CN104052711B

CN104052711B - Quadrature error correction is carried out using the multinomial model in tone calibration

Info

Publication number: CN104052711B
Application number: CN201410094050.0A
Authority: CN
Inventors: W·安; R·P·舒博特; Y·斯坦
Original assignee: Analog Devices Inc
Current assignee: Analog Devices Inc
Priority date: 2013-03-15
Filing date: 2014-03-14
Publication date: 2018-02-23
Anticipated expiration: 2034-03-14
Also published as: CN104052711A; DE102014103143A1; DE102014103143B4

Abstract

One exemplary embodiment provides a kind of multinomial model using in tone calibration to carry out the systems, devices and methods of quadrature error correction in I/Q receivers.In an exemplary embodiment, there is provided the method for calibrating I/Q receivers, and methods described includes：Unmatched first received between the I passages and Q passages that indicate the I/Q receivers mismatches parameter；With using multinomial model parameter is mismatched to mismatch parameter Estimation second from described first.

Description

Quadrature error correction using polynomial models in pitch calibration

Priority data

This application claims priority to provisional patent application serial No.61/798,197, filed 2013, 3, 15, which is incorporated herein by reference in its entirety.

Technical Field

The present disclosure relates generally to in-phase and quadrature (I/Q) receiver signal processing, and more particularly to an apparatus, method, and system for correcting gain and phase imbalances in an I/Q receiver.

Background

Quadrature modulation is a technique for transmitting communication signals. In quadrature modulation, a transmitter transmits data on both an in-phase (I) channel and a quadrature-phase (Q) channel. Each of the I and Q channels carries a separate data stream and is phase shifted by 90 degrees relative to each other in carrier frequency. In an I/Q receiver, I and Q channel signals are received, down-converted and demodulated at a carrier frequency to recover data from the separate I and Q channels.

The I/Q receiver includes a separate analog processing path for each of the I and Q channels. Each of the I and Q channel paths includes processing components such as, for example, mixers, analog-to-digital converters (ADCs), amplifiers, and filters that process the received analog signals and convert them to digital form. These separate components down-convert and process the channel signal data for each path. The use of separate I and Q analog processing paths in an I/Q receiver creates so-called I/Q imbalance, which is a phase and amplitude mismatch between the I and Q channel signals. A major source of I/Q imbalance in an I/Q receiver is mismatch between individual components in each of the I and Q channel processing paths.

There are two types of I/Q imbalances: frequency independent mismatches and frequency dependent mismatches. The frequency independent mismatch is a phase mismatch generated by a Local Oscillator (LO) in the mixer of each I/Q analog path. These mismatches are spectrally similar and can be further successfully estimated for error correction. The frequency dependent mismatch is a phase and amplitude mismatch generated by baseband components in each of the I and Q channel analog paths. Frequency-dependent mismatch complicates the mismatch estimation technique because the mismatch is spectrally non-uniform.

Techniques have been developed to estimate mismatch parameters of the analog path of an I/Q receiver. These techniques include real-time estimation, where during real-time use of the receiver, the receiver estimates mismatch parameters and compensates for mismatch. The techniques also include off-line estimation, where the receiver is not in use when a mismatch estimation is performed and the correction parameters are input to the receiver. The off-line estimation method may use a tone injection technique to inject test tones into the receiver at the input of the I and Q channels, thereby using a tone calibration method to calibrate for mismatches between the channels. The techniques have proven to be most effective in low noise environments and further when the jitter of the test tone signal is less than a certain level. The effectiveness of these known techniques decreases in high noise level environments or when the test tone signal jitter rises above a certain level. Therefore, a method for estimating I/Q channel mismatch, having improved effect in noisy environments and/or in case of high jitter levels of the test tone signal would provide advantages.

Disclosure of Invention

The present disclosure provides an apparatus, method and system for calibrating I/Q imbalance in an I/Q receiver. In an embodiment according to the present disclosure, a polynomial model is applied to the results of the tone calibration to generate mismatch parameters for calibrating the I/Q receiver. One exemplary embodiment provides a system, apparatus and method for quadrature error correction in an I/Q receiver using a polynomial model in pitch calibration. In one exemplary implementation, a method for calibrating an I/Q receiver is provided and includes: receiving a first mismatch parameter indicating a mismatch between an I channel and a Q channel of an I/Q receiver; and estimating a second mismatch parameter from the first mismatch parameter using the polynomial model.

Initially, tone calibration is used to generate an overall mismatch estimate for the I and Q channels of an I/Q receiver. The results of the pitch calibration can then be processed using the polynomial calibration method of the embodiment to generate frequency-dependent and frequency-independent mismatch parameter estimates. The estimated values of the frequency dependent and frequency independent mismatch parameters generated by the polynomial calibration method can then be used to process the signals in a Quadrature Error Correction (QEC) unit to compensate for the mismatch. One embodiment of the present disclosure results in values of gain and phase mismatch that provide an improved and more accurate calibration process by using a polynomial approach. The method shows a strong noise robustness, since the unknown parameters for the polynomial calibration estimation are significantly reduced compared to the known techniques.

According to another embodiment of the present disclosure, a polynomial model is applied to the results of the tone calibration to generate mismatch parameters for calibrating the I/Q receiver. Initially, tone calibration is used to estimate the total gain and total phase mismatch between the I and Q channels of an I/Q receiver. The results of the pitch calibration are then processed with a polynomial calibration method to generate frequency dependent gain and phase mismatch and frequency independent phase mismatch estimates. The estimated values of the mismatch parameters generated by the polynomial calibration method can then be used to process the signals in a Quadrature Error Correction (QEC) unit to compensate for the mismatch.

According to another embodiment of the present disclosure, a polynomial model is applied to the pitch-calibrated total gain and total phase estimation results to estimate the frequency-dependent gain and phase mismatch and the frequency-independent phase mismatch of the I and Q channels in the RF receiver. When the mismatch is estimated instead of evaluating the mismatch of the frequency pairs by assuming that the mismatch at symmetric positive and negative frequencies is the same, the polynomial model is applied over all frequency pairs in the spectrum. In one exemplary implementation, the positive and negative frequencies are treated as having separate mismatches. The estimates of the frequency dependent gain and phase mismatch and the frequency independent phase mismatch can then be used to process the signals in the error correction unit to compensate for the mismatch.

According to another embodiment of the present disclosure, a polynomial model is applied to tone calibration to estimate frequency-dependent and frequency-independent mismatches between the I and Q channels in an RF receiver. The polynomial model is applied to the pitch aligned total gain and total phase mismatch estimates using Least Squares Error (LSE) optimization to more accurately estimate the frequency dependent gain and phase mismatch and the frequency independent phase mismatch. By assuming that symmetric positive and negative frequencies have the same mismatch, the polynomial model is applied in common across all frequency pairs in the spectrum, rather than evaluating the mismatch individually for each frequency pair.

In accordance with yet another embodiment, the present disclosure provides a calibration system that includes a Polynomial Calibration (PCAL) estimator, a pitch calibration (TCAL) estimator, a pitch generator, and a transform processor. In one exemplary implementation, the calibration system is used to test the I/Q receiver off-line. The TCAL estimator controls the tone generator to input the test tone at a frequency value in the test spectrum into the analog path of the I/Q receiver. The I/Q receiver includes a mixer section having a Local Oscillator (LO) and a mixer, and I and Q paths having an analog-to-digital converter (ADC). The tones are transmitted through the receiver analog path. After passing through the receiver path, the resulting time domain signal is transformed from the time domain to the frequency domain by a transform processor.

The TCAL estimator then operates on the frequency domain signal to generate a mismatch g (f) in overall gain between the I and Q channels_T) Estimate and total phase mismatchAnd (6) estimating. The results of the TCAL tone calibration estimator are then processed by the PCAL estimator. The PCAL estimator uses a polynomial method and a Least Squares (LSE) method to estimate the frequency-dependent gain mismatch g (n) of the modeled magnitude mismatch profile. Polynomial coefficients are estimated and used to calculate the magnitude g (n) for each band using a polynomial. The PAL estimator then uses a polynomial approach to estimate the frequency-independent phase mismatchFrequency dependent phase mismatchThe frequency dependent gain and phase mismatch parameters and the frequency independent phase mismatch parameters are then input to a Quadrature Error Correction (QEC) processor, and the QEC processor can compensate and correct for mismatches in the I/Q receiver.

Drawings

FIG. 1 illustrates an exemplary simplified functional block diagram of a test system according to an example of the present disclosure;

FIG. 2 illustrates exemplary process operations according to embodiments of the present disclosure;

FIG. 3A illustrates exemplary process operations according to embodiments of the present disclosure;

FIG. 3B illustrates exemplary process operations according to embodiments of the present disclosure;

FIG. 4A shows an exemplary graph of performance of an embodiment of the present disclosure at a first noise level;

FIG. 4B shows an exemplary graph of performance of an embodiment of the present disclosure at a second noise level;

FIG. 4C illustrates an exemplary graph of performance of an embodiment of the present disclosure at a third noise level; and

fig. 4D shows an exemplary graph of the performance of an embodiment of the present disclosure at a fourth noise level.

Detailed Description

Referring now to FIG. 1, a simplified functional block diagram of a test system 100 is shown, according to one exemplary embodiment of the present disclosure. The test system 100 includes a test device 101 and a receiver 102. The test apparatus 101 includes a tone generator 104, a transform processor 106, a Tone Calibration (TCAL) estimator 108, and a Polynomial Calibration (PCAL) estimator 110. In fig. 1, the receiver 102 represents a receiver to be calibrated using the test apparatus 101. Receiver 102 includes a Low Noise Amplifier (LNA) 112, a mixer 114, a digital processor block 118, and a Quadrature Error Correction (QEC) processor block 120. Mixer 114 includes a Local Oscillator (LO) 126 and multipliers 128a and 128 b. The analog path of receiver 102 includes separate I-channel and Q-channel processing paths for received in-phase (I) and quadrature-phase (Q) signals. The analog path includes an I-channel analog path through multiplier 128a, transimpedance amplifier (TIA) 124a, and analog-to-digital converter (ADC) 122a, and a Q-channel analog path through multiplier 128b, transimpedance amplifier (TIA) 124b, and ADC122 b. In one exemplary implementation of fig. 1, the test device 101 is used to calibrate the receiver 102 in an offline mode, i.e., the receiver 102 is considered offline and inoperable during calibration. The test device 101 may be implemented in various combinations of hardware and software. For example, a Digital Signal Processor (DSP), Application Specific Integrated Circuit (ASIC), or other suitable software-controlled circuitry may be used to implement the functions of the implementations.

In one exemplary implementation of fig. 1, the PCAL estimator 110 provides an estimate of the mismatch between the I and Q paths of the receiver 102 for the QEC processor 120. The PCAL estimator 110 receives estimates of the total gain mismatch and the total phase mismatch between the I and Q channels from the TCAL estimator 108. The PCAL estimator 110 applies a polynomial model and Least Square Error (LSE) optimization to the total gain and total frequency mismatches to generate mismatch parameter estimates for the frequency dependent gain mismatch, the frequency dependent phase mismatch and the frequency independent phase mismatch. The estimated mismatch parameters in the PCAL estimator 110 may then be used in the QEC processor block 120 to calibrate the receiver 102 to compensate for gain and frequency mismatch.

Referring now to fig. 2, therein is shown process operations according to an exemplary embodiment of the present disclosure. These operations are performed in hardware and software implementing the functional blocks shown in the exemplary embodiment of fig. 1. The test process begins at process operation 200 when an RF test tone is input to mixer 114 by tone generator 104. Each test tone is on a frequency from a set of frequencies ranging over the desired test spectrum. At mixer 114, the test tone is multiplied by the in-phase wave cos (2 π f) at multipliers 128a and 128b, respectively_ct) and orthogonal wave-sin (2 π f)_ct). The test tone can be modeled as a signal z (t):

z(t)＝2A·cos(2πf_ct+2πf_Tt+θ)＝2A·(cos(2πf_ct)cos(2πf_Tt+θ)-sin(2πf_ct)sin(2πf_Tt+θ))，

where 2A is the amplitude of the tone (here, the amplitude of the tone)Convenient to calculate, the factor 2), f is added to the amplitude_cIs the carrier frequency, f_TIs the tone frequency and θ is the relative phase of the tone with respect to LO126 in mixer 114.

The results are passed through low pass filters 124a and 124b, respectively, to each of the I and Q channels of the analog path. The orthogonal waves have magnitude and phase deviations, respectively denoted as g_LOAndthen, the orthogonal wave becomesFor the I channel, the signal is processed as:

similarly, in the Q channel, the signal is:

nominal baseband channel on tone versus frequency f_TIs given by the quantity g_n(f_T) Mismatch and phase shiftThere is no match in the composition. These mismatches are dependent on the pitch frequency. Processing assumes that the I signal passes through the nominal baseband channel, then:

again, the process assumes that baseband mismatch applies only to the Q path. The channel gain and phase shift in the Q path is g_n(f_T)g_m(f_T) Andwherein g is_m(f_T) Andis the baseband gain and phase mismatch. Then, the Q-path signal becomes:

signal y_I(t) and y_Q(t) has a magnitude and phase mismatch between the I and Q paths defined as:

using a common factor:

A(f_T)＝g_n(f_T) A and

baseband I/Q signal y at the end of the analog path and at the input of digital block 118_I(t) and y_Q(t) is given as (a) in the following,

y_I(t)＝A(f_T)·cos(2πf_Tt+θ(f_T))

the baseband signal is then processed by the test apparatus 101 to estimate mismatch parameters for the I and Q channels.

In process operation 202, the transform processor 106 applies a transform to each of the baseband I and Q signals, respectively, to transform each baseband signal from the time domain to the frequency domain. In one exemplary implementation of fig. 1, the transform processor 106 may be capable of performing any one of a Fast Fourier Transform (FFT) or a Discrete Fourier Transform (DFT) as a test alternative. In alternative embodiments, the transform processor 106 may use one or both of an FFT or DFT, or another transform. Either the FFT or DFT may generate a transform result that may be input to the TCAL estimator 108. The FFT and DFT methods can avoid interference from co-existing harmonics. The difference between the two is that the FFT computes frequency points across the entire spectrum, while the DFT computes selected points around the tone frequencies. The latter may be much less computationally complex, especially as the signal length increases. Both methods may follow the same mathematical formula as will be shown herein for the exemplary embodiment of fig. 1.

The transform processor 106 uses a DFT (or FFT) to compute frequency bins by correlating the signal with complex harmonics on the frequency. Applying the transform to the I signal yields:

similarly, applying a transform to the Q signal yields:

the transform processor 106Signal Y converted from rear handle_Q(f_T) And Y_I(f_T) Output to the TCL estimator 108. The TCAL estimator 108 then estimates the total gain and total frequency mismatch parameters using the frequency domain transformed signal at process operation 204. The TCAL estimator 108 measures Y_Q(f_T) Divided by Y_I(f_T) To obtain:

wherein H_D(f_T) Is baseband mismatch andis LO mismatch and the total gain mismatch is given by:

and the total phase mismatch is given by:

the estimated total phase mismatch is the phase mismatch of the combination of LO126 and the baseband analog path, and is given asTwo components are dropped from the total phase mismatch to perform quadrature error correction in the I/Q receiver 102. At this point, the TCAL estimator 108 has information to compute estimates of frequency dependent gain and phase mismatch and frequency independent phase mismatch. An alternative way to perform the calculation would be to assume g (f)_T) Symmetrical between positive and negative frequencies and assuming odd phase mismatch with zero frequency, i.e.This assumption can be used because the impulse response for baseband mismatch is real valued. The LO phase mismatch over the bandwidth of the device swept by the test tone may then be calculated. In addition, the frequency dependent baseband phase mismatch can then also be estimated as the difference between the total phase mismatch and the LO phase mismatch. However, using this assumption does not allow a completely accurate estimation of the mismatch of the receiver, especially in high noise environments or when there is a high jitter level in the test tones.

In one exemplary embodiment, the test device 101 (including the PCAL estimator 110) provides g (f) by using a polynomial approach in common across all frequency pairs to compute mismatch estimates from the total gain and phase mismatch,Andimproved estimation of (2). The frequency dependent I/Q imbalance is mainly caused by small differences in poles and zeros of the transfer functions of the I-channel and Q-channel paths. Thus, the frequency domain mismatch appears as a smoother curve across the spectrum. This property makes it possible to model the mismatch curve with a polynomial function. In the polynomial approach, the key factor in the mismatch parameter estimation is the estimation of the polynomial coefficients.

Referring again to fig. 2, at process operation 206, the PCAL estimator 110 receives the total gain and frequency mismatch parameter estimates from the TCAL estimator 108. The PCAL estimator 110 then jointly estimates the frequency-dependent gain and phase mismatch parameters and the frequency-independent phase mismatch parameters over the frequency pairs (including positive and negative frequencies) of the test spectrum.

Referring to fig. 3A, there is shown process operations performed by the PCAL estimator 110 during the process of operation 206 of fig. 2. At process operation 302, the PCAL estimator 110 receives the total gain and total phase mismatch estimates from the TCAL estimator 108. In one exemplary implementation of the test device 101, N tones are generated over the test spectrum. The PCAL estimator 110 evaluates the total mismatch parameter for each of the N tones to give the following values for the total gain Q (N) mismatch and the total phase W (N) mismatch:

and,

in step 304, the PCAL estimator 110 applies a polynomial model to the gain parameters. In one exemplary implementation, a fourth order polynomial may be used to model the frequency dependent gain mismatch profile g (n). In alternative embodiments, the order of the polynomial used may be of any order, for example three, and is selected based on simulation or any other suitable method suitable for selection. The polynomial model for g (n) is:

S(n)＝P₀+P₁n+P₂n²+P₃n³+P₄n⁴

there are N frequency points to provide N linear equations to estimate the 5 unknown variables. For positive frequencies, the system of equations is:

for negative frequencies, the exponent is modified from the transform, e.g.:

under the matrix format, the linear system becomes:

K·P＝Q

wherein,

P＝[P₀P₁P₂P₃P₄]^T

Q＝[Q(0) ... Q(N-1)]^T

the PCAL estimator 110 then applies the least squares error method and minimizes the matrix, step 306:

J＝(K·P-Q)^T·(K·P-Q)

＝P^TK^TKP-P^TK^TQ-Q^TKP+Q^TQ

the PCAL estimator 110 calculates the derivative and makes it equal to zero, yielding:

P＝(K^TK)^-1K^TQ

the PCAL estimator 110 then calculates a frequency-dependent gain mismatch g (n) at process operation 308. The PCAL estimator 110 uses the estimated polynomial coefficients and uses a polynomial model to calculate the magnitude of the frequency dependent gain mismatch g (n) over each frequency band.

Referring now to fig. 3B, shown therein are process operations performed by the PCAL estimator 110 to estimate frequency-dependent and frequency-independent phase mismatches according to embodiments of the present disclosure. FIG. 3B illustrates estimating frequency dependent phase mismatchFrequency independent phase mismatchThe operation of (2). In process operation 312 of fig. 3B, the PCAL estimator 110 operates on the total phase mismatch parameter W (n) received from the TCAL estimator 108 at step 302.

At process operation 312, the PCAL estimator 110 uses the relational expressionAnd to phase the basebandModeled as a polynomial. The PCAL estimator 110 also uses a relational expressionA fourth order polynomial may be used for the embodiment of fig. 1. The selection of the polynomial may be based on simulations or other selection methods that determine that a particular polynomial is sufficient for use. The PCAL estimator 110Modeling is as follows:

this gives an array of linear equations as:

the PCAL estimator 110 uses 5 unknowns andequation set to solve coefficient and frequency independent phase mismatchAt process operation 314, the PCAL estimator 110 applies a least squares error method for linear regression. The system of equations used can be expressed in a matrix format as:

H·C＝W

wherein,

W＝(W(1) ... W(N/2-1)]^T

least squares error method minimizes J:

J＝(H·C-W)^T·(H·C-W)

＝C^TH^THC-C^TH^TW-W^THC+W^TW

then, the derivative can be taken and made equal to zero, yielding:

C＝(H^ＴH)^-1H^TW

at process operation 315, the PCAL estimator computes values of polynomial coefficients, includingThe value of (c). The PCAL estimator 110 now has frequency-independent phase mismatch parametersThe value of (c).

At process operation 316, the PCAL estimator 110 then estimates a frequency-dependent phase mismatch using the polynomial coefficientsThe value of (a), which can be estimated from the polynomial model using the available coefficient vector C. The PCAL estimator now has a frequency dependent gain mismatch parameter g (n), a frequency dependent phase mismatch parameterFrequency independent phase mismatch parameterThe value of (c).

Referring again to fig. 2, at process operation 208, the PCAL estimator 110 then outputs the estimated frequency dependent gain and phase mismatch parameters and the frequency independent phase mismatch parameter values to the quadrature error correction block 120. The QEC function block may then apply the PCAL estimate to error compensation and correction in the receiver 102.

Referring now to fig. 4A-4B, shown therein are mirror rejection ratio (IRR) graphs showing improved computer simulations that may be provided by the PCAL estimator according to the exemplary embodiment of fig. 1. In fig. 4A-4D, PCAL calibration using a polynomial approach is compared with results given by TCAL calibration gain and phase mismatch parameter estimates without using a polynomial estimation approach (i.e., by assuming that the gain mismatch is symmetric between positive and negative frequencies and assuming that the phase mismatch is odd symmetric with zero frequency).

Fig. 4A shows the improvement in the noise-free case, fig. 4B shows the improvement at 51dB, fig. 4C shows the improvement at 41dB and fig. 4D shows the improvement at 31 dB. It can be seen that as the noise gradually worsens, the PCAL plot is spectrally smoother from the noiseless of fig. 4A to 51dB of fig. 4B, 41dB of fig. 4C, and 31dB of fig. 4D than without the polynomial approach. Simulations show similar performance at low noise. As noise increases, the better performance of PCAL calibration becomes significant. In certain embodiments, as noise increases, PCAL using the polynomial method is estimated at least 10dB better than TCAL without the polynomial method. The method shows strong noise robustness because the unknown estimation parameters are significantly reduced.

In the discussion of the above embodiments, capacitors, buffers, interconnect boards, clocks, tone generators, processors, TCALs, PCALs, receivers, LNAs, mixers, digital processor blocks, QECs, LOs, dividers, inductors, resistors, amplifiers, switches, digital cores, transistors, and/or other components may be readily replaced, substituted, or otherwise modified to suit particular circuit needs. Additionally, it should be noted that the use of complementary electronics, hardware, non-transitory software, etc. provides an equally viable option for implementing the teachings of the present disclosure.

While this disclosure has described at least one illustrative embodiment, various alterations, modifications, and improvements are possible. Such alterations, modifications, and improvements are intended to be within the spirit and scope of the disclosure. Accordingly, the foregoing disclosure is by way of example only and is not intended as limiting. Any number of the functional blocks of the exemplary embodiments may be implemented on a circuit board of an associated electronic device. The circuit board may be a generic circuit board that can hold various components of the internal electronic system of the electronic device and further provide connectors for other peripheral devices. More specifically, the circuit board may provide electrical connections through which other components of the system may electrically communicate. Any suitable processor (including digital signal processors, microprocessors, companion chipsets, and the like), memory elements, and the like may be suitably coupled to the circuit board based on particular configuration needs, processing needs, computer design, and the like. Other components, such as external storage, additional sensors, controllers for audio/video displays, and peripherals, may be connected to the circuit board as a paddle card, connected by a cable, or integrated into the circuit board itself.

In another exemplary implementation, the functions may be implemented as stand-alone modules (e.g., devices having associated components and circuitry configured to perform specific applications or functions) or as plug-in modules into dedicated hardware of an electronic device. Note that certain embodiments of the present disclosure may be readily included, in part or in whole, in a system-on-a-chip (SOC) package. An SOC represents an IC that integrates components of a computer or other electronic system into a single chip. It may contain digital, analog, mixed signal and common radio frequency functions: all of these may be provided on a single chip substrate. Other embodiments may include multi-chip modules (MCMs) in which multiple individual ICs are located in a single electronic package and configured to interact closely with each other through the electronic package. In various other embodiments, the amplification function may be implemented in one or more silicon cores in Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs), and other semiconductor chips.

It should also be noted that all of the described functions and processes are provided for purposes of example and teaching only. The information may vary widely without departing from the spirit of the present disclosure or the scope of the appended claims. The description applies only to one non-limiting example and they should therefore be understood as such. In the description above, exemplary embodiments have been described with reference to specific processors and functional arrangements. Various modifications and changes may be made to such embodiments without departing from the scope of the appended claims. The specification and drawings are, accordingly, to be regarded in an illustrative sense rather than a restrictive sense.

Note that in several examples provided herein, interaction may be described in terms of two, three, four, or more electronic elements. However, this is done for clarity and example purposes only. It should be appreciated that the systems may be combined in any suitable manner. Along similar lines of design, any of the illustrated components, modules, and elements of the drawings may be combined in various possible configurations, all of which are apparent to be within the broad scope of this specification. In some cases, one or more of the functions of a given set of flows may be readily described by reference to only a limited number of elements. It should be understood that the drawings and their teachings are readily scalable and can accommodate a greater number of components and more complex/sophisticated arrangements and configurations. Accordingly, the examples provided should not limit the scope of the disclosure or inhibit the broad teachings of the disclosure as potentially applied to myriad other structures.

Note that in this specification, references to various features (e.g., elements, structures, modules, components, steps, operations, features, etc.) included in "one embodiment", "an exemplary embodiment", "an embodiment", "another embodiment", "some embodiments", "various embodiments", "other embodiments", "alternative embodiments", etc., are intended to mean that any such features are included in one or more embodiments of the present disclosure, but may or may not be combined in the same embodiment.

Numerous other changes, substitutions, variations, alterations, and modifications may be ascertained to one skilled in the art and it is intended that the present disclosure encompass all such changes, substitutions, variations, alterations, and modifications as falling within the scope of the appended claims. To assist the U.S. patent and trademark office (USPTO), and to otherwise assist any reader of any patent issued to this application, in interpreting the appended claims, applicants wish to note that applicants: (a) it is not intended that any appended claims be presented at the filing date to this application as an aid to 35u.s.c. article 112, paragraph six (6), unless the phrase "means for … …" or "step for … …" is explicitly used in a particular claim; and (b) is not intended to limit the disclosure in any way by any statement in this specification that is not otherwise embodied in the appended claims.

Claims

1. A method for calibrating an I/Q receiver, comprising:

receiving a first mismatch parameter indicating a mismatch between an I channel and a Q channel of the I/Q receiver; and

a polynomial model is used to estimate a second mismatch parameter from the first mismatch parameter,

wherein the first mismatch parameters include an overall gain mismatch and an overall phase mismatch, and the second mismatch parameters include a frequency dependent gain mismatch, a frequency dependent phase mismatch, and a frequency independent phase mismatch.

2. The method of claim 1, wherein the estimating comprises: applying a third order polynomial to estimate the second mismatch parameter.

3. The method of claim 2, wherein the estimating comprises: applying a fourth order polynomial to estimate the second mismatch parameter.

4. The method of claim 1, wherein the first mismatch parameter comprises a first plurality of frequency-dependent mismatch parameters at frequencies on a test spectrum, and wherein the estimating comprises: a second plurality of frequency-dependent mismatch parameters is estimated over negative and positive frequency pairs in the test spectrum.

5. The method of claim 1, wherein the estimating comprises: the matrix is minimized using a least squares error, LSE, calculation.

6. The method of claim 1, further comprising:

performing tone calibration by generating N tone signals over a test spectrum;

inputting said tone signal to said I/Q receiver;

performing a pitch calibration in a pitch calibration estimator to generate the first mismatch parameter; and

the first mismatch parameter is input to a polynomial calibration estimator.

7. The method of claim 6, wherein said estimating a second mismatch parameter comprises: a gain mismatch parameter and a phase mismatch parameter for each frequency of the tone signal are estimated.

8. The method of claim 6, wherein said estimating a second mismatch parameter comprises: estimating parameter mismatch on negative and positive frequency pairs in the test spectrum.

9. An apparatus for calibrating an I/Q receiver, the apparatus comprising:

a polynomial estimator configured to:

receiving a first mismatch parameter, wherein the first mismatch parameter indicates a mismatch between an I-channel and a Q-channel of the I/Q receiver; and

wherein the first mismatch parameters received by the polynomial estimator include total gain mismatch and total phase mismatch, and the second mismatch parameters include frequency dependent gain mismatch, frequency dependent phase mismatch, and frequency independent phase mismatch.

10. The apparatus of claim 9, wherein the polynomial estimator estimates the second mismatch parameter from the first mismatch parameter by applying a third order polynomial.

11. The apparatus of claim 10, wherein the polynomial estimator estimates the second mismatch parameter from the first mismatch parameter by applying a fourth order polynomial.

12. The apparatus of claim 9, wherein the first mismatch parameter comprises a first plurality of frequency-dependent mismatch parameters at frequencies of a test spectrum, and wherein the polynomial estimator estimates a second plurality of frequency-dependent mismatch parameters over negative and positive frequency pairs of the test spectrum.

13. The apparatus of claim 9, wherein the polynomial estimator estimates the second mismatch parameter using a polynomial with a matrix and uses Least Squares Error (LSE) calculation.

14. The apparatus of claim 9, further comprising:

a tone generator configured to generate N tone signals over a test spectrum and input the tone signals into an analog path of the I/Q receiver;

a transform block coupled to an end of an I/Q signal analog path, the transform block to transform a signal received on the end of the I/Q signal analog path from a time-domain signal to a frequency-domain signal and output the frequency-domain signal; and

a pitch calibration estimator coupled to the transform block and the polynomial estimator, the pitch calibration estimator configured to receive the frequency domain signal from the transform block, perform a pitch calibration on a frequency domain test signal to generate the first mismatch parameter, and input the first mismatch parameter to the polynomial estimator.

15. The apparatus of claim 14, wherein the second mismatch parameter comprises an overall gain mismatch parameter and an overall phase mismatch parameter for each frequency of the N tone signals.

16. The apparatus of claim 14, wherein the first mismatch parameter comprises a first plurality of frequency-dependent mismatch parameters at frequencies on the test spectrum, and wherein the polynomial estimator estimates a second frequency-dependent mismatch parameter over a negative and positive frequency pair in the test spectrum.

17. An apparatus for calibrating an I/Q receiver, comprising:

a tone calibration estimator configured to inject a test tone at a frequency within a test spectrum into the I/Q receiver, to estimate a first set of mismatch parameters including an overall gain mismatch parameter and an overall phase mismatch parameter, and to output the first set of mismatch parameters;

a polynomial calibration estimator coupled to the pitch calibration estimator, the polynomial calibration estimator configured to receive the first set of mismatch parameters from the pitch calibration estimator and estimate a second set of mismatch parameters from the first set of mismatch parameters using a polynomial model, the second set of mismatch parameters comprising a frequency-dependent gain mismatch parameter, a frequency-dependent phase mismatch parameter, and a frequency-independent phase mismatch parameter.

18. The apparatus of claim 17, wherein the polynomial calibration estimator estimates the second set of mismatch parameters over negative and positive frequency pairs in the test spectrum.