Abstract: A method and system for designing and implementing a finite impulse response (FIR) filter to create a plurality of output signals, each output signal having the same frequency but at a different phase shift from the other output(s), is described. Values are determined for the resistors, or other elements having impedance values, in a FIR filter having a plurality of outputs, such that each output has the same frequency response but a different phase than the other output(s). This is accomplished by the inclusion of a phase factor in the time domain calculation of the resistor values that does not change the response in the frequency domain. The phase shift is constant and independent of the frequency of the output signal.

Abstract: FIR filters for compensating for fixed pattern jitter, and methods of constructing the same, are disclosed. In one embodiment, a FIR filter filters a signal having a desired frequency component, with the coefficients of the FIR filter selected so that the filter is the equivalent of two combined FIR filters, one having the desired frequency at the filter's peak output frequency, and a second in which the signal is delayed by a time equal to half of a period of a different frequency which is desired to be removed from the output signal. In another embodiment, a FIR filter includes a delay line with a total delay longer than the period of the jitter. A signal is passed down the delay line, the number of signal edges that have occurred as the signal passes each delay element in the counted. Drivers corresponding to the delay elements in which a number of signal edges occur at the desired frequency during the period of fixed pattern jitter activate impedance elements attached to those delay elements.

Abstract: A delay circuit in which the delay is independent of variations in the power supply which powers the logic gates of the delay circuit is disclosed. By separating the CMOS transistors that form each logic gate by additional CMOS bias transistors which are biased at a controlled voltage, variations in the gate delay of the inverter transistors due to variations in the power supply voltage for the inverter transistors may be minimized. In one embodiment, the constant bias voltage may be provided by a constant current source comprising a series of amplifiers each having a gain significantly less than one connected to a triple cascode.

Abstract: A method and system is disclosed for simultaneously down-converting multiple selected signals, such as RF signals, into adjacent ranges in an intermediate frequency band so that the total resulting bandwidth, and thus the sampling rate required to digitize the signal, is minimized. A first signal is down-converted into a range starting at a lowest selected frequency in the IF band. The next signal is down-converted, into a range higher than, but near or adjacent to, the down-converted range of the first signal, and so on. A guard band may be left between the signals if desired. In this way, the selected signals occupy the minimum bandwidth required. When the selection of signals to be down-converted is changed, the frequency ranges are dynamically adjusted so that the signals being down-converted always occupy the lowest ranges of the IF band.

Abstract: A method and system for generating and matching complex series and/or parallel combinations of nominally identical initial elements to achieve an arbitrary compound value is disclosed. A recursive algorithm successively adds one or more similar nominal two-terminal elements to generate a series and/or parallel compound combination of nominal elements, the compound combination having a desired impedance. The compound value, and thus the ratio between two compound values, can be determined to almost any desired degree of accuracy, with potential errors greatly reduced from those typical in the construction of individual elements of different values. Since the initial elements are nominally identical, the compound value, and the ratio between values, depends primarily upon the connections of the initial elements, rather than their geometry, and thus remain virtually constant regardless of variations in the manufacturing process.

Abstract: A system and method is disclosed for selecting between two electronic signals, one of high quality, such as music audio, and the other of low quality, such as telephone call audio, in a smart phone, tablet or other device. In one embodiment, when the low quality signal is to be used this is accomplished by disabling the amplifier output to disconnect the high quality audio signal from the output port, rather than by means of a switch between the amplifier and the output port as in the prior art. This eliminates degradation of the signal due to the switch when the high quality signal is to be used. The amplifier typically has an associated feedback resistor network, and this may also be disconnected by means of a switch when the low quality signal is to be used, thus preventing distortion of the low quality signal due to the feedback network being a parallel load to the output port.

Abstract: A method and system is disclosed in which the phase detector in a phase-locked loop is able to run at the fastest speed appropriate for a reference signal. A frequency offset is added to the output frequency of the phase-locked loop, to alter the frequency fed to the frequency divider which would receive the output frequency in a conventional PLL to an intermediate frequency. The frequency offset is selected so that the ratio of the intermediate frequency to the reference frequency is a simple fraction, and preferably an integer, i.e., the intermediate frequency is a multiple of the reference frequency. In cases where the relationship between the output frequency and the reference frequency is largely relatively prime, the phase detector is thus able to receive signals at the frequency of the reference signal and operate at the fastest speed appropriate for the reference signal.

Abstract: Channel select filter circuits are described. One circuit implements a multiplying element and digital-to-analog converter as a differential current mode device. Another circuit implementing a multiplying element and digital-to-analog converter with weighted addition, deferred after multiplication of the digital-to-analog converter and multiplier combination. In one such circuit, substantially equal current source magnitudes are in different columns of the circuit. Another such circuit, with substantially equal current source magnitudes, uses non-radix2. Another such circuit, with substantially equal current source magnitudes, has partial segmentation. Another circuit implements a multiplying element and digital-to-analog converter, with partial segmentation, scrambling bit allocation for elements. One such circuit scrambles bit allocation on equally weighted segments, as described herein.

Abstract: A circuit that provides a rotating coefficient FIR filter with all necessary coefficient sets present at the same time, without the need for delay elements or devices providing for adjustable impedances is described. An input signal is sampled in round robin fashion by a plurality of sample and hold devices. The outputs of the sample and hold devices are connected to sets of impedance devices. Each set of impedance devices implements the coefficients of the desired frequency response of the filter. The impedance devices in each set are connected to the sample and hold devices in a different order from each other set, so that each set of impedance devices will produce the desired frequency response when a different one of the sampling circuits contains a new sample of the input signal. Switches connect the sets of impedance devices to an output, only one switch being closed at a time to provide the output signal.

Abstract: A system and method for filtering an analog signal with a finite impulse response (FIR) filter that does not require analog delay elements are disclosed. An analog signal is pulse-width encoded, and the pulse-width encoded signal passed to a delay line comprising unclocked delay elements, such as logic gates, rather than clocked delay elements such as are used in conventional FIR filters. The propagation of the input signal is thus due only to the delay inherent in each gate, and occurs based upon when a signal reaches the gate rather than being caused by a clock signal. As with a conventional FIR filter, weighting elements having impedance are used to weigh the output of each delay element, and the resulting outputs summed to obtain a filtered output signal. For certain signals, such a circuit and method provides a simpler way of filtering than conventional filters.

Abstract: The present application describes an apparatus and method for reducing distortion in a class-D amplifier. The power output section of the amplifier is driven by an adjusted PWM signal, rather than by a PWM signal created directly from the input analog signal. A reference output, designed to closely track the input analog signal, is compared to the amplifier output. The resulting difference is an error signal which is inverted and summed with a second analog signal corresponding to the directly created PWM signal and changes the timing of the voltage transitions of the second analog signal. The changed voltage transitions are used to create the adjusted PWM signal. The inversion of the error signal causes negative feedback which results in the adjustment of the PWM signal being in a direction which reduces the error signal and thus the distortion of the amplifier.

Abstract: A virtual Weaver architecture filter is implemented using a sampling mixer that successively processes samples of the input signal in round-robin fashion and provides a sum of the samples as multiplied by coefficients emulating quadrature sinusoidal waveforms. A virtual rather than actual local oscillator is reliably implemented without mismatch. Filtering between the Weaver mixers is eliminated in favor of filtering at the sampling input and effective time division multiplexing is achieved by selecting between resistor combinations that implement different scaling coefficients, resulting in an efficient analog implementation of a virtual Weaver architecture.

Abstract: A dynamically selectable resistor network is provided in a star configuration for producing a weighted sum of input values, without attenuation from near zero contributions. Each branch of the star connected network comprises sets of impedance components, preferably resistors, that are actively selectable to produce permutated combinations of effective weighting values. The resistors code digital control bits and the outputs of sets of resistors in respective branches that correspond to the least significant control bits provide their outputs to the summing output node independently of the sets of resistors corresponding to control bits of other significance.

Abstract: An opportunity is apparent to develop alternative circuitry. Simplified circuitry without artifacts tied to the clock that drives a digital frequency generator (DFG) is useful in a variety of tunable electronic devices. The present invention relates to digital frequency generation. In particular, it relates to a method and apparatus for the digital generation of a pulse stream having a desired frequency relative to a reference clock signal and the ratio of two integers. The method applies generally to integers whose ratio is not an integer. The DFG as a device can be integrated onto a simple chip, without need for an off-chip filter.

Abstract: A system and method for filtering an analog signal with a finite impulse response (FIR) filter that does not require analog delay elements are disclosed. An analog signal is pulse-width encoded, and the pulse-width encoded signal passed to a delay line comprising unclocked delay elements, such as logic gates, rather than clocked delay elements such as are used in conventional FIR filters. The propagation of the input signal is thus due only to the delay inherent in each gate, and occurs based upon when a signal reaches the gate rather than being caused by a clock signal. As with a conventional FIR filter, weighting elements having impedance are used to weigh the output of each delay element, and the resulting outputs summed to obtain a filtered output signal. For certain signals, such a circuit and method provides a simpler way of filtering than conventional filters.

Abstract: A method and system is disclosed in which the phase detector in a phase-locked loop is able to run at the fastest speed appropriate for a reference signal. A frequency offset is added, to the output frequency of the phase-locked loop, to alter the frequency fed to the frequency divider which would receive the output frequency in a conventional PLL to an intermediate frequency. The frequency offset is selected so that the ratio of the intermediate frequency to the reference frequency is a simple fraction, and preferably an integer, i.e., the intermediate frequency is a multiple of the reference frequency. In cases where the relationship between the output frequency and the reference frequency is largely relatively prime, the phase detector is thus able to receive signals at the frequency of the reference signal and operate at the fastest speed appropriate for the reference signal.

Abstract: The present application describes an apparatus and method for improving the performance of ?? modulators functioning as ADCs. In one embodiment, the ?? modulator comprises a plurality of quantizers operating in a round-robin fashion, rather than the single quantizer of the prior art. The use of multiple quantizers allows the ?? modulator to appear to be functioning at a significantly higher rate than a single quantizer allows. In another embodiment, a second-order ?? modulator contains a plurality of control loops, rather than the single control loop of the prior art. The use of multiple control loops allows the ?? modulator to have multiple points of maximum signal-to-noise ratio rather than a single such point as in prior art ?? modulators.

Abstract: A method and system is disclosed for designing a radio for down-converting RF signals to IF signals by sampling the signals in a round-robin sampling circuit and multiplying the samples by coefficients that are changed at a fixed rate equal to the rate of operation of each of the sampling circuits. The circuit is able to down-convert multiple channels simultaneously to adjacent positions in the IF band, while rejecting unwanted image signals. The method and system avoids the difficulty and cost of directly digitizing the RF signal, allowing each component to operate at a greatly reduced speed. The coefficients are selected to provide the desired transfer function while keeping the output signal centered at a desired frequency.

Abstract: A method and system is disclosed for simultaneously down-converting multiple selected signals, such as RF signals, into adjacent ranges in an intermediate frequency band so that the total resulting bandwidth, and thus the sampling rate required to digitize the signal, is minimized. A first signal is down-converted into a range starting at a lowest selected frequency in the IF band. The next signal is down-converted, into a range higher than, but near or adjacent to, the down-converted range of the first signal, and so on. A guard band may be left between the signals if desired. In this way, the selected signals occupy the minimum bandwidth required. When the selection of signals to be down-converted is changed, the frequency ranges are dynamically adjusted so that the signals being down-converted always occupy the lowest ranges of the IF band.

Abstract: A circuit that provides a rotating coefficient FIR filter with all necessary coefficient sets present at the same time, without the need for delay elements, devices providing for adjustable impedances, or buffers is described. An input signal is sampled in a round robin fashion by a plurality of switches and capacitors. The capacitors are connected directly to sets of impedance devices. Each set of impedance devices implements the coefficients of the desired frequency response of the filter, adjusted to compensate for the decay of samples in the capacitors between samples. The impedance devices in each set are connected to the capacitors in a different order from each other set, so that each set of impedance devices will produce the desired frequency response when a different one of the capacitor contains a new sample of the input signal. Switches connect the sets of impedance devices to an output and a virtual ground, only one switch being connected to the output at a time to provide the output signal.