Transmitter Circuit, Compensation Value Correction Device, and Compensation Value Correction Method

Abstract

A compensation value correction method for correcting one or more compensation values used by a compensation device of a transmitter, comprising: obtaining a plurality of output signals generated by the transmitter processing a set of input signals according to multiple sets of compensation values in sequence as multiple feedback signals, wherein each feedback signal corresponds to one of the multiple sets of compensation values; determining multiple coefficients of a cost function according to the multiple sets of compensation values and the feedback signals in a correction operation; and determining a set of corrected compensation values according to the coefficients and providing the set of corrected compensation values to the compensation device.

Description

Technical Field

[0001] This invention application relates to a compensation value correction method applicable to a transmitter circuit, so as to reduce the radio frequency impairment of the transmitter circuit. Background Art

[0002] The design of a zero intermediate frequency (Zero-IF) transmitter or a zero intermediate frequency receiver faces basic radio frequency impairment problems because it adopts the technology of direct-up conversion, that is, it converts the baseband signal to radio frequency using only one mixing stage, or directly converts the received radio frequency signal to baseband signal using only one mixing stage. These problems may include in-phase and quadrature-phase (IQ) imbalance and local oscillation (LO) leakage. The reason for the formation of IQ imbalance is that when the responses of the in-phase channel and the quadrature channel are different, the two signals passing through the in-phase channel and the quadrature channel will have unequal amplitude or phase gains. And local oscillation leakage is the interference caused by the local oscillation signal remaining in the output signal.

[0003] Since the causes of IQ imbalance and local oscillation leakage are different, the compensation methods and parameters to be corrected for solving IQ imbalance and local oscillation leakage are different. In order to efficiently find the best parameters to solve radio frequency impairment problems such as IQ imbalance and local oscillation leakage, a compensation value correction method that can be integrally applied to solve different radio frequency impairment problems is needed to efficiently obtain the corresponding compensation value. Summary of the Invention

[0004] An object of the present invention is to provide a compensation value correction method that can be integrally applied to solve different radio frequency impairment problems, so as to efficiently obtain the corresponding compensation value.

[0005] According to an embodiment of the present invention, a transmitter circuit includes a compensation device, at least one transmission signal processing device, and a compensation value correction device. The compensation device is arranged on a transmission signal processing path, and is used to sequentially process a group of input signals according to multiple groups of compensation values to generate multiple groups of compensated signals. At least one transmission signal processing device is arranged on the transmission signal processing path, and is used to sequentially process the multiple groups of compensated signals to generate multiple output signals. The compensation value correction device is coupled to an output end of the transmission signal processing path, and is used to sequentially receive the output signals from the output end as multiple feedback signals, and perform a correction operation according to the feedback signals and the multiple groups of compensation values. The compensation value correction device includes a digital signal processor, which is coupled to the compensation device. In the correction operation, the digital signal processor determines multiple coefficients of a cost function according to the multiple groups of compensation values and the feedback signals, determines a group of corrected compensation values according to the coefficients, and provides the group of corrected compensation values to the compensation device.

[0006] According to another embodiment of the present invention, a compensation value correction device for correcting one or more compensation values used by a compensation device of a transmitter includes a power spectral density estimation device and a digital signal processor. The power spectral density estimation device is used to perform a fast Fourier transform on multiple feedback signals received from the transmitter to generate feedback signals in the frequency domain, and estimate the distribution of the power of the feedback signals in the frequency domain, where each feedback signal corresponds to one of the multiple groups of compensation values respectively. The digital signal processor is coupled to the power spectral density estimation device and the compensation device, and is used to determine multiple coefficients of a cost function according to the multiple groups of compensation values and the feedback signals in a correction operation, determine a group of corrected compensation values according to the coefficients, and provide the group of corrected compensation values to the compensation device.

[0007] According to another embodiment of the present invention, a compensation value correction method for correcting one or more compensation values used by a compensation device of a transmitter includes: obtaining multiple output signals generated by the transmitter sequentially processing a group of input signals according to multiple groups of compensation values as multiple feedback signals, where each feedback signal corresponds to one of the multiple groups of compensation values respectively; determining multiple coefficients of a cost function according to the multiple groups of compensation values and the feedback signals in a correction operation; and determining a group of corrected compensation values according to the coefficients, and providing the group of corrected compensation values to the compensation device. Description of the Drawings

[0008] Figure 1 The transmitter circuit according to an embodiment of the present invention is shown.

[0009] Figure 2 A spectrum example of a radio frequency signal is shown.

[0010] Figure 3 A spectrum example of the feedback signal is shown.

[0011] Figure 4 Shows a graph of the IQ imbalance compensation value versus the energy of the feedback signal at a given frequency.

[0012] Figure 5 Shows a transmitter circuit according to another embodiment of the present invention.

[0013] Figure 6 Shows an exemplary spectrum of a radio frequency signal.

[0014] Figure 7 Shows an exemplary spectrum of the feedback signal.

[0015] Figure 8 Shows a graph of the local oscillator signal leakage compensation value versus the energy of the feedback signal at a given frequency.

[0016] Figure 9 Shows a flowchart of a compensation value correction method according to an embodiment of the present invention.

[0017] Figure 10 Shows a transmitter circuit according to yet another embodiment of the present invention.

[0018] Symbol description:

[0019] 100, 300, 500: Transmitter circuit

[0020] 110, 310, 510: Transmitted signal processing path

[0021] 120, 320, 520: Feedback signal processing path

[0022] 111, 311, 511-1, 511-2: Compensation device

[0023] 112-1, 112-2, 312-1, 312-2, 512-1, 512-2: Digital-to-analog converter

[0024] 113-1, 113-2, 313-1, 313-2, 513-1, 513-2: Filter

[0025] 114-1, 114-2, 124, 314-1, 314-2, 324, 514-1, 514-2, 524: Mixer

[0026] 115, 315, 515: Adder

[0027] 116, 316, 516: Buffer circuit

[0028] 117, 317, 517: Power amplifier

[0029] 121, 321, 521: Power spectral density estimation device

[0030] 122, 322, 522: Analog-to-digital converter

[0031] 123, 323, 523: Programmable gain amplifier

[0032] 125, 325, 525: Digital signal processor

[0033] S902, S904, S906: Steps

[0034] CMP_X, CMP_Y, IDAC_I, IDAC_Q: Compensation values

[0035] LO_Sig: Oscillation signal

[0036] LO: Oscillation frequency

[0037] Sig: Input signal

[0038] Img: Image signal

[0039] Sig*Img: Signal

[0040] ωm: Angular frequency Detailed implementation manners

[0041] Figure 1 Shows a transmitter circuit according to an embodiment of the present invention. The transmitter circuit 100 may include a transmission signal processing path 110 and a feedback signal processing path 120. The transmission signal processing path 110 may include multiple stages of transmission signal processing devices for processing transmission signals. For example, converting a baseband signal into a radio frequency signal. According to an embodiment of the present invention, the transmission signal processing path 110 may include a compensation device 111, digital-to-analog converters (DAC) 112-1 and 112-2 respectively located on the in-phase channel and the quadrature channel, filters 113-1 and 113-2, mixers 114-1 and 114-2, and adders 115 and buffer circuits 116.

[0042] The compensation device 111 is disposed on the transmission signal processing path 110. The compensation device 111 can receive a set of input signals respectively located on the in-phase (I) channel and the quadrature (Q) channel, and perform IQ imbalance compensation on the received signals according to a set of compensation values CMP_X and CMP_Y to generate compensated signals. Among them, the compensation values CMP_X and CMP_Y can be compensation values for compensating the amplitude or phase of the in-phase channel and the quadrature channel. For example, the compensation device 111 can multiply the input signal on the in-phase (I) channel by the compensation value CMP_X, and add the input signal on the in-phase (I) channel multiplied by the compensation value CMP_Y to the input signal on the quadrature (Q) channel to compensate for the IQ imbalance on the transmission signal processing path. In an embodiment of the present invention, the compensation values CMP_X and CMP_Y are real numbers.

[0043] The digital-to-analog converters 112-1 and 112-2 are respectively used to convert the compensated multiple signals from the digital domain to the analog domain on the in-phase channel and the quadrature channel. The filters 113-1 and 113-2 are respectively used to perform filtering operations on the received signals. The mixers 114-1 and 114-2 are respectively used to multiply the received signals by an oscillation signal LO_Sig to convert the received signals from the baseband frequency to the radio frequency signal. The oscillation signals LO_Sig provided to the mixers 114-1 and 114-2 can be two signals with the same frequency and orthogonal phases. In an embodiment of the present invention, the oscillation frequency of the oscillation signal LO_Sig is LO. The adder 115 is used to combine the signals on the in-phase channel and the quadrature channel. The buffer circuit 116 can be a driving circuit of the power amplifier 117, used to buffer the received radio frequency signal and drive the subsequent power amplifier 117. The power amplifier 117 is used to amplify the radio frequency signal before it is transmitted through the antenna.

[0044] In an embodiment of the present invention, the feedback signal processing path 120 may include a plurality of feedback signal processing devices. For example, a mixer 124, a programmable gain amplifier (PGA) 123, an analog-to-digital converter (ADC) 122, a power spectral density estimation device 121, and a digital signal processor 125. The feedback signal processing path 120 may be coupled to an output terminal of at least one transmission signal processing device, for example, the output terminal of the buffer circuit 116, to receive, as a feedback signal, the output signal generated by the transmission signal processing device from the output terminal and process the feedback signal. Among them, the output signal is the signal obtained after being processed by the aforementioned transmission signal processing device, including the signal processing performed on the in-phase channel and the quadrature channel respectively, and the mixer 124 multiplies the received feedback signal by itself to down-convert the feedback signal to a baseband signal. The programmable gain amplifier 123 is used to amplify / attenuate the received feedback signal. The analog-to-digital converter 122 is used to convert the feedback signal from the analog domain to the digital domain. The power spectral density estimation device 121 may perform a fast Fourier transform (FFT) on the received feedback signal to generate a feedback signal in the frequency domain and estimate the distribution of the power of the feedback signal in the frequency domain.

[0045] The digital signal processor 125 is coupled to the power spectral density estimation device 121 and the compensation device 111 to perform a calibration operation according to the feedback signal in the frequency domain to calibrate one or more compensation values used by the compensation device 111.

[0046] In an embodiment of the present invention, the components on the transmission signal processing path 110, together with the power amplifier 117 and the antenna, may be regarded as a whole as a transmitter, and one or more components on the feedback signal processing path 120 may be regarded as a whole as a compensation value calibration device to assist in the execution of the calibration operation.

[0047] As described above, due to the possible IQ imbalance on the transmission signal processing path, the output signal generated by the transmission signal processing device may contain the energy of unwanted image signals.

[0048] Figure 2 A spectrum example of a radio frequency signal is shown, where this radio frequency signal is the radio frequency signal output at the output terminal of the buffer circuit 116. Assume that the input signal Sig is a single-frequency signal with an angular frequency of ω m For example Figure 1 shown as a pair of sine signals cosω with an angular frequency of ω m of mt and the cosine signal sinω m t. Due to the presence of IQ imbalance in the transmission signal processing path, the spectrum of the radio frequency signal generated on the transmission signal processing path 110 not only contains the energy of the input signal Sig at the frequency (LO + ω m ), but also contains the energy of the unwanted image signal Img at the frequency (LO - ω m ).

[0049] Figure 3 FIG. shows an example of the spectrum of the feedback signal, where this feedback signal is the feedback signal at the output of the power spectral density estimation device 121. Due to the presence of the image signal Img, the signal generated after the feedback signal is processed by the mixer 124 still has non-negligible energy at the frequency (2ω m ), as shown in the signal Sig*Img at the frequency (2ω m ), which has energy proportional to the amplitude of Sig*LO_Sig.

[0050] Therefore, according to an embodiment of the present invention, the compensation device 111 is used to perform IQ imbalance compensation on the received signal according to a set of compensation values CMP_X and CMP_Y to reduce or minimize the energy of an image signal generated on the transmission signal processing path, so as to solve the IQ imbalance problem. When dealing with the RF impairment problem of IQ imbalance, an attempt can be made to minimize the energy of the feedback signal generated after being processed by the mixer 124 at a predetermined frequency, and the predetermined frequency is twice the input frequency of the input signal. For example, when the angular frequency of the test signal is ω m , the predetermined frequency is twice the angular frequency, 2ω m . It should be noted that the predetermined frequency can be changed according to the design of the mixer 124. For example, when the mixer 124 is designed to multiply the feedback signal by a signal with another frequency, the predetermined frequency can be adjusted to another value obtained by adding or subtracting the input frequency and that frequency.

[0051] When the difference between the energy of the radio frequency signal at the frequency (LO + ω m ) and the energy at the frequency (LO - ω m ) is larger, it represents better image rejection ability of the transmitter circuit 100. Therefore, the compensation device 111 compensates for the IQ imbalance on the transmission signal processing path through the compensation values CMP_X and CMP_Y to suppress the generation of the image signal, so that the energy of the feedback signal at the frequency 2ω m can be reduced, and the compensation value correction device is used to correct one or more compensation values used by the compensation device 111 to make the compensation result reach the best.

[0052] Figure 4A graph showing the relationship between the compensation values CMP_X and CMP_Y of the transmitter circuit 100 and the energy of the feedback signal at the aforementioned predetermined frequency, where the predetermined frequency is twice the input frequency of the input signal. As described above, since the energy of the feedback signal at the predetermined frequency is positively correlated with the image signal Img, the smaller the energy of the feedback signal at the predetermined frequency, the better the image rejection ability of the transmitter circuit 100. Therefore, in the embodiments of the present invention, by finding the compensation values CMP_X and CMP_Y that can minimize the energy of the feedback signal at the predetermined frequency, the problem of IQ imbalance of the transmitter circuit 100 can be effectively solved, and the compensation result can reach the best. From Figure 4 As can be seen from the shown graph, the relationship between the compensation values CMP_X and CMP_Y and the energy of the feedback signal at the predetermined frequency can be approximated as a quadratic function.

[0053] Figure 5 A transmitter circuit according to another embodiment of the present invention is shown. The transmitter circuit 300 may include a transmit signal processing path 310 and a feedback signal processing path 320. The transmit signal processing path 310 may include multiple stages of transmit signal processing devices for processing transmit signals. For example, converting a baseband signal into a radio frequency signal. According to an embodiment of the present invention, the transmit signal processing path 310 may include digital-to-analog converters (DACs) 312-1 and 312-2, filters 313-1 and 313-2, a compensation device 311, mixers 314-1 and 314-2, and adders 315 and buffer circuits 316 respectively located on the in-phase channel and the quadrature channel.

[0054] The components included in the transmitter circuit 300 are substantially the same as those in the transmitter circuit 100, and the structures are similar. The difference is only that the compensation device 311 is coupled between the filters 313-1 / 313-2 and the mixers 314-1 / 314-2.

[0055] In this embodiment, digital-to-analog converters 312-1 and 312-2 are respectively used to convert an input signal from the digital domain to the analog domain on the in-phase channel and the quadrature channel. Filters 313-1 and 313-2 are respectively used to perform filtering operations on the received signals. Compensation device 311 is disposed on the transmit signal processing path 310. Compensation device 311 can receive a set of input signals located on the in-phase (I) channel and the quadrature (Q) channel from filters 313-1 and 313-2 respectively, and perform compensation on the received signals according to a set of compensation values IDAC_I and IDAC_Q to generate compensated signals. Among them, the compensation values IDAC_I and IDAC_Q can be compensation values for compensating the amplitude or phase of the in-phase channel and the quadrature channel. For example, compensation device 311 can add the compensation value IDAC_I to the input signal on the in-phase (I) channel, and add the compensation value IDAC_Q to the input signal on the quadrature (Q) channel to reduce the leakage of the local oscillator signal. In an embodiment of the present invention, the compensation values IDAC_I and IDAC_Q are real numbers.

[0056] Mixers 314-1 and 314-2 are respectively used to multiply the received signals, such as the aforementioned compensated signals, by an oscillation signal LO_Sig to convert the received signals from the baseband frequency to the radio frequency signal. The oscillation signals LO_Sig provided to mixers 314-1 and 314-2 can be two signals with the same frequency and orthogonal phases. In the embodiment of the present invention, the oscillation frequency of the oscillation signal LO_Sig is LO. Adder 315 is used to combine the signals on the in-phase channel and the quadrature channel. Buffer circuit 316 can be a driving circuit of power amplifier 317, and is used to buffer the received radio frequency signal and drive the subsequent power amplifier 317. Power amplifier 317 is used to amplify the radio frequency signal before it is transmitted through the antenna.

[0057] In an embodiment of the present invention, the feedback signal processing path 320 may include a plurality of feedback signal processing devices. For example, a mixer 324, a programmable gain amplifier (PGA) 323, an analog-to-digital converter (ADC) 322, a power spectral density estimation device 321, and a digital signal processor 325. The feedback signal processing path 320 may be coupled to an output terminal of at least one transmission signal processing device, such as the output terminal of the buffer circuit 316, to receive, as a feedback signal, the output signal generated by the transmission signal processing device from the output terminal and process the feedback signal. Wherein, the output signal is the signal obtained after signal processing by the foregoing transmission signal processing device, including signal processing performed on the in-phase channel and the quadrature channel respectively, and the mixer 324 multiplies the received feedback signal by itself to down-convert the feedback signal to a baseband signal. The programmable gain amplifier 323 is used to amplify / attenuate the received feedback signal. The analog-to-digital converter 322 is used to convert the feedback signal from the analog domain to the digital domain. The power spectral density estimation device 321 performs a fast Fourier transform on the received feedback signal to generate a feedback signal in the frequency domain and estimate the distribution of the power of the feedback signal in the frequency domain. The digital signal processor 325 is coupled to the power spectral density estimation device 321 and the compensation device 311 to perform a calibration operation according to the feedback signal in the frequency domain to calibrate one or more compensation values used by the compensation device 311.

[0058] In an embodiment of the present invention, the components on the transmission signal processing path 310, as well as the power amplifier 317 and the antenna, may be regarded as a whole as a transmitter, and one or more components on the feedback signal processing path 320 may be regarded as a whole as a compensation value calibration device to assist in the execution of the calibration operation.

[0059] As described above, in addition to the RF impairment problem of IQ imbalance, there may also be an RF impairment problem of local oscillator signal leakage on the transmission signal processing path, that is, the output signal generated by the transmission signal processing device may contain the energy of the unwanted local oscillator signal.

[0060] Figure 6 A spectrum example of a radio frequency signal is shown, where this radio frequency signal is the radio frequency signal output at the output terminal of the buffer circuit 316. Assuming that the input signal Sig is a single-frequency signal with an angular frequency of ω m For example, Figure 5 as shown, a pair of sine signals cosω m with an angular frequency of ω m t and a cosine signal sinω m t, then in the case of local oscillator signal leakage, not only at the frequency (LO + ω m) contains the energy of the input signal Sig, and also contains the energy of the oscillation signal LO_Sig at the oscillation frequency LO.

[0061] Figure 7 shows a spectrum example of the feedback signal, where this feedback signal is the feedback signal at the output of the power spectral density estimation device 321. Due to the presence of the oscillation signal LO_Sig, the signal generated after the feedback signal is processed by the mixer 324 has non-negligible energy at the frequency ω m as shown by the signal Sig*LO_Sig at the frequency ω m which has energy proportional to the amplitude of Sig*LO_Sig.

[0062] Therefore, according to an embodiment of the present invention, the compensation device 311 is used to perform compensation on the received signal according to a set of compensation values IDAC_I and IDAC_Q to reduce or minimize the energy of the local oscillation signal leakage generated on the transmission signal processing path, so as to solve the problem of local oscillation signal leakage. When dealing with the RF impairment problem of local oscillation signal leakage, an attempt can be made to minimize the energy of the feedback signal generated after being processed by the mixer 324 at a given frequency, and the given frequency is equal to the input frequency of the input signal. For example, when the test signal angular frequency is ω m , the given frequency is equal to the angular frequency ω m . It should be noted that the given frequency can be changed according to the design of the mixer 324. For example, when the mixer 324 is designed to multiply the feedback signal with a signal having another frequency, the given frequency can be adjusted to another value obtained by adding or subtracting the input frequency and that frequency.

[0063] When the difference between the energy of the RF signal at the frequency (LO + ω m ) and the energy at the frequency LO is larger, it means that the local oscillation signal leakage is smaller. Therefore, the compensation device 311 compensates the signal on the transmission signal processing path according to the compensation values IDAC_I and IDAC_Q to reduce the energy of the RF signal at the frequency LO, and further reduce the energy of the feedback signal at the frequency ω m , and the compensation value correction device is used to correct one or more compensation values used by the compensation device 311 to make the compensation result optimal.

[0064] Figure 8It shows a graph of the relationship between the compensation values IDAC_I and IDAC_Q of the transmitter circuit 300 and the energy of the feedback signal at a given frequency, where the given frequency is equal to the input frequency of the input signal. As described above, the energy of the feedback signal at the given frequency is positively correlated with the leakage of the local oscillation signal. When the energy of the feedback signal at the given frequency is smaller, it means that the leakage of the local oscillation signal of the transmitter circuit 300 is smaller. Therefore, in the embodiments of the present invention, by finding the compensation values IDAC_I and IDAC_Q that can minimize the energy of the feedback signal at the given frequency, the problem of the leakage of the local oscillation signal of the transmitter circuit 300 can be effectively solved, and the compensation result can reach the best. From Figure 8 As can be seen from the shown graph, the relationship between the compensation values IDAC_I and IDAC_Q and the energy of the feedback signal at the given frequency can also be approximated as a quadratic function.

[0065] Since the relationship between the compensation values CMP_X and CMP_Y and the energy of the feedback signal at a given frequency that is twice the input frequency ω m can be approximated as a quadratic function, therefore, in the embodiments of the present invention, setting the compensation for IQ imbalance as an optimization problem, using a compensation value correction method to obtain the optimal solution, so as to find the better or optimal compensation values CMP_X and CMP_Y applicable to the transmitter circuit 100. Similarly, since the relationship between the compensation values IDAC_I and IDAC_Q and the energy of the feedback signal at a given frequency equal to the input frequency ω m can also be approximated as a quadratic function, therefore, in the embodiments of the present invention, the leakage of the local oscillation signal can also be set as an optimization problem, and the same compensation value correction method can be used to obtain the optimal solution, so as to find the better or optimal compensation values IDAC_I and IDAC_Q applicable to the transmitter circuit 300.

[0066] According to an embodiment of the present invention, the optimization problem can be expressed as the following formula (1):

[0067]

[0068] where the cost function y(x i , x q ) is a quadratic function including two variables x i and x q , and are two parameters that can obtain the minimum y value, and the cost function y(x i , x q ) can be expressed as the following formula (2):

[0069]

[0070] where a, b, c, d, e, and f are multiple coefficients of the cost function y(x i , x q ).

[0071] In an embodiment of the present invention, by corresponding the compensation value CMP_X to x i , corresponding the compensation value CMP_Y to x q , and corresponding the energy of the feedback signal at a predetermined frequency to y, the optimal solution for compensating IQ imbalance can be obtained by using the compensation value correction method proposed by the present invention. Similarly, by corresponding the compensation value IDAC_I to x i , corresponding the compensation value IDAC_Q to x q , and corresponding the energy of the feedback signal at a predetermined frequency to y, the optimal solution for compensating the leakage of the local oscillator signal can be obtained by using the compensation value correction method proposed by the present invention. It should be noted that, as described above, in an embodiment of the present invention, when dealing with the RF impairment problem of IQ imbalance, the predetermined frequency is twice the input frequency of the input signal, and when dealing with the RF impairment problem of the leakage of the local oscillator signal, the predetermined frequency is equal to the input frequency of the input signal.

[0072] The compensation value correction method proposed by the present invention is described as follows:

[0073] Taking the partial derivatives of the cost function y(x i , x q ) with respect to x i and x q respectively, the following equations (3) and (4) are obtained:

[0074]

[0075]

[0076] If the simultaneous equations are solved using equations (3) and (4), the optimal solution can be obtained. Therefore, equations (3) and (4) are rewritten in matrix form as equation (5) below:

[0077]

[0078] Further, equation (6) for calculating the optimal parameters and is derived from equation (5) as follows:

[0079]

[0080]

[0081] As can be seen from equation (6), if the cost function y(xi ,x q ) multiple coefficients a, b, c, d, e, the optimal parameters can be obtained and Therefore, in the embodiment of the present invention, if six sets of parameters (x i1 ,x q1 )、(x i2 ,x q2 )、(x i3 ,x q3 )、(x i4 ,x q4 )、(x i5 ,x q5 )、(x i6 ,x q6 ), and assuming that the corresponding output values obtained by substituting the six sets of parameters into the cost function are y1, y2, y3, y4, y5, and y6, and then substituting these parameters into the cost function of formula (2) to obtain six mathematical formulas, the matrix function of the following formula (7) is obtained by combining them:

[0082]

[0083] Then, using Cramer's Rule, we can get the calculation method of coefficients a, b, c, d, and e as shown in the following formula (8):

[0084]

[0085] The determinant Δ can be calculated using the following formula (9):

[0086]

[0087] Determinant Δ a It can be calculated by the following formula (10):

[0088]

[0089] Determinant Δ b It can be calculated by the following formula (11):

[0090]

[0091] Determinant Δ c It can be calculated by the following formula (12):

[0092]

[0093] Determinant Δ d It can be calculated by the following formula (13):

[0094]

[0095] Determinant Δ e can be calculated through the following formula (14):

[0096]

[0097] According to an embodiment of the present invention, after substituting the coefficients a, b, c, d, e calculated using formula (8) into formula (6), the optimal parameter values can be obtained. and

[0098] Specifically, according to an embodiment of the present invention, the digital signal processors 125 and 325 can first set multiple different compensation values as the aforementioned parameters (x i , x q ), and sequentially provide the multiple compensation values to the corresponding compensation devices 111 and 311. The compensation devices 111 and 311 sequentially process a set of input signals according to the received multiple compensation values. For example Figure 1 , 5 a pair of sine signals cosω m t and cosine signal sinω m t shown in, to generate multiple compensated signals. The transmission signal processing devices arranged on the transmission signal processing paths 110 and 310 sequentially process the multiple compensated signals to generate multiple output signals. The compensation value correction devices arranged on the feedback signal processing paths 120 and 320 sequentially receive the multiple output signals as multiple feedback signals from an output end of the transmission signal processing paths 110 and 310, process the multiple feedback signals, and perform a correction operation according to the multiple feedback signals and the multiple compensation values.

[0099] The digital signal processors 125 and 325 obtain the above-mentioned matrix function of formula (7) according to a matrix composed of the multiple compensation values and a row vector composed of the energies of the multiple feedback signals at a predetermined frequency, calculate multiple determinants using the above-mentioned formulas (9) to (14), and finally apply the obtained determinants to the above-mentioned formula (8) to determine the multiple coefficients a, b, c, d, e of the cost function y(x i , x q ).

[0100] According to an embodiment of the present invention, the set of input signals can be a set of single-frequency signals having an input frequency. For example, a pair of sine signals cosω m with an angular frequency of ω m t and cosine signal sinω mt. As described above, in the embodiments of the present invention, when dealing with the RF impairment problem of IQ imbalance, the predetermined frequency can be set to twice the input frequency of the input signal, and when dealing with the RF impairment problem of local oscillator signal leakage, the predetermined frequency can be set to be equal to the input frequency of the input signal.

[0101] Therefore, in the embodiments of the present invention, the power spectral density estimation devices 121 and 321 are used to perform a fast Fourier transform on a plurality of feedback signals received from the transmitter after being processed by the mixer to generate feedback signals in the frequency domain, and estimate the distribution of the power of the feedback signals in the frequency domain. When correcting the compensation values of IQ imbalance to obtain the optimal solution, the digital signal processor 125 can obtain the energy of the feedback signal at twice the input frequency from the power spectral density estimation device 121 as the output value corresponding to the cost function, such as y1, y2, y3, y4, y5, y6 in Equation (7), and use the above Equation (8) to determine the coefficients a, b, c, d, e of the cost function y(x i ,x q ), and determine the corrected compensation values CMP_X and CMP_Y according to Equation (6). Similarly, when correcting the compensation values of local oscillator signal leakage to obtain the optimal solution, the digital signal processor 325 can obtain the energy of the feedback signal at the input frequency from the power spectral density estimation device 321 as the output value corresponding to the cost function, such as y1, y2, y3, y4, y5, y6 in Equation (7), and use the above Equation (8) to determine the coefficients a, b, c, d, e of the cost function y(x i ,x q ), and determine the corrected compensation values IDAC_I and IDAC_Q according to Equation (6).

[0102] Therefore, in the embodiments of the present invention for correcting the compensation values of IQ imbalance, the obtained set of corrected compensation values is a set of compensation values that can minimize the energy of the feedback signal corresponding to the output signal generated in response to the input signal on the transmission signal processing path of the transmitter at a predetermined frequency of twice the input frequency. In the embodiments of the present invention for correcting the compensation values of local oscillator signal leakage, the obtained set of corrected compensation values is a set of compensation values that can minimize the energy of the feedback signal corresponding to the output signal generated in response to the input signal on the transmission signal processing path of the transmitter at a predetermined frequency equal to the input frequency.

[0103] After obtaining the corrected compensation values, the compensation devices 111 and 311 can further process the subsequently received input signals according to the corrected compensation values to achieve the best compensation result.

[0104] Figure 9The figure shows a flowchart of a compensation value correction method according to an embodiment of the present invention, for correcting one or more compensation values used by a compensation device of a transmitter, including the following steps:

[0105] Step S902: Obtain a plurality of output signals generated by the transmitter processing a set of input signals according to multiple sets of compensation values in sequence as a plurality of feedback signals. Each feedback signal corresponds to one of the multiple sets of compensation values.

[0106] Step S904: Determine a plurality of coefficients of a cost function according to the multiple sets of compensation values and the plurality of feedback signals in a calibration operation.

[0107] Step S906: Determine a set of calibrated compensation values according to the coefficients of the cost function, and provide the set of calibrated compensation values to the compensation device of the transmitter.

[0108] It should be noted that although Figure 1 and Figure 5 show examples of transmitter circuits, it should be understood that Figure 1 , Figure 5 For simplicity, the schematic diagram of the transmitter circuit only shows the components related to the present invention. Those skilled in the art can understand that the transmitter circuit may include many components not shown in the figure to implement functions of wireless communication and related signal processing. Therefore, the present invention is not limited thereto. For example, the compensation value correction method provided by the present invention can also be applied to a transmitter circuit configured with multiple compensation devices, and the compensation values used by each compensation device can be corrected by using the compensation value correction method.

[0109] Figure 10 The figure shows a transmitter circuit according to another embodiment of the present invention. The transmitter circuit 500 may include a transmit signal processing path 510 and a feedback signal processing path 520. The transmit signal processing path 510 may include multiple stages of transmit signal processing devices for processing transmit signals. For example, the transmit signal processing path 510 may include compensation devices 511-1 and 511-2, digital-to-analog converters (DACs) 512-1 and 512-2 located on the in-phase channel and the quadrature channel respectively, filters 513-1 and 513-2, mixers 514-1 and 514-2, and adders 515 and buffer circuits 516.

[0110] The feedback signal processing path 520 may include a plurality of feedback signal processing devices, for example, a mixer 524, a programmable gain amplifier (PGA) 523, an analog-to-digital converter (ADC) 522, a power spectral density estimation device 521, and a digital signal processor 525. One or more components on the feedback signal processing path 520, for example, the power spectral density estimation device 521 and the digital signal processor 525, may be regarded as a whole as a compensation value correction device to assist in the execution of the correction operation.

[0111] Figure 10 The illustrated transmitter circuit 500 and the components included in the transmitter circuits 100 and 300 are substantially the same and have a similar structure, and the difference is only that the transmitter circuit 500 includes a plurality of compensation devices. Similar component symbols refer to the same components, so the detailed description of the same components will not be repeated here. In an embodiment of the present invention, the compensation value correction device may sequentially apply the foregoing compensation value correction method to correct the compensation values used by each compensation device. For example, the compensation value correction device may first apply the foregoing compensation value correction method to correct the compensation values CMP_X and CMP_Y used by the compensation device 511-1 to obtain better or optimal compensation values CMP_X and CMP_Y, and then apply the foregoing compensation value correction method to correct the compensation values IDAC_I and IDAC_Q used by the compensation device 511-2 to obtain better or optimal compensation values IDAC_I and IDAC_Q. In addition, the compensation value correction device may also iteratively correct the compensation values used by the compensation device 511-1 and / or 511-2. For example, the obtained better compensation value is used as one of a plurality of groups of compensation values, and the compensation value correction is performed again. Therefore, the present invention is not limited to performing the compensation value correction only once, and the compensation value correction device may also perform multiple corrections on the compensation values used by the compensation device 511-1 and / or 511-2 to obtain better or optimal compensation values.

[0112] As described above, the present invention provides a compensation value correction method that can be integrally applied to solve different radio frequency impairment problems, so as to efficiently obtain corresponding compensation values, and better or optimal compensation values can be obtained through the correction of the compensation values.

[0113] The above are only the preferred embodiments of the present invention, and all equivalent changes and modifications made according to the claims of the present invention should fall within the scope covered by the present invention.

Claims

1. A transmitter circuit, characterized in that, The transmitter circuit includes: A compensation device disposed on a transmission signal processing path for sequentially processing a set of input signals according to multiple sets of compensation values to generate multiple sets of compensated signals; At least one transmission signal processing device disposed on the transmission signal processing path for sequentially processing the multiple sets of compensated signals to generate multiple output signals; and A compensation value correction device coupled to an output end of the transmission signal processing path for sequentially receiving the multiple output signals as multiple feedback signals from the output end and performing a correction operation according to the multiple feedback signals and the multiple sets of compensation values, wherein the compensation value correction device includes: A digital signal processor coupled to the compensation device. In the correction operation, the digital signal processor determines multiple coefficients of a cost function according to the multiple sets of compensation values and the multiple feedback signals, determines a set of corrected compensation values according to the multiple coefficients, and provides the set of corrected compensation values to the compensation device.

2. The transmitter circuit according to claim 1, wherein The set of input signals is a set of single-frequency signals having an input frequency. The set of corrected compensation values is a set of compensation values that makes a feedback signal corresponding to an output signal generated in response to the set of input signals on the transmission signal processing path have the minimum energy at a predetermined frequency, and the predetermined frequency is equal to the input frequency.

3. The transmitter circuit according to claim 1, wherein The set of input signals is a set of single-frequency signals having an input frequency. The set of corrected compensation values is a set of compensation values that makes a feedback signal corresponding to an output signal generated in response to the set of input signals on the transmission signal processing path have the minimum energy at a predetermined frequency, and the predetermined frequency is twice the input frequency.

4. The transmitter circuit according to claim 1, characterized in that, After receiving the set of corrected compensation values, the compensation device further processes the subsequently received set of input signals according to the set of corrected compensation values.

5. A compensation value correction device for correcting one or more compensation values used by a compensation device of a transmitter, characterized in that The compensation value correction device includes: A power spectral density estimation device for performing a fast Fourier transform on the multiple feedback signals received from the transmitter to generate the multiple feedback signals in the frequency domain and estimating the distribution of the power of the multiple feedback signals in the frequency domain, wherein each feedback signal corresponds to one of the multiple sets of compensation values; and A digital signal processor coupled to the power spectral density estimation device and the compensation device for determining multiple coefficients of a cost function according to the multiple sets of compensation values and the multiple feedback signals in a correction operation, determining a set of corrected compensation values according to the multiple coefficients, and providing the set of corrected compensation values to the compensation device.

6. The compensation value correction device according to claim 5, characterized in that, The set of corrected compensation values is a set of compensation values that makes a feedback signal corresponding to an output signal generated in response to an input signal by the transmitter have the minimum energy at a predetermined frequency, and the predetermined frequency is equal to an input frequency of the input signal.

7. The compensation value correction device according to claim 5, wherein The set of corrected compensation values is a set of compensation values for which a feedback signal corresponding to an output signal generated by the transmitter in response to an input signal has minimum energy at a predetermined frequency, and the predetermined frequency is twice an input frequency of the input signal.

8. A compensation value correction method for correcting one or more compensation values used by a compensation device of a transmitter, characterized in that, The compensation value correction method includes: Obtaining a plurality of output signals generated by the transmitter processing a set of input signals according to a plurality of sets of compensation values in sequence as a plurality of feedback signals, wherein each feedback signal respectively corresponds to one of the plurality of sets of compensation values; Determining a plurality of coefficients of a cost function according to the plurality of sets of compensation values and the plurality of feedback signals in a correction operation; and Determining a set of corrected compensation values according to the plurality of coefficients, and providing the set of corrected compensation values to the compensation device.

9. The compensation value correction method according to claim 8, wherein The set of input signals is a set of single-frequency signals having an input frequency, the set of corrected compensation values is a set of compensation values for which a feedback signal corresponding to an output signal generated by the transmitter in response to the set of input signals has minimum energy at a predetermined frequency, and the predetermined frequency is equal to the input frequency.

10. The compensation value correction method according to claim 8, wherein The set of input signals is a set of single-frequency signals having an input frequency, the set of corrected compensation values is a set of compensation values for which a feedback signal corresponding to an output signal generated by the transmitter in response to the set of input signals has minimum energy at a predetermined frequency, and the predetermined frequency is twice the input frequency.

Images