A statistical multiplexer supporting a number of bursty sources is modeled as a discrete time, single server queueing system with an infinite buffer. The probability generating function (PGF) method is used to analyze the queueing behavior. The PGF method requires the determination of a large number of boundary values and, hence, the roots of the characteristic equation. An iterative algorithm to evaluate the characteristic roots is proposed. When the arrival process is a superposition of independent processes, a decomposition approach is used to reduce the state space involved in the computational algorithm. Additionally, the computational algorithm is made efficient through the establishment of conditions under which all the roots are either real or nonnegative real numbers. A set of equations to recursively compute the moments of the queue length are established. Sample applications of the computational methods to evaluate the performance of a multiplexer supporting voice and video sources, modeled by two-state Markov and L-state MMPP processes, respectively, demonstrate the viability of the proposed methods.

Buffer allocation to provide an efficient and fair use of the available buffer spaces is critically important for ATM networks. A complete sharing with virtual partition (CSVP) strategy for buffer management at a multiplexer or an output port of an output buffered switch is proposed and analyzed. The total buffer space is partitioned based on the relative traffic loads (measured or estimated). Virtual partition allows a newly arriving cell belonging to an oversubscribed type to occupy the spare space of an undersubscribed type, and to be overwritten when necessary. Using a fluid flow approach, a set of partial differential equations with a triangular stability region is established to characterize the dynamics of a system supporting two traffic flows. Under a buffer full condition, the system behavior is described by a set of non-homogeneous ordinary differential equations. The cell loss probability for each traffic type is obtained by solving the ordinary differential equations. Numerical examples indicate that by appropriately selecting the partitions, the CSVP method may be used to provide differential cell loss rate requirements by the different traffic types.

The integrated transmission of voice and data at a time-division multiplexer (TDM) is discussed and analyzed. The system operates in a frame format and the channel capacity is governed by the frame size. The allocation of channel capacity for the transmission of voice and data is performed by a controller. Digital speech interpolation (DSI) and embedded coding techniques are used to enhance the transmission efficiency and to facilitate the implementation of multiplexing. Using a dynamic programming approach, a capacity allocation policy which jointly optimizes the voice/data performance is introduced. Numerical results indicate that the aggregate throughput of the system can be improved with a slight degradation in voice quality.>

A model of multiuser variable rate subband coding of speech incorporating digital speech interpolation (DSI) is presented and analyzed. A dynamic bit assignment algorithm which minimizes the mean squared reconstruction error is presented. A backward estimation of subband variances which makes the bit assignment algorithm backward-adaptive and greatly reduces the need to send side information is proposed. A novel switched-stepsize adaptation algorithm is used at the coders to improve system performance. With adaptive buffer control, buffer overflow is completely prevented. It has been shown by simulation means that a 2:1 DSI gain can be obtained with a relatively small buffer size.>

The integrated transmission of voice and data at a time-division multiplexer (TDM) is discussed and analyzed. The system operates in a frame format and the channel capacity is governed by the frame size. The allocation of channel capacity for the transmission of voice and data is performed by a controller. Digital speech interpolation and embedded coding techniques are used to enhance the transmission efficiency and to facilitate the implementation of multiplexing. Using a dynamic programming approach, a capacity allocation policy that jointly optimizes the voice/data performances is introduced. Numerical results indicate that the aggregate throughput of the system can be improved with a slight degradation in voice quality.>