Selection of network coding nodes for minimal playback delay in streaming overlays

📝 Abstract

Network coding permits to deploy distributed packet delivery algorithms that locally adapt to the network availability in media streaming applications. However, it may also increase delay and computational complexity if it is not implemented efficiently. We address here the effective placement of nodes that implement randomized network coding in overlay networks, so that the goodput is kept high while the delay for decoding stays small in streaming applications. We first estimate the decoding delay at each client, which depends on the innovative rate in the network. This estimation permits to identify the nodes that have to perform coding for a reduced decoding delay. We then propose two iterative algorithms for selecting the nodes that should perform network coding. The first algorithm relies on the knowledge of the full network statistics. The second algorithm uses only local network statistics at each node. Simulation results show that large performance gains can be achieved with the selection of only a few network coding nodes. Moreover, the second algorithm performs very closely to the central estimation strategy, which demonstrates that the network coding nodes can be selected efficiently in a distributed manner. Our scheme shows large gains in terms of achieved throughput, delay and video quality in realistic overlay networks when compared to methods that employ traditional streaming strategies as well as random network nodes selection algorithms.

💡 Analysis

Network coding permits to deploy distributed packet delivery algorithms that locally adapt to the network availability in media streaming applications. However, it may also increase delay and computational complexity if it is not implemented efficiently. We address here the effective placement of nodes that implement randomized network coding in overlay networks, so that the goodput is kept high while the delay for decoding stays small in streaming applications. We first estimate the decoding delay at each client, which depends on the innovative rate in the network. This estimation permits to identify the nodes that have to perform coding for a reduced decoding delay. We then propose two iterative algorithms for selecting the nodes that should perform network coding. The first algorithm relies on the knowledge of the full network statistics. The second algorithm uses only local network statistics at each node. Simulation results show that large performance gains can be achieved with the selection of only a few network coding nodes. Moreover, the second algorithm performs very closely to the central estimation strategy, which demonstrates that the network coding nodes can be selected efficiently in a distributed manner. Our scheme shows large gains in terms of achieved throughput, delay and video quality in realistic overlay networks when compared to methods that employ traditional streaming strategies as well as random network nodes selection algorithms.

📄 Content

1 Selection of network coding nodes for minimal playback delay in streaming overlays Nicolae Cleju∗, Nikolaos Thomos† and Pascal Frossard† †Signal Processing Laboratory (LTS4), Ecole Polytechnique F´ed´erale de Lausanne (EPFL), Lausanne, Switzerland ∗The “Gheorghe Asachi” Technical University of Iasi, Iasi Romania. nikcleju@etti.tuiasi.ro, {nikolaos.thomos,pascal.frossard}@epfl.ch Abstract—Network coding permits to deploy distributed packet delivery algorithms that locally adapt to the network availability in media streaming applications. However, it may also increase delay and computational complexity if it is not implemented efficiently. We address here the effective placement of nodes that implement randomized network coding in overlay networks, so that the goodput is kept high while the delay for decoding stays small in streaming applications. We first estimate the decoding delay at each client, which depends on the innovative rate in the network. This estimation permits to identify the nodes that have to perform coding for a reduced decoding delay. We then propose two iterative algorithms for selecting the nodes that should perform network coding. The first algorithm relies on the knowledge of the full network statistics. The second algorithm uses only local network statistics at each node. Simulation results show that large performance gains can be achieved with the selection of only a few network coding nodes. Moreover, the second algorithm performs very closely to the central estimation strategy, which demonstrates that the network coding nodes can be selected efficiently in a distributed manner. Our scheme shows large gains in terms of achieved throughput, delay and video quality in realistic overlay networks when compared to methods that employ traditional streaming strategies as well as random network nodes selection algorithms. Index Terms—Network coding, delay minimization, through- put maximization, overlay networks. I. INTRODUCTION The recent development of overlay networks offers interest- ing perspectives for multimedia streaming applications, since network diversity can be used advantageously for improved quality of service. The traditional streaming systems based on ARQ or channel coding techniques however generally fail to efficiently exploit this diversity. They either suffer from relatively high computational cost, require coordination between network nodes or lead to suboptimal performance in large scale networks where local channel conditions are hard to estimate. A different paradigm has been initiated recently with network coding [1], [2], where some processing is requested from the network nodes in order to improve the packet delivery performance. Specifically, network coding nodes combine buffered packets before forwarding them to next hop nodes. This coding strategy is particularly appealing in distributed streaming systems, as it removes the need for reconciliation between nodes. It locally adapts to the available bandwidth and packet loss rate and even permits to approach This work has been partly supported by the Swiss National Science Foundation, under grant PZ00P2-121906. max-flow min-cut bound of the network graph. Overall, the network coding systems have shown improved resiliency to dynamics, delays, scalability and buffer capacities in networks with diversity [3]. The application of network coding algorithms in multimedia streaming systems is however not straightforward. Specifically, multimedia streaming imposes strict timing constraints that impact the design of network coding algorithms. A practical network coding system has been presented in [4] and ad- dresses the specific characteristics of streaming applications. It implements randomized network coding (RNC) techniques [5] in the network nodes and devises a protocol to deal with buffering issues and timing constraints. Moreover, it introduces the concept of generations that restricts coding operations to packets that share similar decoding deadlines. However, network coding systems still face important issues in practical systems due to the decoding delays imposed by successive network coding operations. This delay as well as the com- putational overhead in the system grow with the number of network coding nodes. It becomes therefore important to select efficiently the subset of nodes that perform network coding in order to control delay and complexity and still exploit efficiently the diversity in the network. In this paper, we discuss solutions for the selective place- ment of a few network coding nodes in order to reduce the delay for video delivery. The nodes in the network are cate- gorized into network coding (NC) and store and forward (SF) nodes. The network coding nodes use the practical network coding algorithm described in [4], which has been selected for its effectiveness and simplicity. Similarly, we adopt the concept of coding generation and buffer models [4] for proper handling of the timing constraints in the s

This content is AI-processed based on ArXiv data.