KR20050030807A - A method and a apparatus of frequency offset estimation for mobile phone - Google Patents

Abstract

A device and a method for estimating an offset frequency of a portable terminal are provided to easily detect and remove an error generation symbol by using a decision region, and to interpolate a phase value of an adjacent symbol by averaging the phase value, thereby minimizing a least square error. A de-rotation unit(100) processes a symbol signal into a training sequence symbol signal, and offsets a phase difference between QPSK modulation signals. A phase calculator(110) inputs the offset signals, and detects a phase value. A phase unwrapper(120) compensates for an error caused during a function calculation of the detected phase value. An error detector(130) analyzes a signal inputted from the unwrapper(120), and removes a symbol signal detected in other region except a decision region. An interpolator(140) obtains an average value of symbol signals adjacent to the removed symbol signal. An offset estimator(150) operates an offset frequency estimation value having the smallest least square error.

Description

Offset frequency estimating apparatus and method for mobile terminal {A METHOD AND A APPARATUS OF FREQUENCY OFFSET ESTIMATION FOR MOBILE PHONE}

The present invention relates to an automatic frequency control (AFC) of a receiver of a QPSK modulation type mobile terminal, and more particularly, to improve accuracy of automatic frequency control by eliminating the maximum symbol error and replacing the average value with an adjacent symbol. Offset frequency estimation apparatus and method.

The mobile communication system secures as many channels as possible by using limited radio frequency resources and allows multiple mobile terminals to occupy and communicate at the same time. Among the various methods developed to meet the above requirements, CODE DIVISION MULTIPLE ACCESS (CDMA) allows multiple radios to be used simultaneously.

In the above code division multiple access (CDMA) mobile communication system, the QPSK (QUADRATURE PHASE SHIFT KEYING) modulation method is used, and if the cell area is changed due to the characteristics of the mobile communication system, the allocated channel is changed. Therefore, since the corresponding frequency is changed together, the receiving side of the mobile communication system must receive the signal by matching the corresponding frequency according to the channel change of the transmitting side without error.

When there is an obstacle on a path from which a signal transmitted from a transmitting side of the mobile communication system is transmitted to a receiving side, multipath occurs due to reflection and the like, and each path has a difference in transmission length, and a difference in length of the path. As a result of the phase difference of the reception frequency, the reception side detects that the reception frequency is fluctuated. In addition, the reception frequency fluctuates due to noise or the like.

In the case of receiving and analyzing a signal at the receiving side, a training sequence is a method of sharing information about a position at which a received signal starts or a reference signal with the same value as the transmitting side.

In the QPSK modulation type mobile communication system, an automatic frequency control (AFC) method can be used to automatically change a transmission frequency of a channel that is changed according to a movement of a cell region of a mobile terminal in order to improve reception performance and increase reliability of a received signal. There is a need for the development of a technology for receiving tuned signals and accurately detecting corresponding communication signals.

Hereinafter, an apparatus for estimating a frequency offset of a mobile terminal according to the prior art will be described with reference to the accompanying drawings.

Attached to explain the prior art, FIG. 1 is a functional configuration diagram of a frequency offset estimator for a mobile terminal according to the prior art.

Referring to FIG. 1, the de-rotation unit 10 cancels the phase difference of the QPSK method by DE-ROTATION of the training sequence signal received in the code division multiple access method by the original training sequence signal; A phase calculation unit 20 for detecting a phase value of the received signal whose phase difference is canceled by the derotation unit 10; A phase unwrapping unit 30 for correcting an error of a function value generated in calculating a phase value of a training sequence detected by the phase calculator 20; The offset estimator 40 analyzes the frequency signal applied from the phase unwrapping unit 30 and estimates and outputs an offset frequency (OFFSET FREQUENCY) value.

Hereinafter, a portable terminal frequency offset estimation apparatus according to the related art having the above configuration will be described in detail with reference to the accompanying drawings.

The derotation unit 10 inputs a received training sequence signal received through a multipath from a transmitting side and modified by noise, and also inputs an original training sequence shared with the transmitting side in the same manner. Enter it.

By analyzing the original training sequence, the QPSK phase value of the received signal can be confirmed, and the phase of the QPSK signal is canceled because DE-ROTATION is performed with the checked phase value.

The canceling of the phase difference by the training sequence is obtained by multiplying the received training sequence signal by a value conjugated to the original training sequence value as follows.

Trot (k) = Tr (k)-T ^* o (k), k = 1, ..., N

: Trot (k); Derotation Value

Tr (k); Received Training Sequence Value

T ^* o (k); Conjugate Value of Original Training Sequence Signal

N; Length of training sequence (144 symbols)

As described above, the reception signal Trot (k) whose phase value is canceled by the derotation unit 10 is applied to the phase calculator 20, and the phase error OFFSET generated by the multipath or the noise is applied. It detects and calculates as follows.

θ (k) = ARCTAN 2 (Q (k), I (k))

; Q is an imaginary part

I is the real part

Since the calculated phase value θ (k) has a range value of −π <θ (k) <π by the ARCTAN2 function, a calculation difference of ± 2π may occur in the θ (k) sequence. have.

As described above, the phase value θ (k) calculated by the ARCTAN2 function calculated and output from the phase calculator 20 is applied to the phase unwrapping unit 30 so that the phase value θ (k) is the previous phase. Check whether the difference from the value θ (k-1) exceeds the value of π, and if it exceeds the value of π in the above check, unwrapping is added to add ± 2π. Correct the difference by

The phase value θ (k) output from the phase unwrapping unit 30 is applied to the offset estimating unit 40 to calculate and output an offset frequency FREQUENCY OFFSET Δf, which is not shown in the figure. The offset frequency is processed to perform automatic frequency control (AFC).

The AFC scheme of the related art calculates a frequency deviation FREQUENCY DIFFERENCY based on an offset frequency by simply using a received training training signal.

Therefore, in the prior art, when a large error (ERROR) is included in a training sequence signal of a received frequency, the offset frequency calculation value is greatly affected, so that the accuracy of the received frequency is reduced and the reliability of the mobile communication system is increased. There is a problem of deterioration.

The present invention, in the calculation of the offset frequency for automatic control of the frequency of the receiver of the KP UE-modulated mobile communication system portable terminal, by removing the received training sequence signal that causes an error exceeding the allowable range and interpolating the average value of the adjacent symbols with high reliability It is an object of the present invention to provide an offset frequency estimating apparatus and method for a portable terminal for outputting an offset frequency.

In order to achieve the above object, the present invention includes a derotation unit for processing a radio received symbol signal as a training sequence symbol signal to cancel the phase difference of the QP-SC modulation signal; A phase calculator for inputting a signal in which the phase difference of the derotation unit is canceled and detecting a phase value; A phase unwrapping unit configured to correct an error occurring during a function calculation process of the phase value detected by the phase calculator; An error detector for analyzing a signal input from the phase unwrapping unit and removing a symbol signal detected in a region other than a determination region; An interpolation unit for obtaining and replacing an average value of the symbol signals adjacent to the symbol signal removed by the error detector; And an offset estimator configured to calculate and output an offset frequency estimate value closest to a straight line having the smallest list square error of the symbol signal applied from the interpolation unit.

In addition, the present invention has been made in order to achieve the above object, the process of receiving the code division multiple access method KPK escape modulation signal by the portable terminal and the de-rotation process of the signal received as the original training sequence signal; A functional process of detecting and calculating a phase of the derotated symbol signal in the process, obtaining a phase value by function processing, and correcting an error caused by the function processing; An error process of removing if the symbol signal of the process is detected outside a predetermined predetermined region; A replacement step of calculating an adjacent symbol signal average value of the symbols removed in the error process and performing interpolation processing; An estimation process of calculating and estimating an offset frequency signal having a smallest list square error and close to a straight line to the symbol signal of the above process is performed.

Hereinafter, an apparatus and method for estimating offset frequency for automatic reception frequency control of a portable terminal according to the present invention will be described with reference to the accompanying drawings.

2 is a functional block diagram of an offset frequency estimating apparatus of a mobile terminal according to the present invention. FIG. 3 is a state diagram of a determination region for distinguishing error symbol signal detection according to the present invention. 4 is a flowchart of an offset frequency estimation method of a mobile terminal according to the present invention.

Referring to FIG. 2, an offset frequency estimator for automatic terminal frequency control (AFC: AUTOMATIC FREQUENCY CONTROL) according to the present invention processes a radio received symbol signal as a training sequence symbol signal, thereby providing a QPSK. The phase difference of the modulated signal is canceled, and the QPSE modulation modulated phase difference set in the received and received training sequence symbol signal is analyzed using the phase value identified by analyzing the original training sequence symbol signal shared with the transmitter. Derotation unit 100 to offset,

The derotation unit 100 inputs a symbol signal of which phase difference of the QPSK modulation scheme is canceled and detects and calculates a phase value. The fading and noise of the input symbol signal are calculated. A phase calculation unit 110 for detecting and calculating the phase value generated by

A phase unwrapping unit 120 that compensates for an error generated by a function calculation process of a phase value detected by the phase calculator 110, and compensates for an error generated by calculating an arc tangent (ARCTAN2) function of each symbol signal; ,

The signal input from the phase unwrapping unit 120 is analyzed and " | θ (k)-θ (k-1) |> π / 4 in the regions other than the determination region, where θ (k) is a symbol detection. The detected phase value, k is an error detection unit 130 for removing the detected symbol signal divided by a reference such as "number of training sequence symbol signal",

The average value of the symbol signal adjacent to the symbol signal removed by the error detection unit 130 is obtained and replaced. When the removed symbol signal is not continuous, " hat θ (k) = (θ (k−1) + θ (k + 1)) / 2, where θ (k) represents the detected phase value of the symbol and k is the average calculated based on the same criteria as "number of training sequence symbol signal" instead of the removed symbol signal value If the removed symbol signal is continuous, " error symbol phase; &amp;thetas; (k + 1), θ (k + 2), ..., θ (k + E), θ ( k + i) = θ (k) + I ^* Δθ, provided that Δθ = (θ (k + E + 1)-θ (k)) / (E + 1), I is 1,2, .., E where θ (k) represents the detected phase value of the symbol, and k is the interpolation unit 140 for interpolating and replacing the calculated value instead of the removed symbol signal value based on a reference such as “number of training sequence symbol signals”. ,

The least square error (LEAST SQUARE ERROR) of the symbol signal applied from the interpolation unit 140 is the smallest and calculates and outputs an offset frequency close to a straight line, and the list square error of the applied and input symbol signal is the smallest. The offset frequency close to the straight line is "Δf = Σ k = 1 to N (Ak + B) θ (k), where A = 12 / (T (N-1) N (N + 1)), B = -6 / (T (N-1) N), N is the length of the training sequence symbol signal, k is the configuration consisting of an offset estimator for estimating based on a reference such as "number of training sequence symbol signal".

Hereinafter, according to the present invention having the above-described configuration, an offset frequency estimating apparatus for automatic terminal receiver frequency automatic control (AFC) will be described in detail with reference to the accompanying drawings.

The DE-ROTATION unit 100 inputs a training sequence symbol signal received through a radio path and inputs an original training sequence symbol signal shared with the transmitting side at the same value.

The training sequence symbol signal is divided into four phase difference signals by the QPSK modulation scheme, and is transmitted. The information on the sequence having phase differences as described above is shared between the transmitting side and the receiving side. .

That is, the training sequence method is a method in which the receiving side already knows that the transmitting side wirelessly transmits the training sequence symbol signal having a certain phase PHASE in advance.

The training sequence symbol signal wirelessly transmitted from the transmitting side in the QPSK modulation scheme may have an error or offset frequency that slightly changes in frequency during transmission due to fading and noise of a radio path. ) Occurs.

In addition, even if the receiving side exactly matches the frequency signal transmitted from the transmitting side, the conditions of each equipment characteristic, component element characteristics, aging degree, use environment, etc. are different. (OFFSET) will occur.

Therefore, as described above, the receiving side receiving the frequency of the transmitting side by the fading and noise of the wireless transmission path, the characteristics of each device, the characteristics of each component, the degree of aging, the surrounding environment, etc., the error or offset in the channel frequency In order to raise the frequency and to improve the quality of the communication signal, a control that tunes to the frequency where the error occurs as described above is required, and automatic frequency control is performed to automatically perform the tuning. AFC).

The present invention relates to a problem in which a wireless signal transmitted from a base station (BS) constituting a mobile communication system, or vice versa, on a wireless transmission path, characteristics of equipment, components, and aging of the mobile terminal MS. And an offset frequency estimating apparatus and method for detecting an offset frequency caused by a change in the surrounding environment using the equipment and accurately tuning the received frequency in an automatic frequency control (AFC) apparatus. To provide.

In the QPSK modulation scheme, a transmission signal is modulated and transmitted in four phases (PHASE), and there are a real part and an imaginary part, and a transmitting side and a receiving side pre-set the phase of the modulated signal. A method of transmitting in a shared symbol (SYMBOL) order is called a training sequence method.

The training sequence symbol signal shared with the training sequence symbol signal wirelessly transmitted to the transmitting side is applied to the derotation unit 100, and the derotation unit 100 analyzes the shared training sequence symbol signal. The transmitter confirms the phase difference in which the phase change is set by the QPSK method.

The derotation unit 100 cancels the phase difference of the received signal by using the phase difference value identified as described above.

The cancellation of the phase difference as described above is processed by conjugating the phase of the original training sequence symbol signal by the phase of the received training sequence symbol signal.

That is, if the received training sequence symbol signal is Tr (k), and the shared original training sequence symbol signal is To (k), and k is a sequence number of symbols from 1 to 144, the derotated signal Trot (k) is , Trot (k) = Tr (k) .T ^* o (k).

As described above, each symbol signal of the QP-ESC modulation method in which the phase difference is canceled by the derotation unit 100 is detected by the phase calculator 110 based on a constant reference value of the phase value θ (k). Is calculated.

Since the phase value θ (k) of each symbol signal detected by the phase calculator 110 includes a real part I (k) and an imaginary part Q (k), the real part I (k)) and the imaginary part Q (k) must be calculated as a function, and the phase unwrapping part 120 calculates the function of ARCTAN2 (I (k), Q (k)).

The functional calculation of the ARCTAN2 (I (k), Q (k)) is because the range of the result value is -π <θ (k) <π, the phase value θ (k) may occur by ± 2π.

Therefore, it is checked whether or not the difference between the current phase value θ (k) and the previous phase value θ (k-1) exceeds the value of π, and when the value of the value exceeds the value of π, the current phase value θ by ± 2π. It should be added (+) or minus (-) according to the sign of (k), and the above process is called unwrapping process and is processed by the phase unwrapping unit 120.

When analyzing the phase value θ (k) of the Kewpie escape symbol processed by the phase unwrapping unit 120, the offset frequency Δf can be estimated. There may be an error symbol that is out of the allowable range, and the error detection unit 130 detects and removes the offset frequency Δf due to the error symbol as described above.

In order to detect the QPSK error symbol as described above, a DECISION REGION having a predetermined phase region is set, and symbols detected in the phase region outside the determination region are divided into error symbols.

3, a determination region for distinguishing the error symbol signal detection of the present invention is shown, and a phase region outside the determination region is divided by the following criteria.

| θ (k)-θ (k-1) | > π / 4

Where k is a symbol from 1 to 144

The criterion is limited to the estimated range of the normal symbol within ± π / 4 to easily distinguish the error symbol, the symbol corresponding to the criterion is divided into a symbol having an error phase, detected by the error detector 130 And removed.

For example, in the case of the TD-SCDMA (TIME DIVISION-SYNCHRONOUS CDMA) method selected as a standard for a mobile communication system in China, since 144 chips (CHIP) are used as a training sequence symbol, 144 phase values (θ (k)) are obtained. Can be.

However, in order to use the LEAST SQUARE ERROR method, since the total phase value of the phase sequence is required, it is necessary to replace the phase value removed by the error detector 130.

The interpolation unit 140 calculates a phase value replacing the symbol deleted by the error detection unit 130 and interpolates and replaces the phase value. When the phase symbol of the deleted symbol is one, the previous symbol is replaced. The phase value θ (k-1) and the symbol phase value θ (k + 1) are averaged to estimate the normal phase value, and the averaged phase value is interpolated.

As described above, the process of estimating the normal phase value by averaging the phase values of the previous symbol and the subsequent symbol by the interpolation unit 140 is expressed as follows.

Hat θ (k) = (θ (k-1) + θ (k + 1)) / 2

In addition, when the error detector 130 detects a symbol error that occurs continuously, the interpolation unit 140 detects the difference between the phase value of the previous symbol and the phase value of the next symbol to equally estimate the phase. Interpolate by value.

For example, when an error occurs continuously and the number of detected symbols is E, it is estimated as follows.

Phase of error symbol: θ (k + 1), θ (k + 2), ..., θ (k + E)

θ (k + i) = θ (k) + i ^* Δθ

I is 1,2, .. E

Δθ is (θ (k + E + 1) -θ (k)) / (E + 1)

As described above, the phase value of the symbol estimated by the interpolation unit 140 is interpolated to replace the position of the corresponding symbol, and the phase values of all 144 symbols output from the interpolation unit 140 are offset as described above. Is applied to the estimator 150.

The offset frequency estimator of the KPSC receiver uses a phase sequence θ (k) for the received training sequence, and if there is no fading or noise in the radio path through which the signal is transmitted, The sequence of phases being linear has a linear characteristic proportional to the offset frequency (FREQUENCY OFFSET) Δf and is expressed as follows.

θ (k) = 2πΔfT + θo

Where θo is the value of ARBITRARY PHASE OFFSET

T is the symbol period (SYMBOL PERIOD, approximately 0.78125 ㎲)

However, in general, since fading or noise exist in the radio path or channel, the linear characteristics are distorted, and thus, the offset frequency closest to the straight line, that is, the LEAST SQUARE ERROR is the least, FREQUENCY OFFSET) Δf is estimated, and the LEAST SQUARE ERROR is expressed as follows.

ε = Σ k = 1 to N [θ (k)-(ΔfkT + hat θ)] square

In order to find the offset frequency Δf that minimizes ε, a derivative in the form of ∂ε / ∂Δf is taken, and the offset estimator 150 estimates the offset frequency Δf through the following process. do.

Δf = Σ k = 1 to N (Ak + B) θ (k),

Where A = 12 / (T (N-1) N (N-1))

B = -6 / (T (N-1) N)

According to the present invention having the above-described configuration, in a receiving terminal automatic frequency control (AFC) of a QPSK modulation type mobile terminal, an error symbol is easily removed, and an interpolation (INTERPOLATION) to an average value of neighboring symbols leads to LEAST. Since the offset frequency Δf with the smallest squaring error is secured, frequency control with high accuracy at the receiving end can be performed.

Hereinafter, with reference to the accompanying Figure 4, the offset frequency estimation method of the mobile terminal according to the present invention will be described.

When a signal is wirelessly received by a mobile terminal that performs QPSK modulation in a code division multiple access (CDMA) or TD-SCDMA mobile communication system, the signal received by the original training sequence signal is decoded. Determining whether the mobile terminal receives a KPSP modulation signal by performing a rotation process (S100); If it is determined in step S100 that the UE receives the Kewpiesuke modulation signal, a preparation process consisting of a process of derotation (DE-ROTATION) to cancel the phase difference by the original training sequence signal (S110),

Detecting and calculating the phase PHASE of the derotated symbol signal in the preparation process, obtaining a phase value by function processing, and correcting the error by the function processing, detecting and calculating the phase of the input signal. , Arctangent 2 (ARCTAN2) function to obtain the phase value (θ (k)), and the function process has a range of -π <phase value <π to 2π according to the sign of the phase value for error correction A process of unwrapping (UNWRAPPING) adding or subtracting a value (S120),

If the symbol signal of the functional process (S120) is detected outside the predetermined predetermined region is removed, determining whether the unwrapped symbol signal in the functional process (S120) is detected outside the predetermined determination region ( S130); If the symbol signal is detected out of the determination region in the step (S130) and the error process comprising the step of determining and removing (S140) as an error symbol,

The interpolation process is performed by calculating an average value of adjacent symbol signals of symbols removed in the error process. When the symbol signals removed in the error process are not continuous, " hat θ (k) = (θ (k-1) + θ) (k + 1)) / 2, where θ (k) calculates an average value based on the detected phase value of the symbol, and k is the number of the training sequence symbol signal, and the symbol signal removed in the error process is continuous. If the error symbol phase is θ (k + 1), θ (k + 2), θ (k + E), θ (k + i) = θ (k) + I ^* Δθ, Where θθ = (θ (k + E + 1)-θ (k)) / (E + 1), i is 1,2, .. E, θ (k) is the detected phase value of the symbol, k Since the average value is calculated based on the same reference number as "the number of training sequence symbol signals", calculating and outputting an adjacent symbol average value (S150); An alternative process consisting of interpolating (S160) the symbol value position removed with the average value calculated in the process (S150);

Computing and estimating the offset frequency signal having the smallest list square error and close to the straight line to the symbol signal of the replacement process. = Σ k = 1 to N (Ak + B) θ (k), where A = 12 / (T (N-1) N (N + 1)), B = -6 / (T (N-1 N), N is the length of the training sequence symbol signal, k is the estimation process (S170) for estimating based on a reference such as "number of training sequence symbol signal".

Hereinafter, the method of estimating the offset frequency of the mobile terminal according to the present invention having the above configuration will be described in detail with reference to the accompanying drawings.

For example, in the TD-SCDMA mobile communication system, since information of a certain period is composed of 144 chips (CHIP) or symbols (SYMBOL), the training sequence information is configured in units of 144 symbols.

The training sequence frequency signal wirelessly transmitted from the transmitting side of the mobile communication system generates a phase difference at the receiving side due to fading and noise generated through the wireless transmission path, characteristics of the transmitting and receiving communication equipment, and changes in the use environment. do.

When the phase difference generated as described above is detected, the difference value between the transmission frequency and the currently set reception frequency, that is, the offset frequency value Δf is confirmed, and accurate tuning (using the difference value or offset frequency value Δf) as described above is performed. TUNING), and is generally used for automatic reception frequency control (AFC), and the present invention is to find an offset frequency for accurate tuning as described above.

The portable terminal registers and communicates with a mobile communication system. For example, the portable terminal subscribes to a code division multiple access (CDMA) or TD-SCDMA mobile communication system and wirelessly communicates with the other party.

The portable terminal determines whether to receive and input a KewPie-escapsulated modulation signal transmitted from the transmitting side (S100), and when the KPY-Skew modulated received signal is input in the determination (S100), the mobile terminal determines a phase value according to the original training sequence. The received and input signal is subjected to a de-rotation (DE-ROTATION) process to cancel the phase value by the derotation unit 100 (S110).

The CDA or TD-SCDMA mobile communication system uses a training sequence method for sharing phase information of signals transmitted from a transmitting side and a receiving side by a QPSK modulation scheme, and transmits a signal to be transmitted. By converting into four phases by the QP escaping method, the transmitting side and the receiving side share the information of the converted phase. Therefore, the receiving side knows in advance the phase information of the corresponding signal transmitted by the transmitting side. The sharing of phase information of the same KP-CS is called a training sequence.

By the derotation process (S110) as described above, the 144 symbol signals having the four phase values have the same phase value, and each phase value is detected and calculated by the phase calculator 110. The real time and the imaginary parts are processed by the arctangent 2 function by the phase unwrapping unit 120 and unwraped by the error occurring in the function processing (S120).

Since the signal processed through the process (S120) has an abnormal or error phase value due to heavy fading or heavy noise in the process of being transmitted through the wireless transmission path, check and detect whether there is an error symbol having such an error phase value. (S130).

In detecting the error symbol, the determination region of FIG. 3 is used, and a symbol detected in an area other than the DECISION REGION becomes an error symbol.

When an error symbol having an error phase value is detected as described above, the erase symbol is deleted to remove the corresponding error symbol (S140), and the interpolation unit 140 removes the removed or deleted symbol to minimize the LEAST SQUARE ERROR. An arithmetic operation is performed to obtain an average value from the adjacent symbol values (S150), and the interpolation is performed to interpolate the deleted symbol value with the averaged value calculated as described above (S160).

In the process of obtaining the average value of the adjacent symbols, when only one symbol detected and deleted as an error symbol is not consecutive, the average value of the adjacent symbols is calculated based on the following criteria.

Hat θ (k) = (θ (k-1) + θ (k + 1)) / 2

: θ (k); The detected phase value of the symbol

k; Number of training sequence symbol

In the process of obtaining the average value of the adjacent symbols, when the symbols detected and deleted as the error symbols are consecutive, the average value of the adjacent symbols is calculated based on the following criteria.

Error symbol phase; θ (k + 1), θ (k + 2), ..., θ (k + E)

θ (k + i) = θ (k) + i ^* Δθ

: Δθ = (θ (k + E + 1)-θ (k)) / (E + 1)

i; 1,2, .., E

: θ (k); The detected phase value of the symbol

k; Number of training sequence symbol

The signal processed as described above and interpolated is applied to the offset estimator 150 to estimate an offset frequency (FREQUENCY OFFSET) that minimizes LEAST SQUARE ERROR based on the following reference.

ε = Σ k = 1 to N [θ (k)-(ΔfkT + hat θ)] square

Δf = Σ k = 1 to N (Ak + B) θ (k)

: A = 12 / (T (N-1) N (N + 1))

: B = -6 / (T (N-1) N)

: N; Length of training sequence symbol signal

k; Number of training sequence symbol

Therefore, the present invention can easily detect and eliminate error occurrence symbols by using DECISION REGION in the estimation of offset frequency Δf generated by the radio transmission path, and the phase of adjacent symbols. Since the values are averaged and interpolated, the LEAST SQUARE ERROR is minimized, so the offset frequency is improved.

The present invention having the above-described configuration has an industrial use effect of easily detecting and removing abnormally received symbol signals because the error symbol signals of the QSPACE modulation method received signals are detected by the determination region.

In addition, since the adjacent symbol phase values of the detected and removed error symbols are averaged and interpolated, there is an industrial use effect of offset frequency estimation that minimizes the list square error.

In addition, since error symbols are easily detected and removed, and average phase values of adjacent symbols are calculated and interpolated to estimate an offset frequency at which the list square error is minimized, a reception frequency accurately tuned to a transmission frequency is estimated. It has a convenient effect on improving reliability and improving communication quality.

1 is a functional block diagram of a frequency offset estimation apparatus for a portable terminal of the prior art;

2 is a functional block diagram of an offset frequency estimation apparatus for a mobile terminal according to the present invention;

3 is a state diagram of a determination region for distinguishing error symbol signal detection according to the present invention;

4 is a flowchart illustrating an offset frequency estimation method of a mobile terminal according to the present invention.

          ** Explanation of symbols on the main parts of the drawing **

100: derotation unit 110: phase calculation unit

120: phase unwrapping unit 130: error detection unit

140: interpolation unit 150: offset estimation unit

Claims (16)

A derotation unit for processing a radio received symbol signal as a training sequence symbol signal to cancel a phase difference of a QP-SC modulation signal; A phase calculator which inputs a signal in which the phase difference of the derotation unit is canceled and detects a phase value; A phase unwrapping unit configured to correct an error occurring during a function calculation process of the phase value detected by the phase calculator; An error detector which analyzes a signal input from the phase unwrapping unit and removes a symbol signal detected in a region other than a determination region; An interpolation unit for obtaining and replacing an average value of the symbol signals adjacent to the symbol signal removed by the error detector; And an offset estimator configured to calculate and output an offset frequency estimate value closest to a straight line having the smallest list square error of the symbol signal applied from the interpolation unit. According to claim 1, The derotation unit cancels the phase difference of the QPSE modulation scheme set in the received and received training sequence symbol signal by using the phase difference identified by analyzing the original training sequence symbol signal shared with the transmitter, The phase calculating unit detects and calculates a phase value generated by fading and noise of a symbol signal inputted by canceling a phase difference modulation method of the QPSE modulation scheme from the derotation unit. And the phase unwrapping unit is configured to compensate for an error generated by calculating an arc tangent function of each symbol signal. The method of claim 1, wherein the error detection unit, And an offset frequency estimating apparatus of the portable terminal, characterized in that the symbol signal is input from the phase unwrapping unit to remove the symbol signal when detected in a region other than the set determination region. The method of claim 3, wherein the error detection unit, An offset frequency estimating apparatus for a portable terminal, characterized in that it is configured to detect and remove symbol signals other than the set determination region based on the following criteria. | θ (k)-θ (k-1) | > π / 4 : θ (k); The detected phase value of the symbol k; Number of training sequence symbol The method of claim 1, wherein the interpolation unit, And a symbol signal value adjacent to the symbol signal removed by the error detector, and averaging the symbol signal value and replacing the averaged value instead of the removed symbol signal value. The method of claim 5, wherein the interpolation unit, If the symbol signal removed from the error detector is not continuous, the offset frequency estimation apparatus of the mobile terminal, characterized in that for the replacement of the symbol signal value instead of the average value obtained by the following criteria. Hat θ (k) = (θ (k-1) + θ (k + 1)) / 2 : θ (k); The detected phase value of the symbol k; Number of training sequence symbol The method of claim 5, wherein the interpolation unit, If the symbol signal removed by the error detection unit is continuous, the offset frequency estimation apparatus of the mobile terminal, characterized in that the configuration is configured to replace the average value obtained in accordance with the following criteria instead of the removed symbol signal value. Error symbol phase; θ (k + 1), θ (k + 2), ..., θ (k + E) θ (k + i) = θ (k) + i ^* Δθ : Δθ = (θ (k + E + 1)-θ (k)) / (E + 1) i; 1,2, .., E : θ (k); The detected phase value of the symbol k; Number of training sequence symbol The method of claim 1, wherein the offset estimator, And an offset frequency of the list signal having the smallest square error of the symbol signal inputted from the interpolation unit and estimating an offset frequency close to a straight line based on the following criteria. Δf = Σ k = 1 to N (Ak + B) θ (k) : A = 12 / (T (N-1) N (N + 1)) : B = -6 / (T (N-1) N) : N; Length of training sequence symbol signal k; Number of training sequence symbol A process of receiving the code division multiple access method Kewpiescape modulation signal by the mobile terminal and derotating the signal received as the original training sequence signal; A functional process of detecting and calculating a phase of the derotated symbol signal in the process, obtaining a phase value by function processing, and correcting an error caused by the function processing; An error process of removing if the symbol signal of the process is detected outside the predetermined predetermined region; An alternative process of calculating an adjacent symbol signal average value of symbols removed in the error process and performing interpolation; And an estimation process of calculating and outputting an offset frequency signal having a smallest list square error and close to a straight line to the symbol signal of the process, and outputting the offset frequency signal. The method of claim 9, wherein the preparation process, Determining whether a code division multiple access method mobile terminal receives a QPSC modulation signal; And determining a reception of the KPCU-esqueued modulated signal in the above process. The method of claim 9, wherein the functional process, Detects and calculates the phase of the input signal in the preparation process, calculates the phase value by performing the arc tangent 2 function, and calculates the phase value of the phase value for error correction caused by the function process having a range of −π <phase value <π. The method of estimating the offset frequency of a mobile terminal, characterized in that the unwrapping process of adding or subtracting 2π value according to the sign. The method of claim 9, wherein the error process, Determining whether an unwrapped symbol signal in the function process is detected outside a predetermined determination region; If the symbol signal is detected out of the determination region in the process, the offset frequency estimation method of the mobile terminal comprising the step of determining and removing the error symbol. The method of claim 9, wherein the replacement process, Calculating and outputting an adjacent symbol average value of symbols removed in the error process; And offsetting the symbol value position removed by the average value calculated in the above process. The method of claim 13, wherein the average value calculation output process, And calculating an average value based on a following criterion when the removed symbol signals are not continuous. Hat θ (k) = (θ (k-1) + θ (k + 1)) / 2 : θ (k); The detected phase value of the symbol k; Number of training sequence symbol The method of claim 13, wherein the average value calculation output process, And calculating an average value based on a following criterion when the removed symbol signals are continuous. Error symbol phase; θ (k + 1), θ (k + 2), ..., θ (k + E) θ (k + i) = θ (k) + i ^* Δθ : Δθ = (θ (k + E + 1)-θ (k)) / (E + 1) i; 1,2, .., E : θ (k); The detected phase value of the symbol k; Number of training sequence symbol The method of claim 9, wherein the estimating process, And an offset frequency closest to a straight line having the smallest list square error of the input symbol signal subjected to interpolation in the replacement process based on the following criteria. Δf = Σ k = 1 to N (Ak + B) θ (k) : A = 12 / (T (N-1) N (N + 1)) : B = -6 / (T (N-1) N) : N; Length of training sequence symbol signal k; Number of training sequence symbol