Receiving station, wireless communication system, and wireless communication method

Abstract

To provide a reception station, a radio communication system, and a radio communication method capable of accurately receiving a signal transmitted by a moving transmission station.SOLUTION: A reception station includes: a reception unit for receiving a control signal transmitted in a predetermined frequency band by a moving transmission station, the control signal including a known signal unique to the station, and a data signal transmitted in another frequency band in synchronization with the control signal by the transmission station; a division filter unit that divides the known signal received by the reception unit into signals in a plurality of sections and calculates mutual correlation values of the respective signals obtained through division; an error detection unit that, on the basis of each of the mutual correlation values calculated by the division filter unit, calculates a frequency error in the control signal; a channel estimation unit that, on the basis of the frequency error detected by the error detection unit, estimates a channel of the data signal; and a machine learning unit that performs machine learning based on the frequency error, the mutual correlation values, and positional information on the transmission station. The error detection unit detects the frequency error in the control signal by reflecting a result of the machine learning performed by the machine learning unit.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to a receiving station, a wireless communication system, and a wireless communication method. [Background technology]

[0002] A low Earth orbit (LEO) satellite system is a wireless communication system that uses satellites in a lower orbit (800-2000km) than geostationary satellites. This low Earth orbit satellite system has the advantage of lower latency and attenuation in communications at terminal stations compared to geostationary Earth orbit (GEO) systems at an altitude of approximately 36,000km.

[0003] In addition, a terminal station with multiple receiving antennas can improve communication capacity by using MIMO (Multi-Input Multi-Output) transmission using signals from multiple LEO satellites compared to receiving signals from a single LEO satellite.

[0004] The signals received by terminal stations are affected by the Doppler shift caused by the movement of LEO satellites. Therefore, in MIMO transmission using multiple LEO satellites, it is necessary to individually estimate the Doppler shift of each LEO satellite and assign a unique frequency band to each LEO satellite so that the control signals specific to each LEO satellite can be separated from each other.

[0005] However, allocating a unique frequency band to each LEO satellite reduces the system bandwidth, lowering the communication capacity of the LEO-MIMO system.

[0006] For example, in order to improve the frequency utilization efficiency of a low-earth-orbit satellite system, a downlink control method is known that performs dynamic MIMO transmission and frequency allocation from multiple LEO satellites in a Doppler shift environment (see, for example, Non-Patent Document 1).

[0007] In this system, the control signals and data signals, which are downlink signals from LEO satellites, are allocated to different frequencies. Furthermore, the control signals of multiple LEO satellites that can be received by terminal stations within the service area are assigned to different frequency bands.

[0008] The terminal station synchronizes using control signals from multiple LEO satellites and determines the number of LEO satellites it can receive from.The terminal station then performs channel estimation using the control signals and, based on the estimated information for all LEO satellites it can receive, notifies the base station via an uplink signal of the information on the multiple LEO satellites that are targets for MIMO transmission.

[0009] The base station, upon receiving the notification from the terminal station, performs MIMO transmission to the terminal station using the multiple LEO satellites. At this time, the base station sets a unique MIMO band as the data signal band based on the number of satellites that each terminal station can receive, and begins transmission to each LEO satellite.

[0010] Furthermore, the base station notifies each LEO satellite of the frequency, bandwidth, and satellite information of the data signal of that satellite by using a control signal for that satellite. Note that the control signal and data signal transmitted via each LEO satellite are generated from the same baseband signal and are synchronized.

[0011] The terminal station receives the data signal, performs equalization processing, and demodulates the data according to the notified information. More specifically, the terminal station uses the channel estimation information of the control signal to calculate a weight matrix for equalizing the MIMO signal, and separates and demodulates the signals of each LEO satellite.

[0012] In this downlink control method, to prevent erroneous detection of received signals due to frequency asynchronousity caused by the different Doppler shifts of each LEO satellite, the control signals of LEO satellites that can be received by terminal stations within the service area are assigned to different frequency bands.

[0013] However, as the number of LEO satellites increases, the number of control signals also increases, raising concerns that frequency utilization efficiency will deteriorate. In other words, even with MIMO transmission using multiple LEO satellites, the increase in control signals may prevent the transmission capacity from improving.

[0014] Furthermore, in order to improve the frequency utilization efficiency of low-earth-orbit satellite systems even when the number of LEO satellites increases, a downlink control method is known in which dynamic MIMO transmission and frequency allocation are performed from multiple LEO satellites in a Doppler shift environment, and control signals are allocated to the same frequency band between LEO satellites (see, for example, Non-Patent Document 2).

[0015] Here, a receiver (terminal station) that receives a control signal superimposed on the same frequency band detects each control signal by applying machine learning to the received signal. [Prior art documents] [Non-patent literature]

[0016] [Non-Patent Document 1] D. Goto, H. Shibayama, F. Yamashita, T. Yamazato, "LEO-MIMO Satellite Systems for High Capacity Transmission", IEEE GLOBECOM 2018, Dec 2018. [Non-patent document 2] R. Okema, D. Goto, T. Yamazato, F. Yamashita, and T. Shibayama, "Detection Method of Two-Wave LEO-MIMO Control Signals Using Deep Learning," Institute of Electronics, Information and Communication Engineers, May 2019, vol. 119, no. 27, pp. 13-17 [Non-patent document 3] I.Ali, N.Al-Dhahir and JEHershey, "Doppler characterization for LEO satellites", IEEE Transactions on Communications, March 1998, vol.46, no.3, pp.309-313 [Non-patent document 4] R. Okema, D. Goto, T. Yamazato, F. Yamashita, H. Shibayama, "Deep Learning Detection for superimposed control signal in LEO-MIMO", IEEE GLOBECOM 2020, Dec 2020. [Non-Patent Document 5] Masanobu Mizutani, Masaaki Katayama, Takaya Yamazato, Akira Ogawa, "Parallel Acquisition of Spread Spectrum Signals with Carrier Frequency Offset," IEICE Transactions on Electronics, Information and Communication Engineers, Vol. J79-A, No. 1, pp. 98-104, January 1996. Summary of the Invention [Problem to be solved by the invention]

[0017] However, since the above-described downlink control method can only perform local frequency estimation, it may be difficult to improve the estimation accuracy.

[0018] For example, the fluctuation of a satellite's Doppler frequency is not a characteristic that is uncorrelated on the time axis, but is largely determined by the satellite's orbit and the position of the ground station, and is defined by a trigonometric function with time as a variable (see, for example, Non-Patent Document 3).

[0019] On the other hand, errors occur in the Doppler frequency due to variations in the Earth's orbit caused by the gravitational forces of the sun and moon (the three-body problem), variations in the satellite's orbit and the position of the ground station, and frequency variations in the transmitter itself. Therefore, it is difficult to sufficiently improve the accuracy of channel estimation using only peripheral information such as orbital information and positional information.

[0020] Furthermore, a technique is known in which a received signal is extracted into multiple time intervals, and a spectrogram obtained by performing a short-time Fourier transform is used as an input layer to perform machine learning using a neural network (CNN: Convolutional Neural Network), and estimation is performed based on the local Doppler frequency obtained from the waveform and the amount of change before and after, thereby further reducing estimation errors (see, for example, Non-Patent Document 4).

[0021] However, the technology described in Non-Patent Document 4 had the risk that as uncertainties in the waveform increased, for example, when the number of satellites was increased, the estimation accuracy would be limited using only the information obtained from the spectrogram.

[0022] Furthermore, in the spread spectrum technology, a technology has been proposed in which the despread sequence on the receiving side is divided into multiple sections to improve the accuracy of synchronization acquisition (see, for example, Non-Patent Document 5).

[0023] However, the technology described in Non-Patent Document 5 is limited to the study under the presence of a single transmitter, and does not go as far as to consider frequency error estimation.

[0024] An object of the present invention is to provide a receiving station, a wireless communication system, and a wireless communication method that can accurately receive a signal transmitted from a moving transmitting station. [Means for solving the problem]

[0025] A receiving station according to one aspect of the present invention includes a receiving unit that receives a control signal including a known signal specific to the transmitting station, which is transmitted by a moving transmitting station in a predetermined frequency band, and a data signal that is transmitted by the transmitting station in another frequency band in synchronization with the control signal; a dividing filter unit that divides the known signal received by the receiving unit into a plurality of sections and calculates a cross-correlation value for each of the divided signals; a machine learning unit that performs machine learning based on each of the cross-correlation values calculated by the dividing filter unit and position information of a moving transmitting station; The cross-correlation values calculated by the division filter unit are and reflecting the results of machine learning performed by the machine learning unit in the position information of the moving transmitting station. and an error detection unit that detects a frequency error in the control signal. and With do It is characterized by:

[0026] Furthermore, a wireless communication system according to one aspect of the present invention is a wireless communication system including a receiving station that receives a signal transmitted by a moving transmitting station, the receiving station including a receiving unit that receives a control signal that the transmitting station transmits in a predetermined frequency band and that includes a known signal specific to the receiving station, and a data signal that the transmitting station transmits in another frequency band in synchronization with the control signal; a dividing filter unit that divides the known signal received by the receiving unit into a plurality of sections and calculates a cross-correlation value for each of the divided signals; a machine learning unit that performs machine learning based on each of the cross-correlation values calculated by the dividing filter unit and position information of a moving transmitting station; The cross-correlation values calculated by the division filter unit are and reflecting the results of machine learning performed by the machine learning unit in the position information of the moving transmitting station. an error detection unit that detects a frequency error of the control signal; and a channel estimation unit that estimates a channel of the data signal based on the frequency error detected by the error detection unit. and With do It is characterized by:

[0027] Furthermore, a wireless communication method according to one aspect of the present invention includes a receiving step of receiving a control signal including a known signal specific to the mobile transmitting station, which is transmitted in a predetermined frequency band by the mobile transmitting station, and a data signal synchronized with the control signal and transmitted in another frequency band by the transmitting station; a division filtering step of dividing the received known signal into a plurality of sections and calculating a cross-correlation value for each of the divided signals; a machine learning step for performing machine learning based on the calculated cross-correlation values and the position information of the moving transmitting station; and a division filtering step for performing machine learning based on the calculated cross-correlation values and the position information of the moving transmitting station. The calculated cross-correlation values are and reflecting the results of machine learning performed by the machine learning step in the position information of the moving transmitting station. an error detection step of detecting a frequency error of the control signal; and a channel estimation step of estimating a channel of the data signal based on the detected frequency error. and Including nothing It is characterized by: [Effects of the Invention]

[0028] According to the present invention, it is possible to receive a signal transmitted from a moving transmitting station with high accuracy. [Brief explanation of the drawings]

[0029] [Figure 1] 1 is a diagram illustrating an example of an overview of a wireless communication system according to an embodiment. [Figure 2] FIG. 2 is a functional block diagram illustrating functions of a base station. [Figure 3] FIG. 10 is a diagram illustrating a frame format of an uplink control signal. [Figure 4] FIG. 2 is a diagram illustrating a frame format of a downstream control signal. [Figure 5] FIG. 2 is a functional block diagram illustrating functions of a relay station. [Figure 6] FIG. 2 is a functional block diagram illustrating functions of a terminal station. [Figure 7] FIG. 10 is a diagram schematically illustrating a state in which a known signal included in a control signal is divided. [Figure 8] FIG. 2 is a diagram illustrating a schematic example of machine learning performed by a machine learning unit. [Figure 9] FIG. 1 is a diagram illustrating an example of the operation of a wireless communication system. DETAILED DESCRIPTION OF THE INVENTION

[0030] An embodiment of a wireless communication system will be described below with reference to the drawings. Fig. 1 is a diagram illustrating an overview of a wireless communication system 100 according to an embodiment. The wireless communication system 100 includes, for example, a base station 10, multiple relay stations A, B, C,...X, and multiple terminal stations 1, 2, and 3. The base station 10, the relay stations A, B, C,...X, and the terminal stations 1, 2, and 3 each function as a transmitting station that transmits signals and as a receiving station that receives signals.

[0031] Each of relay stations A, B, C, X is, for example, an artificial satellite such as a LEO satellite or a mobile station moving on the ground, and while moving, relays signals between base station 10 and terminal stations 1, 2, and 3. Here, it is assumed that relay stations A, B, and C are located within a range where they can relay signals to terminal stations located within a predetermined service area.

[0032] The base station 10 is, for example, a ground station located on the ground, and multiplexes and transmits signals to multiple terminal stations located within a predetermined service area using the FDMA (Frequency-Division Multiple Access) method. The base station 10 also performs downlink multiple access, performing MIMO (spatial multiplexing) transmission to terminal stations that can perform equalization and separation for each frequency band.

[0033] At this time, the base station 10 knows the location (location information) of each of the relay stations A, B, C,...X, and identifies the relay stations from which the terminal station can receive signals within a specified service area based on the location of each relay station.

[0034] If the relay stations are artificial satellites moving on a predetermined orbit, the base station 10 grasps the current location of each relay station based on information indicating the orbit and the time. If the relay stations are mobile stations moving on the ground, the base station 10 grasps the current location of each relay station by obtaining location information acquired by the relay stations using a GPS (Global Positioning System) or the like via a terrestrial network or the like.

[0035] Terminal stations 1, 2, and 3 are each equipped with one or more antennas and are terminals that receive communication services within a predetermined service area. Here, for example, terminal station 1 has one antenna, terminal station 2 has two antennas, and terminal station 3 has three antennas. In other words, MIMO transmission is possible for terminal stations 2 and 3 that have two or more antennas.

[0036] The wireless communication system 100 will be described below using a downlink as an example in which relay stations A, B, and C relay downstream signals transmitted by the base station 10 to terminal stations 1, 2, and 3.

[0037] The base station 10 transmits control signals A, B, and C allocated to the same frequency band to the terminal stations 1, 2, and 3 via relay stations A, B, and C whose signals the terminal stations 1, 2, and 3 can receive within a predetermined service area.

[0038] Furthermore, base station 10 allocates frequency bands for data signals to the request relay stations of the terminal stations indicated by the uplink control signals transmitted by terminal stations 1, 2, and 3. Then, base station 10 notifies terminal stations 1, 2, and 3 of information indicating the allocated frequency bands of data signals by downlink control signals (control signals A, B, and C), and starts MIMO transmission via the corresponding relay stations (request relay stations).

[0039] It should be noted that the control signal and the data signal are generated from the same baseband signal and are therefore synchronized with each other. In other words, it is assumed that the data signal can also be synchronized by synchronizing using the control signal.

[0040] The control signal includes a known signal specific to each relay station (known signal specific to each relay station) that is transmitted by each relay station A, B, and C in a predetermined frequency band.

[0041] Relay stations A, B, and C receive signals transmitted by base station 10, and by performing frequency conversion and amplification, convert the signals into a state that can be received by terminal stations 1, 2, and 3, and then relay and transmit the converted signals. Specifically, relay stations A, B, and C allocate control signals and data signals, which are downlink signals, to different frequency bands. The frequency band of the control signal is fixed. Terminal stations 1, 2, and 3 located within the service area are assumed to know the frequency band (center frequency, etc.) of the control signal. The frequency bands W1, W2, and W3 of the data signals are dynamically changeable.

[0042] Here, there is a one-to-one correspondence between relay stations A, B, and C and the alphabets of control signals A, B, and C and data signals A, B, and C. For example, relay station A transmits control signal A in a predetermined frequency band and transmits data signal A in frequency bands W1, W2, and W3. Relay station B transmits control signal B in a predetermined frequency band and transmits data signal B in frequency bands W2 and W3. Relay station C transmits control signal C in a predetermined frequency band and transmits data signal C in frequency band W3.

[0043] In the example shown in Fig. 1, a data signal addressed to terminal station 1 is transmitted from relay station A in frequency band W1. A data signal addressed to terminal station 2 is transmitted from relay stations A and B in frequency band W2 using MIMO. A data signal addressed to terminal station 3 is transmitted from relay stations A, B, and C in frequency band W3 using MIMO.

[0044] The terminal stations 1, 2, and 3 receive the control signals A, B, and C that they can receive from the relay stations A, B, and C, perform synchronization processing, and grasp the number of relay stations that they can receive.

[0045] Furthermore, terminal stations 1, 2, and 3 estimate the channel of the data signal between the relay station and each terminal station using the synchronized control signal, and based on information (channel information) indicating the receivable relay stations and channels, notify base station 10 via any uplink of information indicating the relay station (requesting relay station) that requests MIMO transmission.

[0046] Furthermore, terminal stations 1, 2, and 3 utilize the synchronization between the control signal and the data signal to receive and equalize the data signal according to the notified information, and demodulate the data. Furthermore, terminal stations 1, 2, and 3 calculate a weight matrix for equalizing the MIMO signal from the channel information of the control signal, separate the spatially multiplexed signals of each relay station, and demodulate the data.

[0047] In addition, when the wireless communication is in a line-of-sight environment and has channel characteristics with little frequency selectivity, such as when relay stations A, B, and C are artificial satellites, synchronization and channel information by control signals can be reflected in the equalization of the data signal.

[0048] Furthermore, by setting the guard band (G shown in FIG. 1) between each FDMA signal to a value twice the maximum Doppler frequency Δfmax expected in wireless communication system 100, it becomes possible to avoid inter-carrier interference.

[0049] Next, specific examples of the functions possessed by the base station 10, the relay stations A, B, C, . . . X, and the terminal stations 1, 2, and 3 will be described.

[0050] Fig. 2 is a functional block diagram illustrating functions of base station 10. As shown in Fig. 2, base station 10 includes a receiver 11, a terminal station detector 12, a relay station detector 13, a selector 14, a position calculator 15, a band allocator 16, a signal generator 17, and a transmitter 18.

[0051] The receiver 11 receives uplink control signals transmitted by the terminal stations 1, 2, and 3. Fig. 3 is a diagram illustrating a frame format of the uplink control signal. As shown in Fig. 3, the uplink control signal includes, for example, a known synchronization signal used for synchronization, a terminal station ID for identifying the terminal station, and a request relay station ID for identifying the request relay station.

[0052] The terminal station detection unit 12 (FIG. 2) detects terminal stations located within a predetermined service area based on the terminal station ID included in the uplink control signal received by the receiving unit 11. The relay station detection unit 13 detects, for each terminal station detected by the terminal station detection unit 12, a request relay station to which each terminal station requests signal relay for MIMO transmission or the like, based on the uplink control signal.

[0053] The selection unit 14 selects a relay station to relay the data signal for each of the transmission data addressed to terminal station 1, the transmission data addressed to terminal station 2, and the transmission data addressed to terminal station 3, based on the requested relay station of each terminal station detected by the terminal station detection unit 12.

[0054] For example, as shown in the example of Figure 1, the selection unit 14 selects relay station A for transmission data addressed to terminal station 1, selects relay stations A and B for transmission data addressed to terminal station 2, and selects relay stations A, B, and C for transmission data addressed to terminal station 3.

[0055] The position calculation unit 15 calculates the current position of each relay station based on the predetermined orbits and times of relay stations A, B, C, etc. (or position information of relay stations A, B, C, etc. by GPS, etc.), and outputs the calculated position to the bandwidth allocation unit 16.

[0056] Based on the current location of each relay station calculated by the location calculation unit 15, the band allocation unit 16 allocates a frequency band for transmitting signals to each relay station that can relay the signal transmitted by the base station 10 to a terminal station located within a specified service area.

[0057] Based on the combination of each relay station selected by the selection unit 14 and the frequency band allocated by the band allocation unit 16, the signal generation unit 17 generates downlink control signals and data signals to be transmitted to the terminal stations 1, 2, and 3 via each relay station, and outputs them to the transmission unit 18.

[0058] FIG. 4 is a diagram illustrating an example of a frame format of a downlink control signal. As shown in FIG. 4, the downlink control signal includes, for example, a known signal for synchronization and estimation, a relay station ID, a target terminal station ID, and data signal frequency band information. The known signal for synchronization and estimation is information used, for example, for synchronization and channel estimation. The relay station ID is information identifying the relay station to be relayed. The target terminal station ID is information identifying the target terminal station that is the target of downlink. The data signal frequency band information is information indicating the frequency band for transmitting the data signal. In other words, the downlink control signal also indicates the correspondence between the data signal and the frequency band of the data signal.

[0059] 1, signal generation unit 17 generates, as signals addressed to terminal station 1, a control signal A to be transmitted in a predetermined frequency band via relay station A and a data signal to be transmitted in frequency band W1 via relay station A. The data signal addressed to terminal station 1 is transmitted by allocating data signal A to be transmitted by relay station A to frequency band W1.

[0060] Furthermore, signal generator 17 generates, as signals addressed to terminal station 2, a control signal B to be transmitted in a predetermined frequency band via relay station B, and a data signal to be transmitted in frequency band W2 via relay stations A and B. The data signal addressed to terminal station 2 is transmitted using MIMO by multiplexing data signals A and B transmitted by relay stations A and B onto frequency band W2.

[0061] Furthermore, signal generator 17 generates, as signals addressed to terminal station 3, a control signal C to be transmitted in a predetermined frequency band via relay station C, and a data signal to be transmitted in frequency band W3 via relay stations A, B, and C. The data signal addressed to terminal station 3 is transmitted via MIMO by multiplexing data signals A, B, and C transmitted by relay stations A, B, and C into frequency band W3.

[0062] In this way, the signal generating unit 17 essentially assigns a frequency band to a relay station that relays a data signal for each terminal station.

[0063] The transmitter 18 (FIG. 2) transmits the downlink control signal and data signal generated by the signal generator 17 to each of the relay stations A, B, and C, and performs communication with the terminal stations 1, 2, and 3 via the relay stations.

[0064] 5 is a functional block diagram illustrating the functions of relay station A (or relay stations B, C). As shown in FIG. 5, relay station A (or relay stations B, C) has a signal receiving unit 21, a frequency converting / amplifying unit 22, and a signal transmitting unit 23.

[0065] The signal receiving unit 21 receives a signal transmitted by the base station and outputs it to the frequency conversion and amplification unit 22. The frequency conversion and amplification unit 22 performs processing such as frequency conversion and amplification on the signal received by the signal receiving unit 21. The signal transmitting unit 23 transmits the signal processed by the frequency conversion and amplification unit 22 to the terminal station.

[0066] By receiving the downlink control signal, relay station A (or relay stations B and C) grasps information such as the frequency band (center frequency, etc.) of the downlink control signal and the relay station ID.

[0067] 6 is a functional block diagram illustrating functions of the terminal station 1 (or the terminal stations 2 and 3). As shown in FIG. 6, the terminal station 1 (or the terminal stations 2 and 3) includes, for example, a request transmission station (relay station) selection unit 30 and a data signal demodulation unit 40.

[0068] The request transmission station selection unit 30 includes a downlink control signal receiving unit 301, a division filter unit 302, a memory unit 303, an error detection unit 304, a machine learning unit 305, a synchronization unit 306, a relay station detection unit 307, a channel estimation unit 308, a channel correlation calculation unit 309, a selection unit 310, a signal generation unit 311, and a transmission unit 50.

[0069] The downlink control signal receiving unit 301 receives receivable downlink control signals transmitted by the base station 10 via the relay station, and outputs the signals to the division filter unit 302. More specifically, the downlink control signal receiving unit 301 receives a plurality of control signals transmitted by a plurality of mobile relay stations (transmitting stations) superimposed on the same predetermined frequency band (see FIG. 1).

[0070] The dividing filter unit 302 has a plurality of matched filters (MF), divides each of the known signals received by the downlink control signal receiving unit 301 into a plurality of sections, calculates the cross-correlation value of each of the divided signals using the plurality of matched filters, and stores the calculated cross-correlation values in the memory unit 303.

[0071] 7 is a diagram schematically illustrating a state in which the division filter unit 302 divides the known signal included in the control signal. The division filter unit 302 divides the known signal of each relay station into signals of multiple intervals, for example, by dividing it by time. For example, as shown in FIG. 7, the division filter unit 302 divides the known signal of the nth relay station into k intervals.

[0072] Here, MF nkindicates the k-th sequence (symbol sequence) after dividing the known signal of the n-th relay station. Then, dividing filter unit 302 calculates the cross-correlation value of each divided signal using k matched filters, and stores the calculated cross-correlation values (cross-correlation sequences) in storage unit 303. Note that dividing filter unit 302 also shifts the output timing for each divided section as shown in FIG. 7.

[0073] The error detector 304 (FIG. 6) reads out each of the cross-correlation values calculated by the dividing filter unit 302 from the storage unit 303, and detects the frequency error (or Doppler frequency) of each of the divided known signals (and control signals) based on each of the cross-correlation values and the reference information. Then, the error detector 304 calculates the frequency error (or Doppler frequency) of each of the detected known signals (and control signals). of The signal is output to the synchronization unit 306 .

[0074] The reference information used by the error detection unit 304 includes previously acquired transmission signal information related to the transmission signal transmitted by the base station 10, and information other than the transmission signal. The transmission signal information includes, for example, the center frequency of the control signal, a primary modulation method for the digital signal such as QPSK or QAM, and a secondary modulation method such as OFDM modulation. The information other than the transmission signal includes, for example, location information, azimuth angle information, speed information, and location information for the relay station of the terminal station.

[0075] The error detection unit 304 also has a function of detecting a frequency error in the control signal by reflecting the results of machine learning performed by the machine learning unit 305.

[0076] Machine Learning Department 305 , minutes Machine learning is performed based on the cross-correlation value calculated by the dividing filter unit 302 and the position information of the moving relay station (transmitting station), and the result of the machine learning is output to the error detection unit 304.

[0077] FIG. 8 is a diagram illustrating a schematic example of machine learning performed by the machine learning unit 305. For example, the machine learning unit 305 generates two-dimensional information based on intervals (time) and filter responses based on cross-correlation values calculated by a plurality of matched filters after dividing the input signal r(t) by the dividing filter unit 302. Then, the machine learning unit 305 performs machine learning using a neural network (CNN: Convolutional Neural Network) with the generated two-dimensional information as the input layer 70. For example, the input layer 70 may be configured as a layer with the horizontal axis representing time (interval MF n1 ~MF nk ) and the vertical axis is two-dimensional information representing the filter response.

[0078] The machine learning unit 305 then performs feature detection and compression on the input layer 70 using a convolutional layer 701, a pooling layer 702, a convolutional layer 703, a pooling layer 704, and so on, performs classification using a fully connected layer 80, and estimates the Doppler frequency by calculating an output value from the output layer 90. The output layer 90 includes each frequency for each interval and an estimated probability.

[0079] △f n (t) denotes an estimated value of the Doppler frequency of the nth relay station at time t. Then, the machine learning unit 305 estimates the Doppler frequency of each section divided by the dividing filter unit 302 for the known signal of each relay station in parallel (simultaneously), for example, for each relay station.

[0080] Furthermore, the machine learning unit 305 may estimate the Doppler frequency of the corresponding section from the received signal using deep learning (DNN: deep neural network).

[0081] Even if the downlink control signals transmitted by each relay station are superimposed on the same frequency band, the reception patterns are limited to some extent in the channel model when the relay station is an artificial satellite, etc., or in an environment where the communication conditions are simpler than on the ground.

[0082] In this case, even if the machine learning unit 305 performs blind estimation by performing machine learning using a hierarchical neural network composed of, for example, an input layer, a hidden layer, and an output layer, the error detection unit 304 can detect the frequency error (Doppler frequency) of the downlink control signal of each relay station.

[0083] The synchronization unit 306 (FIG. 6) synchronizes each of the plurality of downlink control signals based on the frequency error of each control signal detected by the error detection unit 304.

[0084] The relay station detection unit 307 detects a relay station that should relay the signal transmitted by the base station 10 based on the relay station ID included in the plurality of control signals synchronized by the synchronization unit 306 .

[0085] Based on the frequency error detected by the error detection unit 304 and the plurality of control signals synchronized by the synchronization unit 306, the channel estimation unit 308 estimates the channel (downlink channel information) of each of the plurality of data signals transmitted by each relay station, and outputs the estimated channel information to the channel correlation calculation unit 309. For example, the channel estimation unit 308 estimates (detects) the frequency band of the data signal allocated to the own station based on the downlink control signal addressed to the own station.

[0086] Furthermore, since the channel estimation unit 308 estimates the channel based on the estimated Doppler frequency, it is possible to compensate for frequency errors due to the Doppler effect and estimate the channel with high accuracy.

[0087] The channel correlation calculation unit 309 calculates the correlation of the channel estimated by the channel estimation unit 308 and outputs the calculation result to the selection unit 310 .

[0088] The selection unit 310 selects a relay station (request relay station) that requests downlink relay (MIMO transmission, etc.) based on the channel correlation calculated by the channel correlation calculation unit 309, and outputs the selected relay station to the signal generation unit 311. For example, the selection unit 310 selects a relay station that has low channel correlation and requests MIMO transmission.

[0089] As a specific example, if H is a 3×3 channel matrix generated from channel information between relay stations A, B, and C and the three antennas of terminal station 3, the larger the determinant det|H|, the lower the channel correlation and the larger the transmission capacity tends to be. For this reason, terminal station 3 may set whether to perform MIMO transmission based on the value of the determinant det|H'| of matrix H' generated by estimating channel information.

[0090] Furthermore, each terminal station does not necessarily have to select all of the relay stations from which it can receive signals. For example, in the example shown in Fig. 1, even if terminal station 2 can receive downlink control signals A, B, and C transmitted from relay stations A, B, and C, if the correlation between relay stations B and C is high, it may select relay stations A and B (or a combination of relay stations A and C) and request MIMO transmission.

[0091] Furthermore, since the number of antennas owned by terminal stations 1 and 2 is less than the number of relay stations that can be received, it is necessary for terminal stations 1 and 2 to select relay stations with the number of antennas equal to or less than the number of relay stations that can be received.

[0092] The signal generator 311 (FIG. 6) generates an uplink control signal including information indicating the request relay station selected by the selector 310, and outputs the signal to the transmitter 50.

[0093] The transmitter 50 transmits the uplink control signal generated by the signal generator 311 to the base station 10. The transmitter 50 may transmit the uplink control signal to the base station 10 in any manner. That is, the transmitter 50 may transmit the uplink control signal to the base station 10 via a relay station, or may transmit the uplink control signal to the base station 10 via another terrestrial station (not shown).

[0094] The data signal demodulation unit 40 includes a data signal receiving unit 401 , a target terminal station detection unit 402 , an equalization matrix generation unit 403 , and an equalization and demodulation unit 404 .

[0095] The data signal receiving unit 401 receives data signals (or multiple data signals superimposed on a frequency band different from that of the control signal) transmitted by multiple relay stations (transmitting stations) in synchronization with each control signal, and outputs the received data signals to the target terminal station detection unit 402 and the equalization / demodulation unit 404.

[0096] The downlink control signal receiving unit 301 and the data signal receiving unit 401 constitute a receiving unit 60 that receives downlink control signals and data signals. The receiving unit 60 receives, for example, a plurality of control signals each including a known signal specific to the respective relay stations (transmitting stations) superimposed on the same predetermined frequency band, and a plurality of data signals superimposed on other frequency bands that are transmitted by the plurality of relay stations in synchronization with the respective control signals.

[0097] The target terminal station detection unit 402 detects, based on the data signal received by the data signal receiving unit 401, whether the data signal is addressed to the own station (whether the own station is a target terminal station that should receive the data signal), and outputs the data signal addressed to the own station to the equalization matrix generation unit 403.

[0098] The equalization matrix generation unit 403 generates an equalization matrix (reception weight matrix) based on the data signal input from the target terminal station detection unit 402 and the channel information estimated by the channel estimation unit 308, and outputs it to the equalization and demodulation unit 404.

[0099] The equalization and demodulation unit 404 uses the equalization matrix generated by the equalization matrix generation unit 403 to equalize and demodulate the data signal received by the data signal reception unit 401 and addressed to the own station.

[0100] Next, a description will be given of an operation example of the wireless communication system 100. Fig. 9 is a diagram showing an operation example of the wireless communication system 100. As shown in Fig. 9, first, the base station 10 identifies (detects) relay stations that can be received by the terminal stations 1, 2, and 3 within the service area, based on the positions of the respective relay stations (S100).

[0101] Next, the base station 10 generates and transmits a downlink control signal in the same frequency band to all relay stations (S102).

[0102] The terminal stations 1, 2, and 3 each receive a receivable downlink control signal among the downlink control signals transmitted from each relay station (S200). At this time, a different frequency error (Doppler frequency) occurs in each of the downlink control signals.

[0103] Next, terminal stations 1, 2, and 3 perform division filtering by dividing each known signal of the downlink control signal into multiple sections and calculating the cross-correlation values of each divided signal using multiple matched filters (S202), and store each calculated cross-correlation value (S204).

[0104] Then, the terminal stations 1, 2, and 3 perform machine learning using the cross-correlation values of the multiple sections and the above-mentioned reference information, and detect the frequency error of the downlink control signal of each relay station (S206).

[0105] Furthermore, the terminal stations 1, 2, and 3 synchronize each of the plurality of downlink control signals based on the frequency error of each control signal (S208).

[0106] For example, in the example shown in Fig. 1, terminal station 1 estimates the frequency error ΔfA of the downlink control signal A transmitted by relay station A and achieves frequency synchronization. Terminal station 2 estimates the frequency errors ΔfA and ΔfB of the downlink control signals A and B transmitted by relay stations A and B and achieves frequency synchronization. Terminal station 3 estimates the frequency errors ΔfA, ΔfB, and ΔfC of the downlink control signals A, B, and C transmitted by relay stations A, B, and C and achieves frequency synchronization.

[0107] Furthermore, the terminal stations 1, 2, and 3 estimate the channel between the terminal station and the relay station that received the downlink control signal (S210).

[0108] Then, the terminal stations 1, 2, and 3 select a relay station to request MIMO transmission (S212), and transmit uplink control information including information identifying the requested relay station (request relay station ID) to the base station 10 (S214).

[0109] The base station 10 detects the requesting relay station for each terminal station based on the terminal station ID included in the uplink control signal transmitted by each terminal station (S104), and assigns a relay station and a frequency band to transmit a data signal to each terminal station (S106).

[0110] Then, the base station 10 generates a downlink control signal and a data signal for each terminal station, and transmits them to each terminal station via the corresponding relay station (S108).

[0111] The terminal stations 1, 2, and 3 detect the frequency band allocated to the data signal based on the downlink control signal addressed to the terminal station (estimate the channel), and demodulate the data signal addressed to the terminal station using the generated reception weight matrix (S216).

[0112] In addition, the terminal stations 1, 2, and 3 determine whether or not it is necessary to change the relay station from which they can receive signals (S218), and if it is necessary to change the relay station (S218: Yes), they proceed to the processing of S210, and if it is not necessary to change the relay station (S218: No), they continue the processing of S216.

[0113] That is, the terminal stations 1, 2, and 3 monitor the downlink control signals transmitted by the base station 10, and when the terminal station changes the requested relay station from which it can receive signals due to the movement of the relay station or the terminal station, the terminal stations 1, 2, and 3 re-estimate the channel and transmit uplink control information to the base station 10.

[0114] In this way, the terminal stations 1, 2, and 3 estimate the channel of the data signal based on the frequency error detected by the error detection unit 304, and are therefore able to receive the signal transmitted from the mobile relay station (transmitting station) with high accuracy.

[0115] Furthermore, even if the number of mobile relay stations (transmitting stations) increases, the terminal stations 1, 2, and 3 can accurately receive signals transmitted from the mobile relay stations (transmitting stations) without reducing frequency utilization efficiency.

[0116] In addition, each function possessed by the base station 10, relay stations A, B, C,...X, and terminal stations 1, 2, and 3 may be configured in part or in whole by hardware such as a PLD (Programmable Logic Device) or an FPGA (Field Programmable Gate Array), or may be configured as a program executed by a processor such as a CPU.

[0117] For example, the terminal stations 1, 2, and 3 according to the present invention can be realized using a computer and a program, and the program can be recorded on a storage medium or provided via a network.

[0118] Although the embodiments of the present invention have been described above with reference to the drawings, it is clear that the above-described embodiments are merely examples of the present invention and that the present invention is not limited to the above-described embodiments. Therefore, addition, omission, substitution, and other modifications of components may be made without departing from the technical spirit and scope of the present invention. [Explanation of symbols]

[0119] 1, 2, 3... terminal station, 10... base station, 11... receiver, 12... terminal station detection unit, 13... relay station detection unit, 14... selection unit, 15... position calculation unit, 16... band allocation unit, 17... signal generation unit, 18... transmitter, 30... requested transmission station selection unit, 40... data signal demodulation unit, 50... transmitter, 60... receiver, 100... wireless communication system, 301... downlink control signal reception unit, 302... division filter unit, 303... memory unit, 304... error detection unit, 305... machine learning unit, 306... synchronization unit, 307... relay station detection unit, 308... channel estimation unit, 309... channel correlation calculation unit, 310... selection unit, 311... signal generation unit, 401... data signal reception unit, 402... target terminal station detection unit, 403... equalization matrix generation unit, 404... equalization and demodulation unit

Claims

1. a receiving unit that receives a control signal including a known signal specific to the mobile transmitting station that is transmitted in a predetermined frequency band by the mobile transmitting station, and a data signal that is transmitted in another frequency band by the transmitting station in synchronization with the control signal; a division filter unit that divides the known signal received by the receiver unit into a plurality of sections and calculates a cross-correlation value for each of the divided signals; a machine learning unit that performs machine learning based on each of the cross-correlation values calculated by the dividing filter unit and position information of a moving transmitting station; an error detection unit that detects a frequency error of the control signal by reflecting the results of machine learning performed by the machine learning unit on each of the cross-correlation values calculated by the division filter unit and on position information of a moving transmitting station; a channel estimation unit that estimates a channel of a data signal based on the frequency error detected by the error detection unit; Having A receiving station characterized by:

2. The receiving unit receiving a plurality of control signals each including a known signal specific to the respective transmitting stations superimposed on the same predetermined frequency band transmitted by the plurality of transmitting stations, and a plurality of data signals each superimposed on another frequency band transmitted by the plurality of transmitting stations in synchronization with the respective control signals; The division filter unit Dividing each of the known signals received by the receiving unit into a plurality of sections, and calculating a cross-correlation value for each of the divided signals; The error detection unit Detecting a frequency error of each control signal based on the cross-correlation values calculated by the dividing filter unit; a synchronization unit that synchronizes each of the plurality of control signals based on the frequency error of each of the control signals detected by the error detection unit; The channel estimation unit estimating channels of each of the plurality of data signals based on the plurality of control signals synchronized by the synchronization unit; 2. A receiving station according to claim 1, wherein:

3. In a wireless communication system having a receiving station for receiving a signal transmitted from a moving transmitting station, The receiving station: a receiving unit that receives a control signal including a known signal specific to the transmitting station and transmitted by the transmitting station in a predetermined frequency band, and a data signal that is transmitted by the transmitting station in another frequency band in synchronization with the control signal; a division filter unit that divides the known signal received by the receiver unit into a plurality of sections and calculates a cross-correlation value for each of the divided signals; a machine learning unit that performs machine learning based on each of the cross-correlation values calculated by the dividing filter unit and position information of a moving transmitting station; an error detection unit that detects a frequency error of the control signal by reflecting the results of machine learning performed by the machine learning unit on each of the cross-correlation values calculated by the division filter unit and on position information of a moving transmitting station; a channel estimation unit that estimates a channel of a data signal based on the frequency error detected by the error detection unit; Having A wireless communication system comprising:

4. a receiving step of receiving a control signal including a known signal specific to the mobile transmitting station transmitted in a predetermined frequency band by the mobile transmitting station, and a data signal transmitted in another frequency band by the mobile transmitting station in synchronization with the control signal; a division filtering step of dividing a received known signal into a plurality of sections and calculating a cross-correlation value for each of the divided signals; a machine learning step of performing machine learning based on each of the calculated cross-correlation values and position information of the moving transmitting station; an error detection step of detecting a frequency error of the control signal by reflecting the results of the machine learning performed in the machine learning step on each of the cross-correlation values calculated in the division filtering step and on position information of the moving transmitting station; a channel estimation step of estimating a channel of the data signal based on the detected frequency error; Contains A wireless communication method comprising:

5. In the receiving step, receiving a plurality of control signals each including a known signal specific to the respective transmitting stations superimposed on the same predetermined frequency band transmitted by the plurality of transmitting stations, and a plurality of data signals each superimposed on another frequency band transmitted by the plurality of transmitting stations in synchronization with the respective control signals; In the division filtering step, Dividing each of the known signals received in the receiving step into a plurality of sections and calculating a cross-correlation value for each of the divided signals; In the error detection step, Detecting a frequency error of each control signal based on each cross-correlation value calculated by the division filtering step; a synchronization step of synchronizing each of the plurality of control signals based on the frequency error of each of the control signals detected in the error detection step; In the channel estimation step, estimating channels of each of the plurality of data signals based on the plurality of control signals synchronized by the synchronization step; 5. The wireless communication method according to claim 4, wherein:

Images