Chapter 9
BICM Receivers for Trellis Codes

In this chapter we study a trellis encoder and a bit-interleaved coded modulation (BICM) receiver, i.e., the coded modulation (CM) encoder is a serial concatenation of a convolutional encoder (CENC) and a memoryless mapper, and the bitwise interleaver is not present. This in fact corresponds to the model in Fig. 2.4, where the maximum likelihood (ML) decoder is replaced by a BICM decoder. The study of this somehow unusual combination is motivated by numerical results that show that for the additive white Gaussian noise (AWGN) channel and convolutional encoders (CENCs), removing the interleaver in a BICM transceiver results in improved performance.

This chapter is organized as follows. In Section 9.1 we introduce the idea of BICM with “trivial interleavers” (i.e., no interleaving) and in Section 9.2 we study the optimal selection of CENCs for this configuration. Motivated by the results in Section 9.2, we study binary labelings for trellis encoders in Section 9.3.

9.1 BICM with Trivial Interleavers

The design philosophies behind trellis-coded modulation (TCM) and BICM for the AWGN channel are quite different. On one hand, TCM is constructed coupling together a CENC and a constellation using a set-partitioning (SP) labeling. On the other hand, BICM is typically a concatenation of a binary encoder, a bit-level interleaver, and a mapper with constellation labeled by the binary reflected Gray code (BRGC). The BRGC is often used in BICM because it maximizes the bit-interleaved coded modulation generalized mutual information (BICM-GMI) for medium and high signal-to-noise ratios, as we showed in Section 4.3. In TCM, the selection of the CENC is done so that the minimum Euclidean distance (MED) of the resulting trellis code is maximized, while in BICM the CENCs are designed for binary transmission, i.e., to have maximum free Hamming distance (MFHD) or—more generally—to have optimal distance spectrum (ODS).

In this chapter we consider the BICM transceiver where the interleaver c09-math-0001 in Fig. 2.7 (see also Fig. 8.2) is not present. Equivalently, we might say that the interleaving vector defined in Section 2.7 is “trivial,” i.e., c09-math-0002. Owing to this interpretation, we will refer to this transceiver as BICM with trivial interleavers (BICM-T). For clarity of presentation, we consider the simplest 1D case, i.e., a rate c09-math-0003 (c09-math-0004 and c09-math-0005) CENC and a (4PAM) (c09-math-0006) constellation labeled by the BRGC over the AWGN channel. The resulting transceiver is shown in Fig. 9.1. Furthermore, we do not consider a multiplexer block (as we did in Chapter 8), as in this case the two most relevant cases can be obtained by swapping the order of the generator polynomials of the encoder.

A quick examination of Fig. 9.1 reveals that the structure of the transmitter of BICM-T corresponds to the trellis encoder shown in Fig. 2.4. This type of encoders are typically used with a decoder that implements the ML detection rule in Definition 3.2. In BICM-T, however, the receiver in Fig. 9.1 is composed of a demapper and a binary decoder, i.e., it is a suboptimal decoder that implements the BICM decoding rule in Definition 3.4.

c09f001

Figure 9.1 BICM-T system under consideration: a trellis encoder and a BICM decoder

When the (nontrivial) interleaver is present (e.g., in BICM-S or BICM-M), the use of a BICM decoder is mandatory as there is no efficient way of implementing the ML rule. Nevertheless, the receiver in Fig. 9.1 uses the BICM decoding rule, even though the interleaver is not present anymore. As we will see later, this in fact gives an advantage over BICM-S. Therefore, the receiver of BICM-T in Fig. 9.1 corresponds to a conventional BICM receiver, where L-values for each bit are computed and then fed to a binary decoder. Here we consider a CENC and a binary decoder based on the Viterbi algorithm.

9.1.1 PDF of the L-values in BICM-T

At each time instant c09-math-0007, two bits c09-math-0008 and c09-math-0009 are transmitted using the same symbol c09-math-0010. The two corresponding L-values c09-math-0011 and c09-math-0012 are then obtained from the same observation c09-math-0013, and thus, c09-math-0014 and c09-math-0015 are dependent. In principle, we might design a decoder, which is aware of the fact that the L-values are not independent in BICM-T. However, this would go against the BICM principle of using a binary decoder operating “blindly” on the L-values and we thus continue to use the BICM decoder, which applies the decision rules defined in Section 6.2.4.

In particular, the decoding metric (6.183) for BICM-T is expressed as

where because of the absence of the interleaver c09-math-0017 and c09-math-0018.

The pairwise-error probability (PEP) in (6.188) for the metric in (9.1) is then given by

where

and

9.4 equation

indicates that the L-value c09-math-0022 contributes to the PEP only for the values of c09-math-0023 where c09-math-0024. Because the channel is memoryless, the metrics in (9.3) for different time instants c09-math-0025 are independent, and thus, without loss of generality, from now on we drop the time index c09-math-0026.

Depending on values of c09-math-0027 and c09-math-0028, the metric c09-math-0029 in (9.3) is calculated as specified in Table 9.1. We will now compute the probability density function (PDF) of c09-math-0030 in (9.3) for the 12 cases in which c09-math-0031 or c09-math-0032.

Table 9.1 Possible values of the metric c09-math-0033 in (9.3)

c09-math-0034 c09-math-0035 c09-math-0036 c09-math-0037
c09-math-0038 0 c09-math-0039 c09-math-0040 c09-math-0041
c09-math-0042 c09-math-0043 0 c09-math-0044 c09-math-0045
c09-math-0046 c09-math-0047 c09-math-0048 0 c09-math-0049
c09-math-0050 c09-math-0051 c09-math-0052 c09-math-0053 0

We start with the simplest case, i.e., c09-math-0079 and c09-math-0080. We use the zero-crossing model (ZcM) from Section 5.5.2 to approximate the PDF of the L-values, and thus, we can use the results in Example 5.19 for the cases c09-math-0081 and c09-math-0082. In particular, from (5.198), we obtain

and from (5.197) we obtain

9.6 equation

where c09-math-0086 (see (5.7)).

The expressions (9.5)–(9.7) give the parameters of 8 out of the 12 PDFs. These parameters are shown in Table 9.2 (not highlighted). In what follows, we consider the problem of computing the remaining four PDFs in Table 9.1, i.e., the PDFs c09-math-0087, c09-math-0088, c09-math-0089, and c09-math-0090.

Table 9.2 Mean values and variances of the Gaussian approximation for c09-math-0054 in Table 9.1. Eight out of the 12 relevant cases follow directly from Example 5.19. The remaining four (highlighted) are given by (9.9)–(9.10). To indicate that the Gaussian distributions correspond to the same c09-math-0055, we use circles (for c09-math-0056), stars (for c09-math-0057), and diamonds (for c09-math-0058)

c09-math-0059 c09-math-0060 c09-math-0061 c09-math-0062
c09-math-0063 c09-math-0064 c09-math-0065 c09-math-0066
c09-math-0067 c09-math-0068 c09-math-0069 c09-math-0070
c09-math-0071 c09-math-0072 c09-math-0073 c09-math-0074
c09-math-0075 c09-math-0076 c09-math-0077 c09-math-0078

As shown in Section 5.1.2, the max-log L-values are piece-wise linear functions of c09-math-0091, and thus, the metrics c09-math-0092 in (9.3), being a linear combinations of c09-math-0093 and c09-math-0094, are also piece-wise linear functions of c09-math-0095. In Fig. 9.2, we show c09-math-0096 in (9.3) for the four possible cases when c09-math-0097. These functions are obtained by linearly combining the piece-wise functions in Fig. 5.10 (see also (5.53) and (5.54)).

c09f002

Figure 9.2 Piece-wise relation between the metric c09-math-0098 in (9.3) and the received signal c09-math-0099 for all the possible values of c09-math-0100 and c09-math-0101, with c09-math-0102. The four functions (solid lines) are obtained by linearly combining the piece-wise functions in (5.53) and (5.54). The constellation symbols are shown with black circles and the binary labelings of the corresponding symbols relevant for the PDF computation are also shown. The dashed lines represent the linear approximation made by the ZcM approximation. (a) c09-math-0103, c09-math-0104; (b) c09-math-0105, c09-math-0106; (c) c09-math-0107, c09-math-0108; and (d) c09-math-0109, c09-math-0110

For a given transmitted symbol c09-math-0111, the received signal c09-math-0112 is a Gaussian random variable with mean c09-math-0113 and variance c09-math-0114. Therefore, the PDF of c09-math-0115 in (9.3) is a sum of piece-wise truncated Gaussian functions. In order to obtain expressions that are easy to work with, we use again the ZcM, which replaces all the Gaussian pieces by a single Gaussian function. To clarify this, we show in Fig. 9.2 the linear approximation the ZcM does. For example, for c09-math-0116 (i.e., c09-math-0117, and c09-math-0118), we have

9.8 equation

and because c09-math-0120, we obtain

A similar analysis can be done for the remaining three cases, yielding the same result, i.e.,

The parameters of the Gaussian PDFs in (9.9) and (9.10) are the ones we highlight in Table 9.2. This completes the task of finding the 12 densities for the metrics in Table 9.1.

9.1.2 Performance Analysis

Now that we have analytical approximations for the PDF of c09-math-0123 in (9.3), we return to the problem of computing the PEP in (9.2). First, from the results in Table 9.2 we see that there are three Gaussian PDFs with different parameters, where the number of Gaussian functions needed to evaluate the PEP in (9.2) depends on the transmitted codeword c09-math-0124 and the competing codeword c09-math-0125. In principle, this is not a problem, and the PEP can be computed for each possible pair c09-math-0126. However, the disadvantage of such an approach is that the final PEP expression will have to consider a distance spectrum of the CENC that takes into account all possible pairs of coded sequences that leave a trellis state and remerge after an arbitrary number of trellis stages. In other words, in such a case, the generalized input-dependent weight distribution (GIWD) we considered in Section 8.1.4, which assumes the all-zero codeword was transmitted, cannot be used. This is not surprising as the trellis encoder we consider now is not linear.

To overcome the problem described above, we assume the pseudorandom scrambler in Fig. 3.3 is used. At each time instant c09-math-0127, four scrambler outcomes are possible: c09-math-0128, c09-math-0129, c09-math-0130, and c09-math-0131. For any given pair c09-math-0132, the distribution of the c09-math-0133th metric contribution is given in Table 9.2. It can be shown that for each scrambler outcome, the distribution of the c09-math-0134th metric contribution c09-math-0135 changes over the four distributions shown with the same marker in Table 9.2. For example, for c09-math-0136 and c09-math-0137, we see that the c09-math-0138th metric contribution is distributed as c09-math-0139, which is marked with a circle. For c09-math-0140 no additional analysis is necessary (the bits are not scrambled), but if c09-math-0141, then c09-math-0142 and c09-math-0143, and thus, the c09-math-0144th metric contribution is given by the entry c09-math-0145 in the table (marked with a circle), i.e., c09-math-0146. If c09-math-0147 the relevant entry in the table is c09-math-0148 and if c09-math-0149 the entry is c09-math-0150. The key observation here is that all these four densities (marked with a circle) correspond to the same binary difference c09-math-0151. In the case we just described, c09-math-0152.

A similar analysis can be done for other cases. For c09-math-0153, all the four distributions marked with a star are covered by changing the scrambler outcome. In this case, the distribution is always given by c09-math-0154.

When c09-math-0155 is considered, we have c09-math-0156, and then, unequal error protection (UEP) effect is apparent: in two out of the four cases the distribution c09-math-0157 is obtained (low protection), while in the other two cases we obtain the distribution c09-math-0158 (high protection). This means that while scrambler does not affect the distribution for the cases c09-math-0159 and c09-math-0160, it does affect the case c09-math-0161.

We can summarize the previous discussion by expressing the PDF of the metric in (9.3) (after marginalization over the scrambler outcomes) conditioned on the random variable c09-math-0162 modeling the error c09-math-0163, i.e.,

where

For any linear code, the PEP can then be expressed as

which means that we can assume the all-zero codeword was transmitted and use the PDFs in (9.11)–(9.14) to evaluate the PEP.

To take into account the contribution of the errors c09-math-0169 in the PEP, we make a “super-generalization” of the Hamming weight (HW) of c09-math-0170 via the vector c09-math-0171, where

9.16 equation
9.17 equation
9.18 equation

Then, the PEP in (9.15) can be expressed as

9.19 equation

Using the PDF of the L-values in (9.12)–(9.14), we can now calculate the PEP in (9.20). The elements in the integral in (9.20) can be expressed as

9.22 equation

where we used c09-math-0180 and where the binomial coefficient in (9.21) appears because of the convolution of the sum of the two Gaussian functions in (9.12).

Using the relationships (9.21)–(9.23) in (9.20) yields

where

By using c09-math-0184 (see (2.47)) and (9.25) and (9.26) in (9.24), we obtain

In analogy to (8.22), the bit-error probability (BEP) for BICM-T can be approximated as

which used with (9.27) gives the final PEP approximation for BICM-T:

Similarly, an expression for the word-error probability (WEP) for BICM-T can be found, namely,

In (9.29) and (9.30), we use c09-math-0189 and c09-math-0190, which are the GIWD and generalized weight distribution (GWD) of the CENC. These GIWD and GWD for BICM-T should consider neither the total HW c09-math-0191 (needed for BICM-S) nor the generalized Hamming weight (GHW) c09-math-0192 (needed for BICM-M), but a GHW c09-math-0193. In this super-generalization of the HW, c09-math-0194 counts the number of columns of the divergent sequence in the trellis where c09-math-0195 and c09-math-0196, c09-math-0197—the number of columns where c09-math-0198 and c09-math-0199, and c09-math-0200—the number of columns where c09-math-0201. In other words, it generalizes the GIWD used in BICM-M in the sense that it considers the case c09-math-0202 as a special case. This new GIWD of the encoder is such that for a divergent codeword with HW c09-math-0203,

The next example shows how to calculate the new, super-generalized c09-math-0205 and c09-math-0206.

c09f003

Figure 9.3 Generalized state machine for c09-math-0207 used in the analysis of BICM-T. The path that generates the FHD event is c09-math-0208

c09f004

Figure 9.4 BEP approximations (lines) and simulation results (markers) for BICM-T, BICM-M, and BICM-S, 4PAM labeled by the BRGC, and for the ODS CENC in Table 2.1: c09-math-0209 (c09-math-0210) and c09-math-0211 (c09-math-0212). The BEP approximations are given by (9.29) for BICM-T, by (8.52) for BICM-M, and (6.398) for BICM-S. The asymptotic bounds given by (9.36) are shown with dashed lines

We conclude this section by comparing the results for BICM-T with those obtained for BICM-M and BICM-S. First of all, we note that unlike BICM-M where for c09-math-0253 only two variables are needed (c09-math-0254 and c09-math-0255, see Section 8.1.4), in BICM-T three variables are needed (c09-math-0256, c09-math-0257, and c09-math-0258). If we now consider BICM-S, owing to the interleaver, the case c09-math-0259 in (9.11) is ignored because it does not occur frequently (see Fig. 6.13 (a)). In such a case, only circles and diamonds in Table 9.2 should be considered, and thus, only two Gaussian distributions appear in the final density, namely, c09-math-0260 and c09-math-0261, with probabilities c09-math-0262 and c09-math-0263, respectively. This can also be seen from (9.12) and (9.13) (the PDF in (9.14) is not considered). As expected, this corresponds to the model in Example 6.34, where the probabilities c09-math-0264 and c09-math-0265 correspond to the values of c09-math-0266 and c09-math-0267 studied in that example.

9.2 Code Design for BICM-T

The results from the previous section show relatively large gains offered by BICM-T over BICM-S. In this section, we quantify these gains and we analyze the performance of BICM-T for asymptotically high signal-to-noise ratio (SNR). Asymptotically optimum encoders are defined and BICM-T is also compared with Ungerboeck's TCM.

9.2.1 Asymptotic Bounds

The BEP approximation in (9.29) is a sum of weighted Q-functions, and thus, as c09-math-0268, the BEP is dominated by the Q-functions with the smallest argument. For each c09-math-0269, the dominant Q-function in (9.28) is obtained for c09-math-0270, and thus, the BEP of BICM-T can be (asymptotically) approximated as

where

9.38 equation

In Fig. 9.4, we show asymptotic approximations for c09-math-0274. For BICM-T we used (9.36), for BICM-S we used (6.399), and for BICM-M we used (8.53). All of them are shown to follow well the simulation results. Similar results can be obtained for the encoder with c09-math-0275; however we do not show them so that the figure is not overcrowded.

The asymptotic gain (AG) provided by using BICM-T instead of BICM-S, denoted by c09-math-0276, is obtained directly from (9.36) and (6.399), as stated in the following corollary.

Corollary 9.3 holds for any CENC; however, because ODS CENCs are the asymptotically optimal encoders for BICM-S, studying their loss is of particular interest. To clarify this, consider the following example.

Although the gains offered by BICM-T over BICM-S are quite large, they are bounded, as shown in the following theorem.

The results in Theorem 9.5 are valid for any CENC. In particular, they also hold for the ODS CENCs. As expected, the AG obtained in (9.40) is consistent with (9.43). The AG offered by BICM-T over BICM-S for all the ODS CENCs c09-math-0316 (up to c09-math-0317) are shown in the eighth column of Table 9.3. The values obtained are around c09-math-0318, which is consistent with Theorem 9.5.

Table 9.3 Optimal encoders for BICM-T, their FHD, as well as their coefficients c09-math-0319 and c09-math-0320. New encoders found, better than the ODS CENCs in Table 2.1, are highlighted. The asymptotic gains are also shown

c09-math-0321 Optimal ODS AG [dB]
c09-math-0322 c09-math-0323 c09-math-0324 c09-math-0325 c09-math-0326 c09-math-0327 c09-math-0328 c09-math-0329 c09-math-0330
c09-math-0331 c09-math-0332 5 9 c09-math-0333 9 c09-math-0334 c09-math-0335 c09-math-0336 c09-math-0337
c09-math-0338 c09-math-0339 6 10 c09-math-0340 10 c09-math-0341 c09-math-0342 c09-math-0343 c09-math-0344
c09-math-0345 c09-math-0346 7 11 c09-math-0347 11 c09-math-0348 c09-math-0349 c09-math-0350 c09-math-0351
c09-math-0352 c09-math-0353 7 13 c09-math-0354 12 c09-math-0355 c09-math-0356 c09-math-0357 c09-math-0358
c09-math-0359 c09-math-0360 9 14 c09-math-0361 14 c09-math-0362 c09-math-0363 c09-math-0364 c09-math-0365
c09-math-0366 c09-math-0367 10 16 c09-math-0368 15 c09-math-0369 c09-math-0370 c09-math-0371 c09-math-0372

9.2.2 Optimal Encoders for BICM-T

ODS CENCs are the asymptotically optimal CENCs for binary transmission and for BICM-S, as shown in (6.385). For BICM-M, we showed in Section 8.2.3 that ODS CENCs are not optimal anymore; the optimal joint interleaver and encoder design for BICM-M is given in Definition 8.11. If BICM-T is considered, and as a direct consequence of (9.36), optimal encoders for BICM-T can be defined too. These encoders are asymptotically optimal for BICM-T and guarantee gains for high SNR; however, we will also show that gains appear for low and moderate SNR.

We performed an exhaustive numerical search based on Definition 9.6 and found the optimal encoders for BICM-T. The results are shown in Table 9.3, where for comparison we also show the respective coefficients c09-math-0378 and c09-math-0379 of the ODS CENCs. The search is done in lexicographic order and considering a truncated spectrum c09-math-0380, where c09-math-0381 is the FHD of the ODS CENC for that memory c09-math-0382. If there is more than one optimal encoder for a given c09-math-0383, we present the first one in the list. As shown in Table 9.3, new encoders (highlighted) are found in four out of six analyzed cases.

We emphasize that, during the search, the optimal encoders for BICM-T are not assumed to be encoders with MFHD, i.e., c09-math-0384 (and not c09-math-0385). The results in Table 9.3 confirm this by showing that in general the FHD of the encoder is not the adequate criterion in BICM-T, i.e., encoders with small FHD could perform better than the ODS CENCs (which are obtained by maximizing the FHD, see Section 2.6.1). For example, for c09-math-0386 the FHD of the optimal encoder is c09-math-0387 while for the ODS CENC the FHD is larger (c09-math-0388). The same happens for c09-math-0389. In fact, only for c09-math-0390 the ODS CENC is also optimum for BICM-T.3 The results in Table 9.3 also show that the gains offered by the optimal encoders in four out of six cases come from a reduced multiplicity c09-math-0395, i.e., both the optimal and the ODS CENC have the same c09-math-0396, and thus, the same asymptotic behavior. On the other hand, for c09-math-0397 and c09-math-0398, there is a differencein c09-math-0399 that will result in nonzero asymptotic gains. Namely, for c09-math-0400

9.47 equation

and for c09-math-0402

9.48 equation

where we used the notation c09-math-0404 to show the asymptotic gain of BICM-T using the optimal encoders versus BICM-T using ODS CENCs.

9.2.3 BICM-T versus TCM

As explained before, the difference between the system in Fig. 9.1 and TCM is the receiver used. In what follows we compare the asymptotic performance of BICM-T and the optimum ML decoder.

We have previously defined in (9.39) the AG of BICM-T over BICM-S. It is also possible to define the AG of BICM-T compared to uncoded transmission with the same spectral efficiency (uncoded 2PAM). This AG is

9.49 equation

which follows from (9.36) and (6.33).

The AG in (9.49) is shown in Table 9.3. For c09-math-0406, c09-math-0407 is equal to c09-math-0408, which is the same as c09-math-0409. This is because BICM-S with c09-math-0410 does not offer any AG compared to uncoded 2PAM.

For comparison, consider the TCM design proposed by Ungerboeck. This is shown in Table 9.4, where the CENCs with their corresponding FHD for different values of c09-math-0429 are shown. This table also includes the AG offered by TCM over uncoded transmission with the same spectral efficiency. Analyzing the values of c09-math-0430 in Table 9.3, we find that they are the same as those in Table 9.4. These results show that if BICM-T is used with the correct CENC, it performs asymptotically as well as TCM, and therefore, it should be considered as a good alternative for CM in nonfading channels.4 This is, however, not the case if BICM-S is used, or if BICM-T is used with the ODS CENC. For example, BICM-T for c09-math-0431 and an ODS CENC gives an AG

9.50 equation

Table 9.4 Ungerboeck's trellis encoders for c09-math-0411 and 4PAM labeled by the natural binary code (NBC) for different c09-math-0412

c09-math-0413 c09-math-0414 c09-math-0415 c09-math-0416 [dB]
2 c09-math-0417 3 c09-math-0418
3 c09-math-0419 4 c09-math-0420
4 c09-math-0421 4 c09-math-0422
5 c09-math-0423 4 c09-math-0424
6 c09-math-0425 5 c09-math-0426
7 c09-math-0427 8 c09-math-0428

which is less than the AG of c09-math-0433 offered by BICM-T with an optimal encoder (or by TCM).

For completeness, in Table 9.3, we also show the AG offered by BICM-S over uncoded transmission, defined as c09-math-0434. The performance of uncoded transmission does not depend on c09-math-0435, but increasing c09-math-0436 changes the performance of BICM-T and BICM-S. This is reflected in an increase of the asymptotic gains (AGs) c09-math-0437 and c09-math-0438 in Table 9.3 when c09-math-0439 increases. Finally, we also note this is not the case for c09-math-0440, where the use of BICM-S with the ODS CENC results in both cases in an asymptotic gain of c09-math-0441 compared to uncoded transmission (see c09-math-0442 in Table 9.3). This can be explained from the fact that the ODS CENCs for c09-math-0443 have the same c09-math-0444.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset