摘要:
A wide band, wide operating range, general purpose digital phase locked loop (PLL) runs in the digital domain except for the associated Time Digitizer (T2D) and Digitally-Controlled-Oscillator (DCO). By calibrating the T2D and DCO on the fly, a constant PLL loop BW is achieved by using the calibrated Phase Frequency Detection (PFD) and DCO information to normalize the control loop correction regardless of the input clock frequency, power supply voltage, processing and temperature variations. PLL loop BW is completely decoupled from the operating conditions and semiconductor device variation. This means that the PLL loop BW can be chosen very aggressively to reject the noise, thus achieving a low jitter, high performance PLL. Furthermore, since this PLL can reliably operate over a wide operating range, it is a one-design-fits-all general purpose PLL.
摘要:
A high PSRR, low power semiconductor digitally controlled oscillator (DCO) architecture employs only one simple current steering D/A converter directly on top of a multi-stage current controlled oscillator. The architecture provides a good building block for many circuit applications, e.g., all digital phase lock loops, direct modulation transmitters for wireless devices, and the like.
摘要:
A symmetric glitch free clock multiplexing circuit allows the input clock to a digital or analog processing unit to be switched from one frequency to the other at any moment during the operation, assuming the respective clocks themselves are stable. There exist no restrictions on the clocks or the switch control signal to be synchronous in any fashion. This circuit guarantees a glitch free output and also prevents short cycling of the output clock. Since all the related clocks and switch control signal are asynchronous, this circuit further eliminates meta-stability problems. Its symmetrical architecture allows the circuit to function with the output clock being switched from slow clock to fast clock and vise versa. More importantly, the complete switch over only takes two cycles of the targeted clock in the best case once the active clock is turned off, when switching from slow to fast clock; and four target clock cycles in the worst case once the active clock is turned off, when switching from fast to slow clock.
摘要:
A flash analog-to-digital converter having precise differential voltage interpolation without the use of silicide-blocked resistors. A reference conversion voltage output portion converts an analog input voltage on the basis of a plurality of reference voltages into a plurality of reference conversion voltages. An intermediate voltage generating portion includes a predetermined number of non-linear resistance units respectively provided between one voltage and the other voltage in pairs of a predetermined number of the plurality of reference conversion voltages to generate a plurality of intermediate voltages by resistance division using the predetermined number of non-linear resistance units. In addition, the intermediate voltage generating portion generates a plurality of conversion voltages. A digital data output portion outputs the digital output voltage on the basis of the plurality of conversion voltages using double interpolation. Each of the predetermined number of non-linear resistance units includes a first input terminal connected to the one voltage, a second input terminal connected to the other voltage, and a plurality of non-linear resistor elements having the same resistance value connected in series between the first and second input terminals. The plurality of intermediate voltages includes at least part of voltages obtained from one end of each of the plurality of non-linear resistor elements.
摘要:
Embodiments of the invention provide a system and method for chip to chip communications in electronic circuits. A router or switch receives data packets at input port ASICs. A routing table on the input port ASIC or on a routing ASIC is used to identify a destination port ASIC based upon header information in the data packet. The data packet is transmitted from the input port ASIC to the destination port ASIC using millimeter wave signals that are transmitted across a waveguide or a wireless interface.
摘要:
A method for determining the position of a target is provided. Several emitted pulses of terahertz radiations are emitted from a phased array (which has several transceivers) in consecutive cycles (typically). These emitted pulses are generally configured to be reflected by a target so as to be received by the phased array within a scan range (which includes a digitization window with several sampling periods). Output signals from each of the transceivers are then combined to generate a combined signal for each cycle. The combined signal in each sampling period within the digitization window for emitted pulses is averaged to generate an averaged signal for each sampling period within the digitization window. These averaged signals are then digitized.
摘要:
In some developing interconnect technologies, such as chip-to-chip optical interconnect or metal waveguide interconnects, misalignment can be a serious issue. Here, however, a interconnect that uses an on-chip directional antenna (which operates in the sub-millimeter range) to form a radio frequency (RF) interconnect through a dielectric waveguide is provided. This system allows for misalignment while providing the increased communication bandwidth.
摘要:
At very high frequencies, generally above 100 GHz, the performance of traditional radio frequency (RF) circuitry begins to significantly limit performance. An example is the hybrid coupler, which can have a relatively narrow 90° bandwidth in these frequency ranges. Here, however, a branch-line hybrid coupler (which has been integrated into a quadrature downconversion mixer) has been modified. Namely, an adjustable impedance network has been coupled to isolation port (which has traditionally been terminated) to substantially increase the tuning range and expand the bandwidth of the quadrature mixer within these very high frequency ranges.
摘要:
One aspect of the invention provides a semiconductor device that includes a microchip having an outermost surface. First and second bond pads are located on the microchip and near the outermost surface. A first UBM contact is located on the outermost surface and between the first and second bond pads. The first UBM contact is offset from the first bond pad. A second UBM contact is located on the outermost surface and between the first and second bond pads. The second UBM contact is offset from the second bond pad, and a capacitor supported by the microchip is located between the first and second UBM contacts.
摘要:
An architecture for testing a plurality of circuits on an integrated circuit is described. The architecture includes a TAP Linking Module located between test pins on the integrated circuit and 1149.1 Test Access Ports (TAP) of the plurality of circuits to be tested. The TAP Linking Module operates in response to 1149.1 scan operations from a tester connected to the test pins to selectively switch between 1149.1 TAPs to enable test access between the tester and plurality of circuits. The TAP Linking Module's 1149.1 TAP switching operation is based upon augmenting 1149.1 instruction patterns to affix an additional bit or bits of information which is used by the TAP Linking Module for performing the TAP switching operation.