Ultrafast Packet Research Articles

Label Stacking in Photonic Packet-Switched Networks With Spectral Amplitude Code Labels

Label stacking is used for hierarchical addressing to reduce the size of lookup tables and to increase the speed of the routing process. We propose an optical label stacking using spectral-amplitude codes (SAC) as labels to accomplish ultrafast packet forwarding. We discuss the advantages of this label architecture compared to other proposals in the literature and present experimental results. We experimentally examine two types of optical packets, one with separable SAC labels and the other one with SAC-encoded payloads. In the first case, the SAC label is a collection of spectral tones modulated at the packet rate (low rate), and the payload is on a separate wavelength modulated at the data rate (fast rate). In the second case, the payload data modulates the collection of wavelengths that constitute the code. We implement a network with two forwarding nodes, and we transmit the packets with two labels in the label stack over 80 km of fiber and measure the bit error rate (BER) after two hops. We achieve error-free transmission (BER<10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-9</sup> ) for the packets with SAC labels and SAC-encoded payload at payload bit rates of 10 and 2.5 Gb/s, respectively. This is the first experimental demonstration of optical label stacking to our knowledge

Ultrafast all-optical signal processing and packet switching

Ultrafast all-optical signal processing and packet switching

Packet clock recovery using a bismuth oxide fiber-based optical power limiter

We demonstrate an optical clock recovery circuit that extracts the line rate component on a per packet basis from short data packets at 40Gb/s. The circuit comprises a Fabry-Perot filter followed by a novel power limiting configuration, which in turn consists of a 5m highly nonlinear bismuth oxide fiber in cascade with an optical bandpass filter. Both experimental and simulation-based results are in close agreement and reveal that the proposed circuit acquires the timing information within only a small number of bits, yielding a packet clock for every respective data packet. Moreover, we investigate theoretically the scaling laws for the parameters of the circuit for operation beyond 40 Gb/s and present simulation results showing successful packet clock extraction for 160 Gb/s data packets. Finally, the circuit's potential for operation at 320 Gb/s is discussed, indicating that ultrafast packet clock recovery should be in principle feasible by exploiting the passive structure of the device and the fsec-scale nonlinear response of the optical fiber.

A rack-mounted all-optical packet-switching system using a terahertz optical asymmetric demultiplexer (TOAD)

We introduce a novel all-optical packet-switching system using a terahertz optical asymmetric demultiplexer (TOAD) based on optical time-division multiplexing (OTDM). Ultra-fast packet switching is performed without undergoing optical-to-electronic conversion. An optical packet at 10 Gb/ps is generated with a high-definition video signal. All-optical packet-level clock extraction for synchronization of the switch node is performed using a modified TOAD. The extracted packet-level clock is used for detecting the header and the payload data. By integrating the above all-optical processors, we set up a rack-mounted all-optical packet-switching system and successfully demonstrate exchanging the video signals from the source to the destination. © 2006 Wiley Periodicals, Inc. Microwave Opt Technol Lett 48: 577–581 2006; Published online in Wiley InterScience (www.interscience.wiley.com) DOI 10.1002/mop.21412

Fast and Widely Tunable Optical Packet Switching Scheme Based on Tunable Laser and Dual-Pump Four-Wave Mixing

An optical packet switching scheme based on a fast electronically controllable tunable laser dual-pump four-wave mixing for wavelength conversion and an N/spl times/N arrayed waveguide grating for passive routing is examined. This scheme can realize ultrafast packet switching and large port switch fabrication, required in next-generation core optical packet routers. The performance is examined in terms of packet-based bit-error-rate measurements for packet-by-packet switching. Design issues are also discussed.

Ultrafast optical packet switching using low optical pulse energies in a self-synchronization scheme

A detailed numerical treatment is undertaken to investigate the operating characteristics of a self-synchronization scheme based on a modified terahertz optical asymmetric demultiplexer (TOAD) device for ultrafast packet switching. The simulations agree with published experimental measurements. It is shown that input picosecond optical pulse energies as low as /spl sim/100 fJ can be utilized to achieve a larger than 10 dB contrast ratio of the first to the remaining pulses of a transmitted signal packet. The utilization of low input pulse energies in the scheme also provides significant advantages including insensitivity of contrast ratio to pulse energies and pulse widths as well as distortionless extracted pulse shapes. Means for achieving a maximum contrast ratio are identified.

Ultrafast multihop packet-switched optical time-division multiplexing: components and systems

Ultrafast processing techniques for the implementation of high-speed all-optical packet-switched networks are reviewed. Some of the key technologies required for building the next generation of high-speed interconnection networks include ultrafast packet generation, high-speed packet demultiplexing, and efficient packet routing. A review of recent component, subsystem, and system development as well as experimental demonstrations being conducted at Princeton University is presented to illustrate some of the capabilities of these technologies. © 1998 Society of Photo-Optical Instrumentation Engineers.

Routing of 100 Gb/s words in a packet-switched optical networking demonstration (POND) node

This paper presents the design and experimental results of an optical packet-switching testbed capable of performing message routing with single wavelength time division multiplexed (TDM) packet bit rates as high as 100 Gb/s. The physical topology of the packet-switched optical networking demonstration (POND) node is based on an eight-node ShuffleNet architecture. The key enabling technologies required to implement the node such as ultrafast packet generation, high-speed packet demultiplexing, and efficient packet routing schemes are described in detail. The routing approach taken is a hybrid implementation in which the packet data is maintained purely in the optical domain from source to destination whereas control information is read from the packet header at each node and converted to the electrical domain for an efficient means of implementing routing control. The technologies developed for the interconnection network presented in this paper can be applied to larger metropolitan and wide area networks as well.

Self-routing of 100 Gbit/s packets using 6 bit ‘keyword’address recognition

The authors report the first demonstration of all-optical header recognition and self-routing of ultrafast packets with multibit addresses. A single optical AND gate recognises 6 bit ‘keyword’ codes, allowing self-routing of 100 Gbit/s packets. [This Letter first appeared in print in 1995, issue 17, p.1475–1476, and has been reprinted because of the significance of the corrections.]

Time-of-flight cross correlation on a detector array for ultrafast packet detection

We describe and demonstrate an interferometric technique for measuring the first-order cross correlation of ultrafast optical pulses. This technique may permit single-shot pulse detection and is applicable to receivers for time-domain optical communications.

Self-clocking scheme for bit synchronisation in ultrafast packet switching transparent optical networks

A challenging issue in ultrafast packet switching transparent optical networks is fast acquisition of synchronisation at the bit level for packet demultiplexing. A novel scheme is proposed, in which a timing reference is transmitted along with each data packet in the form of a clock comb. This comb is extracted at each node reached by the packet for header recognition and, at the destination, for reading of that particular packet. The synchronisation problem is solved with no need for a phase locked loop, at the expense of an increased effective packet length.