This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. IEEE TRANSACTIONS ON BROADCASTING 1 Offset and Normalized Min-Sum Algorithms for ATSC 3.0 LDPC Decoder Seho Myung, Sung Ik Park, Senior Member, IEEE, Kyung-Joong Kim, Jae-Young Lee, Member, IEEE, Sunhyoung Kwon, Member, IEEE, and Jeongchang Kim, Senior Member, IEEE Abstract—Offset min-sum algorithm (OMSA) and normalized min-sum algorithm (NMSA) are widely used in commercial LDPC decoders due to low complexity and reasonable performance. In this paper, we pro- vide optimized offset and scaling values for those LDPC decoders based on Advanced Television Systems Committee (ATSC) 3.0 LDPC codes. Furthermore, according to extensive computer simulations, it is shown that the performance of the OMSA and NMSA with the obtained values is close to that of sum-product algorithm, even though the NMSA-based decoder may show a serious error floor due to a channel mismatch effect. Consequently, we recommend that the ATSC 3.0 transmitters use a concatenation of BCH outer code and LDPC inner code as channel coding, considering the use of NMSA for LDPC decoder in ATSC 3.0 receivers. Index Terms—ATSC 3.0, LDPC decoder, min-sum, modified min-sum, normalized min-sum, offset min-sum. I. I NTRODUCTION T HE NEXT generation digital broadcasting standard, known as Advanced Television Systems Committee (ATSC) 3.0, has been developed with distinct features and capabilities over the existing standards [1]. In the physical layer protocol standard of ATSC 3.0, different broadcasters’ needs and requirements have been taken into account, and as a result, this new standard provides delivery of high data rate services (e.g., ultra-high definition), robustness under mobile or indoor environments, and flexibility according to variety of parameter choices [2], [3]. The ATSC 3.0 physical layer pro- tocol standard includes the following technology features: Given orthogonal frequency division multiplexing, bit-interleaved coded modulation (BICM) that consists of forward error correction codes, bit interleavers, and symbol mappers provides efficiency and channel error mitigation of transmitted broadcasting signals [4]. The latest BICM technologies of low-density parity-check (LDPC) codes and non-uniform constellations allow to further reach the Shannon limit in performance, and moreover, wide ranges of code rates and constel- lations provide flexible choices of broadcasting parameters [5], [6]. Manuscript received February 6, 2017; revised March 3, 2017; accepted March 6, 2017. This work was supported by the Institute for Information and communications Technology Promotion through the Korea Government (MSIP; Transmission/Reception Environment Analysis and Network Configuration Development for Terrestrial UHD) under Grant 2017-0-00442. (Corresponding author: Jeongchang Kim.) S. Myung and K.-J. Kim are with Samsung Electronics Company Ltd, Suwon 16677, South Korea (e-mail: seho.myung@samsung.com; kj1981.kim@samsung.com). S.-I. Park, J.-Y. Lee, and S. Kwon are with Broadcasting System Research Department, Electronics and Telecommunications Research Institute, Daejeon 34129, South Korea (e-mail: psi76@etri.re.kr; jaeyl@etri.re.kr; shkwon@etri.re.kr). J. Kim is with the Division of Electronics and Electrical Information Engineering, Korea Maritime and Ocean University, Busan 49112, South Korea (e-mail: jchkim@kmou.ac.kr). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TBC.2017.2686011 The standard also adopts the latest physical layer technologies such as signaling and framing, layered division multiplexing, channel bonding, enhanced multiple-input single-output and multiple-input multiple-output [7]–[14]. As these distinct features and capabili- ties are integrated in the ATSC 3.0 physical layer protocol, the overall physical layer system provides good robustness under both mobile and indoor environments, efficiency in multiple services con- figuration, and flexibility given broadcasters’ intention of service scenarios. LDPC codes, first introduced by Gallager [15], are known as capacity-approaching code over a symmetric memoryless channel. Because of the performance advantage approaching the Shannon limit and low complexity in encoding and decoding, the LDPC codes have been adopted in many wireless broadcasting standards including ATSC 3.0 [16]–[19]. Recently, efficient decoding algo- rithms of LDPC codes have been intensively studied in order to be used in practice. The sum-product algorithm (SPA) is known as the best performing and the most complex decoding algorithm for LDPC codes. The min-sum algorithm (MSA) is a simplified method that can greatly reduce decoding complexity compared to SPA [20], but it may introduce substantial performance degrada- tion in terms of bit error rate (BER) or frame error rate (FER). The modified MSAs have been introduced to improve the BER and FER performance of the MSA [21]–[24], and there are normalized min-sum algorithm (NMSA) and offset min-sum algorithm (OMSA), which use normalization and offset terms for the output of check node operations, respectively. In this paper, we present the performance comparison of the ATSC 3.0 LDPC codes according to the use of different decoding algorithms such as the SPA, MSA, OMSA, and NMSA. Since the OMSA and NMSA are widely used in commercial LDPC decoders due to low complexity and reasonable performance, we also presents optimized offset and scaling values for the MSA-based LDPC decoders where those values are obtained by density evolution (DE) algorithm. Furthermore, we provides the extensive simulation results to verify that the performance of the OMSA and NMSA with the obtained offset and scaling values is close to that of the SPA. The remainder of the paper is organized as follows. In Section II, we briefly review the LDPC codes in ATSC 3.0. LDPC decoding algo- rithms such as SPA and modified MSAs are described in Section III and the optimized offset and scaling values are obtained and analyzed in Section IV. Finally, conclusions are drawn in Section V. II. LDPC CODE FOR ATSC 3.0 In the physical layer standard of ATSC 3.0 [1], two different lengths of LDPC codes are adopted: the short length (16200) and long length (64800) codes, and two different structures, called Type-A and Type-B, are used for each code length and rate. In Section IV, the different types of LDPC codes are summarized in Tables I and II. During the standardization process of ATSC 3.0, one struc- ture against the other structure for each code rate was selected 0018-9316 c  2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.