JP4621046B2 - Frequency error detection device and reception device - Google Patents

Description

  The present invention relates to a frequency error detection device that detects a frequency error of a modulation signal.

  A receiving apparatus incorporated in a wireless communication device adopting a digital modulation system includes a digital demodulation unit that demodulates a digital quadrature modulation signal (hereinafter simply referred to as a quadrature modulation signal). The orthogonal modulation signal is obtained by associating a symbol code defined as an information unit by a digital code with a phase or phase change or amplitude of a carrier wave. When the quadrature modulation signal is represented by a complex vector, the component value of the complex vector changes while staying at a predetermined coordinate value for a predetermined time according to the period for associating the symbol code, that is, the symbol period.

  In general, a receiving apparatus extracts an in-phase component signal (hereinafter referred to as I signal) and a quadrature component signal (hereinafter referred to as Q signal) from a quadrature modulation signal by a quadrature detector. When the I signal value and Q signal value are plotted on the symbol plane with the I signal value on the horizontal axis and the Q signal value on the vertical axis, the locus with respect to time changes is drawn. The complex vector to be represented is similar to the locus drawn on the complex plane. That is, it can be said that the vector locus drawn on the symbol plane with reference to the origin is a mapping of the locus drawn by the complex vector of the orthogonal modulation signal. For this reason, in this specification, the deflection angle of the vector locus drawn by the I signal and the Q signal on the symbol plane is referred to as a phase on the symbol plane. The locus drawn by the I signal and the Q signal is referred to as a coordinate point that can be taken on the symbol plane for each symbol period, and the declination of a vector representing the symbol point on the symbol plane is referred to as a symbol point phase. And

  The digital demodulator reproduces the symbol code based on the phase of the symbol point represented by the I signal and Q signal, then performs code detection such as synchronous detection and delay detection, error correction, etc., and the symbol code obtained for each symbol period A digital code is obtained from the digital signal, and a digital signal in which the digital code is arranged in time series is output.

  The frequency of the quadrature modulation signal includes a frequency error caused by the fact that the frequency accuracy of the transmission source device of the quadrature modulation signal is not sufficient and the nonlinearity of the propagation path of the electromagnetic wave. This frequency error appears as a phase error of the symbol point and causes a decoding error of the digital code. Therefore, if the frequency error is large, the error rate of the extracted digital signal tends to increase. Therefore, in order for the digital demodulation unit to perform processing after code detection, it is preferable to reduce the included frequency error as much as possible. It can be said.

  The error rate is an index indicating the goodness of the reception state of a wireless communication device such as a mobile phone terminal, and reception performance such as reception sensitivity and interference wave resistance is generally defined by an error rate. For example, the reception sensitivity is defined based on the received signal power that can guarantee that the error rate is below a predetermined value, and the interference immunity is the error rate when a certain level of received signal is received. Is defined as the interference power that can be guaranteed to be below a predetermined value. Therefore, it can be said that the reception performance can be improved by reducing the frequency error and suppressing the increasing tendency of the error rate.

  Therefore, a frequency error signal generation unit and a frequency control unit are provided in front of the code detection unit that performs code detection such as synchronous detection and delay detection, and the frequency error signal generation unit generates a frequency error signal using the I signal and the Q signal. In general, the frequency controller compensates for the phase error of the symbol points represented by the I and Q signals. Here, the phase error of the symbol point is caused by the frequency error. The frequency error signal is a signal having information on the frequency error.

  FIG. 6 shows a partial configuration of the digital demodulator. Here, a case will be considered in which an MSK modulation method (Minimum Shift Keying) in which the phase transitions at ± π / 2 is adopted as the modulation method.

  The frequency error signal generator 42 calculates the phase transition Δθ between adjacent symbol points based on the input I signal and Q signal, and calculates the frequency error Δf according to the equation (1).

(Equation 1)
Δf = (2Δθ + π) / (4πT) (1)

Here, T is a period when a symbol point is associated with a phase change of the quadrature modulation signal, that is, a symbol period. The reason why the frequency error Δf can be calculated by the equation (1) will be described later.

  The frequency error signal generation unit 42 generates a frequency error signal having information on the frequency error Δf and inputs the frequency error signal to the frequency control unit 20. The frequency control unit 20 compensates the phase error of the symbol point represented by the separately input I signal and Q signal based on the frequency error signal input from the frequency error signal generation unit 42, and the phase error is compensated. The I signal and the Q signal are input to the code detector 22. The code detection unit 22 extracts and outputs a digital signal from the I signal and the Q signal input from the frequency control unit 20.

  According to such a configuration, since the phase error of the symbol point due to the frequency error is compensated for in the frequency control unit 20, an increasing tendency of the error rate can be suppressed and reception performance can be improved.

Next, the reason why the frequency error Δf can be calculated by the equation (1) will be described. Δθ is expressed as in equation (2) by the true value Δθ _m of the phase transition and the phase error Δθ _f caused by the frequency error.

(Equation 2)
Δθ = Δθ _m + Δθ _f (2)

The term relating to the frequency error on the right side of the equation (2) is only Δθ _f . Therefore, if derive the relationship between the Erase and [Delta] [theta] and [Delta] [theta] _f the [Delta] [theta] _m, it is possible to calculate the [Delta] [theta] _f using the [Delta] [theta], it is possible to calculate the further frequency error from [Delta] [theta] _f. Under such a policy, if both sides of equation (2) are doubled, equation (3) is obtained.

(Equation 3)
2Δθ = 2Δθ _m + 2Δθ _f = 2Δθ _f + π or 2Δθ _f −π (3)

Here, in the calculation from the center expression of the expression (3) to the right side, Δθ _m of the expression at the center of the expression (3) takes either a value of π / 2 or −π / 2. Was used. Further, when π is added to both sides of the formula (3), the formula (4) is obtained.

(Equation 4)
2Δθ + π = 2Δθ _f (4)

Here, in the process of obtaining the equation (4), 0 and 2π are degenerated to the same point on the symbol plane, and it is used that 2π can be replaced with 0.

(4) reacting a solved for [Delta] [theta] _f, between the [Delta] [theta] _f and the frequency error Δf be used that there is a relationship as shown in the following equation (5), (1) is obtained.

(Equation 5)
Δf = Δθ _f / (2πT) (5)

  The following document includes means for detecting a frequency error based on two symbol sets arranged at some distance in the input signal and controlling the frequency of the input signal based on the frequency error. A wireless communication device is disclosed.

JP 2004-140790 A

  The range of the frequency error that can be detected by the frequency error signal generator 42 shown in FIG. 6 is | Δf | <1 / (4T). That is, it is possible to detect only up to a frequency shift of one quarter of the reciprocal of the symbol period. The reason is as follows.

Equation (4) shows that there is a relationship of 2Δθ + π = 2Δθ _f between Δθ and Δθ _f , which doubles the phase transition Δθ between adjacent symbol points and adds π to that value. This means that a value obtained by doubling the phase error caused by the frequency error is uniquely determined. However, when 2Δθ _f is divided by 2, a region where −π <2Δθ _f <0 is a region where −π / 2 <Δθ _f <0, and a region where 0 <2Δθ _f <π is 0 <Δθ _f < Note that the range of possible values is halved, each mapped to a π / 2 region. That is, the range of values that Δθ _f can take is half of the range of values that 2Δθ _f can take, and the following equation (6) is established.

(Equation 6)
−π / 2 <Δθ _f <π / 2 (6)

By applying the equation (5) to this, the range of the frequency error that can be detected is obtained as the following equation (7).

(Equation 7)
-1 / (4T) <Δf <1 / (4T) (7)

Therefore, when the frequency error is calculated according to the equation (1) and is compensated, the frequency error that can be compensated is a frequency that is a quarter of the reciprocal of the symbol period. The range of the frequency error expressed by equation (7) is a unique value determined by the modulation method, and is based on the fact that the true value Δθ _m of the phase transition takes either π / 2 or −π / 2. Is.

  As described above, it is possible to improve the reception performance by suppressing the phase error of the symbol points represented by the I signal and the Q signal due to the frequency error, and suppressing the increasing tendency of the error rate. In recent years, communication frequency allocation is in an overcrowded state, and sufficient tolerance is required against reception interference due to intermodulation, cross modulation, and the like. In addition, with the spread of digital equipment and the like, it is necessary to increase the tolerance of digital signals to harmonic noise. In view of such a situation, it can be said that a digital demodulator used in a wireless communication device such as a mobile phone terminal is not sufficient to detect and compensate for the frequency error of the degree expressed by equation (7). .

  The present invention has been made for such a problem, and provides a frequency error detection device capable of detecting a frequency error larger than a frequency error determined based on a true value of a phase transition determined by a modulation method. provide.

The present invention is a frequency error detection device for detecting a frequency error of a modulation signal modulated by a code string, wherein an I signal and a Q signal obtained by quadrature detection with respect to the modulation signal are converted into the I signal and the Q signal. A sampling unit that samples at a time interval shorter than the symbol period, and a phase transition of symbol points of the I signal and the Q signal sampled by the sampling unit, a cumulative addition value over a period corresponding to the period of the code string is obtained, A frequency error generation unit that obtains the frequency error based on the cumulative addition value, and the code string has a predetermined period of the cumulative addition value of the phase transition of the symbol point associated with the code string. The frequency error generation unit is sampled by the sampling unit. Cumulatively adding the phase transition between sample points of the signal and Q signal, and wherein the.

In the frequency error detection apparatus according to the present invention, a control code sequence for extracting an I signal and a Q signal in which a phase transition of a symbol point is associated with the code sequence from the I signal and the Q signal sampled by the sampling unit It is preferable that the frequency error generation unit obtains the cumulative addition value for the I signal and the Q signal extracted by the control code string extraction unit .

The receiving device according to the present invention compensates for the phase error of the symbol points of the I signal and the Q signal sampled by the sampling unit based on the frequency error obtained by the frequency error detecting unit and the frequency error generating unit. And a frequency control unit.

  According to the present invention, it is possible to detect a frequency error larger than the frequency error determined based on the true value of the phase transition of the modulation signal. As a result, a digital demodulator capable of compensating for a frequency error larger than the conventional one can be realized, so that a receiving apparatus with improved reception performance can be realized.

  FIG. 1 shows a configuration of a receiving apparatus according to an embodiment of the present invention. This receiving apparatus is incorporated as a receiving means of a wireless communication device such as a mobile phone terminal. The present invention is applicable to a receiving apparatus that receives a signal subjected to general FSK modulation (Frequency Shift Keying) or PSK modulation (Phase Shift Keying), but here, it is a special aspect of the FSK modulation system. Consider a case in which a quadrature modulated signal modulated by a certain MSK modulation method is received.

  The quadrature modulation signal input from the reception signal input terminal 10 is input to the radio circuit unit 12. The radio circuit unit 12 amplifies the input quadrature modulation signal and performs frequency conversion, and then performs quadrature detection to extract an I signal and a Q signal and input them to the A / D conversion unit 14.

  The A / D converter 14 samples the unit symbol period of the input I signal and Q signal n times (n is a natural number) and converts it into a digital signal, and converts the I signal and Q signal as a digital signal into a control code string The data is input to the extraction unit 16 and the frequency control unit 20.

  The control code string extraction unit 16 extracts a signal having control code string information from the input I signal and Q signal. Here, the relationship between the signal input to the control code string extraction unit 16 and the control code string will be described. The signal input to the control code string extraction unit 16 is obtained by reflecting the codes constituting the frame code string 30 as shown in FIG. 2 in the amplitude of the I signal and the amplitude of the Q signal for each symbol code. One frame code sequence 30 includes a plurality of slot code sequences 32, and one slot code sequence 32 further includes a control code sequence 34a and an information code sequence 34b. At the rear end of each slot code string 32, a guard period 34c indicating the end of the slot code string 32 is provided. FIG. 2 shows a case where one frame code sequence 30 is composed of k (k is a natural number of 2 or more) slot code sequences 32. In a general wireless communication system, a plurality of frame code sequences 30 are connected to form a larger code sequence. Here, for simplicity of explanation, a frame code sequence 30, a slot code sequence 32, and a control code sequence 34a are used. , The relationship between the information code string 34b and the guard period 34c. Thus, by dividing the frame code sequence 30 into a plurality of slot code sequences 32, time domain division multiplex communication becomes possible. One slot code string 32 is assigned to one wireless communication device, and each wireless communication device performs communication processing only on the slot code string 32 assigned to itself. The control code string 34a includes control information for the wireless communication device to determine its own operation timing, and the information code string 34b includes information related to a communication target.

  The approximate timing at which the wireless communication device acquires its own slot code sequence 32 is acquired by detecting the rising edge of a burst, which is a collection of slot code sequences 32, and the receiving apparatus acquires its own slot code sequence 32. It is not necessary to recognize in advance the timing to perform.

  Detection of the rising edge of the burst is performed based on detection of the power value. As shown in FIG. 2, most of the guard period 34c exhibits a power value that is about the thermal noise, and the value is smaller than the power value of the burst. Therefore, the control code string extraction unit 16 takes the ratio of both power values at the boundary between the guard period 34c and the information code string 34b having a small power value, and when the ratio exceeds a predetermined threshold, It is considered. By detecting the rising edge of the burst in this way, it is possible to grasp a substantially accurate timing for acquiring the slot code string 32 and to detect the control code string 34a.

  The frame code sequence 30 and the slot code sequence 32 are obtained by arranging codes represented by “0” or “1” at symbol period intervals in symbol code units. Here, a single-digit code is assigned to one symbol code. Therefore, the time from one code to the next code is one symbol period, and here, since it is sampled n times by the A / D converter 14, the time from one code to the next code is reached. In the meantime, n sample values are output from the A / D converter 14. In orthogonal modulation, if the code is “0”, the symbol point on the symbol plane is rotated by π / 2, and if the code is “1”, the symbol point on the symbol plane is rotated by −π / 2. Conversely, from the viewpoint of quadrature demodulation, when the symbol point is rotated by π / 2, the code “0” is acquired, and when the symbol point is rotated by −π / 2, the code “1” is acquired. Will be.

  The control code string extraction unit 16 grasps the timing at which the control code string 34a appears in the assigned slot code string 32, and extracts the I signal and the Q signal corresponding to the control code string 34a. The code pattern constituting the control code string 34a and the correspondence relationship between the code constituting the control code string 34a and the phase transition Δφ of the symbol point exhibited by the I signal and the Q signal are defined in advance in the system. FIG. 3A shows an example of a code pattern constituting the control code string 34a, and FIG. 3B shows a code constituting the control code string 34a and the phase of the symbol point represented by the I ′ signal and the Q ′ signal. The correspondence relationship with the transition Δφ is shown. Here, Δφ = −π / 2 is associated with “1”, and Δφ = π / 2 is associated with “0”.

  FIG. 3C shows the accumulated addition value ΣΔφ of the phase transition of the symbol points and the normalized amplitudes of the I ′ signal and the Q ′ signal in association with each other under such association. For example, when the cumulative addition value ΣΔφ of the phase transition of the symbol point changes like symbol number 0 to symbol number 3 in FIG. 3C, it is indicated by symbol number 0 to symbol number 3 in FIG. The symbol point moves like a point. Here, the prime symbol of the I ′ signal and the Q ′ signal means a signal corresponding to the control code string 34a, and is distinguished from the I signal and the Q signal input from the A / D conversion unit 14. It is attached.

  The code pattern constituting the control code string 34a is defined by paying attention to the regularity of the cumulative addition value ΣΔφ of the phase transitions of the symbol points exhibited by the I ′ signal and the Q ′ signal. FIG. 4 is a graph showing the cumulative addition value ΣΔφ of the phase transition of symbol points under the condition that the code pattern constituting the control code string 34a is appropriately defined in advance. As is apparent from FIG. 4, the cumulative addition value ΣΔφ of the symbol point phase transition has a period of 16 symbols. This means that if the phase angle of a symbol point at a certain time is set to zero and the phase transition of the symbol point is cumulatively added over 16 symbols therefrom, the result becomes zero. Such periodicity can be realized by defining a code constituting the control code string 34a so that the cumulative addition value of the phase transitions of the symbol points associated with the code constituting itself has periodicity. . In the present embodiment, the frequency error is calculated using such regularity defined in the control code string 34a.

  In the above description, it has been mentioned that the timing of obtaining the slot code string 32 can be grasped by detecting the rising edge of the burst, and the control code string 34a can be detected. Here, since the control code string 34a has a feature that the accumulated addition value between 16 symbol points becomes the same at an arbitrary timing, even if there is a deviation in the timing at which the control code string 34a is detected, the frequency error Can be detected.

The control code string extraction unit 16 extracts the I ′ signal and the Q ′ signal from the input I signal and Q signal, and inputs them to the frequency error signal generation unit 18. The frequency error signal generation unit 18 uses one of the symbol points of the I ′ signal and the Q ′ signal as a reference, and calculates a cumulative addition value Φ of the phase transition of the symbol points over 16 symbols therefrom. If the phase transition between each sample point is Δφ _i (i is an integer from 1 to 16n, n is the number of samples between symbol periods), the cumulative addition value Φ of the phase transition at the symbol point Is calculated as in the following equation (8).

(Equation 8)
Φ = ΣΔφ _i (8)

Here, Σ means that addition and summation is performed for i = 1 to i = 16n, and is the same in the following equations. The frequency error signal generation unit 18 further calculates a frequency error based on the following equation (9).

(Equation 9)
Δf = Φ / (32πT) (9)

  The frequency error signal generation unit 18 inputs a frequency error signal including frequency error information calculated according to the equation (9) to the frequency control unit 20. The frequency control unit 20 compensates the phase error of the symbol point represented by the separately input I signal and Q signal based on the frequency error signal input from the frequency error signal generation unit 18, and the phase error is compensated. The I signal and the Q signal are input to the code detector 22. The code detection unit 22 extracts and outputs a digital signal from the I signal and the Q signal input from the frequency control unit 20.

  As described above, in this embodiment, the I ′ signal and the Q ′ signal reflecting the code of the control code string 34a are extracted from the I signal and the Q signal sampled n times, and the I ′ signal and the Q ′ signal are extracted. The frequency error is calculated based on the cumulative addition value of the phase transitions of the symbol points represented by. By calculating the frequency error in this way, it is possible to detect a frequency error larger than the frequency error determined based on the phase transition determined by the modulation method. The reason will be described below.

The phase transition Δφ _i between each sample point represented by the I ′ signal and the Q ′ signal is divided into a true value Δφ _mi and a phase error Δφ _f caused by a frequency error, and the following equation (10) is expressed. It becomes like the formula.

(Equation 10)
Φ = ΣΔφ _i = Σ (Δφ _mi + Δφ _f ) = Σ (Δφ _mi ) + Σ (Δφ _f ) (10)

Here, the first term on the right side of equation (10) becomes zero due to the nature of the control code string 34a. Therefore, the cumulative addition value Φ of the phase transition of the symbol point is only the contribution of the phase error Δφ _f caused by the frequency error, and Φ is expressed as the following equation (11).

(Equation 11)
Φ = Σ (Δφ _f ) = 16 nΔφ _f (11)

The above equation (9) is obtained by solving the equation (11) with respect to Δφ _f and further using the relationship of the following equation (12).

(Equation 12)
Δf = nΔφ _f / (2πT) (12)

Next, the range of frequency errors that can be detected according to the equation (9) will be examined. Since Δφ _mi can take only one value of π / (2n) or −π / (2n), Δφ _i is expressed as in equation (13).

(Equation 13)
Δφ _i = Δφ _mi + Δφ _f
= Δφ _f + π / (2n) or Δφ _f −π / (2n) (13)

Here, Δφ _f can take a value in an arbitrary range, while Δφ _i can only have a value in a range of −π <Δφ _i <π in detecting a frequency error. This is because Δφ _i is a value to be actually measured on the symbol plane. That is, theoretically, the value of Δφ _i can be made to correspond to an arbitrary Δf in the range of −jπ <Δφ _i <jπ (j is a natural number), but since there is one symbol plane, j can not determine the value, the range defined in the [Delta] [phi _i is regarded to be a j = 1 is because you are -π <Δφ _i <π. Therefore, since the value of Δφ _{f in} the range where Δφ _i is equal to or larger than π cannot be detected, when Δφ _i = Δφ _f + π / (2n) in the equation (13) is satisfied, The range determined by the equation (14) is a range in which Δφ _f cannot be detected.

(Equation 14)
Δφ _f ≧ π−π / (2n) (14)

In addition, since it is impossible to detect the value of Δφ _{f in} the range where Δφ _i is equal to or less than −π, when Δφ _i = Δφ _f −π / (2n) in the equation (13) is satisfied. The range determined by the following equation (15) is a range in which Δφ _f cannot be detected.

(Equation 15)
Δφ _f ≦ −π + π / (2n) (15)

In order for any of the right sides of Equation (13) to hold, the range in which Δφ _f can be detected is the range excluding the union of Equations (14) and (15). 16) or (17).

(Equation 16)
−π + π / (2n) <Δφ _f <π−π / (2n) (16)

(Equation 17)
_{-Π + | Δφ mi | <Δφ} f <π- | Δφ mi | (17)

This equation means that the range in which the phase error due to the frequency error can be detected is eroded from the symbol plane by twice the phase transition between unit sample points. From another point of view, the boundary point of the range where the phase angle can be expressed by a value within the range of one rotation on the symbol plane is ± π, but the phase transition from that point to the unit sample point It can be said that there is an uncertainty that the range that is just before and after is not uniquely determined by the value within the range of one rotation on the symbol plane. This uncertainty is, for example, when Δφ _f takes a value of −π + | Δφ _mi |, and cannot be distinguished from π + | Δφ _mi |, or Δφ _f is a value of −π− | Δφ _mi |. This means that it cannot be distinguished from π− | Δφ _mi |.

Here, because of the above-described equation (12) relationship between the frequency error Δf and [Delta] [phi _f, the detectable range of the frequency error is be expressed by the following equation (18).

(Equation 18)
− (2n−1) / (4T) <Δf <(2n−1) / (4T) (18)

According to the equation (18), it can be seen that a larger frequency error can be detected as the number of samples n is increased. This is because the range in which the phase error Δφ _f due to the frequency error can be detected becomes wider as the absolute value | Δφ _mi | of the phase transition between the sample points is smaller, and | Δφ _mi | This is because the smaller the number n, the smaller.

  Comparing equation (18) with equation (7) representing the range of frequency errors that can be detected by the conventional frequency error signal generator 42, (2n-1) / (4T) / (1 / 4T) = It can be said that the present embodiment can detect a frequency error that is 2n-1 times larger than 2n-1.

  Here, the code string shown in FIG. 3A is taken as an example of the control code string 34a. This code string is obtained by performing an exclusive OR between adjacent codes in the 26-bit original control code string 36 (training sequence) defined in GSM (Global System for Mobile Communications) shown in FIG. Are arranged. FIG. 5B shows a control code string 34a obtained by such an arrangement. The present invention is not limited to the embodiment to which the control code string 34a shown in FIG. 3 (a) or FIG. 5 (b) is applied, and the cumulative addition value of the phase transition of the symbol point associated with the code is periodic. Any code string can be applied as long as it has regularity.

Here, when deriving the control code string 34a shown in FIG. 5B, the reason for the exclusive OR operation between adjacent codes in the training sequence is associated with “code change”. This is because the training sequence has a regularity that the accumulated addition value of the phase transitions of the symbol points has periodicity. Therefore, it can be said that the code sequence arranged by performing exclusive OR between adjacent codes of the training sequence has a periodicity in the accumulated addition value of the phase transitions of the symbol points associated with the “code”. Due to such regularity, the phase angle of a symbol point at a certain point in time is set to zero, and the phase transition of the symbol point is cumulatively added from there, and when that cumulative addition is made for one period of the cumulative addition value of the phase transition, the value is obtained. Can be made zero. That is, as shown in the equation (10), a value based on the phase error Δφ _f caused by the frequency error can be extracted from a value based on the phase transition Δφ _i between the respective sample points.

When the control code string extraction unit 16 extracts the I ′ signal and the Q ′ signal, it is preferable that faster processing is possible. Therefore, a code string itself in which the cumulative addition value of the phase transition of the symbol point associated with the code has periodicity is arranged in the control code string 34a constituting the slot code string 32 of FIG. If the code string has such a property, it is possible to give periodicity to the cumulative addition value of the phase transition, and based on the phase transition Δφ _i between the respective sample points as shown in the equation (10). It is possible to separate the value based on the phase error Δφ _f caused by the frequency error from the value. Such code arrangement design of the slot code string 32 is preferably performed at the stage of designing the radio communication system.

It is a figure which shows the structure of the receiver which concerns on embodiment of this invention. It is a figure which shows a frame code sequence, a slot code sequence, a control code sequence, an information code sequence, and a guard period. It is a figure which shows a mode that a control code sequence, the phase transition of a symbol point, the cumulative addition value of a phase transition, the normalized amplitude value of I 'signal and Q' signal, and a symbol point move. It is a figure which shows the cumulative addition value of the phase transition of a symbol point. It is a figure which shows the relationship between the original control code sequence (the training sequence defined in GSM) and the control code sequence arranged based on it. It is a figure which shows the structure of a part of digital demodulator.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 Received signal input terminal, 12 Radio circuit part, 14 A / D conversion part, 16 Control code sequence extraction part, 18, 42 Frequency error signal generation part, 20 Frequency control part, 22 Code detection part, 30 Frame code string, 32 Slot code string, 34a control code string, 34b information code string, 34c guard period, 36 original control code string.

Claims (3)

A frequency error detection device for detecting a frequency error of a modulation signal modulated by a code string,
A sampling unit that samples the I and Q signals obtained by quadrature detection on the modulated signal at a time interval shorter than the symbol period of the I and Q signals;
For the phase transitions of the symbol points of the I signal and Q signal sampled by the sampling unit, a cumulative addition value over a period corresponding to the period of the code string is obtained, and the frequency error is obtained based on the cumulative addition value An error generator;
With
The code string has regularity that the cumulative addition value of the phase transitions of the symbol points associated with the code string takes the same value for each predetermined period,
The frequency error generation unit cumulatively adds phase transitions between sample points of the I signal and the Q signal sampled by the sampling unit;
A frequency error detection apparatus characterized by the above. The frequency error detection device according to claim 1,
A control code string extraction unit that extracts an I signal and a Q signal in which a phase transition of a symbol point is associated with the code string from the I signal and the Q signal sampled by the sampling unit;
With
The frequency error detection device, wherein the frequency error generation unit obtains the cumulative addition value for the I signal and the Q signal extracted by the control code string extraction unit . The frequency error detection device according to claim 1 or 2 ,
A frequency controller that compensates for a phase error of symbol points of the I signal and the Q signal sampled by the sampling unit, based on the frequency error obtained by the frequency error generating unit;
Receiving device, characterized in that it comprises a.

Images