WO2010031433A1

WO2010031433A1 - Error compensation device

Info

Publication number: WO2010031433A1
Application number: PCT/EP2008/062472
Authority: WO
Inventors: Mats HÖGBERG; Björn Ulf Anders SIHLBOM
Original assignee: Telefonaktiebolaget L M Ericsson (Publ)
Priority date: 2008-09-18
Filing date: 2008-09-18
Publication date: 2010-03-25

Abstract

An error compensation device for a radio transmitter adapted to receive a digital input signal (13), modulate the input signal with a reference frequency signal from an oscillator (20) and generate an output signal (5). The error compensation device (2) comprising a feedback element (7) arranged to receive part of the output signal (5) and generate a feedback signal (11) by demodulation, an error calculation element (8) arranged to receive part of the input signal and the feedback signal from the feedback element (7) and generate a digital error signal (12) representative of the phase error introduced to the input signal (13) by the radio transmitter (1). An associated method is also disclosed.

Description

ERROR COMPENSATION DEVICE

TECHNICAL FIELD

This invention relates to an error compensation device for a radio transmitter. In particular, it relates to a phase error compensation device for compensating for phase error of an oscillator in a radio transmitter. It also relates to a radio transmitter incorporating such a device and to its method of operation.

BACKGROUND

Radio transmitters receive a baseband input signal, convert it to a radio frequency signal, amplify it and transmit the resulting signal. It is important that the amplified transmitted signal is an accurate reproduction of the input signal to minimise interference on other channels, for example. When an input signal is converted to a radio frequency signal, frequency errors are introduced by an oscillator of the radio transmitter. In particular, the frequency reference oscillator in the transmitter is not perfect, which causes band-limited phase error (correlated phase errors) on the generated signal.

To compensate for such errors it is known to improve the accuracy of the frequency reference oscillator. However, using a sufficiently accurate oscillator for a transmitter that transmits signals in accordance with Wideband Code Division Multiple Access (WCDMA) and High Speed Packet Access (HSPA) is expensive. Such an oscillator is required to operate accurately at frequencies of 15.36 MHz and 30,72MHz.

SUMMARY According to a first aspect of the invention we provide an error compensation device for a radio transmitter adapted to receive a digital input signal, modulate the input signal with a reference frequency signal from an oscillator and generate an output signal, the error compensation device comprising a feedback element arranged to receive part of the output signal and generate a feedback signal by demodulation, an error calculation element arranged to receive part of the input signal and the feedback signal from the feedback element and generate a digital error signal representative of the phase error introduced to the input signal by the radio transmitter.

This is advantageous as the phase error imparted by the radio transmitter on the input signal can be determined digitally by the comparison of the input signal to the feedback signal. This approach is particularly accurate as the feedback signal is directly compared to the input signal. Further, the calculation of an error signal in the digital domain is cost-effective in terms of hardware requirements.

The feedback element may include a demodulator adapted to demodulate the output signal using the reference frequency signal from the oscillator received via a delay element. The delay element may be tuneable such that the error calculation element can more accurately detect rapid changes in the phase error or more accurately detect the phase error. This is advantageous as the error calculation element can be set to obtain the most advantageous results. If an accurate measurement of the phase error is required, then the delay element can be tuned for accuracy. Alternatively, if the phase error frequently changes, then the delay element can be tuned so that the error calculation element responds quickly to those changes.

The error calculation element may include an analogue-to-digital convertor arranged to receive the feedback signal and digitize it for processing by the error calculation element. The error calculation element may receive the part of the input signal via an input signal delay element and the feedback element receives the part of the output signal delayed by the radio transmitter, the input signal delay element being arranged to operate in the digital domain, wherein the input signal delay element is tuned to match the delay experienced by the output signal through the radio transmitter. This is advantageous as the delays experienced by the signal as it travels through the radio transmitter in the analogue domain can be matched by the input signal delay element in the digital domain. Thus, the error calculation element can accurately extract the phase error caused by the transmitter digitally.

The error compensation device may include a correction element arranged to receive the digital error signal and calculate an error correction function that is applied to the input signal to compensate for the phase error of the oscillator. This is advantageous as the input signal, which is a digital signal, can be modified by application of a digital error correction function such that the phase error of the oscillator can be compensated for in the digital domain. Further, as the error compensation device is able to effectively compensate for frequency errors caused by an imperfect oscillator of a radio transmitter. Thus, the error correction element ensures that the transmitted signal is an accurate representation of the input signal. Accordingly, the bit error rate and packet error rate of the transmitted output signal are kept to a minimum. The error compensation device of the invention is particularly effective at compensating for frequency errors that introduce a vector error into the output signal.

The correction element may include a regulator adapted to receive the error signal and control the error correction function in response to the error signal such that the error signal is minimized. Thus, the regulator can modify the error correction function to compensate for changes in the phase error imparted by the radio transmitter.

According to a second aspect of the invention we provide a method of compensating for phase error caused by an oscillator of a radio transmitter, the transmitter adapted to receive a digital input signal, modulate the input signal with a reference frequency signal from an oscillator and generate an output signal, the method comprising the steps of; generating a feedback signal by demodulation of the output signal; and calculating a digital error signal using the feedback signal and the input signal.

This is advantageous as the digital error signal is obtained by a comparison of the input signal with the feedback signal .

The method may include the additional steps of; calculating an error correction function from the error signal; applying the error correction function to the input signal.

This is advantageous as the digital error signal can be used to calculate an error correction function to compensate for the phase error introduced to the input signal by the oscillator. Thus, the input signal can be modified to minimize the effect of the phase error.

BRIEF DESCRIPTION OF THE DRAWINGS

There now follows by way of example only a detailed description of the present invention with reference to the accompanying drawings in which: Figure 1 shows a radio transmitter incorporating an embodiment of the error compensation device of the invention; and

Figure 2 shows a flow chart illustrating an example of the operation of the error compensation device.

DETAILED DESCRIPTION

A radio transmitter assembly 1 is shown in Figure 1 that includes an error compensation device 2. The assembly receives an input signal shown generally at 6 and passes an output signal 5 to an antenna 4.

The assembly 1 comprises a modulator 17 that modulates the input signal 6 with a reference frequency signal from an oscillator 20. The modulator is connected to a power amplifier 3 that amplifies the modulated signal and supplies the antenna 4 with the amplified output signal 5.

The error compensation device 2 comprises a feedback element 7, an error calculation element 8 and a correction element 10. The feedback element 7 receives a part of the amplified output signal 5 and passes a baseband feedback signal at 1 1 to the error calculation element 8. The error calculation element 8 also receives the input signal 6 and outputs an error signal at 12 that represents the phase error imparted by the oscillator 20 to the input signal 6. The error signal is received by the correction element 10, which applies an error correction function to the input signal 6. As represented by the dashed line through Figure 1 , parts of the assembly operate in the digital domain and other parts in the analogue domain. The error calculation element 8 and correction element 10 perform operations on digital signals and therefore operate in the digital domain. The feedback element 7 and the remainder of the radio transmitter assembly 1 perform operations on analogue signals and therefore operate in the analogue domain.

The input signal 6 undergoes several modifications before it is received by the power amplifier 3. In particular, the radio transmitter assembly 1 first receives a digital data signal represented by the function x(t) and shown as 13 in Figure 1 . The digital data signal 13 is a baseband signal. The digital data signal 13 is modified by the correction element 10 to form a compensated digital signal shown at 14. This signal is then converted to an analogue signal shown at 15 by digital-to-analogue convertor 16. The analogue signal 15 is modulated with a carrier wave by a modulator 17 comprising an analogue mixer to form a modulated signal shown at 18 for receipt by the power amplifier 3. The reference frequency signal, represented by function f(t), from oscillator 20 is input to the mixer 17. As discussed above, the power amplifier 3 outputs the amplified signal 5, which is passed to the transmitter 4. As it will be appreciated by those skilled in the art, each of the components 16, 17, 3 introduce a time delay. This output signal time delay is represented in Figure 1 by a time delay element 21 .

The feedback element 7 comprises a demodulator 22 comprising an analogue mixer which receives a part of the amplified output signal 5. The mixer 22 also receives the reference frequency signal from the oscillator 20 via a second delay element 23. The mixer 22 is adapted to demodulate the amplified signal 5 back to baseband using the reference frequency signal 20. The delay element 23 is chosen so as to preserve the phase error introduced by the oscillator 20 to the output signal 5. The mixer 22 outputs the feedback signal 1 1 , which is received by the error calculation element 8. The error calculation element 8 includes an analogue-to-digital convertor 24, which digitises the baseband feedback signal 1 1 and directs it to a phase error extractor 25 comprising a digital mixer. The mixer 25 also receives part of the digital data signal 13 via a third delay element 26. The third delay element 28 thus forms an input signal delay element. The digital mixer 25 demodulates the digital feedback signal 1 1 with the original digital data signal 13, which leaves the phase error z(t), which is output at 27. The phase error z(t) is received by an error calculator 28 which generates an error signal e(t).

The error signal is received by a regulator 30 of the correction element 10. The regulator 30 is arranged to output a regulation signal represented by function α(t) at 31 . The function α(t) is chosen to minimize the error signal e(t) calculated by the error calculator 8, as described in more detail below.

The signal α(t) is received by complex number generator 32. The complex number generator 32 converts an input angle "x" to a complex number e^JX=cos(x)+j.sin(x). An error correction function represented by g(t) is output from the complex number generator 32 at 33 and forms the input to an error correction function applicator 34 as well as the data signal 13. The signal g(t) is such that when combined with the data signal x(t), the output signal 5 is an accurate representation of the data signal 13. Thus, the error compensation device 2 minimises the phase error in the transmitted signal due to the non-perfect oscillator 20.

The output signal y(t) is a function of the input signal x(t), the error correction function g(t) and the reference frequency signal f(t), as shown in equation 1 . y(t) = x(t-t_l)g(t-t_l)f(t-t_l) (1)

The reference frequency signal f(t) is ideally a constant frequency F, but in practice the reference frequency signal includes a frequency error component, e^φ(t). The function φ(t) represents the phase slope error. Thus, f(t) also includes the phase error added by the oscillator 20 to the reference frequency signal. Accordingly;

Further, g(t) has the following equivalence with α(t) once processed by complex number generator 32. Thus, equation 3 shows that g(t) can be written as;

g(t) = e^ (3)

Thus, using equations 2 and 3, equation 4 can be derived;

Therefore, y(t) of Equation 1 can be written as shown in equation 5.

Further, the phase error z(t) at 27 is a function of the input signal, the error correction function and the frequency reference signal, taking account of the delay elements. Accordingly;

_Z(ή = x(t-t_ι)_g(t-t_ι)f(t-t_ι)f*(t-t_o)x*(t-t_ι) (6) f* and x* represent the complex conjugate of functions f and x respectively, as applied by demodulator 22 and phase error extractor 25.

Further, a function angle(x), which calculates the argument of a complex number, is applied to Equation 6;

angle(z(t)) = angle(x(t -t_γ)) + angle(g(t - t_γ)) + angle(f(t - t_γ))- angle(f(t - t₀)) - angle(x(t - t_γ )) = a(t - t_ι) + (2πF(t - t_ι) +ψ(t - t_ι))- (ψ(t - t₀) + 2πF(t - t₀)) = a(t - t_ι) + 2πF(t₀ - t_ι) + [ψ(t - ^) -Vf(I - I₀)]

(7)

Using the approximation f(x+Ax) ~f(x)+Ax.f(x) and equation 7, angle(z(t)) can be written as follows;

angle{z{t)) » α(f - J₁) + (t₀ - t_γ){2πF + φ'(f - I₁)) (8)

Thus, the error signal e(t) can be calculated as;

Therefore, the regulator 30 generates the regulation signal α(t) such that the error signal e(t) is minimized. The regulator shall increase α(t) in a controlled way, if e(t) is larger than zero, and decrease α(t) if e(t) is less than zero. Thus, e(t) is naturally dependent on the regulation signal α(t). If the error signal e(t) received by the regulator 30 is zero, then the frequency f(t) output by the oscillator 20 is constant within a bandwidth limit.

Thus, in summary, the output signal y(t) is the input signal x(t) converted to a radio frequency for transmission. A frequency error is added to the output signal by the oscillator 20. The frequency error is called e^φ(t). With an appropriately chosen delay to and filtering in the baseband processing, the required frequency error can be detected by error calculator 28 and an error signal e(t) can be determined. Accordingly, the regulator 30 can generate a regulation signal that can be used to minimize the frequency errors introduced by the oscillator 20.

Figure 2 shows a flow chart illustrating an embodiment of the method of the invention. Step 60 shows generating a feedback signal by demodulating the output signal. Step 61 shows calculating a digital error signal from the feedback signal and the input signal. Step 62 shows calculating an error correction function. Step 63 shows applying the error correction function to the input signal to form a compensated input signal to correct for the calculated phase error introduced by the oscillator 20.

Claims

1 . An error compensation device for a radio transmitter adapted to receive a digital input signal (13), modulate the input signal with a reference frequency signal from an oscillator (20) and generate an output signal (5), the error compensation device (2) comprising a feedback element (7) arranged to receive part of the output signal (5) and generate a feedback signal (1 1 ) by demodulation, an error calculation element (8) arranged to receive part of the input signal and the feedback signal from the feedback element (7) and generate a digital error signal (12) representative of the phase error introduced to the input signal (13) by the radio transmitter (1 ).

2. An error compensation device according to claim 1 , in which the feedback element (7) includes a demodulator (22) adapted to demodulate the output signal (5) using the reference frequency signal received from the oscillator (20) via a delay element (23).

3. An error compensation device according to claim 2, in which the delay element (23) is tuneable such that the error calculation element can more accurately detect rapid changes in the phase error or more accurately detect the phase error.

4. An error compensation device according to any preceding claim, in which the error calculation element (8) includes an analogue-to-digital convertor (24) arranged to receive the feedback signal (11 ) and digitize it for processing by the error calculation element (8).

5. An error compensation device according to any preceding claim, in which the error calculation element (8) receives the part of the input signal (13) via an input signal delay element (26) and the feedback element (7) receives the part of the output signal (5) delayed by the radio transmitter, the input signal delay element (26) being arranged to operate in the digital domain, wherein the input signal delay element (26) is tuned to match the delay (21 ) experienced by the output signal (5) through the radio transmitter.

6. An error compensation device according to any preceding claim, in which the error compensation device (2) includes a correction element (10) arranged to receive the digital error signal (12) and calculate an error correction function g(t) that is applied to the input signal (6) to compensate for the phase error of the oscillator (20).

7. An error compensation device according to claim 6, in which the correction element (10) includes a regulator (30) adapted to receive the error signal (12) and control the error correction function in response to the error signal such that the error signal is minimized.

8. A method of compensating for phase error caused by an oscillator (20) of a radio transmitter (1 ), the transmitter adapted to receive a digital input signal (13), modulate the input signal with a reference frequency signal from the oscillator (20) and generate an output signal (5), the method comprising the steps of; generating a feedback signal by demodulation of the output signal; and calculating a digital error signal using the feedback signal and the input signal.

9. A method according to claim 8, including the additional steps of; calculating an error correction function from the error signal; applying the error correction function to the input signal.