Information processing device and program

Abstract

An information processing device (11) comprises: an acquisition unit (111) that acquires the modulation scheme, the BER, the SNR, and the symbol of a reception signal in a processing flow of an uplink process; a correction unit (112) that corrects the symbol of the reception signal in the uplink processing flow on the basis of the output of a trained model (M) created by using machine learning; and a determination unit (113) that determines the processing content of a demodulation process in the processing flow of the uplink process on the basis of the correction content for the symbol of the reception signal corrected by the correction unit (112).

Description

Information processing device and program

[0001] The present invention relates to an information processing device and a program.

[0002] Patent Document 1 describes that by compensating for a constant phase shift between the I and Q channels in the process of IQ demodulation, it is possible to ensure a constant demodulation quality in all bands.

[0003] Japan Special Table No. 2011-530256

[0004] Digital modulation methods such as QPSK and 16QAM are used in wireless sections of mobile communication networks. To meet the growing demand for wider communication bandwidths, modulation methods with high IQ compression rates are being adopted to increase the number of bits per symbol. With modulation methods with high IQ compression rates, it is necessary to minimize the required EVM (Error Vector Magnitude). For example, 3GPP TS38.521 specifies the EVM required for each modulation method.

[0005] When a modulation method with a high IQ compression rate is adopted, the EVM is likely to deteriorate. Therefore, in order to satisfy the EVM required by the above-mentioned 3GPP TS38.521, the current limit is compression up to BFP 9 bits.

[0006] An object of one aspect of the present invention is to enable the adoption of a modulation scheme with a high IQ compression rate, thereby realizing a wider communication band, while also realizing accurate demodulation of symbols.

[0007] In order to solve the above problem, the information processing device of the present invention is an information processing device arranged in a radio access network having a DU connected to a core network, an RU to which an antenna is connected, and a fronthaul connecting the DU and the RU, and is equipped with an acquisition unit that acquires the modulation method, BER, SNR, and symbols of the received signal in the processing flow of uplink processing in which the DU processes the received signal from the RU and forwards the processed data to the core network, a correction unit that corrects the symbols of the received signal in the uplink processing flow based on the output of a trained model constructed by machine learning, and a decision unit that determines the processing content of the demodulation processing in the processing flow of the uplink processing based on the correction content of the symbols of the received signal in the processing flow of the uplink processing corrected by the correction unit, wherein the input of the trained model is the modulation method, BER, SNR, and symbols of the received signal forwarded by the DU from the RU, and the output of the trained model is the correction content of the symbols of the received signal.

[0008] The information processing device according to each aspect of the present invention may be realized by a computer. In this case, the information processing device program that realizes the information processing device on a computer by causing the computer to operate as each part (software element) of the information processing device, and the computer-readable recording medium on which the program is recorded, also fall within the scope of the present invention.

[0009] According to one aspect of the present invention, a modulation method with a high IQ compression rate can be adopted, enabling a wider communication band to be realized while also achieving accurate demodulation of symbols.

[0010] It is a schematic diagram showing a schematic configuration of a radio base station according to an embodiment of the present invention. It is a block diagram showing an example of an information processing device according to an embodiment of the present invention. It is a processing flow diagram in a DU (Distributed Unit) included in the radio base station according to an embodiment of the present invention.

[0011] Hereinafter, one embodiment of the present invention will be described in detail.

[0012] (Base Station) Fig. 1 is a schematic diagram showing the general configuration of a base station 100 according to this embodiment. As shown in Fig. 1, the base station 100 includes a DU (Distributed Unit) 10, an RU (Radio Unit) 20, an antenna 12, and a fronthaul 30. The DU 10 is an O-DU (Open Distributed Unit) in the O-RAN (Open Radio Access Network) fronthaul specification, and the RU 20 is an O-RU (Open Radio Unit) in the O-RAN fronthaul specification. The fronthaul 30 is an open fronthaul in the O-RAN fronthaul specification. The DU 10 and the RU 20 share functions in accordance with Split Option 7-2x, which is a provision regarding the division of functions in the O-RAN fronthaul specifications.

[0013] The DU 10 is a distributed unit of the radio access network and is responsible for signal processing functions, including packet scheduling, error correction, encryption, and user data and control data processing. The DU 10 is responsible for data processing and radio resource management for the base station 100, and also serves as an interface with the core network 200.

[0014] The DU 10 also manages radio resources based on instructions from the core network 200 and prepares data to be transmitted to the RU 20. That is, the DU 10 executes a downlink (DL) processing flow. The DU 10 also processes signals received from the RU 20 and transfers the processed data to the core network 200. That is, the DU 10 executes an uplink (UL) processing flow.

[0015] Furthermore, the DU 10 has an information processing device 11 for suppressing EVM to a low level in the demodulation process in the processing flow of the above-mentioned uplink process. The configuration and functions of the information processing device 11 will be described later.

[0016] The RU 20 is a radio unit of the radio access network that transmits and receives radio signals. The RU 20 is connected to the antenna 12 and performs frequency conversion and analog-to-digital conversion of the radio signals. Based on instructions from the DU 10, the RU 20 transmits radio signals from the antenna 12 into the air and receives signals from the air using the antenna 12.

[0017] The fronthaul 30 is an interface connecting the DU 10 and the RU 20. The fronthaul 30 defines the communication protocol used for data transmission between the DU 10 and the RU 20. The fronthaul 30 enables real-time data transmission from the DU 10 to the RU 20 and from the RU 20 to the DU 10, and realizes advanced wireless network functions.

[0018] The fronthaul 30 may use, for example, an optical fiber cable or an Ethernet (registered trademark) cable. The fronthaul 30 may use a transmission technology such as eCPRI (enhanced Common Public Radio Interface). The eCPRI is a protocol that enables efficient digital transmission of wireless signals and is suitable for data transmission between the DU 10 and the RU 20.

[0019] In the following, this embodiment will be described using the example of a configuration in which the DU 10 and the RU 20 are located in the same base station, i.e., the base station 100, as shown in Figure 1. The configuration shown in Figure 1 is particularly suitable for small-scale networks and applications that require minimal delay. Furthermore, when the DU 10 and the RU 20 are physically close to each other, hardware resources can be efficiently utilized and network management can be simplified.

[0020] However, this embodiment is not limited to a configuration in which the DU 10 and the RU 20 are located in the same base station 100. In an O-RAN, there are cases in which the O-DU and the O-RU are physically separated. For example, by locating the O-DU in an edge location close to the user, network delay can be significantly reduced and the performance of real-time or delay-sensitive applications can be improved. Furthermore, by sharing O-DU resources among multiple O-RUs, resource utilization efficiency can be improved, and capital investment and operational costs can be reduced. This embodiment can also be applied to cases in an O-RAN in which the O-DU and the O-RU are physically separated and cases in which O-DU resources are shared among multiple O-RUs.

[0021] 1 may be physically separated in the radio access network, rather than being located in the same base station 100. Also, in the radio access network, one DU 10 may be shared among a plurality of RUs 20.

[0022] (Information Processing Device) The information processing device 11 is realized by computational resources arranged around the RU 20 and the base station 100. The information processing device 11 is, for example, a server device that functions as a RAN Intelligent Controller (RIC). The information processing device 11 uses techniques such as machine learning to correct symbols based on the estimation results of a machine-learned model in the demodulation process in the processing flow of the uplink process executed by the DU 10, thereby suppressing EVM. This makes it possible to adopt a modulation method with a high IQ compression rate, thereby realizing a wider communication bandwidth and accurate demodulation of symbols.

[0023] The information processing device 11 may have a function other than the RIC. For example, the information processing device 11 may further have a function as an MEC (Multi Access Edge Computing, Mobile Edge Computing).

[0024] 2 is a block diagram showing an example of an information processing apparatus according to an embodiment of the present invention. The information processing apparatus 11 functions as an acquisition unit 111, a correction unit 112, and a determination unit 113.

[0025] When a received signal, which is U-plane data, is transferred from the RU 20 to the DU 10, the acquisition unit 111 acquires the modulation scheme, SNR (Signal to Noise Ratio) and symbols of the received signal. Furthermore, when the acquisition unit 111 extracts the original symbols from the received signal, it acquires the BER (Bit Error Rate) of the received signal.

[0026] The modulation scheme is a method of changing the characteristics of a carrier wave, specifically, amplitude, frequency, and phase, to transmit the received signal. Examples of the modulation scheme include QAM and QPSK. The BER is the rate of erroneously transmitted bits in the received signal. The SNR is the ratio of the strength of the received signal to the strength of background noise. The symbol is an individual data unit that constitutes the received signal. In other words, the symbol is the smallest unit of the received signal.

[0027] The modulation scheme, the BER, the SNR, and the symbol acquired by the acquisition unit 111 are input to the correction unit 112 .

[0028] The compensation unit 112 compensates the symbols of the received signal to ensure smooth transmission of the received signal transmitted at a high compression rate. The EVM is an index representing the error between the ideal modulation symbol of the received signal and the symbol of the actually received signal, and the lower the EVM, the higher the quality of the received signal. The compensation unit 112 improves the quality of the received signal, increases transmission efficiency, and ensures the accuracy of the received signal.

[0029] Furthermore, the correction unit 112 uses the output value of the trained model M to correct the symbols. The trained model M is a model that has been machine-learned in advance to determine the relationship between the modulation scheme, BER, SNR, and symbols of the received signal and the correction content of the symbols of the received signal. The trained model M may be constructed and updated by an arbitrary learning device provided outside the information processing device 11 or the base station 100.

[0030] The inputs to the trained model M are the modulation scheme, BER, SNR, and symbols of the received signal transferred from the RU 20 to the DU 10, acquired by the acquisition unit 111. The output from the trained model M is the correction content of the symbols of the received signal. Here, correcting the symbols is a process of improving the quality of the received signal and bringing it closer to an ideal signal. Correcting the symbols of the received signal can mitigate distortion and information loss due to compression, thereby improving the quality of the received signal. Correcting the symbols of the received signal improves the quality of the compressed received signal, thereby enabling the received signal transmitted at a high compression rate, i.e., the U-plane data, to be transferred more efficiently and accurately.

[0031] The determination unit 113 determines the processing content of the demodulation processing in the processing flow of the above-mentioned uplink processing based on the correction content of the symbols of the received signal from the correction unit 112.

[0032] The processing performed by the information processing device 11 will be described in more detail below with reference to Fig. 3, which is a processing flow diagram for the DU 10. Furthermore, among the downlink processing (DL) and uplink processing (UL) shown in Fig. 3, the description of processing that is not directly related to the processing performed by the information processing device 11 will be simplified or omitted.

[0033] First, downlink processing will be described. In downlink processing, the DU 10 performs encoding (S101), scrambling (S102), modulation (S103), layer mapping (S104), and resource element mapping (S105) on the received signal. In Split Option 7-2x, the DU 10 is responsible for functions up to resource element mapping (S105). The received signal that has undergone resource element mapping (S105) is transferred to the RU 20 via the fronthaul 30.

[0034] Next, uplink processing will be described. When a received signal is transferred from the RU 20 to the DU 10 via the fronthaul 30, the acquisition unit 111 of the information processing device 11 acquires the modulation method, SNR, and symbol of the received signal.

[0035] The DU 10 performs resource element demapping (S201) and IDFT (Inverse Discrete Fourier Transform) channel estimation (S202) on the received signal, and then performs demodulation (S203). Note that, once the original symbols are extracted from the received signal, the acquisition unit 111 acquires the BER of the received signal. The information processing device 11 inputs the modulation scheme, BER, SNR, and symbols acquired by the acquisition unit 111 for the received signal to the trained model M. The trained model M outputs the EVM of the received signal to the correction unit 112.

[0036] When the DU 10 performs the demodulation (S203), the correction unit 112 of the information processing device 11 corrects the symbols of the received signal on which IDFT channel estimation (S202) has been performed, based on the output from the trained model M. The decision unit 113 determines the processing content of the demodulation processing in the processing flow of the uplink processing, based on the symbol correction content from the correction unit 112. More specifically, in the demodulation processing, the decision unit 113 determines to correct the symbols of the received signal on which IDFT channel estimation has been performed. This correction is performed so as to reduce the error between the symbols of the received signal on which IDFT channel estimation has been performed and the ideal symbols of the received signal.

[0037] The demodulation process for correcting the symbols of the received signal is, for example, as follows: Since the presence of I / Q imbalance causes degradation of EVM, the demodulator is instructed to perform more accurate I / Q imbalance correction using parameters (e.g., correction coefficients for amplitude imbalance and phase imbalance) inferred by the trained model M, thereby improving demodulation accuracy.

[0038] The DU 10 performs descrambling (S204) and decoding (S205) on the received signal that has been demodulated (S203). In Split Option 7-2x, the DU 10 is responsible for functions starting from resource element demapping (S201).

[0039] (Variant) In the above embodiment, the information processing device 11 corrects the symbols of the received signal in the demodulation process in the processing flow of the uplink processing performed by the DU10, but the information processing device 11 may also correct the symbols of the received signal in the modulation process in the processing flow of the downlink processing performed by the DU10.

[0040] In the above embodiment, functions are shared between the DU 10 and the RU 20 in accordance with Split Option 7-2x. If this functional sharing is revised so that the RU 20 is also responsible for the demodulation process in the processing flow of the uplink process, the RU 20 will have the information processing device 11.

[0041] [Summary] An information processing device according to aspect 1 of the present invention is an information processing device arranged in a radio access network including a DU connected to a core network, an RU connected to an antenna, and a fronthaul connecting the DU and the RU, and includes: an acquisition unit that acquires a modulation scheme, a BER, an SNR, and symbols of the received signal in a processing flow of uplink processing in which the DU processes a received signal from the RU and transfers the processed data to the core network; a correction unit that corrects the symbols of the received signal in the uplink processing flow based on the output of a trained model constructed by machine learning; and a determination unit that determines the processing content of a demodulation processing in the processing flow of the uplink processing based on the correction content of the symbols of the received signal in the processing flow of the uplink processing corrected by the correction unit, wherein the input of the trained model is the modulation scheme, BER, SNR, and symbols of the received signal transferred from the RU to the DU, and the output of the trained model is the correction content of the symbols of the received signal.

[0042] According to the above configuration, when demodulating a received signal, the symbols of the received signal can be corrected based on the estimation results of the machine learning model, so that the EVM of the received signal can be kept small.

[0043] In an information processing device according to aspect 2 of the present invention, in the above aspect 1, the determination unit may determine the processing content of the demodulation processing so as to reduce an error between the symbol of the received signal and an ideal symbol of the received signal in the demodulation processing.

[0044] A program according to aspect 3 of the present invention is a program for causing a computer to function as the information processing device described in aspects 1 and 2 above, and may be configured to cause the computer to function as the acquisition unit, the correction unit, and the determination unit.

[0045] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. Furthermore, new technical features can be formed by combining the technical means disclosed in each embodiment.

[0046] REFERENCE SIGNS LIST 10 DU 11 Information processing device 12 Antenna 20 RU 30 Fronthaul 100 Base station 111 Acquisition unit 112 Correction unit 113 Determination unit M Trained model

Claims

1. An information processing device disposed in a radio access network having a DU connected to a core network, an RU to which an antenna is connected, and a fronthaul connecting the DU and the RU, comprising: an acquisition unit that acquires the modulation method, BER, SNR, and symbols of the received signal in the processing flow of uplink processing in which the DU processes the received signal from the RU and forwards the processed data to the core network; a correction unit that corrects the symbols of the received signal in the processing flow of the uplink processing based on the output of a trained model constructed by machine learning; and a determination unit that determines the processing content of the demodulation processing in the processing flow of the uplink processing based on the correction content of the symbols of the received signal in the processing flow of the uplink processing corrected by the correction unit, wherein the input of the trained model is the modulation method, BER, SNR, and symbols of the received signal forwarded by the DU from the RU, and the output of the trained model is the correction content of the symbols of the received signal.

2. The information processing device according to claim 1, wherein the determination unit determines the processing content of the demodulation processing so as to reduce an error between the symbols of the received signal and ideal symbols of the received signal in the demodulation processing.

3. A program for causing a computer to function as the information processing device according to claims 1 and 2, the program causing the computer to function as the acquisition unit, the correction unit, and the determination unit.

Images