Next Mobile Network research aspects Distributed network monitoring and configuration Adaptive media streaming over the internet Media-aware rate allocation on multipath networks Forward error correction for multipath streaming applications Media packet scheduling over multiple transmission paths Next Mobile Network research aspects For more information on the research projects please visit the company's official website. Distributed network monitoring and configuration We are interested in the efficient monitoring of local variables in large scale scenarios, where centralized solutions are unpractical due to the huge overhead incurred at the monitoring station. Distributed alternatives for network monitoring and configuration include tree-based or gossiping solutions, where nodes self-organize into meaningful management overlays which relay aggregate information. Such solutions, together with error filter or rate control schemes exhibit good trade-offs between the accuracy of the monitoring objective and management overhead. We present a protocol for the continuous monitoring of a local network state variable. Our aim is to provide a management station with the value distribution of the local variables across the network, by means of partial histogram aggregation, with minimum protocol overhead. Our protocol is decentralized and asynchronous to achieve robustness and scalability, and it executes on an overlay interconnecting management processes in network devices. On this overlay, the protocol maintains a spanning tree and updates the histogram of the network state variables through incremental aggregation. The protocol allows to control the trade-off between protocol overhead and a global accuracy objective. This functionality is implemented by a dynamic configuration of local error filters that control whether an update is sent towards the management station or not. We evaluate our protocol by means of simulations. Our results demonstrate the controllability of our method in a wide selection of scenarios, and the scalability of our protocol for large-scale networks. Dan Jurca and Rolf Stadler Adaptive media streaming over the internet With the advances in audio-visual encoding standards and broadband access networks, multimedia communications (MMC) is becoming quite popular. The continuing expansion of the Internet further stimulates the demand for MMC services. However, real-time media streaming over lossy networks (such as Internet), still poses several challenging problems because of the adverse effects caused by complex network dynamics. Within the general framework of a best-effort network, my research goal is to develop adaptive media streaming scenarios that manage an improved media quality at the end receiver. In a general streaming scenario, a streaming server must send stored or live media to the client. The information is transmitted over the internet, a best effort network, that does not offer any service guarantees. The client must be able to consume the received media after an initial playback delay, without suffering interruptions or severe quality degradation. The underlying network represents a challenging transmission environments for the media content, which normally requires fixed quality of service guarantees. Time varying transmission rates, fluctuating delays or variable losses inside the network pose a serious threat for any "stupid" media delivery system. However, an in-depth analysis of these network parameters, combined with a smart manipulation of the media content can provide sufficient information in order to create new adaptive delivery frameworks that can greatly enhance the received media quality at the receiver. Possible solutions can contain a combination of flexible video encoding techniques (MDC, layers, FGS), error concealment or error correction strategies, and transmission techniques (packet scheduling, rate smoothing). Dan Jurca, Jakob Chakareski, Jean-Paul Wagner and Pascal Frossard Media-aware rate allocation on multipath networks We address the problem of joint path selection and rate allocation in multipath streaming in order to optimize a media specific quality of service. An optimization problem is proposed, which aims at minimizing a video distortion metric based on sequence-dependent, and transmission channel parameters, for a given overlay network infrastructure. Even if in general, optimal path selection and rate allocation is an NP complete problem, an in-depth analysis of the media distortion evolution allows to define a low complexity algorithm for an optimal streaming strategy. In particular, we show that a greedy allocation of rate along paths with increasing error probability leads to an optimal solution for the most common network scenarios. We argue that a network path shall not be chosen for transmission, unless all other available paths with lower error probability have been chosen. Moreover, the chosen paths should be used at their maximum end-to-end bandwidth. Simulation results show that the optimal rate allocation carefully trades off total encoding/transmission rate, with the end-to-end transmission error probability and the number of chosen paths. In many cases, the optimal rate allocation provides more than 20% improvement in received video quality, compared to heuristic-based algorithms. This justifies the use of a low complexity rate allocation algorithm, that optimizes media specific quality of service, and simultaneously saves network resources. Dan Jurca and Pascal Frossard Dan Jurca and Pascal Frossard Forward error correction for multipath streaming applications We address the problem of joint optimal rate allocation and scheduling between media source rate and error protection rate in scalable streaming applications over lossy multipath networks. Starting from a distortion representation of the received media information at the client, we propose a novel optimization framework in which we analyze the performance of the most relevant FEC and scheduling techniques. We describe both optimal and heuristic algorithms that find solutions to the rate allocation and scheduling problem, and emphasize the main characteristics of the compared techniques. Our results show that efficient unequal error protection schemes that better protect the most important media information greatly improve the quality of the streaming process. At the same time we emphasize the importance of priority scheduling of the information over the best available network paths, which outperforms traditional FIFO or network flooding mechanisms. Dan Jurca, Pascal Frossard and Aleksandar Jovanovic Media packet scheduling over multiple transmission paths We address the problem of packet scheduling in a multipath network scenario, with the final goal of optimizing the received quality of the client. A network topology with multiple transmission paths offers several benefits to the streaming process, including aggregated bandwidth on the multiple disjoint paths, loss decorrelation and flexibility in dealing with delays and network jitter. We use a video encoding abstraction that takes into account the video packet sizes, importances in terms of distortion, packet dependencies and individual decoding deadlines. By developing a detailed analysis on the multipath packet scheduling, we can derive optimal and heuristic algorithms that achieve the transmission of the video packets over the network taking into account a given playback delay constrained imposed by the client. We show that our heuristic approach performs close to optimal, even in adverse network conditions. In the same time, our solutions complexity stays low. We show significant results improvements compared with other state of the art scheduling mechanisms. Dan Jurca and Pascal Frossard |
