Joint Shaping of Geometry and Probability based on Mutual Information Neural Estimation

LIANG Jia-xi; NIU Ze-kun; HU Wei-sheng; YI Li-lin

doi:10.13756/j.gtxyj.2022.03.001

您当前的位置：

首页 >

文章列表页 >

Joint Shaping of Geometry and Probability based on Mutual Information Neural Estimation

更新时间：2022-06-15

- Joint Shaping of Geometry and Probability based on Mutual Information Neural Estimation
- Study on Optical Communications Issue 3, Pages: 1-6(2022)
- 作者机构：
  
  上海交通大学　区域光纤通信网与新型光通信系统国家重点实验室，上海　200240
- 作者简介：
- 基金信息：
- DOI：10.13756/j.gtxyj.2022.03.001
  CLC： TN919.3
- Published：10 June 2022，
  
  Received：13 December 2021，
扫描看全文
LIANG Jia-xi, NIU Ze-kun, HU Wei-sheng, et al. Joint Shaping of Geometry and Probability based on Mutual Information Neural Estimation. [J]. Study on Optical Communications (3):1-6(2022)
DOI：

LIANG Jia-xi, NIU Ze-kun, HU Wei-sheng, et al. Joint Shaping of Geometry and Probability based on Mutual Information Neural Estimation. [J]. Study on Optical Communications (3):1-6(2022) DOI： 10.13756/j.gtxyj.2022.03.001.

摘要

针对正交振幅调制在高信噪比的加性高斯白噪声(AWGN)信道与香农极限有1.53 dB容量间隙的问题，文章提出了一种基于互信息量估计的几何与概率联合整形的方法，将几何整形与概率整形相结合，提升通信系统的互信息量。文章将互信息量估计作为计算系统互信息量的方式，以最大化互信息量为目的训练发端的编码器，实现几何整形与概率整形。通过在不同信噪比下AWGN信道中的仿真，验证了基于互信息量估计的几何与概率联合整形系统的性能要优于单独进行几何整形或概率整形的性能。在信噪比为10 dB的AWGN信道中，系统的互信息量与几何整形相比有0.041 7 bit/symbol的增益，与概率整形相比有0.027 9 bit/symbol的增益。

Abstract

In view of the problem that quadrature amplitude modulation has 1.53 dB capacity gap at the Shannon limit in high signal noise ratio

the paper proposes a joint shaping method of geometry and probability based on mutual information neural estimation. Geometric shaping and probability shaping are combined to improve the mutual information of the communication system. In this paper

the mutual information neural estimation is used to calculate the mutual information of the system

and the encoder of the transmitter is trained for the purpose of maximizing the mutual information

so as to realize geometric shaping and probability shaping. Through the simulation of Additive White Gaussian Noise (AWGN) channels with different signal noise ratios

it is proved that the performance of joint shaping of geometry and probability based on mutual information neural estimation is better than that of geometric shaping or probabilistic shaping separately. In AWGN channel with signal-to-noise ratio of 10 dB

the mutual information of the system has a gain of 0.041 7 bit/symbol over geometric shaping

and a gain of 0.027 9 bit/symbol compared over probability shaping.

关键词

深度学习互信息量估计物理层通信

Keywords

deep learningmutual information neural estimationphysical layer communication

references

Chen B, Okonkwo C, Hafermann H, et al. Increasing Achievable Information Rates via Geometric Shaping[C]//2018 European Conference on Optical Communication (ECOC). Rome, Italy: IEEE, 2018: 1-3.

Zhang S, Yaman F, Mateo E, et al. A Generalized Pairwise Optimization for Designing Multi-Dimensional Modulation Formats[C]//Optical Fiber Communications Conference & Exhibition. Los Angeles, CA, USA: IEEE, 2017: 1-3.

Kschischang F R, Pasupathy S. Optimal Nonuniform Signaling for Gaussian Channels[J]. IEEE Transactions on Information Theory, 2002, 39(3): 913-929.

O’shea T, Hoydis J. An Introduction to Deep Learning for the Physical Layer[J]. IEEE Transactions on Cognitive Communications & Networking, 2017, 3(4): 563-575.

Goldsmith A. Joint Source/Channel Coding for Wireless Channels[C]//Vehicular Technology Conference, 1995 IEEE 45th. Chicago, IL, USA: IEEE, 1995: 614-618.

Stark M, Aoudia F A, Hoydis J. Joint Learning of Geometric and Probabilistic Constellation Shaping[C]//2019 IEEE Globecom Workshops (GC Wkshps). Waikoloa, HI, USA: IEEE, 2019: 1-6.

Belghazi M I, Baratin A, Rajeswar S, et al. MINE: Mutual Information Neural Estimation[J]. arXiv preprint arXiv: 1801.04062, 2018.

Paninski L. Estimation of Entropy and Mutual Information[J]. Neural Computation, 2014, 15(6): 1191-1253.

Donsker M, Varadhan S. Asymptotic Evaluation of Certain Markov Process for Large Time[J]. Communications on Pure and Applied Mathematics, 1975, 36(2): 183-212.

Jang E, Gu S, Poole B. Categorical Reparameterization with Gumbel-Softmax[J]. arXiv preprint arXiv: 1611.01144, 2016.

Fehenberger T, Alvarado A, Böcherer G, et al. On Probabilistic Shaping of Quadrature Amplitude Modulation for the Nonlinear Fiber Channel[J]. Journal of Lightwave Technology, 2016, 34(21): 5063-5073.

更多指标>

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Deep Learning based Channel Estimation in PLC Communication

Application of Deep Learning in Impairments Compensation of Optical Fiber Communication

Electro-optic Modulation Programmable Optical Frequency Comb based on Deep Learning

Research on Modulation Format Recognition Algorithm based on Machine Learning

Traffic Data Synthesis Algorithm for Optical Network based on Topological Link Recognition

Related Author

No data

Related Institution

School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications

Beijing Smartchip Microelectronics Technology Co., Ltd.

School of Information Science and Technology, Southwest Jiaotong University

College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications

The Ministry of Education Key Laboratory of Electromagnetic Wave Information Science, Department of Communication Science and Engineering, Fudan University

⁰