Novel monolithic optical receivers in 0.8 micrometers standard MOS technology with NRZ data rates in excess of 531 Mb/s at 850 nm are introduced. At 638 nm the data rates are above 622 Mb/s. The innovative integrated double photodiode allowing this high speed without any modification in a CMOS process is described and results for their quantum efficiency and for their transient response are presented. Furthermore the improvement of the sensitivity of the photodiodes and the OEICs by an antireflection coating is discussed. The circuit topology of the OEICs is described. A transimpedance input stage with an active MOS feedback resistor is implemented. Analytical calculations for the bandwidth and for the effective transimpedance of the amplifiers as well as measured results are presented. Finally, an OEIC in a 1.0 micrometers CMOS technology with a polysilicon resistor instead of a MOS feedback resistor and implementing a pin photodiode achieves a data rate of 1 Gb/s for 638 nm. The sensitivity of this OEIC is improved by 9 dB m compared to that of a published OEIC.

A demonstration of hybrid integration of a 1 X 12 metal- semiconductor-metal (MSM) photo-detector array and polyimide channel waveguides via 45 degree(s) total-internal-reflection micro-coupler is reported here. MSM photo-detector array was fabricated on a semi-insulated GaAs wafer. The two-layer polyimide waveguide array was constructed using Ultradel 9120D for core and Ultradel 9020 for lower cladding layer. The photodetectors are integrated to operate in the conventional vertical illumination mode. We have measured the external quantum-efficiency and 3 dB bandwidth of the integrated MSM photo-detectors to be 0.4 A/W and 2.648 GHz, respectively. The aggregate 3 dB bandwidth of the 12-channel integrated system is 32 GHz.

We report a generic method to construct 3D wavelength routers by adapting a novel design for multi-optical wavelength interconnects (MOWI's). Optical wavelength- selective (WS) interconnections are realized by resorting to layered diffractive phase elements. Besides, we simultaneously carry out several other integrated operations on the incident beams according to their wavelengths. We demonstrate an 4 X 4 inline 3D WS optical crossconnect and a 1D 1 X 8 WS perfect shuffler. The devices are well feasible for mass production by using current standard microelectronics technologies. It is plausible that the proposed WS MOWI scenario will find critical applications in module-to-module and board-to-board optical interconnect systems, as well as in other devices for short-link multi- wavelength networks that would benefit from function integration.

The PMD-technology opens up a wide area of solutions for all detector problems where a high phase accuracy is required. One important application of the new Photonic Mixer Device is optical distance measurement based on the time-of-flight principle. The high integration of PMD smart-pixels in an array based on CMOS-technology means a breakthrough in optical 3D-imaging. Additionally to the inherent mixing feature, even more functionality could be integrated according to the well-known active pixel sensor concept.

The design of some communication systems requires the implementation of time delays within the system. As a result, the system design often becomes more complex and/or the system size is increased when the delay is accomplished by digitizing the RF analog signal or using RF transmission lines cut to the length necessary to accomplish the time delay. This time delay can be accomplished with a variety of optics technologies, which could be readily fabricated and integrated into the communication system without significant impacts on the system design. This paper describes three different potential applications of optics designs, which could be implemented to accomplish the time delay requirements associated with communication system.

We describe a new method for broadcasting signals in an optical backplane bus system based on guided-wave interconnects. By introducing a distributor (an active coupler) in the center board, which consists of a receiver, a doubly multiplexed hologram, and a transmitter, all signals coming from boards are collected in a receiver in the distributor and then are rebroadcasted from transmitter in the distributor to all boards. This method shows a variety of advantages such as equalized fan-out powers, increased the interconnect distance, reduced number of holograms, and easier system assembly. We show the design concepts for a centralized optical backplane, and resulting experimental performance advantages over previously- developed free-space and guided-wave optical backplane bus systems.

A high-bandwidth, free-space integrated optoelectronic interconnect system was built for high-density, parallel data transmission and processing. Substrate-emitting 980 nm vertical-cavity surface-emitting laser (VCSEL) arrays and photodetector arrays, both driven by complimentary metal- oxide-semiconductor (CMOS) circuitry, were employed as a transmitter and receiver. We designed, fabricated, hybridized, and packaged the VCSEL transmitter and photoreceiver arrays. Data rates above 1 Gbs for each channel on the VCSEL/CMOS emitter and 500 MHz for each channel on photoreceiver were measured, respectively. We integrated the optical interconnects using free-space optical alignment and demonstrated serial and parallel transmissions of digital data and video images.

We demonstrate high packing density true-time-delay lines using 3D integrated polymer waveguides. Although several relatively new delay-line approaches based on the use of optical fibers offer excellent performance, they require complicated, bulky, and costly packaging. The demonstrated 3D integrated polymer waveguide delay lines will overcome these shortcomings. Packaging is greatly simplified, and the layered integration of the waveguides offers greater compactness.

We present a novel design for optoelectronic-multi-chip- modules based small segments of rigid imaging fiber bundles. These packages have very small (chip-scale) geometries and support bandwidth and latency comparable to on-chip interconnections.

The proposed 3D lightwave circuits (LWCs) with square-shaped intersection geometry are transformed into 3D LWCs of another geometry (rectangles) with a lower number of layers (n₁) by grid embeddings (GEs). In the limit, 2-layer sandwich architectures are obtained by GEs and by space filling curves, respectively. The purpose of applying GEs is to control the dilation (number of edges between two modes of the rectangle which are adjacent in the square-shaped grid) of the binary topology, related to the 3D LWCs and introduced by embeddings into cubes.

Large-scale computer and data-communication systems have reached a bottleneck in performance in recent years due to the limitations of electronic interconnections for data transfer. One potential solution is based on the use of optoelectronic device arrays for free space optical interconnects. In this paper, we present the design and implementation of a 16 X 16 3D distributed optoelectronic crossbar switch.

We demonstrate a setup with 10 optically interconnected chips,k which can perform a distributed radix-2-butterfly calculation for fast Fourier transformation. The setup consists of a motherboard, five multi-chip-modules (MCMs, with processor/transceiver chips and laser/detector chips), four plug-on-top optics modules that provide the bi- directional optical links between the MCMs, and external control electronics. The design of the optics and optomechanics satisfies numerous real-world constraints, such as compact size (< 1 inch thick), suitability for mass-production, suitability for large arrays (up to 10³ parallel channels), compatibility with standard electronics fabrication and packaging technology, and potential for active misalignment compensation by integrating MEMS technology.

We present a high-bandwidth parallel optical link that operates from CMOS chip to CMOS chip, for applications like chip-to-chip, board-to-board and rack-to-rack interconnects. The optical channel is an oversampling imaging fiber bundle of 1.9 mm diameter. The light sources are a 10 X 10 2D array of high-speed, high-efficiency light-emitting diodes, design for flip-chip mounting onto CMOS driver circuits. Detectors and receivers are integrated together in standard CMOS, as an array of 10 X 10 detector/receiver cells.

Image fiber is an important candidate for 2D parallel optical transmission medium for high-speed and high-density optical interconnects. To realize one-to-many image fiber link for optical bus systems, image fiber coupler which branches image fibers is a key device. Image fibers can be branched simply by using beam-splitters (BS's) and lenses which image one end to the other ends of the image fibers. However, the compact and reliable assembly and ease of optical alignment are required. We have developed a novel image fiber coupler using graded index rod lenses, a miniature cube BS, and V-grooves.

We design and implement a system demonstrator based on vertical-cavity surface-emitting lasers, polymeric hologram grating couplers, and metal-semiconductor-metal photodetectors. As a preliminary experiment, we show the feasibility of board-to-board level substrate-guided wave optoelectronic interconnections in the real electrical system. First, we introduce a new architecture--centralized optical backplane--for board-to-board level interconnections. Second, the optoelectronic data channel is constructed compatible with standard PECL and capable of operating a 1.25 Gbps. Finally, it is employed to replace the conventional electrical data channel in a microprocessor system. We describe the performance of the entire system and discuss the future application of our centralized optical backplane in other electrical systems.

We report the design and formation of a high-performance polymer waveguide array with 45-degree micro-mirror couplers for achieving fully-embedded board-level optoelectronic interconnects. We have used Si CMOS process compatible polymer as the fabrication material, which is relatively easy to process and has low propagation loss at 850 nm wavelength. 45-degree total interior reflection micro-mirror couplers fabricated within the channel waveguides provide surface-normal light coupling between the waveguide and the optoelectronic devices, thus forming a fully-embedded 3D optoelectronic interconnect. We have demonstrated a hybrid optoelectronic integrated system of GaAs MSM photodetector array and polymer channel waveguide array with 45-degree micro-mirror couplers, showing an aggregate 3 dB bandwidth of 32 GHz.

In this paper, we demonstrate a thin-film polymeric waveguide beam deflector using a new device concept, an electrode of prism-array pattern on top of a three-layer planar waveguide. The three-layer planar waveguide was composed of UV15 as the top cladding layer, a polyimide as the core layer, and SiO₂ as the bottom cladding layer on a silicon substrate. A gold layer was deposited on the top of the waveguide by e-beam deposition and then patterned into a prism-array as the heating electrode by photolithography.

A narrowband optical amplifier but bandwidth-expandable is demonstrated. Future amplifier-based sub-system can be constructed to perform non-linear dispersion compensation and to flatten gain while the input level changes in a DWDM optical networking system. The amplifier is designed to inherently provide two networking features: narrowband accessibility and bandwidth scalability. For a 100 GHz- spaced DWDM application, each amplifier can process 4 channels in an array structure. The crosstalk between each amplifier is better than 58 dB. The capability to flatten gain and improve the transient response is experimentally demonstrated.

Extremely broadband emission is obtained from superluminescent diodes/semiconductor laser amplifiers with nonidentical quantum wells made of InGaAsP/InP materials. Two opposite sequences of nonidentical multiple quantum wells (MQWs), consisting of three In_0.67Ga_0.33As_0.72P_0.28 quantum wells and two In_0.53Ga_0.47As QWs, are designed, fabricated, and measured. Nonuniform carrier distribution inside MQWs is further verified experimentally.

We present mushroom-type TWEAM, which has improved velocity mismatch, with optimized impedance match compared to conventional modulators by reducing the distance between signal and ground metal line. In this paper, the layer structure of mushroom-type TW MQW EAM is designed for the operation of 1.55 um and optical index of active layer is designed to be 3.6. Also, we simulate an 1.55 um InGaAs/InGaAsP traveling-wave multiple quantum well electro- absorption modulator using 3D Finite Difference Time Domain method. Also, we investigate microwave characteristics in detail.

A 2D true-time delay (TTD) module for phased-array antennas (PAAs) is presented, which is based on substrate-guided wave elements. We use substrates of different thickness to get time delays as small as needed. We also use coated wedge to substitute input hologram element to decrease losses up to 60%. Also, a step is designed to obtain 0 degree steering which is impossible for packaging in the previous design. We designed special delay steps to avoid the side lobe caused by the limitation of the distance between adjacent antenna elements. The simulation results were given for the far- field radiation pattern of PAA controlled by this module. Experimental results of the TTD module are shown in both 2D and 3D images.

As processor speeds enter the Gigahertz regime, the disparity between processing time and memory access time plays an increasingly important role in the overall limitation of processor performance. Furthermore, as the components continue to shrink in size, the limitations in interconnect density and bandwidth serve to exacerbate communication bottlenecks. To address these issues, we propose a 3D architecture based on through-wafer vertical optical interconnects. Our system is monolithically fabricated on a single host substrate and preserves the VLSI-scale of integration by using meso-scopic diffractive optical elements for beam fan-out and signal distribution at the chip level.

The microjet printing method is being used to fabricate microlenses for free-space optical interconnects in telecommunication devices. This low-cost, data-driven process, based on `drop-on-demand' inkjet technology, involves the dispensing and placing of precisely sized micro-droplets of 100%-solids, UV-curing optical epoxy pre- polymers at elevated temperatures onto optical substrates and components. The method offers significant cost advantages in the fabrication of components such as precision lens arrays and fiber ribbon collimators for use in certain configurations of DWDM multiplexing- demultiplexing devices and optical switches.

In this paper we have developed a model of an all optical router based on the terahertz optical asymmetric demultiplexer (TOAD). The model architecture is based on a system which has as its input on OTDM packet containing header and payload information. The model simulates extraction of header information from the data stream using one TOAD, which is subsequently used to make a routing decision. The payload information is routed through a second TOAD according to the information contained in the header.