There is currently considerable research interest in all-optical signal processing for telecommunication applications.
In this paper it is shown that a range of all-optical signal processing functions may be carried out using passive
waveguides and phase-shifting elements as basic building blocks. Two types of multimode interference (MMI) photonic
circuits, based on silicon-on-insulator (SOI) technology, are discussed in detail. In the first type, phase-shifting elements
are added to the ports of MMI couplers. In the second type, the waveguide structure of the MMI coupler itself is
modified to achieve a range of signal processing functions. Designs are verified using 2D and 3D simulations.
Microring resonators are promising candidates for photonic signal processing applications. However, almost all
resonators that have been reported so far use directional couplers or 2×2 multimode interference (MMI) couplers as the
coupling element between the ring and the bus waveguides. In this paper, instead of using 2×2 couplers, novel structures
for microring resonators based on 3×3 MMI couplers are proposed. The characteristics of the device are derived using
the modal propagation method. The device parameters are optimized by using numerical methods. Optical switches and
filters using Silicon on Insulator (SOI) then have been designed and analyzed. This device can become a new basic
component for further applications in optical signal processing. The paper concludes with some further examples of
photonic signal processing circuits based on MMI couplers.
There is a demand for reliable photonic switches and routers, of moderate switching or routing order, that can be
integrated easily into photonic integrated circuits. For some time, low order MMI devices have been employed to
provide a variety of circuit routing and switching functions. However, good performance of higher order MMI devices
has been more difficult to achieve. This paper reviews the design criteria and design primitives for MMI structures and
highlights their advantages and shortcomings.
The most recent impetus for the convergence of photonics and silicon integrated circuit technology has been the looming communications bottleneck associated with chip-to-chip and on-chip high-speed data transfer. Whilst there have been significant separate improvements in the materials and technologies for both integrated optics and integrated electronics, there is now a real commercial interest in putting these pieces together to achieve functional circuits that take advantage of both technologies. It is the purpose of this paper to review the recent developments in both microelectronics and photonics that are causing these fields to merge in the area of on-chip and off-chip data links.
Couplers based on the multimode interference (MMI) principle form the basis for a variety of integrated optics devices that can achieve routing, switching and other telecommunication functions. Whilst small order MMI couplers show good performance, the imaging properties deteriorate as the order increases. In order to better understand the performance capabilities of these devices there is interest in devising improved analytical models of MMI structures. This paper reviews various approaches to modeling integrated optics devices based on MMI couplers and attempts to highlight the advantages and the pitfalls associated with a number of models, and suggests some avenues for improving these models.
Integrated optical switches based on multimode interference (MMI) structures in a generalized Mach-Zehnder configuration can be used to achieve space-division switching. This paper examines the fundamental limitations of these optical switches and assesses their suitability for use as optical cross-connects.
The use of optical waveguides as the data link for off-chip and on-chip interconnects is attracting considerable interest. This paper examines the requirements and some possible solutions suitable for FPGA applications.
The increasing complexity of silicon integrated circuits are making global interconnects more difficult to achieve. It is predicted that, even at the high clock rates envisaged in future generations of chips, the propagation delays across the whole chip will amount to multiples of the clock period. Clock distribution and data transfer via optical waveguides offer a solution to this global interconnect problem. This paper examines the optical waveguide structures that would be suitable for these applications.
The use of multimode interference (MMI) structures within Mach-Zehnder configurations have showed some promise for switching and power splitting functions in integrated optics circuits. In this paper a matrix model is presented that can aid in assessing the switching and power-splitting capabilities of these structures. The paper also examines some of the factors that may limit the performance of these devices when high port counts are attempted.
VLSI/ULSI and the evolutions being driven by the International Technology Roadmap for Semiconductors (ITRS) are once again presenting severe challenges to the metal interconnect. Clock skew and other timing delays are becoming application critical design factors. The RC induced delays as well as parasitics (due to the trace density) are causing severe limitations to designs. Unfortunately these issues are very difficult to deal with using conventional computer aided design tools although efforts are being made, notably via DARPA funded programmes. We shall review techniques (and design elements) for on-chip optical communications. Through this we will present a new proposition for optical interconnects integrated upon otherwise conventional CMOS devices. We believe that the illustrated methodologies can be developed to provide very effective optical functionality appropriate to alleviating high-speed communications and timing issues.
Photonic space switches are useful elements in optical fibre communication links. This paper describes the design of 2Nx2N photonic switches based on the generalized Mach-Zehnder interferometer. Although these switches are limited in dimension, they have the desirable attributes of low loss, good loss uniformity, small size, robustness and ease of fabrication and integration.
Compact integrated optical circuits for signal routing and signal processing are currently the subjects of much active research. Multimode interference (MMI) couplers are widely used as splitters and combiners since they possess the desirable attributes of small size, low excess loss, well- defined slitting, dimensional tolerance and ease of fabrication. Recently there has been renewed interest in employing MMI devices within Mach-Zehnder structures to achieve splitting and switching functions. However, the extent of the switching capabilities achieved so far has been quite limited. This paper highlights how the switching capabilities of these Mach-Zehnder switches can be extended and presents design techniques for this type of photonic switch.
The metal-semiconductor-metal (MSM) photodetector has the desirable attributes of large bandwidth and ease of fabrication. The lateral structure of the MSM detector allows easy incorporation into optoelectronic integrated circuits. In this paper, a new, simplified, broad-band model of the MSM detector is presented. Practical MSM detectors often exhibit an undesirable low frequency gain that is bias-dependent. It is shown in this paper that the trapping process involved in producing this gain can be modeled, in part, by including a passive equivalent circuit within the circuit model of the detector. The components of the equivalent circuit are related to the trap lifetime and the probability of a hole becoming trapped. The nonlinear effect resulting from the saturation of the electron velocity is modeled as a nonlinear current source whose magnitude is a function of bias voltage. The distributed nature of the interdigitated structure is modeled by a single set of coupled transmission lines. The trapping model and associated current sources are incorporated at the ends of the transmission lines to produce the overall model for the compete detector. Tests have shown that the proposed model provides good agreement with previously published experimental results.
The Metal-Semiconductor-Metal (MSM) photodetector has attracted a great deal of interest because of its potential for high speed operation and low fabrication cost. In this paper, the parameters affecting the capacitance, sensitivity and bandwidth of the interdigitated MSM photodiode are examined. Models are presented for the detector capacitance, gain and transmission line effects. The proposed circuit models have been verified by experimental results.
A scanning near-field optical microscope (SNOM) has been used to measures directly the evanescent field distribution surrounding an optical fiber taper. The SNOM interaction with the fiber taper is explain for the first time using a wave optics approach. Result of evanescent field measurements with varying wavelengths and surrounding refractive index media are presented. Experimental results are compared with theoretical data produced by the Finite Difference Beam Propagation Method.
The metal-semiconductor (MSM) photodetector attracts a great deal of interest as a result of its high bandwidth and low fabrication costs. In this paper a broad-band circuit model for the interdigitated MSM photodetector is presented. The circuit model can be used for both design and simulation purposes. The circuit model can also take into account nonlinear effects so that the practical behavior of the photodetector can be more faithfully represented.
A feature of metal-semiconductor-metal (MSM) photodetectors is their low capacitance. In this paper, a new design formula is presented for the capacitance per unit area of an MSM photodetector. A comprehensive circuit model is also described which incorporates the optoelectronic conversion within a larger model which in turn accommodates the transmission line effects of the interdigitated structure. These models can be used to optimize the design of a receiver employing an MSM photodetector.
At low bit rates most current image coding techniques will generate a smooth image where some of the important details of the original image are lost. For example, the compression of an image by JPEG down to a bit rate of 0.1 to 0.2 bit/pel usually results in a significant loss of detail and an annoying blocking effect. The basic idea in this paper is that in applications where the detail information in a low bit rate coded image is of more concern than the gray scale information, it is necessary to encode the former and disregard the latter. Based on the logarithmic image processing model, we have recently proposed a novel algorithm to generate a binary image (called the sketch image) from a gray scale image. We have shown that the sketch image retains much more information of the original image than those images generated by other image binarization techniques such as simple thresholding, ordered dither and error diffusion. In this paper, we further study the sketch image algorithm and propose a simple data compression scheme which incorporates a decimation and interpolation algorithm and the JBIG algorithm. Experimental results have shown that at low bit rate (about 0.1 bit/pel) the proposed algorithm preserves the fine details of the original image at the expense of loss of grey scale information.