Patent Application
20120140685
Over the past decade, the use of wireless communication devices has witnessed enormous growth to become commonplace in the daily lives of many. Many modern wireless communication devices (e.g., cell phones, PDAs, etc.) utilize transceivers having both a transmitter path (i.e., transmission chain) configured to transmit data and a receiver path (i.e., receiver chain) configured to receive data over radio frequencies.
Intermodulation noise or distortion may arise during the operation of such wireless communication devices. For example, second order intermodulation noise may occur in receiver chain when a modulated blocker passes a component with a nonlinear characteristic to form a spurious signal in the receiver chain. In such a case, the spurious signal within a receiver chain may contain harmful signal components that are detrimental to operation of the transceiver device.
The present invention will now be described with reference to the attached drawing figures, wherein like reference numerals are used to refer to like elements throughout, and wherein the illustrated structures and devices are not necessarily drawn to scale.
During full duplex mode operation, the transmitter section 104 typically uses one carrier frequency in a given frequency band (e.g., 900 MHz, 1800 MHz, etc.) while the receiver section 102 uses another carrier frequency in the frequency band. Despite using different frequencies, intermodulation distortion may generate spurious signals (e.g., additional signals at frequencies that are not at harmonic frequencies of a received signal, but are instead at a sum and difference of the original signal frequency) in a receiver path that interfere with the operation of receiver signals. Therefore, a receiver 102 may be susceptible to intermodulation distortion (i.e., intermodulation noise) from a transmitted signal.
Intermodulation distortion can have a harmful effect on the operation of modern wireless communication systems. For example, in direct conversion receivers, second order intermodulation distortion (IM2) from the transmitted signal is a significant source of interference since it falls in the baseband occupied by a down converted receive signal.
Therefore, to minimize intermodulation noise in modern wireless communication systems an adaptive filter 110 (e.g., operating a least mean square (LMS) algorithm) may be configured to cancel intermodulation noise induced by the transmitter. The adaptive filter 110 is configured to estimate an intermodulation noise that is present in the receiver path 102, based upon an input signal from the transmission path 104, and to cancel the estimated noise from the receiver path 102. For example, the adaptive filter 110 may be used to operate a LMS algorithm that estimates and cancels the transmitter induced second-order intermodulation distortion (IM2). However, since the transmission path comprises a complex transmission baseband signal, normally an adaptive filter configured to implement a complex adaptive algorithm is needed.
Mathematically, this is described by generating an adaptive filter output signal y(n) that is equal to the input signal u(n) convolved with a tap weight vector w(n) (i.e., y(n)=wH(n)×u(n)). The iterative nature of the process relies upon iteratively searching for a tap weight vector w(n) that optimizes the filtering operation. Therefore, a tap weight vector w(n) may be iteratively updated by stepping it by a value that is equal to the product of a step size μ (i.e., a convergence factor controlling the rate of adaption), the input signal u(n), and an error e(n) that is equal to the instantaneous difference between the output signal y(n) and a desired signal d(n) (i.e., w(n+1)=w(n)+μu(n)e(n), wherein e(n)=d(n)−y(n)).
As shown in
The inventors have appreciated that removal of even-order TX intermodulation noise in a transceiver configured to operate in a full-duplex mode can be simplified through simplification of the adaptive filtering. Accordingly, a simplified adaptive filtering method and apparatus are provided herein. In one embodiment, the adaptive filtering apparatus comprises first and second real valued adaptive filters, respectively configured to receive an adaptive filter input signal based upon a baseband transmission signal in a transmission path. The first real valued adaptive filter is configured to operate a real valued adaptive filter algorithm on the input signal to estimate a first intermodulation noise component in a desired signal (e.g., a noise component that is distorting the in-phase component of a received baseband signal) and to cancel the estimated noise. The second real valued adaptive filter is configured to operate a real valued adaptive filter algorithm on the same input signal to estimate a second intermodulation noise component in the desired signal (e.g., a noise component that is distorting the quadrature phase component of a received baseband signal) and to cancel the estimated noise. Accordingly, each filter operates a real valued adaptive algorithm to cancel a component of an intermodulation noise generated by a transmission baseband signal, thereby removing complex cross terms between the components from the adaptive filtering process.
In one particular embodiment, an adaptive filtering apparatus is configured to comprise a first adaptive filter and a second adaptive filter respectively configured to receive an input signal and estimate in-phase (I) and quadrature phase (Q) components of the intermodulation noise. The first adaptive filter is configured to operate a real valued adaptive algorithm on the input signal to estimate an in-phase component of the intermodulation noise (e.g., a noise component that is distorting the in-phase component of a received baseband signal). The second adaptive filter is configured to operate the real valued adaptive algorithm on the input signal to estimate a quadrature-phase component of the intermodulation noise (e.g., a noise component that is distorting the quadrature-phase component of a received baseband signal). Therefore, the adaptive filters are configured to act independently of each other to filter the I and Q components of the intermodulation noise, so that the noise can be cancelled and filter coefficients can be updated for each filter independent of the other filter.
In particular, the transceiver 300 comprises an antenna 302 that is shared between a receiver and a transmitter section. The antenna 302 is coupled to a duplexer 304 that allows the transceiver 300 to operate in a continuous transmission/reception mode (e.g., a full-duplex mode). The duplexer may also be configured to reduce interference between the receiver and transmitter sections by selectively providing isolation (e.g., 50-60 dB of isolation) between the transmitter and receiver sections.
The receiver section may comprise one or more low noise amplifiers 314 and a mixer 316 (e.g., down conversion module) configured to convert an inbound RF signal into the inbound baseband or near baseband signal. The RF transmitter section may comprise an up-conversion module 320 and one or more power amplifiers 322 configured to convert an outbound baseband or near baseband signal into an outbound RF signal.
An adaptive filtering system 306 is configured to generate a plurality of system output signals SOUT that are corrected to remove components of the intermodulation noise (e.g., an in-phase component and a quadrature phase component) generated in the non-ideal mixer 316 of the receiver path by a transmitted signal (e.g., parts of the TX signal leak via the duplexer into the receiver path and second order intermodulation noise is produced by the nonlinear characteristic of 316).
In one embodiment, the adaptive filtering system 306 comprises a first filtering path and a second filtering path configured to respectively generate system output signals, which can be collectively considered as a baseband signal, having different noise components removed. In one embodiment, the first filtering path is configured to generate a first system output signal, comprising a signal that is corrected to remove a first intermodulation noise component (e.g., a noise component that is distorting the in-phase component of a desired signal) generated by the transmission signal, and a second filtering path is configured to generate a second system output signal, comprising a signal that is corrected to remove a second intermodulation noise component (e.g., a noise component that is distorting the quadrature phase component of a desired signal) generated by the transmission signal.
It will be appreciated that by the use of first and second adaptive filtering paths that are independent of each other, intermodulation noise introduced into one path will have no affect on the other path. For example, intermodulation noise leaked into the I path will not be “seen” by the adaptive filter in the Q path (i.e., the filter in the Q path is not influenced by this noise). In contrast, if one complex filter (with cross-coupling between I and Q) is used, the filter output for the Q path will show a reaction on changes of the noise leaked into in the I path. Therefore, the use of two adaptive filters having no cross-coupling between the filtering paths can be detected if changing IM2 noise provided to one filtering path has no influence on the output signal of the other path.
In one embodiment, the adaptive filtering system 306 may be comprised within a digital front end (DFE) 308 having a first filtering path comprising adaptive filter 312a and a second filtering path comprising adaptive filter 312b. The adaptive filters 312a and 312b are respectively configured to estimate an intermodulation noise component based upon an adaptive filter input signal SIN from the transmitter path and to subtract the estimated intermodulation noise component from a receiver signal SRX. For example, the input signal SIN is provided to a first adaptive filter 312a configured to operate a real valued adaptive algorithm to iteratively correct for a first noise component caused by the transmission signal. The input signal is also provided to a second adaptive filter 312b configured to operate a real valued adaptive algorithm to iteratively correct for a second noise component caused by the transmission signal.
In one embodiment, the DFE 308 may be configured to use an LMS algorithm to cancel the IM2 noise generated by the transmission signal at the output from mixer 316. LMS algorithms can be used by adaptive filters to find filter coefficients that produce the least mean squares of the error signal (e.g., difference between the desired and the actual signal). In particular, LMS adaptive algorithms may generate a compensation signal, which may be used to eliminate intermodulation distortion by appending the compensating signal to the receiver signal SRX.
It will be appreciated that the even ordered intermodulation noise is proportional to the envelope of the transmission signal (e.g., second order intermodulation noise distorting the receivers' baseband signal is proportional to the squared envelope of the TX signal, fourth order intermodulation noise is proportional to the 4th power of the envelope of the TX signal, etc.) Therefore, the adaptive filter input signal SIN may be computed by an exponential circuit component 310 configured to generate an even power (e.g., n=2, 4, 6, 8, etc.) of the envelope of the transmission signal. In one embodiment, wherein the DFE is configured to cancel second order intermodulation noise (IM2) the exponential circuit component 310 may be configured to square the envelope of the transmitted signal (which is linear proportional to the magnitude of the baseband signal STX) to generate an input signal SIN that is provided to the adaptive filters 312a and 312b, since IM2 noise is proportional to the squared envelope of the modulated blocker. In alternative embodiments, the exponential circuit component 310 may be configured to raise the magnitude of the transmitted baseband signal to alternative powers (e.g., n=4, 6, 8, etc.) to generate an input signal SIN that is provided to adaptive filters to cancel intermodulation noise. In one embodiment, wherein the transmission signal comprises an I/Q signal, an envelope generator 309 may be configured to generate an envelope, from the I/Q signal, which is provided to the exponential circuit component 310.
Therefore, the adaptive filtering system of
In particular, an adaptive filter input signal u(n) (e.g., corresponding to SIN of
The simplified adaptive filtering system as provided herein may be implemented into a variety of transceiver systems.
The transmission path of the polar modulation transceiver 500 comprises a processing unit 526 (e.g., a DFE, a baseband processor) configured to beak a transmitted signal unto an amplitude component A(t) and a phase component Φ(t). In one embodiment the processing unit may comprise a DFE 528 and a polar to rectangular conversion circuit 530. The output of the DFE 528 may comprise a transmission signal which can be separated into in-phase I(t) and quadrature phase Q(t) components. The in-phase I(t) and quadrature phase Q(t) components are then provided to the rectangular to polar conversion circuit 530 configured to convert the in-phase I(t) and quadrature phase Q(t) components into an amplitude component A(t) and a phase component Φ(t). A digital to analog converter 532 is configured to provide the phase component Φ(t) to a phase modulator 536 that modulates a radio frequency carrier signal having a constant signal envelope, and the amplitude component A(t) to an amplitude modulator 534 that varies a transmission signal envelope. A power amplifier 538 amplifies the modulated signal prior to transmission by antenna 502.
Since second order intermodulation noise (IM2) is proportional to the squared envelope of the transmission signal A(t), a DFE 506 comprises a squaring block 508 configured to square the amplitude component A(t) of the baseband transmission signal to generate an adaptive filter input signal u(n), which is provided to the first adaptive filter 510a and the second adaptive filter 510b as the input signal u(n).
The first adaptive filter 510a is configured to adapt the real parts of a weight vector wI(n) at each sampling instant while minimizing the real error signals eI(n). For example, the first adaptive filter 510a is configured to operate a real valued adaptive algorithm (e.g., LMS adaptive algorithm) that generates an output signal yI(n) which is an estimate of a first component of intermodulation noise. The output signal is provided to an adder 512a that produces a real valued error signal eI(n). The error signal eI(n) is fed back to the adaptive filter 510a to iteratively update a tap weight vector wI(n). Iterative operation of such an adaptive filtering process estimates the intermodulation noise distorting the in-phase component of the receiver signal yI(n), and by cancelling the estimated noise generates an in-phase output signal eI(n) having substantially no in-phase intermodulation noise.
The second adaptive filter is configured to adapt the its weight vector wQ(n) at each sampling instant while minimizing error signals eQ(n) in the imaginary component of the received signal. For example, the second adaptive filter 510b is configured to operate a real valued adaptive algorithm (e.g., LMS adaptive algorithm) that generates an output signal yQ(n) that is an estimate of a second component of intermodulation noise. The output signal is provided to an adder 512b that produces an imaginary valued error signal eQ(n). The error signal eQ(n) is fed back to the adaptive filter 510b to iteratively update a tap weight vector wQ(n). Iterative operation of such an adaptive filtering process estimates the intermodulation noise distorting the quadrature phase component of the receiver signal yQ(n), and by cancelling the estimated noise generates a quadrature phase output signal eQ(n) having substantially no quadrature phase intermodulation noise.
In particular, the transmission path of the I/Q transceiver 600 comprises a DFE 626 configured to generate an in-phase signal component I(t) and a quadrature phase signal component Q(t). The in-phase and quadrature phase components are provided to an digital to analog converter configured to provide an analog signal to upconverters 630a and 630b that generate an upconverted signal to a power amplifier 632 configured to amplify the upconverted signal prior to transmission by antenna 602. The in-phase and quadrature phase signal components are also provided to a logic circuit 608 that is configured to generate an adaptive filter input signal u(n) therefrom. In one embodiment, the logic circuit 608 may be configured to generate a magnitude from the I/Q signals and then to raise the magnitude to an even power.
As described above, the first adaptive filter 610a is configured to adapt the real parts of a weight vector wI(n) at each sampling instant while minimizing the real error signals eI(n). The second adaptive filter is configured to adapt the imaginary parts of a weight vector wan) at each sampling instant while minimizing the imaginary error signals eQ(n).
Because a different adaptive filter is used to separately filter the I and Q components of the reference signal, there are no cross coupling effects between the filters in the I and Q paths. For example, estimations of the IM2 noise generated in one path (e.g., the in-phase path) have no effect on estimations of IM2 noise generated in the other path (e.g., the quadrature phase path). Accordingly, the apparatus provided herein reduces computational complexity of the adaptive filtering algorithm.
Typically, the simultaneous adaptation of the real and imaginary parts of a weight vector operates with a cross relation between the real and imaginary components of an input reference signal u(n) (e.g., as shown in
In particular, the flow diagram illustrates the simplified adaptive filtering system use of two separate adaptive filters, illustrated by boxes 702 and 704, wherein each respective adaptive filter is configured to operate a real valued adaptive algorithm. The first adaptive filter 702 is configured to compensate for intermodulation noise using a real valued adaptive algorithm that generates an output signal yI(n), while the second separate adaptive filter 704 is configured to compensate for intermodulation noise using a real valued adaptive algorithm that generates an output signal yQ(n). Since the adaptive filters separately compensate for intermodulation noise the separate filters can be used, without interaction, to cancel noise to the output signal eI(n) and the output signal eQ(n) separately. This removes cross coupling between real and imaginary parts of the noise cancellation so that intermodulation noise cancellation can be performed for the in-phase component and the quadrature phase component respectively using a real valued adaptive algorithm for each path.
Operation of the adaptive algorithm as shown in
The adaptive filtering algorithm shown in
Therefore, the signal flow diagram 700 illustrates how the use of two real valued adaptive filters can simplify an adaptive algorithm so that the in-phase and quadrature phase components of an intermodulation noise can be filtered separately and so that updating the filter coefficients wI(n) and wQ(n) can be done without the knowledge of signals in the other branch.
Although the simplified adaptive filters provided herein are illustrated as acting upon an amplitude of a transmission signal raised to a power (e.g., a squared amplitude of a transmission signal, as shown in
Although
In particular, the transceiver 900 comprises an adaptive filtering system 904 comprising a first real valued adaptive filter 906a and a second real valued adaptive filter 906b configured to respectively operate real valued adaptive algorithms on analog signals to iteratively cancel components of an intermodulation noise caused by a transmission signal. The adaptive filtering system 904 is configured between the downconverter 902 and an analog to digital converter 908 configured to convert the analog filtered signal to a digital signal for digital signal processing by DFE 910. A squaring block 912 is configured downstream of a digital to analog converter 914 in the transmission chain so that the implementation of the real valued adaptive filters 906a and 906b is fully analog.
It will be appreciated that in additional embodiments, a transceiver, as provided herein, may be configured cancel transmitter induced even ordered intermodulation noise utilizing a circuitry that combines analog and digital processing. For example, as shown in the transceiver circuit of
While these methods are illustrated and described below as a series of acts or events, the present disclosure is not limited by the illustrated ordering of such acts or events. For example, some acts may occur in different orders and/or concurrently with other acts or events apart from those illustrated and/or described herein. In addition, not all illustrated acts are required and the waveform shapes are merely illustrative and other waveforms may vary significantly from those illustrated. Further, one or more of the acts depicted herein may be carried out in one or more separate acts or phases.
Furthermore, the claimed subject matter may be implemented as a method, apparatus, or article of manufacture using standard programming and/or engineering techniques to produce software, firmware, hardware, or any combination thereof to control a computer to implement the disclosed subject matter (e.g., the circuits shown in
At 1102 an adaptive filter input signal is generated. The adaptive filer inputs signal may be generated from a transmission signal in the baseband (e.g., x(t)=I(t)+jQ(t)) to comprise a first component (e.g., an in phase component) and a second component (e.g., a quadrature phase component). In one embodiment, wherein the method is configured to cancel second order intermodulation noise, the input signal may be set equal to the square of the magnitude of the transmission baseband signal.
At 1104 a first real valued adaptive filtering algorithm is applied to the input signal. The first adaptive filtering algorithm is configured to iteratively determine a first intermodulation noise component caused by a transmission signal.
In one embodiment the adaptive filtering algorithm may comprise an LMS algorithm configured to estimate an in-phase component of intermodulation noise (e.g., a noise component that is distorting the in-phase component of a desired signal). In such an embodiment, the method may comprise iteratively calculating a filter output yI(n) (at 1106), estimating an error signal eI(n) (at 1108), and adjusting tap weights wI(n) (at 1110).
At 1112 intermodulation noise is cancelled from a first (e.g., in-phase) component of the desired (receiver) signal. In one embodiment, cancellation of the first intermodulation noise generates a first output signal comprising a signal that is corrected to remove a first intermodulation noise component generated by the transmission signal.
At 1114 a second real valued adaptive filtering algorithm is applied to the input signal. The second adaptive filtering algorithm is configured to iteratively determine a second intermodulation noise component caused by a transmission signal.
In one embodiment the adaptive filtering algorithm may comprise an LMS algorithm configured to estimate a quadrature phase component of intermodulation noise (e.g., a noise component that is distorting the quadrature phase component of a desired signal). In such an embodiment, the method may comprise iteratively calculating a filter output yQ(n) (at 1116), estimating an error signal eQ(n) (at 1118), and adjusting tap weights wQ(n) (at 1120).
At 1122 intermodulation noise is cancelled from a second (e.g., quadrature phase) component of the desired (receiver) signal. In one embodiment, cancellation of the second intermodulation noise generates a second output signal comprising a signal that is corrected to remove the second intermodulation noise component generated by the transmission signal.
Although the invention has been illustrated and described with respect to one or more implementations, alterations and/or modifications may be made to the illustrated examples without departing from the spirit and scope of the appended claims. In particular regard to the various functions performed by the above described components or structures (assemblies, devices, circuits, systems, etc.), the terms (including a reference to a “means”) used to describe such components are intended to correspond, unless otherwise indicated, to any component or structure which performs the specified function of the described component (e.g., that is functionally equivalent), even though not structurally equivalent to the disclosed structure which performs the function in the herein illustrated exemplary implementations of the invention. In addition, while a particular feature of the invention may have been disclosed with respect to only one of several implementations, such feature may be combined with one or more other features of the other implementations as may be desired and advantageous for any given or particular application. Furthermore, to the extent that the terms “including”, “includes”, “having”, “has”, “with”, or variants thereof are used in either the detailed description and the claims, such terms are intended to be inclusive in a manner similar to the term “comprising”.
