Relay device and relay method

Abstract

The present invention provides a relay device and relay method that can accurately estimate transmission path characteristics. [Solution] The relay device according to the embodiment comprises a receiving unit that receives radio waves and obtains a received signal, a first calculation unit that calculates the transmission path characteristics of the radio wave transmission path based on the received signal and calculates a correction coefficient based on the transmission path characteristics, a second calculation unit that calculates the calculation time of the first calculation unit, and a correction unit that corrects the received signal based on the correction coefficient. The first calculation unit determines the correction coefficient based on the transmission path characteristics and the calculation time.

Description

[Technical Field]

[0001] Embodiments of the present invention relate to a relay device and a relay method. [Background technology]

[0002] To expand the communication area, a relay device is placed between the transmitting and receiving devices. Signal degradation occurs during transmission due to distortions in the characteristics of the signal transmission path. The relay device compensates for this signal degradation.

[0003] The relay device estimates the transmission path characteristics from the received signal, calculates the inverse characteristics of the transmission path, corrects the received signal based on the inverse characteristics, and reproduces the signal transmitted by the transmitting device. The relay device transmits this signal to the receiving device. As a result, the receiving device can receive the signal transmitted by the transmitting device.

[0004] Since estimating transmission path characteristics takes time, the received signal used for estimating transmission path characteristics and the received signal being corrected are different signals. In other words, the received signal at a certain point in time is corrected based on transmission path characteristics estimated based on a received signal at a point in time prior to that point. For this reason, if the estimation time is long, the relay device cannot accurately estimate the transmission path characteristics of the received signal during transmission. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Patent No. 4237423 [Non-patent literature]

[0006] [Non-Patent Document 1] Information and Communications Council, Information and Communications Technology Subcommittee, Broadcasting Systems Committee (77th Meeting), Report of the Working Group on the Advancement of Terrestrial Digital Broadcasting Systems (Document 77-3), April 28, 2023. [Overview of the Initiative] [Problems that the invention aims to solve]

[0007] The objective of the present invention is to provide a relay device and a relay method that can accurately estimate transmission path characteristics. [Means for solving the problem]

[0008] The relay device according to this embodiment comprises a receiving unit that receives radio waves and obtains a received signal, a first calculation unit that calculates the transmission path characteristics of the radio wave transmission path based on the received signal and calculates a correction coefficient based on the transmission path characteristics, a second calculation unit that calculates the calculation time of the first calculation unit, and a correction unit that corrects the received signal based on the correction coefficient. The first calculation unit determines the correction coefficient based on the transmission path characteristics and the calculation time. [Brief explanation of the drawing]

[0009] [Figure 1] A diagram illustrating an example of a system including an example of a relay device according to the first embodiment. [Figure 2] A diagram illustrating an example of a single Orthogonal Frequency Division Multiplexing (OFDM) segment transmitted by a transmitting device according to the first embodiment. [Figure 3] A diagram illustrating an example of the configuration of an OFDM frame according to the first embodiment. [Figure 4] A block diagram illustrating an example of a relay device according to the first embodiment. [Figure 5] A block diagram illustrating an example of a correction unit according to the first embodiment. [Figure 6] A diagram illustrating an example of a received signal from a relay device according to the first embodiment. [Figure 7] A diagram illustrating the degradation of the received signal due to frequency drift in the transmitting device according to the first embodiment. [Figure 8] A diagram illustrating the degradation of the received signal due to frequency drift in the transmitting device according to the first embodiment. [Figure 9]A diagram for explaining an example of a system including an example of a relay device according to the second embodiment. [Figure 10] A diagram for explaining an example of the learning of a relay device according to the second embodiment. [Figure 11] A diagram for explaining an example of a system including an example of a relay device according to the third embodiment. [Figure 12] A diagram for explaining an example of a system including an example of a relay device according to the fourth embodiment.

Embodiments for Carrying Out the Invention

[0010] Hereinafter, embodiments will be described with reference to the drawings. The following description illustrates devices and methods for embodying the technical idea of the embodiments, and the technical idea of the embodiments is not limited to the arrangement of the components described below. Modifications that can be easily conceived by those skilled in the art are naturally included in the scope of disclosure. In a plurality of drawings, corresponding elements may be denoted by the same reference numerals and redundant descriptions may be omitted. In some cases, a plurality of names may be given to a single element, but these are merely examples and do not deny the assignment of other names to these elements. Nor does it deny the assignment of other names to elements that do not have a plurality of names. "Connection" may include not only direct connection but also connection via other elements. Unless otherwise specified, an element may be a single element or a plurality of elements.

[0011] [First Embodiment] FIG. 1 is a diagram for explaining an example of a system including an example of a relay device 20 according to the first embodiment. Examples of the system include a broadcast system and a communication system. In the embodiment, a broadcast system is described as an example.

[0012] The system includes at least one (here, two) transmission devices 12, 14, at least one (here, one) reception device 16, and at least one (here, one) relay device 20. The transmission devices 12, 14 constitute a Single Frequency Network (SFN) system that transmits the same signal S at the same carrier frequency. However, there may be a deviation (Δf) in the carrier frequencies transmitted by the transmission devices 12, 14. When receiving the same signal from the transmission devices 12, 14 with a deviation in the carrier frequency, even if the deviation of one carrier frequency can be canceled by the automatic frequency control unit, the deviation of the other carrier frequency cannot be canceled. As a result, the output signal of the automatic frequency control unit includes a residual frequency component other than 0 (Hz). Due to this residual frequency component, the phase of the received signal changes with time. Since the phase of the received signal changes with time, even if the received signal is corrected using the past transmission path estimation value, there is a possibility that the received signal may deteriorate (details will be described later). The first embodiment compensates for this deterioration. One of the transmission devices 12, 14 is referred to as the master station, and the other is referred to as the slave station. Generally, the slave station is connected to the master station by wire.

[0013] The transmission devices 12, 14 modulate and transmit the signal S. In the case of a broadcast system, examples of the signal S transmitted by the transmission devices 12, 14 are video signals, audio signals, Transmission and Multiplexing Configuration Control (TMCC) signals, or Low latency Channel (Lch) signals. The video signal and the audio signal are the data to be transmitted. The TMCC signal is the control information necessary for reception. The Lch signal is low-latency data. The TMCC signal and the Lch signal are signals necessary for the reception of the video signal and the audio signal. The video signal and the audio signal are the data to be transmitted.

[0014] An example of the modulation scheme for transmitters 12 and 14 is OFDM modulation. OFDM modulation includes symbol mapping using carrier modulation schemes such as Binary Phase Shift Keying (BPSK), Differential BPSK (DBPSK), Quadrature PSK (QPSK), Differential QPSK (DQPSK), and Quadrature Amplitude Modulation (QAM), and carrier mapping the mapped symbols to a specific carrier.

[0015] Transmitting devices 12 and 14 upconvert the OFDM-modulated signal into a high-frequency (RF) signal and transmit it as a wireless signal.

[0016] The relay device 20 receives signals transmitted from the transmitters 12 and 14, processes the received signals, and transmits the processed signals as signals transmitted from the transmitters 12 and 14. The receiving device 16 receives the signals processed by the relay device 20.

[0017] Between the transmitter 12 and the relay device 20, there may be a direct wave path p1 and a reflected wave path p2. Generally, the existence of multiple paths is called multipath. Interference of radio waves due to this multipath degrades the transmission path characteristics of the transmission line. Similarly, between the transmitter 14 and the relay device 20, there may also be multipath with a direct wave path p3 and a reflected wave path p4. For the sake of explanation, we will assume that the relay device 20 and the receiver 16 are close together, and that a direct wave path p5 exists between them, but no reflected wave path exists. However, a reflected wave path may also exist between the relay device 20 and the receiver 16.

[0018] If multiple relay devices are in place, a signal transmitted by one relay device may be received by another relay device.

[0019] The relay device 20 compensates for the degradation of the transmission path characteristics of the transmission path by correcting the transmission path characteristics of the direct wave received signal and the reflected wave received from the transmitters 12 and 14. As a result, the relay device 20 obtains the signal S transmitted by the transmitters 12 and 14 (a signal with undegraded transmission path characteristics). The correction of the transmission path characteristics of the received signal is also called equalization. If the transmission path characteristics of the received signal are degraded due to the effects of multipath, there is a possibility that the transmitted signal may be received incorrectly. Equalization can reduce errors in the transmitted data.

[0020] Figure 2 illustrates an example of an OFDM segment transmitted by transmitters 12 and 14 according to the first embodiment. In a subframe section, the OFDM segment is composed of a scattered pilot (SP) signal, a boundary (BP) pilot signal, an Lch signal, and data. Blank carrier symbols are data carriers. The SP signal is a pilot signal for transmission path estimation. The symbols at the beginning and end of the subframe are called boundary symbols, and the pilot signal that is the demodulation reference for the transmission main symbol by synchronous modulation and is placed only at boundary symbols is called the BP signal. The pilot signal for broadband frequency synchronization and noise estimation, which is assigned to a specific subcarrier within the segment, is called the Lch signal. There may also be a continuous pilot (CP) signal that is the demodulation reference for the transmission main symbol by synchronous modulation. The SP signal is placed every few symbols at a specific carrier number, while the CP signal is placed at every symbol at a specific subcarrier.

[0021] Figure 3 is a diagram illustrating an example of the frame configuration according to the first embodiment. The horizontal axis represents time, and the vertical axis represents frequency.

[0022] A frame consists of a frame synchronization signal, a TMCC signal, and one or more subframes. In Figure 3, two subframes, subframe 1 and subframe 2, are transmitted via time multiplexing. Each subframe consists of multiple OFDM symbols with the same GI length and the same FFT size. Different subframes may have different GI lengths and different FFT sizes. Frames are transmitted sequentially in time. The scale of Figure 3 is not necessarily accurate.

[0023] Figure 4 is a block diagram illustrating an example of a relay device 20 according to the first embodiment. The relay device 20 includes a receiving antenna 32, a wireless unit (hereinafter referred to as the RF unit) 34, an analog-to-digital conversion unit (hereinafter referred to as the ADC unit) 36, a quadrature demodulation unit 38, an automatic frequency control unit (hereinafter referred to as the AFC unit) 40, a finite impulse response filter (hereinafter referred to as the FIR filter) 42, a quadrature modulation unit 44, a digital-to-analog conversion unit (hereinafter referred to as the DAC unit) 46, an RF unit 48, a transmitting antenna 50, a fast Fourier transform unit (hereinafter referred to as the FFT unit) 52, a transmission path estimation unit 54, an inverse characteristic calculation unit 56, an inverse fast Fourier transform unit (hereinafter referred to as the IFFT unit) 58, a correction unit 60, a TMCC demodulation unit 62, and a correction delay calculation unit 64. The receiving antenna 32, RF section 34, ADC section 36, quadrature demodulation section 38, and AFC section are also referred to as the receiving section. The FIR filter 42 is also referred to as the correction section. The quadrature modulation section 44, DAC section 46, RF section 48, and transmitting antenna 50 are also referred to as the transmitting section. The FFT section 52, transmission path estimation section 54, inverse characteristic calculation section 56, IFFT section 58, and correction section 60 are also referred to as the first calculation section. The TMCC demodulation section 62 and correction delay calculation section 64 are also referred to as the second calculation section.

[0024] The receiving antenna 32 receives signals transmitted from the transmitting devices 12 and 14 and outputs the received radio signals to the RF unit 34.

[0025] The RF unit 34 receives signals from the transmitters 12 and 14. The RF unit 34 down-converts the radio signals to signals in the intermediate frequency band (hereinafter referred to as the IF band). The IF band signals are supplied to the ADC unit 36. The RF unit may be divided into multiple parts. Examples of multiple parts include a down-converter unit, a filter unit, and an amplifier unit. The down-converter unit converts the RF frequency radio signals into analog IF band signals. The filter unit removes out-of-band signals. The amplifier unit amplifies the level of the IF band signals.

[0026] The ADC unit 36 ​​converts the analog signal in the IF band into a digital signal in the IF band. The digital signal in the IF band is supplied to the quadrature demodulation unit 38.

[0027] The quadrature demodulation unit 38 quadrature demodulates the IF band digital signal and converts it into a digital baseband signal for channel I and a digital baseband signal for channel Q. The converted digital baseband signals for channels I and Q are supplied to the AFC unit 40.

[0028] The AFC unit 40 controls the residual frequency of the carrier wave remaining in the input signal to be kept at 0 Hz. The output of the AFC unit 40 is supplied to the FIR filter unit 42, the FFT unit 52, and the TMCC demodulation unit 62.

[0029] The FFT unit 52 removes GI from the digital signal output from the AFC unit 40, performs an FFT, and converts the input signal into a frequency domain signal (referred to as the carrier signal). The GI length required to remove GI is input from the TMCC demodulation unit 62 (not shown). The FFT size required for the FFT processing is also input from the TMCC demodulation unit 62 (not shown). The carrier signal is supplied to the transmission path estimation unit 54.

[0030] The transmission path estimation unit 54 extracts a pilot signal, which is a known signal, from the carrier signals input from the FFT unit 52, and estimates the transmission path characteristics (vectors) of the transmission path based on the change in the pilot signal from the known signal. The pilot arrangement pattern necessary to extract the pilot signal is input from the TMCC demodulation unit 62 (not shown). The estimation results are supplied to the inverse characteristic calculation unit 56.

[0031] The inverse characteristic calculation unit 56 calculates the inverse characteristic (reciprocal of the vector) of the transmission line characteristics of the transmission line. The inverse characteristic is supplied to the IFFT unit 58.

[0032] The IFFT unit 58 performs IFFT processing on the inverse characteristics to convert them into a time-domain signal, which is then supplied to the correction unit 60. The IFFT size used for IFFT processing is the same as the FFT size used for FFT processing in the FFT unit 52.

[0033] During IFFT processing, the null carrier portion corresponding to the guard band may be padded with zeros, or it may be extrapolated in some way.

[0034] The TMCC demodulation unit 62 extracts the TMCC signal portion from the digital signal output from the AFC unit 40 and demodulates the TMCC signal. The TMCC signal includes information on the GI length of each subframe, the pilot placement pattern, and the FFT size. The time required for FFT and IFFT differs depending on their size. The larger the FFT size, the longer the time required for FFT processing. Similarly, the larger the IFFT size, the longer the time required for IFFT processing. The TMCC signal is supplied to the correction delay calculation unit 64.

[0035] The correction delay calculation unit 64 calculates the time required for correction from the FFT unit 52 to the correction unit 60 (referred to as the correction delay time) based on the FFT size information, GI length, and pilot pattern included in the TMCC signal. The correction delay calculation unit 64 calculates the time required in the FFT unit 52 based on the GI length and FFT size information. The correction delay calculation unit 64 calculates the time required in the transmission path estimation unit 54 based on the FFT size information and the pilot arrangement pattern. The correction delay calculation unit 64 calculates the time required in the inverse characteristic calculation unit 56 based on the FFT size information. The correction delay calculation unit 64 calculates the time required in the IFFT unit 58 based on the FFT size information. The time required in the correction unit 60 is constant and known. The correction delay calculation unit 64 calculates the sum of these required times to obtain the correction delay time. The correction delay time is supplied to the correction unit 60.

[0036] Furthermore, the correction delay calculation unit 64 may calculate the correction delay time based on the FFT size information, since the time required for the FFT unit 52 and the IFFT unit 58 is dominant among the time required for each unit. For example, the correction delay calculation unit 64 may calculate the correction delay time based on the time required for the FFT and IFFT using the FFT size information and a known constant time.

[0037] The correction unit 60 corrects the inverse characteristics supplied from the IFFT unit 58 based on the correction delay time and calculates the correction coefficient (also called the tap coefficient) of the FIR filter 42.

[0038] The FIR filter 42 performs a convolution operation on the output signal of the AFC unit 40 based on the tap coefficients, thereby correcting (equalizing) the output signal of the AFC unit 40. Since the tap coefficients are calculated based on the corrected inverse characteristics, equalizing the received signal distorted by multipath interference yields a received signal (digital baseband signal) with the multipath distortion corrected. The output of the FIR filter 42 is supplied to the quadrature modulation unit 44.

[0039] The quadrature modulation unit 44 quadrature modulates the received signal, which has been corrected for multipath distortion, and converts it into an IF band signal. The converted IF band signal is supplied to the DAC unit 46.

[0040] The DAC section 46 converts the output of the quadrature modulation section 44 into an analog signal in the IF band. The analog signal in the IF band is supplied to the RF section 48.

[0041] The RF unit 48 converts the signal output from the DAC unit 46 into an RF signal and outputs the RF signal to the transmitting antenna 50. The RF unit 48 may be divided into multiple parts. Examples of multiple parts include an upconverter unit, a filter unit, and an amplifier unit. The upconverter unit converts the IF band analog signal into an RF frequency radio signal. The filter unit removes out-of-band radio signals. The amplifier unit amplifies the level of the radio signal. The RF unit 48 transmits the signal to the receiving device 16. The transmitting antenna 50 radiates the radio signal.

[0042] Figure 5 is a block diagram illustrating an example of the correction unit 60. The correction unit 60 includes a path extraction unit 72, a polar coordinate transformation unit 74, an amplitude averaging unit 76, a correction processing unit 78, a phase difference calculation unit 80, and a phase difference averaging unit 82. Each of these units operates for each OFDM symbol.

[0043] The output of the IFFT unit 58 is supplied to the path extraction unit 72. The output of the IFFT unit 58 is a complex signal (a complex number with the value of Ich on the real axis and the value of Qch on the imaginary axis) per OFDM symbol, where n is the number of IFFT samples (n is an integer satisfying the following equation: n=2 m(where m is a natural number). This complex number is called a sample. Of these n samples, a samples (where a is an integer satisfying the following equation: 1 ≤ a ≤ n) are samples that cancel out the predetermined paths p1, p2, p3, and p4 in Figure 1. The remaining b samples (where b is an integer satisfying the following equation: b = n - a) are samples that arise due to the influence of noise, etc. Among these a samples, there may be some samples with a small level (magnitude of the complex number) and a level close to 0. Replacing the values ​​of these samples with a level close to 0 with 0 has almost no effect on the equalization performance. Also, the levels of the b samples that arise due to the influence of noise, etc. are often small. Therefore, from these n samples, samples above a certain level can be extracted, corrected, and the values ​​of the other samples can be replaced with 0. These can then be combined as the tap coefficients of the FIR filter section 42 to equalize the received signal.

[0044] The path extraction unit 72 extracts predetermined samples from the n input samples that are above a certain level, and outputs the complex number of the extracted sample and its sample number. The sample number is a number indicating which of the n samples input from the IFFT unit 58 it is, and is between 1 and n. The extracted sample is called a path. Each unit from the polar coordinate transformation unit 74 to the correction processing unit 78 performs processing for each path. At least one path output from the path extraction unit 72 is supplied to the polar coordinate transformation unit 74.

[0045] The path extraction unit 72 may also extract a predetermined number of c samples (where c is a natural number) from the n samples input from the IFFT unit 58, centering on the sample number of the sample with the highest level. In the above, it is not necessarily required to center on the sample number of the sample with the highest level; c samples may be extracted so as to include that sample number. Subsequently, as described above, predetermined samples higher than a certain level may be extracted, and the complex number of the extracted sample and its sample number may be output.

[0046] In the correction unit 60, sample number information is transmitted between each unit, and when it is output to the FIR filter unit 42, the sample number is associated with the tap number of the FIR filter unit 42. The tap number of the FIR filter unit 42 corresponding to the sample number that was not extracted by the path extraction unit is set to 0.

[0047] The polar coordinate transformation unit 74 converts the Ich and Qch values ​​of each path into a polar coordinate signal Aexp(jθ) and detects the amplitude A and phase θ. The amplitude is supplied to the amplitude averaging unit 76. The phase is supplied to the phase difference calculation unit 80.

[0048] The amplitude averaging unit 76 has a memory that stores the amplitudes of multiple past passes. The amplitude averaging unit 76 calculates a moving average of the amplitudes of multiple past passes. When the moving average is calculated, noise components are removed. The moving average of the pass amplitudes output from the amplitude averaging unit 76 is supplied to the correction processing unit 78.

[0049] The phase difference calculation unit 80 includes a memory that stores the phase of the previously received path. The phase difference calculation unit 80 calculates the phase difference between the phase of the previous path and the phase of the current path. The phase difference output from the phase difference calculation unit 80 is supplied to the phase difference averaging unit 82.

[0050] The phase difference averaging unit 82 also includes a memory for storing the phase differences of multiple past passes. The phase difference averaging unit 82 calculates a moving average of the phase differences of multiple past passes. The moving average of the phase differences output from the phase difference averaging unit 82 is supplied to the correction processing unit 78.

[0051] The correction processing unit 78 calculates the tap coefficients of the FIR filter unit 42 based on the output of the correction delay calculation unit 64, the output of the amplitude averaging unit 76, and the output of the phase difference averaging unit 82. The tap coefficients are then set in the FIR filter unit 42.

[0052] Details of the processing performed by the correction processing unit 78 will be described later.

[0053] Transmitter 12 in Figure 1 transmits a signal S to relay device 20 at carrier frequency f1 via direct wave path p1 and reflected wave path p2. Transmitter 14 transmits a signal S to relay device 20 at carrier frequency f1+Δf via direct wave path p3 and reflected wave path p4.

[0054] Figure 6 is a diagram illustrating an example of a signal received by the relay device 20. The relay device 20 receives path p1 at timing t1, path p2 at timing t2, path p3 at timing t3, and path p4 at timing t4.

[0055] Paths p1 and p2 have a carrier frequency f1, and paths p3 and p4 have a carrier frequency f1 + Δf.

[0056] The relay device 20 receives the direct and reflected waves from the transmitter 12 and the direct and reflected waves from the transmitter 14. In each process from antenna 32 to AFC unit 40 in Figure 4, the control is performed so that the residual frequency of the carrier wave remaining at the output of AFC unit 40 becomes 0 (Hz). Here, the carrier frequency f1 is canceled, and the residual frequencies of paths p1 and p2 are assumed to be 0 (Hz), and the residual frequencies of paths p3 and p4 are assumed to be Δf.

[0057] Note that the carrier frequency f1 + Δf / 2 can be canceled, and the residual frequencies of paths p1 and p2 can be -Δf / 2, and the residual frequencies of paths p3 and p4 can be Δf / 2, or they can be other values.

[0058] The relay device 20 equalizes the output of the AFC unit 40 to obtain a single signal S with carrier frequency f1. The relay device 20 transmits the signal S.

[0059] Figures 7 and 8 illustrate the degradation of the received signal due to the carrier frequency shift Δf of the transmitters 12 and 14. Figures 7 and 8 show the received baseband signal on the IQ plane.

[0060] Figure 7 illustrates an example of a reception point on the IQ plane of the pilot signal contained in the first symbol 1 of the received signal in the relay device 20. The reception point is represented by Aexp(jθ), where A represents amplitude and θ represents phase. Typically, the amplitude and phase of each path are different.

[0061] Figure 8 illustrates an example of a reception point on the IQ plane of the pilot signal contained in the nth symbol n of the received signal in the relay device 20. If the transmission path when symbol 1 is received is the same as the transmission path when symbol n is received, and there is no residual frequency remaining in the output signal of the AFC unit 40 (0 Hz), then the amplitude and phase of each path will not change. However, if the transmission path when symbol 1 is received is different from the transmission path when symbol n is received, the amplitude and phase of each path will change. Also, if there is a residual frequency remaining, the phase will change. Here, we will explain assuming that the transmission paths are the same. In reality, due to the influence of noise, it is rare for the amplitude and phase of each path to remain completely unchanged.

[0062] If there is a difference Δf in the carrier frequencies of transmitters 12 and 14, it is not possible to make the residual frequency of all paths zero in the relay device 20. If the residual frequency of each path received from transmitter 12 is set to 0 (Hz), the residual frequency of each path received from transmitter 14 will be Δf. Therefore, the phases of paths p3 and p4 included in symbol n are different from the phases of paths p3 and p4 included in symbol 1. Specifically, if the time difference between receiving symbol 1 and symbol n is Δt, then the phase of paths p3 and p4 of symbol n is rotated by 2πΔfΔt [rad] relative to paths p3 and p4 of symbol 1. In this case, even if the signal of symbol n is corrected based on tap coefficients that are based on the amplitude and phase of paths p1, p2, p3, and p4 estimated when receiving symbol 1 transmitted from transmitter 14, it cannot be correctly equalized. Note that the amplitude and phase of the reception point of the pilot signal included in symbol n transmitted from transmitter 12 are the same as the amplitude and phase of the reception point of the pilot signal included in symbol 1 transmitted from transmitter 12. The amplitude of the reception point of the pilot signal included in symbol n transmitted from transmitter 14 is not affected by the carrier frequency shift Δf and is the same as the amplitude of the reception point of the pilot signal included in symbol 1 transmitted from transmitter 14. However, it may change due to the influence of noise.

[0063] The correction processing unit 78 includes a memory for storing the moving average of amplitude and the moving average of phase difference. The correction processing unit 78 receives the correction delay time supplied from the correction delay calculation unit 64. The correction processing unit 78 predicts the amplitude at future timing 2 from the moving average of amplitude at past timing 1. By calculating the moving average, future amplitudes can be predicted based on slow fluctuations, ignoring fast fluctuations in amplitude. Fast fluctuations include fluctuations due to noise. The time difference between timing 1 and timing 2 is the correction delay time. The correction processing unit 78 predicts the phase at timing 2 from the moving average of phase difference at timing 1. By calculating the moving average, future amplitudes can be predicted based on slow fluctuations, ignoring fast fluctuations in phase difference. For example, if the correction delay time is t d , the amplitude average is A ave , the average phase difference is θave Let the phase at timing 1 be θ1, the amplitude at timing 2 be A2, the phase at timing 2 be θ2, and the time between symbols be t. s Then, the amplitude A2 and phase θ2 at timing 2 may be calculated as follows.

[0064] A2 = A ave θ2 = θ1 + θ ave ×t d / t s The correction processing unit 78 sets the predicted complex number as the tap coefficient of the FIR filter unit 42.

[0065] The FIR filter unit 42 performs a convolution operation on the received signal supplied from the AFC unit 40. From this, the FIR filter unit 42 can compensate for the characteristics of the received signal deteriorated by the transmission line characteristics and correctly equalize the received signal.

[0066] Each of the amplitude averaging unit 76 and the phase difference averaging unit 82 calculates the average value of a plurality of past inputs. The averaging process has the same effect as the process of a low-pass filter in the sense that it cuts off high-frequency components. Calculating the average value of more inputs has the same effect as setting the cut-off frequency of the low-pass filter lower. The number of inputs to be averaged may be changed according to the FFT size. Note that a low-pass filter may be realized by means different from the averaging process. In that case, the cut-off frequency of the low-pass filter may be changed according to the FFT size.

[0067] For taps where the level of the complex number predicted within the correction processing unit 78 is smaller than a predetermined threshold value, the correction processing unit 78 may use the complex number output from the IFFT unit 58 as the tap coefficient as it is without predicting future values and correcting the tap coefficient.

[0068] [Second Embodiment] Figure 9 is a diagram illustrating an example of a system including an example of a relay device 20 according to the second embodiment. The relay device 20 according to the second embodiment differs from the relay device 20 according to the first embodiment in that a storage unit 92 and a learning unit 94 are added. The storage unit 92 stores the tap coefficients output from the correction unit 60. The learning unit 94 receives the tap coefficients from the storage unit 92 and the input of the correction unit 60 (i.e., the time-domain signal representing the transmission path inverse characteristics output from the IFFT unit 58). The learning unit 94 may be configured as a neural network.

[0069] The learning unit 94 receives tap coefficients from the storage unit 92 and inputs the correction unit 60 (i.e., the time-domain signal representing the inverse characteristics of the transmission path, which is the output of the IFFT unit 58). The learning unit 94 learns the tap coefficients output from the correction unit 60 using the output of the IFFT unit 58 as training data. Once the learning unit 94 has learned the tap coefficients, it outputs the learned tap coefficients to the correction unit 60 for the output of the IFFT unit 58. The correction unit 60 sets the learned tap coefficients output from the learning unit 94 to the FIR filter 42.

[0070] Figure 10 is a diagram illustrating an example of the processing timing of the learning unit 94. The upper part of Figure 10 shows an image of the processing timing of the signal output from the AFC unit 40 at timing t11. The signal output from the AFC unit 40 is processed sequentially by the FFT unit 52, the transmission path estimation unit 54, the inverse characteristic calculation unit 56, and the IFFT unit 58. The output timing of the IFFT unit 58 is set to t12. The signal output from the IFFT unit 58 is processed by the correction unit 60. The output timing of the correction unit 60 is set to t13. The signal output from the correction unit 60 is input to the storage unit 92. The stored signal from the storage unit 92 is then input to the learning unit 94.

[0071] The lower part of Figure 10 shows an image of the processing timing of the signal output from the AFC unit 40 at timing t14. The signal output from the AFC unit 40 is processed sequentially by the FFT unit 52, the transmission path estimation unit 54, the inverse characteristic calculation unit 56, and the IFFT unit 58. The output timing of the IFFT unit 58 is set to t15. The signal output from the IFFT unit 58 is processed by the correction unit 60. The output timing of the correction unit 60 is set to t16.

[0072] If t13=t14, the data output from the IFFT unit 58 at timing t15 becomes the training data for the data input to the memory unit 92 at timing t14. The time t15-t13 is calculated by the correction delay calculation unit 64, and the correction delay calculation unit 64 may output the calculated time to the learning unit 94. The learning unit 94 learns the stored signal of the memory unit 92 based on the training data selected based on this time.

[0073] Note that since each part from the FFT section 52 to the correction section 60 processes on a symbol-by-symbol basis, t13 may not always equal t14. In such cases, the learning section 94 may learn by inputting data from the memory section 92 that makes t13 and t14 as close as possible. In that case, it may learn by assuming t13=t14, or by considering the difference between t13 and t14.

[0074] According to the relay device of the second embodiment, the relationship between the transmission path inverse characteristics and the tap coefficient can be learned. Since the tap coefficient can be determined from the transmission path inverse characteristics based on the learning results, the tap coefficient of the FIR filter 42 that can correctly equalize the received signal can be determined.

[0075] [Third Embodiment] In the first and second embodiments, the tap coefficient was calculated and then corrected by the correction unit 60. In other words, the correction was applied to the transmission line correction value. In the third and fourth embodiments, a method of correcting the transmission line estimated value will be described.

[0076] Figure 11 is a diagram illustrating an example of a system including an example of a relay device 20 according to the third embodiment. The third embodiment is a modification of the first embodiment. In the third embodiment, the output of the transmission path estimation unit 54 is input to the IFFT unit 58. The output of the IFFT unit 58 is input to the correction unit 60. The output of the correction unit 60 is input to the FFT unit 96. The output of the FFT unit 96 is input to the inverse characteristic calculation unit 56. The output of the inverse characteristic calculation unit 56 is input to the IFFT unit 98. The output of the IFFT unit 98 is input to the FIR filter 42.

[0077] The correction unit 60 receives a time-domain signal representing past transmission path characteristics and estimates the transmission path characteristics of the signal that the FIR filter 42 equalizes from the past transmission path characteristics. The correction unit 60 in the first and second embodiments estimates the inverse characteristics of the transmission path characteristics. The correction unit 60 in the third embodiment differs from the correction unit 60 in the first and second embodiments in that the target of estimation is the transmission path characteristics. The prediction processing of the correction unit 60 in the third embodiment is the same as the processing in the first and second embodiments.

[0078] The FFT unit 96 converts the time-domain signal representing the estimated transmission path characteristics output from the correction unit 60 into a frequency-domain signal.

[0079] The inverse characteristic calculation unit 56 calculates the inverse characteristics of the transmission path from the frequency domain signal representing the transmission path characteristics output from the FFT unit 96.

[0080] The IFFT unit 98 converts the frequency domain signal representing the transmission path inverse characteristics output from the inverse characteristics calculation unit 56 into a time domain signal, and sets the conversion result as the tap coefficient of the FIR filter 42.

[0081] The correction delay calculation unit 64 calculates the correction delay time from the FFT unit 52 to the correction unit 60 based on the FFT size information, GI length, and pilot pattern included in the TMCC signal. The correction delay calculation unit 64 calculates the required time in the FFT unit 52 based on the GI length and FFT size information. The correction delay calculation unit 64 calculates the required time in the transmission path estimation unit 54 based on the FFT size information and the pilot placement pattern. The correction delay calculation unit 64 calculates the required time in the IFFT unit 58 based on the FFT size information. The required time in the correction unit 60 is constant and known. The correction delay calculation unit 64 calculates the sum of these required times to obtain the correction delay time. The correction delay time is supplied to the correction unit 60.

[0082] Similar to the first embodiment, the correction delay calculation unit 64 may calculate the correction delay time based on the FFT size information. For example, the correction delay calculation unit 64 may calculate the correction delay time based on the FFT and IFFT time required times and a known constant time based on the FFT size information.

[0083] The relay device according to the third embodiment also provides the same effects and advantages as the first embodiment.

[0084] [Fourth Embodiment] Figure 12 is a diagram illustrating an example of a system including an example of a relay device 20 according to the fourth embodiment. The fourth embodiment is a modification of the third embodiment. The relay device 20 according to the fourth embodiment differs from the relay device 20 according to the third embodiment in that a storage unit 92 and a learning unit 94 are added. The storage unit 92 stores the tap coefficients output from the correction unit 60. The learning unit 94 receives the tap coefficients from the storage unit 92 and the input of the correction unit 60 (i.e., a time-domain signal representing the transmission path characteristics output from the IFFT unit 58). The learning unit 94 may be configured as a neural network.

[0085] The learning unit 94 receives tap coefficients from the storage unit 92 and inputs the correction unit 60 (i.e., the time-domain signal representing the transmission path characteristics, which is the output of the IFFT unit 58). The learning unit 94 learns the time-domain signal representing the transmission path characteristics output from the correction unit 60, using the output of the IFFT unit 58 as training data. The learned learning unit 94 outputs the learned time-domain signal representing the transmission path characteristics to the correction unit 60 in relation to the output of the IFFT unit 58. The correction unit 60 outputs the signal supplied by the learning unit 94 to the FFT unit 96.

[0086] According to the relay device of the fourth embodiment, the tap coefficients of the FIR filter 42 that can correctly equalize the received signal can be determined, similar to the second embodiment.

[0087] In the above description, an example is given in which each functional unit included in the relay device 20 is configured with dedicated hardware. However, at least one of these functional units may be implemented in software by at least one processing unit. For example, the processing unit may be realized by one or more processing circuits such as a CPU, microprocessor, GPU (Graphics Processing Unit), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), or a combination thereof. Alternatively, the processing unit may be realized by an information processing device such as a computer, a computer system configured by multiple computers or servers communicating with each other via a network, or a PC cluster in which multiple computers cooperate to perform information processing.

[0088] It should be noted that the present invention is not limited to the embodiments described above, and the components can be modified and implemented in practice without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining the multiple components disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiments. Moreover, components from different embodiments may be appropriately combined. [Explanation of Symbols]

[0089] 20…Relay device, 38…Quaternary demodulation unit, 42…FIR filter, 44…Quaternary modulation unit, 52…FFT unit, 54…Transmission path estimation unit, 56…Inverse characteristic calculation unit, 58…IFFT unit, 60…Correction unit, 62…TMCC demodulation unit, 64…Correction delay calculation unit

Claims

1. A receiving unit that receives radio waves and obtains a received signal, A first calculation unit calculates the transmission path characteristics of the radio wave transmission path based on the received signal and calculates a correction coefficient based on the transmission path characteristics. A second calculation unit that calculates the calculation time of the first calculation unit, The system comprises a correction unit that corrects the received signal based on the correction coefficient, The first calculation unit is a relay device that determines the correction coefficient based on the transmission line characteristics and the calculation time.

2. The first calculation unit is, A first conversion unit that converts the received signal into a signal in the frequency domain, A third calculation unit that calculates the transmission line characteristics from the signal in the frequency domain, A fourth calculation unit for calculating the inverse characteristics of the transmission line characteristics, A second conversion unit that converts the aforementioned inverse characteristics into a time-domain signal, The system comprises a processing unit that determines the time-domain signal as a correction coefficient, The second calculation unit calculates the total time of the conversion time of the first conversion unit, the processing time of the third calculation unit, the processing time of the fourth calculation unit, the conversion time of the second conversion unit, and the processing time of the processing unit as the calculation time. The first calculation unit estimates the correction coefficient for the second timing, which is a calculation time from the first timing, based on the correction coefficient for the first timing obtained by the processing unit. The relay device according to claim 1, wherein the correction unit corrects the received signal based on the estimated correction coefficient.

3. The first transformation unit comprises a Fourier transform unit, The second transformation unit comprises an inverse Fourier transform unit, The second calculation unit is, Based on the size of the Fourier transform in the Fourier transform unit, the conversion time of the first transform unit is determined. The relay device according to claim 2, wherein the conversion time of the second transform unit is determined based on the size of the inverse Fourier transform of the inverse Fourier transform unit.

4. The received signal includes first information representing the size of the Fourier transform and second information representing the size of the inverse Fourier transform. The relay device according to claim 3, wherein the second calculation unit determines the conversion time of the first conversion unit based on the first information and determines the conversion time of the second conversion unit based on the second information.

5. The relay device according to claim 1, wherein the correction unit comprises a finite impulse response filter that performs a convolution operation on the received signal using the correction coefficient as a tap coefficient.

6. The relay device according to claim 1, wherein the first calculation unit determines the correction coefficient based on the phase component of the transmission line characteristics and the calculation time.

7. The relay device according to claim 1, wherein the first calculation unit determines the correction coefficient based on the average of the changes in the phase component of the transmission line characteristics and the calculation time.

8. The relay device according to claim 1, wherein the first calculation unit determines the average time of the change in the phase component of the transmission line characteristics based on the calculation time and the correction coefficient based on the calculation time.

9. The system further comprises a learning unit that stores the output of the first calculation unit, receives the output of the first calculation unit at a first timing and the input of the first calculation unit at a first timing, and learns the output of the first calculation unit using the input of the first calculation unit as training data. The relay device according to claim 1, wherein the correction unit corrects the received signal based on the learning results of the learning unit.

10. The first calculation unit is, A first conversion unit that converts the received signal into a signal in the frequency domain, A third calculation unit that calculates the transmission line characteristics from the signal in the frequency domain, A second conversion unit that converts the transmission path characteristics into a time-domain signal, A processing unit that determines the output of the second conversion unit as a correction coefficient, A third conversion unit converts the correction coefficient output from the second conversion unit into a frequency domain signal, A fourth calculation unit calculates the reciprocal of the output of the third conversion unit, The system comprises a fourth conversion unit that converts the output of the fourth calculation unit into a time-domain signal, The relay device according to claim 1, wherein the correction unit corrects the received signal using the output of the fourth conversion unit as the correction coefficient.

11. The system further comprises a learning unit that stores the output of the fourth conversion unit, receives the output of the fourth conversion unit at a first timing and the input of the processing unit at a first timing, and learns the output of the fourth conversion unit using the input of the processing unit as training data. The relay device according to claim 10, wherein the processing unit determines the learning result of the learning unit as the correction coefficient.

12. By receiving radio waves and obtaining the received signal, Based on the received signal, the transmission path characteristics of the radio wave transmission path are calculated. The transmission line characteristics are estimated, The estimated time for the transmission line characteristics is calculated, A correction coefficient is determined based on the transmission path characteristics and the estimated time. A relay method for correcting the received signal based on the correction coefficient.

Images