“…Other specification languages include PREMO 8 and Object-Oriented Programming. 9 We find that the above models are not able to specify all the four categories of synchronization scenarios. Their key deficiencies are: various types of media objects and dynamic module structure are not supported; they cannot manage higher layers synchronization control operations; there is no reference clock in the system; and they do not provide explicit temporal constraints for complex synchronization scenario specifications.…”
SUMMARYMultimedia synchronization scenario modelling can be classified into four categories: axes-based model; synchronization point control; event-based model; and interval-based model.1 Existing formal languages do not support the specifications and verifications of all these four categories of synchronization scenarios. Estelle, 2 an internationally standardised Formal Description Technique (FDT), unfortunately does not have enough expressive power to specify the time-dependent behaviours of a multimedia system. To address this limitation, we have developed an extended Estelle, called Time-Estelle 3 to express multimedia QoS parameters, to identify various types of media objects, to describe temporal and spatial relationships, and to specify a range of synchronization scenarios. This paper discusses how Time-Estelle can be used for specifying the four categories of synchronization scenarios, and describes how these scenarios can be verified for correctness, thus enabling any potential temporal inconsistencies to be identified.
“…Other specification languages include PREMO 8 and Object-Oriented Programming. 9 We find that the above models are not able to specify all the four categories of synchronization scenarios. Their key deficiencies are: various types of media objects and dynamic module structure are not supported; they cannot manage higher layers synchronization control operations; there is no reference clock in the system; and they do not provide explicit temporal constraints for complex synchronization scenario specifications.…”
SUMMARYMultimedia synchronization scenario modelling can be classified into four categories: axes-based model; synchronization point control; event-based model; and interval-based model.1 Existing formal languages do not support the specifications and verifications of all these four categories of synchronization scenarios. Estelle, 2 an internationally standardised Formal Description Technique (FDT), unfortunately does not have enough expressive power to specify the time-dependent behaviours of a multimedia system. To address this limitation, we have developed an extended Estelle, called Time-Estelle 3 to express multimedia QoS parameters, to identify various types of media objects, to describe temporal and spatial relationships, and to specify a range of synchronization scenarios. This paper discusses how Time-Estelle can be used for specifying the four categories of synchronization scenarios, and describes how these scenarios can be verified for correctness, thus enabling any potential temporal inconsistencies to be identified.
“…The work was dedicated to intra-media synchronization for stored information and did not provide higher level mechanisms for interrelations between different media streams supporting inter-media synchronization. Blair et al [35] approached support real-time intraand intermedia synchronization in a distributed environment relying on two key concepts, Reactive Objects (for real-time control and synchronization) and QoS Controlled Bindings (for predictable communication between objects). The real-time synchronization aspects can be programmed using a synchronous language.…”
This paper presents experiences from the implementation of Multimedia Synchronization Mechanisms related to the support ol' the many-to-one inter-media synchronization and describes the proposed Synchronization Architecture and its Dcsign Objectives. The main goal of thc communication system in the proposed architecture is to maintain the asynchrony lcvels between the media streams belonging in a Synchronization Group between the levels speciticd by the Multimcdia Application. This is achieved through appropriate Synchronization Primitives related to the mnnagemcnt and monitoring of the synchronization. The pilper emphasizes on the implcmentation issues and problems, and the solutions adopted. concluding to thc lirst experiences from the use of the synchronization. 24-1 ETT
“…A real-time presentation system (RPS) collects the data of various source entities, transports the data through the underlying network, and delivers the data at destination entities for consumption [1]. During a processing of data, maintaining the required temporal association between the data of various streams is necessary in the presence of system induced delays and asynchrony (e.g., multimedia synchronization).…”
Abstract. In this paper, we propose a programming model based on 'timed event dissemination' for structuring a distributed real-time multimedia presentation. In this model, event notifications capture programgenerated actions and/or user-level object accesses on a multimedia window. A coherent effect of these actions requires enforcing deadlines on the event processing over prescribed time intervals. To meet this requirement, the paper advocates an integration of the 'flow of time' as part of the semantics of data presentation on a multimedia window. The paper explores a programming paradigm for event processing: causal ordering of timed messages, to realize multimedia data presentations. This yields simplicity and uniformity in the programming structure of multimedia applications. The presentation specifications in our model can be easily and accurately mapped onto system-level QOS parameters (such as network delays and play-out buffer delays) for scheduling purposes. This in turn may lead to an optimal use of the system resources by a multimedia presentation protocol. The generality of our event-oriented programming interface also allows reducing the multimedia system development costs through software reuse.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.