Robert B. Magill

A high-throughput scheduling algorithm for a buffered crossbar switch fabric

We examine high-throughput scheduling algorithms for buffered crossbar switch fabrics containing one buffer per crosspoint. We propose a scheduling system that uses longest queue first (LQF) scheduling for virtual output queues (VOQs) at the inputs and round-robin (RR) scheduling for the crosspoints. It is shown, through fluid model techniques, that this system achieves 100% throughput for input traffic that satisfies the strong law of large numbers and that produces a load /spl les/1/N for any input/output pair of an N/spl times/N switching fabric. Simulations indicate that 100% throughput may be attained for a much larger class of admissible loads.

A bandwidth efficient self-healing ring for B-ISDN

This paper examines required capacity in self-healing rings (SHRs) as a function of the protection scheme, spare bandwidth allocation, routing and the traffic demand pattern. Existing SONET bidirectional rings and proposals for B-ISDN bidirectional rings assume a fixed allocation of 50% of the available bandwidth for protection traffic. This paper presents a method of minimum spare bandwidth allocation on bidirectional rings which reduces the required capacity for both symmetric and asymmetric demand between node pairs. It is shown that a bidirectional ring with 1:1 path protection and minimum spare bandwidth allocation requires equal or less ring capacity than existing SONET SHRs for any traffic demand pattern.

A simple shaping scheme for frame-based bandwidth allocation

Packet switching devices that rely on input buffering to prevent cell loss in the switch fabric must regulate the arrival pattern of data packets to the input links of the switch fabric. Frame-based bandwidth allocation can be used in cell-based switch fabrics to regulate the number of cells from a flow entering the fabric during a fixed-length frame of cell slots. We present a simple cell slot assignment algorithm for distributing each flows cell slot allocations within a frame. The algorithm controls the inter-cell spacing for each flow and bounds fabric queues to a small number independent of the frame length. The simplicity of the algorithm supports per-frame changes in the allocated rates.

Encoder reverberations in adaptive predictive coding

In the past, the nonlinear effects of quantization in differential speech coders either were modeled as additive white Gaussian noise or were ignored in analysis and quantified experimentally. Describing functions are used to model the nonlinear quantizer effects of coarsely quantized difference signals found primarily in adaptive predictive coding (APC) systems. The analysis predicts marginal instabilities in APC encoders. The marginal instabilities result in ringing distortions in the reconstructed signal which, in the past, may have been attributed to pitch and spectral predictor mismatches. Adaptive order prediction is introduced, analyzed, and offered as a method for increasing the robustness of APC encoders. >

T1.EMB single encoding distortion improvements using synchronous code correction

The authors report the results of an investigation into the single transcoding performance effects of synchronous code correction (SCC) on the draft American National Standard embedded ADPCM (adaptive differential pulse code modulation) algorithm (T1.EMB). It is established through analysis and simulations that the inclusion of the SCC property in T1.EMB will guarantee no loss in PCM-to-PCM signal-to-noise ratio (SNR) measurements and for many input signals will improve the PCM-to-PCM SNR. The SNR improvements are shown to be modest, which leaves as an open issue the question of whether the SCC complexity is justified.<<ETX>>