IEEE/CIC ICCC Workshop on
Information Theory and Coding for Future Wireless
11th August 2022 Room 3
Keynote: 13:00-14:00
[13:00-13:30] Keynote 1
Speaker: Jinhong Yuan, University of New South Wales
Title: Capacity achieving generalized spatially coupled turbo-like codes
Abstract: Turbo codes or parallel concatenated convolutional codes (PCCs) are the standard channel codes used in 3G and 4G mobile communication systems. It is well known that they can approach the Shannon limit under low-complexity iterative Bahl-Cocke-Jelinek-Raviv (BCJR) decoding. To close the gap between their performance and Shannon limit, several works applied the technique of spatial coupling (SC) to turbo codes, resulting in spatially coupled parallel concatenated codes (SC-PCCs), partially information coupled turbo codes (PIC-TCs), etc. SC turbo-like codes outperform their uncoupled counterparts in terms of waterfall and error floor. Particularly, it was proved that SC-PCCs exhibit threshold saturation: the belief-propagation (BP) decoding threshold can saturate to the maximum a posterior probability (MAP) decoding threshold of the underlying uncoupled codes. However, the threshold of SC-PCCs is still away from the capacity when punctured. In contrast, PIC-TCs have better threshold but the effect of threshold saturation is not observed. Despite this progress, it is unclear whether the class of turbo-like codes can achieve capacity.
In this talk, we introduce generalized spatially coupled PCCs (GSC-PCCs). The codes are constructed by applying spatial coupling on PCCs with a fraction of information bits repeated q times. GSC-PCCs can be seen as a combination of PIC-TCs and SC-PCCs. By using density evolution and the potential function arguments, threshold saturation is numerically observed and analytically proven. Most importantly, we rigorously prove that any rate-R GSC-PCC ensemble with 2-state convolutional component codes achieves at least a fraction 1-R/(R+q) of the binary erasure channel (BEC) capacity for repetition factor q ≥ 2 and this multiplicative gap vanishes as q tends to infinity. To the best of our knowledge, this is the first class of turbo-like codes that are proven to be capacity-achieving. In addition, the decoding threshold can be further improved by using convolutional component codes with larger states.
[13:30-14:00] Keynote 2
Speaker: Jincheng Dai, Beijing University of Posts and Telecommunications
Title: Semantic Communications: Why, What, and How?
Abstract: With the deep fusion of wireless communication and artificial intelligence, the concept of “semantic communications” becomes an emerging research direction. However, the viewpoints about semantic information generation, transmission, and usage in communication systems are still divergent. This speech gathers ideas to formulate an explicit definition of semantic communication and summarize what formats of semantics can be utilized in different communication scenarios. In particular, we justify that in end-to-end wireless communication systems, the semantics of information is not necessarily defined as the meaning of data that serves for human understanding, but as the significance of both data and environment relative to the goal attainment. Semantic communications are optimized for a match between source data diversity and wireless channel diversity. We discuss several typical formats of semantics used for data transmission and generation, including embedded features, compact feature keypoints, or even neural network models.
Session 1: 14:00-15:30
[14:00-14:15] Deep Learning Based Constellation Rearrangement Design in Hybrid ARQ System
Bo Cao and Ming Jiang (Southeast University, China); Chunming Zhao (National Mobile Communications Research Laboratory, Southeast University, China); Longhao Zou (Southeast University, China)
[14:15-14:30] A Spatially Coupled PAC Coding Scheme and Its List Decoding
Siyu Ying, Shan Jin and Zhaoyang Zhang (Zhejiang University, China)
[14:30-14:45] Polar Codes for Multilevel Steganography
Jinsheng Li and Ling Liu (Shenzhen University, China); Yinfei Xu (Southeast University, China)
[14:45-15:00] Monomial Codes With Predefined Automorphisms
Kirill Shabunov (Huawei Technologies, Russia)
[15:00-15:15] Design of Non-punctured Protograph LDPC Codes for JSCC Systems over Partial Response Channels
Ying Sun, Shiwei Zhu, Sanya Liu and Chen Chen (Huaqiao University, China)
[15:15-15:30] Design of 4-cycles Free Short-length LDPC Codes Based on the Kirkman Triple Systems
Dongliang Guo (Sun Yat-sen University, China); Yuejiao Feng (Sun Yat-Sen University, China); Congduan Li (Sun Yat-sen University, China)
Coffee Break: 15:30-16:00
Session 2: 16:00-17:30
[16:00-16:15] A Finite Blocklength Coding Scheme for the Two-Way Full-duplex Fading Channel
Dengfeng Xia (Southwest Jiaotong University, China); Chuanchuan Yang (State Key Laboratory of Advanced Optical Communication Systems and Networks, Peking University, China); Bin Dai (Southwest Jiaotong University, China)
[16:15-16:30] Reduced Complexity Ordered Statistics Decoding of Linear Block Codes
Lijia Yang and Wenhao Chen (Sun Yat-Sen University, China); Li Chen (Sun Yat-sen University, China)
[16:30-16:45] Parameters and Decoding Performance of Subfield Subcodes of One-Point Elliptic Codes
Jianguo Zhao (Sun Yat-Sen University, China); Li Chen (Sun Yat-sen University, China)
[16:45-17:00] BP Flip Decoding Algorithm of Polar Code Based on Convolutional Neural Network
Xiaojun Zhang, Yue Qiu, Wenxiu Kong, Jianming Cui and Yimeng Liu (Shandong University of Science and Technology, China)
[17:00-17:15] An Iterative Receiver for Coded Faster-than-Nyquist Signaling over Rayleigh Fading Channels
Chenxi Si (Xidian University, China); Shuangyang Li (University of New South Wales, Australia & Xidian University, China); Mengmeng Liu and Baoming Bai (Xidian University, China); Jiaqing Wang (CICT Mobile Communication Technology Co., Ltd, China)
[17:15-17:30] An SVD-based Construction of Doppler Resilient Complementary Waveforms
Jiahuan Wang and Pingzhi Fan (Southwest Jiaotong University, China); Desmond McLernon (The University of Leeds, United Kingdom (Great Britain)); Zhiguo Ding (University of Manchester, United Kingdom (Great Britain))