Towards Achieving Ultra Reliable Low Latency Communications Using Guessing Random Additive Noise Decoding

Download or read book Towards Achieving Ultra Reliable Low Latency Communications Using Guessing Random Additive Noise Decoding written by Marwan Jalaleddine and published by . This book was released on 2021 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: "Ultra-reliable and low latency communications (URLLCs) is one of the key pillars of the 5G communications standard which is used to enable applications ranging from the smart grid to robot control. In the upcoming communication standards, more stringent requirements are being established on the end-to-end latency and reliability of data.In an effort to build upon the current advancements in URLLC and guessing random additive noise decoding (GRAND), we develop the partitioned GRAND (PGRAND) which uses the quantized reliability information from the channel to generate the most likely test error patterns. We assess the performance of PGRAND on 5G NR CA-polar code, random linear code, and cyclic redundancy check codes. PGRAND provides superior performance to that of ordered reliability bit GRAND at high signal-to-noise ratios (SNRs) by achieving a 0.2dB gain at a frame error rate (FER) of 10^(-4) and a 50% reduction in the average queries per frame performance at Eb/N0 > 5.5dB. Additionally, PGRAND approaches the FER performance of soft maximum likelihood GRAND at high SNRs with less scheduling complexity. This makes PGRAND a desirable candidate as a near maximum likelihood code agnostic decoder for any short, high rate code. Alternatively, we also develop guessing random additive noise-assisted decoding (AGRAND) that can be used alongside any conventional decoder to improve the decoder latency. If AGRAND succeeds to find a version of the codeword that belongs to the codebook, the decoder terminates early, saving latency and power. This decoding scheme can reduce latency by up to 84% at Eb/N0 = 5.5dB when used with successive cancellation list decoding on CA-polar code. As such, AGRAND enables maximum likelihood low latency decoding of CA-polar codes"--