摘要:
This disclosure concerns low insertion loss optical circulators. In one example, the optical circulator has four ports and includes a polarization dividing and combining element that is positioned adjacent the first and fourth ports and is adapted to divide a beam of light into two beams of light of orthogonal polarizations. The polarization dividing and combining element is also adapted to combine two beams of light of orthogonal polarizations into one beam of light. The optical circulator also includes a Faraday rotator positioned near the second port, and a Faraday rotator positioned near the third port. The Faraday rotator rotates beams of light before or after the pass through the polarization dividing and combining elements.
摘要:
An optical interleaver for use in a range of telecommunications applications including optical multiplexers/demultiplexers and optical routers. The optical device includes an optical processing loop which allows multi-stage performance characteristics to be achieved with a single physical filtration stage. Optical processing on the first leg and second legs of the loop is asymmetrical thereby improving the integrity of the optical signals by effecting complementary chromatic dispersion on the first and second legs. A fundamental filter cell within the interleaver filters optical signals propagating on each of the two legs of the optical loop which intersects the fundamental filter cell.
摘要:
A package for maintaining alignment of components includes a frame and a beam with the beam attached to the frame and one end of the beam. One or more components are mounted to the beam. The frame and a portion of the beam are separated from each other by a channel which allows portions of the beam to flex substantially independently of the frame when a force is applied to the frame. Thus, the effects resulting from forces applied to the frame are not experienced by the beam, or are at least attenuated, so that the alignment of the components mounted on the beam is substantially preserved.
摘要:
A package for maintaining alignment of components includes a frame and a beam with the beam attached to the frame and one end of the beam. One or more components are mounted to the beam. The frame and a portion of the beam are separated from each other by a channel which allows portions of the beam to flex substantially independently of the frame when a force is applied to the frame. Thus, the effects resulting from forces applied to the frame are not experienced by the beam, or are at least attenuated, so that the alignment of the components mounted on the beam is substantially preserved.
摘要:
A method of reducing power consumption in a computing platform is disclosed. An endpoint-device that is coupled to the computing platform includes a first data buffer and a second data buffer. The first data buffer buffers outgoing data to be transmitted to an external device via a first communications medium. The second data buffer buffers incoming data received from the external device via the first communications medium. At least one of the first or second data buffers may selectively communicate with the computing platform, via a second communications medium, during an active window of the other data buffer. The active window may be requested by the first and/or second data buffers based, at least in part, on a system idle signal. For some embodiments, the first communications medium is an Ethernet link. Further, for some embodiments, the second communications medium is a Peripheral Component Interconnect Express (PCIe) link.
摘要:
Techniques are disclosed that allow for power conservation in integrated circuit memories, such as SRAM. The techniques can be embodied in circuitry that allows for floating of bitlines to eliminate or otherwise reduce power leakage associated with precharging bitlines. For instance, the techniques can be embodied in a bitline floating circuit having a single logic gate for qualifying the precharge control signal with a wake signal, so that precharging of the bitline does not occur if the wake signal is not in an active state. The techniques further allow for the elimination or reduction of unnecessary power consumption by the I/O circuitry or the memory array, such as when the memory array is not being accessed or when the array or a portion thereof is permanently disabled for yield recovery.
摘要:
A process of forming a semiconductive capacitor device for a memory circuit includes forming a first capacitor cell recess and a second capacitor cell recess that are spaced apart by a capacitor cell boundary of a first height. The process includes lowering the first height of the capacitor cell boundary to a second height. A common plate capacitor bridges between the first recess and the second recess over the boundary above the second height and below the first height.
摘要:
An array of memory cells is disclosed. The memory cell includes a fuse and at least one transistor. The transistor is used to control the programming or sensing of the fuse. A program voltage is applied to a stack of first and second conductive layers. A first portion of the stack couples the program voltage to a terminal of the transistor in a cell. A second portion of the stack couples the program voltage to a terminal of the transistor in another cell.
摘要:
An apparatus, a method, and a system for fuse cells are disclosed herein. In various embodiments, a fuse cell may include circuitry to adjust a sensing margin. A fuse cell may include first and second fuse cells, and first and second resistance devices. The first resistance device may be configured to adjust a first voltage output from the first fuse cell, and the second resistance device may be configured to adjust a second voltage output from the second fuse cell. The first and second resistance devices may be configured adjust the first and second voltages asymmetrically.