Dominik Kaspar

Improving the performance of quality-adaptive video streaming over multiple heterogeneous access networks

Devices capable of connecting to multiple, overlapping networks simultaneously are becoming increasingly common. For example, most laptops are equipped with LAN- and WLAN-interfaces, and smart phones can typically connect to both WLANs and 3G mobile networks. At the same time, streaming high-quality video is becoming increasingly popular. However, due to bandwidth limitations or the unreliable and unpredictable nature of some types of networks, streaming video can be subject to frequent periods of rebuffering and characterised by a low picture quality. In this paper, we present a client-side request scheduler that distributes requests for the video over multiple heterogeneous interfaces simultaneously. Each video is divided into independent segments with constant duration, enabling segments to be requested over separate links, utilizing all the available bandwidth. To increase performance even further, the segments are divided into smaller subsegments, and the sizes are dynamically calculated on the fly, based on the throughput of the different links. This is an improvement over our earlier subsegment approach, which divided segments into fixed size subsegments. Both subsegment approaches were evaluated with on-demand streaming and quasi-live streaming. The new subsegment approach reduces the number of playback interruptions and improves video quality significantly for all cases where the earlier approach struggled. Otherwise, they show similar performance.

Using HTTP Pipelining to Improve Progressive Download over Multiple Heterogeneous Interfaces

Today, mobile devices like laptops and cell phones often come equipped with multiple network interfaces, enabling clients to simultaneously connect to independent access networks. Even though applications, such as multimedia streaming and video-on-demand delivery systems, could potentially benefit greatly from the aggregated bandwidth, implementation and standardization challenges have so far hindered the deployment of multilink solutions. Previously, we have explored the benefits of collaboratively using multiple Internet connections to progressively download and play back large multimedia files. In this paper, we present an improved version of our approach that utilizes HTTPs capability of request pipelining in combination with range retrieval requests. While, in our earlier work, the optimal choice of file segmentation size presented a tradeoff between throughput and startup latency, the enhanced solution is able to overcome this tradeoff. The use of very small segments no longer impairs the efficiency of throughput aggregation, which additionally makes our solution robust against link variances and agnostic to network heterogeneity.

A network-layer proxy for bandwidth aggregation and reduction of IP packet reordering

With todays widespread deployment of wireless technologies, it is often the case that a single communication device can select from a variety of access networks. At the same time, there is an ongoing trend towards integration of multiple network interfaces into end-hosts, such as cell phones with HSDPA, Bluetooth and WLAN. By using multiple Internet connections concurrently, network applications can benefit from aggregated bandwidth and increased fault tolerance. However, the heterogeneity of wireless environments introduce challenges with respect to implementation, deployment, and protocol compatibility. Variable link characteristics cause reordering when sending IP packets of the same flow over multiple paths. This paper introduces a multilink proxy that is able to transparently stripe traffic destined for multihomed clients. Operating on the network layer, the proxy uses path monitoring statistics to adapt to changes in throughput and latency. Experimental results obtained from a proof-of-concept implementation verify that our approach is able to fully aggregate the throughput of heterogeneous downlink streams, even if the path characteristics change over time. In addition, our novel method of equalizing delays by buffering packets on the proxy significantly reduces IP packet reordering and the buffer requirements of clients.

Mobile video streaming using location-based network prediction and transparent handover

A well known challenge with mobile video streaming is fluctuating bandwidth. As the client devices move in and out of network coverage areas, the users may experience varying signal strengths, competition for the available resources and periods of network outage. These conditions have a significant effect on video quality. In this paper, we present a video streaming solution for roaming clients that is able to compensate for the effects of oscillating bandwidth through bandwidth prediction and video quality scheduling. We combine our existing adaptive segmented HTTP streaming system with 1) an application layer framework for creating transparent multi-link applications, and 2) a location based QoS information system containing GPS coordinates and accompanying bandwidth measurements, populated through crowd-sourcing. Additionally, we use real-time traffic information to improve the prediction by, for example, estimating the length of a commute route. To evaluate our prototype, we performed real-world experiments using a popular tram route in Oslo, Norway. The client connected to multiple networks, and the results show that our solution increases the perceived video quality significantly. Also, we used simulations to evaluate the potential of aggregating bandwidth along the route.

An analysis of the heterogeneity and IP packet reordering over multiple wireless networks

With the increasing deployment of wireless technologies, such as WLAN, HSDPA, and WiMAX, it is often the case that simultaneous coverage of several access networks is available to a single user device. In addition, devices are also often equipped with multiple network interfaces. Thus, if we can exploit all available network interfaces at the same time, we can obtain advantages like the aggregation of bandwidth and increased fault tolerance. However, the heterogeneity and dynamics of the links also introduce challenges. Due to different link delays, sending packets of the same flow over multiple heterogeneous paths causes the reordering of packets.

Quality-adaptive scheduling for live streaming over multiple access networks

Video streaming ranks among the most popular services offered through the Internet today. At the same time, accessing the Internet over public WiFi and 3G networks has become part of our everyday lives. However, streaming video in wireless environments is often subject to frequent periods of rebuffering and characterized by low picture quality. In particular, achieving smooth and quality-adaptive streaming of live video poses a big challenge in mobile scenarios. Building on the observation that the subjective video experience on mobile devices decreases when quality changes are more frequent than every 1 to 2 seconds, we present a client-side scheduler that retrieves segments of several video encodings over heterogeneous network interfaces simultaneously. By extending the DAVVI streaming platform with support for multiple interfaces, the proposed schedulers performance is experimentally evaluated. The results show that our scheduler reduces the video interruptions and achieves a higher and more stable average quality over multiple, truly heterogeneous wireless interfaces.

Enhancing Video-on-Demand Playout over Multiple Heterogeneous Access Networks

Multimedia streaming is increasing in popularity and has become one of the dominating services on the Internet today. Even though user devices are often equipped with multiple network interfaces and in reach of several access networks at the same time, media streams are normally communicated over only one of the available Internet connections. In this paper, we explore the challenges and potential benefits of using multiple access networks simultaneously. Exploiting HTTPs capability of handling requests for specific byte ranges of a file, we present the implementation of a lightweight, application-layer, on-demand streaming service that requires no changes to existing servers and infrastructure. Based on real-world experiments with a multihomed host, we investigate the potential performance gains of video-on-demand playout. We achieve a bandwidth aggregation efficiency of 90% when downloading over 3 heterogeneous access networks in parallel. In addition, we analyze the effect of file segmentation on the buffer requirements and the startup latency.

Using bandwidth aggregation to improve the performance of quality-adaptive streaming

Devices capable of connecting to multiple, overlapping networks simultaneously is becoming increasingly common. For example, most laptops are equipped with LAN- and WLAN-interface, and smart phones can typically connect to both WLANs and 3G mobile networks. At the same time, streaming high-quality video is becoming increasingly popular. However, due to bandwidth limitations or the unreliable and unpredictable nature of some types of networks, streaming video can be subject to frequent periods of rebuffering and characterized by a low picture quality. In this paper, we present a multilink extension to the data retrieval part of the DAVVI adaptive, segmented video streaming system. DAVVI implements the same core functionality as the MPEG DASH standard. It uses HTTP to retrieve data, segments video, provides clients with a description of the content, and allows clients to switch quality during playback. Any DAVVI-data retrieval extensions can also be implemented in a DASH-solution. The multilink-enabled DAVVI client divides video segments into smaller subsegments, which are requested over multiple interfaces simultaneously. The size of each subsegment is dynamic and calculated on the fly, based on the throughput of the different links. This is an improvement over our earlier subsegment approach, which divided segments into fixed size subsegments. The quality of the video is adapted based on the measured, aggregated throughput. Both the static and the dynamic subsegment approaches were evaluated with on-demand streaming and quasi-live streaming. The new subsegment approach reduces the number of playback interruptions and improves video quality significantly for all cases where the earlier approach struggled. Otherwise, they show similar performance.

Using multiple links to increase the performance of bandwidth-intensive UDP-based applications

Networked devices often come equipped with multiple network interfaces, and bandwidth aggregation is one of the many possible benefits of using multiple interfaces simultaneously. Real-world networks introduce several challenges that have often been ignored by related work on bandwidth aggregation. The challenges include limited connectivity due to NAT-boxes, link heterogeneity and link variability. In this paper, we present a transparent solution for proxy-based bandwidth aggregation that is able to overcome the different deployment and link heterogeneity challenges present in real-world networks. Our focus has been on increasing the performance of bandwidth-intensive UDP-based applications, and through evaluation we show that our solution efficiently aggregates bandwidth and increases the in-order throughput. Previously, we introduced a multi-link UDP proxy solution that improves in-order throughput. This paper presents a significant extension and improvement in terms of support for middle-boxes (NAT), congestion control, a client-based resequencer and support for all operating systems.

Demo: quality-adaptive video streaming with dynamic bandwidth aggregation on roaming, multi-homed clients

Video streaming has become one of the most popular services on the Internet today. At the same time, mobile devices capable of displaying high-quality multimedia content have become common. In addition, most of these devices are equipped with multiple network interfaces. Smartphones and tablets can typically connect to both 3G networks (HSDPA) and WLANs, while laptops at least have a LAN and a WLAN interface, amongst others. Even though a mobile device is capable of displaying highquality video, it will in many cases be connected to networks unable to meet the requirements of the stream or the user’s expectations. For example, the bitrate of H.264 encoded 1080p video is around 6-8 Mbit/s. This might not be a problem when connected to a fixed network. However, wireless networks are often unable to support this bandwidth, causing periods of rebuffering and playback interruptions. Also, mobile devices typically roam between networks. Unless supported by the application, the streaming will stop when the connection to the previous network is lost. The deployment of different wireless networks have increased rapidly. A device is therefore often within coverage range of multiple networks simultaneously. For example, several telephone providers allow their subscribers to access WiFi hotspots. Using networks simultaneously offers several benefits, for example, bandwidth aggregation and increased connection reliability. Bandwidth aggregation means merging the bandwidth of several interfaces. We have developed a video streaming solution for mobile, roaming devices that makes use of the benefits offered by multiple links. To the best of our knowledge, no such solution currently exists. We use HTTP Adaptive Streaming, also used by for example Microsoft and Apple, which allows a client to change quality based on the available bandwidth. In order to do bandwidth aggregation, each video segment is divided into smaller (logical) subsegments using HTTP range-requests, and the subsegments are requested over the different links. Our client application automatically detects when links are made available or become unavailable, and reacts accordingly. Bandwidth measurements, used to select video quality, are requested from a lookup database, using the client’s GPS coordinates as parameter. This database is populated using crowdsourcing and is consulted at given intervals, as the coverage range of wireless networks and