Streaming Protocol Research Articles

All-in-one computation vs computational-offloading approaches: a performance evaluation of object detection strategies on android mobile devices

Object detection gives a computer ability to classify objects in an image or video. However, high specified devices are needed to get a good performance. To enable devices with low specifications performs better, one way is offloading the computation process from a device with a low specification to another device with better specifications. This paper investigates the performance of object detection strategies on all-in-one Android mobile phone computation versus Android mobile phone computation with computational offloading on Nvidia Jetson Nano. The experiment carries out the video surveillance from the Android mobile phone with two scenarios, all-in-one object detection computation in a single Android device and decoupled object detection computation between an Android device and an Nvidia Jetson Nano. Android applications send video input for object detection using RTSP/RTMP streaming protocol and received by Nvidia Jetson Nano which acts as an RTSP/RTMP server. Then, the output of object detection is sent back to the Android device for being displayed to the user. The results show that the android device Huawei Y7 Pro with an average FPS performance of 1.82 and an average computing speed of 552 ms significantly improves when working with the Nvidia Jetson Nano, the average FPS becomes ten and the average computing speed becomes 95 ms. It means decoupling object detection computation between an Android device and an Nvidia Jetson Nano using the system provided in this paper successfully improves the detection speed performance.

Verifiable data streaming with efficient update for intelligent automation systems

The wide deployment of Internet of Things (IoT) devices enables the controller to continuously collect massive volume data in automation systems, and makes it possible to make intelligent decisions based on machine learning techniques. In fact, data-driven intelligent automation systems have been common in the industrial community. Nevertheless, how to effectively store the collected stream data and ensure their integrity is still challenging. To this end, the notion of verifiable data streaming (VDS) protocol, which enables a client to outsource the stream data to an untrusted server in a verifiable manner, was introduced. However, we argue that existing VDS protocols based on the chameleon authentication tree (CAT) are inefficient in the data update, since the whole CAT must be updated accordingly to avoid acute exposure of chameleon hashing. Thus, they are infeasible for intelligent automation systems that need to frequently update data. In this article, we first introduce a new primitive called double-trapdoor chameleon hash tree (DCHT) based on the double-trapdoor chameleon hash families, where each leaf of DCHT is calculated and fixed by using a double-trapdoor chameleon hash family, making the entire DCHT always unchanged. Furthermore, we propose a novel VDS protocol based on the DCHT. Due to the distinctive properties of the underlying DCHT, the proposed VDS protocol has a constant update cost and more efficient than previous VDS protocols based on CAT. Besides, we prove that the proposed VDS protocol is secure in the standard model.

Design and analysis of a fair, efficient and stable multi-server adaptive streaming protocol

Due to the surging bandwidth demand for applications like 360° video streaming, hypertext transfer protocol (HTTP) adaptive video streaming has been extended to fetch contents from a single server to multiple servers. However, when players share a bottleneck link, it fails to hold three critical properties: fairness, efficiency, and stability. Effective algorithms have been proposed to tackle the problem in the single-server case. We revisit the three properties and show that they degenerate in multi-server streaming. In this paper, we propose MultiFast, a protocol that offers the best user experience by achieving fair, efficient, and stable multi-server video streaming. MultiFast consists of (1) a bitrate selection algorithm that ensures fairness with concurrent connections; (2) a connection management algorithm that ensures efficiency and stability by managing connections based on network conditions. Our real-world experiments show that MultiFast achieves the best performance in fairness, efficiency, and stability metrics, outperforming its closest alternative.

Prediction of Crime Scene Objects Using Deep Learning

The crime rates today can inevitably put a civilian's life in danger. While consistent efforts are being made to alleviate crime, there is also a dire need to create a smart and proactive surveillance system. Our project implements a smart surveillance system that would alert the authorities in real-time when a crime is being committed. During armed robberies and hostage situations, most often, the police cannot reach the place on time to prevent it from happening, owing to the lag in communication between the informants of the crime scene and the police. We propose an object detection model that implements deep learning algorithms to detect objects of violence such as pistols, knives, rifles from video surveillance footage, and in turn send real-time alerts to the authorities. There are a number of object detection algorithms being developed, each being evaluated under the performance metric mAP. On implementing Faster R-CNN with ResNet 101 architecture we found the mAP score to be about 91%. However, the downside to this is the excessive training and inferencing time it incurs. On the other hand, YOLOv5 architecture resulted in a model that performed very well in terms of speed. Its training speed was found to be 0.012 s / image during training but naturally, the accuracy was not as high as Faster R-CNN. With good computer architecture, it can run at about 40 fps. Thus, there is a tradeoff between speed and accuracy and it's important to strike a balance. We use transfer learning to improve accuracy by training the model on our custom dataset. This project can be deployed on any generic CCTV camera by setting up a live RTSP (real-time streaming protocol) and streaming the footage on a laptop or desktop where the deep learning model is being run.

Crowdsourcing Framework for QoE-Aware SD-WAN

Quality of experience (QoE) is an important measure of users’ satisfaction regarding their network-based services, and it is widely employed today to provide a real assessment of the service quality as perceived by the end users. QoE measures can be used to improve application performance, as well as to optimize network resources and reallocate them as needed when the service quality degrades. While quantitative QoE assessments based on network parameters may provide insights into users’ experience, subjective assessments through direct feedback from the users have also gathered interest recently due to their accuracy and interactive nature. In this paper, we propose a framework that can be used to collect real-time QoE feedback through crowdsourcing and forward it to network controllers to enhance streaming routes. We analyze how QoE can be affected by different network conditions, and how different streaming protocols compare against each other when the network parameters change dynamically. We also compare the real-time user feedback to predefined network changes to measure if participants will be able to identify all degradation events, as well as to examine which combination of degradation events are noticeable to the participants. Our aim is to demonstrate that real-time QoE feedback can enhance cloud-based services and can adjust service quality on the basis of real-time, active participants’ interactions.

Real-time multiple damage mapping using autonomous UAV and deep faster region-based neural networks for GPS-denied structures

An autonomous unmanned aerial vehicle (UAV) system integrated with a modified faster region-based convolutional neural network (Faster R-CNN) is proposed to identify various types of structural damage and to map the detected damage a GPS-denied environment. The proposed method reduces the number of false positives significantly using a real-time streaming protocol and multi-processing, particularly in the case of very small cracks in blurry videos due to the UAV vibrations. In comparative studies, the modified Faster R-CNN using ResNet-101 as the base network showed superior performance in detecting small and blurry defects with a mean average precision of 93.31% and mean intersection-over-union of 92.16% in video frames captured by the low-cost autonomous UAV. The autonomous flights of the UAV were tested in a real large-scale parking structure to account for the high wind effects during flight. The UAV successfully followed the desired trajectories, and the Faster R-CNN detected defects accurately.

The Art of the Meta Stream Protocol: Torrents of Streams

The rise of streaming libraries such as Akka Stream, Reactive Extensions, and LINQ popularized the declarative functional style of data processing. The stream paradigm offers concise syntax to write down processing pipelines to consume the vast amounts of real-time data available today. These libraries offer the programmer a domain specific language (DSL) embedded in the host language to describe data streams. These libraries however, all suffer from extensibility issues. The semantics of a stream is hard-coded into the DSL language and cannot be changed by the user of the library. We introduce an approach to modify the semantics of a streaming library by means of meta-programming at both run-time and compile-time, and showcase its generality. We show that the expressiveness of the meta-facilities is strong enough to enable push and pull semantics, error handling, parallelism, and operator fusion. We evaluate our work by implementing the identified shortcomings in terms of a novel stream meta-architecture and show that its design and architecture adhere to the design principles of a meta-level architecture. The state of the art offers plenty of choice to programmers regarding reactive stream processing libraries. Expressing reactive systems is otherwise difficult to do in general purpose languages. Extensibility and fine-tuning should be possible in these libraries to ensure a broad variety of applications can be expressed within this single DSL.

Factors influencing streaming to General Practitioners in emergency departments: A qualitative study

BackgroundEmergency Department attendance is increasing internationally, of which a significant proportion could be managed in general practice. In England, policies backed by substantial capital funding require such patients attending Emergency Departments be directed or ‘streamed’ to General Practitioners working in or parallel to Emergency Departments. However, evidence for streaming is limited and the processes of streaming patients attending Emergency Departments to General Practitioners lacks exploration. ObjectivesThis paper explores streaming to General Practitioners in and alongside Emergency Departments at ten sites across England. It highlights positive streaming practice, as well as issues that may contribute to poor streaming practice, in order to inform future service improvement. MethodsA longitudinal qualitative study was conducted with data collected between October 2017 and December 2019 across 10 case study sites throughout England as part of a broader mixed methods study. 186 non-participant observations and 226 semi-structured interviews with 191 health professionals working in Emergency Departments or related General Practitioner Services were thematically analysed in relation to streaming processes and experiences. ResultsSix interconnected themes influencing streaming were identified: implementing and maintaining structural support; developing and supporting streaming personnel; implementing workable and responsive streaming protocols; negotiating primary/secondary care boundaries; developing and maintaining interprofessional relationships and concerns for patient safety. Streaming was considered central to the success of General Practitioners in/parallel to Emergency Departments. The importance of the skills of streaming nurses in delivering an optimal and safety critical service was highlighted, as was the skillset of General Practitioners and interprofessional relationships between streamers and General Practitioners. There was no distinct streaming model or method associated with good streaming practice to General Practitioners in/alongside Emergency Departments, instead factors for success were identified and key recommendations suggested. ‘Inappropriate’ streaming was identified as a problem, where patients streamed to General Practitioners in or parallel to Emergency Departments required Emergency Department management, or patients suitable for General Practitioner care were kept in the Emergency Department. ConclusionDespite adopting differing methods, commonalities across case sites in the delivery of good streaming practice were identified, leading to identification of key recommendations which may inform development of streaming services.Study Registration: ISRCTN51780222.Tweetable abstract: Workplace culture and the skillset of streamers and General Practitioners is crucial to streaming of patients to General Practitioners in Emergency Departments

Simplified Stream Reservation Protocol Over Software-Defined Networks for In-Vehicle Time-Sensitive Networking

As the automotive industry adopts Ethernet for in-vehicle network technology, automotive Ethernet has to handle bounded delay traffic. IEEE std 802.1Q introduces the stream reservation protocol (SRP) to reserve resources for time-sensitive traffics. SRP uses bridge-by-bridge propagation for its messages and registers VLAN ID and group address before making Audio-Video (AV) stream reservations. In this study, we propose a simplified SRP scheme over the Software-defined Network (SDN) for in-vehicle bridged networks. The proposed scheme exploits the topology and bridge information of bridged networks. A centralized stream reservation procedure in the scheme replaces the bridge-by-bridge propagation for SRP reservations. The reservation procedure does not require VLAN ID and group address registration. We implemented a prototype of the proposed scheme and showed that reserved flows are well protected when links of bridges are overloaded.

Adaptive live video streaming on low-cost wireless multihop networks for road traffic surveillance in smart cities

Traffic surveillance is an important issue for Intelligent Transportation Systems (ITS) that helps detect incidents automatically, such as wrong-way drivers, still-standing vehicles and jams. Sometimes these systems require a fast and short-term deployment of video cameras. In these cases, ad-hoc networks could be a low-cost and feasible option, but they have poor performance for video delivery. Thus, we propose a smart live video adaptive streaming technique in order to transport video streams from the cameras to the external road traffic monitoring servers. To achieve this goal, these networks need a thorough study in order to analyze video quality under their inherent constraints, given by their specific topologies (once deployed), video streaming protocol, nodes’ performance and routing protocols. From our results both in simulated and real scenarios, we demonstrate the feasibility of our proposal in these specific scenarios, validated with several performance metrics in terms of Quality of Service and video quality.

A reliable peer-to-peer streaming protocol in low-capacity networks

The recent global demand for video streaming applications has paved the way for peer-to-peer streaming system (P2PSS). Strategic scheduling scheme and dynamic overlay topology are essential to maintain quality of service (QoS) and quality of experience (QoE) in P2PSS. The concept of P2PSS was tailored towards relying on active peers’ bandwidth to achieve cheap and scalable means of distribution over the Internet, such that peers with highest bandwidth serve as backbones for others. However, selecting backbone peers in low-capacity network environment is challenging due to insufficient bandwidth and poor infrastructure, thereby resulting in poor QoS and unpleasant user’s QoE. In this paper, we conducted a survey on users’ experiences with live video in selected locations in Nigeria. We designed an adaptive P2P streaming protocol and performed a packet-level simulation in Network Simulator 3(NS-3). Diverse simulation scenarios were set up to evaluate the proposed streaming protocol. Trace files data were analysed to measure end-to-end delay, start-up delay, and throughput. Furthermore, the proposed streaming protocol was benchmarked against selected existing schemes. The evaluation results revealed a 7.4% and 28% reduction in start-up and in end-to-end delays and 9% increase in throughput.

Pengiriman Video secara Live Streaming Menggunakan Dynamic Adaptive Streaming over HTTP (DASH)

Keterbatasan bandwidth dan fluktuasi trafik pada jaringan menjadi kendala dalam proses pengiriman video. Pada saat ketersediaan bandwidth kecil dan traffic jaringan yang tinggi, proses pengiriman video akan memakan waktu yang lama dikarenakan proses transmisi video yang lambat atau lebih umum disebut dengan buffering. Penelitian ini melakukan evaluasi pengiriman video secara live streaming menggunakan metode Dynamic Adaptive Streaming over HTTP (DASH) pada jaringan LAN. DASH merupakan sebuah protokol streaming adaptive yang mampu menyajikan video dengan kualitas video sesuai kemampuan jaringan pada klien. Implementasi ini dilakukan menggunakan video kompresi H.265 dan proses streaming dilakukan pada jaringan LAN secara live streaming sebanyak 20 klien. Hasil penelitian ini diperoleh nilai throughput pengiriman video mencapai 1,4 Mbps dan nilai delay terbesar 12,5 ms. Persentase permintaan klien terhadap kualitas video tertinggi untuk 1 klien sebanyak 98%. Sedangkan penggunaan 20 klien, permintaan untuk video kualitas tinggi berkurang menjadi 46% yang membuktikan bahwa DASH berhasil diimplementasikan secara live streaming.
 Kata kunci: Video, Live, Streaming, H.265, DASH

QoS Enhancement In VoD Systems: Load Management And Replication Policy Optimization Perspectives

AbstractThe amount of online video content is exponentially increasing, which spurs its access demands. Providing optimal quality of service (QoS) for this ever-increasing video data is a challenging task due to the number of QoS constraints. The system resources, the distributed system platform and the transport protocol thus all need to collaborate to guarantee an acceptable level of QoS for the optimal video streaming process. In this paper, we present a comprehensive survey on QoS management for the video-on-demand systems. First, we focus on load management and replication algorithms in content delivery networks and peer-to-peer (P2P) networks for their shortcomings. We also address the problem of admission control and resource allocation with the objectives of congestion avoidance and frame-loss reduction. Besides, we introduce and discuss various replication schemes. For both the client–server architecture and P2P networks, we highlight the need for a specific storage management policy to preserve system reliability and content availability. We also focus on content distribution and streaming protocols scaling. We deduce that content availability is linked to the characteristics and the performance of the streaming protocols. Finally, we create a comparison table that presents the different contributions of the discussed approaches as well as their limitations. We believe that such a comprehensive survey provides useful insights and contributes to the related domains.

GeoQoE-Vanet: QoE-Aware Geographic Routing Protocol for Video Streaming over Vehicular Ad-hoc Networks

Video streaming is one of the challenging issues in vehicular ad-hoc networks (VANETs) due to their highly dynamic topology and frequent connectivity disruptions. Recent developments in the routing protocol methods used in VANETs have contributed to improvements in the quality of experience (QoE) of the received video. One of these methods is the selection of the next-hop relay vehicle. In this paper, a QoE-aware geographic protocol for video streaming over VANETs is proposed. The selection process of the next relay vehicle is based on a correlated formula of QoE and quality of service (QoS) factors to enhance the users’ QoE. The simulation results show that the proposed GeoQoE-Vanet outperforms both GPSR and GPSR-2P protocols in providing the best end-user QoE of video streaming service.

Ethernet TCP/IP-Based Audio Interface for DSP System Verification

With advances in digital technologies, development and verification of DSP audio systems are becoming more and more challenging. Due to high communication requirements introduced with surround audio channel- and latest object-based technologies such as DTS:X and Dolby Atmos, computer's sound card capabilites are expanded with an external audio interface for exchanging high-resolution content between computer and DSP device under test. In this article, the Ethernet TCP/IP-based audio interface solution is proposed. The main idea was to come with the cost-effective communication solution using standard Ethernet IEEE 802.3 hardware, free open-source network stacks, and finally implementing TCP application protocol in point-to-point topology instead of porting custom UDP streaming protocol. Experiments conducted in order to observe overall performance of the system indicate that expected throughput is achieved, but at the cost of long delays and high memory consumption.

Indoor propagation models for emulated LTE networks

The growth in mobile network data traffic is even more noticeable in networks compatible with the Long-Term Evolution (LTE) standard. As such, both academia and industry seek mechanisms aimed at improving new communications systems, therefore, new test benches are required. The emulators are candidates to validate the proposals without affecting the real central systems. However, many papers that seek to address LTE emulation ignore the not-so-insignificant effect of wireless media propagation features. In this work, a characterization of the main propagation models used by typical emulation systems has been carried out. The network simulator three tool using indoor propagation models, has compared the outcomes achieved compared with the results obtained in a real test bench. Since most mobile data in the coming years is forecast to be mobile video traffic, this study focuses on a live video streaming service, considering both real-time message protocol and real-time streaming protocol transmission protocols. From the results, we demonstrate that only one of the propagation models under study is able to equate with a real LTE system performance. Thereby, the emulation system under this specific setup is well suited to conducting performance tests of LTE networks supporting video streaming services in indoor scenarios.

Region of interest–based segmented tiled adaptive streaming using head-mounted display tracking sensing data

To support 360 virtual reality video streaming services, high resolutions of over 8K and network streaming technology that guarantees consistent quality of service are required. To this end, we propose 360 virtual reality video player technology and a streaming protocol based on MPEG Dynamic Adaptive Streaming over HTTP Spatial Representation Description to support the player. The player renders the downsized video as the base layer, which has a quarter of the resolution of the original video, and high-quality video tiles consisting of tiles obtained from the tiled-encoded high-quality video (over 16K resolution) as the enhanced layer. Furthermore, we implemented the system and conducted experiments to measure the network bandwidth for 16K video streaming and switching latency arising from changes in the viewport. From the results, we confirmed that the player has a switching latency of less than 1000 ms and a maximum network download bandwidth requirement of 100 Mbps.

Improving Adaptive Video Streaming through Session Classification

With internet video gaining increasing popularity and soaring to dominate network traffic, extensive studies are being carried out on how to achieve higher Quality of Experience (QoE) with the delivery of video content. Associated with the chunk-based streaming protocol, Adaptive Bitrate (ABR) algorithms have recently emerged to cope with the diverse and fluctuating network conditions by dynamically adjusting bitrates for future chunks. This inevitably involves predicting the future throughput of a video session. Some of the session features like Internet Service Provider (ISP), geographical location, and so on, could affect network conditions and contain helpful information for this throughput prediction. In this article, we consider how our knowledge about the session features can be utilized to improve ABR quality via customized parameter settings. We present our ABR-independent, QoE-driven, feature-based partition method to classify the logged video sessions so that different parameter settings could be adopted in different situations to reach better quality. A variation of Decision Tree is developed for the classification and has been applied to a sample ABR for evaluation. The experiment shows that our approach can improve the average bitrate of the sample ABR by 36.1% without causing the increase of the rebuffering ratio where 99% of the sessions can get improvement. It can also improve the rebuffering ratio by 87.7% without causing the decrease of the average bitrate, where, among those sessions involved in rebuffering, 82% receives improvement and 18% remains the same.

Light-weight streaming protocol for the Internet of Multimedia Things: Voice streaming over NB-IoT

Despite the significant potential of the Internet of Multimedia Things (IoMT), the integration of multimedia services and devices into the Internet of Things (IoT) still stays a challenge due to the limited capabilities of IoT devices and distinct characteristics of multimedia applications. In this paper, we investigate a novel, efficient and light-weight streaming protocol for IoMT to enable numerous multimedia-based IoT applications and services. Further, the use of standardized IoT architectures, interaction and information models will provide syntactic, structural and semantic interoperability for IoMT which will permit heterogeneous multimedia things to communicate and interact with each other and with other connected things seamlessly. Additionally, we also validate the proposed design by means of a real-life use case and implementation, which targets the ability of end-to-end bidirectional voice communication over NB-IoT technology, a long-range limited bit rate communication technology. Also, we assess the suitability of NB-IoT for handling less-demanding IoT multimedia applications by means of theoretical calculations and practical experiments.

TimeStream: Exploiting video streams for clock synchronization

Network Time Protocol (NTP) is the most commonly used protocol for clock synchronization on the Internet.In the wake of frequent news about attacks that use Network Time Protocol (NTP), we propose a TimeStream algorithm, which – in its plain version (without any extra settings) – successfully synchronizes computer clocks up to seconds, by processing data found in video streams that use HTTP Live Streaming (also known as HLS) protocol. Further increase of precision is possible with a proper setup of the origin video servers.TimeStream provides secure time synchronization without requiring new ports, new hardware, or expensive cryptography, and uses information which – currently – is already making at least 70% of all Internet traffic.We observe that some intrinsic properties of video streaming over HTTP and congestion control of TCP yield some useful properties that facilitate good performance of the proposed method even in adverse network conditions. These make TimeStream to be robust against packet delay manipulation attack in which, an attacker adds specifically crafted delays to packets carrying time information.To the best of our knowledge, TimeStream may be one of the first methods of its kind that uses structure and core features of video streams and underlying network transportation protocol, to extract information about time and opens a new line of research on almost real-time extraction of useful information from video content.