Streaming Delay Research Articles

Video Content Placement At the Network Edge: Centralized and Distributed Algorithms

In the traditional video streaming service provisioning paradigm, viewers typically request video content through a central Content Delivery Network (CDN) server. However, because of the uncertain wide area network delays, the (remote) viewers usually suffer from long video streaming delay, which affects the quality of experience. Multi-Access Edge Computing (MEC) offers a way to shorten the video streaming delay by building small-scale cloud infrastructures at the network edge, which are in close proximity to the viewers. In this paper, we present novel centralized and distributed algorithms for the video content placement problem in MEC. In the proposed centralized video content placement algorithm, we leverage the Lyapunov optimization technique to formulate the video content placement problem as a series of one-time-slot optimization problems and apply an Alternating Direction Method of Multipliers (ADMM)-based method to solve each of them. We further devise a distributed Multi-Agent Reinforcement Learning (MARL)-based method with value decomposition mechanism and parallelization policy update method to solve the video content placement problem. The value Decomposition mechanism deals with the credit assignment among multiple agents, which promotes the cooperative optimization of the global target and reduces the frequency of information exchange. The parallelization of policy network can speed up the convergence process. Simulation results verify the effectiveness and superiority of our proposed centralized and distributed algorithms in terms of performance.

Delay optimization strategy based on aperiodic traffic in time-sensitive networking

Time-Sensitive Networking (TSN) is a cluster of protocols working in the data link layer of the OSI reference model, which has been introduced into the industrial Internet due to its ability to ensure the deterministic transmission of data. Through IEEE 802.1AS clock synchronization and IEEE 802.1 Qbv gating mechanism, TSN networks allocate separate transmission time slots for periodic signals to ensure low delay and jitter, but this increases the delay of aperiodic stream to some extent. In this paper, a scheduling scheme of Grouping transmission is proposed. By changing the sending time of periodic data stream, the periodic data stream which is originally continuously transmitted is divided into several groups to be transmitted in different time periods in a cycle, so as to reduce the real-time aperiodic data stream delay. Experimental results show that the proposed scheduling scheme can reduce the transmission delay of the non-periodic traffic without affecting the quality of service (QoS) of the periodic traffic.

Real Time Image Encoding for Fast IOT (Internet of Things) Based Video Vigilance System

In recent times, technology has played an important role in almost every field as it advances. Law and order situation in the third world countries are extensively poor, and it is imperative to aid the institutions providing security by incorporating systems capable of identifying perpetrators and catch them. It is a well-known fact that security and surveillance industry is evolving, and most of the system are based on video surveillance rather than the basic alarm based solutions. Such systems are integrated with multiple cameras which provide live video surveillance features and in turn increases the visibility of the security systems. The Internet of Things (IOT) is helping to create an efficient security system capable of acquiring live video streams from multiple sensory sources. Due to the lack of available bandwidth, it may not be possible to acquire a video stream with decent quality. In this paper, a compression algorithm is proposed based on discrete cosine transform (DCT) and temporal reduction using motion vectors extracted from the incoming frames in order to achieve best compromise between quality and streaming time to properly indicate the perpetrator. Most of the time, it becomes impossible to seize the perpetrator due to the inferior video quality which makes him unidentifiable. Degraded video quality is mostly due to over compression. Implementation of the proposed algorithm is performed in MATLAB environment by acquiring the live video streams of the videos in real time. Mean Squared Error (MSE) and Peak Signal to Noise Ratio are calculated in order to elaborate the quality difference between the reconstructed frames by conventional means and with the incorporated Temporal Masking technique. By incorporating the proposed algorithm in the compression technique, it was discovered that it actually reduces the stream delay by 0.104 s with an acceptable Structural Similarity Index (SSIM) difference between the reconstructed frames acquired by conventional means and by proposed Temporal Masking Technique.

Material flow influence on the quality of molded parts

This paper focuses on a new occurrence mechanism of the flow marks, which appear on the surface of injection-molded parts, through characterization of the part surface structure. Injection-molded parts involving long flow lengths can exhibit visible flow defects characterized by alternating glossy and dull bands on the polymer surface, which have approximately the same direction with the material flow front. These flow defects are considered to have strong negative effect on the appearance of the molded component and thus on the perceived quality. The reasons for the appearance of these stripes are not entirely clear. Defect flow marks (shiny areas) were cut from the molded part and analyzed using optical microscopy, Fourier transform infrared spectroscopy, laser Raman spectroscopy and scanning electron microscopy coupled with energy-dispersive X-ray (SEM/EDX) spectroscopy. Raman spectra detected differences between spectra of defect area and of the reference one. The EDX analysis revealed that the defect area contains slightly more quantity of talc (Mg and Si). This happened because of material flow morphology in the areas where the molded part presents ribs. The flow of melted resin through an enlargement (rib) results in a decrease in flow velocity, and consequently, a pressure rise will occur and it also involves a vortex development. Therefore, the shear stress in this area is higher than upstream and downstream ones, and it will cause a delay of the flow stream, a premature solidification of the melted resin and ribs flow marks occurrence.

Major Stream Delay under Limited Priority Conditions

AbstractUnder a limited priority situation, U-turning vehicles attempt to accept smaller gaps, which compels approaching through vehicles to slow down and experience delays. Delay within the median...

Investigation of major stream delays at unsignalized intersections

Investigation of major stream delays at unsignalized intersections

Low-Latency Routing Scheme for a Fronthaul Bridged Network

A fronthaul bridged network has attracted attention as a means of efficiently constructing the centralized radio access network (C-RAN) architecture. If we change the functional split of C-RAN and employ timedivision duplex (TDD), the data rate in fronthaul will become variable, and the global synchronization of front-haul streams will occur. This feature results in an increase in the queuing delay in fronthaul bridges among fronthaul flows. To address this problem, this paper proposes a novel low-latency routing scheme designed to minimize the worst-case delay in fronthaul networks with path-control protocols. The proposed scheme formulates the worst-case delay of each fronthaul stream based on the distribution of nodes, the propagation delay, and metric of the links. It selects the set of paths that minimizes the maximum value of the worst-case delay from the candidate path sets for fronthaul flows generated with the k-th shortest path algorithm. We confirmed with computer simulations that the proposed scheme can adequately minimize the worst-case delay, irrespective of the network topology. The maximum queuing delay is minimized by considering the time synchronization between fronthaul flows and the burst size determined by the TDD subframe length.

Modelling the major stream delay due to U-turns

This study assesses the delay faced by approaching through vehicles at uncontrolled median openings due to U-turning movement. Under limited priority situation, the U-turning vehicles accept smaller gaps which compel the through vehicles to reduce their speeds and experience delays. First, the slowdown section is identified and the delay faced by through vehicles from the starting of slowdown section to start of median opening is estimated. This is termed as approach delay. The approaching vehicles shift laterally as they approach toward the median opening. The concept of zone has been introduced to study this lateral movement. This study shows that the vehicles moving through the zone adjacent to curb face more delay as compared to the vehicles moving through the zone adjacent to median. Finally, a regression model is developed for estimating approach delay faced by any vehicle category traveling through a particular zone at a given U-turning traffic volume.

Dependency- and similarity-aware caching for HTTP adaptive streaming

There has been significant interest in the use of HTTP adaptive streaming for live or on-demand video over the Internet in recent years. To mitigate the streaming transmission delay and reduce the networking overhead, an effective and critical approach is to utilize cache services between the origin servers and the heterogeneous clients. As the underlying protocol for web transactions, HTTP has great potentials to explore the resources within state-of-the-art CDNs for caching; yet distinct challenges arise in the HTTP adaptive streaming context. After examining a long-term and large-scale adaptive streaming dataset as well as statistical analysis, we demonstrate that the switching requests among the different qualities frequently emerge and constitute a significant portion in a per-day view. Consequently, they have substantially affected the performance of cache servers and Quality-of-Experience (QoE) of viewers. In this paper, we propose a novel cache model that captures the dependency among the segments in the cache server for adaptive HTTP streaming. Our work does not assume any specific selection algorithm on the client’s side and hence can be easily incorporated into existing streaming cache systems. Its centralized nature is also well accommodated by the latest DASH specification. Moreover, we extend our work to the multi-server caching context and present a similarity-aware allocation mechanism to enhance the caching efficiency. The performance evaluation shows our dependency- and similarity-aware strategy can significantly improve the cache hit-ratio and QoE of HTTP streaming as compared to previous approaches.

Extent of Priority Stream Delays at Midblock Median Opening Zone

The study is concerned with investigating delays and queues in both flow directions of exclusive U-turn median openings zone. Note that other turning movements are not allowed at exclusive U-turn median openings zone. Travel time delay study was carried out at four sites in Johor, Malaysia. Resultsshow that average delay of 5s can be expected at divergingsection and no significant delay to conflicting traffic flows at the merging section. Delayswere recorded for vehicles waiting to merge at the exit lane. The paper concluded that weaving may be called to account for travel time delay at the diverging section. And also that since vehicle at the merging section must give way, there will be no delay to conflicting traffic flow.

An Architecture for QoS-Enabled Mobile Video Surveillance Applications in a 4G EPC and M2M Environment

Mobile video surveillance applications are used widely nowadays. They offer real-time video monitoring for homes, offices, warehouses, airports, and so on with live and pre-recorded on-demand video streaming. Quality of service (QoS) remains a key challenge faced by most of these applications. In this article, we propose an architecture for mobile video surveillance applications with a guaranteed and differentiated QoS support. The architecture relies on the 3GPP 4G evolved packet core (EPC). The main components are the QoS enabler, media server, and machine-to-machine gateway and surveillance application. To demonstrate its feasibility, a proof of concept prototype has been implemented and deployed. We also took measurements to evaluate the performance. Several lessons were learned. For instance, multimedia frameworks must allow for buffering controls in media streaming to reduce live streaming delay. In addition, we have learned that publicly available materials related to the EPC prototyping platform we have used (i.e., OpenEPC) are scarce. This has made our prototyping task rather difficult.

Beyond 1Mbps Global Overlay Live Streaming

In order to provide live streaming over the global Internet, a content provider often deploys an overlay network consisting of distributed proxies placed close to user pools. Streaming of multi-Mbps video over such an overlay is challenging because of bandwidth bottlenecks in paths. To effectively overcome these bottlenecks, we consider employing proxy helpers in the overlay to provide rich path diversity. The helpers do not have any attached users, and hence may forward partial video streams (or not at all) if necessary. In this way, the helpers serve as stepping stones to supply full streams to the servers. The issue is how to involve the helpers in the overlay to achieve low streaming delay meeting a certain high streaming bitrate requirement. To address the issue, we first formulate the problem which captures various delay and bandwidth components, and show that it is NP-hard. We then propose an efficient algorithm called Stepping-Stones (SS) which can be efficiently implemented in a controller. Given the encouraging simulation results, we develop a novel streaming testbed for SS and explore, through sets of Internet experiments, the effectiveness of helpers to achieve high bitrate (multi-Mbps) global live streaming. In our experiments, proxies are deployed with a reasonably wide global footprint. We collect more than a hundred hours of streaming traces with bitrate ranging from 500kbps to a few Mbps. Our experimental data validates that helpers indeed play an important role in achieving high bitrate in today's Internet. Global multi-Mbps streaming is possible due to their multihop and multipath advantages. Our experimental trials and data also provide valuable insights on the design of a global push-based streaming network. There are strong benefits of using proxy helpers to achieve high bitrate and low delay.

Numerical Simulation of Oxy-coal Combustion for a Swirl Burner with EDC Model

The characteristics of oxy-coal combustion for a swirl burner with a specially designed preheating chamber are studied numerically. In order to increase the accuracy in the prediction of flame temperature and ignition position, eddy dissipation concept (EDC) model with a skeletal chemical reaction mechanism was adopted to describe the combustion of volatile matter. Simulation was conducted under six oxidant stream conditions with different O2/N2/CO2 molar ratios: 21/79/0, 30/70/0, 50/50/0, 21/0/79, 30/0/70 and 50/0/50. Results showed that O2 enrichment in the primary oxidant stream is in favor of combustion stabilization, acceleration of ignition and increase of maximum flame temperature, while the full substitution of N2 by CO2 in the oxidant stream delays ignition and decreases the maximum flame temperature. However, the overall flow field and flame shapes in these cases are very similar at the same flow rate of the primary oxidant stream. Combustion characteristics of the air-coal is similar to that of the oxy-coal with 30% O2 and 70% CO2 in the oxidant stream, indicating that the rear condition is suitable for retrofitting an air-coal fired boiler to an oxy-coal one. The swirl burner with a specially designed preheating chamber can increase flame temperature, accelerate ignition and enhance burning intensity of pulverized coal under oxy-coal combustion. Also, qualitative experimental validation indicated the burner can reduce the overall NOx emission under certain O2 enrichment and oxy-coal combustion conditions against the air-coal combustion.

Cross-Layer-Based Adaptive Video Transport Over Low Bit-Rate Multihop WSNs

A low bit-rate ZigBee-based sensor network carrying scalable video streams will be considered. A cross-layer priority algorithm that combines the state of the channel and the quality of the video traffic will be suggested to optimize the required quality of service of the video streams. The proposed algorithm prioritizes the transport of the basic frames, and thus minimizes the carried enhancement orphan frames. It also avoids transmitting the enhancement frames related to a dropped basic frame. A noticeable improvement in the end-to-end delay of the video stream was achieved, an increase in the carried base frames, and a decrease in the orphan frames provided both an enhanced quality of the received video stream and an improvement in the channel utilization.

Estimation of major stream delays with a limited priority merge

The purpose of this paper is to develop an estimation model for major stream delays at an unsignalized intersection under limited priority conditions. The key idea of this paper is that, when the minor stream drivers accept smaller major stream gaps, some drivers on the major road have to slow down their speeds and adjust their relative positions to avoid traffic accidents, incurring some delays. Assuming that headways in the major stream follow the M3 distribution, the delay for the first major stream vehicle after a limited priority merge is presented using acceptance gap theory and probability theory, and then recursive models are developed for the following major stream vehicles before the next merge. Finally, the precision of the method proposed in this paper is calibrated using field survey data from Changchun city, China, and the results show that the maximum, minimum, and average relative error of the 12 samples are approximately 30.38%, 9.04%, and 18.97%, respectively.

A Real Time Video Transmission Routing Protocol in Multi-Interface Multi-Channel Ad Hoc Based on Queue Length and Delay Contraint

Real-time video transmission demands tremendous bandwidth, throughput and strict delay. For transmitting real-time video in the multi-interface multi-channel Ad hoc, firstly, we applied multi-interface multi-channel extension methods to the AOMDV (Ad-hoc On-demand Multipath Distance Vector) routing protocol, and improved extant channel switching algorithm, called MIMC-AOMDV (Multi-Interface Multi-Channel AOMDV) routing protocol. Secondly, we proposed video streaming delay QoS(Quality of Service) constraint and link-quality metrics, which used the multi interface queue’s total used length to get QMMIMC-AOMDV (Quality metric MIMC -AOMDV) routing protocol. The simulations show that the proposed QMMIMC-AOMDV can reduce the frame delay effectively and raise frame decodable rate and peak signal to noise ratio (PSNR), it is more suitable for real-time video streams.

연결 정보를 이용한 P2P 스트리밍 네트워크 구조의 개선

본 논문에서는 P2P스트리밍에서 peer간 연결 정보를 이용하는 방법을 기반으로 topology 최적화를 연구한다. 제안하는 방법은 mesh-network에서 사용된 link와 사용되지 않은 link의 수로 peer의 업로드 용량을 추정하는 방법을 기반으로 한다. 연결된 link의 정보를 사용하는 기존 방법은 peer의 자원 관리 측면에서 메시지 과부하를 줄이는데 효과적이다. 하지만 업로드 대역폭이 고려되지 않는 topology를 구성할 우려가 있다. 또한 서버에 가까운 peer에서 네트워크 오류 발생시 네트워크 전송성능이 저하될 수 있다. 본 논문에서는 기존 방법의 단점을 보완하는 방법을 제안한다. 기존 방법과 제안하는 방법을 시뮬레이션 하고 결과를 비교 분석한다. In P2P streaming management, peer's churning and finding efficient topology architecture optimization algorithm that reduces streaming delay is important. This paper studies a topology optimization algorithm based on the P2P streaming using peer's link information. The proposed algorithm is based on the estimation of peer's upload bandwidth using peer's link information on mesh-network. The existing algorithm that uses the information of connected links is efficient to reduce message overload in the point of resource management. But it has a risk of making unreliable topology not considering upload bandwidth. And when some network error occurs in a server-closer-peer, it may make the topology worse. In this paper we propose an algorithm that makes up for the weak point of the existing algorithm. We compare the existing algorithm with the proposed algorithm using test data and analyze each simulation result.

On Reducing Delay in Mesh-Based P2P Streaming: A Mesh-Push Approach

The peer-assisted streaming paradigm has been widely employed to distribute live video data on the internet recently. In general, the mesh-based pull approach is more robust and efficient than the tree-based push approach. However, pull protocol brings about longer streaming delay, which is caused by the handshaking process of advertising buffer map message, sending request message and scheduling of the data block. In this paper, we propose a new approach, mesh-push, to address this issue. Different from the traditional pull approach, mesh-push implements block scheduling algorithm at sender side, where the block transmission is initiated by the sender rather than by the receiver. We first formulate the optimal upload bandwidth utilization problem, then present the mesh-push approach, in which a token protocol is designed to avoid block redundancy; a min-cost flow model is employed to derive the optimal scheduling for the push peer; and a push peer selection algorithm is introduced to reduce control overhead. Finally, we evaluate mesh-push through simulation, the results of which show mesh-push outperforms the pull scheduling in streaming delay, and achieves comparable delivery ratio at the same time.

QoS Control for Guaranteed Service Bundles Over Fiber-Wireless (FiWi) Broadband Access Networks

The integration of ethernet passive optical networks (EPONs) and IEEE 802.16 has been lately presented as a promising candidate for next-generation broadband access. This paper investigates the performance of the integrated carrier when specific service bundles (SBs) with bandwidth guarantee and diverse quality of service requirements are provisioned. We first present a novel framework that fairly distributes the upstream network capacity among the different SBs. We then propose a novel delay-based admission control scheme to increase the network utilization and enable an inter-SB statistical multiplexing. To obtain the expected delay of each SB stream, a generic analytical model is developed. Based on this model, we further formulate some important performance measures such as queuing delay, end-to-end (from wireless user to the central office of PON) delay, average queue size and optimal orthogonal frequency division multiplexing frame length. Numerical results validate our analysis and highlight the advantages of the proposed solution.

Stochastic performance analysis of non-feedforward networks

Many Internet applications are both delay and loss sensitive, and need network performance guarantees that include bandwidth, delay/delay jitter, and packet loss rate. It is very important to quantify and exploit the capabilities of guaranteed service provisioning of communication networks. In this paper, we study the queueing behaviors of non-feedforward networks (a non-feedforward network is a network in which at least one set of acyclic traffic routes forms a cycle; a feedforward network is a network in which any set of acyclic traffic routes does not form a cycle) with FIFO scheduling discipline and Regulated, Markov On-Off, and Fractional Brownian traffic sources. We develop a new methodology to analyze the probabilistic bounds on the delays experienced by traffic. By leveraging the large deviations and fixed-point techniques, we turn probability problems into deterministic optimization problems and translate a probabilistic delay bound into a fixed point of a non-linear real function. Our contribution in this paper is the derivation of a probabilistic bound on the delays experienced by traffic in non-feedforward networks, based on an assumption, i.e., the tail probability of the difference between the beginning time of a busy interval of a server and the earliest arriving time at the corresponding network ingress of the traffic arrivals that arrive at this server during this busy interval can be bounded by the maximum of the violation probabilities of the accumulative upper stream delay bound suffered by this server`s traffic arrivals. Consequently, our new results not only consummate the theory of stochastic analysis of network performance, but also facilitate the design of protocols and algorithms for non-feedforward networks to provide performance guarantees to various applications with diverse performance requirements.