DE4101802C1 - In-phase and quadrature detector for QAM receiver - demodulates using unregulated orthogonal carriers and controls carrier and quadrature phase offsets at baseband - Google Patents

Abstract

The circuit recovers the communication signal transmitted, according to quadrature amplitude modulation techniques, in a receiver. The QAM signal received is demodulated with unregulated orthogonal carriers. In this way four carrier functions (v1...v4) are produces, with parameters psi, rho and sigma. These parameters are regulated in a control circuit (D1, D2, D3, LF1, LF2, LF3) so that one of the transmitted communications signals is allotted, in a single-valued fashion, to each of the demodulated signal components (As, Bs) coming from the control circuit. ADVANTAGE - Improvement in reliability, by reducing errors, using inexpensive equipment.

Description

The present invention is based on a circuit arrangement for determining the in-phase and quadrature signal components a quadrature amplitude modulated (QAM) signal in one QAM receiver, the QAM signal with two orthogonal Carrier converts into the baseband, the signal components

a _s (t) = a (t) cos (Δωt + Φ - R + δ + ε) + b (t) sin (Δωt + Φ - R - δ + ε)

b _s (t) = -a (t) sin (Δωt + Φ - R + δ - ε) + b (t) cos (Δωt + Φ - R - δ - ε)

arise with the orthogonal QAM signal carriers modulated message signals a and b, the Carrier frequency difference Δω between the carrier frequency of the QAM signal and the carrier frequency generated in the receiver, the Carrier phase Φ of the QAM signal, the receiver carrier phase R, the quadrature phase error δ of the QAM signal carriers and the Quadrature phase error ε of the receiver carrier.

Such a circuit arrangement is for example from the Conference proceedings of the "2nd European Conference on Radio-Relay Systems, 17th-21st April 1989, Albano Terme-Padua, Italy, p. 267 to 274 known. To the transmitted message signals to be able to recover as flawlessly as possible  usually a regulation, the frequency, the phase and the Carrier quadrature phase performed in the receiver. That I that is, the regulation in the high-frequency range can play digital circuit components hardly used will.

From US 47 12 222 a circuit arrangement emerges with which the carrier phase of a recipient on the carrier phase of Received signal can be tracked. It will Received signal converted into a complex baseband, and this becomes complex through an analog-digital conversion Samples generated. The phase of every complex Sampling value is with the phase of the previous one complex sample value compared and the phase offset determined. The phase shift signal controls a phase shifter so that the phase offset between the receiver carrier and the Received signal carrier disappears. This circuit arrangement is unable to receive all when receiving a QAM To regulate signals and frequency offsets occurring so that finally the transmitted message signals be clearly recovered.

The invention has for its object a Specify circuit arrangement of the type mentioned at the outset, which with little circuitry the through a QAM signal transmitted message signals as pure as possible regained.  

According to the invention, this object is achieved through the features of Claim 1 solved. Advantageous versions of the invented Circuit arrangement emerge from the subclaims.

Since according to the invention the QAM signal received with two unregulated carriers is implemented in the baseband and only then (in the low frequency range) the influences of faulty carrier frequency, the carrier phase and the Quadrature phase can be corrected, the necessary for this Circuit arrangement with little complex digital Circuit elements can be realized.

Using several shown in the drawing Exemplary embodiments of the invention will now be explained in more detail. It shows

Fig. 1 shows a part of a QAM receiver,

Fig. 2 circuit means for determining the in-phase and the quadrature signal,

FIGS. 3 and 4 embodiments, a discriminator for the carrier frequency and phase,

FIGS. 5 and 6 embodiments, a discriminator for the quadrature phase error of the QAM signal carrier,

Phase Figs. 7 and 8 embodiments of a common discriminator for the carrier frequency and and for the quadrature phase error of the QAM signal carrier,

Fig. 9, 10 and 11 embodiments, a discriminator for the quadrature phase error of the receiver carrier,

Fig. 12 a support function generator.

Has a quadrature amplitude modulated (QAM) signal the shape is known

y (t) = a (t) cos (ω₀t + Φ) + b (t) sin (ω₀t + Φ) (1)

Here a (t) and b (t) represent two mutually independent Message signals represent the two orthogonal carriers cos (ω₀t + Φ + δ) and sin (ω₀t + Φ-δ) modulated in the QAM transmitter will. With ω₀ is the carrier frequency and with Φ the phase of Carrier designated. The orthogonality (quadrature phase of 90 °) between the carriers can usually not be adhered to exactly will. The deviation from the orthogonality is expressed in the quadrature phase error δ.

In the QAM receiver, the QAM signal is transmitted using two carriers

c₁ (t) = 2 cos (ω₁t + R - ε) (2)

c₂ (t) = 2 sin (ω₁t + R + ε) (3)

whose frequency is ω₁ and whose phase is R, demodulated, d. H. in implemented the baseband. The receivers also have one Quadrature phase error; it is denoted by ε. After a Subsequent low-pass filtering results in the two Signal components

a _s (t) = a (t) cos (Δωt + Φ - R + δ + ε) + b (t) sin (Δωt + Φ - R - δ + ε) (4)

b _s (t) = -a (t) sin (Δωt + Φ - R + δ - ε) + b (t) cos (Δωt + Φ - R - δ - ε) (5)

where Δω = ω₀-ω₁ is the difference between the carrier frequency ω₀ of the QAM signal y (t) received and that in the receiver generated carrier frequency ω₁ represents.

In order to obtain a first signal component a _s (t), which only depends on the message signal a (t), and a second signal component b _s (t), which contains only the message signal b (t), the carriers are formed in accordance with known methods Receiver controlled so that ω₁ = ω₀, R = Φ and ε = δ applies.

With the circuit arrangements described below are used to determine the actually transmitted Message signals a (t) and b (t) not regulated the carriers, but only after the received QAM signal with the unregulated carriers has been implemented in the baseband, there is a control process that is reliable Detection of the message signals a (t) and b (t) guaranteed. Because the regulation after implementation in the low frequency range takes place, the regulation by digitally working Circuit elements can be realized. Because with a digital Signal processing only time-discrete signal states occur, the time dependency in the  following are omitted. Therefore the time variable t occurs no longer appears in the following signal displays.

Fig. 1 shows a circuit arrangement which b from the sampled values a _s and b _s of the converted in the baseband signal components a _s (t) and b _s (t) is a message signal a re-imaging in-phase signal A _s and a the message signal, reproducing Quadrature signal B _s wins.

The in-phase signal A _s and the quadrature signal B _s are obtained from a module PWK from the signal components a _s , b _s and four carrier functions v₁, v₂, v₃, v₄ by the following links:

A _s = a _s · v₁ - b · _s v₃ (6)

B _s = a _s · v₄ + b _s · v₂ (7)

The four carrier functions have the form:

v₁ = cos (ψ - ρ - σ) (8)

v₂ = cos (ψ + ρ + σ) (9)

v₃ = sin (ψ - ρ + σ) (10)

v₄ = sin (ψ + ρ - σ) (11)

Like these carrier functions and their parameters ψ, ρ and σ arise, will be explained below.

An embodiment of the module for generating the in-phase signal A _s and the quadrature signal B _{s is} shown in FIG. 2.

Four multipliers M1. . . M4 form the products a _s · v₁, a _s · v₄, b _s · v₃ and b _s · v₂.

The additive and subtractive combination of the four products according to equations (6), (7) provide an adder A and Subtractor S. Stand at the two outputs of the PWK module then the in-phase signal

A _s = a [cos (ξ + δ + ε) cos (ψ - ρ - σ) + sin (ξ + δ - ε) sin (ψ - ρ + σ)] + b [sin (ξ - δ + ε) cos (ψ - ρ - σ) - cos (ξ - δ - ε) sin (ψ - ρ + σ)] (12)

and the quadrature signal

B _s = a [cos (ξ + δ + ε) sin (ψ + ρ - σ) - sin (ξ + δ - ε) cos (ψ + ρ + σ)] + b [sin (ξ - δ + ε) sin (ψ + ρ - σ) + cos (ξ - δ - ε) cos (ψ + ρ + σ)] (13)

available, where the variable ξ = Δωt + Φ-R. Succeeds in a subsequent control circuit the conditions ψ = ξ, ρ = δ and σ = ε, there is finally an in-phase and a quadrature signal

A _s = aCos (2δ) cos (2ε) (14)

B _s = bcos (2δ) cos (2ε) (15)

It turns out that there are no longer any "crosstalk terms" between the in-phase and quadrature signals, because the in-phase signal A _s is solely dependent on the message signal a and the quadrature signal B _{s is} only dependent on the message signal b. Only an amplitude distortion dependent on the quadrature phase errors δ and ε of the carriers remains, which can be completely eliminated in a suitable level control.

As can be seen in FIG. 1, the PWK module is followed by two decision-makers E 1 and E 2 , which assign the most likely transmitted message signals to the in-phase signal A _s and the quadrature signal B _s , which here are used as estimation signals a and b be designated. In addition, the deposits A _s -â and B _s - of the in-phase and quadrature signals A _s , B _s from the assigned estimation signals â are determined by two subtractors S 1 and S 2 . The mentioned control circuit of the circuit arrangement shown in FIG. 1 contains three discriminators D 1 , D 2 and D 3 . The first discriminator D 1 provides an estimate of the sum of the carrier frequency difference Δω = ω₁-ω₀ and the carrier phase difference Φ-R. For this purpose, this first discriminator D 1 forms an output signal of the form from the in-phase signal A _s , the quadrature signal B _s and the estimation signals a and b:

d₁ = â (B _s -) - (A _s - â) (16)

A downstream of the first discriminator D 1 control filter LF 1 filters out the output signal d 1 by the carrier frequency difference .DELTA..omega. And the carrier phase difference .DELTA..times .RTM. Size .DELTA..omega. . . v₄ generating carrier function generator TFG. The expected value E {d₁} of the output signal takes effect through the control process.

E {d₁} = 2S [sin ((Δω - Ω) t + ϕ - η) · cos (δ - ρ) · cos (ε + σ) + sin ((Δω + Ω) t + ϕ + η) · sin (δ + ρ) · sin (ε - σ)], (17)

with ϕ = Φ-R and ψ = Ωt + η, where S = E {a²} = E {b²} is the constant Power of the message signals a, b represents. There one statistical independence between the message signals a  and b can be assumed, the disappears Expected value of the product of the two message signals (E {abb} = 0).

For small quadrature phase errors, | ε-δ | «1 applies. So that's the Expected value

E {d₁} ≈ 2Ssin ((Δω - Ω) t + ϕ - η) cos (δ - ρ) cos (ε + σ) (18)

strongly from the carrier frequency difference Δω-Ω and the Carrier phase difference δ-η dependent, but comparatively weak of the angles δ, ρ, ε and σ, if only small Quadrature errors are assumed.

The first discriminator D 1 , which produces the output signal d 1 according to equation (16), can have the embodiment shown in FIG. 3. In it, two multipliers M 5 and M 6 produce the products â (B _s -) and (A _s -â), and a subtractor S 3 forms the difference between the two products.

In many cases it is sufficient if the control information formed by the discriminators D 1 , D 2 , D 3 only the direction (+, -) for the readjustment of the parameters ψ, ρ, σ of the four carrier functions, v₁, v₂, v₃, v₄ indicates. In this case, the multipliers M 5 and M 6 (see FIG. 3) in the first discriminator D 1 can be replaced by EXOR gates EX 1 and EX 2 , to which the two-stage quantized sign functions of the signals â,, A _s - â and B _s - are fed.

The second discriminator D 2 provides an estimate of the offset between the quadrature phase error δ of the QAM signal carrier and the parameter ρ of the four carrier functions v₁, v₂, v₃, v₄. For this purpose, this discriminator D 2 generates the in-phase signal A _s , the quadrature signal B _s and the estimation signals â and an output signal of the form:

d₂ = (A _s - â) + â (B _s -) (19)

A downstream of the second discriminator D 2 control filter LF 2 derives the difference δ-ρ between the quadrature phase error δ of the QAM signal carriers and the parameter ρ from the output signal d₂ and causes the parameter to be updated to the quadrature phase error δ.

The following applies to the expected value E {d₂} of the output signal d₂ of the second discriminator D 2 :

E {d₂} = 2S [- cos (Δω - Ω) t + ϕ - η) sin (δ - ρ) cos (ε - σ) + cos ((Δω + Ω) t + ϕ + η) cos (δ + ρ) · sin (ε - σ)] (20)

with ϕ = Φ-R and ψ = Ωt + η.

The following applies to | ε-σ} «1:

E {d₂} ≈ -2Scos ((Δω -Ω) t + ϕ - η) sin (δ - ρ) cos (ε + σ) (21)

Here the strong dependence of the expected value E {d₂} from the offset between the quadrature phase error δ of the QAM Signal carrier and the parameter ρ clearly.

FIG. 5 shows an embodiment for the second discriminator D2. To form the output signal d₂ according to equation (19) serve two multipliers M 7 and M 8 , which produce the products (A _s -â) and â (B _s -), and an adder A 1 , which the two products to the output signal d₂ additively summarized. Here, too, the multipliers M 7 , M 8 can be replaced by EXOR gates EX 3 , EX 4 (see FIG. 6) if the output signal d₂ is to contain only directional information for the readjustment of the parameter ρ.

Comparing the first discriminator D 1 from FIG. 3 and the second discriminator D 2 from FIG. 5, it is found that both differ only in the use of a subtractor S 3 or an adder A 1 . It is therefore advisable to implement a combined discriminator according to FIG. 7. The two multipliers M 9 and M 10 form the products â (B _s -) and (A _s -â). The subtractor S 4 generates the output signal d 1 of the first discriminator from the difference between the two products, and the adder A 2 summarizes the two products to the output signal d 2 of the second discriminator. As Fig. 8 shows, instead of the multiplier M 9, M 10 EXOR gate EX 5 EX 6 are inserted when d₁ of the output signals and d₂ only directional information is required.

The third discriminator D 3 provides an estimate for the offset between the quadrature phase error ε of the receiver carrier and the parameter σ of the four carrier functions v₁, v₂, v₃, v₄. This discriminator D 3 generates the following output signal from the signal components a _s , b _s , the estimation signals â, and the four carrier functions v₁, v₂, v₃, v₄ converted into the baseband:

d₃ = (a _s v₂ + b _s v₄) - â (a _s v₃ - b _s v₁) (22)

A third discriminator D 3 downstream control filter LF 3 derives from the output signal d₃ the difference ε-σ between the quadrature phase error ε of the receiver carrier and the parameter σ forth and causes a tracking of the parameter σ to the quadrature phase error ε.

The following applies to the expected value E {d₃} of the output signal d₃ of the third discriminator D 3 :

E {d₃} = 2S [cos ((Δω - Ω) t + ϕ - η) · cos (δ + ρ) · sin (ε - σ) - cos ((Δω + Ω) t + ϕ + η) · sin (δ - ρ) · cos (ε + σ)] (23)

with ϕ = Φ-R and ψ = Ωt + η.

For | δ-ρ | «1 applies

E {d₃} ≈ 2S [cos ((Δω - Ω) t + ϕ - η) · cos (δ + ρ) · sin (ε - σ)

In FIG. 9, a third embodiment of the discriminator D 3 is shown. First, four multipliers M 11 , M 12 , M 13 , M 14 form the four products a _s · v₂, b _s · v₁, a _s · v₃ and b _s · v₄. Two adders A 3 and A 4 each link two of these four products, and the resulting sum signals are multiplied by means of two multipliers M 15 and M 16 with the estimation signals a and b. Finally, a subtractor S 5 generates the output signal d₃ from the partial signals d₃₁ and d₃₂ supplied by the multipliers M 15 , M 16 according to equation (2).

In the event that only one direction information for the regulation of the parameter σ is required from the output signal d₃, as shown in FIG. 10, the multipliers M 15 and M 16 by EXOR gates EX 7 and EX 8 and also also as shown in FIG . 11, shows the four multipliers M 11, M 12, M 13 and M 14 by EXOR gate EX 9, EX 10, EX 11 and EX 12 are replaced.

A carrier function generator TFG, which consists of the parameters ψ, ρ and σ according to equations (8). . . (11) the four carrier functions v₁. . . v₄ generated, can be seen in FIG. 12. This carrier function generator TFG has four adders ADD 1 , ADD 2 , ADD 3 and ADD 4 , which are those in equations (8). . . (11) produce the specified linear combinations between the parameters ψ, ρ and σ. Finally, two cosine generators FG 1 , FG 2 and two sine generators FG 3 , FG 4 form the four carrier functions v 1 from the linear combinations of the parameters ψ, ρ, σ. . . v₄, which, as described above, the module PWK and the third discriminator D 3 are supplied.

Claims (5)

1. Circuit arrangement for determining the in-phase and quadrature signal components of a quadrature amplitude modulated (QAM) signal in a QAM receiver, which converts the QAM signal into the baseband with two orthogonal carriers, the signal components a _s (t) = a ( t) · cos (Δωt + Φ - R + δ + ε) + b (t) · sin (Δωt + Φ - R - δ + ε) b _s (t) = -a (t) · sin (Δωt + Φ - R + δ - ε) + b (t) · cos (Δωt + Φ - R - δ - ε) arise with the message signals a and b modulated onto the orthogonal QAM signal carriers, the carrier frequency difference Δω between the carrier frequency of the QAM signal and the carrier frequency generated in the receiver, the carrier phase Φ of the QAM signal, the receiver carrier phase R, the quadrature phase error δ of the QAM signal carriers and the quadrature phase error ε of the receiver carriers, characterized in that circuit means (PWK) are present which record the samples a _s and b _s of the signal components converted into the baseband a _s (t) and b _s (t) with four carrier functions v₁ = cos (ψ - ρ - σ) v₂ = cos (ψ + ρ + σ) v₃ = sin (ψ - ρ + σ) v₄ = sin (ψ + ρ - σ) link in such a way that an in-phase signal A _s = a _s v₁ - b _s v₃ and a quadrature signal B _s = a _s v₄ + b _s v₂ arise, and that control circuits (D 1 , D 2 , D 3 , LF 1 , LF 2 , LF 3 ) the parameters ψ, ρ and σ of the four carrier functions v₁. . . Set v₄ so that ψ = Δωt + Φ - R
ρ = δ
σ = εwith which neither the in-phase signal A _s nor the quadrature signal B _s depends on both message signals a and b, but rather the in-phase signal A _s clearly represents the message signal a and the quadrature signal B _s clearly represents the message signal b. 2. Circuit arrangement according to claim 1, characterized characterized in that a first discriminator (D₁) is present which is an estimate of the sum Δωt + Φ-R from the Carrier frequency difference Δω and the carrier phase difference Φ-R provides that a second discriminator (D₂) is present, the an estimate of the filing between the Quadrature phase error δ of the QAM signal carrier and the parameter delivers, and that a third discriminator (D₃) is present,  which is an estimate of the filing between the Quadrature phase error ε of the receiver carrier and the parameter σ delivers. 3. A circuit arrangement according to claim 2, characterized in that the first discriminator (D₁) the in-phase signal A _s , the quadrature signal B _s and the most likely transmitted message signals derived from these signals by threshold decision (E 1 , E 2 ) and linked to an output signal d₁ = â (B _s -) - (A _s - â) and that a downstream control filter (LF₁) filters out from this output signal d₁ the quantity Δωt + Φ-R determined by the carrier frequency difference and the carrier phase difference Φ-R and readjustment of the parameter Ψ to this variable Δωt + Φ-R causes. 4. Circuit arrangement according to claim 2, characterized in that the second discriminator (D₂) the in-phase signal A _s , the quadrature signal B _s and derived from these signals by threshold decision (E 1 , E 2 ) most likely transmitted message signals â and linked to an output signal d₂ = (A _s - â) + â (B _s -), and that a downstream control filter (LF₂) filters out the difference δ-ρ from this quadrature phase error δ of the QAM signal and the parameter from this output signal d₂ and causes readjustment of the parameter ρ to the quadrature phase error δ. 5. Circuit arrangement according to claim 3, characterized in that the third discriminator (D₃) the four carrier functions v₁. . . v₄, the sample values of the signal components a _s , b _s converted into the baseband and the most likely transmitted message signals â derived from the in-phase signal A _s and the quadrature signal B _s by threshold value decision (E 1 , E 2 ) and to one Output signal d₃ = (a _s v₂ + b _s v₄) - â (a _s v _s - b _s v₁) linked, and that a downstream control filter (LF₃) from this output signal the difference ε-σ from the quadrature phase error ε of the receiver carrier and filter out the parameter σ and readjust the parameter σ to the quadrature phase error ε.