Wireless communication device and frequency offset correction method

Abstract

To improve both of frequency offset estimation accuracy and possible range of estimation in burst transmission.SOLUTION: In a wireless communication system 1, a wireless communication device 2 performs frequency offset correction on a received signal in three runs based on three frequency offset estimation values calculated by first to third frequency offset estimation units. The first to third frequency offset estimation units calculates estimated frequency offset values using different autocorrelation symbol intervals τ1 to τ3 (τ is an integer), the symbol interval τ being set such that τ1<τ2<τ3. Among the first to third frequency offset estimation units, the first frequency offset estimation unit estimates a frequency offset amount included in the received signal. The second and third frequency offset estimation units each estimate a frequency offset amount included in a received signal after a frequency offset correction that is based on the estimated frequency offset value calculated by the immediately preceding frequency offset estimation unit.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to a wireless communication device and a frequency offset correction method for correcting a frequency offset in burst transmission. [Background technology]

[0002] Burst communication is a communication method in which discontinuous signals are transmitted at irregular intervals, and is used in time division multiplexing (TDM) communication, etc. When burst communication is a many-to-one communication, the other station may change for each communication, and in such a case, it is necessary to estimate correction values ​​such as frequency offset and clock error every time a burst signal is received.

[0003] Incidentally, a data frame used in burst communication is composed of a data signal and a preamble signal added to the beginning of the data signal, as shown in Fig. 8(A). This preamble signal is generally used to correct frequency offsets, clock errors, and the like.

[0004] As shown in Fig. 8(B), the frequency offset correction is performed in an automatic frequency control (AFC) 100. Specifically, a preamble signal is extracted from a received signal output upon receiving a data frame, a frequency offset estimation unit 101 calculates a frequency offset estimation value, a numerically controlled oscillator (NCO) 102 converts the frequency offset estimation value into a complex number, and a mixer 103 receives an input of the received signal and also receives an input of the complex number output from the NCO 102 and performs complex multiplication processing on the received signal.

[0005] A method of estimating using autocorrelation of a preamble signal is known as a conventional frequency offset estimation method in frequency offset estimation unit 101 (see, for example, Patent Document 1). In this method, a frequency offset estimate is obtained by the following equation: where Δfest is a frequency offset estimate, fsym is a symbol frequency (symbol rate), τ is a symbol interval of autocorrelation, atan is an arctangent, M is an autocorrelation length, x(i) is a preamble signal, * is a complex conjugate, and Rm is an autocorrelation (where τ and M are integers).

number

[0006] [Patent Document 1] JP 2008-167230 A Summary of the Invention [Problem to be solved by the invention]

[0007] However, in the conventional frequency offset estimation method, while it is possible to estimate the frequency offset with high estimation accuracy by increasing the symbol interval τ of the autocorrelation, there is a problem that the estimable range of the frequency offset is narrowed. Therefore, even if a sufficient estimation accuracy can be ensured in a system with a small frequency offset, it is not possible to ensure the estimation accuracy in a system with a large frequency offset.

[0008] Therefore, an object of the present invention is to provide a wireless communication apparatus and a frequency offset correction method that can improve both the estimation accuracy and the estimable range of a frequency offset in burst transmission. [Means for solving the problem]

[0009] In order to solve the above problem, the invention described in claim 1 is a wireless communication device comprising: receiving means for receiving a data frame that has been burst-transmitted with a preamble signal added to the beginning of a data signal, and outputting a received signal; and automatic frequency control means for calculating a frequency offset estimate value from an autocorrelation of the preamble signal, and performing frequency offset correction of the received signal based on the frequency offset estimate value, wherein the automatic frequency control means performs the frequency offset correction of the received signal n times based on n frequency offset estimate values ​​calculated by n frequency offset estimation means (n ​​is an integer equal to or greater than 2), and Each of the estimation means calculates the frequency offset estimation value using a different symbol interval τ (τ1 to τn symbols) of autocorrelation (τ is an integer), and the symbol interval τ is set to satisfy τ1<τ2<τ3<...<τn. A first of the n frequency offset estimation means estimates a frequency offset amount contained in the received signal, and second to nth frequency offset estimation means estimate a frequency offset amount contained in the received signal after the frequency offset correction based on the frequency offset estimation value calculated by the frequency offset estimation means immediately preceding it.

[0010] The invention described in claim 2 is characterized in that, in the wireless communication device described in claim 1, only the number of symbols of the preamble signal included in the data frame required by the first frequency offset estimation means are added to the beginning of the data signal, and the remainder are inserted at a predetermined position in the section formed by the data signal.

[0011] The invention recited in claim 3 is a frequency offset correction method for receiving a data frame which has been burst-transmitted with a preamble signal added to the beginning of a data signal, outputting a received signal, calculating a frequency offset estimation value from an autocorrelation of the preamble signal, and performing frequency offset correction of the received signal based on the frequency offset estimation value, characterized in that the frequency offset correction of the received signal is performed n times based on n frequency offset estimation values ​​calculated in n frequency offset estimation steps (n is an integer of 2 or more), and in each of the n frequency offset estimation steps, the frequency offset estimation value is calculated using a different symbol interval τ (τ1 to τn symbols) (τ is an integer), and the symbol interval τ is set so as to satisfy τ1<τ2<τ3<···<τn, and in a first frequency offset estimation step among the n frequency offset estimation steps, an amount of frequency offset included in the received signal after the frequency offset correction is estimated based on the frequency offset estimation value calculated in the frequency offset estimation step immediately preceding the first frequency offset estimation step, Effect of the Invention

[0012] According to the inventions recited in claims 1 and 3, a plurality of frequency offset estimation means (steps) are provided, the symbol interval τ of the autocorrelation used in each frequency offset estimation means (step) is set so as to gradually increase (τ1<τ2<τ3<···<τn) to calculate a plurality of frequency offset estimation values, and frequency offset correction is performed in a plurality of times based on the plurality of calculated frequency offset estimation values, so that a wide estimable range can be ensured in the first frequency offset estimation means (step), and the estimation accuracy, which is in a trade-off relationship with the estimable range, can be ensured in the subsequent second to nth frequency offset estimation means (steps), making it possible to achieve both a wide estimable range and high estimation accuracy. Therefore, it becomes possible to ensure sufficient estimation accuracy even in a system in which the frequency offset becomes large.

[0013] According to the invention recited in claim 2, when estimating and correcting a frequency offset in multiple steps while gradually increasing the estimation accuracy using multiple frequency offset estimation means (steps), the number of symbols of the preamble signal required increases compared to the case of estimating and correcting the frequency offset in one step. However, by adding only the number of symbols required for the first frequency offset estimation means (step) to the beginning of a data signal and inserting the remaining increased amount of the preamble signal at a predetermined position in a section formed by a data signal, the increased number of symbols of the preamble signal can be suppressed, and therefore a decrease in throughput can be suppressed. As a result, it is possible to simultaneously achieve three goals: high estimation accuracy, a wide estimation range, and high throughput. [Brief description of the drawings]

[0014] [Figure 1] 1 is a diagram showing a schematic configuration of a wireless communication system according to a first embodiment of the present invention. [Diagram 2] 2 is a functional block diagram showing a schematic configuration of a transmission unit of the wireless communication device shown in FIG. [Diagram 3] 2 is a functional block diagram showing a schematic configuration of a receiving section of the wireless communication device shown in FIG. [Figure 4] 1 is a functional block diagram showing a schematic configuration of an AFC in accordance with an embodiment 1 of the present invention. [Diagram 5] 5A is a table showing a range in which the frequency offset can be estimated by the first to third frequency offset estimators shown in FIG. 4, and FIG. 5B is a graph showing the estimation accuracy of the frequency offset by the first to third frequency offset estimators shown in FIG. 4. [Figure 6] 5 is a flowchart showing a procedure for frequency offset correction in the AFC shown in FIG. [Figure 7] FIG. 1A is an explanatory diagram showing the frame structure of a conventional data frame when used in a first embodiment of the present invention, and FIG. 1B is an explanatory diagram showing the frame structure of a data frame used in a second embodiment of the present invention. [Figure 8]FIG. 1A shows the frame structure of a data frame used in conventional burst communication, and FIG. 1B is an explanatory diagram showing a schematic configuration of conventional AFC. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] The present invention will be described below based on the embodiments shown in the drawings. Note that, in the following, only the characteristic configuration of the present invention will be described, and a description of the same mechanism as the conventional one for performing wireless communication will be omitted.

[0016] (Embodiment 1) 1 is a diagram showing a schematic configuration of a wireless communication system 1 using a wireless communication device 2 and a frequency offset correction method according to an embodiment of the present invention. A wireless communication device 2 and an antenna 3 are arranged in each of wireless communication transmitting and receiving stations constituting the wireless communication system 1. The wireless communication devices 2 are connected to each other by a wireless line 4 via the antenna 3. Note that a communication satellite or the like may be used as a relay station in the wireless line 4.

[0017] First, in this embodiment, a schematic configuration of a wireless communication device 2 according to the present invention will be described.

[0018] The wireless communication device 2 includes an interface unit 5, a transmitter 6, a splitter 7, and a receiver 8 (see the lower part of FIG. 1).

[0019] Here, the wireless communication device 2 is a device that is equipped with a transmission mechanism and a reception mechanism and transmits and receives data. In the following description, the wireless communication device 2 when performing processing related to transmission using the transmission mechanism is referred to as the "transmitting side," and the wireless communication device 2 when performing processing related to reception using the reception mechanism is referred to as the "receiving side."

[0020] The interface unit 5 mainly includes a data circuit-terminating device 51 (including devices called data communication devices and data circuit devices). The interface unit 5 receives transmission data to be transmitted, and outputs the transmission data to the transmission unit 6 via the data circuit-terminating device 51.

[0021] The transmitter 6 receives the transmission data output from the interface 5, combines the data signal obtained by mapping the transmission data with a preamble signal placed at the beginning of the data signal to generate a data frame, and further superimposes a carrier signal of a predetermined frequency on the data frame to perform digital modulation. The preamble signal includes an unmodulated CW signal used for frequency offset correction and an alternating signal used for clock error correction. The modulation method used in the wireless communication system 1 is not limited to a specific method, but for example, quadrature amplitude modulation (QAM) is used.

[0022] The transmitter 6 converts the digitally modulated data frame into an analog signal, converts the frequency of the signal into a high-frequency signal higher than a predetermined frequency, and outputs the signal as a transmission signal after amplifying it with a power amplifier. The transmission signal is guided from the transmitter 6 to the antenna 3 via the splitter 7, and is transmitted in bursts as radio waves from the antenna 3 via the wireless line 4 to the antenna 3 of the other wireless communication device 2 (in other words, the receiving side in this communication).

[0023] In addition, when a transmission signal is transmitted in bursts as radio waves from the antenna 3 of the other wireless communication device 2 (in other words, the transmitting side in this communication) to the antenna 3 of the wireless communication device 2 (in other words, the receiving side in this communication) via wireless line 4, the antenna 3 converts the received radio waves into an electrical signal (received signal) and outputs it.

[0024] The received signal output from the antenna 3 is guided to the receiving unit 8 via the branching filter 7. The receiving unit 8 receives the received signal, passes it through a channel filter that passes only signals in a predetermined frequency band, and converts it into a signal with a frequency lower than the high frequency. The receiving unit 8 further converts the frequency-converted received signal into a digital signal and outputs it.

[0025] The receiver 8 performs quadrature detection processing (demodulation) on the digital received signal to generate an in-phase component (Ich) baseband signal and a quadrature component (Qch) baseband signal, the phases of which are orthogonal to each other. In the following explanation, the in-phase component and the quadrature component will not be particularly distinguished from each other and will be explained as being common to both, except when it is necessary to focus on the in-phase component and the quadrature component separately, and in the drawings, the in-phase component signal and the quadrature component signal are represented by a single signal line.

[0026] The receiver 8 performs burst detection based on the demodulated baseband signal. When the arrival of a data frame is detected by this burst detection, the receiver 8 performs frequency offset correction, clock error correction, phase correction, etc. on the received data frame. The receiver 8 then separates the preamble signal from the data frame that has been subjected to various corrections, performs demapping on the data signal to generate transmission data, and outputs it to the interface 5.

[0027] 2 is a functional block diagram showing a schematic configuration of the transmission unit 6. The transmission unit 6 includes a FIFO memory 61, a mapping unit 62, a preamble signal generation unit 63, a combining unit 64, a transmission ROF 65, a quadrature modulation unit 66, a DAC (Digital Analog Converter) 67, a mixer 68, a local oscillator 69, and a power amplifier 610.

[0028] The FIFO memory 61 is a memory that temporarily stores the transmission data output from the interface unit 5 and outputs it to the mapping unit 62, and transfers the transmission data to the mapping unit 62 on a so-called first-in first-out basis.

[0029] The mapping unit 62 performs mapping processing on the binary data string of the transmission data so as to obtain a predetermined signal point arrangement, generates a data signal consisting of a symbol string, and outputs the data signal to the combining unit 64. The preamble signal generator 63 generates a CW signal, which is an unmodulated continuous wave used for frequency offset correction, and an alternating signal used for clock error correction, and outputs the generated signals to the combining unit 64.

[0030] The combining unit 64 combines the preamble signal input from the preamble signal generation unit 63 with the beginning of the data signal input from the mapping unit 62 to generate a data frame DF (see FIG. 7 ) and outputs it to the transmission ROF 65. The transmission ROF 65 has a roll-off filter function and performs band limiting processing on the data frame DF input from the combining unit 64 and outputs it to the orthogonal modulation unit 66.

[0031] The quadrature modulation unit 66 superimposes a carrier signal of a predetermined frequency on the data frame DF input from the transmission ROF 65, digitally modulates it, and outputs it to the DAC 67. Note that the modulation method used in the quadrature modulation unit 66 is not limited to a specific method, but for example, quadrature amplitude modulation is used.

[0032] The DAC 67 converts the data frame DF input from the quadrature modulation unit 66 into an analog transmission signal and outputs it to the mixer 68. The local oscillator 69 generates a local oscillation signal having a predetermined fixed frequency and outputs the generated local oscillation signal to the mixer 68. The mixer 68 mixes the local oscillation signal with the transmission signal input from the DAC 67 to convert it into a signal with a higher frequency than the predetermined frequency.

[0033] The power amplifier 610 amplifies the transmission signal frequency-converted by the mixer 68 and outputs it to the antenna 3. The antenna 3 transmits the transmission signal amplified by the power amplifier 610 as a radio wave to the antenna 3 of the other wireless communication device 2 (in other words, the receiving side in this communication). Although not shown, a splitter 7 is connected between the power amplifier 610 and the antenna 3.

[0034] 3 is a functional block diagram showing a schematic configuration of the receiving unit 8. The receiving unit 8 includes a channel filter 81, a mixer 82, a local oscillator 83, a variable ATT (attenuator) 84, an ADC (Analog Digital Converter) 85, a quadrature detection unit 86, a burst detection unit 87, a timing control unit 88, an AFC (Automatic frequency control) 89, a receiving ROF 810, a symbol recovery unit 811, an APC (Automatic Phase Control) 812, a separator 813, a demapping unit 814, an AGC (Automatic Gain Control) 815, and a DAC 816.

[0035] The antenna 3 converts the received radio waves into an electric signal (received signal) and outputs it to the channel filter 81. The splitter 7 is connected between the antenna 3 and the channel filter 81. The channel filter 81 passes a predetermined frequency band of the received signal input from the antenna 3 and outputs it to the mixer 82. The local oscillator 83 generates a local oscillation signal having a predetermined fixed frequency and outputs the generated local oscillation signal to the mixer 82. The mixer 82 mixes the local oscillation signal with the received signal input from the channel filter 81 to convert it into a signal with a frequency lower than the predetermined frequency, and outputs it to the variable ATT 84.

[0036] The variable ATT 84 includes an attenuator, and adjusts the attenuation amount of the received signal output from the mixer 82 in response to an external signal to attenuate the signal and output the attenuated signal to the ADC 85. The attenuation amount in the variable ATT 84 changes based on a control signal of the AGC 815 supplied via the DAC 816.

[0037] The ADC 85 converts into a digital signal the received signal input from the variable ATT 84. The quadrature detection unit 86 performs quadrature detection processing on the received signal to generate an in-phase component (Ich) baseband signal and a quadrature component (Qch) baseband signal whose phases are orthogonal to each other.

[0038] The burst detection unit 87 performs burst detection from the baseband signal input from the quadrature detection unit 86. When the burst detection unit 87 detects the arrival of a data frame DF by burst detection, it outputs a detection flag to the timing control unit 88. The timing control unit 88, to which the detection flag has been input, outputs an enable signal to each module to cause it to process the data frame DF.

[0039] In response to an enable signal from the timing control unit 88, the AFC 89 performs frequency offset correction on the data frame DF of the baseband signal and outputs the result to the receive ROF 810. The receive ROF 810 has a roll-off filter function, and performs band limiting processing on the baseband signal input from the AFC 89 and outputs the result to the symbol recovery unit 811.

[0040] The symbol recovery unit 811 corrects the clock error using an alternating signal included in the data frame DF of the baseband signal input from the reception ROF 810, and outputs the corrected signal to the APC 812. The APC 812 performs phase correction on the data frame DF of the baseband signal input from the symbol recovery unit 811, and outputs the corrected signal to the separation unit 813.

[0041] The separator 813 separates the data signal from the data frame DF and outputs the data signal to the demapping unit 814. The demapping unit 814 performs demapping processing (decoding processing) on ​​the signal (each of the in-phase component and the quadrature component) consisting of the symbol sequence data input from the separator 813, converts the symbol sequence data into transmission data of a binary data sequence, and outputs the transmission data to the interface unit 5.

[0042] 4 is a functional block diagram showing a schematic configuration of an AFC 89 of a wireless communication device 2 according to this embodiment. The AFC 89 includes a plurality of frequency offset estimation units (frequency offset estimation means) 891, a plurality of delay adjustment units 892, a plurality of NCOs 893, and a plurality of mixers 894.

[0043] The present invention is characterized in that the AFC 89 includes n (n is an integer equal to or greater than 2) frequency offset estimating units 891, and performs frequency offset correction n times using the n frequency offset estimated values ​​calculated by each frequency offset estimating unit 891. While n is most preferably 2, this embodiment will be described taking the case of n=3 as an example.

[0044] The frequency offset estimation units 891A, 891B, and 891C are mechanisms for calculating a frequency offset estimation value from the autocorrelation of the preamble signal. As a calculation method, a conventional method can be used, and the frequency offset estimation value is expressed by the following formula. Here, Δfest is the frequency offset estimation value, τ is the symbol interval of the autocorrelation, atan is the arctangent, M is the autocorrelation length, x(i) is the preamble signal, * is the complex conjugate, and Rm is the autocorrelation (where τ and M are integers).

number

[0045] Each of the frequency offset estimation units 891A, 891B, and 891C calculates a frequency offset estimation value using a different autocorrelation symbol interval τ (τ1, τ2, τ3). This enables each of the frequency offset estimation units 891A, 891B, and 891C to estimate a frequency offset with different estimation accuracy and estimable range.

[0046] Each symbol interval τ1 to τ3 is set so as to satisfy τ1<τ2<τ3. By setting in this manner, as shown in Fig. 5(A), the estimable range of the frequency offset in the first frequency offset estimator 891A is the widest, and the estimable range of the frequency offset gradually narrows as it progresses to the second frequency offset estimator 891B and the third frequency offset estimator 891C. In other words, as shown in Fig. 5(B), the estimation accuracy of the frequency offset gradually increases as it progresses from the first frequency offset estimator 891A to the subsequent second frequency offset estimator 891B and the third frequency offset estimator 891C (in Fig. 5(B), the "estimation error range" on the horizontal axis of the graph is a parameter representing the estimation accuracy). Therefore, a wide estimable range is ensured in the first frequency offset estimator 891A, and the estimation accuracy, which is in a trade-off relationship with the estimable range, can be ensured in the subsequent second frequency offset estimator 891B and third frequency offset estimator 891C, making it possible to achieve both a wide estimable range and high estimation accuracy.

[0047] Each of the symbol intervals τ1 to τ3 is determined, for example, as follows. To explain using the examples of Fig. 5(A) and (B), first, the number of symbols P_LEN of a target preamble signal is determined. Next, the value of the symbol interval τ1 of the first frequency offset estimation unit 891A is calculated from the following formula (1) so that the estimation possible range (±X1) of the first frequency offset estimation unit 891A is wider than the expected frequency offset. Here, Freq_est represents the estimation possible range. ±Freq_est[%]=100 / (2×τ)················································Formula (1)

[0048] Next, the expected CNR (Carrier to Noise Ratio) and the target estimation accuracy are determined, and the symbol interval τ3 and autocorrelation length M3 of the third frequency offset estimator 891C are obtained by simulation so that the target estimation accuracy (estimated error range) can be obtained. At this time, a combination that can make M3 as small as possible is searched for. Note that the estimated error range varies depending on various disturbance factors, and is calculated by simulation at the time of design after determining the three factors of the symbol interval τ, the autocorrelation length M, and the expected CNR, and is expressed as a percentage of the symbol frequency.

[0049] Next, the symbol interval τ3 of the third frequency offset estimator 891C is substituted into equation (1) to find the estimation possible range (±X3) of the third frequency offset estimator 891C.

[0050] Next, the autocorrelation length M1 of the first frequency offset estimation unit 891A is calculated from the following equation (2): Here, P_LEN represents the number of symbols of the preamble signal. M1= P_LEN - M3 - τ1············································Formula (2)

[0051] Next, a simulation is performed using the symbol interval τ1 and autocorrelation length M1 of the first frequency offset estimator 891A and an assumed CNR to obtain the estimation error range (±Y1) of the first frequency offset estimator 891A. Here, when the estimation error range (±Y1) of the first frequency offset estimator 891A is wider than the estimable range (±X3) of the third frequency offset estimator 891C, the second frequency offset estimator 891B is provided. In other words, when the estimation error range (±Y1) of the first frequency offset estimator 891A is narrower than the estimable range (±X3) of the third frequency offset estimator 891C, there is no need to provide the second frequency offset estimator 891B, i.e., it is sufficient to provide only two frequency offset estimators, which is the most preferable aspect.

[0052] When the second frequency offset estimator 891B is provided, the symbol interval τ2 of the second frequency offset estimator 891B is calculated so that the estimation range (±X2) of the second frequency offset estimator 891B is wider than the estimation error range (±Y1) of the first frequency offset estimator 891A.

[0053] If τ2>τ1+M1, the autocorrelation length M2 of the second frequency offset estimation unit 891B is calculated by simulation so as to satisfy the following conditions 1 and 2. Here, P_LEN represents the number of symbols of the preamble signal. Condition 1: The estimation error range (±Y1) of the first frequency offset estimator 891A is narrower than the estimable range (±X2) of the second frequency offset estimator 891B. Condition 2: P_LEN > τ1+M1+M2+M3...Equation (3)

[0054] On the other hand, if τ2≦τ1+M1, the autocorrelation length M2 of the second frequency offset estimation unit 891B is calculated from the following equation (4): Here, P_LEN represents the number of symbols of the preamble signal. M2= P_LEN - M3 - τ2··················································Formula (4)

[0055] Next, a simulation is performed using the symbol interval τ2 and autocorrelation length M2 of the second frequency offset estimator 891B and an assumed CNR to find the estimation error range (±Y2) of the second frequency offset estimator 891B. If the estimation error range (±Y2) of the second frequency offset estimator 891B becomes wider than the estimable range (±X3) of the third frequency offset estimator 891C, the number of frequency offset estimators (i.e., n) is increased (n≧4).

[0056] Of the frequency offset estimation means, the first frequency offset estimation unit 891A estimates the amount of frequency offset included in the received signal and calculates it as a first frequency offset estimation value. The second frequency offset estimation unit 891B estimates the amount of frequency offset included in the received signal after the first frequency offset correction has been provisionally performed, as described below, based on the first frequency offset estimation value calculated in the first frequency offset estimation unit 891A, and calculates it as a second frequency offset estimation value. The third frequency offset estimation unit 891C estimates the amount of frequency offset included in the received signal after the second frequency offset correction has been provisionally performed, as described below, based on the second frequency offset estimation value calculated in the second frequency offset estimation unit 891B, and calculates it as a third frequency offset estimation value.

[0057] The delay adjustment units 892A, 892B, and 893C are mechanisms for performing delay adjustment processing on received signals.

[0058] NCOs 893A to 893C receive the frequency offset estimation values ​​output from frequency offset estimators 891A, 891B, and 891C, convert the frequency offset estimation values ​​into complex numbers, and output the complex numbers.

[0059] The mixers 894A to 894C receive the delay-adjusted received signals output from the delay adjustment units 892A, 892B, and 893C, as well as the complex numbers output from the NCOs 893A to 893C, and perform complex multiplication on the delay-adjusted received signals using the complex numbers, and output the result.

[0060] Next, the operation of the AFC 89 in this embodiment will be described with reference to the flow chart of FIG.

[0061] The AFC 89 in this embodiment is characterized by including three frequency offset estimating units 891, namely, a first frequency offset estimating unit 891A, a second frequency offset estimating unit 891B, and a third frequency offset estimating unit 891C, and performing frequency offset estimation and correction three times. Specifically, first, the first frequency offset estimating unit 891A extracts a preamble signal from a received signal and calculates a first frequency offset estimated value from the autocorrelation of the preamble signal (step S1), the NCO 893A converts the first frequency offset estimated value into a complex number, and the mixer 894A performs a complex multiplication process based on the received signal delay-adjusted by the first delay adjusting unit 892A and the complex number output from the NCO 893A, thereby performing a tentative first frequency offset correction (step S2).

[0062] Next, in the second frequency offset estimation unit 891B, a preamble signal is extracted from the received signal after the first frequency offset correction, and a second frequency offset estimation value is calculated from the autocorrelation of the preamble signal (step 3). The second frequency offset estimation value is converted to a complex number in the NCO 893B, and a complex multiplication process is performed in the mixer 894B based on the received signal after the first frequency offset correction, which has been delay-adjusted by the second delay adjustment unit 892B, and the complex number output from the NCO 893B, thereby performing a provisional second frequency offset correction (step S4).

[0063] Finally, in the third frequency offset estimation unit 891C, a preamble signal is extracted from the received signal after the second frequency offset correction, and a third frequency offset estimation value is calculated from the autocorrelation of the preamble signal (step S5). In the NCO 893C, the first frequency offset estimation value, the second frequency estimation value, and the third frequency estimation value are added together and converted into a complex number. In the mixer 894C, a complex multiplication process is performed based on the received signal whose delay has been adjusted by the third delay adjustment unit 892C and the complex number output from the NCO 893C, thereby performing a final third frequency offset correction, and the corrected received signal is output to the receiving ROF 810 (step S6).

[0064] As described above, according to the wireless communication system 1 of this embodiment, a plurality of frequency offset estimators 891A, 891B, and 891C are provided, and the symbol intervals τ1, τ2, and τ3 of the autocorrelation used in each of the frequency offset estimators 891A, 891B, and 891C are set so as to gradually increase (τ1<τ2<τ3) to calculate the frequency offset estimation value, and the frequency offset correction is performed in a plurality of times based on each of the calculated frequency offset estimation values, so that a wide estimable range can be ensured in the first frequency offset estimator 891A, and the estimation accuracy, which is in a trade-off relationship with the estimable range, can be ensured in the second and subsequent frequency offset estimators 891B and 891C at the subsequent stage, making it possible to achieve both a wide estimable range and high estimation accuracy. Therefore, it becomes possible to ensure sufficient estimation accuracy even in a system in which the frequency offset increases.

[0065] (Embodiment 2) Next, a wireless communication system 1 according to a second embodiment using the wireless communication device 2 and the frequency offset correction method of the present invention will be described. The wireless communication system 1 according to this embodiment differs from the first embodiment in that, of the preamble signal included in the data frame, only the number of symbols required by the first frequency offset estimation unit 891A are added to the beginning of the data signal, and the remaining preamble signal is inserted at a predetermined position in the section consisting of the data signal. Note that, in the following, the same components as those in the wireless communication system 1 according to the first embodiment are denoted by the same reference numerals and detailed description thereof will be omitted.

[0066] In the first embodiment, in order to obtain a wide estimation range and high estimation accuracy, a plurality of frequency offset estimation units 891 are provided, and the frequency offset is estimated and corrected in a plurality of times. However, in the first embodiment, in order to obtain a desired estimation accuracy, the symbol interval τ of the preamble signal must be increased toward the latter stage, and a preamble signal with a longer number of symbols is required compared to the case where the frequency offset is estimated and corrected in one time. However, if such a preamble signal with a longer number of symbols is added to the beginning of a data signal as in the conventional method, there is a problem that the throughput decreases. Therefore, in this embodiment, instead of adding all the preamble signals included in the data frame to the beginning of the data signal, only the preamble signals with the number of symbols required by the first frequency offset estimation unit 891A are added to the beginning of the data signal, and the remaining (increased) preamble signals are inserted at a predetermined position in the section consisting of the data signal, thereby suppressing the increase in the preamble signals.

[0067] Fig. 7(A) shows a frame structure of a conventional data frame when used in the first embodiment. As shown in Fig. 7(A), if the wireless communication device 2 according to the first embodiment receives a conventional data frame and estimates and corrects the frequency offset, the number of symbols of the preamble signal added to the head of the data signal is (τ3+M3) symbols required for the third frequency offset estimation performed by the third frequency offset estimation unit 891C. Here, the number of symbols of the preamble signal required for the first frequency offset estimation performed by the first frequency offset estimation means 891A is (τ1+M1), and generally, (τ1+M1)<(τ3+M3). Therefore, in the above example, a long preamble signal is added to the front of the data signal, and the throughput is relatively reduced, compared to the case where the frequency offset is estimated and corrected in one go using the frequency offset estimation means 891A.

[0068] On the other hand, Fig. 7(B) shows the frame structure of a data frame used in this embodiment. As shown in Fig. 7(B), in the data frame used in the wireless communication device 2 according to this embodiment, only the preamble signal of (τ1+M1) symbols required by the first frequency offset estimation unit 891A is added to the beginning of the data signal, and the remaining preamble signals are inserted at predetermined positions in the section consisting of the data signal. Here, if the “predetermined position” is preferably a position (i.e., the τ3 symbol) away from the beginning of the data frame (the symbol interval required for the third frequency offset estimation), and the “remaining preamble signal” is preferably M3 symbols (the autocorrelation length required for the third frequency offset estimation), the third frequency offset estimation can be performed using a preamble signal of M3 symbols from the beginning (0) to the (M3−1)th symbol, and a preamble signal of M3 symbols from the τ3th symbol to the (τ3+M3−1)th symbol, which are τ3 symbols away from the beginning (note that τ3>τ1+M1), it is possible to suppress the number of increased symbols of the preamble signal to only the number of symbols (M3) of the autocorrelation length required for the third frequency offset estimation.

[0069] As described above, in the wireless communication system 1 according to this embodiment, when the frequency offset is estimated and corrected in multiple steps while gradually increasing the estimation accuracy using multiple frequency offset estimation units 891, the number of symbols of the preamble signal increases compared to the case where the frequency offset is estimated and corrected in one step. However, by adding only the number of symbols required by the first frequency offset estimation unit 891A to the beginning of the data signal and inserting the remaining increased amount of the preamble signal at a predetermined position in the section formed by the data signal, the increased number of symbols of the preamble signal can be suppressed, and therefore a decrease in throughput can be suppressed. As a result, it is possible to achieve three things at the same time: high estimation accuracy, a wide estimation range, and high throughput.

[0070] Although the embodiment of the present invention has been described above, the specific configuration is not limited to the above embodiment, and even if there are design changes and the like within the scope of the gist of the present invention, they are included in the present invention. [Explanation of symbols]

[0071] 1. Wireless communication systems 2. Wireless communication devices 3 Antennas 4 Radio circuit 5 Interface section 6. Transmitter 7 Duplexer 8 Receiving unit (receiving means) 81 Channel Filter 82 Mixer 83 Local Oscillator 84 Variable ATT 85 ADC 86 Quadrature detection section 87 Burst detection section 88 Timing control section 89 AFC (automatic frequency control means) 891 Frequency offset estimation unit (frequency offset estimation means) 891A First Frequency Offset Estimation Unit 891B Second Frequency Offset Estimation Unit 891C Third Frequency Offset Estimation Unit 892 Delay Adjustment Unit 892A First Delay Adjustment Unit 892B Second Delay Adjustment Unit 892C 3rd Delay Adjustment Unit 893, 893A, 893B, 893C NCO 894, 894A, 894B, 894C mixer 810 Received ROF 811 Symbol Regeneration Unit 812 APC 813 Separation section 814 Demapping section 815 AGC 816DAC DF Data Frame

Claims

1. a receiving means for receiving a data frame which has a preamble signal added to the beginning of the data signal and is transmitted in bursts, and outputting a received signal; and an automatic frequency control unit that calculates a frequency offset estimation value from the autocorrelation of the preamble signal and performs frequency offset correction of the received signal based on the frequency offset estimation value, The automatic frequency control means performing the frequency offset correction of the received signal n times based on n frequency offset estimation values ​​calculated by n frequency offset estimation means (n ​​is an integer equal to or greater than 2); Each of the n frequency offset estimation means calculates the frequency offset estimate value using a different autocorrelation symbol interval τ (τ1 to τn symbols), where τ is an integer; The symbol interval τ is set to satisfy the relationship τ1<τ2<τ3< ... <τn, Among the n frequency offset estimation means, a first frequency offset estimation means estimates a frequency offset amount included in the received signal, and second to nth frequency offset estimation means estimate a frequency offset amount included in the received signal after the frequency offset correction based on the frequency offset estimated value calculated by the immediately preceding frequency offset estimation means.

23. A wireless communication device comprising:

2. The preamble signal included in the data frame is only the number of symbols required by the first frequency offset estimation means is added to the beginning of the data signal, and the remaining symbols are inserted at predetermined positions in a section formed by the data signal; 2. The wireless communication device according to claim 1 .

3. receiving a data frame that has a preamble signal added to the beginning of the data signal and is transmitted in bursts, and outputting the received signal; A frequency offset correction method for calculating a frequency offset estimation value from an autocorrelation of the preamble signal, and correcting the frequency offset of the received signal based on the frequency offset estimation value, comprising: performing the frequency offset correction of the received signal n times based on n frequency offset estimation values ​​calculated in n frequency offset estimation steps (n is an integer equal to or greater than 2); In each of the n frequency offset estimation steps, a different symbol interval τ (τ1 to τn symbols) is used to calculate the frequency offset estimate, where τ is an integer; The symbol interval τ is set to satisfy the relationship τ1<τ2<τ3< ... <τn, Among the n frequency offset estimation steps, a first frequency offset estimation step estimates a frequency offset amount included in the received signal, and second to n frequency offset estimation steps estimate a frequency offset amount included in the received signal after the frequency offset correction based on the frequency offset estimated value calculated in the immediately preceding frequency offset estimation step.

2. A method for correcting a frequency offset comprising:

Images