Apparatus and method for a wireless transceiver

Abstract

To provide an apparatus and a method for transceivers.SOLUTION: In a first aspect, an observation receiver is implemented to detect common-mode local oscillator (LO) leakage while the observation receiver performs other functions. In a second aspect, a transceiver is implemented with an LO leakage compensation circuit that compensates a transmitter for LO leakage using a digital filter, such as a Kalman filter, that combines differential LO leakage observations with common-mode LO leakage observations. In a third aspect, a transmitter includes multiple local loopback paths including a first loopback path after a variable gain amplifier (VGA) and a second loopback path before the VGA.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] FIELD OF THE INVENTION Embodiments of the present invention relate to electronic systems, and more particularly to transceivers for radio frequency (RF) communication systems. [Background technology]

[0002] Wireless transceivers can be used in a wide variety of radio frequency (RF) communication systems. For example, transceivers can be included in base stations or mobile devices to transmit and receive signals associated with a wide variety of communication standards, including, for example, cellular and / or wireless local area network (WLAN) standards. Transceivers can also be used in radar systems, instrumentation, industrial electronics, military electronics, laptop computers, digital radios, and / or other electronic devices. Summary of the Invention

[0003] In one aspect, a transceiver integrated circuit (IC) is provided. The transceiver IC includes a first transmitter including at least one mixer and a local oscillator (LO), where the first transmitter is configured to generate a first radio frequency (RF) transmit signal based on controlling the at least one mixer so that the LO provides frequency upconversion. The transceiver IC further includes an observation receiver configured to receive an observation signal for observing at least the first transmitter, where the observation receiver includes an observation data path configured to process the observation signal to detect a transmit power of the first RF transmit signal, and an LO leakage observation circuit configured to generate leakage observation data based on processing data captured from the observation data path. The first transmitter is configured to process the leakage observation data to compensate for LO leakage.

[0004] In some embodiments, the LO leakage observation circuitry does not prevent the observation data path from detecting the transmit power.

[0005] In various embodiments, the observation data path is further configured to process the observation signal to perform digital pre-distortion (DPD), and the LO leakage observation circuitry does not prevent the observation data path from performing DPD.

[0006] In certain embodiments, the first transmitter is further configured to compensate for LO leakage based on local transmit observation data from one or more local observation paths of the transceiver IC. According to some embodiments, the first transmitter is configured to detect differential LO leakage from the local leakage observation data and common-mode LO leakage from the leakage observation data from the observation receiver. According to some embodiments, the first transmitter further includes a variable gain amplifier (VGA) configured to amplify the first RF transmit signal, and the one or more local observation paths include a first local observation path after the VGA and a second local observation path before the VGA. According to various embodiments, the first transmitter comprises a digital filter configured to process both the local leakage observation data and the leakage observation data from the observation receiver. According to some embodiments, the digital filter is a Kalman filter. According to some embodiments, the first transmitter is configured to collect the local leakage observation data and the leakage observation data from the observation receiver in multiple batches. According to various embodiments, the first transmitter is configured to update a value of an LO leakage compensation signal for compensating for LO leakage after receiving the local leakage observation data and a given number of samples of the leakage observation data from the observation receiver.

[0007] In some embodiments, the LO leakage observation circuit includes a digital mixer configured to generate frequency-shifted data based on mixing data captured from the observation data path, and a digital accumulator configured to generate leakage observation data based on accumulating the frequency-shifted data. According to some embodiments, the LO leakage observation circuit further includes a digital oscillator configured to provide a digital clock signal to the digital mixer. According to some embodiments, the digital oscillator is a numerically controlled oscillator (NCO).

[0008] In various embodiments, the transceiver IC further includes a plurality of transmitters, including the first transmitter, and the observation signal is configured to observe one of the plurality of transmitters at a given time. According to some embodiments, the LO leakage observation circuit is configured to separately track leakage observation data for each of the plurality of transmitters. According to some embodiments, the transmitter of the plurality of transmitters selected for observation can be changed without interfering with operation of the LO leakage observation circuit.

[0009] In certain embodiments, the observation data path includes an analog-to-digital converter (ADC) configured to generate a digital observation signal based on the observation signal, and a digital circuit configured to process the digital observation signal.

[0010] In another aspect, a method for compensating for local oscillator (LO) leakage in a transceiver is provided, the method including: generating a first radio frequency (RF) transmit signal using a first transmitter including at least one mixer and a local oscillator (LO) controlling the at least one mixer, generating an observation signal based on a loopback path from the first transmitter through an RF front end to an observation receiver, processing the observation signal to detect a transmit power of the first RF transmit signal using an observation data path of the observation receiver, generating leakage observation data based on processed data captured from the observation data path, and processing the leakage observation data to compensate for LO leakage.

[0011] In various embodiments, the leakage observation data is generated without interfering with the detection of the transmit power.

[0012] In some embodiments, processing the observation signal includes performing digital predistortion (DPD) on the first RF transmit signal, and the leakage observation data is generated without disturbing the DPD.

[0013] In certain embodiments, the method further includes compensating for LO leakage based on local transmit observation data from one or more local observation paths that do not pass through the RF front end. According to some embodiments, the method further includes detecting differential LO leakage from the local leakage observation data and detecting common-mode LO leakage from leakage observation data from the observation receiver. According to some embodiments, the method further includes amplifying the first RF transmit signal using a variable gain amplifier (VGA) at the first transmitter and generating local leakage observation data using a first local observation path after the VGA and a second local observation path before the VGA. According to various embodiments, the method further includes merging the local leakage observation data and the leakage observation data from the observation receiver using a digital filter. According to some embodiments, the digital filter is a Kalman filter. According to some embodiments, the method further includes collecting the local leakage observation data and the leakage observation data from the observation receiver in multiple batches. According to various embodiments, the method further includes updating a value of an LO leakage compensation signal for compensating for LO leakage after receiving the local leakage observation data and a given number of samples of the leakage observation data from the observation receiver.

[0014] In some embodiments, generating the leak observation data includes mixing the captured data from the observation data path using a digital mixer and accumulating the mixed data to generate the leak observation data. According to various embodiments, generating the leak observation data further includes controlling the digital mixer using a digital oscillator. According to some embodiments, the digital oscillator is a numerically controlled oscillator (NCO).

[0015] In some embodiments, the method further includes generating a plurality of RF transmit signals using a plurality of transmitters and observing one of the plurality of transmitters at a given time using an observation receiver. According to various embodiments, the method further includes separately tracking leakage observation data for each of the plurality of transmitters. According to some embodiments, generating the observation signal includes amplifying the plurality of RF transmit signals using a plurality of power amplifiers, sensing a plurality of output powers of the plurality of power amplifiers using a plurality of directional couplers, and multiplexing the plurality of directional couplers to obtain the observation signal.

[0016] In various embodiments, processing the observation signal includes generating a digital observation signal from the observation signal using an analog-to-digital converter (ADC) and processing the digital observation signal using digital circuitry.

[0017] In a particular embodiment, generating the observation signal includes amplifying the first RF transmit signal using a power amplifier and sensing an output power of the power amplifier using a directional coupler.

[0018] In another aspect, a radio frequency (RF) communication system is provided. The RF communication system includes a transceiver including an RF front end, a first transmitter configured to provide a first RF transmit signal to the RF front end, and an observation receiver configured to receive an observation signal via a loopback path from the first transmitter through the RF front end to the observation receiver. The observation receiver includes an observation data path for processing the observation signal to detect a transmit power of the first RF transmit signal, and a local oscillator (LO) leakage observation circuit configured to generate leakage observation data based on processed data captured from the observation data path. The first transmitter processes the leakage observation data to compensate for LO leakage of the first transmitter.

[0019] In some embodiments, the LO leakage observation circuitry does not prevent the observation data path from detecting the transmit power.

[0020] In various embodiments, the observation data path processes the observation signal to perform digital pre-distortion (DPD), and the LO leakage observation circuitry does not prevent the observation data path from performing DPD.

[0021] In certain embodiments, the first transmitter is further configured to compensate for LO leakage based on local transmit observation data from one or more local observation paths of the transceiver. According to some embodiments, the first transmitter is configured to detect differential LO leakage from the local leakage observation data and common-mode LO leakage from the leakage observation data from the observation receiver. According to some embodiments, the first transmitter further includes a variable gain amplifier (VGA) configured to amplify the first RF transmit signal, and the one or more local observation paths include a first local observation path after the VGA and a second local observation path before the VGA. According to various embodiments, the first transmitter comprises a digital filter configured to process both the local leakage observation data and the leakage observation data from the observation receiver. According to some embodiments, the digital filter is a Kalman filter. According to some embodiments, the first transmitter is configured to collect the local leakage observation data and the leakage observation data from the observation receiver in multiple batches. According to various embodiments, the first transmitter is configured to update a value of an LO leakage compensation signal for compensating for LO leakage after receiving the local leakage observation data and a given number of samples of the leakage observation data from the observation receiver.

[0022] In some embodiments, the LO leakage observation circuit includes a digital mixer configured to generate frequency-shifted data based on mixing data captured from the observation data path, and a digital accumulator configured to generate leakage observation data based on accumulating the frequency-shifted data. In some embodiments, the LO leakage observation circuit further includes a digital oscillator configured to provide a digital clock signal to the digital mixer. According to some embodiments, the digital oscillator is a numerically controlled oscillator (NCO).

[0023] In various embodiments, the transceiver includes a plurality of transmitters, including a first transmitter, and the observation signal is configured to observe one of the plurality of transmitters at a given time. According to some embodiments, the LO leakage observation circuit is configured to separately track leakage observation data for each of the plurality of transmitters. According to some embodiments, the transmitter selected for observation among the plurality of transmitters is changeable without interfering with operation of the LO leakage observation circuit. According to some embodiments, the RF front end includes a plurality of power amplifiers configured to amplify a plurality of RF transmit signals, a plurality of directional couplers configured to sense a plurality of output powers of the plurality of power amplifiers, and a multiplexer configured to couple to the plurality of directional couplers and output the observation signal.

[0024] In certain embodiments, the observation data path includes an analog-to-digital converter (ADC) configured to generate a digital observation signal based on the observation signal, and a digital circuit configured to process the digital observation signal.

[0025] In some embodiments, the RF front end includes a power amplifier configured to amplify the first RF transmit signal and a directional coupler configured to generate an observation signal based on sensing an output power of the power amplifier.

[0026] In another aspect, a transceiver includes a transmitter configured to generate a radio frequency (RF) transmit signal, the transmitter configured to generate first observation data based on observing the RF transmit signal on one or more local observation paths within the transmitter. The transceiver further includes an observation receiver configured to generate second observation data based on processing the observation signal. The transceiver further includes a local oscillator (LO) leakage compensation circuit configured to compensate the transmitter for transmitter LO leakage, the LO leakage compensation circuit including a digital filter configured to process the first observation data and the second observation data.

[0027] In some embodiments, the digital filter is a Kalman filter. According to various embodiments, the Kalman filter is configured to estimate a correction for the LO leakage using a plurality of nonlinear equations. According to some embodiments, the Kalman filter is further configured to modify the correction based on a linear update of the Kalman filter using an observation matrix. According to some embodiments, the observation matrix corresponds to a Jacobian matrix. According to various embodiments, a first portion of the plurality of nonlinear equations is a function of first observation data, and a second portion of the plurality of nonlinear equations is a function of second observation data. According to some embodiments, the LO leakage compensation circuit includes a processor, and the Kalman filter is implemented as software running on the processor.

[0028] In particular embodiments, the transmitter includes a digital transmitter circuit configured to process a digital in-phase (I) signal and a digital quadrature-phase (Q) signal, and the digital filter is configured to compensate the transmitter for LO leakage based at least in part on controlling the digital transmitter circuit. According to some embodiments, the digital filter is configured to control a first DC offset of the digital I signal and a second DC offset of the digital Q signal. According to various embodiments, the digital transmitter circuit includes a programmable finite impulse response (PFIR) filter, and the digital filter is configured to control the PFIR. According to some embodiments, the transmitter includes an I path digital-to-analog converter (DAC) coupled to the digital transmitter circuit and configured to generate an analog I signal, a controllable I path filter configured to generate a filtered I signal based on filtering the analog I signal, a Q path DAC coupled to the digital transmitter circuit and configured to generate an analog Q signal, and a controllable Q path filter configured to generate the filtered Q signal based on filtering the analog Q signal. According to some embodiments, the digital filter controls settings of the controllable I path filter and the controllable Q path filter. According to some embodiments, the transceiver further includes an I path mixer configured to receive the filtered I signal and controlled by the LO, and a Q path mixer configured to receive the filtered Q signal and controlled by the LO, and the RF transmit signal is generated based on combining the output of the I path mixer and the output of the Q path mixer.

[0029] In some embodiments, the first observed data is indicative of differential LO leakage of the LO, and the second observed data is indicative of common-mode LO leakage of the LO.

[0030] In some embodiments, the transmitter includes a variable gain amplifier (VGA) configured to amplify the first RF transmit signal, and the one or more local observation paths include a first local observation path after the VGA and a second local observation path before the VGA.

[0031] In another aspect, a method for local oscillator (LO) leakage compensation in a transceiver is provided, the method including generating a radio frequency (RF) transmit signal using a transmitter, generating first observation data based on observing the RF transmit signal on one or more local observation paths within the transmitter, generating second observation data using an external loopback path from the transmitter through a front-end system to an observation receiver, and compensating the transmitter for transmitter LO leakage based on processing the first observation data and the second observation data using a digital filter.

[0032] In particular embodiments, the digital filter is a Kalman filter. According to various embodiments, the method further includes estimating a correction value for the LO leakage using a plurality of nonlinear equations. According to some embodiments, the method further includes modifying the correction value based on a linear update of the Kalman filter using an observation matrix. According to some embodiments, the observation matrix corresponds to a Jacobian matrix. According to various embodiments, a first portion of the plurality of nonlinear equations is a function of first observed data, and a second portion of the plurality of nonlinear equations is a function of second observed data.

[0033] In some embodiments, the method further includes processing the digital in-phase (I) signal and the digital quadrature-phase (Q) signal using a digital transmitter circuit of the transmitter, and compensating the transmitter for LO leakage based at least in part on controlling the digital transmitter circuit using a digital filter. According to some embodiments, controlling the digital transmitter circuit includes adjusting a first DC offset of the digital I signal and adjusting a second DC offset of the digital Q signal. According to some embodiments, the digital transmitter circuit includes a programmable finite impulse response (PFIR) filter, and controlling the digital transmitter circuit includes adjusting the PFIR.

[0034] In various embodiments, the first observed data is indicative of differential LO leakage of the LO, and the second observed data is indicative of common-mode LO leakage of the LO.

[0035] In some embodiments, the method further includes amplifying the RF transmit signal using a variable gain amplifier (VGA), and the one or more local observation paths include a first local observation path after the VGA and a second local observation path before the VGA.

[0036] In another aspect, a radio frequency (RF) communication system is provided. The RF communication system includes an RF front end and a transceiver. The transceiver includes a transmitter configured to provide an RF transmit signal to the RF front end, the transmitter configured to generate first observation data based on observing the RF transmit signal on one or more local observation paths within the transmitter. The transceiver further includes an observation receiver configured to generate second observation data based on processing an observation signal received from an external loopback path from the transmitter through the RF front end to the observation receiver. The transceiver further includes a local oscillator (LO) leakage compensation circuit configured to compensate the transmitter for transmitter LO leakage. The LO leakage compensation circuit includes a digital filter configured to process the first observation data and the second observation data.

[0037] In some embodiments, the digital filter is a Kalman filter. According to some embodiments, the Kalman filter is configured to estimate a correction for the LO leakage using a plurality of nonlinear equations. According to various embodiments, the Kalman filter is configured to modify the correction based on a linear update of the Kalman filter using an observation matrix. According to some embodiments, the observation matrix corresponds to a Jacobian matrix. According to some embodiments, a first portion of the plurality of nonlinear equations is a function of first observation data, and a second portion of the plurality of nonlinear equations is a function of second observation data. According to various embodiments, the LO leakage compensation circuit comprises a processor, and the Kalman filter is implemented as software running on the processor.

[0038] In particular embodiments, the transmitter comprises a digital transmitter circuit configured to process a digital in-phase (I) signal and a digital quadrature-phase (Q) signal, and the digital filter is configured to compensate the transmitter for LO leakage based at least in part on controlling the digital transmitter circuit. According to some embodiments, the digital filter is configured to control a first DC offset of the digital I signal and a second DC offset of the digital Q signal. According to various embodiments, the digital transmitter circuit includes a programmable finite impulse response (PFIR) filter, and the digital filter is configured to control the PFIR. According to some embodiments, the transmitter includes an I path digital-to-analog converter (DAC) coupled to the digital transmitter circuit and configured to generate an analog I signal, a controllable I path filter configured to generate a filtered I signal based on filtering the analog I signal, a Q path DAC coupled to the digital transmitter circuit and configured to generate an analog Q signal, and a controllable Q path filter configured to generate the filtered Q signal based on filtering the analog Q signal. According to some embodiments, the digital filter controls settings of the controllable I path filter and the controllable Q path filter. According to various embodiments, the RF communication system further includes an I path mixer configured to receive the filtered I signal and controlled by the LO, and a Q path mixer configured to receive the filtered Q signal and controlled by the LO, and the RF transmit signal is generated based on combining the output of the I path mixer and the output of the Q path mixer.

[0039] In some embodiments, the first observed data is indicative of differential LO leakage of the LO, and the second observed data is indicative of common-mode LO leakage of the LO.

[0040] In some embodiments, the transmitter includes a variable gain amplifier (VGA) configured to amplify the first RF transmit signal, and the one or more local observation paths include a first local observation path after the VGA and a second local observation path before the VGA.

[0041] In various embodiments, the RF front end includes a power amplifier configured to amplify the RF transmit signal and a directional coupler configured to generate an observation signal based on sensing the output power of the power amplifier.

[0042] In another aspect, a transceiver integrated circuit (IC) includes a plurality of mixers, including a first mixer configured to upconvert an analog in-phase (I) signal and a second mixer configured to upconvert an analog quadrature-phase (Q) signal, wherein an output of the first mixer and an output of the second mixer are combined to generate a radio frequency (RF) transmit signal. The transceiver further includes a variable gain amplifier (VGA) configured to amplify the RF transmit signal to generate an amplified RF transmit signal, and a plurality of local loopback circuits, including a first local loopback circuit coupled to the output of the VGA and a second local loopback circuit coupled to the input of the VGA.

[0043] In various embodiments, the transceiver IC further includes at least one sub-sampling analog-to-digital converter (ADC) configured to process the output of at least one of the multiple local loopback circuits. According to some embodiments, the at least one sub-sampling ADC includes a shared sub-sampling ADC configured to digitize the output of a selected local loopback circuit selected from the first local loopback circuit or the second local loopback circuit. According to some embodiments, the first mixer and the second mixer are implemented as harmonic rejection mixers. According to some embodiments, the first mixer and the second mixer are configured to reject at least third and fifth harmonics.

[0044] In certain embodiments, at least one of the first local loopback circuit or the second local loopback circuit includes an RF harmonic rejection filter. According to some embodiments, the RF harmonic rejection filter is configured to reject at least seventh-order harmonics. According to some embodiments, the transceiver IC further includes a digital transmitter circuit configured to process the plurality of digital transmit samples captured by the at least one sub-sampling ADC. According to some embodiments, the digital transmitter circuit is configured to estimate aliasing effects of the at least one sub-sampling ADC. According to some embodiments, the digital transmitter circuit includes a frequency selection circuit configured to generate a plurality of frequency-shifted digital transmit samples based on shifting the frequency of the plurality of digital transmit samples. According to some embodiments, the digital transmitter circuit further includes an accumulator circuit configured to separately accumulate the plurality of frequency-shifted digital transmit samples and the plurality of non-frequency-shifted digital transmit samples. According to some embodiments, the transceiver IC further includes a local oscillator configured to control the first mixer and the second mixer, and the frequency shift circuit is configured to shift the plurality of digital transmit samples based on a frequency difference between the local oscillator and a sampling rate of the at least one sub-sampling ADC. According to some embodiments, the transceiver IC further includes an observation receiver configured to provide the plurality of digital observation samples to the digital transmitter circuit, and the accumulation circuit is further configured to accumulate the plurality of digital observation samples after frequency shifting by the frequency shift circuit. According to various embodiments, the transceiver IC further includes a local oscillator configured to control the first mixer and the second mixer, and the frequency shift circuit is configured to shift the plurality of digital observation samples based on a frequency of the local oscillator.According to some embodiments, the at least one sub-sampling ADC is configured to sample the outputs of the plurality of loopback circuits at a sampling rate that is less than the carrier frequency of the RF transmit signal.

[0045] In various embodiments, the transceiver IC further includes a digital transmitter circuit configured to generate a digital I signal and a digital Q signal, a first DAC configured to generate an analog I signal based on the digital I signal, and a second DAC configured to generate an analog Q signal based on the digital Q signal. According to some embodiments, the transceiver IC further includes a local oscillator configured to control the first mixer and the second mixer, and the digital transmitter circuit is configured to perform local oscillator leakage compensation based on a plurality of digital transmit samples captured by at least one of the plurality of local loopback circuits. According to some embodiments, the transceiver IC further includes an observation receiver configured to generate a plurality of digital observation samples, and the digital transmitter circuit is further configured to compensate for local oscillator leakage based on the plurality of digital observation samples. According to some embodiments, the digital transmitter circuit is configured to provide quadrature error correction (QEC) based on the plurality of digital transmit samples.

[0046] According to various embodiments, the first local loopback circuit includes a controllable attenuator.

[0047] In some embodiments, at least one of the first local loopback circuit or the second local loopback circuit includes a controllable gain circuit.

[0048] In another aspect, a method of loopback in a transceiver is provided. The method includes generating a radio frequency (RF) signal based on upconverting an analog in-phase (I) signal using a first mixer, upconverting an analog quadrature-phase (Q) signal using a second mixer, and combining an output of the first mixer and an output of the second mixer. The method further includes amplifying the RF transmit signal using a variable gain amplifier (VGA) to generate an amplified RF transmit signal, and providing loopback using multiple local loopback paths, including a first local loopback circuit coupled to an output of the VGA and a second local loopback circuit coupled to an input of the VGA.

[0049] In some embodiments, the method further includes processing an output of at least one of the plurality of local loopback circuits using at least one sub-sampling analog-to-digital converter (ADC). According to various embodiments, the method further includes digitizing an output of a selected local loopback circuit, selected from the first local loopback circuit or the second local loopback circuit, using a shared sub-sampling ADC. According to some embodiments, the first mixer and the second mixer are implemented as harmonic rejection mixers. According to some embodiments, the first mixer and the second mixer are configured to reject at least third and fifth harmonics.

[0050] In certain embodiments, at least one of the first local loopback circuit or the second local loopback circuit includes an RF harmonic rejection filter. According to some embodiments, the RF harmonic rejection filter is configured to reject at least seventh-order harmonics. According to some embodiments, the method further includes processing the plurality of digital transmit samples captured by the at least one sub-sampling ADC using a digital transmitter circuit. According to some embodiments, the method further includes estimating aliasing effects of the at least one sub-sampling ADC using the digital transmitter circuit. According to various embodiments, the method further includes generating a plurality of frequency-shifted digital transmit samples based on shifting the frequency of the plurality of digital transmit samples. According to some embodiments, the method further includes separately storing the plurality of frequency-shifted digital transmit samples and the plurality of non-frequency-shifted digital transmit samples. According to some embodiments, the method further includes controlling the first mixer and the second mixer using a local oscillator, wherein the plurality of frequency-shifted digital transmit samples are generated based on a frequency difference between the local oscillator and a sampling rate of the at least one sub-sampling ADC. According to some embodiments, the method further includes frequency-shifting the plurality of digital observation samples from the observation receiver and accumulating the plurality of digital observation samples after the frequency shift. According to some embodiments, the method further includes controlling the first mixer and the second mixer using a local oscillator, wherein the plurality of digital observation samples are frequency-shifted based on a frequency of the local oscillator. According to some embodiments, the method further includes operating at least one sub-sampling ADC at a sampling rate that is less than a carrier frequency of the RF transmit signal.

[0051] In certain embodiments, the method further includes controlling the first mixer and the second mixer using a local oscillator and compensating for local oscillator leakage based on a plurality of digital transmit samples captured by at least one of the plurality of local loopback circuits. According to some embodiments, the method further includes compensating for local oscillator leakage based on a plurality of digital observation samples from an observation receiver.

[0052] In some embodiments, the method further includes correcting a quadrature error between the analog I signal and the analog Q signal based on a plurality of digital transmit samples captured by at least one of the plurality of local loopback circuits.

[0053] In another aspect, a radio frequency (RF) communication system is provided. The RF communication system includes an RF front end configured to receive an amplified RF transmit signal and a transceiver. The transceiver includes a plurality of mixers, including a first mixer configured to upconvert an analog in-phase (I) signal and a second mixer configured to upconvert an analog quadrature-phase (Q) signal, where an output of the first mixer and an output of the second mixer are combined to generate the RF transmit signal. The transceiver further includes a variable gain amplifier (VGA) configured to amplify the RF transmit signal to generate an amplified RF transmit signal. The transceiver further includes a plurality of local loopback circuits, including a first local loopback circuit coupled to an output of the VGA and a second local loopback circuit coupled to an input of the VGA.

[0054] In various embodiments, the transceiver further includes at least one sub-sampling analog-to-digital converter (ADC) configured to process the output of at least one of the plurality of local loopback circuits. [Brief explanation of the drawings]

[0055] [Figure 1] 1 is a schematic diagram of a radio frequency (RF) communication system according to one embodiment. [Figure 2A] 2 is a schematic diagram of a portion of an RF communication system according to another embodiment. [Figure 2B] 2 is a schematic diagram of a portion of an RF communication system according to another embodiment. [Figure 3A] FIG. 10 is a schematic diagram of an observation receiver according to another embodiment. [Figure 3B] This is a continuation of Figure 3A. [Figure 4] FIG. 1 is a schematic diagram of a local oscillator (LO) leakage compensation circuit according to another embodiment. [Figure 5A] FIG. 2 is a schematic diagram of a first example of LO leakage. [Figure 5B] FIG. 10 is a schematic diagram of a second example of LO leakage. [Figure 5C] FIG. 10 is a schematic diagram of a third example of LO leakage. [Figure 5D] FIG. 10 is a schematic diagram of a fourth example of LO leakage. [Figure 5E] FIG. 10 is a schematic diagram of a fifth example of LO leakage. [Figure 5F] FIG. 10 is a schematic diagram of a sixth example of LO leakage. [Figure 5G] FIG. 10 is a schematic diagram of a seventh example of LO leakage. [Figure 6A] FIG. 2 is a schematic diagram of one embodiment of a digital transmitter circuit for processing transmitted signal samples and observed receiver samples. [Figure 6B] 6B is a schematic diagram of one embodiment of the data flow of transmit signal samples and observation receiver samples for the digital transmitter circuit of FIG. 6A. [Figure 7] FIG. 10 is a schematic diagram of a transmitter according to another embodiment. [Figure 8] FIG. 2 is a schematic diagram of one embodiment of a digital transmitter circuit. [Figure 9] FIG. 10 is a schematic diagram of a transmitter according to another embodiment. [Figure 10A]1 is a schematic diagram of a circuit including a controllable oscillator, a variable gain amplifier, and a harmonic rejection mixer according to one embodiment. [Figure 10B] 10B is a graph of the circuit of FIG. 10A operating in a divide-by-two mode. [Figure 10C] 10B is a graph of the circuit of FIG. 10A operating in a divide-by-four mode. DETAILED DESCRIPTION OF THE INVENTION

[0056] In the following detailed description of the embodiments, various illustrations of specific embodiments of the present invention are provided. However, the present invention can be embodied in many different ways. In this description, reference is made to the drawings, in which like reference numerals may indicate identical or functionally similar elements. It will be understood that the elements illustrated in the figures are not necessarily drawn to scale. Furthermore, it will be understood that particular embodiments can include more elements and / or a subset of the elements illustrated in the drawings. Furthermore, some embodiments can incorporate any suitable combination of features from two or more drawings.

[0057] Transceivers are used in radio frequency (RF) communication systems to transmit and receive signals associated with a wide variety of communication technologies, such as cellular and / or wireless local area network (WLAN) technologies.

[0058] Examples of RF communication systems having one or more transceivers include, but are not limited to, base stations, mobile devices (e.g., smartphones or handsets), laptop computers, tablets, and wearable electronic devices.

[0059] A transceiver includes a transmitter for transmitting RF signals and a receiver for receiving RF signals. To improve the performance of the transceiver, the transceiver can be calibrated to compensate for impairments. Calibrating the transceiver in this manner may reduce error vector magnitude (EVM), reduce out-of-band emissions, and / or otherwise improve the performance of the transceiver.

[0060] Without calibration, transceiver impairments can lead to degraded performance.

[0061] In one example, a direct conversion quadrature radio includes a zero intermediate frequency (zero IF) transmitter used to transmit an RF signal. The zero IF transmitter upconverts the baseband signal using a local oscillator (LO) frequency that is at or near the carrier frequency. While zero IF transmitters can have significant power and cost advantages compared to IF and superheterodyne transmitters, zero IF transmitters suffer from the performance-limiting obstacle of in-band undesired emissions.

[0062] One such transmitter impairment is LO leakage, where a portion of the LO signal used for mixing appears at the transmitter output. LO leakage can result from finite isolation between the LO port of the mixer and the signal port of the mixer. Without compensation or calibration, LO leakage can lead to failure of tests monitoring undesired transmit emissions.

[0063] For example, an unwanted DC offset at the transmitter's baseband may mix with the LO signal, thereby generating LO power at the transmitter's output. LO leakage may exist at frequencies that have certain limits on unwanted emissions from the transmitter. Therefore, LO leakage may limit transmitter performance in applications with relatively strict emissions specifications. For example, certain zero-IF transmitters support discontinuous carrier aggregation, where the LO frequency may not be within the desired frequency channel. In such implementations, emissions limits may be relatively strict.

[0064] To reduce or limit such unwanted emissions, the transmitter may be calibrated to reduce the level of LO leakage.

[0065] The transmitter may include a non-inverting voltage output Vtx_p and an inverting voltage output Vtx_n for providing a transmitter output signal. Additionally, the transmitter output signal has a differential component approximately equal to (Vtx_p-Vtx_n) and a common-mode component approximately equal to (Vtx_p+Vtx_n) / 2. Differential LO leakage may affect the differential component of the transmitter output signal, and common-mode LO leakage may affect the common-mode component of the transmitter output signal.

[0066] While the differential LO leakage may be locally observable to the transceiver, the common-mode LO leakage component may not be locally observable. Differential LO leakage is also referred to herein as internal LO leakage, and common-mode LO leakage is also referred to herein as external LO leakage.

[0067] Common-mode LO leakage can degrade the performance of transmitters, such as zero-IF transmitters. For example, a balun can be included to convert the transmitter's differential output signal to a single-ended signal suitable for amplification by a power amplifier (PA) and subsequent transmission via an antenna. Ideally, the balun rejects the common-mode component of the LO leakage contained in the transmitter's differential output signal. However, the finite common-mode rejection ratio (CMRR) of the balun can result in a portion of the LO leakage common-mode signal reaching the output of the balun.

[0068] On the other hand, the CMRR of a typical observation receiver used for transmitter observation and calibration is relatively high, so the observation receiver does not observe common-mode LO leakage. Even in implementations where the CMRR of the observation receiver is finite, the CMRR of the observation receiver and the CMRR of the balun may not match. This makes such an observation receiver unsuitable for estimating the amount of common-mode LO leakage and for compensating for it.

[0069] SUMMARY OF THE INVENTION Apparatus and methods for a transceiver are provided herein.

[0070] In a first aspect, an observation receiver is implemented to detect common-mode LO leakage. Such detection may occur while the observation receiver performs other functions, such as observing a portion of the RF transmit spectrum for purposes of power control and / or digital predistortion (DPD). Thus, the observation receiver is implemented to observe common-mode LO leakage as a background process (e.g., as a batch collection) without having to interfere with power control and / or DPD.

[0071] In a second aspect, a transceiver is implemented with an LO leakage compensation circuit that compensates the transmitter for LO leakage based on combining differential LO leakage observations (which may be observed locally within the transceiver) with common-mode LO leakage observations (which may be observed via an off-chip path from the transmitter through an RF front-end system to an observation receiver). Thus, the LO leakage compensation circuit can operate based on observations from a local (on-chip) loopback path and an external (off-chip) loopback path. The LO leakage compensation circuit may include a digital filter, such as a Kalman filter, that processes both the differential and common-mode LO leakage observations to generate a combined LO correction signal for compensating the transmitter.

[0072] In a third aspect, a transmitter includes a pair of mixers for generating an RF transmit signal based on upconverting analog I and Q signals, an LO for providing a clock signal to the pair of mixers, a variable gain amplifier (VGA) for amplifying the RF transmit signal to generate an amplified RF transmit signal, and multiple local loopback paths including a first loopback path after the VGA and a second loopback path before the VGA. The first and second loopback paths can be used to compensate for LO leakage and for other functions such as quadrature error correction (QEC). In a specific implementation, observations from the local loopback paths are digitized by at least one sub-sampling ADC, the mixers are implemented as harmonic rejection mixers, and / or the loopback paths include RF filters for filtering high-order harmonics. Thus, the transceiver is implemented to reduce harmonics (e.g., third, fifth, and / or seventh harmonics) that can cause problems due to aliasing in the sub-sampling ADC.

[0073] Example of an RF communication system that includes a transceiver 1 is a schematic diagram of a radio frequency (RF) communication system 40 according to one embodiment. The RF communication system 40 includes a transceiver 1, an RF front-end system 2, and an antenna 3.

[0074] Although one embodiment of RF communication system 40 is depicted, the teachings herein are applicable to RF communication systems implemented in a wide variety of ways. Accordingly, other implementations are possible.

[0075] In the illustrated embodiment, transceiver 1 includes a receiver 4, a transmitter 5, an observation receiver 6, and a local oscillator (LO) leakage compensation circuit 7. Although an example having one transmit channel, one receive channel, and one observation channel is depicted, the teachings herein are applicable to transceivers having other numbers of transmit, receive, and / or observation channels. Furthermore, the numbers of transmit, receive, and observation channels need not be equal.

[0076] Transceiver 1 receives in-phase (I) and quadrature-phase (Q) transmit data from a baseband processor (not shown in FIG. 1) and provides I and Q receive data to the baseband processor.

[0077] As shown in FIG. 1, the transmitter 5 includes an I-path digital-to-analog converter (DAC) 21a, an I-path filter 22a, an I-path mixer 23a, a Q-path DAC 21b, a Q-path filter 22b, a Q-path mixer 23b, and an LO 24.

[0078] I path DAC 21a functions to process the I transmit data to generate an analog I signal, which is filtered by I path filter 22a to generate a filtered I signal. The filtered I signal is upconverted to radio frequency by I path mixer 23a. In particular, I path mixer 23a mixes the filtered I signal with a first LO clock signal from LO 24 to generate the upconverted I signal.

[0079] Similarly, Q-path DAC 21b functions to process the Q transmit data to generate an analog Q signal, which is filtered by Q-path filter 22b to generate a filtered Q signal. Additionally, Q-path mixer 23b mixes the filtered Q signal with a second LO clock signal from LO 24 to generate an upconverted Q signal.

[0080] The LO 24 can be implemented in a variety of ways, including, but not limited to, using a frequency synthesizer such as a fractional-N phase-locked loop (PLL). The first and second LO clock signals can have a phase difference suitable for upconverting the filtered I and Q signals. For example, a quadrature or 90-degree phase difference can be used.

[0081] The upconverted I signal and the upconverted Q signal are combined to generate a differential transmit signal TX, which is fed to an RF front-end system 2.

[0082] In the illustrated embodiment, RF front-end system 2 includes a balun 11, a power amplifier (PA) 12, a low-noise amplifier (LNA) 13, antenna access circuitry 14, and a directional coupler 15. While one example of front-end circuitry is depicted, other implementations are possible. For example, RF front-end system 2 may include additional structures, such as filters, amplifiers, attenuators, duplexers, diplexers, switches, and / or control circuitry. Furthermore, the components may be configured in other ways, including, but not limited to, using separate antennas for transmission and reception.

[0083] Balun 11 receives a differential transmit signal TX, which is converted by balun 11 to a single-ended transmit signal for amplification by PA 12. PA 12 outputs an amplified RF transmit signal, which is provided to antenna 3 by antenna access circuitry 14. Antenna access circuitry 14 may include switches, duplexers, diplexers, and / or other structures suitable for controlling the access of the transmit and receive paths to antenna 3.

[0084] 1, a directional coupler 15 is included between the output of PA 12 and antenna access circuitry 14 and functions to sense the amplified RF transmit signal and generate an observation signal OBS, which is provided to an observation receiver 6 of transceiver 1, as shown in FIG.

[0085] The LNA 13 receives an RF receive signal from the antenna 3 via the antenna access circuit 14. The LNA 13 amplifies the RF receive signal to generate an amplified receive signal RX, which may be single-ended or differential. The receiver 4 processes the amplified receive signal RX to generate I and Q receive data.

[0086] As shown in FIG. 1, transceiver 1 includes LO leakage compensation circuitry 7 that compensates transmitter 5 for LO leakage based on observation of both differential LO leakage and common-mode leakage.

[0087] For example, in the illustrated embodiment, the LO leakage compensation circuit 7 includes a differential LO leakage observation circuit 31 for observing the amount of differential LO leakage present in the differential transmit signal TX. Because the transceiver 1 is typically formed on a semiconductor die separate from the semiconductor die used to form the RF front-end system 2, the observation of the differential LO leakage may be based on a local observation path of the transceiver 1.

[0088] 1, the LO leakage compensation circuit 7 further includes a common-mode LO leakage observation circuit 32 that observes the common-mode LO leakage based on data passing through the observation receiver 6. In particular implementations, the observation receiver 6 is implemented to allow observation of the common-mode LO leakage without interfering with other operations of the observation receiver 6, such as observing a portion of the frequency spectrum of the amplified RF transmit signal for purposes of transmit power control and / or DPD.

[0089] In the illustrated embodiment, LO leakage compensation circuit 7 further includes a digital filter 33 that combines observations from differential LO leakage observation circuit 31 and common-mode LO leakage observation circuit 32 to generate an LO leakage compensation signal for transmitter 5. In a particular implementation, digital filter 33 includes a Kalman filter.

[0090] The LO leakage compensation circuit 7 may LO leakage compensate the transmitter 5 in a variety of ways. In one example, digital compensation may be used. For example, a digital summer may be used to add appropriate DC offsets to the I and Q paths. In a second example, analog compensation may be used. For example, a controllable oscillator signal (with controllable differential and / or common mode components) may be combined with the differential transmit signal TX to provide LO leakage compensation. Furthermore, a combination of digital and analog techniques may be used.

[0091] An exemplary observation receiver for observing common-mode LO leakage 2A is a schematic diagram of a portion of an RF communication system 110 according to another embodiment. The RF communication system 110 includes an observation receiver 101 and an RF front-end system 102. The observation receiver 101 is part of a transceiver (not shown in FIG. 2A).

[0092] The RF front-end system 102 includes a power amplifier 108 and a directional coupler 109. The power amplifier 108 outputs an amplified RF transmit signal, which is sensed by the directional coupler 109 to generate an RF observation signal OBS. For clarity of illustration, only the power amplifier 108 and the directional coupler 109 of the RF front-end system 102 are depicted. However, the RF front-end system 102 may include other components.

[0093] In the illustrated embodiment, observation receiver 101 includes an RF analog-to-digital converter (RF ADC) 103 for generating a digital observation signal based on an RF observation signal OBS from RF front-end system 102. While an example with an RF ADC 103 is shown, other implementations are possible, such as a configuration using a cascade of observation mixers and ADCs.

[0094] 2A, observation receiver 101 further includes a digital observation path 104, which includes various digital circuits for processing the digital observation signal. Digital observation path 104 is also referred to herein as a digital observation circuit. Digital observation path 104 can be used to perform observations for purposes of transmit power control, DPD, and / or other functions.

[0095] 2A, also included for observing common-mode LO leakage are digital mixer 105, digital oscillator 106, and digital accumulator 107. Digital mixer 105 taps off a portion (or portions) of digital observation path 104, thus observing the digital data stream of observation receiver 101 without having to interfere with operations of observation receiver 101, such as transmit power control and / or DPD.

[0096] Digital mixer 105 is used to frequency-shift the extracted observation data based on a digital oscillation signal from digital oscillator 106. Thus, the frequency content for the common-mode LO leakage observation can be offset in frequency relative to the data passing through digital observation path 104. Digital accumulator 107 accumulates the observation data, which can be further processed by the LO leakage compensation circuit (shown here in FIG. 2A). In a specific implementation, digital oscillator 106 corresponds to a numerically controlled oscillator (NCO).

[0097] 2B is a schematic diagram of a portion of an RF communication system 120 according to another embodiment. The RF communication system 120 includes an observation receiver 111 and an RF front-end system 112. The observation receiver 111 is part of a transceiver (not shown in FIG. 2B).

[0098] RF front-end system 112 includes power amplifiers 108a, 108b, ... 108n, directional couplers 109a, 109b, ... 109n, and multiplexer 115. As indicated by the ellipsis, any number of power amplifiers and directional couplers may be included. Additionally, power amplifiers 108a, 108b, ... 108n may transmit at the same frequency or at different frequencies. Multiplexer 115 is used to select particular sensed signals from directional couplers 109a, 109b, ... 109n to provide as RF observation signal OBS.

[0099] By multiplexing the directional couplers 108a, 108b, ... 108n, a common observation receiver can be used to observe multiple transmit paths passing through the RF front-end system 120. However, other implementations are possible.

[0100] 2A, observation receiver 111 includes RF ADC 103 and digital observation path 104. Observation receiver 111 further includes digital mixers 105a, 105b, ... 105n, digital oscillators 106a, 106b, ... 106n (any or all of which may be NCOs), and digital accumulators 107a, 107b, ... 107n.

[0101] The digital mixers 105a, 105b, . . . 105n take data from the digital observation path 104 and thus observe the digital data stream of the observation receiver 111 without having to interfere with the operation of the observation receiver 111, such as transmit power control and / or DPD.

[0102] Furthermore, multiple instantiations of mixers, oscillators, and accumulators are provided to separately track common-mode LO leakage associated with multiple transmit paths. For example, each of PAs 108a, 108b, ... 108n may amplify an RF transmit signal generated using a different LO, and thus the common-mode LO leakage will be different for each transmit channel.

[0103] Thus, based on which directional coupler 109a, 109b, . . . 109n is selected by multiplexer 115, the corresponding accumulator 107a, 107b, . . . 107n of observation receiver 111 may be activated.

[0104] By accumulating frequency-shifted samples indicative of common-mode LO leakage in the background, a user can change the observation path of RF front-end system 112 and / or perform other desired operations without having to shut it down for the purpose of observing the common-mode LO leakage. Thus, the common-mode LO leakage compensation process can be transparent to the user.

[0105] 3A and 3B are schematic diagrams of an observation receiver 270 according to another embodiment. The observation receiver 270 includes an RF ADC 201, an ADC interface with a first-in-first-out (FIFO) 202, an input multiplexer 203, a digital downconverting mixer 204, a fine NCO 205, an I-path decimator 206a, a first I-path finite impulse response (FIR) filter 207a, a second I-path FIR filter 208a, a first I-path interpolator 211a, a second I-path interpolator 212a, a third I-path FIR filter 208b, a fourth I-path interpolator 211b, a fifth I-path interpolator 212c, a fifth I-path interpolator 212d, a sixth I-path interpolator 212e, a sixth I-path interpolator 212f, a sixth I-path interpolator 212g, a sixth I-path interpolator 212h, a sixth I-path interpolator 212i, a sixth I-path interpolator 212j, a sixth I-path interpolator 212m ... I path interpolator 213a, first I path multiplexer 221a, second I path multiplexer 222a, third I path multiplexer 223a, fourth I path multiplexer 224a, fifth I path multiplexer 225a, sixth I path multiplexer 226a, first I path half-band (HB) filter 231a, second I path HB filter 232a, I path mixer 235a, Q path decimator 206b, first Q path FIR filter 207b, a second Q-path FIR filter 208b, a first Q-path interpolator 211b, a second Q-path interpolator 212b, a third Q-path interpolator 213b, a first Q-path multiplexer 221b, a second Q-path multiplexer 222b, a third Q-path multiplexer 223b, a fourth Q-path multiplexer 224b, a fifth Q-path multiplexer 225b, a sixth Q-path multiplexer 226b, a first Q-path HB filter 231b, It includes a second Q-path HB filter 232b, a Q-path mixer 235b, a peak detection power measurement circuit 241, a detection multiplexer 242, an NCO 245, a formatter 246, a crossbar link 248, a framer 249, an observation capture random access memory (RAM) 251, an observation capture multiplexer 252, a leakage observation multiplexer 254, a leakage observation NCO 255, a leakage observation mixer 256, and a leakage observation accumulator 257.

[0106] Although one embodiment of an observation receiver is depicted, the teachings herein are applicable to observation receivers implemented in other ways.

[0107] Exemplary Kalman Filter for Processing Local and External LO Leakage Observations 4 is a schematic diagram of an LO leakage compensation circuit 310 according to another embodiment. The LO leakage compensation circuit 310 includes a differential LO leakage observation circuit 301, a common-mode LO leakage observation circuit 302, and a Kalman filter 303.

[0108] In the illustrated embodiment, the Kalman filter 303 processes both the differential and common-mode LO leakage observations to generate a combined LO correction signal for compensating the transmitter of the transceiver.

[0109] 5A-5G depict various transmitter LO leakage cases that may occur during operation of a transceiver. s For all figures, f s is f s / 2, where f s is the sampling rate of the ADC. Furthermore, data collection can be performed around a single alias point if desired.

[0110] 5A is a schematic diagram of a first example of LO leakage, in which there is no transmit signal but there is internal transmitter LO leakage.

[0111] In certain implementations, since no signal is present, a perturbation is introduced in this scenario (e.g., to find the channel for correction purposes). Furthermore, if a transmitter LO leakage correction loop is operated (e.g., in LO leakage correction circuit 7 of FIG. 1), at the end of each pass, an adjustment is made to the transmitter DC (e.g., a digital adjustment to the I and Q transmit signals input to the transmitter) to attempt to correct for the LOL. This adjustment can be used with the pre-adjustment and post-adjustment measurements as a perturbation.

[0112] When starting from an initial correction point, channel information can be inherited from a previously performed calibration or other valid channels. If such information is not available, an initial guess can be used as the transmitter LO leakage correction loop iteratively adjusts and learns from the results. For example, in the absence of channel information, the digital filter (e.g., a Kalman filter) of the LO leakage compensation circuit can be initialized to a general guess for a range with high uncertainty. At the end of the first data collection pass, the Kalman filter can estimate the correction.

[0113] In certain implementations, the correction is limited to a reasonable amount, for example, 128 least significant bits (LSBs). The estimated correction may be incorrect, but in subsequent passes, the estimated correction can act as a perturbation to help settle on the actual correction value. For example, this scenario may correspond to an edge case of the algorithm and typically occurs rarely in practice due to the presence of a transmitted signal or the availability of inherited channel information.

[0114] Continuing with reference to FIG. 5A, in a particular implementation, the correction is implemented using the following equation: LB′ corresponds to the frequency-shifted loopback sample, as discussed further below:

number

[0115] 5B is a schematic diagram of a second example of LO leakage, which corresponds to a scenario where the LO leakage is below the carrier frequency.

[0116] In such a scenario, a very accurate channel measurement is desirable, since one will observe a large intentional TX signal on the LO, which needs to be measured at LPBCK and removed from the correction. To reference the loopback measurement (observation receiver) to the transit signal, the measurement can be divided by the channel, and therefore accurate channel is desired for accurate cancellation. In the digital transmitter electronics, an accumulator can provide such information.

[0117] 5C is a schematic diagram of a third example of LO leakage. In this scenario, the carrier image is on the LO frequency. In the third scenario, the loopback ADC receives a signal at frequency f s so that the transmit spectrum at 2f0-LO appears in the loopback spectrum at the same location as the LO frequency.

[0118] 5D is a schematic diagram of a fourth example of LO leakage, in which the carrier is on the LO frequency and the carrier image is on the -LO frequency.

[0119] 5B-5D, measurements of the TX at LO and -LO can be made along with measurements of the observation receiver at the LO frequency. To do this, a Kalman filter can be used that operates on three variables: g0 (gain at LO frequency), g1 (gain at -LO frequency), and u (correction).

[0120] For the second case depicted in Figure 5B, g can be calculated by cross-correlation. For the third case depicted in Figure 5C, both g and g can be calculated by cross-correlation. In the third case, the Kalman filter trains g separately from g. Thus, in this case, g is trained using a large signal and g is trained via correction.

[0121] In the absence of a signal, the Kalman filter knows the corrections made previously and their corresponding effect on the current measurements, so it can automatically use the corrections as perturbations to generate channel information.

[0122] 5E is a schematic diagram of a fifth example of LO leakage, in which internal transmitter LO leakage is present and the carrier is far from both the LO and −LO frequencies.

[0123] 5F is a schematic diagram of a sixth example of LO leakage, in which external transmitter LO leakage is present and the carrier is on the LO frequency.

[0124] 5G is a schematic diagram of a seventh example of LO leakage, in which external transmit LO leakage is present and the carrier is far from the LO frequency.

[0125] Although seven examples of LO leakage are depicted, other LO leakage scenarios are possible, for example, in another example, no transmit signal is present, but external LO leakage is present.

[0126] Figure 6A is a schematic diagram of one embodiment of a digital transmitter circuit 410 for processing transmit signal samples (TX) and observation receiver samples (LPBCK). Figure 6B is a schematic diagram of one embodiment of a transmit signal sample and observation receiver sample data flow 420 for the digital transmitter circuit 410 of Figure 6A.

[0127] The digital transmitter circuit 410 receives digital samples of the RF transmit signal TX indicative of the differential LO leakage signal. The digital transmitter circuit 410 also receives digital samples of the observation signal LPBCK received via loopback from the observation receiver.

[0128] In the illustrated embodiment, the digital transmitter circuit 410 includes a frequency selection circuit (FSC) 401 for observing samples at different frequencies. The FSC 401 shifts the TX and LPBCK samples to generate unshifted TX samples, TX samples shifted by 2*(fs-LO), and LPBCK samples shifted by LO (to downshift the LO content to DC). These shifts can be achieved in any suitable manner, for example, using digital mixers 403 and 404 controlled by NCOs 405 and 406. In a particular implementation, the FSC 401 includes a digital accumulator 407 for accumulating each of the three shifted samples.

[0129] For internal LO leakage monitoring, the TX / LPBCK signal may come from FSC 401. However, for external LO leakage monitoring, it is preferable that the TX signal not be routed through FSC 401, but instead may be independently routed to a separate LOL accumulator.

[0130] External LOL observations depend on the transmitter to observation receiver mapping, which can be controlled by the user. Internal (quadrature error correction (QEC) and LO leakage) observations do not depend on the transmitter to observation receiver mapping, so there are times when internal observations are possible even when external observations are not. If external observations were dependent on FSC, they could block internal observations until the transmitter to observation receiver mapping is changed.

[0131] To avoid external LO leakage observations blocking internal QEC observations, the TX accumulator for external LOL observations may see the TX signal directly (without going through the FSC) and allow the TX QEC to access the FSC at the same time.

[0132] In a particular implementation, a separate I / Q accumulator is built in the digital transmit path that is unconnected and independent from the FSC 401. Similar I / Q accumulators can be built in the observation path for each of the transmit channels. The I / Q accumulators can perform accumulation based on control from the transmitter LO leakage block. Such loopback data can be shifted in frequency using an NCO, as discussed above, so that the LO is at baseband DC.

[0133] In a specific implementation, batch collection is used for collection of TX and LPBCK samples, allowing updates to the accumulator only when both are collected. To implement this, for TX samples, an active accumulator, a holding register, and a final accumulator can be used. For LPBCK samples, an active accumulator and a final accumulator can be used. During data collection, the active accumulator sums the samples of the TX stream. Once the batch size is reached, the active accumulator can be transferred to the holding register, and the active accumulator is zeroed to start the next batch of samples. As soon as the active accumulator completes a batch, its contents can be added directly to the final accumulator, so the ORX holding register does not need to be used.

[0134] Here, one embodiment of a Kalman filter model for processing transmitter leakage observation data is described.

[0135] The transmitter (Tx) LO leakage (e.g., differential LO leakage) is observed through the internal loopback path. Equation 2 is shown below, where y int where x[n] is the internal loopback output after rotating the TxLO frequency to DC, x0[n] is the user Tx signal near DC, x1[n] is the interfering user Tx signal that aliases to the TxLO at the loopback output, g0 is the complex channel from Tx to loopback at the LO frequency (DC), g1 is the complex channel from Tx to loopback at the interfering frequency, and μc is the current TxLO leakage correction value, and μ int is the Tx LO leakage observable at the loopback output but fed back to the Tx input, and u[n] is uncorrelated zero-mean white noise from the Tx, loopback, or both.

number

[0136] The Tx LO leakage (e.g., common-mode LO leakage) is also observed through an external path that is looped back to the observation receiver (ORx). Equation 3 is shown below, where y ext [n] is the external ORx output after rotating the Tx LO frequency to DC, x0[n] is the user Tx signal near DC, h0 is the complex channel from Tx to ORx at the LO frequency (DC), and μ c is the current TxLO leakage correction value, and μ int is the TxLO leakage observable at the loopback output but returned to the Tx input, and μ ext is the additional Tx LO leakage observable at the ORx output fed back to the Tx input, and v[n] is the uncorrelated zero-mean white noise from Tx, ORx, or both.

number

[0137] In a particular implementation, the Kalman filter performs frequent internal observations while making less frequent external observations to optimize the combined μ at the Tx / ORx outputs. int +μ ext To cancel out μ c It is implemented to learn the internal leakage μ int can be assumed to capture the time-varying component of the LO leakage, while μ ext can be assumed to be a static bias that changes fairly infrequently over time.

[0138] The following equations 4 and 5 define the correlation and sum operations.

number

number

[0139] With FSC, for observed internal LO leakage, the Tx input can be cross-correlated with the internal loopback output to quickly learn the unknown channel, as shown in Equations (6) and (7).

number

number

[0140] Since the user Tx signals may have near-zero mean, the cross-correlation is performed with the unknown leakage μ int is not suitable for direct estimation. The following equation (8) is a function of the calculated sum (known constant reference signal μ c (similar to the cross-correlation for

number

[0141] Therefore, from the observation of the internal path, three unknowns (g0, g1, μ int ) and three noise contributions (C(u,x0 * ),C(u,x1 * ),S(u)) are obtained. These observation equations are int terms, is nonlinear in the unknowns.

[0142] For observed external LO leakage, an external LOL accumulator block can be used to perform a time-synchronous summation of the Tx input and ORx output as shown in Equation 9 below.

number

[0143] For the observed external LO leakage, there are three unknowns (h0, μ int ,μ ext ) and noise contribution (S(v)). This observation equation is given by int Term and h0μ ext The term is nonlinear in the unknowns.

[0144] While the above observation equations can be made linear by treating the channel and LOL unknowns jointly (e.g., estimating LOL at the loopback / ORx output instead of referencing the Tx input), the nonlinear form is the μ int Since the observation equations are nonlinear, a nonlinear estimation algorithm (e.g., a Kalman filter) can be used.

[0145] Combining the above internal and external observation equations, there are a total of five unknown state variables, as shown in Equation 10.

number

[0146] To allow for process noise to introduce variations over time, an identity state transition matrix may be used. The prediction step of the Kalman filter may be given by Equations (11) and (12) shown below.

number

number

[0147] The Kalman filter update step may include two steps: (i) estimating what the observation values ​​should be given the current values ​​of the unknowns using the full nonlinear observation equations, and (ii) performing a linear Kalman filter update by defining the Jacobian as the observation matrix.

[0148] For the first step, we can use Equations (13), (14), (15), (16) and (17) below to estimate what the observed values ​​should be given the current values ​​of the unknowns.

number

number

number

number

number

[0149] For the second step, a linear Kalman filter update can be performed using the Jacobian of the observation system as the observation matrix, as shown in Equation (18) below.

number

[0150] If a particular observation is missing, the corresponding column can be removed from the Jacobian and the Kalman update is performed with reduced observation dimension. For example, if external observation data is not available at the current time step, the observation dimension is reduced from 4 to 3.

[0151] Exemplary Transmitter Implemented with Multiple Local Loopbacks 7 is a schematic diagram of a transmitter 540 according to another embodiment. The transmitter 540 includes a digital transmitter circuit 520, an I path DAC 521 a, an I path controllable filter 522 a, an I path mixer 523 a, a Q path DAC 521 b, a Q path controllable filter 522 b, a Q path mixer 523 b, a variable gain amplifier (VGA) 525, a first local loopback circuit 531, a second local loopback circuit 532, and a sub-sampling ADC 533.

[0152] As shown in FIG. 7, digital transmitter circuit 520 receives I transmit data and Q transmit data, which are provided to I path DAC 521a and Q path DAC 521b, respectively, after digital processing.

[0153] The I path DAC 521a functions to convert the I transmit data to an analog I signal, which is filtered by the I path controllable filter 522a to generate a filtered I signal. The filtered I signal is upconverted to radio frequency by the I path mixer 523a. In particular, the I path mixer 523a mixes the filtered I signal with a first LO clock signal from the LO 524 to generate the upconverted I signal.

[0154] Similarly, Q path DAC 521b functions to process the Q transmit data to generate an analog Q signal, which is filtered by Q path controllable filter 522b to generate a filtered Q signal. Additionally, Q path mixer 523b mixes the filtered Q signal with a second LO clock signal from LO 524 to generate an upconverted Q signal.

[0155] The LO 524 can be implemented in a variety of ways, including, but not limited to, using a frequency synthesizer such as a fractional-N PLL. The first and second LO clock signals can have a phase difference suitable for upconverting the filtered I and Q signals. For example, a quadrature phase difference can be used.

[0156] The upconverted I signal and the upconverted Q signal are combined to generate a differential transmit signal, which is amplified by VGA 525 to generate an amplified differential transmit signal that is provided between pins TX+ and TX−. The amplified differential transmit signal is provided to an RF front-end system (e.g., RF front-end system 2 in FIG. 1).

[0157] 7, the digital transmitter circuit 520 further includes an LO leakage compensation circuit 534 for adjusting the I transmit signal and the Q transmit signal to compensate for LO leakage due to the LO 524. Such LO leakage may include differential LO leakage and common-mode LO leakage.

[0158] The LO leakage compensation is based on transmit samples TX of the RF transmit signal generated by the transmitter 540 and observation samples OBS captured from an observation receiver (e.g., see observation receiver 6 in FIG. 1) using an external loopback path through the RF front-end system.

[0159] In the illustrated embodiment, the transmitter 540 operates using multiple (two in this example) local loopback paths in addition to an external loopback path through the observation receiver. In particular, a first local loopback circuit 531 is included to observe the RF transmit signal after amplification by the VGA 525, and a second local loopback circuit 532 is included to observe the RF transmit signal prior to amplification by the VGA 525.

[0160] Including multiple local loopback paths provides many advantages. For example, the first local loopback path 531 includes the effects of the VGA 525 and any noise coupled through the RF transmit signal output pin. However, if the VGA 525 is operated at a low gain setting, the amplified RF transmit signal may have a low signal-to-noise ratio (SNR), which results in delay due to the large number of samples taken for averaging. Furthermore, any calibration for LO leakage using the first local loopback path 531 also exposes downstream circuitry (e.g., a power amplifier) ​​to the RF transmit signal generated by the transmitter 540.

[0161] In contrast, the second local loopback path 532 can be used even when the VGA 525 is turned off to prevent the RF transmit signal from propagating to (and potentially damaging) downstream circuitry. Furthermore, the second local loopback path 532 is insulated from the effects of output impedance matching and exhibits good SNR even at low gain settings of the VGA 525. However, the second local loopback path 532 may overlook the effects of LO leakage that occurs after the VGA 525.

[0162] 7, the RF transmit signal may have a relatively high frequency, for example, up to about 7.125 GHz for cellular communications using Frequency Range 1 (FR1) of fifth generation (5G) technology. The sub-sampling ADC 533, in this embodiment, operates at a sampling rate that is less than the carrier frequency of the RF transmit signal. Using a sub-sampling ADC provides many advantages, such as reduced area, power consumption, and / or cost.

[0163] To help prevent aliasing effects (e.g., image problems) from interfering with the operation of the LO leakage compensation, I path mixer 523a and Q path mixer 523b can be implemented as harmonic rejection mixers. In a particular implementation, I path mixer 523a and Q path mixer 523b are at least third-order harmonic rejection mixers, and more preferably third- and fifth-order harmonic rejection mixers.

[0164] Implementing the mixer in this manner reduces problems due to aliasing of the third and fifth harmonic components.

[0165] In certain implementations, higher harmonics, such as the seventh harmonic, are accounted for using RF filters included in the first local loopback circuit 531 and the second local loopback circuit 532. Thus, the combination of a harmonic rejection mixer and an RF filter can serve to reduce problems due to aliasing.

[0166] To further account for aliasing, LO leakage compensation circuitry can be implemented to model and predict the effects of aliasing. For example, FSC (see, e.g., FIGS. 6A and 6B) can be included in digital transmitter circuitry to help achieve such functionality. Thus, digital processing can act as another layer of compensation for aliasing.

[0167] Figure 8 is a schematic diagram of one embodiment of a digital transmitter circuit 570. The digital transmitter circuit 570 may be used, for example, in the digital transmitter circuit 520 of Figure 7. As shown in Figure 8, the digital transmitter circuit 570 includes a QEC / LO leakage compensation circuit 561, a programmable finite impulse response filter (PFIR) 562, an interpolation / filtering circuit 563, a frequency selection and storage circuit 564, and a processor 565 that includes software stored in memory and operable to implement a Kalman filter 566.

[0168] The digital transmitter circuit 570 processes the digital I and Q transmit signals and provides the processed digital I and Q transmit signals to a pair of DACs (not shown in FIG. 8) that are received from a baseband processor (not shown in FIG. 8).

[0169] In the illustrated embodiment, the frequency selection and storage circuit 564 receives transmit samples TX of the locally derived RF transmit signal (through one or more on-chip local loopback paths) and observation samples OBS captured from the observation receiver (through the RF front end) through one or more external loopback paths. The frequency selection and storage circuit 564 can shift the frequency of the samples (e.g., using a digital mixer and NCO) and store the frequency-shifted samples, for example, as described above with respect to Figures 6A and 6B.

[0170] Processor 565 receives the accumulated samples, and Kalman filter 566 operates to process the accumulated samples to control settings for LO leakage compensation in at least QEC / LO leakage compensation circuit 561. Although shown as controlling only QEC / LO leakage compensation circuit 561 (e.g., a DC offset added to each of the digital I and Q transmit signals), Kalman filter 566 may control other circuitry.

[0171] For example, the Kalman filter 566 can control the PFIR 562 to provide equalization for the I and Q path controllable filters (e.g., I path controllable filter 522a and Q path controllable filter 522b in FIG. 7) and / or settings for the controllable filters themselves (such as the values ​​of the inductors and / or capacitors used for filtering). Additionally or alternatively, the Kalman filter 566 can control the frequency, amplitude, and / or phase settings used to calibrate the transmitter for LO leakage. For example, in certain embodiments, the calibration cycle is operated using a test tone that can be injected when the transmit signal is not present or added to the transmit signal when it is present. To this end, the Kalman filter 566 can provide dithering or other perturbations to assist in the calibration.

[0172] Although one example of a digital transmitter circuit is depicted, other implementations are possible.

[0173] 9 is a schematic diagram of a transmitter 620 according to another embodiment. The transmitter 620 includes various components, including a pair of DACs 601, a pair of controllable filters 602, a pair of mixers 603, a VGA 604, a first local loopback path LB1, a second local loopback path LB2, and a sub-sampling ADC 614.

[0174] 9, the first loopback path LB1 is after the VGA 604 and includes a controllable attenuator 609 and a first gain-controlled harmonic filtering circuit 611. Additionally, the second loopback path LB2 is before the VGA 604 and includes an input switch 610 (capable of inverting the polarity of the differential RF transmit signal) and a second gain-controlled harmonic filtering circuit 612.

[0175] 9, the first loopback path LB1 is coupled to the third gain control harmonic filtering circuit 613 by a first pair of switches 616a, and the second loopback path LB2 is coupled to the third gain control harmonic filtering circuit 613. The output of the third gain control harmonic filtering circuit 613 is provided to a sub-sampling ADC 614. The sub-sampling ADC 614 outputs samples of the RF transmit signal and / or the amplified RF transmit signal.

[0176] To help prevent aliasing effects from interfering with the operation of the LO leakage compensation, the pair of mixers 603 can be implemented as harmonic rejection mixers, for example, mixers that reject the third and fifth harmonics. Implementing the mixers in this manner mitigates problems caused by aliasing the third and fifth harmonic components. Such aliasing can occur because the sub-sampling ADC 614 has a sampling rate that is less than the carrier frequency of the RF transmit signal.

[0177] In certain implementations, higher order harmonics, e.g., the seventh harmonic, are accounted for using the depicted RF filters of gain controlled harmonic filtering circuits 611-613. Thus, the combination of harmonic rejection mixers and RF harmonic filters can serve to reduce problems due to aliasing.

[0178] To further account for aliasing, the digital transmitter circuitry that processes the transmit samples from the sub-sampling ADC can be implemented to model and predict the effects of aliasing. For example, an FSC (see, e.g., FIGS. 6A and 6B) can be included in the digital transmitter circuitry to help achieve such a function. Thus, the digital processing can act as another layer of compensation for aliasing.

[0179] Figure 10A is a schematic diagram of a circuit 720 including a controllable oscillator 701, a VGA 707, and a harmonic rejection mixer 700 according to one embodiment. Figure 10B is a graph of the circuit 720 of Figure 10A operating in a divide-by-2 mode. Figure 10C is a graph of the circuit 720 of Figure 10A operating in a divide-by-4 mode.

[0180] 10A, a controllable oscillator 701 (e.g., a fractional-N PLL) outputs a pair of clock signals CKP and CKN. The circuit 720 further includes a controllable divider 702, which can divide the pair of clock signals CLKP and CKN using a controllable division factor, e.g., 1, 2, or 4.

[0181] The control circuit 720 further includes a divide-by-2 mode circuit 703 and a divide-by-4 circuit 704 for processing the output of the controllable divider 702 to generate multiple LO clock signal phases that are mixed with the I and Q signals by a mixer 705. In this example, six LO clock signal phases (10, II, 12, QO, QI, Q2) are used for purposes of harmonic rejection of the third and fifth harmonics. The outputs of the mixers 705 are combined by a combiner 706 to generate the RF transmit signal that is fed to a VGA 707.

[0182] By using multiple mixers 705 combined in this manner, third and fifth harmonic rejection is provided.

[0183] If the controllable oscillator 701 is operating at a relatively low frequency (eg, 3.5 GHz or less), a divide-by-four circuit 704 can be used to generate multiple clock signal phases, as shown in FIG. 10B.

[0184] However, if the desired LO frequency is relatively high (e.g., greater than 3.5 GHz), it may be impractical to operate controllable oscillator 701 at four times this LO frequency and use divide-by-four circuit 704. For this reason, divide-by-two circuit 703 is included, which provides a divide-by-two and uses interpolation to generate other LO clock signal phases as desired to provide harmonic rejection of the third and fifth harmonics. An exemplary result of the divide-by-two mode is shown in FIG. 10C.

[0185] conclusion The transceiver in this specification is not limited to RF signals in the range of 100 MHz to 7 GHz, but also includes signals in the X band (approximately 7 GHz to 12 GHz), K u band (approximately 12GHz to 18GHz), K band (approximately 18GHz to 27GHz), K a The present invention can handle signals of various frequencies, including higher frequency signals such as the V-band (approximately 27 GHz to 40 GHz), the V-band (approximately 40 GHz to 75 GHz), and / or the W-band (approximately 75 GHz to 110 GHz). Accordingly, the teachings herein are applicable to a wide variety of RF communication systems, including microwave systems.

[0186] The above description may refer to elements or features as being "connected" or "coupled." As used herein, unless expressly stated otherwise, "connected" means that one element / feature is directly or indirectly connected to another element / feature, but not necessarily mechanically connected. Similarly, unless expressly stated otherwise, "coupled" means that one element / feature is directly or indirectly coupled to another element / feature, but not necessarily mechanically coupled. Thus, while the various schematic diagrams shown in the figures depict example configurations of elements and components, in an actual embodiment, additional intervening elements, devices, features, or components may be present (provided the functionality of the depicted circuit is not adversely affected).

[0187] While specific embodiments have been described, these embodiments are presented by way of example only and are not intended to limit the scope of the present disclosure. Indeed, the novel apparatus, methods, and systems described herein may be embodied in a variety of other forms, and various omissions, substitutions, and changes in the form of the methods and systems described herein may be made without departing from the spirit of the present disclosure. For example, while the disclosed embodiments are presented in a given configuration, alternative embodiments may perform similar functions using different components and / or circuit topologies, and some elements may be deleted, moved, added, subdivided, combined, and / or modified. Each of these elements may be implemented in a variety of different ways. Any suitable combination of elements and operations of the various embodiments described above may be combined to provide further embodiments. Accordingly, the scope of the present invention is defined solely by reference to the appended claims.

[0188] It should be understood that any claim may depend on any preceding claim of the same type unless it is clearly not technically feasible.

Claims

1. A transceiver, a transmitter configured to generate a radio frequency (RF) transmit signal, the transmitter configured to generate first observation data based on observing the RF transmit signal on one or more local observation paths within the transmitter; an observation receiver configured to generate second observation data based on processing an observation signal received from an external loopback path from the transmitter through an RF front end to the observation receiver; a local oscillator (LO) leakage compensation circuit configured to compensate for leakage of the transmitter LO, the LO leakage compensation circuit comprising a digital filter configured to process the first observation data and the second observation data.

2. 2. The transceiver of claim 1, wherein the first observed data indicates a differential LO leakage of the LO and the second observed data indicates a common mode LO leakage of the LO.

3. 2. The transceiver of claim 1, wherein the digital filter is a Kalman filter.

4. The transceiver of claim 3 , wherein the Kalman filter is configured to estimate a correction for the LO leakage using a plurality of non-linear equations.

5. The transceiver of claim 4 , wherein the Kalman filter is further configured to modify the correction value based on a linear update of the Kalman filter using an observation matrix.

6. The transceiver of claim 5 , wherein the observation matrix corresponds to a Jacobian matrix.

7. 5. The transceiver of claim 4, wherein a first portion of the plurality of nonlinear equations is a function of the first observed data and a second portion of the plurality of nonlinear equations is a function of the second observed data.

8. 2. The transceiver of claim 1, wherein the transmitter includes digital transmitter circuitry configured to process digital in-phase (I) and quadrature-phase (Q) signals, and the digital filter is configured to compensate the transmitter for leakage of the LO based at least in part on controlling the digital transmitter circuitry.

9. 9. The transceiver of claim 8, wherein the digital filter is configured to control a first DC offset of the digital I signal and a second DC offset of the digital Q signal.

10. 9. The transceiver of claim 8, wherein the digital transmitter circuitry includes a programmable finite impulse response (PFIR) filter, the digital filter configured to control the PFIR.

11. 9. The transceiver of claim 8, wherein the transmitter comprises: an I-path digital-to-analog converter (DAC) coupled to the digital transmitter circuit and configured to generate an analog I signal; a controllable I-path filter configured to generate a filtered I signal based on filtering the analog I signal; a Q-path DAC coupled to the digital transmitter circuit and configured to generate an analog Q signal; and a controllable Q-path filter configured to generate a filtered Q signal based on filtering the analog Q signal.

12. 12. The transceiver of claim 11, wherein the digital filter controls settings of the controllable I path filter and the controllable Q path filter.

13. 12. The transceiver of claim 11, further comprising: an I path mixer configured to receive the filtered I signal and controlled by the LO; and a Q path mixer configured to receive the filtered Q signal and controlled by the LO, wherein the RF transmit signal is generated based on combining an output of the I path mixer and an output of the Q path mixer.

14. 2. The transceiver of claim 1, wherein the transmitter includes a variable gain amplifier (VGA) configured to amplify the RF transmit signal, and the one or more local observation paths include a first local observation path after the VGA and a second local observation path before the VGA.

15. 1. A method for compensating for local oscillator (LO) leakage in a transceiver, comprising: generating a radio frequency (RF) transmit signal using a transmitter; generating first observation data based on observing the RF transmission signal at one or more local observation paths within the transmitter; generating second observation data using an external loopback path from the transmitter through a front-end system to an observation receiver; and compensating a transmitter for leakage of the transmitter LO based on processing the first observation data and the second observation data using a digital filter.

16. The method of claim 15 , wherein the first observed data indicates differential LO leakage of the LO and the second observed data indicates common mode LO leakage of the LO.

17. The method of claim 15, wherein the digital filter is a Kalman filter.

18. 18. The method of claim 17, further comprising estimating a correction for the LO leakage using a plurality of nonlinear equations, a first portion of the plurality of nonlinear equations being a function of the first observed data, and a second portion of the plurality of nonlinear equations being a function of the second observed data.

19. 1. A radio frequency (RF) communication system comprising: an RF front end; a transmitter configured to provide an RF transmit signal to the RF front end, the transmitter configured to generate first observation data based on observing the RF transmit signal on one or more local observation paths within the transmitter; an observation receiver configured to generate second observation data based on processing an observation signal received from an external loopback path from the transmitter through the RF front end to the observation receiver; a local oscillator (LO) leakage compensation circuit configured to compensate for leakage of the transmitter, the LO leakage compensation circuit comprising a digital filter configured to process the first observation data and the second observation data.

20. 20. The RF communication system of claim 19, wherein the digital filter is a Kalman filter.

Images