Set Of Fiber Delay Lines Research Articles

Performance modeling of QoS differentiation in optical packet switching via FDL access limitation

We present an exact analytical model for single-wavelength quality of service (QoS) differentiation in a two-class optical packet switch. In this system, QoS differentiation is achieved by limiting the set of fiber delay lines (FDLs) to the low-priority class, whereas the high-priority class is allowed to access the entire FDL bank. The analytical model is based on multi-regime Markov fluid queues and is extensible to multi-class systems with more than two classes. Markovian arrival process packet arrivals and phase-type distributed packet sizes are considered for the purpose of generality. The proposed analytical model is validated through simulations. The numerical results provide insight into determining appropriate subsets of FDLs allowed for the access of the low-priority class. The results also show that it is possible to direct almost all the packet losses to the low-priority class under moderate loads.

Construction of Polynomial-Size Optical Priority Queues Using Linear Switches and Fiber Delay Lines

One of the main problems in all-optical packet switched networks is the buffering of packets. A popular solution for buffering packets is to use a set of fiber delay lines attached to a switch. A priority queue is one of the most general buffering schemes that allows the packet with the highest priority to depart the buffer on a departure request and dropping of the lowest priority packet if a new packet arrives when the buffer is full. We present a recursive algorithm for constructing optical priority queues of polynomial size from a switch with linear number of inputs/outputs and fiber delay lines. The best known lower bound allows the construction of a priority queue of size $2^{M}$ using a switch of size $M$ . However, the best known upper bound constructs a priority queue of size $M^{3}$ using a switch of size $M$ . We show that it is possible to construct a priority queue of size $O(M^{c})$ for a fixed constant $c$ using a switch of size $O(M)$ .

A Wavelength-Routed Multicast Packet Switch With a Shared-FDL Buffer

In this paper, we propose a wavelength-routed multicast packet switch using a set of fiber delay lines as a shared buffer for both unicast and multicast packets. To avoid multicast packet contention, a multicast packet could be replicated in multiple timeslots by means of multi-wavelength conversion in conjunction with the shared-FDL buffer. Moreover, a packet scheduling technique is designed to assist the operation of the newly proposed switch. We investigate the traffic performance and the complexity in terms of component counts of the newly proposed switch in comparison with our previously reported multicast switch and the broadcast-and-select switch by simulation. Performance evaluation shows that the proposed switch with assistance of the packet scheduling technique can effectively avoid the performance deterioration caused by the increase of multicast traffic, leading to a much better performance than our previously reported multicast switch. In addition, the results show that the proposed switch exhibits an efficient and cost-effective configuration for packet multicasting over other switch architectures.

Exact performance analysis of a single-wavelength optical buffer with correlated inter-arrival times

Providing a photonic alternative to the current electronic switching in the backbone, optical packet switching (OPS) and optical burst switching (OBS) require optical buffering. Optical buffering exploits delays in long optical fibers; an optical buffer is implemented by routing packets through a set of fiber delay lines (FDLs). Previous studies pointed out that, in comparison with electronic buffers, optical buffering suffers from an additional performance degradation. This contribution builds on this observation by studying optical buffer performance under more general traffic assumptions. Features of the optical buffer model under consideration include a Markovian arrival process, general burst sizes and a finite set of fiber delay lines of arbitrary length. Our algorithmic approach yields instant analytic results for important performance measures such as the burst loss ratio and the mean delay.

The impact of synchronization on the performance of FDL buffers

Circumventing the speed bottleneck of electronic switching, novel switching approaches like optical burst switching (OBS) and optical packet switching (OPS) handle the switching of bursts (or packets) in backbone nodes optically, and include a set of fiber delay lines (FDLs) for optical buffering. While previous work acknowledges the performance difference between optical FDL buffers and electronic RAM buffers, the important role of synchronization herein has received little attention to date. Addressing this, we developed a single-wavelength FDL model that allows to capture the interplay between FDL buffer performance, burst size distribution and synchronization. Relying on generating functions, we constructed a discrete-time queueing model with batch arrivals that is valid for general burst and batch size distribution, and allows to quantify how synchronization impacts performance. This paper presents our model and compares its output for different synchronization levels. Whether or not buffer performance benefits from synchronization, turns out to depend strongly on the burst size distribution.

Stability of single‐wavelength optical buffers

AbstractOptical burst switching (OBS) provides a future‐proof alternative to the current electronic switching in the backbone, but has buffering implemented with a set of fiber delay lines (FDLs). The resulting buffering system fundamentally differs from a classic one, in that the set of possible waiting times is not a continuum (like in the classic case) but rather a denumerable set, each value corresponding to the length of a delay line. As a result, arriving bursts in general have to wait longer than they would in a classic buffer, since their waiting time has to be in that denumerable set. The additional delay results in overall longer waiting times, when compared to a classic infinite buffer system with a continuous waiting room.While previous work already focused on the performance evaluation of finite optical buffers, the stability problem of the infinite system received only little attention. This contribution is the first to present a complete proof of sufficient stability conditions in the case of a general infinite FDL set, general arrival process and general service times. The key elements of analysis is the exploiting of the regenerative property of the waiting‐time process and a characterisation of the limiting forward renewal time process. The given bound on the traffic load guarantees stability for a wide class of GI/G/1 optical buffers, and poses no restriction on the FDL lengths. Copyright © 2009 John Wiley & Sons, Ltd.

A light-weight performance model for optical buffers

Providing a photonic alternative to the current electronic switching in the backbone, optical burst switching (OBS) requires that also the buffering is implemented optically, with a set of fiber delay lines (FDLs). Previous performance studies pointed out that the resulting optical buffer suffers additional performance loss (when compared to RAM), but did not allow an exact quantification of this difference for general traffic assumptions. The optical buffer model we constructed considers general inter-arrival times and burst sizes, and is valid for general line lengths. Important features include exactness (no approximations) and minimal numerical complexity (light-weight). This document provides formulas for the waiting times and the loss performance. Also, performance graphs illustrate the paramount impact of correlation, that can be determined exactly and with minimal computational effort with the developed model.

Architecture and Performance of a Broadcast and Select Photonic Switch

A new photonic packet switch based on wavelength encoding of information and on the broadcast-and-select principle for routing is presented. A suitable management of a set of fiber delay lines to allow all-optical output queuing is performed through electronic control. Implementation and traffic aspects are analyzed. Experimental results are given to demonstrate the feasibility of the switch and its future evolution, achieved by a rack-mounted subequipped demonstrator. Different design options are discussed in order to achieve a global terabit per second capacity. Performance is evaluated by simulation under the typical traffic conditions that arise in the broadband communication environment.

Optical Fiber Buffer for High‐Performance Broadband Switching

AbstractThis paper presents an optical buffer shift register based on a set of fiber delay lines electrically‐controlled, to provide cell storage and queueing capabilities in high‐performance broadband switches satisfying ATM service requirements. The concept of multi‐wavelength cell buffering is adopted, to realise a partially shared buffer with a storage capacity equivalent to a set of single shift registers. The final system is well‐suited for photonic switches adopting time‐ and wavelength‐division multiplexing techniques to perform the routing functions inside the switching matrix.We have demonstrated both the suitability of optical fibers as signal delay lines and the use of semiconductor optical amplifiers as gates associated to passive splitters, to realise cell buffering and time‐switching. An experimental set‐up has been implemented at 600 Mbit/s, operating without guardbands between packets. Good performances were obtained in terms of gain and speed of the whole system: the slight penalty and time shifts observed show that the system synchronisation is very good, thus capable of allowing higher operating bit‐rates.