Ozgur Oyman

Quality of experience for HTTP adaptive streaming services

The growing consumer demand for mobile video services is one of the key drivers of the evolution of new wireless multimedia solutions requiring exploration of new ways to optimize future wireless networks for video services towards delivering enhanced quality of experience (QoE). One of these key video enhancing solutions is HTTP adaptive streaming (HAS), which has recently been spreading as a form of Internet video delivery and is expected to be deployed more broadly over the next few years. As a relatively new technology in comparison with traditional push-based adaptive streaming techniques, deployment of HAS presents new challenges and opportunities for content developers, service providers, network operators and device manufacturers. One of these important challenges is developing evaluation methodologies and performance metrics to accurately assess user QoE for HAS services, and effectively utilizing these metrics for service provisioning and optimizing network adaptation. In that vein, this article provides an overview of HAS concepts, and reviews the recently standardized QoE metrics and reporting framework in 3GPP. Furthermore, we present an end-to-end QoE evaluation study on HAS conducted over 3GPP LTE networks and conclude with a discussion of future challenges and opportunities in QoE optimization for HAS services.

Multihop Relaying for Broadband Wireless Mesh Networks: From Theory to Practice

We summarize capacity results to show merits of multihop relaying in broadband cellular mesh networks. Under the guidance of these results, we provide design perspectives on relay deployment, spectrum allocation, and end-to-end optimization of certain QoS measures such as throughput, coverage, reliability, and robustness. We conclude with an overview of recent standardization activities and remarks on remaining open problems and design challenges.

Toward enhanced mobile video services over WiMAX and LTE [WiMAX/LTE Update]

Wireless networks are on the verge of a third phase of growth. The first phase was dominated by voice traffic, and the second phase, which we are currently in, is dominated by data traffic. In the third phase we predict that the traffic will be dominated by video and will require new ways to optimize the network to prevent saturation. This increase in video traffic is one of the key drivers of the evolution to new mobile broadband standards like WiMAX IEEE 802.16m and 3G LTE and LTE-Advanced, motivating the need to enhance the video service capabilities of future cellular and mobile broadband systems. Therefore, it is important to understand both the potential and limitations of these networks for delivering video content in the future, which will include not only traditional video broadcasts, but also video streaming and uploading in the uplink direction. In that vein this article provides an overview of technology options for enabling multicast and unicast video services over WiMAX and LTE networks, quantifies and compares the video capacities of these networks in realistic environments, and discusses new techniques that could be exploited in the future to further enhance the video capacity and quality of user experience.

Video capacity and QoE enhancements over LTE

Quality of Experience (QoE) has taken a center stage in the performance evaluation of multimedia delivery technologies. The Third Generation Partnership Project (3GPP) Long Term Evolution (LTE) system is the latest generation of wireless cellular technology expected to deliver higher data rates and meet the burgeoning data demand. With the projected dominant share of video services in mobile traffic, providing satisfactory QoE to video users is a key objective for LTE system design. In this paper, we present a QoE-based evaluation methodology to assess the LTE system video capacity in terms of the number of unicast video consumers that can be simultaneously supported for a given target QoE. We define and use the notion of rebuffering outage capacity to quantify the video service capacity. Our evaluation further incorporates adaptive streaming, a promising technology for video delivery over wireless, and presents its consequent QoE-capacity tradeoff. The impact of QoE-based outage criteria is also investigated on the downlink video capacity. Finally, we propose a QoE-aware radio resource management (RRM) framework which allows the network operator to further enhance the video capacity. Our results demonstrate that there is a significant potential to optimize video capacity through QoE awareness both at the application level and radio access network (RAN) level.

Opportunistic scheduling and spectrum reuse in relay-based cellular networks

In order to understand the key merits of multiuser diversity, opportunistic scheduling and spectrum reuse techniques in downlink relay-based cellular networks, this paper analyzes the spectral efficiency performance over a fading multihop broadcast channel (MBC) composed of a base station (BS), a relay station (RS), U far users and V near users in the asymptotic regime of large number of users. Using tools from extreme-value theory, we characterize the average spectral efficiency of the MBC as a function of the number of users and physical channel parameters. Our analysis yields very accurate formulas even for moderately low values of U and V , specified in detail and verified (via Monte Carlo simulations) for the case of Rayleigh fading. Next, we consider a relay-based broadband multi-cellular network in the downlink mode; with special focus on orthogonal frequency-division multiple-access (OFDMA) resource allocation and investigate the spectral efficiency performance of opportunistic multiuser scheduling and spectrum reuse techniques based on capacity analysis and system-level simulations. This empirical study validates our analytical insights and helps to further identify design tradeoffs associated with spectrum reuse, interference management and multiuser diversity techniques in relay-based cellular networks.

A Shannon-Theoretic Perspective on Fading Multihop Networks

We consider a frequency-flat fading multihop network with a single active source-destination pair terminals communicating over multiple hops through a set of intermediate relay terminals. We use Shannon-theoretic tools to analyze the tradeoff between energy efficiency and spectral efficiency (known as the power-bandwidth tradeoff) for a simple communication protocol based on time-division decode-and-forward relaying in meaningful asymptotic regimes of signal-to-noise ratio (SNR) under a system-wide power constraint on source and relay transmissions. The impact of multi-hopping and channel fading on the key performance measures of the high and low SNR regimes is investigated to shed new light on the possible enhancements in power/bandwidth efficiency and link reliability. In contrast to the common belief that in fading environments communicating over multiple hops suffers significantly in performance due to the worst link limitation, our results indicate that hopping could significantly improve the outage behavior over slow-fading networks and stabilize links against the random channel fluctuations. In particular, we prove that there exists an optimal number of hops that minimizes the end-to-end outage probability and characterize the dependence of this optimal number on the fading statistics and target energy and spectral efficiencies.

OFDM2A: A Centralized Resource Allocation Policy for Cellular Multi-hop Networks

We study the problem of joint scheduling and routing in the context of multi-hop communication over a relay-assisted broadband cellular network from an information-theoretic view. We propose orthogonal frequency division multi-hop multiple-access (OFDM2A) as a novel resource allocation policy; under which the time and frequency resources of the network are orthogonally shared among multiple users over multiple hops. This scheme provides a very simple and reliable framework for resource allocation over cellular multi-hop networks, ensuring interference-free multi-user communication. Furthermore, the novel principle of separating subcarrier allocation and multi-hop route selection allows for the design of low-complexity centralized opportunistic scheduling algorithms based on end-to-end route cost metrics, simultaneously realizing multi-user diversity and multi-hop diversity gains and thereby enhancing cellular capacity and coverage.

Opportunism in Multiuser Relay Channels: Scheduling, Routing and Spectrum Reuse

In order to understand the key merits of multiuser diversity techniques in relay-assisted cellular multihop networks, this paper analyzes the spectral efficiency of opportunistic (i.e., channel-aware) scheduling algorithms over a fading multiuser relay channel with K users in the asymptotic regime of large (but finite) number of users. Using tools from extreme-value theory, we characterize the limiting distribution of spectral efficiency focusing on Type I convergence and utilize it in investigating the large system behavior of the multiuser relay channel as a function of the number of users and physical channel signal- to-noise ratios (SNRs). Our analysis results in very accurate formulas in the large (but finite) K regime, provides insights on the potential performance enhancements from multihop routing and spectrum reuse policies in the presence of multiuser diversity gains from opportunistic scheduling and helps to identify the regimes and conditions in which relay-assisted multiuser communication provides a clear advantage over direct multiuser communication.

Video-QoE aware radio resource allocation for HTTP adaptive streaming

We consider the problem of scheduling multiple adaptive streaming video users sharing a time-varying wireless channel such as in modern 3GPP Long Term Evolution (LTE) systems. HTTP Adaptive Streaming (HAS) framework is used by the video users for video rate adaptation. HAS is a client driven video delivery solution that has been gaining popularity due to its inherent advantages over other existing solutions. Quality of Experience (QoE) has become the prime performance criterion for media delivery technologies and wireless resource management is a critical part in providing a target QoE for video delivery over wireless systems. We propose a novel cross-layer Video-QoE aware optimization framework for wireless resource allocation that constraints rebuffering probability for adaptive streaming users. We propose a Re-buffering Aware Gradient Algorithm (RAGA) to solve this optimization problem. RAGA relies on simple periodic feedback of media buffer levels by adaptive streaming clients. Our simulation results on an LTE system level simulator demonstrate significant reduction in re-buffering percentage using RAGA without compromising video quality.

QoE evaluation for video streaming over eMBMS

As the multicast standard for Long Term Evolution (LTE), Enhanced Multimedia Broadcast Multicast Service (eMBMS) was introduced by Third Generation Partnership Project (3GPP) to facilitate delivery of popular content to multiple users over a cellular network. If a large number of users is interested in the same content, for e.g., live sports, news clips, etc., multicast transmission can significantly lower the cost and make better use of the spectrum as compared to unicast transmission. In this paper, we study and analyze the quality of experience (QoE) at the end user during live video streaming over eMBMS. We consider a comprehensive end-to-end MBMS streaming frame-work based on H.264/AVC encoded video content delivered in a chunked format over multiple segments using the File Delivery over Unidirectional Transport (FLUTE) protocol, combined with application layer (AL) forward error correction (FEC) based on Raptor and RaptorQ codes. Specifically, our study involves QoE evaluation in terms of startup delay, rebuffering percentage and Peak Signal-to-Noise Ratio (PSNR) metrics and provides performance evaluations to characterize the impact of various MBMS streaming, transport and AL-FEC configurations on the end user QoE. We also propose a decoding strategy that takes into consideration the systematic structure of AL-FEC codes to enhance performance when a source block decoding fails.