Protograph-based LDPC convolutional codes for Continuous phase modulation

2017 IEEE International Symposium on Information Theory (ISIT)

Thomas

et al. 2017

Self Cite

In this paper, we consider a bit-interleaved coded modulation scheme (BICM) composed of an error correcting code serially concatenated with a M-ary non linear modulation with memory. We first compare demodulation strategies for both the coherent and the non coherent cases. Then, we perform an asymptotic analysis and try to show that the design of coding schemes performing well for both the coherent and the non coherent regimes should be done carefully when considering sparse graph based codes such as low-density parity-check (LDPC) codes. It will be shown that optimized coding schemes for the non coherent setting can perform fairly well when using coherent demodulation, while on the contrary, optimized coding schemes for the coherent setting may lead to "non stable" coding schemes in the non coherent setting.

Section: B Exit Charts and Code Designmentioning

confidence: 99%

On sparse graph coding for coherent and noncoherent demodulation

Piat-Durozoi¹,

2017 IEEE International Symposium on Information Theory (ISIT)

Thomas

et al. 2017

Self Cite

“…Following [9], [18], optimization of LDPC code profiles is performed for CPM parameters detailed in Table II using mutual information (EXIT based) evolution equations. It can be efficiently done using linear programming.…”

Section: A Optimized Ldpc Degree Distributions and Protograph Designmentioning

confidence: 99%

New perspectives for coded continuous phase modulations for narrowband waveforms: Iterative versus non-iterative solutions

Naour

Janin

MILCOM 2017 - 2017 IEEE Military Communications Conference (MILCOM)

2017

Self Cite

NarrowBand Waveforms (NBWF) are often used in VHF or UHF tactical communications. For these kinds of waveforms, low latency and robust data rates result in short codeword lengths that are challenging in terms of channel coding. Usually, serially concatenated convolutional code and continuous phase modulation (CC-CPM) schemes are considered in the context of NBWF. When evaluating the achievable rates, CC-CPM schemes show a 1 to 4 dB maximum margin. In this paper, we investigate on two new sparse graph-based channel coding strategies, trying to reduce this degradation for the achievable rate. Some implementation issues with short codeword lengths are also addressed. To this end, we first consider a serially concatenated CPM scheme where the outer code is an optimized Low-Density Parity-Check code (LDPC) based on some recently introduced methods for this kind of applications. The optimized LDPC scheme exhibits a good bit error rate when binary CPMs are used. This new design comes with lower computational complexity and greater flexibility. Second, in order to avoid the cubersome of iterative detection and decoding, we investigate on recently introduced precoded CPM that can achieve near optimal performance, referred to us as Pragmatic CPM (P-CPM). For this kind of precoded CPM schemes, we will show that associated EXIT chart curves have the flat property. It means that, to achieve good performance, P-CPM can be used with standard capacity approaching codes like LDPC Accumulate Repeat Jagged Accumulate (ARJA) or Turbo Codes without the need for iterative decoding. Bit error rate simulations confirm that P-CPM is a reliable alternative NBWF scheme compatible with versatile modern channel codes. As P-CPM is non-iterative, implementation is made easier. In additive white Gaussian noise, both schemes are found practical.

“…Ganesan [23] proposed a bit-interleaved coded-modulation approach to optimize codes for M-ary continuous-phase frequencyshift keying (CPFSK) modulations. Later, structured LDPC codes were considered such as irregular repeat accumulate (IRA) codes [24][25][26] and protograph-based LDPC codes [27,28].…”

Section: Introductionmentioning

confidence: 99%

Spatially coupled turbo-coded continuous phase modulation: asymptotic analysis and optimization

Benaddi

2020

J Wireless Com Network

Self Cite

For serially or parallel concatenated communication systems, spatial coupling techniques enable to improve the threshold of these systems under iterative decoding using belief propagation (BP). For the case of low-density parity-check (LDPC) codes, it has been shown that, under some asymptotic assumptions, spatially coupled ensembles have BP thresholds that approach the bitwise maximum a posteriori (MAP) threshold of the related uncoupled ensemble. This phenomenon is often referred to as threshold saturation, and it has sometimes very important consequences. For example, in the case of regular LDPC code ensembles, spatial coupling enables to achieve asymptotically the capacity for any class of binary memoryless symmetric channels. Since then, this threshold saturation has been conjectured or proved for several other types of concatenations. In this work, we consider a serially concatenated scheme which is the serial concatenation of a simple outer convolutional code and a continuous phase modulator (CPM) separated by an interleaver. Then, we propose a method to do the spatial coupling of several replicas of this serially concatenated scheme, aiming to improve the asymptotic convergence threshold. First, exploiting the specific structure of the proposed system, an original procedure is proposed in order to terminate the spatially coupled turbo-coded CPM scheme. In particular, the proposed procedure aims to ensure the continuity of the transmitted signal among spatially coupled replicas, enabling to keep one of the core characteristics and advantages of coded CPM schemes. Then, based on an asymptotic analysis, we show that the proposed scheme has very competitive thresholds when compared to carefully designed spatially coupled LDPC codes. Furthermore, it is shown how we can accelerate the convergence rate of the designed systems by optimizing the