A Graphical Correlation-Based Method for Counting the Number of Global 8-Cycles on the SCRAM Three-Layer Tanner Graph

TL;DR

This paper introduces a new graphical method to count and analyze 8-cycles in the Tanner graph of SCRAM, a decoding scheme for 6G that improves performance assessment and cycle length estimation.

Contribution

It proposes a methodology to use classical LDPC analysis tools for SCRAM, derives a lower bound on cycle length, and presents a novel cycle counting technique.

Findings

Derived a lower bound on the shortest cycle length in SCRAM Tanner graphs.

Developed a graphical method to count the number of 8-cycles.

Enhanced performance analysis of SCRAM using LDPC tools.

Abstract

This paper presents a novel graphical approach that counts the number of global 8-cycles on the SCRAM three-layer Tanner graph. SCRAM, which stands for Slotted Coded Random Access Multiplexing, is a joint decoder that is meets challenging requirements of 6G. At the transmitter side, the data of the accommodated users is encoded by Low Density Parity Check (LDPC) codes, and the codewords are transmitted over the shared channel by means of Slotted ALOHA. Unlike the state-of-the-art sequential decoders, the SCRAM decoder jointly resolves collisions and decodes the LDPC codewords, in a similar analogy to Belief Propagation on a three-layer Tanner graph. By leveraging the analogy between the two-layer Tanner graph of conventional LDPC codes and the three-layer Tanner graph of SCRAM, the well-developed analysis tools of classical LDPC codes could be utilized to enhance the performance of SCRAM. In essence, the contribution of this paper is three-fold; First it proposes the methodology to utilize these tools to assess the performance of SCRAM. Second, it derives a lower bound on the shortest cycle length of an arbitrary SCRAM Tanner graph. Finally, the paper presents a novel graphical method that counts the number of cycles of length that corresponds to the girth.