METTLE: Efficient Streaming Erasure Code with Peeling Decodability

TL;DR

METTLE introduces a novel streaming erasure code that achieves high coding efficiency with low decoding complexity by integrating hashing-based IBLT concepts, time coupling, and two targeted modifications (MET and TLE). The scheme enables on-the-fly, peeling-based decoding and zero latency under lossless channels, while maintaining robustness to bursty erasures and time-varying Gilbert-Elliott channels. Empirical results show METTLE attains average decoding latencies of $37$–$199$ packets and 95th percentile latencies of $291$–$815$ packets, with decoding speeds around $2.6~\mu$s per packet, delivering $47.7$–$84.6\times$ faster decoding than RaptorQ and competitive coding efficiency (slightly worse than RaptorQ at small erasures, better under bursts). Its hashing-based, nearly stateless design supports continuous rate adaptation via overhead $c$ and window $w$, while effectively mitigating tail loss through tail compression, making METTLE practical for real-time streaming in networks and systems.

Abstract

In this work, we solve a long-standing open problem in coding theory with broad applications in networking and systems: designing an erasure code that simultaneously satisfies three requirements: (1) high coding efficiency, (2) low coding complexity, and (3) being a streaming code (defined as one with low decoding latency). We propose METTLE (Multi-Edge Type with Touch-less Leading Edge), the first erasure code to meet all three requirements. Compared to "streaming RaptorQ" (RaptorQ configured with a small source block size to ensure a low decoding latency), METTLE is only slightly worse in coding efficiency, but 47.7 to 84.6 times faster to decode.

Figures (2)

• Figure 1: Spatial coupling in Walzer's scheme.
• Figure 2: Illustration of MET coupling.