Comparison of LDPC block and LDPC convolutional codes based on their decoding latency

The Four-C Framework for High Capacity Ultra-Low Latency in 5G Networks: A Review

“…Current progress on the detection scheme in mitigating computational complexity is by proposing rate-compatible puncturing polar (RCPP) codes in which two different puncturing methods are identified-quasi uniform puncturing and shortening [186]. Also, another detection scheme for polar codes that can reduce complexity is blind detection based on the cyclic redundancy check (CRC) [187]. The emergence of 5G also introduces the MIMO-non orthogonal multiple access (NOMA) where the low-complexity receiver used LMMSE multi-user detector that operates iteratively with single user message passing decoder [186] for the latest current progress in achieving low-complexity low-latency in 5G.…”

“…In current OFDM, it faces a challenge for opportunistic and dynamic spectrum access [185] which is due to the high out-of-band (OOB) emission [186]. In addition to that, there are various types of modulation schemes such as Filter Bank Multicarrier (FMBC) [16], Universal Filtered Multicarrier (UFMC) [187] and Bi-Orthogonal OFDM (BFDM) [188] has been discussed for 5G, however, the most promising solution of modulation scheme for PHY layer for 5G is Generalized Frequency Division Multiplexing (GFDM) [187]. As discussed in [146], a short transmission time interval in frame structure could be one of the practical solutions in achieving 1ms E2E TTI latency.…”

A Scaling Law to Predict the Finite-Length Performance of Spatially-Coupled LDPC Codes

“…Much less is known about their finite-length behavior in the waterfall region [5]. Even though there are multiple recent papers reporting simulation results for different classes of finite-length SC-LDPC codes [11], [12], [13], [14], [15], to date we lack analytical models to relate the BP block error probability of finite-length SC-LDPC codes to their structural parameters. The present paper addresses this problem by extending existing analytical results to analyze finite-length LDPC codes [16], [17] to a particular family of SC-LDPC codes.…”

Challenges and some new directions in channel coding

“…It has been shown in [27] that for equal structural latency, SC-LDPC codes under window decoding outperform LDPC codes for short to long latency values and outperform convolutional codes from medium to long latency values. For applications requiring very short latency, Viterbi decoded convolutional codes were still found to be the optimal choice [28][29] [27]. Note that only structural latency was considered in all these comparisons which is defined as the number of bits required before decoding can start.…”

“…Another advantage of using a window decoder is the flexibility in terms of decoding latency at the decoder. Since the window size W is a decoder parameter, it can be varied without changing the code, providing a flexible trade-off between performance and latency [27]. In BP decoding, messages are passed between the check and variable nodes according to a parallel (flooding) or serial (ondemand) rule [30].…”