Cal Poly
A Study on Quantization and Hardware Cost Tradeoffs in a 5G NR LDPC Decoder on FPGA
Abstract
dc:description.abstractThe fixed-point implementation of a 5G New Radio LDPC decoder forces a tradeoff between precision and hardware cost, governed by the variable node word length WL, the check node message width WR, and the normalized min-sum correction factor α. This thesis characterizes how these three parameters affect both error correction performance and FPGA resource utilization for an LDPC decoder on an established LDPC decoder architecture. A bit-accurate MATLAB core model records bit error rate (BER) and frame error rate (FER) while a verified HDL Coder model generates synthesizable VHDL for Vivado synthesis, and both are swept over WL, WR, and α at lifting sizes Z = 64 and Z = 384. The results show that α is effectively free in hardware cost, while WL costs far more per bit than WR, because WL largely dominates the decoder’s critical path and memory units. For configurations that reach the 5 × 10-5 BER target, the SNR required to reach the target varies by approximately 0.05 dB, so hardware cost dominates parameter selection and points to a smallest passing operating point of WL = 6 and WR = 5.
Degree
thesis:*- Name thesis:degree_name
- MS in Electrical Engineering
- Discipline thesis:degree_discipline
- Electrical Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Acosta, Giancarlo
- Contributors dc:contributor
-
- Andrew Danowitz
- Electrical Engineering
- College of Engineering
Subjects
dc:subject × 8Identifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.calpoly.edu/theses/3335