Abstract: An adaptive, reduced-complexity soft-output maximum-likelihood detector that is operable to process data by adaptively selecting a processing scheme based on a determination of signal quality. The signal quality is derived as a function of the noise, the modulation format, the channel (the communication environment), the transmit signal power and the receive signal power. If the signal quality is low, the signal is processed using a maximum likelihood detector. If, however, the signal quality is high, a simpler sub-optimal detector is used. By estimating the signal quality and choosing an appropriate detection method, the present invention ensures accurate detection of incoming data signals in a MIMO communication system while maintaining the highest possible processing speed.
Abstract: The present invention is directed to a telephone equipped with multiple microphones that provides improved performance during operation of the telephone in a speaker-phone mode. For example, the multiple microphones can be used to improve voice activity detection, which in turn, can improve echo cancellation. In addition, the multiple microphones can be configured as an adaptive microphone array and used to reduce the effects of (i) room reverberation, when a near-end user is speaking, and/or (ii) acoustic echo, when a far-end user is speaking.
Abstract: A Radio Frequency (RF) circuit includes an active portion and a tuned portion. The active portion receives an RF signal and that operates upon the RF signal. The tuned portion couples to the active portion and includes a first inductor, a second inductor, and an NMOS transistor. The NMOS transistor has a gate, a drain, a source, and a body. The drain and source operably couple/decouple the second inductor to/from the first inductor. The drain and source are biased at respective operating voltages (greater than ground and less than a first voltage supply) by the active portion and the tuning portion. The body couples to a first voltage supply, via a control circuit or switch in some cases. Applying a second voltage supply (that is greater than the first voltage supply) to the gate turns ON the NMOS transistor.
Abstract: A power amplifier power amplifier includes a transconductance stage and a cascode stage. The transconductance stage that is operable to receive an input voltage signal and to produce an output current signal. The transconductance stage includes a first Metal Oxide Silicon (MOS) transistor having a first gate oxide thickness and a first channel length. The cascode stage communicatively couples to the transconductance stage and is operable to receive the output current signal and to produce an output voltage signal based thereupon. The cascode stage includes a second MOS transistor having a second gate oxide thickness and a second channel length.
Abstract: A method and apparatus for searching an electronically stored table of information including a plurality of table entries and facilitating high speed searching of a table to provide a longest matching entry. The table searching method uses at least one memory unit having a table of information including a plurality of data entries. The table of information has a plurality of search keys associated with the plurality of data entries and the plurality of search keys form a tree structure based on a prefix length for each of the search keys. The plurality of search keys are expanded such that each of the plurality of search keys has two lowest level search keys associated therewith that cover a lowest level of the tree structure. A binary search of the lowest level search keys is performed based on a search value to determine a longest prefix match. A data entry of the plurality of data entries is output based on said longest prefix match.
Abstract: A double-conversion tuner receives an RF signal having a number of channels and down-converts a selected channel from the plurality of channels. The double-conversion tuner includes a first mixer configured to up-convert the RF signal to a first IF signal using a first local oscillator signal. A first local oscillator includes a delta-sigma fractional-N phase lock loop to produce the first local oscillator signal. The delta-sigma fractional-N phase lock loop is configured to perform fine-tuning of the first local oscillator signal and to have a wide tuning range sufficient to cover the number of channels. A bandpass filter is configured to select a subset of channels from said first IF signal. A second mixer is configured to down-convert the subset of channels to a second IF signal using a second local oscillator signal. A second local oscillator generates the second local oscillator signal.
Abstract: A printed bandpass filter is mounted on a precision substrate to eliminate the need for post-fabrication tuning. The filter input is capacitively coupled to a series of quarter wavelength resonators and the filter output. The quarter wavelength resonators are printed as spirals to reduce filter size. The resonators define the bandpass characteristics of the filter. The filter also weakly couples the input signal to the filter output in a manner to cancel the signal image. Mechanical clips mitigate thermal stress on solder connections when the precision substrate is mounted on a second printed circuit board.
Type:
Grant
Filed:
November 23, 2004
Date of Patent:
July 4, 2006
Assignee:
Broadcom Corporation
Inventors:
Ramon A. Gomez, Lawrence M. Burns, Sung-Hsien Chang, Carl W. Pobanz
Abstract: The present invention is directed to a system and method of sending packets between ports on trunked network switches. A trunk connection is provided between first and second switches, with the trunk connection including at least two of the plurality of ports from the first switch being connected to at least two of the plurality of ports of the second switch. A packet is received at an ingress submodule of the first switch, and a lookup is performed on one of a source address and a destination address of the packet based upon a lookup table provided in the ingress submodule. A rules tag bit is then compared to a rules table defining which trunk port of the trunk bundle will be used for communication. The rules tag determines which address bits will be used to identify a trunk port for communication.
Abstract: A method and system, compatible with low-voltage CMOS technology, for controlling the charging of a battery. The method includes monitoring a battery voltage with respect to a threshold voltage. The method further includes coupling a charging control logic supply to ground, generating an active low first control signal, inverting the active low first control signal, and charging the battery at a first rate when the battery voltage is below the threshold voltage. The method further includes coupling the charging control logic supply to the battery voltage, generating an active high second control signal, and charging the battery at a second rate when the battery voltage exceeds the threshold voltage. The first charging rate is slower than the second charging rate. The method further includes supplying battery power to a charger line when the battery voltage exceeds the charger voltage, and suppressing a leakage current.
Abstract: A display engine of a video and graphics system includes one or more processing elements and receives graphics from a memory. The graphics data define multiple graphics layers, and the processing elements process two or more graphics layers in parallel to generate blended graphics. Alpha values may be used while blending graphics. The processing elements may be integrated on an integrated circuit chip with an input for receiving the graphics data and other video and graphics components. The display engine may also include a graphics controller for receiving two or more graphics layers in parallel, for arranging the graphics layers in an order suitable for parallel processing, and for providing the arranged graphics layers to the processing elements. The blended graphics may be blended with HDTV video or SDTV video, which may be extracted from compressed data streams such as an MPEG Transport stream.
Type:
Grant
Filed:
January 20, 2005
Date of Patent:
July 4, 2006
Assignee:
Broadcom Corporation
Inventors:
Alexander G. MacInnis, Chengfuh Jeffrey Tang, Xiaodong Xie
Abstract: A circuit is provided for reducing mismatches between the outputs of successive pairs of cells in an analog to digital converter. A voltage input means is coupled to a first input terminal of each cell to introduce and an input voltage. A reference voltage means is coupled to a second input terminal of each cell to introduce progressive fractions of a reference voltage. A low impedance means is coupled between corresponding first output terminals and coupled between corresponding second output terminals in successive cells, to draw load-bearing currents to the successive cells, affecting the relative voltages and thereby reducing the effects of cell mismatches on these output terminals. Lastly, a high impedance means is coupled to the each of the first output terminals and to each of the second output terminals in successive cells.
Abstract: A method and apparatus for dynamically managing power in a wireless interface device while maintaining an acceptable bit error rate. The wireless interface device includes a wireless interface unit, a bit error detection unit that monitors the bit error rate a data stream in the wireless interface unit, a processing unit, and a power management unit. The power management unit operates in a conjunction with the processing unit and the bit error detection unit to monitor the bit error rate and to dynamically adjust the voltage levels in the wireless interface unit to ensure that the bit error rate remains in an acceptable range.
Abstract: Methods and systems for increasing an amplifier circuit's Q factor are disclosed herein. The method may comprise coupling a first LC tank to a source of a single switching transistor and coupling a second LC tank to a drain of the single switching transistor. A gate of the single switching transistor may be controlled by an amplifier core coupled to the first LC tank and the second LC tank. A resistance of the first LC tank and the second LC tank may be decreased by about one half, which increases the Q factor by about two. The gate of the single switching transistor may be controlled by a control signal generator coupled to the amplifier core. The first LC tank and/or the second LC tank may be tuned to a frequency of about 3.4 GHz to 4 GHz. The single switching transistor may comprise an NMOS transistor.
Abstract: A system and method is provided for interleaving data in a communications device. The system includes a memory for storing symbols of a data block, a read module and a write module, each of which is coupled to the memory. The system also includes a interleaving logic module coupled to the read and write modules. The interleaving logic module determines an interleaving sequence comprising a sequence of memory addresses. Each memory address is then communicated sequentially to the read and write modules. When the read module receives the address, the read module reads the stored data symbol. When the write module receives the address, the write module writes a symbol from a next data block to the vacated address. The interleaving logic module repeats these steps until every symbol of the stored block has been read and every symbol of the next data block has been written to memory.
Abstract: Provided are a method and system for reducing glitch in a switch circuit. A system includes a current-steering switch circuit including a main differential pair switch coupled to a first tail current having a first current value. Also included is an auxiliary differential pair switch connected to the main differential pair switch. The auxiliary differential pair switch is coupled to a second tail current and configured to substantially reduce a feed-through current associated with the main differential pair switch.
Abstract: To adjust transmit power of a DSL modem, an estimated electrical length of a loop is first determined when a 1st DSL modem at a 1st location transmits a plurality of signals to a 2nd DSL modem at a 2nd location. Each of the signals transmitted by the 1st DSL modem have a known frequency and are transmitted at a known power level. Upon receiving the signals, the 2nd DSL modem determines the received power level for each of the signals. The 2nd DSL modem then estimates an electrical length of a loop between the 1st and 2nd DSL modems for each of the signals received. The 2nd DSL modem then processes a plurality of estimated electrical lengths in accordance with a function to produce a determined electrical length. Having determined the electrical length, the 2nd DSL modem adjusts its transmit power accordingly.
Abstract: A method of presenting a selected portion of graphical images within a wireless terminal that includes receiving an original graphical image, having a source resolution, to be presented on a display screen. The native pixel resolution of the display screen may differ from that of the original graphical image. A complex decimation pattern when applied to the selected portion of the image allows the selected portion of the original graphical image to be resized from the source resolution to a decimated resolution operable to be displayed within the native resolution of the display screen. Decimated pixels need not be further processed to reduce the processing requirements imposed on the system processor. Operations normally associated with the decimated portions of the image may be offloaded from the processing module to improve system efficiency.
Abstract: Methods and apparatus for improving the current matching within current mirror circuits in applications such as low voltage integrated circuits. Embodiments of the present invention attempt to maintain the proper current ratio between reference and output supplies by adjusting the reference output of the current mirror. An existing reference voltage on the output side of the mirror can be used or a reference voltage can be created to be used for the voltage regulation of the reference side of the current mirror.
Abstract: A differential amplifier is configured in a cascode configuration that includes input transistors that are connected to corresponding cascode transistors. The gates of the cascode transistors are tied together to form a common bias for the cascode devices. The input transistors of the differential amplifier receive a differential input signal that is amplified and outputted to an output circuit that is connected to the cascode transistors. The cascode devices require a bias voltage for proper operation. Preferably, the bias voltage puts the cascode devices into the saturation region. The gates of cascode devices are coupled together and connected to a bias terminal. In embodiments of the invention, the bias terminal is connected to another terminal of the chip to provide the bias for the cascode devices. This can include the input and output nodes if they have a well-defined and relatively fixed voltage.
Abstract: A multi-port transceiver includes a transmitter and receiver for each port. The invention is a test method and apparatus for testing individual components in the transmit and receive paths. Specifically, the invention includes a method of testing the full range of a programmable gain amplifier (PGA) and an analog to digital converter (ADC) in the receive path of each port. This is accomplished by connecting the transmitter of one port directly to the receiver of a second port, and varying the amplitude of the transmitter over a range of gain settings of the PGA while examining if the dynamic range of the receiver has been exceeded.