This paper focuses on live video distribution by video streaming technology. In general, it is an important challenge for video streaming to preserve spatial, temporal, continuous, and real-time quality of video. Pre-roll streaming, which buffers a certain amount of data before video playback, can retain spatial and temporal quality without degradation of continuous quality, while the problem about real-time quality remains because it is difficult to know data amount of video filmed in future on live video distribution. Hence, we propose the method to apply pre-roll sequentially for segmented video data divided by points where continuity of video content breaks, and show it can meet spatial, temporal and continuous quality with few degradation of real-time quality. In the experiment, we show the proposed method met other three qualities with few degradation of real-time quality for lecture video streaming.
In this paper, we describe a new approach to fault diagnosis method of rotating machine by parameter estimation. At first, we derive the moving equations of rotating machine including failure parameters and the observing equation. Next we define the parameter estimation problem to detect symptoms of failure. This parameter estimation problem is devided into two types by terms including failure parameter. In order to apply particle filter to these problems, the moving equations are discretized by Euler-Maruyama method and an approximated parameter estimation problem is defined. Then we conducted numerical simulations, so estimators are effective to detect failure parameters. Moreover log likelihood is comfirmed to decide which failure model is better or not. By using these parameter estimation and model comparison method, we can treat more complex fault diagnosis of rotating machine.
In the discussion of IETF Multi6 Working Group, multihoming by multi-PA (Provider Aggregatable) addressing became rough consensus as the architecture for IPv6 site-multihoming. There are some open issues on this architecture such as source address selection by a host and bypassing ingress filtering of upstream ISPs. In this paper, we propose to introduce a restriction on coverage of source address dependent routing in host centric multihoming to only default route(s). By this proposal, the number of routing entries is increased not multiple of number of upstream ISPs but only the number of upstream ISPs itself. This proposal makes no side effects. Moreover, we have designed and tested a source IPv6 address dependent dynamic routing protocol. This proposed routing gives a consistent route to an exit router with respect to a source address without any modifications on existing routing implementations. This proposal is applicable even in a network which includes commercially-produced routers therefore this proposal improves the applicability of multi-PA address type multihoming in actually operated networks.
In this research, we propose a distributed control method for a multi-vehicle system. Due to the communication range between vehicles, we need to consider the range when vehicles move toward objective positions, and moreover we need to control the network topology. We formulate a multi-vehicle system as a hybrid dynamical system in which we can control both vehicle positions and the network topology. The system can be controled by receding horizon method, but the method may require a long time because there are many discrete variables. Then, we analyze the constraint space and propose a distributed control method with reduced discrete variables. We show the advantage and computatinanal time of our method by a computer simulation.
In this paper, we investigate the game theoretic coverage control whose objective is to lead agents to optimal configurations over a mission space. In particular, the objective of this paper is to achieve the control objective (i) in the absense of the perfect prior knowledge on importance of each point and (ii) in the presence of the action constraints. For this purpose, we first formulate coverage problems with two different global objective functions as so-called potential games. Then, we present a payoff-based learning algorithm determining actions based only on the past actual outcomes. The feature of the present algorithm is to allow an agent to take an irrational action. We also clarify a relation between a design parameter of the algorithm and theprobability which agents take the optimal actions and prove that the probability can be arbitrarily increased. Then, we demonstrate the effectiveness of the present algorithm through experiments on a testbed.