Integrated optical interconnections

The use of optical interconnections between processors, boards, chips, and gates in electronic digital systems to overcome the current performance limitations is described. The advantages of optical interconnections in relation to the interconnection distance, the data capacity, and the interconnection functions are presented. The devices which will support practical implementation of optical interconnections and the integration of optical interconnection devices are discussed. The development of future integrated optoelectronic materials, processing, and fabrication technologies to support integrated optical electronics is also discussed     

Reconfigurable optical interconnections for parallel computing

We describe our research on optically interconnected optoelectronic parallel computing systems. Our architecture is based on a multilayer pipeline of two-dimensional optoelectronic device arrays in which each pixel is composed of an optical input channel, a general-purpose programmable processor, local memory, and a surface-emitting laser diode as an optical output channel. Free-space optics provides parallel, global communication between layers in the pipeline via optical paths that are dynamically reconfigurable. Design and initial realization of a system are described     

Guided-wave optical chip-to-chip interconnections

An optical interconnection technique using a high-silica guided-wave optical circuit composed of channel waveguides and a mixer is proposed for LSI interchip communications. An experimental 4-chip interconnection circuit has 1 Gbit/s transmission capacity performance with an optical power margin of 3 dB.     

Novel broad-band bit-synchronization circuit module for optical interconnections

A novel broad-band and ultrafast bit-synchronization circuit module is proposed and fabricated for optical interconnections. In optical packet switch fabric or optical interconnection between electric circuit boards, instantaneous bit synchronization is crucial to properly retime incoming packets with a random phase and reduce the number of preamble overhead bits. The developed bit-synchronization circuit module has a new clock selection circuit, which is configured with a phase comparator and an amplitude comparator. Since device-dependent delay circuits, such as buffer amplifiers or RC phasors, are not adopted, the newly developed clock selection circuit can operate under broad-band frequencies. The bit-synchronization circuit module was fabricated with a Si-bipolar gate array and it can operate at broad-band bit rates of up to 10.5 Gb/s. It also exhibits a power sensitivity penalty as low as 3 dB for 10-Gb/s input signals. The synchronization acquisition time of less than 9 b over the entire 360/spl deg/ phase range was confirmed by experiment.     

MT-compatible laser-ablated interconnections for optical printed circuit boards

Integration of optical interconnections on a printed circuit board (PCB) is very challenging, since compatibility should be maintained with standard PCB manufacturing technology. This paper describes the use of laser ablation, a technique already used in PCB manufacturing for drilling microvias, as a suitable technique for the fabrication of multimode polymer waveguides, micromirrors, alignment features, and microlenses. A frequency-tripled Nd-YAG laser and a KrF excimer laser are used, both mounted on the same stage, resulting in very high alignment accuracies. This paper demonstrates a parallel optical link over approximately 5-cm-long PCB integrated waveguides, fully connected using a standard MT-based connector. This proves that laser ablation can be a key technology in optical board manufacturing to reach the stringent coupling tolerances.     

Planar packaging of free-space optical interconnections

Planar optics is an approach to build integrated free-space optical systems on single substrates. Computer-aided design, lithographic fabrication, and micro-bonding techniques are used to package the optics in a compact way. This paper reviews recent work on planar optics. It discusses various aspects of the fabrication, the design, and the application of planar optics as an interconnection technology for optoelectronic computing and switching systems     

Low-noise optical receiver for high-speed optical transmission

This paper describes the design of a low-noise optical receiver using Si bipolar transistors for high-speed optical transmission. The conventional common emitter-common collector circuit (CE-CC pair) and Darlingtou circuit (transimpedance amplifiers with parallel feedback) are studied. Optimal CE-CC pair collector-biasing current for attaining minimum noise current with a 400-MHz bandwidth is 2.7 mA, and less than 1.2 mA for the Darlington circuit. It is confirmed that the Darlington circuit is better than the CE-CC pair in signal-to-noise ratio by about 1.5 dB. The low-noise Darlington optical receiver with a Ge-avalanche photodiode has achieved an optical sensitivity of -41 dBm for a 400 Mbit/s RZ pulse with a bit error rate of 10^-10. This is a 2.5-dB improvement in optical sensitivity over that of the conventional CE-CC receiver.     

Low-noise optical receiver for high-speed optical transmission

This paper describes the design of a low-noise optical receiver using Si bipolar transistors for high-speed optical transmission. The conventional common emitter-common collector circuit (CE-CC pair) and Darlington circuit (transimpedance amplifiers with parallel feedback) are studied. Optimal CE-CC pair collector-biasing current for attaining minimum noise current with a 400-MHz bandwidth is 2.7 mA, and less than 1.2 mA for the Darlington circuit. It is confirmed that the Darlington circuit is better than the CE-CC pair in signal-to-noise ratio by about 1.5 dB. The low-noise Darlington optical receiver with a Ge-avalanche photodiode has achieved an optical sensitiyity of -41 dBm for a 400 Mbit/s RZ pulse with a bit error rate of 10^-10. This is a 2.5-dB improvement in optical sensitivity over that of the conventional CE-CC receiver.     

Fully pipelined TSPC barrel shifter for high-speed applications

In this paper we describe a TSPC (True Single-Phase-Clock) barrel shifter for high-speed, real time applications. The circuit consists of a shift and rotate array and a control unit, both having a pipeline structure, which allows on-line programming. Using 1.5 μm CMOS technology, a full custom prototype for 16-b data shift and rotate array has been implemented, with a simplified control unit, and testing circuitry. Worst case HSPICE simulations indicate a maximum operating frequency of over 100 MHz, both for the barrel with on-line programming and for the barrel implemented. The full functionality of the prototype has been tested up to 25 MHz, although each bit path has been tested at 100 MHz     

SMT-compatible large-tolerance "OptoBump" interface for interchip optical interconnections

A new optical interface called OptoBump has been developed to couple optoelectronic packages to an optoelectronic printed circuit board, thus enabling economical chip-to-chip optical interconnections. The optoelectronic packages have vertical-cavity surface-emitting laser (VCSEL) and PD-array chips in their cavity and an large scale integrated (LSI) mounted on top. A package converts high-speed electrical signals from the LSI into an array of optical signals, which are emitted from the bottom. The PCB contains integrated polymer optical waveguides to optically connect packages, and the use of surface-mount technology (SMT) to mount packages on the printed circuit board (PCB) keeps assembly costs low. The key to making the OptoBump interface fully compatible with SMT is the integration of microlens arrays directly into both packages and the PCB. A wide, collimated optical beam couples a package to the board across a narrow air gap and provides a large tolerance to misalignment during the SMT process. This paper explains the concept of the OptoBump interface, the optical coupling design by ray-trace simulation, and the fabrication of polymer microlenses and polymer waveguides. Experimental results revealed that the OptoBump interface provides a large tolerance of /spl plusmn/50 /spl mu/m, which is large enough for use with SMT. The OptoBump interface can replace high-speed electrical wiring at the chip level and also offers the benefit of not having any optical fibers or connectors on the board. Thus, it has the potential to bring about a revolutionary change in optoelectronic packaging.     

Methodology for characterization of high-speed multi-conductor metal interconnections and evaluation of measurement errors

Analysis and design of interconnects in high speed integrated circuits and systems involves models in the form of multiconductor transmission lines. The fundamental parameters of those models are matrices of capacitance, (C), inductance, (L), resistance, (R), and conductance (G). We present a methodology for measurement of entries in capacitance matrix. The entries of capacitance matrices can be calculated using numerical solvers of electrostatic fields established under the assumption of suitable biasing of interconnect structures. Numerical calculations of complete field equations are very complex and expensive in terms of computer time, therefore several approximations are made in constructing interconnect dedicated software packages available on the market. Because of these approximations it is necessary to validate the calculations via measurements. Calculation of the off-diagonal entries of capacitance matrix from measurements of "two-terminal" capacitances is strongly corrupted by the measuring errors. The method involves direct capacitance measurement in multi-conductor structures and provides analysis of accuracy.     

Subcarrier multiplexing for high-speed optical transmission

The performance of high-speed digital fiber-optic transmission using subcarrier multiplexing (SCM) is investigated both analytically and numerically. In order to reduce the impact of fiber chromatic dispersion and increase bandwidth efficiency, optical single-sideband (OSSB) modulation was used. Because frequency spacing between adjacent subcarriers can be much narrower than in a conventional DWDM system, nonlinear crosstalk must be considered. Although chromatic dispersion is not a limiting factor in SCM systems because the data rate at each subcarrier is low, polarization mode dispersion (PMD) has a big impact on the system performance if radiofrequency (RE) phase detection is used in the receiver. In order to optimize the system performance, tradeoffs must be made between data rate per subcarrier, levels of modulation, channel spacing between subcarriers, optical power, and modulation indexes. A 10-Gb/s SCM test bed has been set up in which 4 × 2.5 Gb/s data streams are combined into one wavelength that occupies a 20-GHz optical bandwidth. OSSB modulation is used in the experiment. The measured results agree well with the analytical prediction     

A polymer optical waveguide with out-of-plane branching mirrors forsurface-normal optical interconnections

We have developed a simultaneous fabrication method using temperature control reactive ion etching (RIE) for channel optical waveguides incorporating plural out-of-plane branching mirrors made from polymer film. By using this method, the etching rate can be adjusted locally by controlling the temperature. This technology also enables the formation of trenches of various depths on the same polymer optical waveguide. We noted from scanning electron microscope (SEM) observations that simultaneous control of the mirror tilt angle and a smooth core surface could be achieved. To be specific, a heat treatment temperature of 130-135°C appears to be the optimum to maintain a rectangular cross section and to achieve a sufficiently smooth core surface for a polymethyl methacrylate (PMMA) waveguide. The measured propagation loss is small, in spite of the presence of a high-Δ waveguide (Δ=5.4%). For example, losses of 0.1, 0.3, and 0.7 dB/cm are measured at wavelengths of 650 nm, 850 nm, and 1.3 μm, respectively. From far-field pattern (FFP) measurements, we found that the mirror plane was almost rectilinear, and that the reflected light can be captured efficiently by a photodiode. In operational temperature tests, we showed that intensity fluctuations of the coupling light can be reduced to less than 1.5 dB for the temperature range between -25°C and +85°C by adopting a sandwich structure with glass plates     

Fully differential bilinear integrator for SC filters

A fully differential bilinear SC integrator, which can be used for SC filter realisations, is proposed. It is parasitic-insensitive and particularly useful in SC filters simulating analogue ladder networks. The design of these filters is different from the commonly used ones, and leads to filters with better sensitivity properties than earlier versions. Realisation of a third-order highpass filter, bilinearly transformed from the continuous-time to the discrete-time domain, is shown as an example.     

Single-side-band optical modulation in SCM systems for high-speed optical transmission

In order to decrease dispersion penalty and increase the optical bandwidth efficiency,an optical single-side-band modulation(SSBM) scheme in sub-carrier multiplexing(SCM) is proposed.The principle of the SSBM is analytically presented,and a configuration for generating optical SSB signal is proposed using a balanced Mach-Zehnder electro-optic modulator.     

Packet optical networks for high-speed TCP-IP backbones

This article presents a new proposal for TCP-IP backbone implementation based on optical packet switching technology. The proposed network architecture merges the flexibility in resource management of packet switching with the high capacity offered by full optical technology     

Fully hardware based WFQ architecture for high-speed QoS packet scheduling

A full hardware implementation of a Weighted Fair Queuing (WFQ) packet scheduler is proposed. The circuit architecture presented has been implemented using Altera Stratix II FPGA technology, utilizing Reduced Latency DRAM (RLDRAM) II and Quad Data Rate (QDR) II SRAM memory components. The circuit can provide fine granularity Quality of Service (QoS) support at a line throughput rate of 12.8 Gb/s in its current implementation. The authors suggest that, due to the flexible and scalable modular circuit design approach used, the current circuit architecture can be targeted for a full ASIC implementation to deliver 50 Gb/s throughput. The circuit itself comprises three main components; a WFQ algorithm computation circuit, a tag/time-stamp sort and retrieval circuit, and a high throughput shared buffer. The circuit targets the support of emerging wireline and wireless network nodes that focus on Service Level Agreements (SLA's) and Quality of Experience.     

A new discrete transmission line model for passive model orderreduction and macromodeling of high-speed interconnections

A new, computationally efficient, discrete model is presented for passive model order reduction of high-speed interconnections. The proposed discrete model is based on the use of the theory of compact finite differences for the development of the discrete approximation to the transmission line equations that govern wave propagation on the interconnections. Thus result in a discrete model that utilizes only a few unknowns per wavelength and yet provides highly accurate waveform resolution. In addition to improved computational efficiency, the generated discrete model is passive, and compatible with the passive reduced-order interconnect modeling algorithm (PRIMA). Thus, it is suitable for the development of passive reduced-order models of interconnection networks of high complexity. Numerical experiments from the simulation and model order reduction of coupled interconnections are used to illustrate the validity and efficiency of the proposed model     

High-accurate computation of wave propagation in complex waveguides for microchip optical interconnections

In two-dimensional optical waveguides with the varied refractive index along the transverse direction, the wave propagation is accurately computed. The asymptotic mode solutions are taken as initial guesses for Rayleigh Quotation iteration. More accurate solutions by considering leaky modes and both wave propagation in the transverse electric (TE) and transverse magnetic (TM) are obtained. Numerical simulations demonstrate that the results from Rayleigh Quotation iteration with the asymptotic solutions are better than ones from Lapack solver when the norms of modes are large enough. Thus, it is the best way to combine the two solvers together. This method is particularly useful to the design and simulation of micro waveguide with core layer refractive index fluctuation.