Electrooptical Channel Waveguide Matrix Switch Using Total Internal Reflection*

As product complexity continues to increase, test engineers are challenged with improving test coverage on a variety of devices while keeping costs at a minimum. When finished products incorporate hundreds or thousands of devices, developing and maintaining separate test systems for individual components lacks scalability. An approach to this problem is to standardize on a single platform and develop a small number of test systems that can each test tens or hundreds of different devices. Test engineers are increasingly seeing the benefits of developing, deploying, and supporting a reduced number of test systems that use common components. Although many factors such as system assurance and platform longevity influence the long-term success of these systems, the core requirements consist of modular, reconfigurable hardware. A switching matrix is the main hardware component that gives a single test system the flexibility to test a variety of devices. With a switching matrix, you can connect all of your instruments and test points through a series of relays and close relevant relays via software. Switching matrices provide several benefits such as simple connectivity and test repeatability; however, it's usually not practical to have a switching matrix that completely matches the I/O of your instruments. For example, if you have a 1,000 V DMM, a 2 A power supply, and a 1.5 GHz digitizer, you're unlikely to find a COTS switching matrix that maximizes those I/O points. Selecting a switching matrix involves evaluating various trade-offs between density, power, bandwidth, and so on and optimizing the cost of your test system. This paper examines the trade-offs you should evaluate when choosing a switching matrix along with considerations for reducing cost and size.

This paper proposes new solutions for implementing wideband large switch matrices. These solutions are based on crossbar and L-shaped topologies. This paper introduces a high-performance wideband switch cell to build up scalable NtimesN switch matrices and gives an account of the design, fabrication, and characteristics of the switch cell and a 3times3 crossbar switch matrix. The chosen design procedure is seen to be appropriate since it produces valid measured results. In addition, this paper presents an RF microelectromechanical systems L-shaped switch matrix, which indicates less variation of characteristics for certain types of connectivity. It also demonstrates that for a 4times4 switch matrix, there is a 50% improvement in insertion loss and phase-shift variation.

With the increased commercial availability of high channel count MR coil arrays and the associated higher number of plugs in the patient bed, it has become a common practice to include switch matrices in the receive path of MR systems. These allow the arbitrary routing of a signal from any plug to any receiver in the console. While switch matrices are standard in systems at clinical field strength and have been developed for systems operating up to 4T, they have not yet been implemented at ultra-high field (UHF). Here, we present a switch matrix suitable for operation at UHF. Crossbar switches, which are the most frequently employed forms of a switch matrix, use RF switches to connect horizontal input lines with the desired vertical output line. This leaves transmission line stubs of variable length physically connected with the selected signal path, potentially resulting in elevated signal losses. While this can be tolerated at low frequencies, and only needs partial compensation at intermediate frequencies (4T), a full compensation is required at UHF. In this study, an RF crossbar switch, which uses switchable compensation elements in both horizontal and vertical transmission lines, was implemented for a 9.4T MRI scanner. The prototype developed was evaluated for single channel and multichannel receive performance and benchmarked against a fixed wire connection.

In a generic switching problem, a switching pattern consists of a one-to-one mapping from a set of inputs to a set of outputs (i.e., a permutation). We propose and investigate a wireless switching framework in which a multi-antenna relay is responsible for switching traffic among a set of N stations. We refer to such a relay as a MIMO switch. With beamforming and linear detection, the MIMO switch controls which stations are connected to which stations. Each beamforming matrix realizes a permutation pattern among the stations. We refer to the corresponding permutation matrix as a switch matrix. By scheduling a set of different switch matrices, full connectivity among the stations can be established. In this paper, we focus on “fair switching” in which equal amounts of traffic are to be delivered for all the N(N - 1) ordered pairs of stations. In particular, we investigate how the system throughput can be maximized. In general, for large N the number of possible switch matrices N! is huge, making the scheduling problem combinatorially challenging. We show that for N = 4 and 5, only a subset of N - 1 of the N! switch matrices need to be considered in the scheduling problem to achieve good throughput. We conjecture that this will be the case for large N as well. This conjecture, if valid, implies that for practical purposes, fair-switching scheduling is not an intractable problem.

This paper investigates a novel optical multicast scheme without excess optical splitting loss to be realized using an active vertical coupler (AVC)-based optical switch matrix, due to the optical gain available in the AVC switch cells. Theoretical analysis and computer simulation have been carried out regarding the operational principle and switching characteristics of this scheme. Computer simulation further suggests that the amplified spontaneous emission (ASE)-limited scalability of the scheme can be as high as more than 40. Experiments using an existing 4 /spl times/ 4 switch-matrix device successfully demonstrated the predicted multicast switching operation. Measured switching characteristics and the optical signal-to-noise ratio (OSNR) values are in good agreement with simulation predictions. High switched-signal Q-factor values of 1-to-2 and 1-to-3 multicasting configurations indicate that the scale of the multicasting scheme can be further increased significantly.

SOI (Silicon on Insulator) based photonic devices has attracted more and more attention in the recent years. Integration of SOI optical switch matrix with isolating grooves, total internal reflection (TIR) mirrors and spot size converter (SSC) was studied. A folding re-arrangeable non-blocking 4times4 optical switch matrix and a blocking 16times16 matrix with TIR mirrors and SSC were fabricated on SOI wafer. The performances, including extinction ratio and the crosstalk, are better than before. The insertion loss and the polarization dependent loss (PDL) at 1.55mum increase slightly with longer device length, more bend and intersecting waveguides. The insertion losses decrease 2-3 dB when anti-reflection films are added in the ends of the devices. The rise and fall times of the devices are 2.1mus and 2.3mus, respectively

A folding rearrangeable nonblocking 4×4 optical matrix switch was designed and fabricated on silicon-on-insulator wafer. To compress chip size, switch elements (SEs) were interconnected by total internal reflection (TIR) mirrors instead of conventional S-bends. For obtaining smooth interfaces, potassium hydroxide anisotropic chemical etching of silicon was utilized to make the matrix switch for the first time. The device has a compact size of 20×1.6 mm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> and a fast response of 7.5 μs. The power consumption of each 2×2 SE and the average excess loss per mirror were 145 mW and -1.1 dB, respectively. Low path dependence of /spl plusmn/0.7 dB in total excess loss was obtained because of the symmetry of propagation paths in this novel matrix switch.

A scalable pulsewidth control approach with 7-bit resolution for thermal control of optical ring resonator switch matrices in 0.16- $\mu\text{m}$ Bipolar-CMOS-DMOS (BCD) with a cell size of $79\ast 105 \mu\text{m}^2$ is introduced. One Gray counter at up to 1.2-GHz clock and pulsewidth modulation generation in each matrix cell leads to an electrical power consumption of 330 mW for all electronic control circuits of an optical 1000-switch-node matrix, resulting in a reduction of 11.1% of the power needed by a constant-voltage control approach.

Integrated optics switch matrices are expected to play an important role in optical space switching systems due to their ability to route information practically irrespective of its bandwidth and coding format (” code and frequency transparency”), in marked contrast to electronic switching networks. The integrated optics switches also offer the advantage of integration, as opposed to mechanical optical switches. Accordingly a number of integrated optics matrices have been reported1,2,3. However , all of these have been polarization sensitive, requiring interfacing to non-standard polarization holding fibers, and this can be considered a major factor in complicating systems applications. Ref4describes a polarization independent 1×16 switch in the tree structure employed below, however with strict fabrication tolerance on the switches. One reason for staying with polarization sensitive matrices has been the difficulty to combine integratability (reasonably low voltage -length product) with high yield for low crosstalk switch points. Here we report a matrix based on a switch point which combines these features by electronic tuning of its switching characteristics, relaxing fabrication tolerances5. The switch points are further arranged in a tree structure6,7which combines the capability for communicative as well as distributive switching with relaxed requirements on the crosstalk of each switching point. This is achieved at the expense of the number of switches: The 4×4 matrix reported in this paper (Fig 1) comprises 24 switches as opposed to the 16 switches of a crossbar structure, but the number of cascaded levels is only 4, as opposed to 7 in the crossbar case.

Single-mode integrated-optic directional couplers on electro-optic substrates can be used to realize photonic space-division multiplexers/switches with low cross-talk level and switching voltage. This paper describes briefly the methods which have been used by the present authors to design 2x2 and 4x4 switch matrices oir LiNbO3 substrates. A computer-aided analytical model has been developed to obtain the two-dimensional distribution of Ti-concentration and refractive index profiles for arbitrary waveguide geometry, Ti-film thickness and processing conditions. The matrix method which is a simple analytical tool for layered structures has been employed to study the wave propagation through Ti indiffused LiNb03 waveguides and directional couplers. The data for refractive index profile required as the input for running the computer program implementing the matrix method were obtained from the above computer-aided analysis. The computed results of Ti-concentration, refractive index profiles and the critical coupling length of directional couplers were validated with available experimental data. Due attention has also been paid to reduce the radiation loss from waveguide bends and to incorporate the polarization dependence in the design of directional couplers. The electrode structure and its placement on Ti:LiNb03 waveguides have been optimized for a 5V switching voltage. The effects of buffer layer and I/O fiber-to-waveguide coupling have also been considered to estimate the total loss. Necessary data generation in appropriate format for electron beam masks for Ti-waveguides and electrode patterns has been completed, and the fabrication of masks is now in progress. A simplistic design of 8x8 switch matrix has also been included.

Packet-switching characteristics are optimized across an integrated 4 x 4 optical cross-point switch matrix based on active vertical coupler switch cells. Optical gain is demonstrated across the entire matrix with a <3-dB difference between the shortest and longest switching paths.

Construction of large scale switch matrix by interconnecting integrated optical switch chips with EDFAs

Large-scale optical switching matrix is the core technology of automatically switched optical network. With the significant increase of the matrix size, the study showed that the nonlinear factors, such as the correlation of the wavelength, phase and polarization between the optical switches have major impacts for the system performance. Because the full-size optical simulation is not available now, the research will be a great significance that we establish an overall model with not only considering the performance of optical switch unit, but also reflecting the structural characteristics of the network.In this paper, we calculated the transmission field of 2×2 Mach –Zehnder(MZI) silica on silicon thermo optic(TO) switch with different structure by using the finite difference beam propagation method(BPM). On the basis of optimization design, we constructed and simulated the large-scale 16×16 MZI silicon-based SiO2 Silica on silicon thermo switching matrix. The results are: In TE Mode, for Crossbar and banyan structures, insertion losses are 13.35 dB and 4.3638 dB; crosstalk losses are 27.125~29.0173 dB and 42.2318-43.7159 dB; extinction ratios are 29.57 dB and 26.79 dB respectively.

Current electronic broadband switch matrices employ discrete semiconductor devices or VLSI chips. Isolation between ON- and OFF-states degrades and crosstalk between input and output lines of these electronic switch matrices increases at high operation frequencies. These will limit electronic switch matrices to low-frequency or low-bit-rate operations. In addition, optical-to-electrical (electrical-to-optical) conversions at the input (output) ends are required for these switch matrices to be used in a fiber environment. One way to achieve broadband, low-crosstalk, and high-isolation switching and to eliminate E/O and O/E conversions is to use optical switches. This paper describes an optical broadband switching architecture using optical/optoelectronic logic gates. Input and output signals are optical. No E/O or O/E conversions are needed for this architecture. The optical/optoelectronic logic gates that can be employed in this architecture are reviewed. Performance of this architecture is evaluated by calculating the crosstalk, isolation, insertion loss, switch matrix size, drive power, and switching speed. Technologies needed for hybrid and monolithic implementations of this architecture are also identified.

The optical switch matrix is the one of the key elements in constructing optical switching systems. Many types of optical switch matrix architectures have been proposed1 to improve their characteristics and to construct devices with larger switch dimensions. However, since the substrate for fabricating the optical switch matrix is limited in size, the switching voltage of these structures was still relatively high. In addition, crosstalk characteristics degraded with increasing the number of output and input ports. The architecture wich can be obtained3 by simplifying the structure of tree-structure,2 wich we propose is one of the candidates for solving these problems. We report on the basic characteristics of this structure and some measured results from devices fabricated on LiNbO3 by indiffusion of Ti.