On VBR video teleconferencing over ATM networks

Abstract

This paper examines the problem of video transport over ATM networks using knowledge of both video system design and broadband networks. We find that the following trends hold for all sequences examined. First, increasing the delay in the video system decreases the necessary peak rate and significantly increases the number of calls that can be carried by the network. Second, as an operational traffic descriptor for video, the leaky-bucket algorithm appears to be superior to the sliding-window algorithm. And finally, with a delay in the video system, the statistical multiplexing gain from VBR over CBR video is upper bounded by roughly a factor of three, and to obtain a gain of 1.5 to 2.0 can require the operational traffic descriptor to have a window or bucket size on the order of a thousand cells. We briefly discuss how increasing the complexity of the video system may enable the size of the bucket or window to be reduced.

Issues of quality and multiplexing when smoothing rate adaptive video

“…Hence (7) is replaced by r all (n) = p(n )r req (n ) n > r 0 n (9) Note that this ignores some second order effects: reduction of the allocated rate for a given frame may increase buffer occupancy at the end of that frame time, and hence the rate request on the next frame may be increased to remain within the delay target. Or, this could cause cropping to be increased.…”

“…A segment of trace set A was subjected to contention from the aggregate requested rate of ten traces from set E, as modeled by (8) and (9). The points show the ratio of encoded to ideal rate (i.e., 1 cropping proportion).…”

“…[8], [9], [10], [11], [14]. These require advance knowledge of at least a part of the future sequence of frame sizes.…”

ThinStreams: an architecture for multicasting layered video

“…We propose that the source specify a mapping of thin streams to thick streams in the base layer. For example, it might specify that it can use thin streams grouped as [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]. This means that although the ThinStreams layer in the receiver might have joined 4 groups, the decoder will only receive the first 3 groups (which correspond to a whole layer One of the disadvantages of using ThinStreams is slow convergence.…”

“…Several unicast video distribution systems have studied the problems associated with storing a scaleable video stream on a server [9,10,11,12]. The server receives feedback from the client or the network, and adapts the transmission rate based on a control algorithm.…”