Receiving apparatus

BACKGROUND OF THE INVENTION

1. Technical Field

The present invention relates to a receiving apparatus for receiving a digitally modulated signal and particularly to a correcting technology for a received frequency.

2. Description of the Related art

A receiving apparatus for receiving a digitally modulated signal shown in

FIG. 6. A

digitally modulated signal is received by an antenna

11

, and after the received signal is amplified by a high frequency amplifier

12

, it is mixed with an output of a frequency synthesizer

14

at a frequency converter

13

and requency-converted to a predetermined intermediate frequency signal. After the converted intermediate frequency signal is amplified by an intermediate frequency amplifier

15

and passed through a band limiting filter

16

, it is supplied to an orthogonal detector

17

for orthogonal detection and becomes a base band signal. The orthogonal detection is carried out in synchronization with a frequency of a carrier wave supplied from a frequency synthesizer

18

.

The base band signal converted by the orthogonal detector

17

is supplied to a demodulator

19

and the demodulated data is error-corrected by an error-correcting circuit

20

and supplied to a voice decoder

21

to be subjected to a voice decoding process, the decoded voice signal is supplied to a voice output circuit (not shown) at a later stage.

Here, the frequency which is frequency-converted by the frequency converter

13

and the frequency which is detected by the orthogonal detector

17

is determined by the frequency supplied from the frequency convertor

13

, but frequency correction needs to be carried out for the frequencies supplied from the frequency synthesizers

14

,

18

to synchronize with the frequency of the received signal.

FIG. 2

is a block diagram showing an arrangement for carrying out the conventional frequency correction. The frequency correction is carried out based on a received signal symbol demodulated by the demodulator

19

. That is, a frequency offset of the received signal symbol demodulated by the demodulator

19

is detected by a frequency offset detector

31

and the value of the detected frequency offset is averaged by being integrated by a digital integrator

32

. The averaged value integrated by the integrator

32

is converted to an analog voltage by a digital/analog convertor

33

and the converted voltage is supplied to a control voltage input of a voltage controlled oscillator

34

to control the output frequency of the voltage controlled oscillator

34

. An output of the voltage controlled oscillator

34

is supplied to the frequency synthesizers

14

,

18

to control their output frequencies.

By the control in this circuit, the correction of the received frequency can be carried out. First of all, in regard to the digital modulated-signal to be received, if a digital information signal of an in-phase component is assumed to be I(t) while a digital information signal of an orthogonal component is assumed to be Q(t), a signal X(t) of the information component is defined by a complex number such as X(t)=I(t)+jQ(t). The digitally modulated signal can be expressed in the following.

s

(

t

)=

Re

(

x

(

t

)·

e

j2πfct

) [Equation 1]

Here, fc denotes a carrier wave frequency and the carrier wave is expressed by a complex sinusoidal wave c(t) =e

j2πfct

.

The digitally modulated received signal is, by being subjected to an orthogonal demodulation on the receiving apparatus side, extracted as a signal of the information component as a base band signal. The orthogonal demodulation will be explained hereinafter. The digitally modulated received signal is demodulated in the following way by being multiplied by a received carrier wave with a reverse phase rotation relative to the carrier wave.

\begin{matrix} \begin{matrix} F (s (t) \cdot ⅇ^{π f - 2 π fct}) = \frac{1}{2} F [(x (t) \cdot ⅇ^{j2π fct} + \\ x * (t) \cdot ⅇ^{- j2π fct}) \cdot ⅇ^{- j2π fct}] \\ = \frac{1}{2} F [x (t) + x * (t) \cdot ⅇ^{- j 2 π fct} \\ = \frac{1}{2} (X (f) + x * (- f - 2 fc)) \end{matrix} & [Equation  2] \end{matrix}

Here, X(f) denotes a frequency of x(t) and x*(t) and X*(f) express complex conjugates of x(t) and X(f). By taking out only a low-band signal from the demodulated received signal by a low-pass filter, the base band signal can be obtained.

When there exists a frequency offset f between the carrier frequency used for the digital modulation and the frequency of the complex sinusoidal wave used for the orthogonal demodulation, the reception signal converted by the orthogonal demodulation is expressed in the following equation and the base band signal obtained after the low-band signal is extracted by the low-pass filter is, as shown in

FIG. 3

, observed as time phase rotation of an angular velocity ω=2f on an IQ plane.

\begin{matrix} \begin{matrix} F (s (t) \cdot ⅇ^{- j 2 π (fc + Δ f) t}) = \frac{1}{2} F [(x (t) \cdot ⅇ^{j 2 π fct} + x * (t) \cdot \\ ⅇ^{- j2π (fc + Δ f) t)} \cdot ⅇ^{- j2π (fc + Δ f) t}] \\ = \frac{1}{2} F [(x (t) \cdot ⅇ^{- j2π Δ ft} + x * \\ (t) ⅇ^{- j2π 2 (fc + Δ f) t}) \\ = \frac{1}{2} (X (f - Δ f) X * (- f - 2 (f_{c} + Δ f))) \end{matrix} & [Equation  3] \end{matrix}

Since this causes a serious influence on a code judgement error of the received signal, it is necessary to measure transmission and received frequency offsets as well as correcting a received carrier wave frequency on the receiving apparatus side.

Next, a process for estimating the frequency offset will be explained with reference to the system shown in FIG.

2

. When a received signal symbol r

t

outputted by the demodulator

19

of

FIG. 1 and a

transmission signal symbol corresponding to a previously demodulated received signal symbol r

t-n

are identical, by carrying out an arctan operation to a value obtained by complex conjugate multiplication between r

t

and r

t-n

, a phase rotation angle Δθ can be detected. By obtaining the phase rotation angle Δθ which rotates on the IQ plane at some constant interval Δt, the frequency offset Δf can be obtained by the following.

Δ f = \frac{Δ θ}{2 π \cdot Δ t}

In the conventional process for estimating a frequency offset, the following problems arise. When a phase rotation angle between symbols is detected, an arctan operation is necessary to obtain a value which is obtained by complex conjugate multiplication of the two symbols, but since the amount of calculation of the arctan operation is comparatively large, when a real time process is needed, it can happen that the arctan operation process does not end within a required time. In order to avoid this, the following approximation has been carried out. Here, amplitude r

tk

is denoted as A

t

, a phase angle as θ

t

, an in-phase signal as it and an orthogonal signal as q

t

.

\begin{matrix} \begin{matrix} r_{t}^{″} r_{t - n}^{*} = A_{t} \cdot ⅇ^{j θ^{t}} \cdot A_{t - n} \cdot ⅇ^{- j θ^{t - n}} \\ = A_{t} \cdot A_{t - n} \cdot ⅇ^{j (θ^{t} - θ^{t - n})} \\ = A_{t} \cdot A_{t - n} \cdot ⅇ^{jΔ θ} \\ = A_{t} \cdot A_{t - n} \cdot (\cos Δ θ + j s in Δ θ) \\ = (i_{t} + {jq}_{t}) \cdot (i_{t - n} - {jq}_{t - n}) \\ = (i_{t} \cdot i_{t - n} + q_{t} \cdot q_{t - n}) + j (q_{t} \cdot i_{t - n} - i_{t} \cdot q_{t - n}) \end{matrix} & [Equation  5] \end{matrix}

By utilizing the approximation that Δθ=sin Δθ is true when Δθ is near to zero, the phase rotation angle Δθ can be expressed in the following formula.

Δθ=

q

t

·i

t-n

−i

t

·q

t-n

[Equation 6]

When this approximation is used, however, the phase rotation angle of the frequency offset which can be detected is limited to within a range of ±π/2. When the phase rotation angle exceeds the range of ±π/2, the frequency offset is erroneously detected. Also, according to the equation 5, since a presumed value of the phase rotation angle changes depending on a strength of the received signal, there is a change in a presumed value of the frequency offset, thereby incurring variations in converging time and correction errors. Also, since the interval between the received signal symbols used for detecting the conventional frequency offset is constant, there have been problems that when the interval between the symbols is narrow, although a frequency pulling-in range becomes wide, the frequency correction error becomes large after the pulling-in, while when the interval between the symbols is wide, although the frequency correction error becomes small after the pulling-in, the frequency pulling-in range becomes narrow.

SUMMARY OF THE INVENTION

An object of the present invention is to make it possible to favorably carry out correction of the received frequency when receiving a digitally modulated signal.

The present invention comprises phase judging means for judging a phase of a received signal symbol by the predetermined number of quantization, phase difference detecting means for detecting a phase difference between phase-judged samples judged by the phase judging means and averaging means for averaging a detecting output of the phase difference detecting means, to thereby control an output frequency of a frequency synthesizer based on an output of the averaging means.

According to such an arrangement, the phase judging means digitally judges the phase by the predetermined number of quantization, and a difference between the judged phase value and a phase value immediately before is detected by the phase difference detecting means and the detected phase difference is predeterminately averaged by the averaging means and the output of the frequency synthesizer is controlled based on the averaged frequency error and a frequency subjected to a received process is corrected.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1

is a block diagram showing an arrangement of a known digital receiving apparatus;

FIG.

2

. is a block diagram of a known arrangement for correcting for a frequency offset in the receiver of

FIG. 1

;

FIG.

3

. is an explanatory diagram showing a phase rotation in a signal received by the receiver of

FIG. 1

;

FIG. 4

is a block diagram of a system for correcting for phase shifts in a digital receiver according to an embodiment of the present invention;

FIG. 5

is an explanatory diagram showing a concept of eight phase judgement for eight different phases according to an embodiment of the present invention;

FIG. 6

is a flowchart showing a process of varying a sample interval according to an embodiment of the present invention;

FIG. 7

is a flowchart showing a process of varying a sampling interval according to an embodiment of the present invention; and

FIG. 8

is an explanatory diagram showing a concept of phase judgement for four phases according to the present invention.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

In the following one embodiment of the present invention will be explained with reference to

FIG. 4

to FIG.

8

. In

FIG. 4

to

FIG. 8

, by attaching the same numerals to corresponding portions in diagrams of FIG.

1

and figures thereafter which are explained as the conventional example, detailed explanations thereof are omitted.

In the present embodiment, there is provided a frequency correcting circuit which is applied to a digital signal receiving apparatus which is shown in

FIG. 1

as the conventional example and the arrangement thereof is shown in FIG.

4

. That is, a received signal symbol demodulated by a demodulator

19

is sampled by a sample circuit

41

and the sampled received signal symbol is supplied to a phase judging unit

42

. In this case, sample timing by the sample circuit

41

is set based on a result of comparison by a comparator

45

. The setting process of the sample timing will be explained later.

In the phase judging unit

42

, the received signal symbol is judged by the predetermined number of quantization. Here, the judgement is carried out in terms of eight phases as the number of quantization. That is, as shown in

FIG. 2

, a range of 2π on a IQ plane with I as a in-phase component and Q as an orthogonal component is divided into eight phases at equal spaces (that is, quantization step is π/4) and 3 bits data, for example, “000”, “001”, “010”, “010”, “100”, “101”, “110” and “111” are allocated to each of the ranges.

For example, as shown in

FIG. 5

, when a symbol phase of some sample timing is θ1, let data be “000” at the phase judging unit

42

, and when a symbol phase of the next sample timing is θ2, let data be “001” at the phase judging unit

42

.

Then, data on the phase judged value judged by the phase judging unit

42

is supplied to a subtractor

43

and a difference Δf from a judged value of sample data immediately before is operated. For example, when data on the above-mentioned two sample timings are “000” and “001”, “001” or a difference between both the data is operated. Data on the difference is supplied to a digital integrator

44

for integration and an average value E[Δθ] of phase differences for a predetermined period of time is calculated. The average value E[Δθ] becomes a value proportional to a frequency offset Δf.

The data integrated by the digital integrator

44

is supplied to a digital/analog converter

33

to be converted to an analog voltage signal and the converted analog voltage signal is supplied to a controlling voltage input of a voltage controlled oscillator

34

to control an output frequency of the voltage controlled oscillator

34

. An output of the voltage control oscillator

34

is supplied to the frequency synthesizers

14

,

18

to control output frequencies of both the frequency synthesizers

14

,

18

and the frequency output offset Δf is corrected.

Also, the integrated data of the digital integrator

44

is supplied to the comparator

45

to be compared with a threshold value supplied from a threshold value generator

46

and, based on a result of the comparison, the sample interval in the sample circuit

41

is varied. To be concrete, it is arranged such that in an initial state when reception of a predetermined frequency is started, a comparatively narrow sample interval is set and as the value of frequency offset Δf becomes smaller, it is varied to a comparatively wide sample interval. Its concrete process will be described later.

Next, a process of the phase judging unit

42

for carrying out a phase judgement will be explained in detail. In a case when the phase judgement of eight phases is carried out by the phase judging unit

42

, a phase judging sample r

t

relative to a received signal symbol x

t

=(i

t

, q

t

) can be shown by an algorithm of the following formula. However, a symbol that “+” is encircled by “o” shows here carrying out of an exclusive logical add operation (a so-called E

x

OR operation) at every bit and figures are expressed in terms of hexadecimal notation.

r

t

=0;

iƒ

(

i

t

<0)

r

t

=r

t

⊕0

x

03;

iƒ

(

q

t

<0)

r

t

=r

t

⊕0

x

0ƒ;

iƒ

(|

i

t

|<|q

t

|)

r

t

=r

t

⊕0

x

01; [Equation

7

]

If the algorithm of the formula is to be shown in a flowchart, it should be what is shown in FIG.

6

. That is, first of all, let a phase judging sample be r

t

=0 (step

101

) and whether or not a value of i

t

is less than 0 is judged (step

102

). Here, when it is less than 0, a value obtained by carrying out an E

x

OR operation between the value of i

t

and a predetermined hexadecimal notation “0x03” is a value of i

t

(step

103

). Next, it is judged whether or not a value of q

t

is less than 0 (step

104

). Here, when the q

t

is less than 0, a value obtained by carrying out an E

x

OR operation between the value of q

t

and a predetermined hexadecimal notation “0x0f” is a value of q

t

(step

105

). Further, a comparison is made between an absolute value of i

t

and that of q

t

(step

106

) and when the absolute value of q

t

is larger, as a value of the phase judging sample r

t

, a value obtained by carrying out an E

x

OR operation between the value of rt and a predetermined hexadecimal notation “0x01” is the value of the phase judging sample r

t

(step

107

).

By carrying out the phase judgement in this manner, it is possible for the phase judging unit

42

to precisely quantize the phase of the received signal symbol to eight values.

Next, a process for varying the sample timing will be explained with reference to a flowchart in FIG.

7

. First of all, when reception of a signal by a predetermined channel is started, the sample circuit

41

sets an initial value previously prepared as a sample interval (step

201

). The initial value is set as a comparatively narrow sample interval. In this state, a phase judging process is carried out by the phase judging unit

42

(step

202

) and a phase rotation angle Δθ is calculated (step

203

) and in a comparator

45

, it is judged whether or not the value of the phase rotation angle Δθ judged from an output of an integrator

44

is a less value than a threshold value M set by a threshold generator

46

(step

204

). Here, when it is judged that the value is not less than the threshold value M, the process returns to the step

202

and carries out the phase judging process of the next sample timing.

Then, in a case when it is judged that the value of the phase rotation angle Δθ is less than the threshold value M, it can be judged that the value of the frequency offset Δf has become smaller, and the sample interval in the sample circuit

41

is widened (step

205

), then the process returning to step

202

.

According to the receiving apparatus of the embodiment explained so far, it is possible to precisely judge the frequency offset without respect to the received signal level as well as to carry out an accurate frequency correction. That is, by carrying out a judgement based on a small number of quantization (here, eight phases) at the phase judging unit

42

, it is possible to precisely judge the frequency offset, to simplify a circuit scale of the phase judging unit and the amount of operation processing and further, to carry out a phase judgement of a high precision without respect to the received signal level.

Also, according to the present embodiment, since a phase difference detecting output of the subtractor

43

as a phase difference detecting means is averaged by the integrator

44

, it is possible to reduce any error of the phase rotation angle due to a noise as well as to improve a frequency detecting precision. This is because of a principle that it is possible to improve a precision by increasing the number of average samples although the number of quantization is sparse. Also, in a case of the embodiment where the quantization by eight phases is carried out by the phase judging unit

42

, it becomes possible for the subtractor

43

to detect a frequency offset in a wide range of about −3π/4

˜

+3π/4, and a range capable of detecting the frequency offset becomes wider than the conventional one. That is, compared with a case of the example shown in

FIG. 2

as the conventional one wherein a frequency offset only in a range of about −π/2

˜

+π/2 can be detected, detection of frequency offset in a wider range becomes possible, thereby making it possible to cope with a case of a larger frequency offset.

Also, by varying the sample interval in response to a frequency offset then, it becomes possible to make a pulling-in range of the frequency offset as well as to minimize a frequency correction error after the pulling-in. Meanwhile, in the example shown in the flowchart in

FIG. 7

, the sample interval is made to vary in two steps of an initial state and a state when the frequency offset becomes smaller, but the sample interval may be varied to a number of smaller steps. Also, in a case when the frequency offset becomes larger due to an unknown factor in the midst of receiving a signal, the sample interval may be returned to a narrow state.

Meanwhile, in the above-mentioned embodiment, although the number of quantization at the time of carrying out the phase judgement by the phase judging unit is eight phases, another number of quantization may be used. For example, the number of quantization by the phase judging unit may be four phases. That is, as shown in

FIG. 8

, by dividing a range of 2π on an IQ plane to four phases at an equal space; I being an in-phase component of the received signal symbol and Q being an orthogonal component (that is, a quantum step is π/2), for example, 2 bit data “00”, “0.1.”, “10” and “11” are allocated to each of the ranges to carry out the phase judgement. A phase judging sample rt relative to a received signal symbol X

t

=(i

t

,q

t

) in this case can be shown by algorithm of the following formula. However, the symbol [|] here is a logically add operation at every bit (a so-called OR operation) and a symbol that “+” is encircled by “o” here shows carrying out of an exclusive logical add operation (a so-called E

x

OR operation) at every bit and figures are expressed in terms of hexadecimal notation.

r

t

=0 ;

if

(r

t

<0)r

t

=r

t

10

x

01;

in

(q

t

<0)r

t

=r

t

⊕0

x

03; [Equation 8]

In a case when the phase judgement is carried out by quantization of the four phases, it becomes possible to detect a frequency offset in ranges of about −π/2

˜

+π/2. The detecting ranges of the frequency offset is similar to the detecting ranges of the arrangement shown in FIG.

2

as the conventional example, but since the detecting precision comes to not depend on a received signal level, the detecting precision becomes higher than the conventional one and at the same time, by carrying out the control of the sample interval, there is provided an effect capable of minimizing a frequency error after the pulling-in.

Meanwhile, the example explained up to now is applied as the one for carrying out the control of the sample interval, but there may be an arrangement in which the comparator

45

and the threshold generator

45

are not provided by making the sample interval always constant at the sample circuit

41

.

According to the present invention, by digitally judging the phase in terms of a predetermined number of quantization by the phase judging means, detecting the difference between the judged phase value and the phase value immediately before, and carrying out the frequency correction based on an average value of the phase difference, the favorable detection of the frequency offset without respect to the received signal level can be carried out as well as the favorable frequency correction can be carried out based on the detected frequency offset value.

According to the present invention, by providing sampling interval setting means for varying the sampling interval to be phase-judged by the phase judging means based on an output of the averaging means, it is possible to widen the pulling-in range of the means offset as well as to minimize the correction error at the time of converging.

According to the present invention, by making the number of quantization of the phase judged by the phase judging means eight phases, it is possible to detect the frequency offset in the ranges of −3π/4

˜

+3π/4 of the phase rotation between the phase judged samplings is about as well as to carry out the favorable frequency correction.

According to the present invention, by making the number of quantization of the phase to be judged by the phase judging means four phases, it is possible to detect the frequency offset in the ranges of −π/2

˜

+π/2 of the phase rotation between the phase judged samplings is about as well as to carry out the favorable frequency correction.

Having described preferred embodiments of the present invention with reference to the accompanying drawings, it is to be understood that the present invention is not limited to the above-mentioned embodiments and that various changes and modifications can be effected therein by one skilled in the art without departing from the spirit or scope of the present invention as defined in the appended claims.

Receiving apparatus

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Abstract

Description

Claims

Priority Claims (1)

US Referenced Citations (1)