Abstract: The Floating Diffusion charge detection system has incorporated a signal feedback directly into the charge-detection node. The feedback is coupled to the node from the output of the standard buffer amplifier A1 through a feedback amplifier A3, switching transistors S2 and S3, and capacitors Cf and Ch. The feedback significantly reduces kTC noise, has good linearity, and improves DR.

Abstract: A cost-effective filter consuming low power and occupying minimal space. The filter may be used in a ADSL modem (or CPE) to filter the signal components other than the ADSL signals. A high pass filter first filters the low frequency components to attenuate (or remove) lower frequency components such as that caused by ADSL transmit echo signals and that used for voice transmission. The high pass filter may be modified by adding a few resistors to limit the voltages of the high frequency signals also. The output of the high pass filter is amplified and passed through a low pass filter to filter the high frequency components (HPNA included). Due to earlier filtering operation of the high pass filter, the signal can be amplified substantially before being sent to the low pass filter. The implementation of the low pass filter is simplified due to the prior amplification.

Abstract: A method of forming a transistor (70) in a semiconductor active area (78). The method forms a gate structure (G2) in a fixed relationship to the semiconductor active area thereby defining a first source/drain region (R1) adjacent a first gate structure sidewall and a second source/drain region (R2) adjacent a second sidewall gate structure. The method also forms a lightly doped diffused region (801) formed in the first source/drain region and extending under the gate structure, wherein the lightly doped diffused region comprises a varying resistance in a direction parallel to the gate structure.

Abstract: The present invention can improve and/or modify data retention lifetimes for ferroelectric devices by baking them prior to or during packaging. A ferroelectric device is programmed to a particular state and then baked for a selected period of time at a selected temperature. This pre-baking or imprinting causes the device to be imprinted or have a preference for the particular state and reduces loss of signal margin over time, thereby at least partially preserving data retention capabilities.

Abstract: A method for measuring a test differential voltage across a first output and a second output of a transmitter integrated circuit device, the test differential voltage corresponding to a voltage across the first output and second output appearing while the device is providing an output while being subjected to a voltages applied across a resistor network connected to the differential outputs, the resistor network including first resistor having a value of Ra connected between the first output and a first voltage, a second resistor having a value of Rb connected between the second output and a second voltage, and a third resistor having a value of Rc connected between the first output and the second output.

Abstract: This invention describes a unique high-speed implementation for overflow detection logic to be used in high performance shifter functions. The overflow logic makes use of parallelism in combining shift value decoding and mask generation logic with the logic necessary to propagate data. Designs for both 16-bit and 32-bit shifters are presented and performance improvement of the new designs over conventional overflow detection circuits is demonstrated.

Abstract: A bus bridge is defined to provide an interface between two AHB buses. These busses normally have separate requirements but both must provide high performance. The first is for transfer of data from CPU to memory and peripherals. The second is to support the transfer of a large amount of data by a single peripheral to local memory or other local peripherals. The AHB-to-HTB bus bridge provides a means for the interfacing these two separate AHB buses allowing communication between them and securing data integrity. The bus bridge of this invention is defined to be an AHB memory bus slave but a high performance data transfer bus master.

Abstract: The present invention is directed to a method of forming an FeRAM integrated circuit, which includes forming a multi-layer hard mask. The multi-layer hard mask comprises a hard masking layer overlying an etch stop layer. The etch stop layer is substantially more selective than the overlying masking layer with respect to an etch employed to remove the bottom electrode diffusion barrier layer. Therefore during an etch of the capacitor stack, an etch of the bottom electrode diffusion barrier layer results in a substantially complete removal of the hard masking layer. However, due to the substantial selectivity (e.g., 10:1 or more) of the etch stop layer with respect to the overlying masking layer, the etch stop layer completely protects the underlying top electrode, thereby preventing exposure thereof.

Abstract: A data processing system (e.g., microprocessor 30) for packing register data while storing it to memory and unpacking data read from memory while loading it into registers using single processor instructions. The system comprises a memory (42) and a central processing unit core (44) with at least one register file (76). The core is responsive to a load instruction (e.g., LDW_BH[U] instruction 184) to retrieve at least one data word from memory and parse the data word over selected parts of at least two data registers in the register file. The core is responsive to a store instruction (e.g., STBH_W instruction 198) to concatenate data from selected parts of at least two data registers into at least one data word and save the data word to memory. The number of data registers is greater than the number of data words parsed into or concatenated from the data registers. Both memory storage space and central processor unit resources are utilized efficiently when working with packed data.

Abstract: The present invention forms a nitrided dielectric layer without substantial harm to a semiconductor layer on which the dielectric layer is formed. The invention employs a multi-stage process in which dielectric sub-layers are individually nitrided before formation of a next dielectric sub-layer. The net result is a nitrided multi-layered dielectric layer comprised of a plurality of dielectric sub-layers wherein the sub-layers have been individually deposited and incorporated with nitrogen.

Abstract: A method of improving shallow trench isolation (STI) gap fill and moat nitride pull back is provided by after the steps of growing a pad oxide, depositing a nitride layer on the pad oxide and the steps of moat patterning, moat etching and moat clean, the steps of growing thermal oxide, deglazing a part of a part of the moat nitride; depositing a thin nitride liner, etching the nitride to form a thin side wall nitride in the STI trench; and performing an oxide Hydroflouric (HF) acid deglazing before STI liner oxidating and depositing oxide to fill the trench.

Abstract: A processor is provided that is a programmable digital signal processor (DSP) with variable instruction length, offering both high code density and easy programming. Architecture and instruction set are optimized for low power consumption and high efficiency execution of DSP algorithms, such as for wireless telephones, as well as pure control tasks. A coefficient data pointer is provided for accessing coefficient data for use in a multiply-accumulate (MAC) unit. Monitoring circuitry determines when the coefficient data pointer is modified (step 1104). When an instruction is executed (step 1102) that requires a coefficient datum from memory in accordance with the coefficient data pointer, a memory access is inhibited (step 1108) if the coefficient data pointer has not been modified since the last time a memory fetch was made in accordance with the coefficient data pointer and the previously fetched coefficient datum is reused.

Abstract: A multiplexer containing multiple cells sharing a common output line. The cells select one of multiple input bits. The output line is first charged to a first logical value (e.g., 0), and one of the cells drives the output line to a second logical value (1) if the corresponding input bit does not equal the first logical value. The remaining cells may not affect the output line. Due to such an implementation, the number of transistors may be reduced.

Abstract: A Virtual Frame Buffer control system and method for cascading several display controllers on one LCD panel. The Virtual Frame Buffer is composed of all the memory in all the controller/memory/source driver chips (in a tiled pattern) for the associated processor to read and write in. The control system also includes hardware clipping controls in each of the controller/memory/source driver chips. The Virtual Frame Buffer and hardware clipping control placement substantially reduces programming problems associated with prior art solutions for cascading LCD controller/memory/source driver devices.

Abstract: A circuit 10 is provided that comprises a source resistance transistor 12 connected to a common node 14. A word line driver circuit 18 receives current if it is the word line driver selected from the VDD supply voltage through the source resistance transistor 12. The gate of source resistance transistor 12 is connected to a bond pad 22 which can be alternatively connected to the VDD supply voltage through a bond pad 24 or to ground potential through a bond pad 26. The effective threshold voltage of a transistor 18 within driver 16 can be adjusted depending upon how the gate of transistor 12 is connected. In this manner, a circuit can be adjusted to compensate for process variation or to be more optimum for a selected application by adjustment of the effective threshold voltage of selected transistors.

Abstract: An output circuit for improved ESD protection (FIG. 2) comprising a pMOS pull-up output transistor connected between a signal (I/O) pad 220 and Vdd power supply 240, the pull-up transistor located in a n-well 203 and having at least one gate 210, the gate connected to internal circuitry 230. A dummy pMOS transistor connected in parallel with the pull-up transistor, the dummy transistor also located in the n-well 203, whereby both the pull-up transistor and the dummy transistor participate in protection against an ESD event. The dummy transistor having at least one gate 251, this gate connected through a resistor 260 to the Vdd power supply 240. The n-well 203 connected to the Vdd power supply 240.

Abstract: A differential circuit with linearity correction loop includes a main differential amplifier 30, and a correction amplifier 20 having inputs coupled to the outputs of the main differential amplifier 30 through feedback paths. The output signals from the correction amplifier 20 are combined with the inputs to the main amplifier 30 such that a negative feedback loop is formed around the differential circuit. This feedback loop provides stability with only a minor power increase.

Abstract: The voltage outputs of a charge coupled device (CCD) are examined to determine the hot pixels. A black pixel is determined to be a hot pixel if the voltage level associated with the black pixel exceeds the voltage level of an adjacent (e.g., previous) pixel by a threshold. If the present black pixel is determined to be a hot pixel, a previous black pixel is substituted for a present black pixel in the computation of the offset. However, if the first black pixel is determined to be a hot pixel, the second black pixel is used in lieu of the first black pixel. The offset is iteratively adjusted by an amount proportionate to an error determined based on the black pixels. The adjustment may be clipped by a threshold to avoid bands in the image.

Abstract: A tool and method is described to decide partial wafer sizes to process multiple random sizes of wafers in pick and place equipment for wafermap operation. The tool identifies the wafer and gets wafermap data. The position of one or more cutters is displayed. The position of the cutters relative to the wafer is displayed. The tool generates and displaying the results of the type of dies in each partial that would result from a cut according to said displayed position of the cutters.

Abstract: A 64-state binary convolutional code is disclosed for a high-speed physical layer (PHY) of a communication network. The proposed code provides improved performance in terms of signal to noise ratio (SNR) and multi-path rejection than previously known codes. The proposed system, which includes binary convolutional codes with scrambling in a packet-based system, is referred to herein as “packet binary convolutional coding” (PBCC). The substantial increase in performance that may be achieved by PBCC makes it an ideal solution for high performance forward error correction (FEC) in a high-speed PHY.