JP2003333117A - Coherent adaptive calibration system and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a receiver which is adaptively calibrated by using a coherent reference signal. <P>SOLUTION: The reference signal is selected to be offset from a center frequency of a calibration signal such that the resultant product is offset from baseband by some small amount. The resultant product is used to determine a next value of calibration parameters. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to wireless communication. More particularly, the present invention relates to adaptive calibration of receiver equipment.

[0002]

2. Related Art With the advent and spread of digital communication systems, the demand for low-cost, high-performance wireless receivers continues to increase at an accelerated rate. Those demands have generated a strong interest in the development of direct conversion receiver architectures that achieve a single conversion from the radio link carrier frequency to the baseband frequency. The reduced complexity of adopting a direct conversion receiver ensures great potential for reduced cost and improved performance.

In many modern digital communication devices, radio signals carry in-phase (I) information, which carries information with a relative phase shift of 90 degrees.
A quadrature modulation scheme is used that includes a component and a quadrature (Q) component. Typically, the in-phase and quadrature components are
It is received using two different signal paths within the direct conversion receiver. Any difference in gain or phase between the two paths degrades the information in the signal. In addition, any DC offset or low frequency noise voltage generated by the receiver will degrade the information in the signal.

To reduce this degradation, direct conversion receivers often employ adaptive calibration mechanisms. For example, prior art devices have been proposed that include an adaptive calibration mechanism that is not coherent. However, such non-coherent adaptive calibration schemes exhibit a high noise figure and therefore do not accurately calibrate receiver imperfections.

[0005]

Therefore, there is a need in the art for the development of an adaptive calibration system that provides accurate, low noise calibration.

[0006]

SUMMARY OF THE INVENTION Coherent adaptive calibration receivers and methods are used to adjust for errors in receivers, such as direct conversion receivers used to receive radio frequency signals. The sequence of first channel signal samples is added with a first channel offset modification parameter to produce a modified sequence of first channel samples. The sequence of second channel signal samples is added with a second channel offset correction parameter followed by multiplication with a gain imbalance correction parameter, the product of the corrected sequence of first channel samples and the phase error correction parameter. Is added, the second
A modified set of channel samples of is generated. When modified, the first channel and the second channel are orthogonal to each other.

The modified sequence of first channel samples is multiplied by a first sinusoidal waveform to determine a first product. The modified sequence of second channel samples is the second
Is multiplied by the sinusoidal waveform of to determine the second product. The second sine waveform is 90 degrees out of phase with the first sine waveform.

The first product is filtered to determine a first channel gain imbalance amount. The second product is filtered to determine a second channel gain imbalance amount. The next value of the gain imbalance correction parameter is the first
And a second channel gain imbalance amount. In one embodiment, multiplication and filtering are performed digitally.

Alternatively, or in addition, the modified sequence of second channel samples is multiplied by a first sinusoidal waveform to determine a third product. The third product is filtered to determine the amount of phase error. The next value of the phase correction parameter is determined based on the first gain imbalance amount and the phase error amount.

In one embodiment, the invention is embodied in a receiver. The receiver has a first adder configured to add a first channel offset correction parameter to the series of first channel signal samples to produce a modified series of first channel samples. The receiver is configured to multiply the modified sequence of first channel samples by a phase error correction parameter.
Also has a multiplier of. A second adder is configured to add a second channel offset correction parameter to the series of second channel signal samples. The second multiplier is configured to multiply the output of the second adder by the gain imbalance correction value. A third adder is configured to add the output of the first multiplier and the output of the second multiplier to produce a modified set of second channel samples. Here, the second channel is orthogonal to the first channel. A third multiplier is configured to multiply the modified sequence of first channel samples by the first digitized sine waveform. The fourth multiplier is the second
Is adapted to be multiplied by a second digitized sine waveform that is 90 degrees out of phase with the first digitized sine waveform.
The first digital filter is configured to filter the output of the third multiplier to determine the first channel gain imbalance amount. A second digital filter is configured to filter the output of the fourth multiplier to determine the second channel gain imbalance amount. A calculator is configured to determine a gain imbalance correction parameter based on the first and second channel gain imbalance amounts.

In another embodiment, the invention is configured to add a first channel offset correction parameter to the sequence of first channel signal samples to produce a modified sequence of first channel samples. Embodied by a receiver having a first adder present. A first multiplier is configured to multiply the modified sequence of first channel samples by the phase error correction parameter. The second adder is configured to add a second channel offset correction parameter to the series of second channel signal samples. The second multiplier is configured to multiply the output of the second adder by the gain imbalance correction value. A third adder is configured to add the output of the first multiplier and the output of the second multiplier to produce a modified set of second channel samples. Where the second
Channels are orthogonal to the first channel. A third multiplier is configured to multiply the modified sequence of first channel samples by the first digitized sine waveform. A fourth multiplier is configured to multiply the modified sequence of second channel samples by the first sinusoidal waveform to determine a third product. The first digital filter is configured to filter the output of the third multiplier to determine the first channel gain imbalance amount.
The second digital filter is configured to filter the output of the fourth multiplier to determine the amount of phase error. A calculator is configured to determine a next value of the phase modification parameter based on the first amount of gain imbalance and the amount of phase error.

The features, objects and advantages of the present invention are as follows:
It will become more apparent from the detailed description given below with reference to the drawings, wherein like reference numerals are used throughout.

[0013]

DETAILED DESCRIPTION OF THE INVENTION Adaptive calibration is used to compensate for the corruption introduced by the receiver.
Adaptive calibration involves measuring the current degradation, calculating updated correction parameters, and applying the updated correction parameters to the received signal. This process is iterative and the calibration parameters adapt to the current operating conditions. For example, the degradation introduced by the receiver is usually a function of operating frequency and operating temperature. Therefore, the degradation introduced by the receiver changes as a function of time. The repeatability of adaptive calibration can compensate for such time-varying behavior. The degradation introduced by the receiver can cause errors in the resulting digital bits output by the receiver, thus reducing overall system performance.

FIG. 1 is a block diagram of the communication device 100. This communication device has gain imbalance, quadrature error, which occur in the realization of a direct conversion digital receiver.
Compensate for direct current (DC) and low frequency offset errors,
The adaptive calibration apparatus and method of the present invention are provided. An antenna and a duplexer 102 couple the communication device 100 to a wireless link or disconnect it from the wireless link. In one embodiment, the antenna and duplexer 102 is a band bus that functions to reduce the level of interfering signals outside the range of received frequencies to prevent aliasing from out-of-band power into the signal band. Has a filter. Antenna and duplex switch 102
Is a single-pole double-throw switch 10 for receiving a signal received from the wireless link.
Connect to 6. Alternatively, switch 106 can be embodied as a coupler, as described in more detail below. In either case, the switch 106 couples the received signals from the antenna and duplexer 102 to the direct conversion receiver 110 (described in more detail below). Direct conversion receiver 1
10 receives the modulated signal at the radio carrier frequency (f _c ) and produces in-phase (I) and quadrature (Q) filtered signal sample components. The antenna and duplexer 102 couples the transmitted signal from the transmitter 104 into a wireless link.

The I and Q filtered signal samples are output by the direct conversion receiver 110 to a digital processor 112. Digital processor 112 digitally demodulates the signal samples. Digital processor 112, in addition to many other features
Also outputs frequency control, signal waveform clock and transmission data. In one embodiment, the digital processor 112 is
A general purpose microprocessor that executes software code. Alternatively, the functionality of digital processor 112 may be implemented in an application specific integrated circuit (ASIC) with a combination of discrete (discrete) hardware components or software, or with an ASIC in combination with them.

As mentioned above, the communication device 100 incorporates an adaptive calibration device. This adaptive calibrator is used to compensate for the degradation introduced by the direct conversion receiver 110. The adaptive calibrator measures,
It consists of three functions: calculation and correction. To implement those functions, a calibration signal is generated and sent through the direct conversion receiver 110, and the results are analyzed by adaptive measurement and calibration algorithms within the direct conversion receiver 110 and digital processor 112. For example, in one embodiment, during normal mode operation, the transmit frequency synthesizer 114 may have an RF
(radio frequency) An RF calibration signal having a calibration frequency (f ₀ ) is generated. The single-pole double-throw switch 108 under the control of the digital processor 112 causes the transmission frequency synthesizer 11 to
4 outputs to switch 106 or transmitter 104. (In one embodiment, switches 108 and 106 are embodied in a single integrated component using techniques well known in the switching arts.) During calibration mode:
Switch 108 couples the RF calibration signal to direct conversion receiver 110 via switch 106. The direct conversion receiver 110 downconverts the calibration signal in the same manner as the wireless link RF carrier signal to produce I and Q filtered calibration samples for the measurement. The direct conversion receiver 110 measures the error introduced by the signal path. Based on the measurements, the digital processor 112 calculates a series of calibration parameters and sends the calibration parameters to the direct conversion receiver 110. The direct conversion receiver 100 uses the calibration parameters to correct the measured error.

Communication device 100 operates in one of two operating modes, a standard operating mode and a calibration operating mode. In standard operating mode, the direct conversion receiver 110 uses the current correction parameters to compensate for the degradation. In the standard operating mode, the communication device 100
Does not perform measurement or calculation functions.

In the calibration mode, the direct conversion receiver 110 measures current operating conditions. Three different calibration modes are possible. 1 for any one implementation
One, two or all three calibration modes can be included. The three calibration modes are (1) concurrent (simultaneous) external calibration mode, (2) concurrent internal calibration mode, and (3) exclusive internal calibration mode.

In both internal calibration modes, the communication device 10
0 produces a calibration signal that is coupled to the direct conversion receiver 110. In the concurrent internal calibration mode, the switch 106 functions as a coupler, and the calibration signal is added to the wireless link signal received from the antenna and the duplexer 102, as well as the direct conversion receiver 1.
Combined with 10. In the exclusive internal calibration mode, the switch 106 couples the input of the direct conversion receiver 110 to the calibration signal and disconnects the direct conversion receiver 110 from the antenna and the radio link signal input from the duplexer 102, a single pole double throw. Acts as a switch.

In the external calibration mode, one of the input communication signals is used as the calibration signal. For example, wireless links carry calibration signals in the form of constant waveform (CW) or narrowband data signals. In such a case, switch 106 simply directly couples the output of antenna duplexer 102 to direct conversion receiver 110. Using communication signals for calibration reduces the costs associated with generating separate calibration signals. The disadvantage of such operation is that it may reduce the overall capacity and performance of the device and may increase the noise level associated with the calibration process.

As described above, the measurement function and the calculation function are executed during the calibration mode. Given that accurate measurements and accurate calculations can be achieved while operating in concurrent internal calibration mode, it is the preferred calibration mode. However, one of the other calibration modes can be used, provided that the predominant conditions, such as the current operating level of the radio link signal, prevent accurate measurements in the concurrent internal calibration mode.

In the embodiment shown in FIG. 1, the calibration signal is generated by the transmit frequency synthesizer 114.
Due to the generation of the calibration signal by the transmission frequency synthesizer 114, especially in a device that does not require full-duplex operation,
An efficient architecture is obtained. For example GSM (Glo
In the bal System for Mobile Communications) system, and in the United States Time Division Multiple Access (TDMA) system operating according to IS (Interim Standard) -54, the transmit and receive functions are time division multiplexed and transmit during the receive period. The frequency synthesizer 114 remains idle. Also, in such TDMA systems, the use of an exclusive internal calibration mode during periods when the received signal is not being sent to the communication device 100 provides accurate measurements without sacrificing sufficient system performance. be able to. In other embodiments, the calibration signal is generated by another frequency synthesizer, such as a dedicated calibration signal synthesizer. The dedicated synthesizer is CDMA
It can also be used in full-duplex systems such as and where recalibration is required more frequently than the transmitter duty cycle allows.

FIG. 2 is a block diagram showing the direct conversion receiver 110 in greater detail. The direct conversion receiver 110 includes a direct converter 120 that performs the actual down conversion of the radio link signal and the calibration signal. For example, in one embodiment, conversion is accomplished using delta-sigma modulator translating and decimation filtering. In other embodiments, the direct converter 120 is implemented using a standard balanced mixer or other continuous time element, and the resulting analog signal is an analog-to-analog signal.
Digitized by digital converter. In either case, the direct converter 120 has both an I path and a Q path and outputs an in-phase path digital sample Y _I [m] and a quadrature path digital sample Y _Q [m].

The output terminal of the direct converter 120 is coupled to the correction and measurement circuit 122. The correction and measurement circuit 122 applies the current correction parameters to the digital samples. Correction and measurement circuit 122
Is the modified digital sample U for the in-phase path.
Output _I [k] and the modified digital samples U _Q [k] for the orthogonal path to digital processor 112.

In calibration mode, the correction and measurement circuit 12
2 applies the calibration parameters to the calibration signal. The correction and measurement circuit 122 then measures the resulting signal and measures the measured parameters C _I (j, f _RX ), C.
_Q (j, f _RX ), C _β-I (j, f _RX , f _λ ), C _β-Q (j, f _RX ,)
f _λ ) and C _Φ (j, f _RX , f _λ ) (each of which are described in more detail below) and feed them to digital processor 112.

The receive frequency synthesizer and clock generator 126 provides frequency control information to the digital processor 1.
12 and the clock waveform from the digital processor 112. The synthesizer 126 produces a conversion reference signal having a frequency f _RX that is approximately equal to the RF carrier frequency of the currently received radio link signal or a multiple of that RF carrier frequency. In addition, synthesizer 126 is used in the analog-to-digital conversion process within the direct conversion receiver and is used by correction and measurement circuit 122 to generate a clock that maintains synchronization between its input and its output.

In the preferred embodiment, a calibration signal having a frequency f ₀ is input to the direct converter 120 during any of the calibration modes. The direct converter 120 converts the calibration signal into the baseband frequency f _OBB =
Convert to f _RX −f ₀ . During concurrent mode calibration,
Down-converted calibration signal f _OBB (f _OBB is 0
Higher) is chosen to be outside the bandwidth of the signal of interest so as not to interfere with the signal of interest.

FIG. 3 is a block diagram showing the correction and measurement circuit 122 in greater detail. Correction and measurement circuit 1
In 22, multiplexer and clock distributor 13
0 receives the calibration parameters from the adaptive calibration algorithm and distributes them in circuit 122. Due to the repeatability of the adaptive calibration as indicated by the iteration indicator variable j, the value of the calibration parameter depends on the conversion frequency f _RX and time. An adder 132 _provides an offset correction parameter R _I-Offset (j, f _RX ) for the I path digital sample Y _I [m] determined by the I offset correction block 134 based on the output of the multiplexer and clock distributor 130. Add together. The I offset correction block 134 compensates for DC offsets and low frequency errors in the I path.

The output of the adder 132 is coupled to the delay device (delay unit) 136. The delay device 136 is
A delay is added to the I path to compensate for the additional computational delay associated with the Q path so that the I and Q paths exhibit the same overall delay. The output of summer 132 is also coupled to a multiplier 138 used to provide phase error correction, as described in more detail below.

The output of delay 136 has the modified signal samples in standard mode, internal concurrent calibration mode, and external concurrent calibration mode. Furthermore, in any of the calibration modes, the modified signal sample U
_I [k] contains the modified calibration signal. For example, Equation 1 below is digital sample Y _I [m] and modified sample U _I
The relationship between [k] is shown.

[0031]

## EQU00001 ## U _I (k) = Y _I (m) + R _I-Offset (j, f _RX ) Equation 1 Here, R _I-Offset (j, f _RX ) is the I path at the conversion frequency f _RX . Is the value of the j-th offset correction parameter for.

The Q path digital sample Y _Q [m] is input to the adder 140. Adder 140 provides multiplexer and clock distributor 130 for the Q path samples.
Offset correction parameter R _Q-Offset (j,
f _RX ). Q offset correction block 14
2 compensates for DC offsets and low frequency errors in the Q path.

The output of adder 140 is the Q path gain correction parameter R _QQ (j, f) determined in multiplier 144 by the output of gain imbalance modifier 145 based on the outputs of multiplexer and clock generator 130. _RX ) is loaded. The multiplier 144 compensates for the gain imbalance between the I and Q paths.

Multiplier 138 provides multiplexer and clock distributor 130 for the modified I-path samples.
Multiply the phase error correction parameter R _IQ (j, f _RX ) determined by the output of the phase correction block 139 based on the output of The output of the multiplier 138 is, in the adder 146,
It is added to the output of the multiplier 144. Phase correction block 139, multiplier 138, and adder 146 compensate for an ideal 90 degree phase shift error between the I and Q paths. The resulting relationship between the digital sample Y _Q [m] and the modified sample U _Q [k] is given below by Equation 2.

[0035]

## EQU00002 ## U _Q (k) = R _QQ (j, f _RX ) {Y _Q (m) + R _Q-Offset (j, f _RX )} + R _IQ (j, f _RX ) {Y _I (m) + R _I-Offset (j, f _RX )} Equation 2 Here, R _Q-Offset (j, f _RX ) is the value of the j-th offset correction parameter for the orthogonal path at the conversion frequency f _RX , and R _QQ ( j, f _RX ) is the value of the j-th orthogonal path gain imbalance correction parameter at the conversion frequency f _RX , and R _IQ (j, f _RX ) is the value of the j-th phase correction parameter at the conversion frequency f _RX . Is.

As mentioned above, the modification parameter is a function of the transform frequency and is adaptive and thus time varying. This dependence on frequency and time is as shown in Equations 1 and 2. It is indicated by parentheses following the correction parameter.

The output of adder 146 contains the signal samples in standard mode and concurrent internal and external calibration modes. Further, during either calibration mode, the output of adder 146 contains the down-converted calibration signal at baseband frequency f _OBB .

Equations 3 and 4 are modified and measured circuits 122
It gives a mathematical representation of the samples input to the in-phase and quadrature paths of.

[0039]

Equation 3 Y _I [m] = cos [2πf _OBB T _s m] + N _I (m) + η _I (m) Equation 3 Y _Q [m] = Asin [2πf _OBB T _s m + Δ] + N _Q (m) + Η _Q (m) (4) where A is the Q channel gain normalized with respect to the I channel gain for the calibration signal at f _RX ,
Δ is the phase error of the quadrature path for the I channel at frequency f _RX and N _I (m) is the quantized and filtered signal, the interference, the quantization noise, the high frequency modulation product, and the quantized heat. Is a noisy I-channel disturbing component, N _Q (m) is the _Q- channel disturbing component including the quantized and filtered signal, the disturbing, the quantizing noise, the high frequency modulation producing component, and the quantized thermal noise. , Η _I (m) is the I channel DC and low frequency noise, η _Q (m) is the Q channel DC and low frequency noise, and T _s is Y _I [m] and Y _I [m ] Is the sample rate (sampling rate) and f _OBB is the converted calibration frequency, ie [f _OBB = f _TX −f _RX ].

Equations 3 and 4 assume for simplicity that the initial phase of the calibration signal at the input to the correction and measurement circuit 122 is zero. By substituting Equation 3 and Equation 4 into Equation 1 and Equation 2, respectively, Equation 3A and Equation 4A are obtained.

[0041]

[Equation 4]

From Equations 3A and 4A, the I and Q paths are R _QQ (j, f _RX ) A cos Δ = 1 and R _IQ (j, f)
_RX ) = − R _QQ (j, f _RX ) As R
When _QQ (j, f _{R X)} and R _IQ (j, f _RX) is adjusted, and equilibrated with no quadrature error in f _RX. This adjustment is done using the adaptive algorithm of the present invention.

The I offset measurer 148 has been modified
Sample U_I[k] and a multi that determines the duration of the measurement
Start command from the Plexer and clock distributor 130
And receive a stop command. I offset measuring instrument 148
Uses modified samples using techniques well known in the art.
Filtered to reduce the DC or low
Responses that reflect either or both frequency offsets
Measurement parameter C _I(j, f_RX) Is generated. I office
The measuring instrument 148 is shown in more detail in FIG.
More on this below.

Similarly, to determine the corresponding measurement parameter C _IQ (j, f _RX ) that reflects the DC and / or low frequency offset found in the signal, the Q offset measurer 156 modifies the modified Q. Path sample U _Q [k]
And a start command and a stop command from the multiplexer and clock distributor 130. The Q offset measurer 156 is shown in more detail in FIG. 4 and will be described in more detail below.

In contrast to gain imbalance and phase correction calibrations, which are described in more detail below, DC and low frequency corrections can be made both during calibration mode and during normal operation. Thus, in one embodiment, the DC calibration function performed by the I offset measurer 148 and the Q offset measurer 156 can continue to adapt, independent of the mode of operation.

[System for Correcting Quadrature Gai
n and Phase Error in a Direct Conversion Single Si
deband Receiver Independent of the Character of th
disclosed in U.S. Pat. No. 5,604,929 entitled "e Modulated Signal" (a device for correcting quadrature gain and phase error in a direct conversion single sideband receiver independent of the nature of the modulated signal) ". According to known techniques, such as the technique described above, one way to determine the amplitude of the calibration signal component of the modified sample is to square the modified sample. For example, squaring the modified digital sample U _I [k] for the common-mode path yields a DC component having a signal proportional to the amplitude of the common-mode calibration signal, according to well-known mathematical principles.
However, such non-coherent detection also results in a sequence of DC components due to noise and interference present in the signal. The DC component impairs the measurements made during the calibration process. Due to the noise component, such incoherent measurements produce inaccurate measurements of the amplitude of the calibration signal.

In contrast, the present invention uses a coherent measurement of the amplitude of the downconverted calibration signal. It would be ideal if the Q path modified samples could be subtracted directly from the I path modified samples. However, the I path component of the calibration signal is approximately 90 degrees out of phase with the Q path component. For example, if the I path component contains a cosine term, the Q path component contains a sine term. Thus, the amplitudes of the I and Q path components cannot be directly subtracted from each other.

In order to perform coherent adaptive calibration, I
The path channel sample U _I (k) has a cosine wave (or rectangle with an equivalent phase relationship) with a frequency f _λ approximately (but not exactly) the baseband frequency f _OBB of the down-converted calibration signal. Wave). Furthermore, a modified sample of the Q path U _Q (k)
Is multiplied by a sine wave having a frequency f _λ (or a rectangular wave having an equivalent phase relationship). Calibration reference generator 152 generates a measurement reference waveform at frequency f _λ determined by the parameters received from multiplexer and clock distributor 130.

Modified samples U _I [k] and Q of the I path
Both of the path modified samples U _Q [k] are coupled to the I / Q gain imbalance measurer 150. The I / Q gain imbalance measuring device 150 includes a sine signal generated by the calibration reference generator 152, a multiplexer and a clock distributor 13.
I using start command and stop command from 0
By checking the path calibration signal amplitude and the Q path calibration signal amplitude,
Helps determine the relative amplitudes of the I and Q paths. I /
The Q gain imbalance measurer 150 is shown in more detail in FIG. 4 and will be described in more detail below.

Similarly, the phase error measurer 154 uses the sine signal generated by the calibration reference generator 152 to generate a Q
It receives the modified sample U _Q (k) of the path and the start and stop commands from the multiplexer and clock distributor 130. The quadrature error measuring device 154 is shown in FIG.
In more detail and will be described in more detail below.

As shown in FIG. 4, the I offset measurer 148 is shown to include a digital filter 170 and a sample rate compressor 172. Equation 5 is the I path channel sample U _I [k] input to the I offset measuring instrument 148 and the I offset measuring instrument 148.
Mathematically represents the relationship between the measured parameter C _I [j, f _RX ] output by

[0052]

[Equation 5]

Where h _I (k) is the unit sample response of the digital low pass filter 170 and K is the number of samples used in the measurement by the sample rate compressor 172.

The output of the I offset measuring device 148 is input to the I offset adaptive correction algorithm 200. In one embodiment, the I offset adaptive correction algorithm 200
Implements a stochastic gradient algorithm in which the delay constant and the modified gain constant are set to the noise figure of the receiver. Also, the gain constant can be a function of time or the magnitude of the measured error. For more information on stochastic gradient algorithms, see JG Proakis et al., "Advanced Digital Signal Processing," pages 341-350, Macmillan.
Publishing Co., New York (1992). The following Equation 6 is an offset correction parameter R _I-Offset [j,
A mathematical representation of the operations performed by the I-offset adaptive correction algorithm 200 to determine the next value of f _RX ].

[0055]

R _I-Offset [j, f _RX ] = ρ _dc R _I-Offset [(j−1), f _RX ] + α _dc C _I [j, f _RX ] ... Equation 6 Here, ρ _dc is Is the delay constant and α _dc is the modified gain term.

The I / Q gain imbalance measuring device 150 includes a multiplier 174, a digital filter 176, and a sample rate compressor 178. In the I / Q gain imbalance measuring device 150, the equation 7 is expressed as I / Q gain imbalance measuring device 1
I path channel samples U _I [k] input to 50,
The relationship between the measured parameter C _β-I [j, f _RX , f _λ ] output by the I / Q gain imbalance measuring device 150 is mathematically represented.

[0057]

[Equation 7]

Here, h _β (k) is the digital filter 17
Is the unit sample response of 6, K is the number of samples used in the measurement, and V _I (k) is the cosine wave measurement reference signal.

The I / Q gain imbalance measuring device 150 is a multiplier.
180, digital filter 182, sample rate
It is also shown to include a compressor 184. Equation 8
Is the Q path input to the I / Q gain imbalance measuring device 150
Channel sample U_Q[k] and I / Q gain imbalance measuring instrument
Measurement parameter C output by 150_β-Q[j,
f _RX, f_λ] Is a mathematical expression of the relationship between.

[0060]

[Equation 8]

Here, h _β (k) is the digital filter 18
Is the unit sample response of 2, K is the number of samples used in the measurement, and V _Q (k) is the sinusoidal measurement reference signal.

Applying the above expanded equation for U _I [k], Equation 9 mathematically illustrates the operation performed by multiplier 174. Multiplier 174 multiplies the value of U _I [k] by the cosine wave of f _λ output by calibration reference generator 152, as defined by equation 3A. Similarly, applying the equation expanded above for U _Q [k], Equation 10
Mathematically show the actions performed by. Multiplier 180
Multiplies the value of U _Q [k] by the sine wave of f _λ output by the calibration reference generator 152, as defined by equation 4A.

[0063]

[Equation 9]

Filter 176 filters terms having frequencies higher than f _λ -f _OBB output by multiplier 174, as mathematically shown below by Equation 9A. Filter 182 filters terms having frequencies higher than f _λ −f _OBB output by multiplier 180, as mathematically illustrated by Equation 10A.

[0065]

[Equation 10]

Here, ξ _I (k) is the residual noise after filtering by the filter 176, and ξ _Q (k) is the residual noise after filtering by the filter 182.

The residual noise component can be further reduced by additional bandpass filtering centered on | f _λ -f _OBB |. Since the DC and low frequency contributions are shifted to the high frequency side up to the frequency f _λ, they can be efficiently removed by filtering according to a known technique. Thus, this method allows the I and Q imbalance correction parameters to be determined independently of the effects of DC and low frequency offset correction.

The output of filter 176 is decimated in sample rate compressor 178 and coupled to adaptive IQ imbalance correction algorithm 202. Similarly, filter 1
The output of 82 is decimated in the sample rate compressor 178 and coupled to the adaptive IQ imbalance correction algorithm 202. Adaptive IQ imbalance correction algorithm 202
Implements the above stochastic gradient algorithm in one embodiment. Equation 11 below is a modified parameter R _QQ (j,
Mathematically represents the operations performed by the adaptive IQ imbalance correction algorithm 202 to determine the next value of f _RX ).

[0069]

[Equation 11]

Here, C _β-I (j, f _RX , f _λ ) is the output of the sample rate compressor 178, and C _β-Q (j, f _RX , f)
_λ ) is the output of the sample rate compressor 184 and is ρ _β
Is a delay constant, | ρ _β | <1, which is an α _β modified gain term.

In equation 11, the modified signal is the difference or
Term [C_β-I(j, f_RX, f_λ) -C_β-Q(j, f_RX, f_λ)] And terms
C_β-I(j, f_RX, f_λ) And the product. Both frequencies
[f_OBB-F_λ] Contains the component due to the calibration signal
It [f_OBB-F_λF should be close to 0_OBBAnd f_λTo
By selecting, the low-pass filters 178, 182
Is the frequency [f_OBB-F_λ] Calibration signal component and narrow band heat
It only passes noise. The resulting product is
Complains some degradation of the DC term due to thermal noise in the band
DC and 2 [f_OBB−
f_λ] Ingredients are included. This correction signal has frequency 2
[f_OBB-F_λ] And the repetition rate to remove the components in
If the delay constant is selected, follow equation 11
The adaptive IQ imbalance correction algorithm 202
To be filtered. This adaptive IQ balance corrector
The method expresses the deviation of the zero frequency from the surrounding range [f_OBB
-F _λ] To effectively implement a narrow bandpass filter
The DC offset and the input signal U_I[k] and
U_QImprove the effect of [k] on low frequency degradation. For example,
Equation 12 below is an adaptive IQ imbalance correction algorithm 20
In order to show the bandpass characteristic of the operation performed by
Term {[C_β-I(j, f_RX, f_λ) -C_β-Q(j, f_RX, f _λ)] C
_β-I(j, f_RX, f_λ)} Is expanded.

[0072]

[Equation 12]

Observing that the residual noise ξ _I (k) after filtering by the filter 176 is not correlated with the residual noise ξ _Q (k) after filtering by the filter 182, the low pass filtering is It is possible to remove all but the DC term and the low-frequency part of [ξ _I (k)] ² , as reflected in.

[0074]

[Equation 13]

Then, R _QQ (j, f _RX ) Acos Δ =
Equilibrium is obtained when 1 is limited by the filtered [ξ _I (k)] ² .

The quadrature error measuring unit 154 has a multiplier 186.
And a digital filter 188 and a sample rate compressor 190 are shown. Equation 13 is between the Q path channel sample U _Q [k] input to the quadrature error measuring unit 154 and the measurement parameter C _φ [j, f _RX , f _λ ] output by the quadrature error measuring unit 154. Express the relationship mathematically.

[0077]

[Equation 14]

Here, h _φ (k) is the digital filter 18
8 is a unit sample response, K is the number of samples used for the measurement.

Applying the equation expanded above for U _Q [k], Equation 14 mathematically illustrates the operation performed by multiplier 186. The multiplier 186 multiplies the value of U _Q [k] determined by Expression 4A by the cosine wave of f _λ output by the calibration reference generator 152.

[0080]

[Equation 15]

Filter 188 filters terms having frequencies higher than f _λ -f _OBB output by multiplier 186, as mathematically shown by Equation 14A below.

[0082]

[Equation 16]

Here, ξ _φ (k) is the residual noise after filtering by the filter 188.

The residual noise component can be further reduced by additional bandpass filtering centered on | f _λ -f _OBB |. The DC and low frequency contributions are
Since it is shifted to the high frequency side up to _λ, it is efficiently removed by filtering according to a known technique. Thus, this method allows the phase error correction parameter to be determined independently of the effects of DC and low frequency offset correction.

The output of the sample rate compressor 178 is input to the adaptive phase error correction algorithm 204. In one embodiment, the adaptive phase error correction algorithm 204 implements a stochastic gradient algorithm, as described above. Equation 15 below mathematically represents the operation performed by the adaptive phase error correction algorithm 204 to determine the next value of the correction parameter R _IQ (j, f _RX ).

[0086]

R _IQ [j, f _RX , f _λ ] = ρ _φ R _IQ [(j-1), f _RX , f _λ ] + α _φ C _β-1 (j, f _RX , f _λ ) C _φ (j, f _RX , f _λ ) (15) where ρ _φ is a delay constant and | ρ _φ | <1,
α _φ is a modified gain term.

In equation 15, the correction signal is the term [C
_β-Q (j, f _RX , f _λ )] and the term C _β-I (j, f _RX , f _λ ), both terms being the calibration signal at the frequency [f _OBB −f _λ ]. Contains the components due to. The resulting product is proportional to the quadrature error plus some degradation of the DC term due to residual narrowband thermal noise, DC and 2
The component of [f _OBB −f _λ ] is included. This modified signal is further filtered by Eq. 15 if the repetition rate and decay constant are chosen to remove the components of frequency 2 [f _OBB −f _λ ]. This technique effectively implements a narrow bandpass filter centered on [f _{O BB} −f _λ ] which represents the deviation of the zero frequency from the surrounding range, thereby providing a DC offset and an input signal U _I [k ] And U _Q [k] with low frequency degradation. For example, Equation 16 below illustrates the term {C _φ (j, f _RX , f _λ ) C to show the bandpass nature of the operation performed by the adaptive phase error correction algorithm 204.
Extend _β-Q (j, f _RX , f _λ )}.

[0088]

[Equation 18]

Observing that the residual noise ξ _N (k) after filtering by the filter 188 is not correlated with the residual noise ξ _I (k) after filtering by the filter 176, Equation 16 becomes the following Equation 16A.

[0090]

[Formula 19]

Then, the adaptive phase error correction algorithm
Mu 204 is R_IQ(j, f_RX) =-R _QQ(j, f_RX) Asi
R as nΔ_IQ(j, f_RX) Adjust the next value
It

The Q offset measuring device 156 includes a digital filter 192 and a sample rate compressor 194. Formula 17 is the Q path channel sample UQ [k] input to the Q offset measuring device 156 and the measurement parameter C _Q [j, f] output by the Q offset measuring device 156.
Mathematically expresses the relationship with _RX ].

[0093]

[Equation 20]

Where h _Q (k) is the digital filter 19
2 is the unit sample response of K, where K is the sample rate compressor 1
The number of samples used for the measurement by 94.

The output of the Q offset measurer 156, in one embodiment, is input to the Q offset adaptive correction algorithm 206, which implements the stochastic gradient algorithm as described above. Equation 18 below is a modified parameter R _Q-Offset (j,
Mathematically represent the operations performed by the Q offset adaptive modification algorithm 206 to determine the next value of f _RX ).

[0096]

_RQ-Offset [j, f _RX ] = ρ _{dc RQ-Offset} [(j−1), f _RX ] + α _dc C _Q (j, f _RX ) ... Equation 18 Here, ρ _dc is Is the delay constant and α _dc is the modified gain term.

FIG. 5 is a flowchart showing an embodiment of the present invention. At block 220, a set of I-path digital samples Y _I [m] and Q-path digital samples Y _Q [m] of the calibration signal at baseband frequency f _OBB is generated.
At block 222, the I offset modification parameter is I
The path is added to the digital sample. Block 22
At 4, the Q offset correction parameter is added to the Q path digital samples. This sum is multiplied by the gain imbalance correction parameter and the result is combined with the phase error correction parameter. At block 226, the modified I path sample is multiplied by a cosine wave of f _λ , and the result is filtered and subsampled. At block 228, the modified Q path sample is multiplied by a sinusoid of f _λ and the result filtered and subsampled. At block 230, the next value of the gain imbalance parameter is determined based on the previous value and the results of blocks 226,228.

FIG. 6 is a flow chart showing a second embodiment of the present invention which may be combined with or used separately from the embodiment shown in FIG. In block 240, I path digital samples Y _I [m] and Q path digital samples Y _Q [m] of the calibration signal at baseband frequency f _OBB .
Are generated. At block 242, the I offset correction parameter is added to the I path digital samples. At block 244, the Q offset correction parameter is added to the Q path digital samples. This sum is multiplied by the gain imbalance correction parameter and the result is combined with the phase error correction parameter. At block 246, the modified I-path sample is multiplied by a cosine wave of f _λ , and the result is filtered and subsampled. At block 248, the modified Q path sample is multiplied by a cosine wave of f _λ , and the result is filtered and subsampled. Block 2
At 50, the next value of the gain imbalance parameter is determined based on the previous value and the results of blocks 226, 228.

The functional blocks and method steps described herein may be distributed by a variety of media. In one embodiment, a general purpose microprocessor executing software code is used to perform all or part of the functions. Alternatively, some or all of the functionality may be implemented in an application specific integrated circuit (ASIC). As long as the I and Q channels are orthogonal to each other, the I and Q channel functions can be reversed.

The present invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are exemplary in all respects,
It is intended that the scope of the invention be defined by the appended claims rather than by the foregoing description. All changes that come within the meaning and range of equivalency of the claims are to be embraced within their scope.

[Brief description of drawings]

FIG. 1 is a block diagram of a communication device including a direct conversion receiver and an adaptive calibration device of the present invention.

FIG. 2 is a block diagram of a direct conversion receiver of the present invention.

FIG. 3 is a block diagram showing the modification and measurement circuit of a direct conversion receiver in great detail.

FIG. 4 is a block diagram showing part of a measurement circuit and some of the functions performed by a digital processor.

FIG. 5 is a flowchart for determining a gain imbalance parameter.

FIG. 6 is a flowchart for determining a phase offset error parameter.

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5K004 AA05 AA08 FD05 FH03 FH04                       JD05 JH02 JH03

Claims (12)

[Claims] 1. A coherent adaptive calibration method for calibrating a receiver, wherein a first channel offset correction parameter is added to a sequence of first channel signal samples to obtain a corrected sequence of first channel samples. Generating, and adding a second channel offset correction parameter to the sequence of second channel signal samples and multiplying by a gain imbalance correction parameter, the corrected sequence of the first channel samples and the phase error correction parameter. To produce a modified set of second channel samples,
Said second channel being orthogonal to said first channel; multiplying a modified sequence of said first channel samples with a first sinusoidal waveform to determine a first product; Multiplying the modified sequence of two channel samples by a second sinusoidal waveform that is 90 degrees out of phase with the first sinusoidal waveform to determine a second product; and filtering the first product. ) To determine a first channel gain imbalance amount; to filter the second product to determine a second channel gain imbalance amount; and to determine the first and second channel gain imbalance amounts. Determining a next value of the gain imbalance correction parameter based on the coherent adaptive calibration method. 2. The method of claim 1, wherein the step of determining the next value is performed by directly taking the difference between the first and second channel gain imbalance amounts. 3. The method of claim 1, wherein the sequence of channel samples includes a calibration signal received over the air. 4. The method of claim 1, wherein the sequence of channel samples includes energy received over the air and a calibration signal from an internal source. 5. The method of claim 1, wherein the step of determining the next value of the gain imbalance correction parameter is performed using a stochastic gradient equation. 6. The step of determining the first product, the step of determining the second product, the step of determining the first channel gain imbalance amount, and the step of determining the second channel gain imbalance amount. The method of claim 1, wherein the determining step is performed digitally. 7. A modified sequence of said second channel samples is digitally multiplied by said first sinusoidal waveform to produce a third
A step of determining a phase error amount by digitally filtering the third product, and a step of determining the phase error amount based on the first gain imbalance amount and the phase error amount. The method of claim 6, further comprising: determining a next value. 8. A first summing module configured to add a first channel offset correction parameter to the series of first channel signal samples to produce a modified series of first channel samples. A second channel offset correction parameter is added to the sequence of second channel signal samples and multiplied by a gain imbalance correction parameter and added to the product of the corrected sequence of the first channel sample and the phase error correction parameter. A second summing module configured to generate a modified set of the second channel samples, the second channel being orthogonal to the first channel; and the first channel. Means for multiplying a modified sequence of samples by a first sinusoidal waveform to determine a first product; and the modified of the second channel samples. Means for determining the third product by multiplying the sequence by the first sine waveform; means for determining the first channel gain imbalance amount by filtering the first product; and A receiver comprising: means for determining a phase error amount by filtering; and means for determining a next value of the phase correction parameter based on the first channel gain imbalance amount and the phase error amount. 9. A first summing module configured to add a first channel offset correction parameter to the series of first channel signal samples to produce a modified series of first channel samples. A second channel offset correction parameter is added to the sequence of second channel signal samples and multiplied by a gain imbalance correction parameter and added to the product of the corrected sequence of the first channel sample and the phase error correction parameter. A second summing module configured to generate a modified set of the second channel samples, the second channel being orthogonal to the first channel; and the first channel. Means for multiplying a modified sequence of samples by a first sinusoidal waveform to determine a first product; and a modified system of the second channel samples. Means for determining a second product by multiplying the column by a second sine waveform whose phase is 90 degrees different from the first sine waveform; and filtering the first product to determine a first channel gain imbalance amount. Determining means for filtering the second product to determine a second amount of channel gain imbalance, and determining a second of the gain imbalance correction parameters based on the first and second channel gain imbalance amounts. A receiver having means for determining a value; 10. The means for determining the first product, the means for determining the second product, the means for determining the first channel gain imbalance amount, and the second channel gain imbalance amount. 10. The receiver according to claim 9, wherein the means for determining is digital. 11. A first channel offset correction parameter is added to a sequence of first channel signal samples to obtain a first channel offset correction parameter.
A first adder configured to generate a modified sequence of the first channel samples, and a first adder configured to multiply the modified sequence of the first channel samples by a phase error correction parameter. A first adder, a second adder configured to add a second channel offset correction parameter to the sequence of second channel signal samples, and a gain imbalance correction at the output of the second adder. A second multiplier that is calibrated to multiply by a value, and the output of the first multiplier and the output of the second multiplier are added to produce a modified set of second channel samples. A second adder, the second channel being orthogonal to the first channel, and a first digitized into a modified sequence of the first channel samples. sine A third multiplier configured to multiply a shape; and a second series of modified sequences of the second channel samples that are 90 degrees out of phase with the first digitized sine waveform. A fourth multiplier configured to multiply a digitized sine waveform, and a fourth multiplier configured to filter an output of the third multiplier to determine a first channel gain imbalance amount. A first digital filter, a second digital filter configured to filter an output of the fourth multiplier to determine a second channel gain imbalance amount, the first and second channel gains A calculator that determines a next value of the gain imbalance correction parameter based on the amount of imbalance. 12. A first channel offset correction parameter is added to a sequence of first channel signal samples to obtain a first channel offset correction parameter.
A first adder configured to generate a modified sequence of the first channel samples, the first adder configured to multiply the modified sequence of the first channel samples with a phase error correction parameter. 1
And a second adder configured to add a second channel offset correction parameter to the sequence of second channel signal samples, and a gain imbalance correction value at the output of the second adder. A second multiplier configured to multiply by, and adding the output of the first multiplier and the output of the second multiplier to produce a modified set of second channel samples. A second adder, wherein the second channel is orthogonal to the first channel, a third adder, and a first digitized sine into a modified sequence of the first channel samples. A third multiplier configured to multiply a waveform, and configured to multiply the modified sequence of second channel samples by the first sinusoidal waveform to determine a third product. A fourth multiplier, and the third A first digital filter configured to filter the output of the calculator to determine a first channel gain imbalance amount, and to filter the output of the fourth multiplier to determine a phase error amount A second digital filter configured according to, and a calculator configured to determine a next value of the phase correction parameter based on the first channel gain imbalance measurement and the phase error amount. And a receiver equipped with.