Resource allocation for WWAN video multicast with cooperative local repair
Resource allocation for WWAN video multicast with cooperative local repair
复制标题
DOI：
10.1109/icc.2013.6655572
复制
发表时间：
2013-06
期刊：
影响因子：
--
通讯作者：
Zhi Liu;Yu Mao;N. Lu;Yusheng Ji;Xuemin Shen
中科院分区：
文献类型：
--
作者：
Zhi Liu;Yu Mao;N. Lu;Yusheng Ji;Xuemin Shen
当前的编码标准采用差分编码来降低编码速率。解码错误可能在丢失不可恢复帧之后的后续帧中传播。由于大多数当前的电子设备都是用多个接口实现的，例如3G和WiFi，解决错误传播问题的一种方法是使用合作策略，该策略利用客户端信道的“不相关性”（不同的信道可能经历不同的信道损耗）。协作策略通过相邻客户端通过第二网络接口共享（客户端到客户端）分组来帮助修复主网络的（基站到客户端）丢失，并将调制和编码方案最优地分配给有限的网络资源，可以大幅度提高客户端的接收质量。其内在问题是如何用合作策略解决资源分配问题。在本文中，我们共同考虑资源分配和视频流的合作修复。我们将问题转化为优化问题，即最大化客户端的接收视频质量，并应用结构化网络编码来进一步提高整体性能。仿真结果表明，该方法优于其他竞争方案至少1.5dB的接收视频质量在典型的网络场景。
The current coding standard employs differential coding to reduce the coding rate. Decoding error may propagate in the following frames after a loss-unrecoverable frame. Since most of the current electronic devices are implemented with multiple interfaces, such as the 3G and WiFi, one way to solve the error propagation problem is to use cooperative strategies, which leverage on the `uncorrelatedness' of the clients' channels (different channels may experience different channel losses). The cooperative strategies helping to repair the primary network's (base station to client) losses via the neighboring clients' packets sharing (client to client) through the second network interface and optimally allocating the modulation and coding schemes to the limited network resource can improve the clients' received quality massively. The inherent problem is how to solve the resource allocation problem with the cooperative strategies. In this paper, we jointly consider the resource allocation and the cooperative repair for the video streaming. We formulate the problem into the optimization problem, i.e. maximizing the clients' received video quality, and apply the structured network coding to further improve the overall performance. Simulation results show that the proposed approach outperforms the other competing schemes by at least 1.5dB in term of the received video quality in typical network scenarios.