Deep Learning Polar Convolutional Parallel Concatenated (DL-PCPC) Channel Decoding for 6G Communications

Abstract

The new wireless generation 6G use of intelligent devices, sensors, and new applications like virtual reality and autonomous driving requires higher demands on the network with more users which needs higher data rate networks with minimum delay and less energy consumption. The current state for channel decoding does not meet the 6G requirements. In this paper, we design, evaluate, and proposes deep learning polar convolutional parallel concatenated (DL-PCPC) decoding, a new powerful decoding technique for 6G. The developed decoding technique dynamically reduces errors by 99.8%. It provides up to 80% better system efficiency than iterative decoding algorithms, with a 100% reduction in system delay. The novel proposes design works with a 6G communication frequency range of 300 and 400 GHz with terahertz data rates, providing correctly received data with a minimum amount of detected errors.