We consider the problem of designing multipulse pulse-position modulation (MPPM) constellations whose sizes are powers of two. This problem amounts to selecting a subset from the collection of all (wn) possible signal points of MPPM with w pulses in n time slots. In a previous work, we have tackled this selection using combinatorial heuristics. In this letter, we further explore two new continuous optimization approaches. The first one is a modified Blahut-Arimoto algorithm. The second one is inspired from compressed sensing. Using the constellation-constrained channel capacity as the figure of merit, numerical results from a relevant free-space optical communication example suggest that simple combinatorial heuristics yield practically the best designs.

Bit-interleaved coded modulation (BICM) has become the de facto coding standard for communication systems. Recently, BICM has been cast as a mismatched decoding scheme due to the assumption of independent bit metrics. In addition to this inherent mismatch, practical demodulators may produce mismatched decoding metrics because of implementation constraints, such as clipping and metric approximation to reduce computational complexity. In this paper, we investigate BICM with such metrics. In line with recent works on this topic, we adopt the generalized mutual information (GMI) as the pertinent performance measure. First, we show that level-dependent scaling of logarithmic bit metrics can improve the BICM GMI. Second, we propose a uniform metric scaling which can lead to an improved performance of mismatched sum-product symbol-by-symbol decoding, even if the GMI is not changed. Third, we investigate general metric-mismatch correction methods and analyze their effects in terms of the GMI. By means of three application examples, we illustrate that metric-mismatch correction, including metric scaling, can significantly increase BICM rates.

Multilevel coding (MLC) is the main contender to the celebrated bit-interleaved coded modulation (BICM) technique for combining binary error-control codes with multilevel constellations. Although MLC has a more complex encoding-decoding structure, it can achieve a larger rate in a number of important scenarios such as multiple-input multiple-output (MIMO) and orthogonal modulation transmission. In this paper, we consider two issues related to the application of MLC. First, we examine the use of general decoding metrics in MLC, including mismatched metrics that arise from approximations to reduce detection complexity. We make use of recent advances in the analysis of BICM and apply those techniques to individual MLC transmission layers. Our contributions include rate analysis and metric-mismatch correction to improve throughput performance of MLC. Second, we consider the combination of MLC with binary rateless codes. Such a combination eliminates the need to carefully design code rate for each MLC layer. In slow fading environments, rateless coding can also seamlessly adapt to the instantaneous channel quality and achieve an increased average throughput compared to a fixed-rate MLC transmission. However, due to the MLC structure, we show that a naive combination of MLC and rateless coding can cause a significant rate loss. Thus, we propose a novel rotation rateless scheme which preserves the rate advantage of MLC over BICM. We provide relevant examples with MIMO, frequency-shift keying (FSK), and pulse-position modulation (PPM) signaling to demonstrate that our scheme can achieve throughput gains compared to BICM in a variety of transmission scenarios.

In this letter, we address two questions concerning the application of multipulse pulse-position modulation (MPPM) for free-space optical (FSO) communications: (1) under what condition would MPPM offer a notable advantage over conventional pulse-position modulation (PPM), and (2) what practical coding scheme should be used to realize that advantage. Focusing on bandwidth-efficient FSO transmission, we find decimated MPPM constellations which are suitable for combining with binary codes and offer significant gains in terms of constellation-constrained capacity over PPM under simultaneous peak-power, average-power, and bandwidth constraints. We then consider labeling design for the popular bit-interleaved coded modulation (BICM) and devise a new multilevel coding (MLC) architecture for MPPM, which we refer to as reduced-layer MLC (RLMLC). Considering the Poisson FSO channel model as a relevant example, we provide simulative evidence that RL-MLC MPPM with off-the-shelf low-density parity-check codes can outperform any PPM scheme under the same transmission constraints.

Free-space optical (FSO) transmission systems enable high-speed communication with relatively small deployment costs. However, FSO suffers a critical disadvantage, namely susceptibility to fog, smoke, and conditions alike. A possible solution to this dilemma is the use of hybrid systems employing FSO and radio frequency (RF) transmission. In this paper we propose the application of a rateless coded automatic repeatrequest scheme for such hybrid FSO/RF systems. The advantages of our approach are (a) the full utilization of available FSO and RF channel resources at any time, regardless of FSO or RF channel conditions and temporal variations, and (b) no need for a-priori rate selection at the transmitter. In order to substantiate these claims, we establish the pertinent capacity limits for hybrid FSO/RF transmission and present simulation results for transmission with off-the-shelf Raptor codes, which achieve realized rates close to these limits under a wide range of channel conditions. We also show that in conditions of strong atmospheric turbulence, rateless coding is advantageous over fixed-rate coding with rate adaptation at the transmitter.

We propose a novel communication scheme that combines rateless transmission and multilevel coding (MLC). The scheme is able to attain a throughput that is equal to the constellation-constrained channel capacity. It is particularly useful in cases where the celebrated bit-interleaved coded modulation (BICM) would leave significant gaps to the channel capacity and therefore, MLC is necessary for efficient transmission. We illustrate the application of this scheme to pulse-position modulation (PPM) in free-space optical and multiple-input multiple-output (MIMO) transmission in radio-frequency communication systems. Our numerical results show that, employing an off-the-shelf Raptor code, the proposed scheme achieves a throughput that exceeds the BICM capacity limit in a wide range of signal power.

Multilevel (Q-ary, Q > 2) pulse-position modulation (Q-PPM) with direct detection is a very popular transmission method for power-efficient free-space optical communication systems. The combination of Q-PPM with error-control coding is an effective means to further improve power efficiency. In this paper, we study the application of the multilevel coding (MLC) paradigm to Q-PPM transmission. In particular, we devise a powerful coded Q-PPM scheme which is a simplified version of MLC and which we refer to as reduced-level MLC (RL-MLC). We show how to design and optimize RL-MLC for Q-PPM when using constellation-constrained capacity as the pertinent figure of merit. Furthermore, we provide simulative evidence that RL- MLC with off-the-shelf low-density parity-check codes (LDPC) closely approaches its corresponding capacity limit. For 64-PPM, RL-MLC with only two levels achieves practically the same performance as that of bit-interleaved coded modulation with iterative decoding (BICM-ID), which involves a more difficult design procedure.

Partial transmit sequences (PTS) is a popular technique to reduce the peak-to-average power ratio (PAR) in orthogonal frequency division multiplexing (OFDM) systems. PTS is highly successful in PAR reduction and efficient redundancy utilization, but the considerable computational complexity for the required search through a high-dimensional vector space and the necessary transmission of side information (SI) to the receiver are potential problems for a practical implementation. In this paper, we revisit PTS for PAR reduction and tackle these two problems. To address the complexity issue, we formulate the search problem of PTS as a combinatorial optimization (CO) problem. This enables us to (i) unify various search strategies proposed earlier in the PTS literature and (ii) adapt efficient search algorithms known from the CO literature to PTS. We also propose a modified PTS objective function, which reduces the number of multiplications required for PTS. Numerical results show that, perhaps surprisingly, simple random search yields the best performance-complexity tradeoff for moderate PAR reduction, whereas two novel CO-based methods excel if close-to-optimum PAR reduction is desired. The SI transmission problem is solved by a simple preprocessing of the data stream before PAR reduction. This preprocessing introduces the minimal possible redundancy and allows SI embedding without affecting the PAR reduction capability of PTS or causing peak regrowth.

The application of trellis shaping was proposed to reduce the peak-to-average power ratio (PAR) of orthogonal frequency division multiplexing (OFDM) signals. In this letter, we review the trellis-shaping schemes presented in the literature, and we introduce modifications such as a new decoding metric and the use of sequential decoding. We conduct comprehensive complexity and performance comparisons for the different schemes, and one interesting result of this work is that, in terms of PAR-reduction capability, trellis shaping with time-domain metrics is generally superior to trellis shaping with frequency-domain metrics. Furthermore, the proposed modifications enable trellis shaping for PAR reduction with a flexible performance-complexity tradeoff.

This paper takes a fresh look at peak-to-average power ratio (PAR) reduction for orthogonal frequency-division multiplexing (OFDM) using partial transmit sequences (PTS). In particular, the two major drawbacks of PTS, its high complexity when searching for a "good" solution and the need to transfer side information (SI) to the receiver are addressed. Reformulating PTS as a combinatorial optimization (CO) problem enables us to apply various efficient search strategies from the CO literature to PTS. We also incorporate an approximation of the PTS objective function to further reduce the overall complexity. Finally, we propose an implicit SI embedding scheme. This low-complexity scheme requires minimal redundancy and has minimal effect on the overall performance of the OFDM system.

The application of trellis shaping was recently proposed to reduce the peak-to-average power ratio (PAR) of orthogonal frequency division multiplexing signals. In this paper, we review the trellis shaping schemes presented in the literature and we introduce modifications such as a new decoding metric and the use of sequential decoding, which improve PAR reduction and/or reduce complexity of trellis shaping. Based on implementation onisuggestions, we conduct a complexity and performance comparison for the different schemes. One interesting result of this work is that trellis shaping with time-domain metrics is superior to trellis shaping with frequency-domain metrics.