WO2013162444A2

WO2013162444A2 - An arrangement and a method for carrier signal recovery

Info

Publication number: WO2013162444A2
Application number: PCT/SE2013/050386
Authority: WO
Inventors: Zhongxia HE
Original assignee: He Zhongxia
Priority date: 2012-04-24
Filing date: 2013-04-10
Publication date: 2013-10-31
Also published as: SE536593C2; SE1250404A1; WO2013162444A3

Abstract

The present invention relates to a carrier recovery arrangement (100) for recovering an input modulated signal (R(t)), down- converted by a local oscillator with a frequency different from the carrier frequency of the input modulated signal (R(t)) such that a distorted down-converted signal (R_Q(t)) is provided. It comprises a delay means (14) for introducing a time delay into the distorted signal (R_Q (t)), a mixer (4) adapted to multiply the received distorted signal (R_Q(t)) with the delayed distorted signal and to provide an adjusted output signal, (R_adj(t)), with a distortion term, a low pass filter (5) adapted to extract a signal comprising the second harmonic of the distortion term from the adjusted output signal (R adj (t)) related to twice the frequency difference (2Δω) between the local oscillator and the carrier frequency of the input modulated signal (R(t)), and means for removing modulated data from said signal to provide a recovery signal free from modulated data that can be used to eliminate the distortion.

Description

Title:

AN ARRANGEMENT AND A METHOD FOR CARRIER SIGNAL RECOVERY

TECHNICAL FIELD

The present invention relates to an arrangement for carrier signal recovery having the features of the first part of claim 1. The invention also relates to a method for carrier recovery having the features of the first part of claim 10.

BACKGROUND

Carrier signal recovery and demodulation are needed to be able to estimate and compensate for frequency and phase differences between for example the local oscillator of a receiver in a communication system and the carrier wave of the received signal modulated in a transmitter, particularly such that coherent demodulation can be provided.

On the transmitting side of a communication system a carrier wave is modulated by a baseband signal. The baseband information is extracted from the modulated waveform in a signal received at the receiver. In an ideal communication system, the oscillator carrier frequencies of the transmitter and the receiver would be matched perfectly as far as both frequency and phase are concerned, which in turn would allow for a perfect coherent demodulation of the modulated baseband signal. Normally, however, the carrier frequencies of the oscillators of the transmitter and the receiver are not the same since the receivers and the transmitters normally are completely independent and they have their own oscillators with frequency and phase offsets, frequency instabilities, phase instabilities etc .

In order to be able to permit coherent demodulation, all these frequency and phase variations have to be estimated using information in the received signal to reproduce or recover the carrier signal at the receiver.

There are two major categories of methods for handling carrier recovery or compensation. One of the methods in a first category is based on using a closed-loop compensation structure with a phase extraction block and a loop filter LF, which is schematically illustrated in Fig. 1, and by means of which phase and frequency offset are detected, and the offset is compensated for by tuning the local oscillator, VCO (Voltage Control Oscillator), of the receiver.

Fig. 2 illustrates an example of the other category which is based on an open-loop compensation structure which, instead of implementing a feedback-loop, a closed-loop, uses an open loop, a forward feed-loop, for cancelling the frequency offset by means of a derotator.

For the closed-loop structure (Fig. 1) and the open-loop structure (Fig. 2), there is a phase extraction block which extracts the frequency offset from a received signal given as: r{t) = cos [(ω + Αω)ί + φ_άαία + φ₀ ] Δω being the frequency offset between transmitter and receiver,

(Pdata being the phase containing the modulated data information, and cpo being a random phase caused by the propagation, LF being a loop filter.

Since cpdata is a data related random parameter which changes with time, it is difficult to extract Δω from r(t) . There are different known methods for performing such a phase extraction. A first method is based on data aided extraction which comprises transmitting known data at certain time instants (pilot data) , and to, based on this known information, enable extraction of

Δω . Other methods that do not use data aided extraction are also known. In one such method, here referred to as a multiply- filter-divide method, non-data-aided carrier recovery, a non- linear operation is applied to a modulated signal to create harmonics of the carrier frequency with the modulation removed. The carrier harmonic is then band-pass filtered and frequency divided to recover the carrier frequency, a PLL (Phase Locked Loop) may then follow. Multiply-filter-divide is an example of open-loop carrier recovery, which is favored in burst transactions since the acquisition time is typically shorter than for closed-loop synchronizers.

For example, in a QPSK (Quadrature Phase Shift Keying) case, a received signal given as

R_QPSK(0 = A{f) cosicoxpt + ^ηπ / 2); ^η = 0,1,2,3 is multiplied by four in a frequency multiplier giving: RQPSK(0^: ------- [3 + 4 cos (2

By using this approach, the unknown data-related phase is normalized into n2n. The signal is then divided by four, and GO F can be obtained (see Fig. 3) : HPF indicates a high pass filter, LPF low pass filter.

Another method uses a so called Costas Loop. A Costas Loop is a phase-locked loop used for carrier phase recovery from suppressed-carrier modulation signals, such as from double- sideband suppressed carrier signals. The main application of Costas loops is in wireless receivers. Its advantage over the PLL-based (Phase Locked Loop) detectors is that at small deviations the Costas loop error voltage is sin (2 (6i-6_f ) ) instead of sin(6i-6_f) . This doubles the sensitivity and also makes the Costas loop uniquely suited for tracking Doppler-shifted carriers.

In modern DSP (Digital Signal Processor) based receivers a decision directed method is often adopted. In decision directed carrier recovery the output of a symbol decoder is fed to a comparison circuit and the phase difference/error between a decoded symbol and a received signal is used to control the local oscillator. A common form of decision directed carrier recovery begins with quadrature phase correlators producing in-phase and quadrature signals representing a symbol coordinate in the complex plane. This point should correspond to a location in the modulation constellation diagram. The phase error between the received value and nearest/decoded symbol is calculated using arctangent (or an approximation) . However, arctangent can only compute a phase correction between 0 and n/2. Most QAM (Quadrature Amplitude Modulation) constellations also have n/2 phase symmetry.

In many systems it is not allowed to introduce pilots in transmitted data, which means that data aided methods cannot be used. For a multiply-filter-divide method, when carrier frequency is high (for example over 15 GHz), frequency multiplication by four is difficult to achieve, since means capable of doing that are very complicated to design in practice. Furthermore, for higher modulation formats, i.e. 8-PSK (Phase Shift Keying) , frequency multiplication by eight is required. Such components are extremely difficult to design in practice .

The complexity of a Costas Loop is increased dramatically for higher modulation formats, and it requires that all branches are symmetric, which also is difficult to design in practice.

Decision-directed methods are commonly used with DSP (Digital Signal Processing), however, when the carrier/IF (Intermediate Frequency) is high, it is difficult to find available ADCs (Analog to Digital Converter) for sampling at such frequencies. SUMMARY OF THE INVENTION

It is therefore an object of the present invention to provide an improved arrangement for carrier signal recovery as initially referred through which one or more of the above mentioned problems can be overcome. It is particularly an object of the invention to provide an arrangement for carrier signal recovery which is easy and cheap to design and fabricate. It is also an object to provide an arrangement which is suitable for high data rates, particularly for communication systems with high data rates and for high carrier frequencies. It is particularly an object to be able to provide an arrangement which can be used in systems which do not allow introduction of pilote signals, and which does not require symmetric branches. It is a particular object to provide an arrangement which is suitable for high modulation formats and which imposes lower requirements on involved components than known arrangements and which particularly can be used for high modulation formats without requiring modification.

Therefore an arrangement as initially referred to is provided which has the features of the characterizing part of claim 1.

A receiving arrangement including such an arrangement for carrier signal recovery is also provided, as well as a method for carrier signal recovery, which has the features of the characterizing part of claim 10.

It is an advantage of the invention that an arrangement is provided which is capable of handling high data rates and high carrier frequencies while still being easy to design and fabricate. It is also an advantage that an arrangement is provided which can be used for high modulation formats without requiring any modification and which, in advantageous embodiments, even removes the need for using any frequency multipliers and enables the use of comparatively simple frequency dividers.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will in the following be more thoroughly described, in a non-limiting manner, and with reference to the accompanying drawings, in which:

Fig. 1 shows a state of the art closed loop carrier recovery compensation structure,

Fig. 2 shows a state of the art open loop carrier recovery compensation structure,

Fig. 3 shows a state of the art carrier recovery structure based on multiplication, filtering and division for performing a phase extraction,

Fig. 4 shows a carrier signal recovery arrangement according to a first embodiment of the invention, Fig. 5 shows a carrier signal recovery arrangement according to a second embodiment of the invention,

Fig. 6 shows a receiving arrangement with a carrier signal recovery arrangement according to one embodiment of the invention, Fig. 7 shows a recovery arrangement according to the invention included in a closed loop structure,

Fig. 8 shows a carrier recovery arrangement according to one embodiment of the invention included in an open loop structure, and

Fig. 9 is a schematic flow diagram describing a procedure for carrier signal recovery according to one embodiment of the invention.

DETAILED DESCRIPTION

Fig. 4 shows a block diagram of a first embodiment of the invention wherein a distorted signal R.Q(t) is received in an arrangement 100. The distorted signal is here a signal which has been down-converted in (here) an IQ direct converting mixer and by means of a local oscillator a sinusoid source has been provided at a receiver carrier frequency. Generally any down- converted, distorted, signal or particularly any output distorted signal from the mixer may form the input signal to the arrangement 100, i.e. it could also have been the Ri(t) which had been input and it would then be recovered in a similar manner as shown in Fig. 4. In a delay means 14 Ro(t) is delayed (Δτ) and in a mixer 4 the time delayed signal is multiplied by the distorted down-converted signal Ro(t) giving an adjusted output signal R_ad (t) which then is input to a low pass filter (LPF) 5. In an advantageous embodiment the low pass filter 5 is so designed that only one signal is selected, namely the second harmonic of the distortion term of the distorted signal, which is only related to 2Δω, Δω being difference between the frequency of the down-converted distorted signal cod, the frequency of the received carrier signal ω_α and the frequency of the local oscillator of the receiver, COLO^ i.e. the second harmonic of the distortion term, the second order of the difference signal, which is achieved by giving the low pass filter a cut-off frequency which is much lower than the data symbol rate so that the output only contains low frequency components and no data.

An output signal R"_adj (t) is thus obtained which only relates to twice the frequency offset, 2Δω. This signal can be used to recover an undistorted signal, which can be done in different ways, some of which will be further explained and exemplified below .

Fig. 5 shows an alternative implementation, similar means as in Fig. 4 are indicated by the same reference numerals but are given an index 1 and will therefore not be described any further herein to the extent that they perform the same function. The difference is here that the requirements on the low pass filter 5i are somewhat less stringent, important being that it be given such a cut-off frequency that only the second harmonic of the distortion term of the noise (difference signal) is selected, but the output signal R'_adj(t> does still contain the modulated data. For the purpose of eliminating data, an envelope detector 6i is introduced which has a time constant which is longer than the symbol period and so the data will be eliminated, and an output signal R"_adj{t) as described in the embodiment shown in Fig. 4 is provided which does not contain any data but only the second order difference signal. In this embodiment it is thus the envelope detector 6 that it responsible for removing the data, whereas in the embodiment of Fig. 4 the low pass filter 5 is carefully selected so that also the data is removed, avoiding the need for any envelope detector.

Fig. 6 shows a receiving arrangement 300A with a recovery arrangement 300 substantially as the arrangement shown in Fig. 5.

The receiver arrangement 300A contains an IQ direct down- converting mixer I₂, and a local oscillator 2₂ providing a sinusoid source at carrier frequency. Thus, a signal R(t) from a transmitter with its local oscillator is received (considered undistorted) , down-converted in mixer I₂ in a receiver with local oscillator 2₂ and a frequency COLO- The distortion produced is

o)d being the frequency of the down-converted distorted signal, wherein ω_α is the frequency of the receiver carrier signal, GOLO the frequency of the receiver local oscillator. One output of the mixer I₂ here R.Q(t) and a delayed- by-T version thereof are both input to a mixer 4₂, where they are multiplied to provide an output signal R_ad (t) . Passing R_adj (t) through a low-pass filter 5₂, the output will be R' adj(t) as explained below. R'_adj(t) is fed to an envelope detector 6₂. The output R"adj (t) from the low-pass filter 6₂ represents the second harmonic of the frequency offset between a transmitter oscillator, not shown, and which does not form part of the present invention, and a receiver oscillator 2₂. To obtain the true frequency offset, a balun 8₂ is used to convert a single-end signal to two differential signals which are 180° out of phase. These out-of-phase signals are then input to frequency dividers 9₂, 10₂, giving outputs cos (Aoot) and sin(Aoot) . By using phase detectors lli,ll₂ the modulated I and Q signals can be extracted, thus demodulation is achieved. To explain the behavior of this structure, a mathematic expression of the demodulation process is given according to the following:

It is assumed that the received signal can be represented as R(t) = cos[coi + φ] [ 1 ] where

wherein (po is a propagation related slow change value, and (pdata is data related and changing at symbol rate. After the IQ direct down-converting mixer I₂, the output can be written as:

R_Q (t) = cos(ct»t + φ)^χ cos[(ct» + Αω)ί]

= · {cos[(2co + Ao)t + φ] + cos(Aot - φ)}

[2]

R_j (t) = cos(cot + (p) x sin [(ct> + Δω)ί]

[3]

= i {sin [(2o + Ao)t + φ] + sm(Aot - φ)}

where it is assumed that LO 2₂ gives a frequency fw ( = cos[(ffl + Affl)i] .

In the proposed method, either R._Q(t) or Ri(t) can be used for carrier recovery. Here R._Q(t) is taken as input to the mixer ₂.

Another input R._Q(t) of the mixer ₂ is provided to a delay element 3₂, which provides a real time delay to the signal, the output thus being: R_Q (t + T) = cos[o(t + Τ) + φ]χ cos[(o + Αω)(ί + T)]

[ 4 ]

= {cos[(2o + Αω)(ί + T) + φ] + cos(Ao(t + Τ) - φ)} wherein T is an certain fixed true time delay, which satisfies T « T_Sym , so that φ in R_Q(t) and Ri(t) can be considered to the same.

The output of the mixer 42 can then be calculated as follows:

R_ad] (t) = R_Q (t) R_Q (t + T)

= i {cos[(2o + Ao)t + φ] + cos(Aot - φ)}

x {cos[(2o + Αω)(ί + Τ) + φ]+ cos[Ao(t + T) - φ]}

= - {COS[(2CD + Δω )t + φ ] x cos[(2o + Δω )(t + T) + φ ]}

+ A {COS[(2CD + Αω)ί + φ] x cos[Aco(t + T) - φ]}

+ -{cos(Aot - φ) x COS[(2CD + Δω)(Υ + Τ) + φ]}

+ -ⁱ-{cos(Act>t - φ) x cos[Act>(t + T) - φ]}

= {COS[(4CD + 2Δω)/ + (2ω + Αω)Τ + 2φ] + cos[(2o + Δω)Γ]}

+ A {COS[(2CD + 2Δω)ί + ΑωΤ] + cos(2ot - ΔωΓ + 2φ)}

+ {COS[(2CD + 2Αω)ί + (2ω + Αω)Τ] + cos[2ot + (2ω + Αω)Τ + 2φ]}

+ {cos[2Ao)t + ΑωΤ - 2φ] + cos(AoT)}

[ 5 ]

According to the invention it is a recovery signal based on the underlined term that is extracted (data has to be removed) and to be used for recovery.

By properly designing low-pass filter LPF 52 right after the mixer 42 to have a cut-off frequency which is lower than the data rate, the output of the filter will contain only really low frequency components. Thus the output of the LPF can be represented as (A being the amplitude) :

R_a'_d] t) = A x {cos[2Aot + ΑωΤ - 2φ]+ cos(AcDJ)} [ 6 ]

After filter 5₂ there is in this embodiment an envelope detector 6₂, the output of which is only related to 2Δω, assuming the frequency offset being much lower than data rate: R:_dj (t) = A x cos(2AcoT) [7]

A balun 8₂ provides two outputs with 180° phase difference:

R_Bahm (.t) = A x∞s(2Aa>T) [8]

R_B'_alun (t) = A x cos[2A(oT + n] [9] and passing these signals through frequency dividers 9₂, IO₂ gives :

R_FD (t) = A x cos(Aot) [10]

R_F'_D (t) = A x cos[Aot + π/ 2] = A x sin( Aot) [11] To low pass filters LPF Ί ι , Ί are input the distorted, down- converted signals Ri(t), R_Q(t) giving the outputs:

R_Q' (t) = cos(Aot + (p) [12]

R_j' (t) = sin(Aot + (p) [13] Using the phase detector lli,ll2 the phase differences between [11] and [13] and [10] and [12] respectively are compared. The outputs of phase detectors Hi, II₂ contain the data modulated phase. Thus a demodulation is made.

Fig. 7 schematically illustrates an implementation of a recovery arrangement 400 in a closed loop recovery compensation structure 400A in which it acts as a phase extraction arrangement. Similar components are given the same reference numerals as in the preceding figures but with an index 3. It is schematically illustrated how the recovery signal, with the extracted phase, from the envelope detector 6₃ is input to a loop filter 15₃ in a compensation structure similar to the one discussed with reference to the state of the art structure of Fig. 1.

Fig. 8 illustrates an implementation of a recovery arrangement comprising an arrangement for phase extraction 500 as implemented in an open loop carrier recovery compensation structure 500A. Again components which have been discussed earlier in the description are given the same reference numerals, but with an index 4. The signal output from the envelope detector 6₄, with the extracted phase, is here input to a loop filter 15 and to a derotator I6₄ which uses the signals from the envelope detector and the loop filter to perform a derotation .

It should be clear that also in these embodiments the envelope detector 6₃, 6₄ could have been disposed of if the corresponding loop filter 5₃, 5₄ were properly designed, so as to remove also all data. Fig. 9 is a schematical flow diagram describing the procedure for extracting a phase to be used for recovery of a distorted signal. It is supposed that a distorted input signal, for example from an IQ down converting mixer is received. A time delay, ΔΤ, is introduced into the signal, 100, and the delayed, distorted signal is multiplied by the distorted signal (without delay) in a mixer, 101. Subsequently, using an LP filter, the second harmonic, depending only on 2Δω, the second order of the difference signal as explained above, is extracted, 102, by giving the filter a cut-off frequency which is so selected that only low frequencies are selected, 102. In an appropriate manner modulated data is removed from the extracted signal to obtain a recovery signal. This can be done either in the LP filter if appropriately designed or in an envelope detector or any other appropriate means, 103. Subsequently the recovery signal from the envelope detector can be used for recovery of the distorted down-converted signal and data can be demodulated, 104. It should be clear that the down-converted signal (and hereby distorted) does not have to be down-converted in an IQ down- converter, but the inventive concept is of course also applicable for otherwise distorted, down-converted, signals.

It is an advantage of the invention that no high frequency component is needed, for example as compared to the multiply- filter-divide method, neither any high frequency multiplier, nor any high frequency divider are needed.

Another advantage is that the implementation complexity is very low and no constructional modification is needed for high modulation formats as long as the symbol rate is much higher than the frequency offset.

Another advantage is that it is suitable for high frequency or high data rates without demanding any high performance components, mixers, dividers, ADCs, which suffer from being extremely expensive and difficult to provide.

The invention is not limited to the explicitly illustrated embodiments, but can be varied in a number of ways within the scope of the appended claims.

Claims

1. A carrier recovery arrangement ( 100 ; 200 ; 300 ; 400 ; 500 ) for recovering an input modulated signal (R(t)), particularly down- converted by a local oscillator with a frequency different from the carrier frequency of the input modulated signal (R(t)), such that a distorted down-converted signal (R_Q(t)) is provided, c h a r a c t e r i z e d i n

that it comprises a delay means ( 14 ; 14₃; 14₄) for introducing a time delay into the distorted signal (R_Q(t)), a mixer ( 4 ; 4i; 4₂ ; 4₃; 4₄ ) adapted to multiply the received distorted signal (R_Q(t)) with the delayed distorted signal and to provide an adjusted output signal, (R_adj (t) ) , with a distortion term, a low pass filter (5; 5i; 52," 5₃; 5₄) adapted to extract a signal comprising the second harmonic of the distortion term from the adjusted output signal (R_adj (t) ) related to twice the frequency difference (2Δω) between the local oscillator and the carrier frequency of the input modulated signal (R(t)), and in that means are provided which are adapted to remove modulated data from said signal to provide a recovery signal free from modulated data that can be used to eliminate the distortion.

2. An arrangement (100) according to claim 1,

c h a r a c t e r i z e d i n

that the low pass filter (5) is adapted to select, extract, only the second harmonic of the distortion term, and hence is adapted to remove also the modulated data.

3. An arrangement (100) according to claim 1,

c h a r a c t e r i z e d i n that the low pass filter (5) has a cut off frequency which is much lower than a symbol rate at which data is transmitted in the input signal thus providing a data free recovery signal

4. An arrangement (200; 300; 400; 500) according to claim 1,

c h a r a c t e r i z e d i n

that it further comprises an envelope detector (6;6₂;6₃;6 ) adapted to remove modulated data from the signal output from the low pass filter (5i; 5₂; 5₃; 5₄) to generate a data free recovery signal (R"_adj (t) ) .

5. An arrangement (400) according to any one of the preceding claims ,

c h a r a c t e r i z e d i n

that it is adapted to be arranged in a closed loop compensation structure (400A) comprising a loop filter (15₃) to receive the recovery signal and to tune the local oscillator (2₃) .

6. An arrangement (500) according to any one of claims 1-4, c h a r a c t e r i z e d i n

that it is adapted to be arranged in an open loop compensation structure (500A) and in that it comprises a loop filter (15 ) adapted to receive the recovery signal and to provide a filtered control signal to a derotator (I6₄) adapted to use said low pass filtered control signal and the recovery signal for frequency distortion cancellation.

7. An arrangement (300) according to any one of claims 1-4, c h a r a c t e r i z e d i n that the envelope detector is adapted to be connected to, or comprise, a balun (8), adapted to convert the recovery signal free from data to two differential signals which are 180° out of phase, and frequency dividers (9₂, 10₂) for dividing the out of phase signals by two.

8. An arrangement according to claim 7,

c h a r a c t e r i z e d i n

that the first and a second frequency dividers (9₂, 10₂) are adapted to provide respective first and second signals terms (cos (Acot) , sin(Acot)) to a first and a second phase detector (lli,ll₂) for comparing the respective signal terms with modulating signals Ri ' (t) and R_Q' (t) respectively to provide modulated output signals I_out(t), Q_out(t) .

9. A receiving arrangement in a communications system,

c h a r a c t e r i z e d i n

that it comprises a carrier recovery arrangement (100;200;300;400;500) according to any one of claims 1-8.

10. A method for recovering an input modulated signal (R(t)) down-converted by a local oscillator with a frequency different from the carrier frequency of the input modulated signal (R(t)) such that a distorted down-converted signal (R_Q(t)) is provided, c h a r a c t e r i z e d i n

that it comprises the steps of:

- time delaying a distorted down-converted signal output from a down-converting mixer ( 1 ) ;

- multiplying the delayed distorted down-converted signal with the distorted signal in a mixer ( 4 ; 4 ; 2 ; 4₃; 4 ) to provide an adjusted signal with a distortion term; - extracting, by means of a low pass filter (5; 5i; 5₂; 5₃; 5₄) , a second harmonic of the distortion term related to twice the frequency difference (2Δω) between the frequency of the local oscillator and the carrier frequency of the input, modulated signal (R (t) ) ;

- removing data from the extracted signal to obtain a recovery signal ;

- using the recovery signal to eliminate the distortion.

11. A method according to claim 10,

c h a r a c t e r i z e d i n

that it comprises the step of:

- removing the data by means of setting the cut off frequency of the low pass filter (5) lower than the data rate such that data is removed.

12. A method according to claim 10,

c h a r a c t e r i z e d i n

that it comprises the step of:

- removing the data by means of an envelope detector ( 6i; 6₂; 6₃,· 6₄) comprising:

- inputting the extracted signal to the envelope detector;

- removing the data from the extracted signal by setting the time constant of the envelope detector ( 6₁; 6₂; 6₃; 6₄) to a period which is longer than the symbol period.

13. A method according to any one of claims 10-12,

c h a r a c t e r i z e d i n

that it comprises the steps of:

- converting the recovery signal to differential signals 180° out of phase by using a balun (8); - dividing the frequency of the differential signals by two in respective frequency dividers (92, 102);

- using phase detectors (lli,ll₂) for, by means of the divided differential signals, demodulating the received, distorted down-converted signals.