Photonic Packet Switching Research Articles

Performance Evaluation of the Loop Buffer Switch Under Prioritized Traffic and Optical Regeneration

In this paper, an all-optical regenerator based, photonic packet switch architecture, which consists of the fiber loop for the storage of the contending packets, is considered. In the loop buffer, the available buffer space may not be fully utilized due to the limited re-circulation count of the data placed on buffer. This limit can be counteracted by placing a pool of regenerators inside the buffer. As optical regenerators are costly devices, hence they should be placed optimally in the buffer. The simulations results are presented by consider Prioritized and non - prioritized traffic. It is shown in the results that regeneration of data is essential if prioritized traffic has to be considered.

Scaling Photonic Packet Switches to a Large Number of Ports [Invited

We review our work on optical switching architectures that allow for scaling to a large number (thousands) of ports. We explain that the complexity of the node control critically depends on the node architecture and, hence, critically impacts the end-to-end latency of the system. We introduce node architectures with highly distributed control that allow for systems with very low end-to-end latency. We present integrated devices that support such systems and we present laboratory experiments to illustrate the technical feasibility of such switching systems.

Bias-free, low power and optically driven membrane InP switch on SOI for remotely configurable photonic packet switches

A small footprint integrated Membrane InP Switch (MIPS) on Silicon-On-Insulator (SOI) is demonstrated for use in all-optical packet switching. The device consists of an optically pumped III-V membrane waveguide of only 100 nm thick, coupled to the underlying SOI waveguide circuit. Because of its limited thickness, the optical confinement in the active layers is maximized, allowing for high extinction ratio of over 30 dB when applying a low power optical pump signal, over the entire C-band. The switch has 400/1300 ps on/off switching times and no measurable pattern dependence or switching related power penalties for a bitrate up to 40 Gb/s, using a switching power of only 2 dBm.

Fiber optic loop buffer switch incorporating 3R regeneration

In this paper, an all-optical regenerator based, photonic packet switch architecture, which consists of fiber loop for the storage of the contending packets, is proposed. In the loop buffer, the available buffer space may not be fully utilized due to the re-circulation limit (maximum number of re-circulations that data can take before it can be correctly received at the output). The re-circulation limit arises due to the accumulated ASE (amplified spontaneous emission) noise of the optical amplifier and crosstalk of the optical components. This limit can be counteract by placing a pool of regenerators inside the buffer, which regenerates the signal after some fixed number of re-circulations. As optical regenerators are costly devices, hence they should be placed optimally in the buffer. To do so, in this paper an expression is presented to evaluate the minimum number of required regenerators, to be placed optimally in the buffer for different buffer configurations (B) and re-circulation limits (K), such that full buffer capacity can be utilized without any re-circulation constraints. The simulation results in terms of packet loss probability are presented by considering the detrimental effect of re-circulation limits. Finally, advantages gain due to the regenerator placement inside the buffer are shown.

All-optical router with pulse-position modulation header processing in high-speed photonic packet switching networks

In future high-speed photonic packet switching networks, it is highly desirable to carry out robust all-optical header recognition to provide high-throughput routing. The authors present a pulse-position modulation header processing (PPM-HP) scheme, offering significantly reduced routing table size and employing a single bit-wise AND operation to carry out correlation of the packet header with the entire routing table entries. The downsized routing table also offers multiple transmission modes (unicast, multicast and broadcast) in the optical layer and improves core network scalability where the number of core/edge nodes could be altered without the need for changing the number of routing table entries. The authors present modelling and simulation of the packet switching router based on PPM-HP. Noise propagation and crosstalk incurred in a multiple-hop routing scenario are investigated. The simulation results are presented and compared with the theoretical calculations.

High-Speed Wavelength-Tunable Optical Filter Using Cascaded Mach–Zehnder Interferometers With Apodized Sampled Gratings

A high-speed wavelength-tunable optical filter is a key device for selecting different wavelength optical packets on a packet-by-packet basis in a photonic packet switching system. In this paper, we describe a high-speed tunable filter using apodized sampled gratings which can achieve the required filter characteristics at low cost. A sampled grating structure that exhibits a comb-like filter response is adopted to obtain a wide tuning range, implemented in InGaAsP-InP deep-ridge waveguides with vertical-groove surface gratings. In addition, apodization is applied to the sampled grating to achieve a high extinction ratio. We fabricate the filter by integrating the sampled gratings within Mach-Zehnder interferometers. The filter exhibits a tuning range of 16 nm, an extinction ratio of 20 dB, a 3-dB transmission bandwidth of 32 GHz, and a switching time of less than 8 ns.

Optical Code Division Multiplexing for Packet Labeling in Optical Switching Networks

In this paper, we proposed a photonic packet switching control method by used optical correlator for optical packet label packet-switched networks for next Generation networks. The main advance is rely on using the Optical Code Division Multiplexing (OCDM) code to labeling optical packets based on source routing. Based on OCDM labeling either header modification or any label swapping techniques can be avoids. With advantage of existing OCDM coding called OCDM-labels schemes to encapsulate the packets, together with optical correlator to decode the label in optical domain, which can achieve optical packet switching without header modification/label swapping techniques. The O/E/O conversion procedure at each switching device can also be eliminated. This method not only simplifies the design of switch devices in the optical domain to simplify the packet forwarding process, but also speeds up packet forwarding and increases throughput significantly.

Full $C$-Band Tuning Operation of Semiconductor Double-Ring Resonator-Coupled Laser With Low Tuning Current

A widely tunable double-ring resonator-coupled laser is developed for use in a photonic packet switching system. By using a double-ring resonator as a wavelength-selective filter, full -band tuning operation is achieved with a low tuning current, and the fabrication process is simpler than that used for distributed-Bragg-reflector-type tunable lasers. A tuning range of 50.0 nm with a current injection of less than 7 mA is successfully demonstrated.

Performance evaluation of the Data Vortex photonic switch

The data vortex photonic packet-switching architecture features an all-optical transparent data path, highly distributed control, low latency, and a high degree of scalability. These characteristics make it attractive as a routing fabric in future photonic packet switches. We analyze the performance of the data vortex architecture as a function of its height and angle dimensions, H and A. The investigation is based on two performance measures: the average delay and the maximum throughput of the switch. We present an analytical model assuming uniform traffic and derive closed-form expressions for these measures. Our results obtained demonstrate that as H increases, the saturation throughput decreases and approaches <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> / <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">9</sub> = 0.22 when A is small and H is large. Furthermore, for fixed switch size, the saturation throughput is maximized when A is minimal. We also present simulation results for the maximum throughput under uniform and nonuniform traffic, as well as for the mean number of hops and the mean input-queue packet delay as a function of input load, and address the issue of resequencing delay. The results obtained advocate that to support more ports, it is preferable to increase the height dimension and to keep the angle dimension as small as possible.

Optical loop memory for photonic switching application

Feature Issue on Photonics in SwitchingWe propose an architecture for a WDM-based photonic packet switch. This is a wavelength routed switch in which, once the wavelength of the incoming packet is tuned appropriately, will either be directly transmitted to the output fiber or will be placed in the buffer, and comes out of the loop buffer after the required amount of delay. The main advantage of this architecture is that no controlling is required inside the buffer. The working of the switch is explained in detail and simulations are done to obtain packet loss probability and delay. These results have been verified by the mathematical model.

Optical fiber-delay-line buffer management in output-buffered photonic packet switch to support service differentiation

We study optical fiber-delay-line buffer management schemes for supporting service differentiation with less complexity, with the objective of achieving ultra-high-speed packet forwarding in a photonic packet switch. We propose the vPBSO (Variable-length-packet-capable Partial Buffer Sharing with Overwriting) method, which is a prioritized buffer management scheme for providing DiffServ Assured Forwarding. This method can handle asynchronously arriving variable-length packets. vPBSO is based on a single queue, and its complexity is Oscr(p), where p is the number of priority classes. We show that vPBSO is effective when the arrival rate of higher priority packets is lower than that of lower priority packets, which is a likely situation in a prioritized system. vPBSO provides better performance than vPBS (an extension of partial buffer sharing) for high-priority packets, and the performance for low-priority packets is also better or at least almost the same. vPBSO provides service differentiation in the case of two or more priority classes in which the difference of packet loss probability in each class is made larger than one order of magnitude by appropriate tuning of thresholds

Optical fiber-delay-line buffer management in output-buffered photonic packet switch to support service differentiation

Optical fiber-delay-line buffer management in output-buffered photonic packet switch to support service differentiation

High-speed buffer management for 40 gb/s-based photonic packet switches

We develop a method of high-speed buffer management for output-buffered photonic packet switches. The use of optical fiber delay lines is a promising solution to constructing optical buffers. The buffer manager determines packet delays in the fiber delay line buffer before the packets arrive at the buffer. We propose a buffer management method based on a parallel and pipeline processing architecture consisting of (log/sub 2/N+1) pipeline stages, where N is the number of ports of the packet switch. This is an expansion of a simple sequential scheduling used to determine the delays of arriving packets. Since the time complexity of each processor in the pipeline stages is O(1), the throughput of this buffer management is N times larger than that of the sequential scheduling method. This method can be used for buffer management of asynchronously arriving variable-length packets. We show the feasibility of a buffer manager supporting 128 /spl times/ 40 Gb/s photonic packet switches, which provide at least eight times as much throughput as the latest electronic IP routers. The proposed method for asynchronous packets overestimates the buffer occupancy to enable parallel processing. We show through simulation experiments that the degradation in the performance of the method resulting from this overestimation is quite acceptable.

An Overview of Photonic Packet Switching Architectures

The fast growing field of telecommunication has led to the generation of new switching technology to support a variety of communication services with a wide range of transmission speeds. But the switching speed of electronics cannot keep up with the transmission capacity offered by the optics. On the other hand, Photonic packet switches offer high speed, data rate transparency, and flexibility required by future telecommunication networks. Optical packet switching provides fine granularity to achieve high asynchronous multiplexing gain but faces significant challenges in the processing and buffering of bits at high speed. Optical burst switching (OBS) attempts to minimize the need for processing and buffering by aggregating flow of data packets into bursts. In this paper, we have reviewed the progress in the research on photonic packet switches. We have given a descriptive overview of the major activities in the rapidly growing field of telecommunication and put them in perspective with each other.

Ultra-low latency optical packet switching node

A photonic packet switching node is introduced, and its routing latency is shown to be 15.3 ns. The power penalty of the node at a bit-error rate (BER) of 10/sup -9/ is confirmed to be approximately 0.2 dB across 33 nm of the C-band for 10-Gb/s payload wavelengths. Moreover, multiple-wavelength packets containing 16 payload wavelengths can be switched while maintaining BERs of 10/sup -12/ or better.

Agile and highly efficient wavelength conversion using highly nonlinear fiber for optical code-labeled packets

Wavelength conversion enhances transparency, interoperability, and efficient wavelength resource utilization in wavelength-division-multiplexing networks. Several devices are used for wavelength conversion, such as the electroabsorption modulator, semiconductor optical amplifier, periodically poled lithium niobate modulator, and highly nonlinear fiber (HNLF). HNLF is promising because it offers high conversion efficiency and phase-preservation between the input and output signals; it also provides immunity from pattern effects. Furthermore, a combination of HNLF and a distributed-feedback laser diode (DFB-LD) array used as a pump light source enables agile wavelength conversion for photonic packet switch networks. In this letter, we discuss our demonstration of agile and highly efficient wavelength conversion of optical code-labeled packets by using HNLF and an eight-DFB-LD array module. We obtained a switching time of less than 1 ns, and the wavelength stability of the LD pump was within 0.04 nm.

A High-Speed Tunable Optical Filter Using a Semiconductor Ring Resonator for Photonic Packet Switch

Large-capacity photonic packet switch using the wavelength division multiplexing (WDM) technology is proposed. In this switch, the optical buffers, consisting of the fiber delay lines, are shared in many input ports by using the WDM. Furthermore, to reduce the number of optical buffers, the speed-up effect in the optical switch part is investigated. A high-speed tunable optical filter incorporating a semiconductor ring resonator is a key device in the proposed packet switch because many optical filters are used. Optical ring resonators should be possible to fabricate them at a low cost because of their simple structure. To achieve a wide tuning range we designed a double-ring structure, in which two ring resonators are connected in series, and fabricated it using the InGaAsP-InP material system. This device exhibits a total free spectral range (FSR) of 1.7 THz and contrast ratio of 9.5 dB. The ring radii are 25.2 and 17.8μm, which correspond to FSRs of 340 and 425 GHz, respectively. The switching time of the device is 2.5 ns.

Rewritable High-Density Optical Recording on Azobenzene Polymer Thin Film

A holographic label processor for photonic packet switching is proposed. We observe the process of time-to-space conversion for an optical code consisting of 2 ps pulses at a wavelength of 1.55 $mUm with 100 km fiber transmission. Label recognition using an angular multiplexed spectral hologram (AMSH) as a label bank is experimentally demonstrated. We also investigate the AMSH by CGH and confirm its feasibility in computer simulations.

Multi-Stage Fiber Delay Line Buffer in Photonic Packet Switch for Asynchronously Arriving Variable-Length Packets

We study photonic packet switches to support asynchronously arriving variable-length packets. A scheduler for contention resolution is operated in electrical domain even when data street of the buffer is provided in optical domain. In this scheme, the scheduler may be a bottleneck. To compensate the gap of high-speed optical transmission and slow-speed electronic processing, we propose a multi-stage fiber delay line (FDL) buffer architecture that forms a tree structure in which each node has a block of FDLs and a scheduler. This is especially useful for output-buffer switches in which scheduling complexity is proportional to the number of ports of the packet switch. Through a newly-developed approximate analytical method, we show the optimum unit length of the fiber delay lines to decrease packet loss probability. We also show the sufficient number of FDLs in the two-stage buffer.

Application of single flux quantum technology to a next-generation photonic packet switch core

Internet traffic is still growing rapidly. Link capacity is easily increased by bundling optical fibers, but the packet switching capacity at a node is limited by the performance of a semiconductor switch facing problems of packaging density, operation speed, and power consumption. Photonic technology is therefore emerging from the link technology into the packet switch technology. A photonic packet switch can analyze the packet label optically and control optical switches, but packet scheduling must be done in the non-optical domain because photonic technology lacks arithmetic characteristics. Increases in the line speed and the number of ports will make this scheduling a bottleneck. Because single flux quantum (SFQ) circuits can operate at several tens of gigahertz, a speed comparable to optical link speed, they will be able to eliminate this bottleneck. To improve the performance of the nodes in the photonic network, we propose an SFQ-circuit-controlled optical packet switch core. Here we describe and discuss two photonic packet switch architectures using SFQ-circuit-controlled photonic switches.