Fixed-Throughput GRAND with FIFO Scheduling

2025-02-07

AI Summary

This work addresses the problem that GRAND decoders suffer from variable throughput due to stochastic noise-pattern guessing, thereby failing to meet deterministic latency requirements in real-time systems. To resolve this, we propose the first hardware-friendly decoding architecture supporting strictly fixed throughput. Our method introduces a FIFO-based scheduling mechanism coupled with dynamic priority management of noise sequences, enabling rigorous throughput enforcement while significantly reducing block error rate (BLER). Experiments demonstrate that our scheme achieves lower BLER than conventional truncated GRAND under identical fixed-rate constraints. Moreover, we reveal that average throughput metrics can mask severe performance degradation: restoring unconstrained decoding performance requires reducing the operating rate by approximately one order of magnitude. To the best of our knowledge, this is the first work to jointly optimize deterministic latency and near-optimal error-correction capability, establishing a deployable GRAND decoding paradigm for real-time communication systems.

Technology Category

• Machine Learning: Hardware-aware ML
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Guessing random additive noise decoding (GRAND) is a code-agnostic decoding method that iteratively guesses the noise pattern affecting the received codeword. The number of noise sequences to test depends on the noise realization. Thus, GRAND exhibits random runtime which results in nondeterministic throughput. However, real-time systems must process the incoming data at a fixed rate, necessitating a fixed-throughput decoder in order to avoid losing data. We propose a first-in first-out (FIFO) scheduling architecture that enables a fixed throughput while improving the block error rate (BLER) compared to the common approach of imposing a maximum runtime constraint per received codeword. Moreover, we demonstrate that the average throughput metric of GRAND-based hardware implementations typically provided in the literature can be misleading as one needs to operate at approximately one order of magnitude lower throughput to achieve the BLER of an unconstrained decoder.

Problem

• Achieves fixed throughput in GRAND decoding
• Improves block error rate with FIFO scheduling
• Addresses misleading average throughput metrics

Innovation

• FIFO scheduling architecture
• Fixed-throughput GRAND decoder
• Improved block error rate