EP1989851A2

EP1989851A2 - In phase and quadrature path imbalance compensation

Info

Publication number: EP1989851A2
Application number: EP07705818A
Authority: EP
Inventors: Xuecheng Qian
Original assignee: NXP BV
Current assignee: NXP BV
Priority date: 2006-02-22
Filing date: 2007-02-08
Publication date: 2008-11-12
Also published as: CN101390360A; JP2009527968A; WO2007096800A2; WO2007096800A3; KR20080098535A; TW200836531A; KR100977938B1

Abstract

A direct down conversion receiver (19) has an imbalance compensation parameter determining stage (20) and a phase and amplitude imbalance compensation stage (21). The determining stage (20) is connected to receive I and Q signals I'(t), Q'(t) output from I and Q paths (5, 6) of the receiver (19). The determining stage (20) determines imbalance compensation parameters g1 cos(φ1), g1 sin(φ1), gQ cos(φ3), and gQ sin(φ3) from the training sequence of a received signal. It then calculates mean values for each of the determined imbalance compensation parameters g1 cos(φ1), g1 sin(φ1), gQ cos(φ3), and gQ sin(φ3) and calculates a further imbalance compensation parameter g1 gQ sin(φ3-φ1), from these mean values. The phase and amplitude imbalance compensation stage (21) uses the determined imbalance compensation parameters g1 cos(φ1), g1 sin(φ1), gQ cos(φ3), and gQ sin(φ3), and the calculated further imbalance compensation parameter g1gQ sin(φ3-φ1) to compensate for phase and amplitude imbalances between the I and Q paths (5,6) of the receiver during receipt of a payload of the received signal.

Description

DESCRIPTION

IN PHASE AND QUADRATURE PATH IMBALANCE COMPENSATION

FIELD OF THE INVENTION

This invention relates to In phase (I) and Quadrature (Q) path imbalance compensation. More particularly, but not exclusively, the invention relates to a receiver, such as a direct down conversion radio receiver, incorporating I/Q path imbalance compensation. BACKGROUND ART

Many receivers use Quadrature demodulation, which requires a received signal to be processed over both an In-phase (I) path and a Quadrature (Q) path. For example, referring to Figure 1 , a direct down conversion receiver 1 according to the prior art has an antenna 2 for receiving a signal. The antenna 2 is connected to output the received signal to a Radio Frequency (RF) filter 3, which is a band-pass filter for filtering the received signal. The RF filter 3 is connected to output the filtered signal to a Low Noise Amplifier (LNA) 4 for amplifying the filtered signal. This amplification helps to reduce noise in subsequent processing of the received signal. The LNA 4 is connected to output the amplified signal to both an In-phase (I) path 5 and a Quadrature (Q) path 6 of the receiver 1. The I path 5 comprises an I mixer 7, an analogue low-pass filter 8, and Automatic Gain Controller (AGC) 9, an Analogue to Digital Converter (ADC) 10 and a digital low pass filter 11 connected in series in that order. The Q path 6 is identical to the I path, except that it has a Q mixer 12 in place of the I mixer 7. So, the Q path 6 comprises a Q mixer 12, an analogue low-pass filter 13, an AGC 14, an ADC 15 and a digital low pass filter 16. Signals output by the I and Q paths 5, 6 pass to a demodulation and decoding stage 17 for demodulating and decoding the signals to produce a data output 18. The I mixer 7 and Q mixer 12 each mix the amplified signal output by the LNA 4 with a local oscillator signal. As the receiver 1 is a direct down conversion receiver, the local oscillator signals should have substantially the same frequency as the carrier frequency of the wanted signal modulated in the signal received at the antenna 2. This allows the mixers 7, 12 to down convert the wanted signal to baseband. Ideally, the local oscillator signals used by the I and Q mixers are shifted from one another by 90° in phase so that the wanted signal can be properly demodulated from the signals on the I path 5 and Q path 6. However, this is difficult to implement in practice and often there are variations in the phase difference between the local oscillator signals used by the I and Q mixers 7, 12. This variation is known as phase imbalance. Similarly, the gains over the I path 5 and Q path 6 may not be the same. The signals output by the I and Q paths 5, 6 may therefore have unintended differences in amplitude. This is known as amplitude imbalance. Phase and amplitude imbalance can cause errors in demodulation. It can also cause the aliasing of image interference into the band of the wanted signal.

Many methods have been suggested for correcting phase and amplitude imbalance, most of which involve calibrating a receiver using a calibration signal. Typically, the calibration signal has a known frequency, phase and amplitude. When it is passed through the receiver, the signals output by the I and Q paths can be analysed to determine the phase and amplitude imbalances between them. There are a number of problems with such methods. For example, the calibration signal is most often generated locally, by the receiver to be calibrated, so that its frequency, phase and amplitude can be carefully controlled. However, this tends to require the provision of a dedicated circuit for generating the calibration signal, which can by undesirably expensive and complex. Some methods avoid this by receiving an external calibration signal via the antenna via which the receiver usually receives signals. Whilst this can avoid the added expense and complexity of generating a calibration signal locally at the receiver, it is impractical as occupies space within the frequency spectrum available for communication or such like. All these methods also prevent the receiver from receiving other signals during calibration. So, during calibration, communication via the receiver or reception of a broadcast signal at the receiver may be interrupted. This is clearly undesirable. Even receivers operating in Time Division Multiple Access (TDMA) mode, Time Division Duplex (TDD) mode or such like that might sometimes be able to accommodate short interruptions in signal reception suffer from this problem, as it can be hard to predict when signals might be received and thus difficult to allocate appropriate times for calibration. So, conventional methods of correcting phase and amplitude imbalance are far from ideal.

DISCLOSURE OF INVENTION

The present invention seeks to overcome the problems described above. According to a first aspect of the present invention, there is provided a receiver for receiving a signal including a training sequence and a payload, the receiver comprising an In-phase path and a Quadrature path for converting the received signal to In-phase and Quadrature signals for output to a demodulator; processing means for determining an imbalance compensation parameter from the In-phase and Quadrature signals representing the training sequence; and an imbalance compensation circuit for compensating for phase and amplitude imbalance between the In-phase and Quadrature signals representing the payload using the determined imbalance compensation parameter.

According to a second aspect of the present invention, there is provided a method of receiving a signal including a training sequence and a payload, the method comprising converting the received signal to In-phase and Quadrature signals for output to a demodulator; determining an imbalance compensation parameter from the In-phase and Quadrature signals representing the training sequence; and compensating for phase and amplitude imbalances between the In- phase and Quadrature signals representing the payload using the determined imbalance compensation parameter. In other words, phase and amplitude imbalance compensation during reception of a signal's payload can be based on the signal's training sequence. Training sequences are an inherent part of the signals of many current communication systems. So, the invention has the significant advantage of allowing compensation for phase and amplitude imbalance without the use of a dedicated calibration signal. No calibration signal needs to be generated locally at the receiver. Similarly, communication capacity does not need to be set aside for the reception of a dedicated calibration signal.

So, the training sequence may be also used for a purpose in addition to determining the imbalance compensation parameter. This might include additional signal processing. In other words, the processing means may use the training sequence for additional signal processing. Likewise, the method may include using the training sequence for additional signal processing. In one example, the additional signal processing may comprise training an equaliser. In other examples, the additional signal processing may comprise any one of synchronisation, channel quality estimation, signal level adjustment and so on.

It can be appreciated that the training sequence may take a variety of forms. Fundamentally, the training sequence is usually a portion of a received signal that is known, e.g. the signal expected to be received during the training sequence may be known. The training sequence might comprise a preamble, pilot sequence, access code, synchronisation word, synchronisation sequence or such like. In any event, the training sequence is usually adapted for a particular purpose in addition to amplitude and phase imbalance compensation. For example, it may meet the requirements of a particular conventional communication standard. In particular, the training sequence may be suitable for training an equaliser or for enabling synchronisation. Typically, this means that the training sequence may comprise a sequence having a prominent peak in its autocorrelation function.. The training sequence and the payload may be contained in a data packet. The compensation of the signals representing the payload of the data packet may therefore be performed using the imbalance compensation parameter determined from the signals representing the training sequence of that data packet. In other words, imbalance compensation can be achieved from the signals representing a single data packet. Indeed, imbalance compensation could be carried out on a packet by packet basis, if desired. It is preferred that the compensation parameters are determined in the time domain rather than the frequency domain, as calculations in the time domain are quicker and easier to implement, avoiding the use of Fourier transforms. For example, it is preferred that receiver comprises a selector for selecting portions of the In-phase and Quadrature signals of the training sequence at two different times. Accordingly, the imbalance compensation parameter can be determined from the selected portions of the In-phase and Quadrature signals at the two different times. These times are illustrated as times t_λ and t₂ in the description of the preferred embodiments of the invention below. Typically, the portions of the In-phase and Quadrature signals are selected from a part of the training sequence that is also used for the further processing described above. In other words, there need not be a specific part of the training sequence used only for amplitude and phase imbalance compensation.

It is also useful for the determination of the imbalance compensation parameter to be based on signals representing the training sequence as expected to be received. These are illustrated by the baseband signals baseband signals A(t_λ) , A(t₂) , BQ₁) , B(t₂) in the description of the preferred embodiments of the invention below.

An imbalance compensation parameter may be determined based on the In-phase signal and another imbalance compensation parameter is determined based on the Quadrature signal. Indeed, it is preferred that a pair of imbalance compensation parameters are determined from the In-phase signal and another pair of imbalance compensation parameters are determined from the Quadrature signal. These are illustrated by imbalance compensation parameters g₇ COs(Cp₁) , g₇ Sm(Cp₁) and imbalance compensation parameters g_e cos(φ₃) , g_e sin(φ₃) respectively in the description of the preferred embodiments of the invention below. It is also preferred that a further imbalance compensation parameter is determined from mean values of these imbalance compensation parameters. This is illustrated by further imbalance compensation parameter g_jg_Q SUi(Cp₃ -(P₁) in the description of the preferred embodiments of the invention below.

The phase and amplitude imbalance compensation can use these parameters. More specifically, the compensation can involve multiplying the In-phase and Quadrature signals by the determined imbalance compensation parameter(s). It is preferred that the compensation involves multiplying the In- phase signal by the imbalance compensation parameter(s) determined from the Quadrature signal and Quadrature signals by imbalance compensation parameter(s) determined from the In-phase signal. It is also preferred that the compensation involves multiplying the In-phase and Quadrature signals by the inverse of the determined further imbalance compensation parameter. This is illustrated by equations (6) and (7) in the description of the preferred embodiments of the invention below.

Use of the terms "processing means", "circuit", "selector" and so on is intended to be general rather than specific. The invention may be implemented using such separate components. However, it may equally be implemented using individual processor, such as a digital signal processor (DSP) or central processing unit (CPU). Similarly, the invention could be implemented using a hard-wired circuit or circuits, such as an application- specific integrated circuit (ASIC), or by embedded software. Indeed, it can also be appreciated that the invention can be implemented using computer program code. According to a further aspect of the present invention, there is therefore provided computer software or computer program code adapted to carry out the method described above when processed by a processing means. The computer software or computer program code can be carried by a computer readable medium. The medium may be a physical storage medium such as a Read Only Memory (ROM) chip. Alternatively, it may be a disk such as a Digital Versatile Disk (DVD-ROM) or Compact Disk (CD-ROM). It could also be a signal such as an electronic signal over wires, an optical signal or a radio signal such as to a satellite or the like. The invention also extends to a processor running the software or code, e.g. a computer configured to carry out the method described above. BRIEF DESCRIPTION OF DRAWINGS Preferred embodiments of the invention will now be described with reference to the accompanying drawings, in which:

Figure 1 is a schematic illustration of a direct down conversion receiver according to the prior art;

Figure 2 is a schematic illustration of a direct down conversion receiver according to a first preferred embodiment of the present invention;

Figure 3 is a schematic illustration of an imbalance compensation parameter determining stage of the direct down conversion receiver illustrated in Figure 2;

Figure 4 is a schematic illustration of a phase and amplitude compensation stage of the direct down conversion receiver illustrated in Figure 2; and

Figure 5 is a schematic illustration of a heterodyne receiver according to a second preferred embodiment of the present invention. MODES FOR CARRYING OUT THE INVENTION Referring to Figure 2, a direct down conversion receiver 19 according to a first preferred embodiment of the invention has many components similar to those of the receiver 1 of the prior art described above with reference to Figure 1 and the same reference numerals are used for the similar components. However, between the I and Q paths 5, 6 and the demodulation and decoding stage 17, the receiver 19 according to the first preferred embodiment of the invention has an imbalance compensation parameter determining stage 20 and a phase and amplitude imbalance compensation stage 21.

The imbalance compensation parameter determining stage 20 is connected to receive I and Q signals output from the I and Q paths 5, 6 of the receiver 19 after filtering by the analogue low pass filters 8, 13, conversion from analogue to digital by the ADCs 10, 15 and filtering by the digital low pass filters 11 , 16. The determining stage 20 is shown in more detail in Figure 3, from which it can be seen that it includes a selector 22 for selecting portions of the received I and Q signals I'(t) and Q\t) output by the

I and Q paths 5, 6 and a processor 23 for processing the selected portions of the signals to determine and calculate imbalance compensation parameters g_/ cos(φ₁) , g₇ sin(φ_j) , g_e cos(φ₃) , g_e sin(φ₃) and

The signal received at the antenna 2 can be expressed as S(t) = ^(t)cos(w_ct + φ₁) -5(t)sin(w_ct+φ₁) (1 ) where A(t) and B(t) are baseband signals modulated onto a carrier signal cos(w_ct+φ₁) and sϊn(w J₊^₁) , (P₁ is the phase of the carrier signal and w the carrier frequency f_c =—^£- . The I and Q mixers 7, 12 of the receiver 19 mix

2π the received signal S(t) with two local oscillating signals, shifted in phase with respect to one another by 90°, to convert the signal S(t) to baseband signals on the I and Q paths 5, 6. The signals on I path 5 and Q path 6 after mixing can be expressed as

I(t) = S(t) * {_gl ∞s(w_ct)}

= \ A(t)g_j [COS(Cp₁ ) + cos(2w_ct + Cp₁ )] -i B(t)_gl [sintø_j ) + sin(2 w_ct + % )]

and

Q(t) = S(t) * {g_β sin(w_ct +θ )} = S(t) * {g_β cos(w_ct +θ - 1)}

))] (3)

-i£(0g_β[sin(_φi - <β -^)) + sm(2w_ct +φ_{ι +} φ -^))]

2 2 respectively, where θ represents any variation from the desired 90° phase difference and g₇ and g_e are the overall gain factors for the I path 5 and Q path 6. Ideally, the phase variation θ is zero, i.e. θ = 0 , and the gain factors g_j and g_Q are equal, i.e. g_I = g_ςr However, due to circuit imperfection, temperature variation and such like, the phase variation θ is not zero, i.e. θ ≠ 0 , and the gain factors g₇ and g_Q are not equal, i.e. g₇ ≠ g_Q .

The components of the signals I(t) , Q(t) on I path 5 and Q path 6 after mixing that have a frequency around 2w_c are relatively easily removed by the low pass filters 8, 11 , 13, 16. The signals , QXt) output from the I and Q paths 5, 6 can therefore be expressed more simply as

and

Q'(t) = ±g_βA(t)cos(<p_ι -(θ - ^)) -ⁱg_βi?(0 SIn(Cp₁ -(Θ —))

2 2 (5)

= \ g_QA(t) cos(φ₃ ) - \ g_QB(t) sin(φ₃ )

respectively, where φ₃ = φ_ι - (θ — ) = φ_t -θ + — . So, these equations

(4) and (5) can be considered to represent the portions of the I and Q signals I'(t) and Q\t) selected by the selector 22.

Equations (4) and (5) can be rearranged to give

and

when φ₃ ≠ φ_ι ± nπ , n=0,1 ,2, ....

Whilst φi, φ₃, g₇ and g_β are not known, during the transmission of a training sequence the baseband signals A(t) , B(t) used to modulate the carrier signal are known. In other words, the signals expected to be received during the training sequence are known. So, the selector 22 selects portions

IXt₁) , IXt₂) , QXt₁) , QXt₂) of the signals r(t) , QXt) output from the I and Q paths 5, 6 at two different times t_γ and t₂ during reception on a training sequence. In this embodiment, the training sequence is received at the beginning of a data packet that also contains a payload and is used to train an equalizer (not shown) of the receiver 19. In more detail, the imbalance compensation parameters g_/cos(φ₁), g_j sin((p_j) are determined from the signal portions /'(V₁), ¹Xh) selected by the selector 22 from the signal I'Q) output from the I path 5 and the known baseband signals AQ₁), AQ₂), BQ₁), B(t₂) of the training sequence at the two different times t_λ and t₂ using equations (8) and (9) below (on condition that A(I₁)BQ₂) ≠ A(I₂)B(I₁)).

_g/cos_(φi) = ffl^^^M (8)

AQ₁)BQ₂)- AQ₂)BQ₁)

_glSK9ι)=^^⁽fύzI^^M ₍9)

AQ₁)BQJ-AQ₂)BQ₁)

Similarly, the compensation parameters g_gcos(φ₃) g_esin(φ₃) are determined from signal portions Q¹Q₁) , Q¹Q₂) selected by the selector 22 from the signal Q'Q) output from the Q path 6 and the known baseband signals AQ₁), AQ₂), BQ₁), BQ₂) of the training sequence at the two different times t_λ and t₂ using equations (10) and (11) below (on the same condition that AQ₁)BQ₂) ≠ AQ₂)BQ₁)).

^{Sβ VΨ3}' AQ₁)BQJ-AQ₂)BQ₁)

Furthermore, the imbalance compensation parameters g_/cos(φ₁), g_j sintø_j) g_Q cos(φ₃) and g_Q sin(φ₃) determined using equations (8), (9), (10) and (11) can be averaged over a given time, in this embodiment during the whole training sequence, to give mean values of each parameter g₇ COs(Cp₁) , g_j Sm(Cp₁) g_β cos(φ₃) and g_Q sin(φ₃) . These can then be used to calculate a further imbalance compensation parameter

gig a sin(φ ₃-φ i) = g_β sin(φ ₃)g_/ COs(Cp ₁) - g_e cos(φ ₃)g₇ Sm(Cp ₁) (12)

So, it can be appreciated that substituting the determined imbalance compensation parameters g₇ COs(Cp₁) , g_/ sin(φ₁) , g_e cos(φ₃) , g_g sin(φ₃) and the calculated further imbalance compensation parameter g₇g_g sin(φ₃ -Cp₁) into equations (6) and (7) can yield the baseband signals A(t) , B(t) used to modulate the carrier signal cos(w_ct+φ₁) and . In practice, this allows the imbalance compensation parameters g_/ cos(φ₁) , g₇ Sm(Cp₁) , g_e cos(cp₃) , g_β sin(φ₃) and g₇g_e sin(cp₃ -Cp₁) determined and calculated from the signals I\i) , Q\t) output by the I and Q paths 5, 6 at times when the baseband signals A(t) , B(t) used to modulate the carrier signal cos(w_ct+φ₁) and are known (e.g. during the training sequence) to be used to compensate for phase and amplitude imbalances in the signals I\i) , Q\t) output by the I and Q paths 5, 6 at other times, e.g. when the baseband signals A(t) , B(t) used to modulate the carrier signal cos(w_ct+φ₁) and are not known (e.g. during a payload portion of the received signal). So, the phase and amplitude imbalance compensation stage 21 , which is shown in more detail in Figure 4, uses the imbalance compensation parameters g₇ cos(φ₁) , g_/ sin(φ₁) , g_e cos(φ₃) , g_β sin(φ₃) and g₇g_e SUi(Cp₃ -Cp₁) to compensate for phase and amplitude imbalance in the signals I\i) , Q\t) using equations (6) and (7). In more detail, the phase and amplitude imbalance compensation stage 21 comprises a pair of I signal multipliers 24, 25, to each of which the I signal I'(t) is input, and a pair of Q signal multipliers 26, 27, to each of which the Q signal Q\t) is input. One multiplier 24 of the pair of I signal multipliers 24, 25 multiplies the I signal I'(t) with a first of the imbalance compensation parameters, g_Q cos(φ₃) , determined from the Q signal Q\t) . The other multiplier 25 of the pair of I signal multipliers 24, 25 multiplies the I signal with a second of the imbalance compensation parameters, g_β sin(φ₃) , determined from the Q signal Q\t) .

Similarly, one multiplier 26 of the pair of Q signal multipliers 26, 27 multiplies the Q signal Q\t) with a first of the imbalance compensation parameters, g₇ COs(Cp₁) , determined from the I signal . Again similarly, the other multiplier 27 of the pair of Q signal multipliers 26, 27 multiplies the Q signal Q\t) with a second of the imbalance compensation parameters, g₇ Sm(Cp₁) , determined from the I signal I'(t) .

The phase and amplitude imbalance compensation stage 21 also has a pair of adders 28, 29 and a further pair of multipliers 30, 31. A first adder 28 of the pair of adders 28, 29 receives the signal output by the first I signal multiplier 24 and the first Q signal multiplier 26. A second of the pair of adders 28, 29 receives the signal output by the second I signal multiplier 25 and the second Q signal multiplier 27. The adders 28, 29 add the signals they each receive together and output the resulting added signals to respective multipliers 30, 31 of the pair of multipliers 30, 31. The multipliers 30, 31 each multiply the signals they receive by twice the inverse of the calculated further imbalance compensation parameter g_jg_Q SUi(Cp₃ -(P₁) to output compensated Q and I signals respectively, which should ideally equivalent to the respective baseband signals B(t) , A{t) used to modulate radio frequency (RF) carriers ) . Referring to Figure 5, a heterodyne receiver 32 according to a second preferred embodiment of the invention has many components similar to those of the direct down conversion receiver 19 of the first preferred embodiment of the invention described above with reference to Figures 2 to 4 and the same reference numerals are used for the similar components. However, between the LNA 4 and the I and Q paths 5, 6, the receiver 32 according to the second preferred embodiment of the invention has a conversion stage 33 comprising a mixer 34 for mixing the amplified filtered received signal with a local oscillator signal at a frequency fo, an intermediate frequency filter 35 for filtering the signal after mixing and an AGC 36 for adjusting the gain of the signal for output to the I and Q mixers 7, 12. In this embodiment, the local oscillator signal used by the I and Q mixers 7, 12 is at an intermediate frequency (IF). The combination of the mixing with the local oscillator frequency at the frequency fo and the local oscillator signal at the intermediate frequency (IF) converts the received signal to baseband. Importantly, the wanted signal is therefore at baseband in the I and Q paths 5, 6.

In the present specification and claims the word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. Further, the word "comprising" does not exclude the presence of other elements or steps than those listed. The inclusion of reference signs in parentheses in the claims is intended to aid understanding and is not intended to be limiting.

Claims

1. A receiver (19; 32) for receiving a signal including a training sequence and a payload, the receiver (19, 32) comprising an In-phase path (5) and a Quadrature path (6) for converting the received signal to In-phase and Quadrature signals for output to a demodulator (17); processing means (23) for determining an imbalance compensation parameter from the In-phase and Quadrature signals representing the training sequence; and an imbalance compensation circuit (21 ) for compensating for phase and amplitude imbalance between the In-phase and Quadrature signals representing the payload using the determined imbalance compensation parameter.

2. The receiver (19; 32) of claim 1 , wherein the signal comprises a data packet containing the training sequence and the payload; and the imbalance compensation circuit (21 ) compensates the In-phase and Quadrature signals representing the payload of the data packet using the imbalance compensation parameter determined from the In-phase and Quadrature signals representing the training sequence of that data packet.

3. The receiver (19; 32) of claim 1 or claim 2, wherein the receiver also uses the training sequence for further signal processing.

4. The receiver (19; 32) of any one of the preceding claims, comprising a selector (22) for selecting portions of the In-phase and Quadrature signals at two different times and wherein the processing means (23) determines the imbalance compensation parameter from the portions of the In-phase and Quadrature signals selected at the two different times.

5. The receiver (19; 32) of any one of the preceding claims, wherein the processing means (23) bases the determination of the imbalance compensation parameter on signals representing the training sequence as expected to be received.

6. The receiver (19; 32) of claim 4, wherein the selector (22) selects the portions of the In-phase and Quadrature signals from which the processing means (23) determines the imbalance compensation parameters at times when the signals as expected to be received differ.

7. The receiver (19; 32) of any one of the preceding claims, wherein the processing means (23) determines an imbalance compensation parameter based on the In-phase signal and an imbalance compensation parameter based on the Quadrature signal.

8. The receiver (19; 32) of any one of the preceding claims, wherein the processing means (23) determines a pair of imbalance compensation parameters based on the In-phase signal and a pair of imbalance compensation parameters based on the Quadrature signal.

9. The receiver (19; 32) of claim 7 or claim 8, wherein the processing means (23) calculates a further imbalance compensation parameter from the determined imbalance compensation parameters.

10. The receiver (19; 32) of claim 9, wherein the processing means

(23) calculates a mean of each of the imbalance compensation parameters based on the In-phase signal and the Quadrature signal and calculates the further imbalance compensation parameter based on these calculated mean imbalance compensation parameters.

11. The receiver (19; 32) of any one of the preceding claims, wherein the imbalance compensation circuit (21 ) compensates for phase and amplitude imbalance between the In-phase and Quadrature signals by multiplying the In- phase and Quadrature signals by the determined imbalance compensation parameter(s).

12. The receiver (19; 32) of any one of claims 7 to 10, wherein the imbalance compensation circuit (21 ) compensates for phase and amplitude imbalance between the In-phase and Quadrature signals by multiplying the In- phase signal by the imbalance compensation parameter(s) determined from the Quadrature signal and Quadrature signal by imbalance compensation parameter(s) determined from the In-phase signal.

13. The receiver (19; 32) of claim 9 or claim 10, wherein the imbalance compensation circuit (21 ) compensates for phase and amplitude imbalance between the In-phase and Quadrature signals by multiplying the In- phase and Quadrature signals by the inverse of the calculated further imbalance compensation parameter.

14. A method of receiving a signal including a training sequence and a payload, the method comprising converting the received signal to In-phase and Quadrature signals for output to a demodulator (17); determining an imbalance compensation parameter from the In-phase and Quadrature signals representing the training sequence; and compensating for phase and amplitude imbalance between the In- phase and Quadrature signals representing the payload using the determined imbalance compensation parameter.

15. The method of claim 14, wherein the signal comprises a data packet containing the training sequence and the payload; and the compensation comprises compensating the In-phase and Quadrature signals representing the payload of the data packet using the imbalance compensation parameter determined from the In-phase and Quadrature signals representing the training sequence of that data packet.

16. The method of claim 14 or claim 15, comprising also using the training sequence for further signal processing.

17. The method of any one of claims 14 to 16, comprising: selecting portions of the In-phase and Quadrature signals at two different times and wherein the imbalance compensation parameter is determined from the portions of the In-phase and Quadrature signals selected at the two different times.

18. The method of any one of claims 14 to 17, wherein the determination of the imbalance compensation parameter is based on signals representing the training sequence as expected to be received.

19. The method of claim 17, wherein the portions of the in-phase and quadrature signals from which the imbalance compensation parameter is determined are selected at times when the signals as expected to be received differ.

20. The method of any one of claims 14 to 19, comprising determining an imbalance compensation parameter based on the In-phase signal and an imbalance compensation parameter based on the Quadrature signal.

21. The method of any one of claims 14 to 19, comprising determining a pair of imbalance compensation parameters based on the In- phase signal and a pair of imbalance compensation parameters based on the Quadrature signal.

22. The method of claim 20 or claim 21 , comprising calculating a further imbalance compensation parameter from the determined imbalance compensation parameter(s).

23. The method of claim 22, comprising calculating a mean of each of the imbalance compensation parameters based on the In-phase signal and the Quadrature signal and calculating the further imbalance compensation parameter based on these calculated mean imbalance compensation parameters.

24. The method of any one of claims 14 to 23, wherein the compensation comprises multiplying the In-phase and Quadrature signals by the determined imbalance compensation parameter(s).

25. The method of any one of claims 20 to 23, wherein the compensation comprises multiplying the In-phase signal by the imbalance compensation parameter(s) based on the Quadrature signal and Quadrature signals by imbalance compensation parameter(s) based on the In-phase signal.

26. The method of claims 22 or claim 23, wherein the compensation comprises multiplying the in-phase and Quadrature signals by the inverse of the calculated further imbalance compensation parameter.

27. Computer program code adapted to carry out the method of any one of claims 13 to 24 when processed by computer processing means.