Patent Application
20080239940
The present invention relates generally to communication receivers and, more specifically, to soft bit demapping in orthogonal frequency division multiplexing (OFDM) receivers.
The following disclosure will be described for a digital video broadcasting (DVB) receiver for digital terrestrial television (DTV). The concepts are equally applicable to any other channels of transmission of DTV receivers and to other receivers or standards using orthogonal frequency division multiplexing (OFDM). These may include but not be limited to wireless standards worldwide, such as wireless LAN 802.11a and g, HIPERLAN/2, Digital Audio Broadcasting (DAB), Digital Video Broadcasting Terrestrial (DVB-T), Digital Video Broadcasting for handheld (DVB-H), 802.16 Broadband Wireless Access, etc. The European terrestrial DTV standard DVB-T (ETS 300 744) is based on COFDM technologies to combat multipath fading. See ETSI EN 300 744 V.1.4.1 “Digital Video Broadcasting (DVB): Framing Structures, Channel Coding, and Modulation for Digital Terrestrial Television.”
A DVB OFDM transmitter modulates all the data-bearing subcarriers in one OFDM symbol by either QPSK, 16-QAM, 64-QAM, non-uniform 16-QAM and 64-QAM constellations.
The present method of demapping is in a receiver wherein the input signal is a) demodulated into I real and Q imaginary data which was mapped using a constellation having M bits for each I and Q data, b) demapped, c) deinterleaved and d) decoded. The method of demapping includes deriving M intermediate soft bit values yj (j-0˜M-1) for each of the I and Q data as a function of the spacing in the constellation; and limiting the range of the M values yj. A look-up table index is derived for each of the limited M values yj. A look-up table, having 2N+1 entries for supporting up to N soft bits, is indexed using the derived indices; and K soft bits (K<=N) for each data bit of the I and Q data are outputted.
The range is limited to ±1. The indices are derived by 2K−(2K−1)*yj. The look-up table includes only soft bits.
These and other aspects of the present disclosure will become apparent from the following detailed description of the disclosure, when considered in conjunction with accompanying drawings.
The N bit QAM soft demapping method will be described using the QPSK, uniform 16-QAM and 64-QAM constellations as shown in
The N bit QAM soft demapping method is performed in two steps. The first step is to transform the input complex signal into uniformly ranged intermediate soft values for each output bit. In the second step, the intermediate soft value corresponding to each bit output is used to calculate an index into a look-up table, to generate the output soft bits.
As illustrated in
The step 102 of deriving M intermediate soft bit values yj for each of the I and Q data for the constellations shown in
1. QPSK Constellation carrying 2 data bits (M=2) (
y0,q′=zi
y1,q′=zq
2. 16-QAM Constellation carrying 4 data bits (M=4) (
y0,q′=zi
y1,q′=zq
y
2,q′=abs(zi)−2
y
3,q′=abs(zq)−2
3. 64-QAM Constellation carrying 6 data bits (M=6) (
y0,q′=zi
y1,q′=zq
y
2,q′=abs(zi)−4
y
3,q′=abs(zq)−4
y
4,q′=2−abs(zi), if abs(zi)<4 abs(zi)−6, else
y
5,q′=2−abs(zq), if abs(zq)<4 abs(zq)−6, else
At the end of the first step 102, the intermediate soft values yj are limited to within ±1 range at step 104. In step 106, the intermediate soft values within ±1 range are used to generate or derive an index into a lookup table to get the final demapped soft bits output by steps 108, 110.
The following is a table that will support soft bit demapping up to 4 soft bits:
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};
Generally speaking, to support up to N soft bit demapping, the above demapper table will need to be of size 2N+1. The following operation shows how the lookup table index is generated in step 106:
Since the intermediate soft values yj are limited to within ±1, the table indices for a) K=2 soft bits per data bit are 1 to 7 which correspond to soft bits 0 to 3, b) K=3 soft bits per data bit are 1 to 15 which correspond to soft bits 0 to 7, and c) K=4 soft bits per data bit are 1 to 31 which correspond to soft bits 0 to 15.
The following 2 soft bit per data bit demapping is an example of the elements in the lookup table:
The decision boundaries are y=−⅔, 0, and ⅔, when y is on the boundary, the soft bit output is chosen to be the following values:
The resulting 7 locations are:
In the present method, each data bit is processed independently, namely for each input I/Q signal, the demapped soft bit output will be M=6 outputs for a 64-QAM constellation, for example. The individual soft bits are available prior to the deinterleaver 21. Thus, the deinterleaver will be the same for N soft bit demapping, be it N=1 for hard demapping or N=K for K soft bit per data bit demapping, since the index is set only to access the bits need for N soft bits.
Although the present disclosure has been described and illustrated in detail, it is to be clearly understood that this is done by way of illustration and example only and is not to be taken by way of limitation. The scope of the present disclosure is to be limited only by the terms of the appended claims.
