), digital video broadcast-terrestrial (DVB-T), hand-held digital video broadcast (DVB-H), 802.16 broadband wireless accesss.Etc. European terrestrial DTV standard DVB-T (ETS 300744) be the technology based on COFDM to anti-multipath fading.Refer to ETSI EN300744V1.4.1 digital video broadcasting (DVB): frame structure, the digital terrestrial television of Channel Coding and Modulation.“

Fig. 1 has described a typical DVBT receiver block diagram.

The Digital Signal Processing of DVBT receiver can be divided into three parts.Part I 10 comprises: RF front end 12, ND transducer 14,

ofdm demodulator

16,18 are separated mediation TPS decoder 19.

The signal processing part of this receiver front end is carried out receiver training, comprises that vanous is synchronous, channel guess value and OFDM demodulation, and OFDM demodulation comprises QAM de-mapping device.Part II 20 is signal processing blocks of DVBT receiver rear end, deinterleaver 21 and 22 Viterbi decoders in DVBT uses.Outer channel decoding is used outer deinterleaver 24, RS decoder 26 and deinterleaver 28.Part III 30 is mpeg decoders.

DVB OFDM transmitter is modulated the arbitrary QPSK in an OFDM symbol of all Deta bearer subcarriers, 16-QAM, 64-QAM's, non-homogeneous 16-QAM and 64-QAM planisphere.Fig. 2 A-2C has expressed respectively QPSK, evenly 16-QAM and 64-QAM planisphere.In the OFDM receiver shown in Fig. 1, Deta bearer subcarrier will first pass through channel correction, QAM demapping and interior deinterleaver before entering Viterbi decoder.Soft QAM demapping be can provide with soft input bit, the performance that makes Viterbi decoder is being shown and is being better than hard bit input viterbi decoder.But the complexity of operation is very important, such as the soft demapping of high-order constellation of 64-QAM.The complexity of its operation is directly proportional to the number of the soft bit of required Viterbi decoder.

Marginal data

Fig. 1 is the block diagram of the digital video broadcast terrestrial receiver of prior art.

Fig. 2 A-C is respectively QPSK in prior art, evenly 16-QAM with 64-QAM planisphere.

Fig. 3 is the demapping process in this patent.

Specifically describe

As shown in Figure 2, QAM soft de-mapping method in N position uses QPSK, evenly 16-QAM and 64-QAM planisphere.The method is equally applicable to other qam constellation figure, as QAM demapping heterogeneous.

The soft de-mapping method of QAM of N position comprises two steps, first step is to convert the sophisticated signal of input to evenly distributed centre soft value, in second step, corresponding to the soft value in centre of the output of each, calculate one of generation and index certain look-up table, generate the soft bit of output.

In the de-mapping method shown in Fig. 3, comprise the middle soft place value yi of M that step 102 derives, from step 100, the I/Q value of input is in order to set up the distance function in planisphere.M value yi is defined in step 104, and in step 106, each look-up table index is derived by M value yi.Each look-up table has 2N+1 entry, uses derivative index and the soft position of K (K <=N) in step 108, supports the nearly index of the soft position of N,

Step 102 is to be the process of data in planisphere to I and the soft place value yi of Q derivation M, and as shown in Fig. 2 A-C, the sophisticated signal of input is:

1:QSPK planisphere carrying 2 data bit (M=2) (Fig. 2 A):

y _0，q′＝z _i

y _1，q′＝z _q

2:16-QAM planisphere carrying 4 data bit (M=4) (Fig. 2 B):

y _0，q′＝z _i

y _1，q′＝z _q

y _2，q′＝abs(z _i)-2

y _3，q′＝abs(z _q)-2

3:64-QAM planisphere carrying 6 data bit (M=6) (Fig. 2 C):

y _2，q′＝abs(z _i)-4

y _3，q′＝abs(z _q)-4

y _0，q′＝z _i

y _4，q′＝2-abs(z _i)，if abs(z _i)＜4abs(z _i)-6，

y _1，q′＝z _q y _5，q′＝2-abs(z _q)，if abs(zq ₎＜4abs(z _q)-6，

In step 102 later stage, middleware software value yj by step 104+1 territory limits.In step 106, the index value that the soft value yj in centre being defined is used for deriving in look-up table draws the final soft bit of mapping that connects in step 108,110.

Below for supporting the table of 4 soft bit demappings:

QuanTbl[32]＝{0，0，1，1，2，2，3，3，4，4，5，5，6.，6，7，7，8，8，9，9，10，10，11，11，12，12，13，13，14，14，15，15}；

It has been generally acknowledged that, if support the demapping of the soft bit in N position, demapping table needs 2N+1 position.Following formula has illustrated the method that generates look-up table index in step 106:

4-3*yi, q ', i=0,1,2,3,4,5, each data bit is used 2 soft bits;

8-7*yi, q ', i=0,1,2,3,4,5, each data bit is used 3 soft bits;

16-15*yi, q ', i=0,1,2,3,4,5, each data bit is used 4 soft bits;

2N-(2N-1) * yi, q ', i=0,1,2,3,4,5, each data bit is used N soft bit.

Be below the soft bit de-mapping method of every data bit 2 example:

Y=-1, soft bit is output as=and 3;

Y=-1/3, soft bit is output as=and 2;

Y=1/3, soft bit is output as=and 1;

Y=-1, soft bit is output as=and 0;

As decision boundary y=-2/3,0,2/3 o'clock, y was on border, and the soft bit values of output is selected as follows:

Y=-2/3, soft bit is output as=and 3;

Y=0, soft bit is output as=and 2;

Y=2/3, soft bit is output as=and 1;

Claims

1. for the mapping method that goes of receiver, for input signal, a) use M centre soft bit values demodulation I component of signal and Q component data to b) demapping c) deinterleaving d) decode, this de-mapping method comprises:

For planisphere distance function I and Q derive M in the middle of soft bit values yj;

Limit the scope of M value yj;

For each M value yj derives look-up table index value;

Use derivative index can set up the look-up table index of (N+1) power entry that contains 2 by the M in limited field value.

2. in the method for statement 1, index is wherein border and the location in planisphere.

3. in the method for statement 1, look-up table wherein only comprises soft bit.