Abstract
Reliable data transmission in 5G and Internet of Things (IoT) systems is limited by multipath fading and Intersymbol Interference (ISI), requiring efficient Forward Error Correction (FEC) with low computational complexity. This paper investigates Multi-Threshold Decoders (MTDs) for Self-Orthogonal Codes (SOCs) in Orthogonal Frequency-Division Multiplexing (OFDM) and Multiple-Input Multiple-Output (MIMO) systems with Space-Time Coding (STC) under International Telecommunication Union – Radio Communication Sector (ITU-R) (Outdoor A, TU6, RA6) and 3GPP Spatial Channel Model (SCM) (Urban Macro/Micro) channels. Using OFDM with 512/1024 subcarriers, SOCs decoded by MTDs were compared with turbo and Low-Density Parity-Check (LDPC) codes (WiMAX 2016; DVB-S2 16/200 and 64/800). The results show that MTDs achieve the same Bit Error Rate (BER) as DVB-S2 LDPC while requiring approximately 30–50 times fewer operations for the same code parameters. In deep fades, MTDs provide up to 1.5 dB BER gain over turbo and WiMAX LDPC codes, with mobility up to 50 km/h having negligible impact. A min-sum refinement provides an additional 1–1.5 dB improvement, and antenna diversity yields a 4–7 dB Signal-to-Noise Ratio (SNR) gain at BER = 10⁻⁵. Open Computing Language (OpenCL)-based GPU acceleration increases throughput from 15 Mbit/s (CPU) to 480 Mbit/s (32× speed-up), confirming scalability for software-defined 5G systems. The proposed MTD framework combines high reliability with low complexity, representing a practical and energy-efficient FEC solution for next-generation 5G and IoT systems.