Apparatus and method for calibrating mixer offset

This application claims priority under 35 U.S.C. §119(e) from U.S. Provisional Application Ser. No. 60/720,857, filed on Sep. 26, 2005, the contents of which are herein incorporated by reference in their entirety.

1. Technical Field

This disclosure relates generally to electronic circuits and, in particular, to circuits for frequency mixing.

2. Description of the Related Art

Wireless devices are becoming more and more popular as mobile communications and computing devices become ubiquitous. Some wireless devices use radio frequency communications to operate. In such wireless devices, radio frequency upconverters or mixers are important components.

Conventional upconverter or mixer circuits are used to convert a baseband signal to a transmission band signal by mixing the baseband signal with a carrier signal. In a conventional upconverter/mixer circuit, DC offsets of circuitry driving the mixer, as well as a DC offset in an input stage of the mixer, contribute to generation of carrier leakage. This leakage can corrupt the modulation or otherwise lead to non-compliance with specifications for standardized systems such as wireless networks and cell phones.

In order to reduce such offsets, the devices in the circuit are made large to improve matching. This large size leads to large die area, increased costs, and increased power consumption. Furthermore, in many RF mixers, it is not practical to sense the DC offset at the output, as such circuitry would unduly load the circuit.

As a result, there remains a need for an improved mixer circuit.

An embodiment includes a circuit including a subcircuit having differential signals, and a feedback circuit coupled to the subcircuit. The feedback circuit is configured to measure an offset between the differential signals, to generate a calibration signal in response to the measurement, and to reduce the offset in response to the calibration signal.

FIG. 1 is a diagram showing a mixer with offset calibration circuitry according to an embodiment.

FIG. 2 is a diagram showing another example of the feedback circuit of FIG. 1 .

FIG. 3 is a diagram showing another example of the feedback circuit of FIG. 1 .

FIG. 4 is a diagram showing another example of a feedback circuit of FIG. 1 .

FIG. 5 is a diagram showing a circuit for calibrating offsets according to another embodiment.

FIG. 6 is a flowchart showing how a mixer offset is calibrated according to another embodiment.

FIG. 7 is a flowchart showing how a register is updated to calibrate a mixer offset according to another embodiment.

FIG. 8 is a flowchart showing how a mixer offset is calibrated in various modes according to another embodiment.

FIG. 1 is a diagram showing a mixer with offset calibration circuitry according to an embodiment. The mixer includes a first differential pair 60 and a second differential pair 62 . The first differential pair 60 includes transistors Q 9 and Q 10 , and is coupled to node N 7 . The second differential pair 62 includes transistors Q 11 and Q 12 and is coupled to node N 8 . The differential pairs 60 and 62 are cross-coupled to each other. Thus, the collector of Q 9 is coupled to the collector of Q 11 through node N 11 . Similarly, the collector of Q 10 is coupled to the collector of Q 12 through node N 12 . A differential local oscillator input VLO for the mixer is coupled to the transistors Q 9 -Q 12 . Resistor R 5 is coupled to node N 11 . Resistor R 6 is coupled to node N 12 . Nodes N 11 and N 12 form a differential RF output VRF for the mixer.

An input stage 12 is coupled to nodes N 7 and N 8 , and is configured to receive a calibration signal 61 and an input signal 65 . In the input stage 12 , transistors Q 13 and Q 14 are coupled to node N 7 and node N 8 , respectively, and coupled to current source 68 . A modulation input circuit 64 is coupled to transistors Q 13 and Q 14 . The modulation input circuit 64 is configured to combine an input signal 65 and a calibration signal 61 to generate a first output OUT 1 and a second output OUT 2 . The first output OUT 1 and the second output OUT 2 drive the transistors Q 13 and Q 14 , respectively.

The modulation input circuit 64 may combine the input signal 65 and the calibration signal 61 through a variety of circuit structures. For example, the input signal 65 may be a differential input signal. A resistor network may combine the calibration signal with one of the signals of the differential input signal, introducing an offset. Thus, when the differential input signals are applied to the transistors Q 13 and Q 14 as the outputs of the modulation input circuit 64 OUT 1 and OUT 2 , one will have the offset. As a result of the offset introduced by the calibration signal 61 , an offset between currents I 1 and I 2 due to any component or signal prior to nodes N 7 and N 8 may be reduced or eliminated.

In another example, the calibration signal 61 may be applied to both signals of a differential input signal. For example, the calibration signal 61 may be combined with a first signal of the differential signal to generate the first output OUT 1 . An inverted version of the calibration signal 61 may be combined with a second signal of the differential signal to generate the second output OUT 2 . Thus, the calibration signal is applied differentially to transistors Q 13 and Q 14 .

Although using resistors has been described as combining the input and the calibration signal, one skilled in the art will understand that a variety of other circuit structures including components other than or in addition to resistors may be used to combine the calibration signal with the input signal. For example, other components may include capacitors, transistors, inductors, diodes, couplers, or the like.

A feedback circuit 14 is coupled to nodes N 7 and N 8 , and the input stage 12 . The feedback circuit is configured to generate the calibration signal 61 . The feedback circuit 14 includes transistors Q 15 and Q 16 , resistors R 7 and R 8 , and calibration signal generator 66 . Transistor Q 15 is coupled to node N 9 , a calibration mode circuit 16 , and node N 8 . Transistor Q 16 is coupled to node N 10 , the calibration mode circuit 16 , and node N 7 . Resistor R 7 is coupled between node N 9 and the power supply 67 . Resistor R 8 is coupled between node N 10 and the power supply 67 . The calibration signal generator 66 is coupled to nodes N 9 and N 10 , and the modulation input circuit 64 . The calibration signal generator 66 is configured to generate the calibration signal 61 . In one embodiment, the calibration signal generator 66 is configured to generate the calibration signal 61 in response to a first voltage on node N 9 and a second voltage on node N 10 .

A calibration mode circuit 16 may be coupled to the feedback circuit 14 and configured to generate a calibration mode signal 63 . The calibration mode signal 63 may indicate multiple modes of operation, including a calibration mode and an operating mode. The feedback circuit 14 is configured to operate in response to the calibration mode signal 63 . In this example, the feedback circuit 14 is configured to generate the calibration signal 61 based on the first current and the second current when the calibration mode signal indicates a calibration mode. The feedback circuit 14 is also configured to sustain the calibration signal 61 when the calibration mode signal 63 indicates an operating mode.

The transistors Q 13 and Q 14 may, but need not be relatively large devices. Such an increase in size may be used to improve the mismatch between currents I 1 and I 2 . However, such matching would not affect the mismatch introduced by components in the signal path prior to transistors Q 13 and Q 14 . The feedback circuit 14 may still be used to further reduce a DC offset between currents I 1 and I 2 without increasing the size of other devices to improve matching. Thus, even though some devices have an increased size, the resulting chip area is still reduced as compared with a circuit using additional larger devices within the signal path to reduce an offset.

Resistors R 7 and R 8 may be increased in size to improve matching. Furthermore, although an increase in size has been described as improving matching of various components, one skilled in the art will understand that other matching techniques, such as device placement or layout may be used as desired.

An embodiment for calibrating mixer offsets operates as follows. A local oscillator signal is applied to transistors Q 9 - 12 as VLO. In this example, VLO is the carrier for the desired modulation. An input is applied to the modulation input circuit 64 . The modulation input circuit 64 drives transistors Q 13 and Q 14 according to the input signal 65 and the calibration signal 61 . As a result, currents I 1 and I 2 are modulated according to the input signal 65 .

As described above, there may be a DC offset between the currents I 1 and I 2 . In a calibration mode, a calibration mode signal 63 is driven high from the calibration mode circuit 16 . As a result, transistors Q 15 and Q 16 of the feedback circuit 14 shown in FIG. 4 are turned on. At least a portion of the currents I 1 and I 2 are redirected to the feedback circuit 14 . Although in practice, some to all of the currents I 1 and I 2 will flow to the feedback circuit 14 in the calibration mode, the following discussion will describe operations as if all of the currents I 1 and I 2 were redirected to the feedback circuit 14 . One skilled in the art will understand that references to the currents I 1 and I 2 as redirected to the feedback circuit 14 may refer to any portion of the currents.

In this example, when transistors Q 15 and Q 16 are turned on, currents I 1 and I 2 flow through resistors R 7 and R 8 , respectively. As a result voltages are generated on nodes N 9 and N 10 that are proportional to the currents I 1 and I 2 , respectively. Assuming that resistors R 7 and R 8 are approximately equal in resistance, a difference in voltage at nodes N 9 and N 10 will be approximately proportional to a difference in current flowing through transistors Q 15 and Q 16 . Any offset between the voltages on nodes N 9 and N 10 reflect an offset between the currents I 1 and I 2 that may result in carrier leakage on the VRF output of the output stage. The calibration signal generator 66 generates the calibration signal 61 . The calibration signal 61 is applied to the modulation input circuit 64 of the input stage 12 . As a result, the outputs of the modulation input circuit 64 are adjusted, reducing the DC offset between the voltages on the nodes N 9 and N 10 , and consequently, the currents I 1 and I 2 .

Although resistors R 7 and R 8 have been described as being approximately equal in resistance, resistors R 7 and R 8 may be differently sized as desired for detection of an offset between currents I 1 and I 2 . For example, the calibration signal generator 66 may have different input impedances or scale voltages on nodes N 9 and N 10 differently. Thus, resistors R 7 and R 8 may have resistances selected such that the feedback circuit 14 may compare an offset between currents I 1 and I 2 .

Thus, by diverting the currents I 1 and I 2 , a DC offset between the currents I 1 and 12 may be reduced. Such a DC offset may be a result of a variety of factors, including component mismatches, imbalanced signal levels, or the like. However, regardless of the source or relative contribution of particular sources to the DC offset, the DC offset between the currents I 1 and I 2 may be reduced.

FIG. 2 is an example of a successive approximation circuit 36 as the feedback circuit 14 of FIG. 1 . The successive approximation circuit 36 includes a comparator 30 , a successive approximation register (SAR) 32 , and a digital to analog converter (DAC) 34 . The comparator 30 is coupled to the nodes N 7 and N 8 , and the SAR 32 . The SAR 32 is coupled to the DAC 34 . The DAC 34 generates the calibration signal 61 .

In one embodiment, the comparator 30 compares the currents flowing through the nodes N 7 and N 8 . The comparator 30 may include resistors R 7 and R 8 , and transistors Q 15 and Q 16 of FIG. 1 to generate voltages on node N 9 and node N 10 . The comparator 30 then generates a comparison between the two voltages and outputs the comparison to the SAR 32 .

Although components of FIG. 1 have been used to describe potential components and functionality of the comparator 30 of FIG. 2 , other components and comparison techniques may be used. For example, the comparator 30 may be a current comparator, comparing currents I 1 and I 2 , rather than converting the currents to voltages.

In this example, the comparison generated by the comparator 30 is a digital signal indicating which of the currents I 1 or I 2 is greater. With each comparison of the comparator 30 , a bit of the SAR 32 is updated. The updated SAR 32 drives the DAC 34 . The DAC 34 generates the calibration signal 61 . As a result, the offset between the currents I 1 and I 2 is adjusted. The sequence repeats with a new comparison generated using the updated offset until all bits of the SAR 32 are determined. The end result of the SAR 32 and the resulting calibration signal 61 from the DAC 34 is a reduced DC offset between the currents of node N 1 and N 2 .

FIG. 3 is a diagram showing another example of the feedback circuit 14 of FIG. 1 . The feedback circuit 14 illustrated in FIG. 3 includes an analog control circuit 40 and a track and hold circuit 46 . The analog control circuit includes an amplifier 42 and a filter 44 . The amplifier 42 amplifies a difference between the nodes N 7 and N 8 . As described above, such a difference may be a voltage difference or a current difference. The amplified output is filtered in the filter 44 .

In this example, the filter 44 is illustrated as discrete from the amplifier 42 . However, one skilled in the art will understand that the filter 44 may be integrated with the amplifier 42 such that the frequency response of the amplifier 42 forms the desired frequency response of the filter 44 . Furthermore, although a symbol for a low pass filter has been illustrated in FIG. 3 , the filter may be of any shape as desired to achieve a desired response of the feedback circuit 14 .

In another example, the filtered output drives the track and hold circuit 46 . The track and hold circuit 46 is responsive to the calibration mode signal 63 . In a calibration mode, the track and hold circuit 46 is in track mode. As a result, the output of the track and hold circuit 46 tracks the input. Thus, the output of the filter 44 becomes the calibration signal 61 through the track and hold circuit 46 .

After the calibration signal 61 has settled, the track and hold circuit 46 may be put into hold mode. For example, the calibration mode signal 63 may transition to an operating mode and in response, the track and hold circuit 46 enters the hold mode. Thus, a steady state calibration signal 61 that resulted in a minimum difference from signals on nodes N 7 and N 8 during the calibration mode would be maintained in the operating mode.

The circuit may be switched from the operating mode to the calibration mode so that a new calibration signal 61 may be generated to keep the offset at a minimum. For example, if the mixer is used as an upconverting mixer for a wireless device, such a mode switch may occur during a time between transmissions. In between the transmissions in a calibration mode, the calibration signal 61 is generated, reducing DC offsets. During transmissions, the calibration signal 61 is held in the state from the calibration mode, maintaining the reduced DC offsets.

Although an operating mode and a calibration mode have been described above, the modes of the circuit are not limited to such modes. Furthermore, labels of such modes were chosen merely to distinguish in an embodiment when a calibration signal may be modified. Thus, in a calibration mode the circuit may still be operating, and in an operating mode, the circuit may still be calibrated.

FIG. 4 is a diagram showing another example of a feedback circuit of FIG. 1 . The feedback circuit illustrated in FIG. 4 includes a first and second analog to digital converters (ADC) 50 and 52 , a digital signal processor (DSP) 54 , and a DAC 56 . The ADCs 50 and 52 are coupled to the DSP 54 . The DSP 54 is coupled to the DAC 56 . The DAC 56 generates the calibration signal.

The ADCs 50 and 52 sample the nodes N 7 and N 8 . As described above, the signals from nodes N 7 and N 8 may be voltages or currents. ADCs 50 and 52 and any front end circuitry may be appropriately selected to digitize such signals. For example, if the signals from nodes N 7 and N 8 are currents, the ADCs 50 and 52 may be current input ADCs. Regardless of the format of the signals from nodes N 7 and N 8 , the signals are digitized in the respective ADCs 50 and 52 .

The DSP 54 generates a digitized version of the calibration signal 61 . As described above, the calibration signal 61 may be generated in a calibration mode, and maintained in an operating mode. The DSP 54 may have an input for the calibration mode signal 63 . Thus, the DSP 54 may maintain the digitized calibration signal in the operating mode. The DAC 56 generates the calibration signal 61 from the digitized calibration signal.

Using the DSP 54 , any manner of techniques may be used to reduce the offset between currents I 1 and I 2 . For example, the analog control circuit 40 described above may be approximated with a digital control loop implemented within the DSP 54 . Alternatively, the successive approximation circuit 36 , described above, may be implemented in the DSP 54 . Furthermore, any digital control technique may be implemented in the DSP to adjust the digitized calibration signal in response to the digitized signals from the nodes N 7 and N 8 .

Although the ADCs 50 and 52 , and the DAC 56 have been illustrated as separate from the DSP 54 , any combination of the ADCs 50 and 52 , and the DAC 56 may be part of the DSP 54 . Furthermore, the DSP 54 need not be a processor solely designed for digital signal processing. For example the DSP 54 may be a general purpose processor. In addition, the DSP 54 need not be labeled a digital signal processor. Any circuitry that may manipulate digital signals may be used as a DSP 54 . For example, a field programmable gate array (FPGA), a programmable logic device (PLD), or the like may be used as a DSP 54 .

As described above, the calibration signal 61 may be generated during a calibration mode. When the mixer is in an operating mode, the calibration signal 61 may be maintained. Alternatively, the calibration circuitry may be operated continuously. As described above, various implementations of analog and digital filters may be used. The filters may be implemented such that the offset reducing effect of the control loop does not affect the modulation input signal. For example, if a particular modulation input signal does not have significant frequency components below a particular frequency, 10 kilohertz (KHZ) for example, the bandwidth of the control loop may be selected to have a cutoff frequency of less than 10 KHz. Thus, frequency components of the difference of currents I 1 and I 2 having frequencies from DC to the cutoff frequency of the control loop may be reduced. However, frequency components of the modulation input signal contributing to the currents I 1 and I 2 that are higher than the cutoff frequency would not be affected. As a result, the DC offset between currents I 1 and I 2 may be continuously reduced.

FIG. 5 is a diagram showing a circuit for calibrating offsets according to another embodiment. The circuit includes an output stage 10 coupled to a node N 1 and a node N 2 . An input stage 12 is coupled to the node N 1 and the node N 2 . The input stage 12 is configured to receive an input signal 11 . A feedback circuit 14 is coupled to the node N 1 , the node N 2 , and the input stage 12 . The feedback circuit is configured to generate a calibration signal 13 .

Although the components of FIG. 1 have been described as being part of a mixer, other components may form the input stage 12 and output stage 10 . For example, analog multipliers or other implementations or applications of a Gilbert cells, for uses other than mixers, may include nodes N 1 and N 2 where a DC offset between currents I 1 and I 2 would result in an imbalance in the output signal 17 .

In addition, although throughout this disclosure, in reference to current, the terms “from”, “to”, “into”, and other directional terms may be used, one skilled in the art will understand that components may be used in an embodiment such that the direction of the current is reversed. For example, currents may be described in one embodiment as flowing from the input stage 12 . In another embodiment, currents may flow into the input stage 12 . Furthermore, all currents in an embodiment may, but need not be similarly reversed.

FIG. 6 is a flowchart showing calibrating mixer offsets according to another embodiment. An offset internal to the mixer is measured in 80 . The measurement in 80 may be performed using a variety of techniques. For example, the currents may be converted into voltages and the voltages may be compared. Alternatively, the currents may be directly compared. Furthermore, any such current or voltage used for a comparison may be digitized, with the digitized values being compared. As used in this discussion, comparing a first current and a second current for the measurement in 80 may include a digital measurement. Thus, the measurement may generate a digital output have two or more possible states. Alternatively, the first current and the second current in 80 may include an analog measurement. Thus, the measurement may generate an analog output with a continuous spectrum of possible output states.

In response to the measurement, a calibration signal is generated in 82 . The calibration signal may be generated in a variety of ways. For example, as described above, the calibration signal may be generated in 82 by analog techniques, digital techniques, or a combination of such techniques.

The mixer is controlled to reduce the offset using the calibration signal in 84 . To control the offset of the mixer, at least one signal in the input stage may be modified. Such modification may include adding, subtracting, or otherwise combining the calibration signal to the signal in the input stage. The calibration signal may modify only one signal, or may modify multiple signals in the input stage. For example, with a differential input stage, two input signals drive the input stage. Thus, the two input signals, as they exist in the input stage, may be modified by the calibration signal. In particular, for differential signals, a first one of the differential signals may be increased and a second one of the differential signals may be decreased to reduce the offset in the mixer in 84 .

The measuring of the offset in 80 , the generation of the calibration signal in 82 , and the controlling of the mixer in 84 may be repeated to minimize the offset of the mixer.

FIG. 7 is a flowchart showing how a register is updated to calibrate a mixer offset according to another embodiment. In 86 , a value in a register is updated in response to the measurement in 80 . This value is used to generate the calibration signal in 88 . In one example, as described above, the calibration signal may be generated using a successive approximation register (SAR). Thus, in response to the measurement in 80 , a bit of the SAR is updated. The value in the SAR is used to generate the calibration signal.

FIG. 8 is a flowchart showing how a mixer offset is calibrated in various modes according to another embodiment. As described above, the mixer may be in a variety of modes, including a calibration mode, and an operating mode. It is determined if the mode is a calibration mode in 90 . If the mode is the calibration mode, then as described above, the offset is measured in 80 , the calibration signal is generated in 82 , and the mixer is controlled in 84 .

In contrast, if the mode is not the calibration mode, the calibration signal is maintained in 92 . The calibration mode may be maintained using a variety of techniques. For example, as described above, with an analog control loop generating the calibration signal, an analog track and hold circuit may hold the analog calibration signal in order to maintain the calibration signal in 92 . Alternatively, if a digital control loop is used, the digitized value used to generate the calibration signal may be held constant. Furthermore, a combination of analog and digital techniques may be used. For example, the analog calibration signal may be digitized during the calibration mode. In the operating mode, the last digitized calibration signal may be used to generate the calibration signal, thus, maintaining it.

In the calibration mode, a first current and a second current from an input stage of the mixer flow to an output stage of the mixer. To make the measurement of the offset in 80 , at least a portion of the first and second currents may be diverted to a calibration signal feedback circuit in 94 . For example, when a calibration mode is entered, transistors are turned on so that the first and second currents flow through resistors coupled to the transistors. Thus, the first and second currents are diverted in 94 and measured in 80 .

Although one type of transistor has been illustrated in the drawings, the transistors described above can be any type. For example, the transistors may be bipolar (NPN or PNP), metal-oxide-semiconductor MOS (N- or P-channel), or the like.

For purposes of clarity, many of the details of the improved solution and the methods of designing and manufacturing the same that are widely known and are not relevant to the present invention have been omitted from the following description.

Furthermore, it should be appreciated that reference throughout this specification to “one embodiment” or “an example” means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment and may, but need not be included in all embodiments. Therefore, it is emphasized and should be appreciated that two or more references to “an embodiment” or “one embodiment” or “another embodiment” in various portions of this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures or characteristics may be combined as suitable in one or more embodiments of the invention.

Similarly, it should be appreciated that in the foregoing description of embodiments, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the various inventive aspects. This method of disclosure, however, is not to be interpreted as reflecting an intention that more features are required than are expressly recited in each claim. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention. Furthermore, although particular embodiments have been described, one skilled in the art will appreciate that various modifications may be made without departing from the scope of the following claims.