Search

In this paper, we proposed a method for accelerating the transmission of a series of streaming files for multi-session file delivery scenarios. When streaming files are transmitted, the throughput of data transmission can be increased and the out of order problem of file reception (i.e., file reception delay) can also be reduced at the same time. This is achieved by using a repair file generation mechanism based on AL-FEC, which is triggered while an empty file transmission session appears but there are still uncompleted files waiting to be delivered. Performance evaluation results shows that the file transmission throughput of the proposed method are 28% to 44% higher than the throughput of a traditional file transmission method that is designed for reducing file reception delay. However, the file reception delay is the same as the above-mentioned traditional method.

Air pollution has witnessed that is an important issue regarding with people health. An excellent air pollution forecasting approach can provide warning to user for avoiding too much polluted air inhalation. Many studies have proposed the model of predicting PM2.5 concentration. How to improve the performance of the forecasting models is an important issue. As well known, ensemble learning is proved that it can improve the forecasting performance. In this paper, in order to improve the performance of air pollution forecasting, we explore an approach named hybrid model framework. which exploit the stacking scheme of ensemble learning. In this work, first we proposed the framework used to forecast the air pollution based on traditional machine approach and deep learning; and exploited Pearson correlation coefficient to calculate the correlation of various models, and finally find out the highest correlation among these models. We conducted experiments and made a comparison with four kinds of models to demonstrate the forecasting performance of hybrid model. The experimental results reveal that the hybrid model is superior than single air pollution forecasting model.

Although cloud storage integration services have been developed in recent years that providing a single interface to manage multiple cloud storage accounts across different providers, these cloud storage integration services still remain several issues that do not address. First, because these services do not support arrangement mechanisms for users, they need to keep track of which account they want to upload files to and arrange their storage spaces manually. Second, since each account is divided into different sections, it is impossible to provide a virtual folder that contains files from different cloud storage providers. In addition, when all accounts are nearly full, integration services could not chunk a file to store it across multiple accounts. Third, uploading and moving files across accounts are always done through the integration service's server, which might become a bottleneck when there are a lot of users uploading files at the same time. In this paper, we propose a new cloud storage integration service - Unified Cloud Storage (UCS) for solving the above-mentioned issues, and evaluate its performance by experiments.

ONVIF (Open Network Video Interface Forum) is a popular standard for the video surveillance industries. Since ONVIF specifications do not provide fault-tolerant functions to deal with NVT (Network Video Transmitter) or NVS (Network Video Storage) failure, we propose a fault-tolerant solution for both ONVIF NVT and NVS in this paper. The proposed fault-tolerant extension is also possible to be deployed over cloud-based environments, such as IaaS (Infrastructure as a Service) platforms. The proposed solution is evaluated by the following three performance metrics: the access response time for ONVIF web services, the video recording interrupt duration due to NVS failure, and the video black out time of the surveillance video monitoring system. The experimental results show that all the metrics are within the acceptable range for a real world video surveillance system.

Cloud computing has been recognized as an important advance for virtualizing computing hardware, distributed computing middleware, and communication networks. For simplifying the network OA/&M (Operation, Administration, and Management) tasks of customers, a CaaS (Communications-as-a-Service) operator can provide virtual communication channels tailored for delivering various services inside a communication cloud, such as VoIP, video conferencing and IPTV. For improving the video quality of broadband IPTV services, Raptor code has been adopted in various broadband IPTV systems. In this paper, we propose a systematic Raptor code decoder based on the group parameters for a group of encoding symbols. For a Raptor code application that can frequently use one or several fixed block lengths (i.e., the number of source symbols in a source block) such as IPTV, we could produce the corresponding group parameters in advance and then the proposed decoder can use them to decode the received encoding symbols more efficiently. For the IPTV delivery scenario over a communication cloud based on DSL (Digital Subscriber Line) access networks, the simulation results show that our decoder is faster than the conventional Raptor code decoder which is adopted by DVB and 3GPP specifications.

To facilitate the distribution of video contents, MPEG-DASH (Dynamic Streaming over HTTP) standard is developed to provide standardized manifest format, content format and streaming protocol for adaptive streaming, thereby increasing interoperability of Internet TV services. Although DASH contents are delivered by HTTP without supporting data multicast, some studies shows the possibility of delivering DASH contents over multicast channels for better service scalability or service quality. These studies usually employ scalable videos (e.g., SVC videos) to support efficient adaptive streaming. In this paper, we try to improve the error protection capability for delivering SVC videos by using layer-aware FEC (L-FEC). We apply two conventional L-FEC schemes and found that the improvement in the construction using multiple multicast groups is quite limited. Therefore, we propose a new inter-layer protection scheme with block-based interleaving to improve the performance. We also evaluate the performance of the proposed scheme and compare it with other schemes over real WiFi networks. The result shows that the proposed scheme can improve the source block recovery rate by 0.47%~2.42% when employed in multiple multicast group construction.

Several studies indicate that video contents will be the vast majority of Internet traffic in the near future. To facilitate the distribution of video contents, we have to overcome the issue of interoperability between various servers and devices. MPEG-DASH standard is developed to provide standardized manifest format, content format and streaming protocol for adaptive streaming, thereby increasing interoperability. However, MPEG-DASH contents are delivered by HTTP without supporting data multicast. Since delivering video contents over multicast channels is necessary or useful for some applications, the authors survey the studies of delivering MPEG-DASH contents over multicast channels. It is found that few studies discuss the detailed construction of such a mechanism. This paper proposes three methods enabling the delivery of MPEG-DASH contents over multicast are proposed. And also evaluates the performance of the proposed methods over real WiFi networks. The result shows that each method has its own advantages and therefore suitable for different applications.

Through the wide deployment of high-speed rails around the world, multimedia services on high-speed trains will be highly demanded in the near future. However, network round trip time and packet loss rate for high-speed rail reception are much higher and also of high variance. To facilitate the delivery of a live video streaming service, our previous research suggests that the video streaming service is transmitted as a file delivery service and the generated streaming files are delivered via a UDP-based file delivery protocol. In this paper, we investigate the potentials of UDP-based file delivery protocols for high-speed rail reception. We propose Reliable FLUTE (R-FLUTE) for reliable UDP-based unicast file delivery by introducing feedback mechanisms into the IETF FLUTE protocol. To get higher throughput and better network utilization, we extend R-FLUTE to support multisession delivery and propose an adaptive session-based rate control mechanism. We further conduct our experiments on a real 802.16e mobile WiMAX network for Taiwan High Speed Rail (THSR). The results show that the proposed scheme in comparison with FTP can provide more stable throughput, thereby requiring smaller buffer level in receivers (i.e., can deliver more real-time live video to receivers).

This paper presents an HD (High Definition) streaming system that employs an AL-FEC (Application-Layer FEC) scheme (i.e., DVB AL-FEC) based on Raptor codes to protect the transmission of MPEG-2 TS (Transport Streaming) over RTP streaming packets over error-prone Wi-Fi multicast channels. The employed AL-FEC scheme can effectively recover the missing Wi-Fi multicast packets in a scalable way. In this paper, we describe the prototype of the proposed HD streaming system, which is implemented based on VLC server and VLC client over Linux operating systems. Finally, we show that the proposed system can significantly improve the HD video quality of a Wi-Fi multicast streaming service.

In this paper, we implement our previously proposed inter-layer protection scheme to protect the RTP-based SVC video streaming service based on the DVB AL-FEC. The proposed scheme enables the video data in a more important layer to be recovered with a higher probability. To verify the feasibility of the proposed scheme, we fully design and implement the corresponding interfaces and components. The experimental result shows that the proposed scheme can enhance the lower layer reconstruction for RTP-based SVC video data.

Raptor codes can provide good error correction capability and efficient encoding and decoding rates. Its fountain code property is effective in avoiding packet retransmission for both unicast and multicast service delivery. Hence, there are many wireless and broadband applications adopting it such as mobile multimedia broadcasting and broadband IPTV systems. In this paper, we present an efficient systematic Raptor codes encoder based on operation lists for known block lengths. For a Raptor codes application that can frequently use one or several fixed block lengths (i.e., the number of source symbols in a source block), we could produce the corresponding operation lists in advance and use it to generate the encoding symbols more efficiently. We first introduce the basic architecture of the proposed Raptor encoder, and then describes the details about how to generate operation lists and Raptor intermediate symbols. The simulation results show that our encoder is at least two times faster than the conventional Raptor codes encoder which is adopted by 3GPP and DVB-H.

Scalable Video Coding (SVC) enables the partition of video data into layers of different priorities for adapting to both the diversities of terminal capabilities and the variation of network transmission mediums. It is more desirable that more important video data can be recovered with higher probabilities. In this paper, we propose a lightweight inter-layer protection scheme for SVC transmission, which enables the data in a higher layer (e.g., an enhancement layer) to enhance the data reconstruction in a lower layer (e.g., a base layer) in the receiver side. In contrast to non-inter-layer protection schemes, the simulation result in a DVB-H broadcast channel shows that the scheme exhibits 1% to 45% enhancement on the reconstruction of the SVC video data in the base layer.

AL-FEC (Application-Layer FEC) is already in used for protecting streaming and file delivery services against packet losses in the field of wide-area multimedia broadcasting, such as in DVB-H and 3GPP MBMS (Multimedia Broadcast Multicast Service) mobile TV systems. However, it is still not a well-studied approach for applying AL-FEC to WiFi multicast since guaranteed bandwidth for a multicast service is more difficult to be allocated in a WiFi network in comparison to traditional mobile TV systems. In this paper, we show our experiences of using an AL-FEC code; i.e. the systematic Raptor code as well the IETF FEC Framework to further protect the RTP packets delivered over WiFi multicast carriers inside the cars of a train for high-speed rails, instead of using any feedback and retransmission mechanisms.

In recent years, Application-Layer Forward Error Correction (AL-FEC), especially rateless AL-FEC, receives a lot of attention due to its superior performance in both transmission and computation efficiency. Rateless AL-FEC (e.g., Raptor code or LT code) can protect large data block with the overhead somewhat close to ideal codes. In the meantime, its data processing rates of both encoding and decoding are quite efficient even in software implementations (probably reaching Gbps throughput on a modern PC). Due to its properties, rateless AL-FEC is suggested to be used to protect packet erasure for wireless communications or the environments where retransmissions are infeasible. However, we found that conventional rateless AL-FEC may not be the best candidate when considering streaming over WiMAX networks for high speed rail reception in Taiwan. To facilitate streaming service delivery for high-speed rail reception, we proposed a new AL-FEC scheme based on Chinese Remainder Theorem (CRT). The scheme supports greater symbol size than conventional ideal codes and requires less overhead than conventional rateless codes, making the proposed AL-FEC scheme more feasible to harsh bandwidth-limited networks.

For improving the service quality of the Taiwan High-Speed Rail (THSR), a dedicated mobile WiMAX network (based on IEEE 802.16e in the current stage) is planned to be deployed along the railway of THSR. The mobile WiMAX network is not only for providing Internet access over the THSR trains, but also for delivering live entertainment contents to the passengers, such as mobile TV streaming of live programs. In this paper, we show the experiences about the design, implementation and evaluation of the mobile TV streaming service for high-speed rail reception based on AL-FEC (Application-Layer FEC). The experiment results show that AL-FEC can effectively compensate the packet losses in the high-speed reception scenario if the underlying reliability control schemes in IEEE 802.16e cannot provide error free mobile TV streaming delivery.

For the sake of satisfying the requirement of different receivers and adapting the network condition, a new video coding scheme appears and is called Scalable Video Coding (SVC). Different from traditional coding scheme i.e. H.264, encoding once by using SVC encoder can generate one bitstream with different resolutions as well as frame rate, and the bitstream can be corresponded with the requirement of different receivers. In this paper, we propose two adaptive SVC layer switching algorithms for dedicated Worldwide Interoperability for Microwave Access (WiMAX) network of Taiwan High Speed Rail (THSR). The adaptive layer switching algorithms are based on network condition to adjust to delivering video data. For the first algorithm, we record the Received Signal Strength Indicator (RSSI) trace at the high speed moving environment of THSR in advance, and measure the available network bandwidth with RSSI trace. Then, Streaming Server can deliver the corresponding SVC layer data by measured network bandwidth. As to the second algorithm, Streaming Server estimates network available bandwidth from static information such as packet loss rate sent by Streaming Client. Streaming Server can select and deliver SVC layer data which are adapted to estimated network bandwidth. Our streaming system is an adaptive streaming system because video layer switching adapts to different network conditions.

The recent IP-based mobile TV standards, such as DVB-IPDC and OMA BCAST, employ both unidirectional broadcast networks and mobile interactive networks for hybrid service delivery. For these mobile TV standards, the performance of the file delivery protocol over broadcast channels (i.e., the IETF FLUTE protocol) is well studied in the literature. However, the performance characteristics of the file delivery protocols available in the mobile TV interactive channels (i.e., the HTTP and FLUTE protocols) are still unclear in the current stage. Hence, in this paper, we provide a performance study of the HTTP and FLUTE protocols in the context of mobile TV file delivery over a WiMAX-based interactive channel, for both stationary and vehicular users. The experiment results show that FLUTE with Raptor FEC can exhibit obvious performance gain than HTTP for delivering large files. In addition, FLUTE with Raptor FEC can coping long shadowing when moving through the outage areas of WiMAX signal, but in contrast HTTP connections are possible to be broken under the same signal condition.

Recently, the Pay-TV broadcasting system becomes one of the popular applications because of the advance of modern network technique. The conditional access system is the essential function to provide the channel protection in Pay-TV broadcasting system. It ensures that only the authorized subscribers can precisely receive the TV program and the service provider can charge the subscribers according to their subscription. Several conditional access systems have been proposed for different business models such as pay-per-channel, pay-per-view, and flexible pay-per-channel. In this paper, we define a new business model named Flexible Pay-Per-Group and design a conditional access system for it. Compared with the previous works, subscribers in Flexible Pay-Per-Group still have sufficient flexibility. Besides, through our analysis and evaluation, the proposed conditional access system is efficient. Consequently, it can be applied on a large scale Pay-TV system with lots of subscribers and channels.

FLUTE is a file delivery protocol whose operation can only require a unidirectional communication channel from a sender to a receiver. Due to the nature of unidirectional delivery, FLUTE is designed in a flexible way so that it can cooperate with different kinds of forward error correction codes in application layer, such as systematic Raptor codes or other FEC codes defined in IETF. Besides, since FLUTE is a generic IP-based file delivery protocol, it is already applied for multicast-based file distribution in IP-based mobile TV standards such as DVB-IPDC and OMA BCAST. In addition, the FLUTE protocol can also be applied for long haul, multi-hop file delivery in an IP-based intranet. In this paper, we describe the experience of using the FLUTE protocol for file delivery over a WiMAX unicast network and show the performance gain of employing a systematic Raptor code as the application layer FEC for FLUTE.

Hybrid streaming delivery over converged mobile and broadcast networks can enable operators to transmit multimedia contents in a more efficient and flexible way. Consequently, the functions related to hybrid streaming delivery are recently standardized in the IP-based service layer standards for DVB-H, i.e. DVB-IPDC and OMA BCAST standards. In both standards, seamless service switching is one of the key building blocks for achieving hybrid streaming delivery. In this paper, we describe the experiences for implementing the function of streaming service switching between DVB-H broadcast and WiMAX mobile networks according to the abovementioned standards. In our implementation, instantiating two sets of audio and video codecs in parallel during service switching is avoid for reducing the required hardware resources in a terminal. In addition, the experiment results of streaming service switching are provided to show the performance of our implementation in an extreme test environment where the level of a DVB-H signal could be varied rapidly for simulating the observed room or building penetration loss for a moving terminal.

Forward Error Correction (FEC) allows the receiver to correct erasure without asking retransmission. This advantage is significant for several applications where retransmission is not available or desirable. In recent years, applying FEC codes to Peer-to-peer data distribution networks is an innovation. In this way, the downloader can efficiently reconstruct the data from the encoding symbols multicasted from multiple distributors. However, FEC codes suffer from pollution attacks where the attackers corrupt some encoding symbols such that the data can not correctly be reconstructed. Although Krohn et al. propose a homomorphic hash based verification scheme, the scheme is specifically designed for Fountain codes and the variants. That is, the scheme can not apply to other FEC codes and suffers from another specific attacks, distribution attacks. Thus, in this paper we propose a new verification scheme for polynomial based FEC codes like Vandermonde codes. The scheme can defend against not only pollution attacks but distribution attacks. The analysis also shows the security and feasibility of the proposed scheme.

Multicast transmission enables a sender to distribute data to multiple intended receivers simultaneously. However, the multicast transmission should be reliable in erasure channel. Forward Error Correction (FEC) enables the lost packets to be recovered from some extra redundant packets without any retransmission. In this paper, we propose a new FEC codes, called Polynomial Interpolation codes (PI codes), to provide reliable multicast data distribution. The PI codes which are inspired by secret sharing technology satisfy all requirements of FEC codes and more importantly, require less redundant transmission than other FEC codes. More specifically, a data consisting of k input symbols can be reconstructed from any k of the sufficiently large number of output symbols encoded without overhead. The bit length of an output symbol may be equal to that of an input symbol without any description information. Besides that, our experiments demonstrate the feasibility of PI codes.

In this paper, we provide a brief survey about the recent standardization advances in the field of mobility management for DVB-H mobile TV systems. The mentioned topics include: DVB-H to DVB-H handover, and DVB-H service roaming. In addition, since the DVB-H standard itself specifies only the behavior below the IP layer, other service layer standards over the DVB-H, e.g. OMA BCAST or DVB-IPDC, are required to support IP-based mobile TV services. Hence, the discussion in this paper covers both the mobility issues above or below the DVB-H IP layer.

In recent years, online auction systems such as eBay have been more and more popular. The client-server architecture is usually adopted for the basic architecture of online auction systems. However, the client/server architecture usually causes the problems of bandwidth bottleneck and insufficient storage spaces. Therefore, we propose to replace the client/server architecture of traditional online auction systems to P2P architecture for distributing the loading and storage requirement to peers by sellers and buyers. Also, OpenID is exploited to be integrated into our system so that users do not have to remember multiple accounts and passwords. In this paper, the design of an OpenID integrated P2P auction system is proposed. Also, the comparison of our system design and other related P2P auction systems will be discussed.

Several applications benefited from multicast data distribution because a sender can concurrently transmit data to multiple receivers using fewer bandwidth. However, reliability is an important issue in multicast because retransmission is not very efficient in multicast. In recent years, Forward Error Correction (FEC) codes are proposed to facilitate reliable multicast data distribution and are applied to several applications. FEC provides a more efficient way of reliable multicast data distribution because of no retransmission. In this paper, we propose a new FEC codes (F-CR codes), which are extended from Chinese Remainder codes, require smaller transmission redundancy than conventional FEC codes and support two variants for transmission and computation efficiency. In particular, the r-decomposition variant allows the sender to decompose an output symbols encoded into r smaller sub-symbols. This decomposable property which is firstly proposed in this paper is especially suitable for link unstable networks or link heterogeneity networks like wireless sensor networks. The analysis and experiments also demonstrate the feasibility of F-CR codes.

DVB-H is an IP-based mobile TV standard, which is originated from the Europe, but also planned to be deployed in some Asian and African countries. Since DVB-H itself specifies only the behavior below the IP layer, other service layer specifications over the DVB-H are required for offering IP-based mobile TV applications. Both OMA BCAST and DVB-IPDC specifications are capable for such purpose, and in fact they are of some similarities and differences, which may be interesting to the potential DVB-H operators and related researchers. Notification services are for delivering information about forthcoming events to promptly communicate to terminals or users. Since notification is one of the new functions of the emerging DVB-IPDC phase 2 specification, in this paper we compare the notification services defined by both OMA BCAST 1.0 and DVB-IPDC phase 2 specifications, and identify the important similarities and differences between them.

In this paper, we propose a scalable IP-streaming service framework providing the time-shifting and place-shifting capability to achieve the vision of video everywhere. Traditional PVR (Personal Video Recorder) provides the time-shifting viewing experience. However, when mobile TV and PMP (Portable Media Player) are available, the traditional PVR has to be enhanced with the extra place-shifting viewing experience. Hence, several proprietary IPTV streaming techniques such as orb or SONY location free already can deliver a video stream to various kinds of end user terminals through Internet or wireless LAN. In this paper, we propose a distributed, scalable IP-streaming service framework to enhance the current IPTV streaming techniques. Three important issues are addressed in this paper. First, we propose a suitable metadata format that is capable of being used in the service framework. Second, we discuss how to produce scalable IDTV (Interactive Digital Television) contents that can adapt the display capabilities of diverse terminal devices. Third, we evaluate how to apply the MPGE-4 SVC (Scalable Video Coding) in the service framework.

In this paper, we propose a template-based MHP (Multimedia Home Platform) authoring tool. The temporal and spatial behavior of an MHP application can be authored and stored in an XML-based instance description file. The MHP authoring tool generates the target MHP Java source codes by utilizing the "real programs generators" for the modules used inside the MHP application being authored. The real program generator for a module is generated by a "meta program generator" according to the XML template description for the module. The Java source codes of an instance of the module are generated by the real program generator according to the XML description for the instance. The proposed design on code generation can simplify the way to extend the MHP authoring tool to support new features, such as DRM or PVR. In addition, the MHP applications generated by the authoring tool are also reconfigurable. They can adapt to both the MHP set-top box resource change and the change on transmission bandwidth of a DVB object carousel.

In this paper, we presented an IP-multicast-based Java message service implementation called FJM (fast Java messaging) library. We discussed its architecture, protocols, and implementation issues. For achieving better performance, we revised the original design of the FJM. The details of the refinement were also described in this paper. Then, we compared the performance of both versions of the FJM with the performance of other JMS implementations. After our careful design and tuning, in most cases, the revised version of the FJM can offer the highest throughput among the well known industrial or open source JMS implementations. Therefore, FJM can be used in many innovative or mission critical cluster applications, such as massively multiplayer virtual worlds or military simulator.

Since the fourth quarter of 1999, the authors engaged in a project where Java was employed to build the real-time control program inside a plastic-injection-molding machine. However, at that time, since the official real-time Java standard-the Real-Time Specification for Java (RTSJ) from the Sun Microsystems was still under development, no RTSJ-compliant Java virtual machine was available. To satisfy our requirement, we designed and implemented a real-time extension library for enhancing the Java virtual machine that was already available in the real-time operating system we used. Our extension library has the following advantages: First, the underlying Java virtual machine needs no modification to accommodate it. Second, this extension library is easy to be ported to any other priority-based real-time operating systems. Third, the core of this extension library is basically derived from the subset of the RTSJ public draft, and thus our real-time control program can be move to a forthcoming RTSJ-compliant Java virtual machine without much difficulty. In brief, our work may help suggest ways of implementing RTSJ, or the notion of (degrees of) "minimal" compliance/support of RTSJ based on non-RTSJ Java virtual machines. In this paper, we show the requirement, application programming interface, and implementation of this extension library, and discuss its influence on timing.