Network coding (NC) is a novel forwarding technique that promise to change many aspects of networking. However, the error propagation problem in NC resulting from the encoding on the Network layer makes the error-correcting process performed classically on the Physical layer very expensive in terms of computational resources, especially when the rank of the errors is relatively high. In this paper we define Error Correction Network Coding (ECNC), a new approach to error correction for the Network layer in NC based schemes which introduces redundancy in the time domain instead of the space domain as it is the case for error-correcting solutions proposed for NC. We describe the encoding and decoding algorithms and give examples of both linear and non-linear ECNCs, then we discuss their error detection and correcting capacities. Furthermore, we show that ECNCs can be used jointly with the space domain error-correcting codes proposed for NC.

网络编码 (NC) 是一种新颖的转发技术，有望改变网络的许多方面。然而，网络层编码导致的 NC 中的错误传播问题使得物理层上经典的纠错过程在计算资源方面非常昂贵，尤其是当错误的等级相对较高时。在本文中，我们定义了纠错网络编码(ECNC)，一种在基于 NC 的方案中对网络层进行纠错的新方法，它在时域而不是空间域中引入了冗余，因为它是为 NC 提出的纠错解决方案的情况。我们描述了编码和解码算法，并给出了线性和非线性 ECNC 的例子，然后我们讨论了它们的错误检测和纠正能力。此外，我们表明 ECNC 可以与为 NC 提出的空间域纠错码联合使用。