METHOD AND APPARATUS FOR COMPENSATING FOR TRANSCEIVER IMPAIRMENTS
CROSS-REFERENCE TO RELATED APPLICATIONS
Not applicable.
BACKGROUND
The disclosed subject matter relates generally to telecommunication and  more particularly  to a method and apparatus for compensating for transceiver impairments.
In telecommunication  direct-conversion techniques involve mixing a baseband signal with a carrier signal without using an intermediate frequency. A direct conversion receiver (DCR)  also known as homodyne  synchrodyne  or zero-IF receiver  may then demodulate incoming signals by mixing the incoming signal with a local oscillator signal synchronized in frequency to the carrier signal. The baseband signal may then be obtained simply by low-pass filtering the mixer output  without requiring further detection.
A direct conversion transceiver may be implemented on a single chip  making it inexpensive and versatile. However  the transmission and receive paths for the base band signals (i.e.  in-phase (I) and quadrature (Q)) are independently formed. Small variances in these paths introduce amplitude and phase variations in the respective transmit and receive signals  otherwise referred to as IQ imbalance impairments.
To calibrate a direct conversion transceiver  the impairments must be identified and compensated for prior to use. This compensation is difficult because it is necessary to calibrate the direct conversion transmitter and direct conversion receiver individually using a separate external reference for each path. If an uncompensated receiver is used to calibrate a transmitter over the complete frequency span  the transmitter calibration will be corrupted by the impairments of the receiver  and vice versa.
This section of this document is intended to introduce various aspects of art that may be related to various aspects of the disclosed subject matter described and/or claimed below. This section provides background information to facilitate a better understanding of the various aspects of the disclosed subject matter. It should be understood that the statements in this section of this document are to be read in this light  and not as admissions of prior art. The disclosed subject matter is directed to overcoming  or at least reducing the effects of  one or more of the problems set forth above.
BRIEF SUMMARY
The following presents a simplified summary of the disclosed subject matter in order to provide a basic understanding of some aspects of the disclosed subject matter. This summary is not an exhaustive overview of the disclosed subject matter. It is not intended to identify key or critical elements of the disclosed subject matter or to delineate the scope of the disclosed subject matter. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is discussed later.
One aspect of the disclosed subject matter is seen in a method for compensating a transceiver for impairments. The method includes transmitting a plurality of partial bandwidth training signals using a transmitter. A plurality of response signals of a receiver having a bandwidth and exhibiting receiver impairments is captured. Each response signal is associated with one of the partial bandwidth training signals. Each of the partial bandwidth training signals is associated with a portion of the receiver bandwidth. A plurality of partial compensation filters is generated based on the plurality of response signals. Each partial compensation filter is associated with one of the response signals. The partial compensation filters are combined to configure a receiver compensation filter operable to compensate for the receiver impairments.
Another aspect of the disclosed subject matter is seen a transceiver including a transmitter  a receiver  a receiver capture unit  a receiver compensation filter  and a receiver compensation estimation unit. The transmitter is operable to transmit a plurality of partial bandwidth training signals. The receiver has impairments and is operable to receive the plurality of partial bandwidth training signals. The receiver capture unit is operable to capture a plurality of response signals of the receiver. Each response signal is associated with one of the partial bandwidth training signals. Each of the partial bandwidth training signals is associated with a portion of the receiver bandwidth. The receiver compensation filter is operable to filter signals received by the receiver. The receiver compensation estimation unit is operable to generate a plurality of partial compensation filters based on the plurality of response signals. Each partial compensation filter is associated with one of the response signals. The receiver compensation estimation unit is operable to combine the partial compensation filters to configure the receiver compensation filter to compensate for the receiver impairments.
BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS
The disclosed subject matter will hereafter be described with reference to the accompanying drawings  wherein like reference numerals denote like elements  and:
Figure 1 is a simplified block diagram of a transceiver in accordance with one illustrative embodiment of the present subject matter;
Figures 2A-2C illustrate training signals used for compensating a receiver in the transceiver of Figure 1;
Figure 3 is a diagram modeling receiver impairments for the transceiver of Figure 1; and
Figure 4 is a diagram of a full bandwidth training signal for compensating a transmitter in the transceiver of Figure 1.
While the disclosed subject matter is susceptible to various modifications and alternative forms  specific embodiments thereof have been shown by way of example in the drawings and are herein described in detail. It should be understood  however  that the description herein of specific embodiments is not intended to limit the disclosed subject matter to the particular forms disclosed  but on the contrary  the intention is to cover all modifications  equivalents  and alternatives falling within the spirit and scope of the disclosed subject matter as defined by the appended claims.

DETAILED DESCRIPTION
One or more specific embodiments of the disclosed subject matter will be described below. It is specifically intended that the disclosed subject matter not be limited to the embodiments and illustrations contained herein  but include modified forms of those embodiments including portions of the embodiments and combinations of elements of different embodiments as come within the scope of the following claims. It should be appreciated that in the development of any such actual implementation  as in any engineering or design project  numerous implementation-specific decisions must be made to achieve the developers"" specific goals  such as compliance with system-related and business related constraints  which may vary from one implementation to another. Moreover  it should be appreciated that such a development effort might be complex and time consuming  but would nevertheless be a routine undertaking of design  fabrication  and manufacture for those of ordinary skill having the benefit of this disclosure. Nothing in this application is considered critical or essential to the disclosed subject matter unless explicitly indicated as being “critical” or “essential.”
The disclosed subject matter will now be described with reference to the attached figures. Various structures  systems and devices are schematically depicted in the drawings for purposes of explanation only and so as to not obscure the disclosed subject matter with details that are well known to those skilled in the art. Nevertheless  the attached drawings are included to describe and explain illustrative examples of the disclosed subject matter. The words and phrases used herein should be understood and interpreted to have a meaning consistent with the under¬standing of those words and phrases by those skilled in the relevant art. No special definition of a term or phrase  i.e.  a definition that is different from the ordinary and custom¬ary mean¬ing as understood by those skilled in the art  is intended to be implied by consistent usage of the term or phrase herein. To the extent that a term or phrase is intended to have a special meaning  i.e.  a meaning other than that understood by skilled artisans  such a special defini¬tion will be expressly set forth in the specification in a defini¬tional manner that directly and unequivocally provides the special definition for the term or phrase.
Referring now to the drawings wherein like reference numbers correspond to similar components throughout the several views and  specifically  referring to Figure 1  the disclosed subject matter shall be described in the context of a transceiver 100 including a transmit path 110  a receive path 120  and a sampling receive path 130. The transmit path 110 and the sampling receive path 130 are connected to a directional coupler 140  which is in turn connected to a filter panel 150 and an antenna 160. The directional coupler 140 prevents reflections from the filter panel 150 from affecting the sampling receive path 130. The filter panel 150 (i.e.  commonly referred to as a duplexer) separates the outgoing transmit signal from the incoming receive signal  performs bandpass filtering of the transmit RF output  and performs bandpass filtering of the receiver RF input.
The transmit path 110 includes a transmitter 111 including a modulator 112 and a digital-to-analog converter (DAC) 113  a transmitter IQ compensation filter 114  a transmitter IQ compensation estimation unit 115  a transmitter capture unit 116  a power amplifier 117  and a switch 118 for selecting between normal data traffic or a training signal 119. The receive path 120 includes a receiver 121 including a demodulator 122 and an analog-to-digital converter (ADC) 123  a receiver IQ compensation filter 124  a receiver IQ compensation estimation unit 125  and a receiver capture unit 126. The sampling receive path 130 includes a sampling receiver 131 including a demodulator 132 and an ADC 133  a sampling receiver IQ compensation filter 134  a sampling receiver IQ compensation estimation unit 135  and a sampling receiver capture unit 136. The transmitter 111  receiver 121  and sampling receiver 131 are implemented using hardware  while the various units illustrated in Figure 1 are implemented in the digital portion of the transceiver 100 (i.e.  by a processing unit executing software or firmware). Hence  the units generally describe functions rather than discrete hardware. However  depending on the particular implementation  dedicated hardware may be employed to perform one or more of the functions attributed to the units.
The transmitter 111  receiver 121  and sampling receiver 131 employ inphase (I) and quadrature (Q) signals. The separate paths for these components are not shown for ease of illustration. Generally  the I and Q paths are separate  such that amplitude and phase variations exist between the paths. These variations cause separate impairments in the transmitter 111  the receiver 121  and the sampling receiver 131. The transmitter 111 and the sampling receiver 131 operate in the same frequency band. In the illustrated embodiment  the sampling receiver 131 is employed to compensate for transmitter impairments. After the transmitter 111 is compensated  it may be used to generate a training signal for compensating the receiver 121. However  because the sampling receiver 131 is used to estimate and compensate the transmitter 111  the impairments of the sampling receiver 131 must first be estimated and compensated. As known to those of ordinary skill in the art  the sampling receiver 131 may also be employed for functions such as digital predistortion or time division duplexing (i.e.  functioning as the only receiver).
For a typical IQ transmitter 111  the unimpaired baseband signal may be represented by:
  (1)
where   are the baseband in-phase and quadrature signal components.
The output of the modulator 112 is given by:
  (2)
where
  (3)
and are the modulator imbalance errors (i.e.  impairments).
Equation (2) can be rewritten in the form of a summation of two components comprised of a desired signal and its conjugate 
  (4)
where
. (5)
The complex conjugate of is given by:
  (6)
and by using and as separate observations  the compensation required to minimize the imbalance errors in can be determined by exploiting the conjugate symmetry nature of the IQ imbalance. For purposes of illustration  the imbalance errors are provided here as scalar errors  but generally  the imbalance errors are functions of frequency .
To perform the compensation analysis  this symmetrical nature is employed by generating partial bandwidth training signals and measuring the full bandwidth response signal. Signal components arising from impairments are detected in the portion of the bandwidth not covered by the partial bandwidth training signal. The partial bandwidth process is continued until the full bandwidth has been covered  and the response signals from the partial bandwidth iterations are combined to compensate the receiver over the full bandwidth. For purposes of the following examples  a half bandwidth technique is used (i.e.  two iterations)  however  it is contemplated that more than two partial bandwidth partitions may be employed.
To facilitate compensation  the transmitter 111 is configured to select the training signal 119 using the switch 118 rather than the normal data traffic. Figure 2A illustrates an exemplary transmit signal 200 that covers half of the calibration bandwidth. In the illustrated embodiment  the signal 200 is an arbitrary wideband signal that need not be flat with respect to frequency. Figure 2b shows the response of the receiver 131 to the transmit signal 200. The actual transmit signal 210 transmitted by the modulator 112 and received by the receiver 131 is shown in Figure 2B. Note that a transmitter impairment signal 220 is also present due to the IQ imbalances of the transmitter 111. The transmit signal 210 shown in Figure 2B is transmitted by setting the local oscillator frequency of the modulator 112 to have a negative center frequency offset by approximately one fourth of the calibration bandwidth (i.e.  –BW/4) from the center frequency of the sampling receiver 131  which places the transmit signal 210 into the lower half bandwidth of the sampling receiver 131. Due to the symmetrical nature of the IQ imbalance  the resulting receiver impairment signal 230 occupies the upper half-bandwidth and is separated from the transmitter impairment signal 220. Note that the receiver impairment signal 230 includes an impairment signal 240 representing impairments of the sampling receiver 131 arising from the transmit signal 200 and a receiver impairment signal 250 representing the impairments of the sampling receiver 131 arising from the transmitter impairment signal 220.
A second iteration is performed to cover the second half of the calibration bandwidth  as shown in Figure 2C. The transmit signal 210’ (i.e.  including a transmit signal 200’ and a transmitter impairment signal 220’) is transmitted by setting the local oscillator frequency of the modulator 112 to have a positive center frequency offset by approximately one fourth of the calibration bandwidth (i.e.  +BW/4) from the center frequency of the sampling receiver 131  which places the transmit signal 210’ into the upper half bandwidth of the sampling receiver 131. Again  due to the symmetrical nature of the IQ imbalance  the resulting receiver impairment signal 230’ occupies the lower half-bandwidth and is separated from the transmitter impairment signal 220’. The receiver impairment signal 230’ includes an impairment signal 240’ representing impairments of the sampling receiver 131 arising from the transmit signal 200’ and a receiver impairment signal 250’ representing the impairments of the sampling receiver 131 arising from the transmitter impairment signal 220’.
For each partial bandwidth iteration  the sampling receiver capture unit 136 records the response signal. Each partial bandwidth capture is used to generate a compensation filter for the corresponding bandwidth. The sampling receiver IQ compensation estimation unit 135 uses the capture information to generate coefficients for the sampling receiver IQ compensation filter 134.
Figure 3 illustrates an exemplary model 300 of the sampling receiver 131 impairments. The impairments are modeled by coefficients h1  h2 in blocks 310  320. The modeled impairments h1  h2 are added into the desired signals   by adders 330  340  resulting in the impaired baseband signals   . The compensation conducted by the sampling receiver IQ compensation estimation unit 135 generates filter coefficients w1  w2 in blocks 350  360  which are subtracted from the impaired baseband signals in adders 370  380 to attempt to remove the impairments  resulting in received signals   . The decorrelation criteria block 390 represents the convergence criteria for the decorrelation algorithm. When the decorrelation criteria are met  the iterative adjustment of the filter taps is terminated.
As indicated above  the baseband signal can be written as the summation of the desired signal and its conjugate . The conjugate component after the down conversion process appears as an image opposite DC in the baseband frequency domain. To perform a compensation  the conjugate contribution in may be minimized though a blind adaptive estimation of and shown in Equation (5). One exemplary method to separate mixed signals is to use second order statistics to observe and minimize the correlation between the sources  provided that the source and its conjugate are uncorrelated. The signals are assumed to be zero mean  complex Gaussian random processes. The correlation for these signals is given by:
. (6)
Separating in phase and quadrature components in the correlation of Equation (6)  results in:
(7)
The imaginary part of the correlation is zero if the in-phase and quadrature components of the signal are uncorrelated. The real part of the correlation is zero if the autocorrelation of both in-phase and quadrature components are the same. Although in-phase and quadrature component autocorrelation is not exactly the same for typical 3G and 4G broadband signals  the correlation is small  such that minimization through an adaptive scheme is possible.
The decorrelation of the received signals requires that . This condition further leads to
  (8)
where the mixing coefficients are given by and   and   are autocorrelation coefficients of and   respectively. The mixing coefficients are conjugates of each other  which implies . Assuming a simple case with only one tap to estimate  this is readily accomplished with a straightforward minimization algorithm that is similar to a Newton zero search  where the zero is in the cross correlation domain.
In a simple example  the sampling receiver IQ compensation estimation unit 135 may use the following signal separation procedure.
1. Capture signal into sampling receiver capture unit 136.
2. Form and .
3. Form Signal Estimates.
. (9)
4. Update the coefficient weights.
  (10)
where is a convergence parameter for the LMS algorithm  and is an autocorrelation coefficient of .
The IQ imbalance observed on typical commercial modulators and demodulators is frequency dependent  and a single tap solution is not usually adequate to compensate the IQ imbalance image across a wide bandwidth. To accommodate wide bandwidth compensation  the expressions in equations (6) through (10) can be vectorized generate a compensation filter of length  
    (1)
where
.
If a solution exists to decorrelate and   then
(2)
over the filter span . This implies that the filter response . Although similar in appearance to an LMS form of adaptive equalization  this approach uses the cross correlation in a minimization search.
The steps described above consider the filters and to be odd symmetric  or conjugate filters. Real world demodulators are not symmetric  and exhibit varying frequency response on either side of the receiver center frequency.
To adequately compensate a demodulator using the blind methods outlined above  the sampling receiver IQ compensation estimation unit 135 uses a multiple step approach to build partial bandwidth filters (e.g.  two half-band filters). The partial bandwidth filters are then spliced together in the frequency domain to form the final full band compensation filter.
To generate the full bandwidth filter coefficients for the sampling receiver IQ compensation filter 134  the sampling receiver IQ compensation estimation unit 135 converts the time-domain coefficients of each partial bandwidth filter to the frequency domain by a discrete Fourier transform (DFT). The partial bandwidth response signals are joined in the frequency domain to form a frequency response that is defined over the entire receiver bandwidth. At intersection points (e.g.  f=0 for two half-bandwidth filters) the responses are averaged to define the midpoint of the full-bandwidth response. Then an inverse DFT is performed to convert the full-bandwidth response back to time domain coefficients.
After the sampling receiver IQ compensation filter 134 has been configured using the partial BW captures to build a full BW compensation filter for the sampling receiver 131  the transmitter 111 can be calibrated by using the compensated sampling receiver 131 as a training reference. As shown in Figure 4  the training signal 119 is now configured as a full bandwidth signal 400 including a transmit signal 410 and a transmitter impairment signal 420. However  compensation of the transmitter 111 is possible now that the sampling receiver 131 may be used as a compensated training reference.
The subsequent final compensation of the transmitter 111 may be conducted using conventional techniques known to those of ordinary skill in the art. For example  the full bandwidth training signal is captured by the transmitter capture unit 116 prior to being modulated by the modulator 112 (i.e.  prior to the transmitter impairments being introduced by the modulator 112). The signal received by the sampling receiver 131 is compensated by the sampling receiver IQ compensation filter 134 to remove the sampling receiver impairments. The filtered signal is then provided to the transmitter IQ compensation estimation unit 115  which compares the compensated received signal to the transmit signal captured by the transmitter capture unit 116. Prior to transmitter compensation  the compensated received signal includes transmitter impairments. By comparing the compensated received signal to the transmit signal  the transmitter IQ compensation estimation unit 115 identifies the impairments and generates coefficients for the transmitter IQ compensation filter 114 to compensate for the transmitter impairments. For clarity and to avoid obscuring the present subject matter  these conventional techniques are not described in greater detail herein.
The techniques described herein allow compensation of the transmitter 111 using the compensated receiver 131 without the use of an external training signal. This approach simplifies the compensation techniques  thereby potentially reducing the complexity and cost of the transceiver.
The particular embodiments disclosed above are illustrative only  as the disclosed subject matter may be modified and practiced in different but equivalent manners apparent to those skilled in the art having the benefit of the teachings herein. Furthermore  no limitations are intended to the details of construction or design herein shown  other than as described in the claims below. It is therefore evident that the particular embodiments disclosed above may be altered or modified and all such variations are considered within the scope and spirit of the disclosed subject matter. Accordingly  the protection sought herein is as set forth in the claims below.

CLAIMS
WE CLAIM:
1. A method for compensating a transceiver for impairments  comprising:
transmitting a plurality of partial bandwidth training signals using a transmitter;
capturing a plurality of response signals of a receiver having a bandwidth and exhibiting receiver impairments  each response signal being associated with one of the partial bandwidth training signals  wherein each of the partial bandwidth training signals is associated with a portion of the receiver bandwidth;
generating a plurality of partial compensation filters based on the plurality of response signals  each partial compensation filter being associated with one of the response signals; and
combining the partial compensation filters to configure a receiver compensation filter operable to compensate for the receiver impairments.

2. The method of claim 1  further comprising:
transmitting a full bandwidth training signal using the transmitter  wherein the transmitter has transmitter impairments;
capturing a second response signal of the receiver to the full bandwidth training signal;
filtering the second response signal using the receiver compensation filter to generate a filtered response signal and
comparing the full bandwidth training signal to the filtered response signal and configuring a transmitter compensation filter to compensate for the transmitter impairments based on the comparison.

3. The method of claim 2  further comprising:
filtering data to be transmitted by the transmitter using the transmitter compensation filter; and
filtering data received by the receiver using the receiver compensation filter.


4. The method of claim 1  wherein generating a selected one of the partial compensation filters comprises:
capturing the response signal of the receiver for the associated partial bandwidth training signal;
generating a conjugate signal of the response signal;
generating an estimation signal based on the response signal  the conjugate signal  and a plurality of filter coefficients;
iteratively adjusting the filter coefficients to minimize a contribution of the conjugate signal to the estimation signal; and
configuring the selected partial compensation filter using the adjusted coefficients.

5. The method of claim 1  wherein combining the partial compensation filters comprises:
transforming each partial compensation filter to the frequency domain;
joining the transformed partial compensation filters in the frequency domain; and
transforming the joined partial compensation filters to the time domain to generate the receiver compensation filter.

6. The method of claim 5  wherein joining the transformed partial compensation filters comprises averaging the transformed partial compensation filters at overlapping points.

7. A transceiver  comprising:
a transmitter operable to transmit a plurality of partial bandwidth training signals;
a receiver having impairments and being operable to receive the plurality of partial bandwidth training signals;
a receiver capture unit operable to capture a plurality of response signals of the receiver  each response signal being associated with one of the partial bandwidth training signals  wherein each of the partial bandwidth training signals is associated with a portion of the receiver bandwidth;
a receiver compensation filter operable to filter signals received by the receiver;
a receiver compensation estimation unit operable to generate a plurality of partial compensation filters based on the plurality of response signals  each partial compensation filter being associated with one of the response signals  and combine the partial compensation filters to configure the receiver compensation filter to compensate for the receiver impairments.

8. The transceiver of claim 7  wherein the transmitter has impairments and is further operable to transmit a full bandwidth training signal  the capture unit is operable to capture a second response signal of the receiver to the full bandwidth training signal  the receiver compensation filter is operable to filter the second response signal to generate a filtered response signal  and the transceiver further comprises:
a transmitter compensation filter operable to filter signals sent by the transmitter;
a transmitter capture unit operable to capture the full bandwidth training signal; and
a transmitter compensation estimation unit operable to compare the captured full bandwidth training signal to the filtered response signal and configure the transmitter compensation filter to compensate for the transmitter impairments based on the comparison.

9. The transceiver of claim 7  wherein the receiver compensation estimation unit is operable to generate a selected one of the partial compensation filters by generating a conjugate signal of the response signal captured by the receiver capture unit for the associated partial bandwidth training signal  generating an estimation signal based on the response signal  the conjugate signal  and a plurality of filter coefficients  iteratively adjusting the filter coefficients to minimize a contribution of the conjugate signal to the estimation signal  and configuring the selected partial compensation filter using the adjusted coefficients.

10. The transceiver of claim 7  wherein the receiver compensation estimation unit is operable to combine the partial compensation filters by transforming each partial compensation filter to the frequency domain  joining the transformed partial compensation filters in the frequency domain  and transforming the joined partial compensation filters to the time domain to generate coefficients for the receiver compensation filter.

ABSTRACT OF THE DISCLOSURE

A method for compensating a transceiver for impairments includes transmitting a plurality of partial bandwidth training signals using a transmitter. A plurality of response signals of a receiver having a bandwidth and exhibiting receiver impairments is captured. Each response signal is associated with one of the partial bandwidth training signals. Each of the partial bandwidth training signals is associated with a portion of the receiver bandwidth. A plurality of partial compensation filters is generated based on the plurality of response signals. Each partial compensation filter is associated with one of the response signals. The partial compensation filters are combined to configure a receiver compensation filter operable to compensate for the receiver impairments.