Blind identification of an unknown interleaved convolutional code

TL;DR

This paper presents an efficient method for reconstructing unknown interleavers and recovering convolutional codes from noisy, interleaved codewords without prior assumptions on interleaver structure, demonstrating effectiveness through experiments.

Contribution

The paper introduces a novel, assumption-free approach to identify interleavers and recover convolutional codes from noisy data, improving robustness and applicability.

Findings

Effective reconstruction of interleavers even with moderate noise

Successful recovery of convolutional codes from noisy interleaved data

Method demonstrates high efficiency in experimental tests

Abstract

We give here an efficient method to reconstruct the block interleaver and recover the convolutional code when several noisy interleaved codewords are given. We reconstruct the block interleaver without assumption on its structure. By running some experimental tests we show the efficiency of this method even with moderate noise.