JP7614940B2 - Deterioration determination system, wireless device, determination device, learning device, and inference device - Google Patents

Description

The technology disclosed in this specification relates to a technology for determining deterioration of wireless equipment.

Conventionally, there have been monitoring systems that monitor the operating status of moving objects such as railway vehicles (see, for example, Patent Document 1 or Non-Patent Document 1). One such monitoring system acquires time series data indicating the operating status of a target core device via a high-speed wireless network, and further accumulates the time series data in a data accumulation server at a data center. Then, a data analysis server at the data center automatically detects signs of failure in railway vehicles and diagnoses the lifespan of railway vehicles based on the time series data accumulated in the data accumulation server.

As such, some conventional monitoring systems monitor the core equipment of railway vehicles to detect signs of failure or diagnose the lifespan of the railway vehicles.

Japanese Patent Application Publication No. 3-243094

Mitsubishi Electric News Release, "Development of 'Fault Sign Detection System' for Railway Vehicles," October 5, 2020

Wireless equipment is becoming increasingly important in society as it serves as the foundation for advanced distributed processing systems that apply artificial intelligence and other technologies. Furthermore, communication systems that include wireless equipment are required to have higher reliability and availability than ever before.

To achieve the high reliability and availability described above, measures have been taken to prepare for the possibility of operational abnormalities occurring while the wireless equipment is in operation by configuring the main units of the wireless equipment in a redundant configuration and adding complex controls such as automatic switching or automatic degenerate operation. As a result, there is a problem in that the introduction costs of the communication system become high.

The technology disclosed in this specification was developed in consideration of the problems described above, and is a technology for effectively maintaining wireless equipment while reducing installation costs.

A degradation determination system, which is a first aspect of the technology disclosed in the present specification, includes a wireless device having an amplifier unit for amplifying and outputting an input wireless signal, and a determination unit for determining that the amplifier unit is degraded when the wireless signal output from the amplifier unit exceeds an amplitude threshold and a phase threshold that are determined for each amplitude of the corresponding wireless signal input to the amplifier unit.

According to at least the first aspect of the technology disclosed in the present specification, deterioration of the amplifier unit can be determined based on changes in the amplitude and phase of the radio signal output from the amplifier unit, so that failure of the amplifier unit can be predicted and preventive maintenance can be performed appropriately. This makes it possible to reduce the number of regular inspections of the performance of the wireless equipment, and also to reduce the number of spare parts required to ensure the reliability and availability of the communication system. As a result, the running costs of the communication system that uses the wireless equipment can be reduced.

Furthermore, the objects, features, aspects and advantages associated with the technology disclosed in the present specification will become more apparent from the detailed description and accompanying drawings set forth below.

FIG. 1 is a diagram showing an example of the configuration of a communication system including fixed or mobile wireless equipment according to an embodiment. 2 is a diagram illustrating an example of the configuration of wireless equipment in the communication system illustrated in FIG. 1 . 1 is a diagram illustrating an example of the relationship between degradation of the power amplification characteristics of wireless equipment and degradation of the characteristics of components in the wireless equipment. 2 is a diagram showing in detail an example of a configuration including a frequency conversion section (IF→RF) and an amplification section. FIG. 1 is a diagram showing an example of an AM/AM characteristic that is a characteristic of an output amplitude relative to an input amplitude. 1 is a diagram showing an example of AM/PM characteristics, which are characteristics of an output phase with respect to an input amplitude; FIG. 13 is a diagram illustrating an example of a characteristic of an output amplitude with respect to an input frequency. 1 is a conceptual diagram showing an example of signal modulation accuracy. 1 is a conceptual diagram showing an example of signal modulation accuracy. 10 is a flowchart showing an example of the operation of a control unit, a storage unit in the control unit, a sorting unit in the control unit, a data accumulation unit, and a determination unit in the data accumulation unit. 2 is a diagram illustrating in detail an example of a configuration including a distortion compensation section and an amplifier section. FIG. 1A and 1B are diagrams showing an example of AM/AM characteristics and an example of AM/PM characteristics; 11A and 11B are diagrams illustrating an example of the AM/AM characteristic of a distortion compensator and an example of the AM/PM characteristic of the distortion compensator. FIG. 13 is a diagram illustrating an example of a configuration including a distortion compensation unit. 10 is a flowchart illustrating an example of the operation of a control unit according to the embodiment. 10 is a timing chart showing an example of the operation of wireless equipment when a communication standard specifies a time during which no communication is performed. FIG. 1 is a diagram showing an example of the configuration of fixed or movable wireless equipment according to an embodiment; FIG. 2 is a conceptual diagram showing the concept of converting a time domain signal into a frequency domain signal by FFT. FIG. 2 is a conceptual diagram showing the concept of converting a time domain signal into a frequency domain signal by FFT. FIG. 1 is a diagram illustrating an example of the configuration of fixed or movable wireless equipment when performing feedback type distortion compensation. FIG. 13 is a diagram illustrating an example of a configuration including a distortion compensation and quadrature modulation section. FIG. 1 is a diagram conceptually illustrating an example of the configuration of a learning device that performs machine learning on wireless equipment. FIG. 11 is a diagram illustrating an example of a method for acquiring training data. FIG. 1 is a diagram illustrating an example of a neural network. 13 is a flowchart relating to a learning process of the learning device. FIG. 2 is a diagram conceptually illustrating an example of the configuration of an inference device related to wireless equipment. 13 is a flowchart relating to an inference process of the inference device.

The following describes the embodiments with reference to the attached drawings. In the following embodiments, detailed features are shown to explain the technology, but these are merely examples and are not necessarily all essential features for the embodiments to be feasible.

The drawings are schematic, and for ease of explanation, configurations are omitted or simplified as appropriate in the drawings. Furthermore, the size and positional relationships of the configurations shown in different drawings are not necessarily described accurately, and may be changed as appropriate. Furthermore, hatching may be used in drawings such as plan views that are not cross-sectional views to make it easier to understand the contents of the embodiments.

In addition, in the following description, similar components are illustrated with the same reference symbols, and their names and functions are also similar. Therefore, detailed descriptions of them may be omitted to avoid duplication.

In addition, in the description of this specification, when a certain component is described as "comprising," "including," or "having," unless otherwise specified, this is not an exclusive expression that excludes the presence of other components.

In addition, even if ordinal numbers such as "first" or "second" are used in the description of this specification, these terms are used for convenience to facilitate understanding of the contents of the embodiments, and the contents of the embodiments are not limited to the orders that may result from these ordinal numbers.

In addition, in the explanations given in this specification, expressions indicating an equal state, such as "same," "equal," "uniform," or "homogeneous," unless otherwise specified, include cases indicating a strictly equal state, as well as cases in which differences occur within a tolerance or within a range in which the same degree of functionality can be obtained.

First Embodiment
The deterioration determination system according to this embodiment will be described below.

<Configuration of the Deterioration Judgment System>
FIG. 1 is a diagram showing an example of the configuration of a communication system including fixed or mobile wireless equipment according to this embodiment.

As shown in the example in FIG. 1, the communication system 1 includes an on-board radio device 12, which is a mobile radio device mounted on a moving body such as a railway vehicle, a ground radio device 14, which is a radio device that is fixed on the ground and is not a moving body, a data storage unit 16 that stores data, and a data analysis unit 18 (a learning device or inference device described below) that analyzes the data stored in the data storage unit 16.

The on-board radio device 12 and the ground radio device 14 wirelessly communicate data to be maintained. The data storage unit 16 receives data from the ground radio device 14 via the network and stores the data in chronological order. The data analysis unit 18 determines the deterioration of the on-board radio device 12 or the ground radio device 14, and also predicts failures, based on the results of the data analysis.

Figure 2 is a diagram showing an example of the configuration of the on-board radio device 12 or the ground radio device 14 in the communication system 1 shown in Figure 1.

The radio device 101 shown in FIG. 2 is an on-board radio device 12 or a ground radio device 14, and includes at least a control unit 20, a data generation unit 102, a digital modulation unit 103, an orthogonal modulation unit 104b, a frequency conversion unit (IF → RF) 105, a local oscillation unit 106, an amplification unit 107, a high-frequency transmission unit 108, an antenna 109, a waveform detection unit 110, a feedback amplification unit 110b that amplifies the output from the waveform detection unit 110, a frequency conversion unit (RF → IF) 111 to which the output from the feedback amplification unit 110b is input, an orthogonal demodulation unit 112, an amplitude and phase measurement unit 113, a power supply unit 114, a voltage measurement unit 115, a current sensor 116, a current measurement unit 117, a temperature sensor 118, a power detection unit 119, and a transmission power measurement unit 120. The control unit 20 includes a memory unit 22 and a sorting unit 24.

The data generation unit 102 generates and outputs transmission data (audio, control data, etc.) corresponding to the input user service. The digital modulation unit 103 digitally modulates the transmission data output from the data generation unit 102 and outputs it as a digitally modulated signal.

The orthogonal modulation unit 104b converts the digitally modulated signal output from the digital modulation unit 103 into a two-dimensional complex signal (IQ signal) that is out of phase with the signal by 90 degrees (i.e., orthogonal). The frequency conversion unit (IF→RF) 105 converts the two-dimensional complex signal (IQ signal) output from the orthogonal modulation unit 104b into a carrier frequency by mixing it with a local signal input from the local oscillation unit 106, and outputs a high-frequency signal.

The amplifier 107 amplifies the high-frequency signal output from the frequency converter (IF → RF) 105 to the desired power and outputs it to the antenna 109 via the high-frequency transmission unit 108.

On the other hand, the waveform detection unit 110 extracts a part of the high frequency signal output from the amplifier unit 107 as a detection signal. The frequency conversion unit (RF → IF) 111 down-converts the frequency of the detection signal by mixing the detection signal with the local signal input from the local oscillator unit 106, and outputs an IF (intermediate frequency) signal.

The orthogonal demodulation unit 112 converts the IF (intermediate frequency) signal output from the frequency conversion unit (RF → IF) 111 into a two-dimensional demodulated complex signal (IQ signal) that is out of phase by 90 degrees (i.e., orthogonal).

Here, the wireless device 101 may further include a distortion compensation unit 104a. The distortion compensation unit 104a outputs a distortion compensation signal Se (error complex signal Se), which is the difference between the timing-aligned input complex signal Si (signal input from the digital modulation unit 103) and the demodulated complex signal So (signal input from the orthogonal demodulation unit 112), to the orthogonal modulation unit 104b.

Next, the operation of measuring the characteristics of each component will be described. Note that, although the operation of measuring the characteristics of multiple components will be described as an example below, it is also possible to measure the characteristics of a single component, or to measure the characteristics of only some of the components included in the wireless equipment.

The amplitude and phase measurement unit 113 measures the amplitude and phase of the input complex signal Si input to the distortion compensation unit 104a, and the amplitude and phase of the distortion compensation signal Se output from the distortion compensation unit 104a. The storage unit 22 stores the amplitude and phase of the input complex signal Si or the amplitude and phase of the distortion compensation signal Se measured at a constant period.

The sorting unit 24 sorts the amplitude of the input complex signal Si and the amplitude and phase of the distortion compensation signal Se (error complex signal Se) stored in the memory unit 22 in the order of the amplitude of the input complex signal Si to generate a table of input amplitude versus error amplitude (AM/AM characteristics) and a table of input amplitude versus error phase (AM/PM characteristics).

The determination unit 32 in the external data storage unit 16 determines whether the power amplification characteristics of the wireless device 101 have deteriorated when the error amplitude and error phase for each input amplitude exceed the reference value for each input amplitude (i.e., when the values exceed the predetermined allowable amplitude range or allowable phase range). The determination unit 32 may directly compare the wireless signal output from the amplifier 107 with the corresponding wireless signal input to the amplifier 107, and determine that the amplifier 107 has deteriorated when the wireless signal output from the amplifier 107 exceeds the amplitude reference value and phase reference value determined for each amplitude of the corresponding wireless signal input to the amplifier 107.

The power supply unit 114 supplies the power required for the amplifier unit 107 to amplify and output the input signal as a bias voltage to the amplifier unit 107. The voltage measurement unit 115 measures the bias voltage output by the power supply unit 114.

The current sensor 116 and the current measurement unit 117 measure the supply current consumed by the amplifier unit 107 from the bias voltage. The temperature sensor 118 measures the temperature around the amplifier unit 107, including the heat generated by the amplifier unit 107.

The power detection unit 119 and the transmission power measurement unit 120 measure the transmission power output by the amplifier unit 107.

<Relationship between deterioration of component characteristics and measured values>
Fig. 3 is a diagram showing an example of the relationship between deterioration of the power amplification characteristic of the wireless device 101 and deterioration of the characteristics of the components in the wireless device 101. Fig. 4 is a diagram showing in detail an example of a configuration including a frequency conversion unit (IF → RF) 105 and an amplification unit 107.

Here, degradation of characteristics means a decrease in the capabilities exhibited due to changes over time, and for example, in the case of power amplification characteristics, it means a decrease in the amount of power amplification. Figure 3 shows changes in measured values indicating degradation of the power amplification characteristics of wireless device 101, or changes in measured values indicating degradation of the characteristics of components in wireless device 101.

As shown in the example of FIG. 3, when the power transistor (i.e., amplifier 107) has a primary characteristic of a decrease in mutual conductance gm (or DC current gain hFE), the secondary characteristic is a decrease in operating current Id (Ic), and the output characteristic (Radio Law characteristic) is a decrease in output (measured transmission power). Also, when the power transistor (i.e., amplifier 107) has a primary characteristic of a decrease in saturated power (Psat), the secondary characteristic is a deterioration in AM/AM characteristics or AM/PM characteristics, and the output characteristic (Radio Law characteristic) is a deterioration in adjacent channel leakage power (increase in leakage power).

In addition, for peripheral components such as filters or matching circuits (i.e., frequency conversion unit (IF → RF) 105, frequency conversion unit (RF → IF) 111, or amplifier unit 107, etc.), if the primary characteristic is insulation deterioration (decrease in insulation) of the capacitor or moisture deterioration of the inductor, the secondary characteristic is deterioration of the frequency characteristics, and the output characteristic (Radio Law characteristic) is deterioration of the frequency characteristics. In addition, for peripheral components such as filters or matching circuits, if the primary characteristic is insulation deterioration (decrease in insulation) of the capacitor or moisture deterioration of the inductor, the secondary characteristic is a decrease in saturated power (Psat), and the output characteristic (Radio Law characteristic) is deterioration of adjacent channel leakage power (increase in leakage power), as in the case of power transistors.

When the primary characteristic of the power supply unit 114 is an increase in ripple, the secondary characteristic is the occurrence of ripples in the envelope waveform, and the output characteristic (Radio Law characteristic) is a deterioration in modulation accuracy (i.e., a decrease in modulation accuracy). When the primary characteristic of the power supply unit 114 is a decrease in bias voltage, the secondary characteristic is a decrease in saturated power (Psat), and the output characteristic (Radio Law characteristic) is a deterioration in adjacent channel leakage power (an increase in leakage power), as in the case of a power transistor.

When the primary characteristic of the local oscillator 106 is a decrease in output level, the secondary characteristic is a decrease in saturated power (Psat), and the output characteristic (Radio Law characteristic) is a deterioration in adjacent channel leakage power (increase in leakage power) as in the case of a power transistor. When the primary characteristic of the local oscillator 106 is a decrease in oscillation stability, the secondary characteristic is an increase in phase noise, and the output characteristic (Radio Law characteristic) is a deterioration in modulation accuracy. When the primary characteristic of the local oscillator 106 is a change in structure or impedance, the secondary characteristic is a frequency shift, and the output characteristic (Radio Law characteristic) is a deterioration in output frequency deviation.

In addition, while the primary characteristic of ground-earth is an increase in impedance (change in material or structure), the secondary characteristic is the introduction of noise, and the output characteristic (radio law characteristic) is a deterioration in spurious characteristics (i.e. an increase in unwanted emissions) or a deterioration in modulation accuracy.

In this embodiment, the following are measured: the secondary characteristics of the power transistor, deterioration of frequency characteristics which is a secondary characteristic of peripheral components such as filters or matching circuits, deterioration of adjacent channel leakage power (increase in leakage power) which is an output characteristic of peripheral components such as filters or matching circuits, reduction in bias voltage which is a primary characteristic of the power supply unit 114, output characteristics of the power supply unit 114, deterioration of adjacent channel leakage power (increase in leakage power) which is an output characteristic of the local oscillator 106, deterioration of modulation accuracy which is an output characteristic of the local oscillator 106, and output characteristics of the ground-earth.

As shown in the example of FIG. 4, the frequency conversion unit (IF → RF) 105 includes a frequency mixer 105a and a filter 105b. The amplification unit 107 includes a power transistor 107a and a matching circuit 107b.

As can be seen from the relationship shown in FIG. 3, by measuring the changes in the secondary characteristics of the components and the output characteristics of the wireless device 101, it is possible to narrow down the components whose characteristics have deteriorated. Then, by identifying the deteriorated components early, it is possible to repair or replace the components that show signs of failure.

<Relationship between each characteristic and the corresponding components>
Next, the relationship between each characteristic shown in FIG. 3 and the components that measure the corresponding characteristic will be described.

The AM/AM characteristics, AM/PM characteristics, and saturated power (Psat) in FIG. 3 can be measured by the amplitude and phase measurement unit 113, memory unit 22, and sorting unit 24 in FIG. 2.

First, we will explain the AM/AM characteristics and the AM/PM characteristics. Ideally, the amplifier unit 107 would obtain an output signal with exactly the same waveform as the input signal. However, in reality, the waveform changes due to a nonlinear response (i.e., the waveform becomes distorted).

Figure 5 is a diagram showing an example of AM/AM characteristics, which are the characteristics of the output amplitude relative to the input amplitude in the amplifier unit 107, etc. In Figure 5, the vertical axis indicates the output amplitude, and the horizontal axis indicates the input amplitude. As shown in the example in Figure 5, when the input amplitude is large, the output amplitude saturates (saturation points are Psat1 and Psat2). In Figure 5, the reference AM/AM characteristics (for example, the initial state) are shown by solid lines, and the state in which the AM/AM characteristics have deteriorated (i.e., the state in which the output amplitude has decreased relative to the input amplitude) is shown by dotted lines.

Figure 6 is a diagram showing an example of AM/PM characteristics, which are the characteristics of the output phase relative to the input amplitude in the amplifier unit 107, etc. In Figure 6, the vertical axis indicates the output phase, and the horizontal axis indicates the input amplitude. As shown in the example in Figure 6, the amount of lead (lag) of the output phase changes depending on the input amplitude. In Figure 6, the reference AM/PM characteristics (for example, the initial state) are shown by solid lines, and a state in which the AM/PM characteristics have deteriorated (i.e., a state in which the output phase has decreased relative to the input amplitude) is shown by dotted lines.

The AM/AM characteristics and AM/PM characteristics are characteristics that indicate the relationship between the input to amplifier 107 and the output from amplifier 107, and therefore may be obtained by measuring the input amplitude of amplifier 107, the output amplitude of amplifier 107, and the output phase of amplifier 107.

Here, in order to measure the RF (radio frequency) signal at the input and output of the amplifier unit 107, it is necessary to sample at a frequency higher than the frequency of the signal. However, it is generally difficult to implement a configuration for performing such sampling.

Therefore, instead of measuring the RF signal, it is also possible to measure the IF (intermediate frequency) signal that is the output of the orthogonal modulation unit 104b and the IF (intermediate frequency) signal that is the output of the frequency conversion unit (RF → IF) 111.

Similarly, the baseband complex signal output from the digital modulation unit 103 and the baseband complex signal output from the orthogonal demodulation unit 112 may be measured. Note that in Figures 5 and 6, the above-mentioned baseband complex signal is measured.

The operating current Id (Ic) of the power transistor in FIG. 3 can be measured by the current measuring unit 117. Also, the bias voltage of the power supply unit 114 in FIG. 3 can be measured by the voltage measuring unit 115.

Next, we will explain the frequency characteristics in Figure 3. The amplifier unit 107 or its surrounding components (e.g., frequency conversion unit) generally have a frequency bandwidth that ensures the desired characteristics.

Figure 7 is a diagram showing an example of the characteristics of the output amplitude versus the input frequency in the amplifier unit 107, etc. In Figure 7, the vertical axis indicates the output amplitude, and the horizontal axis indicates the input frequency. In Figure 7, the reference frequency characteristics (for example, the initial state) are shown by solid lines, and the state in which the frequency characteristics are degraded (i.e., the state in which the output amplitude is reduced relative to the input frequency) is shown by dotted lines. Note that Psat is the saturation point.

As shown in the example in Figure 7, the amplifier unit 107 and other components are designed and manufactured so that the characteristics (i.e., output amplitude) are uniform even when the frequency changes within the above frequency bandwidth.

In other words, the frequency characteristics indicate the uniformity of the characteristics of the amplifier unit 107 or its surrounding components for signals of multiple frequency channels. The frequency characteristics can be obtained by measuring the above-mentioned AM/AM characteristics or AM/PM characteristics, etc., for each frequency channel.

Next, the modulation accuracy in FIG. 3 will be described. In the digital multi-level modulation method, the signal to be transmitted is mapped to two-dimensional complex coordinates and then transmitted. If the modulation accuracy of the signal decreases, variation occurs in the two-dimensional complex coordinates of the signal mapped as described above.

Figures 8 and 9 are conceptual diagrams showing examples of signal modulation accuracy in the amplifier unit 107 and the like. As shown in the examples in Figures 8 and 9, the signal to be transmitted is mapped to two-dimensional complex coordinates.

The transmitted signal that is mapped to two-dimensional complex coordinates is ideally one of a finite number of coordinate points (four points in Figure 8: (1, 1), (1, -1), (-1, -1), and (-1, 1)), as shown in the example in Figure 8.

However, when the modulation accuracy of the amplifier unit 107 decreases, the mapped points deviate from a finite number of coordinate points, as shown in the example in Figure 9.

The amplitude and phase measurement unit 113 measures the amplitude and phase of the distortion compensation signal Se output from the distortion compensation unit 104a, and further samples the amplitude and phase at the symbol timing to obtain the amplitude error (phase error) from the expected coordinate point, i.e., the modulation accuracy.

<AM/PM characteristic measurement method 1 (without distortion compensation)>
Next, a specific description will be given of a method for measuring the AM/AM characteristics of amplifier 107 and a method for measuring the AM/PM characteristics of amplifier 107 using wireless device 101. Here, input signal Si(t), output signal So(t), and error signal Se(t) at time t are expressed as follows:

Here, Ai(t) denotes the amplitude of the input signal, φi(t) denotes the phase of the input signal, Ao(t) denotes the amplitude of the output signal, φo(t) denotes the phase of the output signal, Ae(t) = Ao(t) - Ai(t) denotes the amplitude of the error signal, φe(t) = φo(t) - φi(t) denotes the phase of the error signal, and T(t) denotes the temperature.

Figure 10 is a flowchart showing an example of the operation of the control unit 20, the memory unit 22 in the control unit 20, the sorting unit 24 in the control unit 20, the data accumulation unit 16, and the determination unit 32 in the data accumulation unit 16.

First, the memory unit 22 at least temporarily records Ai(t), Ae(t), and Φe(t) measured by the amplitude and phase measurement unit 113, and T(t) measured by the temperature sensor 118 (step ST101).

Next, in step ST102, the control unit 20 determines whether the recording of these measurement values has been repeated for a specific measurement period. If the specific measurement period has elapsed, that is, if the result corresponds to "YES" branching from step ST102, an example of which is shown in FIG. 10, the process proceeds to step ST103, an example of which is shown in FIG. 10. On the other hand, if the specific measurement period has not elapsed, that is, if the result corresponds to "NO" branching from step ST102, an example of which is shown in FIG. 10, the process returns to step ST101, an example of which is shown in FIG. 10.

Next, in step ST103, the sorting unit 24 classifies the temperature T(t), which is time series data, into a number of temperature groups based on the temperature threshold value. The number of temperature groups is set to an appropriate number for extracting and classifying characteristics such as, for example, abnormally low temperature, low temperature, sub-low temperature, normal temperature, sub-high temperature, high temperature, or abnormally high temperature.

Next, in step ST104, the sorting unit 24 classifies the time series data Ai(t), the time series data Ae(t), and the time series data Φe(t) into the above-mentioned temperature groups corresponding to the time t.

Next, in step ST105, the sorting unit 24 sorts Ai(t), Ae(t), and Φe(t) classified by temperature group by Ai(t).

Next, in step ST106, the sorting unit 24 smoothes the ratio between Ai(t) and Ae(t) for each temperature group using the least squares error method or the like. Then, the smoothed value is set as the AM/AM characteristic of the amplifier unit 107.

Next, in step ST107, the sorting unit 24 smoothes the ratio between Ai(t) and Φe(t) for each temperature group using the least squares error method or the like. Then, the smoothed value is set as the AM/PM characteristic of the amplifier unit 107.

Next, in step ST108, the determination unit 32 determines that there is degradation of the amplifier unit 107 when the error amplitude or error phase for each temperature group and input amplitude exceeds the corresponding reference value for each temperature group and input amplitude based on the AM/AM characteristics and AM/PM characteristics acquired as described above.

Next, in step ST109, the data accumulation unit 16 records the AM/AM characteristics and AM/PM characteristics for each measurement period and each temperature group, and accumulates the data.

Next, in step ST110, the control unit 20 determines whether a command to end the operation has been received from a user or the like. If a command to end the operation has been received, that is, if the result corresponds to "YES" branching from step ST110 as shown in FIG. 10, the operation is ended. On the other hand, if a command to end the operation has not been received, that is, if the result corresponds to "NO" branching from step ST110 as shown in FIG. 10, the process returns to step ST101 as shown in FIG. 10.

In FIG. 10, the temperature is classified into multiple temperature groups to eliminate the influence of temperature, but it is obvious that the deterioration of the amplifier unit 107 can be determined even if the temperature measurement value T(t) is not used and the temperature group classification is not performed.

<AM/PM characteristic measurement method 2 (feedback distortion compensation)>
Next, a method for measuring the AM/AM characteristics of the amplifier 207 when performing distortion compensation will be described below, and a method for measuring the AM/PM characteristics of the amplifier 207. Known distortion compensation methods include feedback distortion compensation, feedforward distortion compensation, and predistortion distortion compensation.

Figure 11 is a diagram showing in detail an example of a configuration including a distortion compensation unit 104a and an amplifier unit 207. In Figure 11, a feedback amplifier unit 110c is provided that performs feedback distortion compensation based on a demodulated complex signal So, which is an output signal of the amplifier unit 207.

The amplifier 207 is represented as a model in which distortion D is added to an ideal amplifier 125 with an amplification factor A. The feedback amplifier 110c is an amplifier with an amplification factor of 1/K. The distortion compensation unit 104a is a negative feedback circuit that outputs the difference between the feedback signal input from the feedback amplifier 110c and the input complex signal Si as a distortion compensation signal Se.

Feedback distortion compensation can cancel the distortion component D added to the demodulated complex signal So, which is the output signal of the amplifier 207, by negative feedback. This makes it possible to reduce the distortion.

Figure 12 is a diagram showing an example of AM/AM characteristics and an example of AM/PM characteristics. In Figure 12, the vertical axis indicates the output amplitude or output phase, and the horizontal axis indicates the input amplitude. In Figure 12, the AM/AM characteristics and AM/PM characteristics of the ideal amplifier 125 are shown by dashed lines. On the other hand, in Figure 12, the AM/AM characteristics and AM/PM characteristics of the actual amplifier unit 107 are shown by solid lines.

In FIG. 12, the AM/AM characteristics of the actual amplifier unit 107 are shown in graph 1001, and the AM/AM characteristics of the ideal amplifier 125 are shown in graph 1002. Graph 1002 is a straight line with a slope A. The difference between graphs 1001 and 1002 represents the amplitude of distortion D of the amplifier.

Similarly, the AM/PM characteristics of the actual amplifier 107 are shown in graph 1003, and the AM/PM characteristics of the ideal amplifier 125 are shown in graph 1004. The difference between graph 1003 and graph 1004 represents the phase of the distortion D of the amplifier.

Figure 13 is a diagram showing an example of the AM/AM characteristics of the distortion compensation unit 104a and an example of the AM/PM characteristics of the distortion compensation unit 104a. In Figure 13, the vertical axis represents the distortion compensation signal Se, and the horizontal axis represents the input amplitude.

In FIG. 13, the AM/AM characteristics of the distortion compensation unit 104a are shown in graph 1005. The AM/PM characteristics of the distortion compensation unit 104a are shown in graph 1006.

It is known that the AM/AM and AM/PM characteristics of the distortion compensation unit 104a are the inverse characteristics of the AM/AM and AM/PM characteristics of the amplifier unit 207 in order to cancel out the distortion D of the amplifier unit 207.

Therefore, measuring the AM/AM and AM/PM characteristics of the distortion compensation unit 104a is equivalent to measuring the AM/AM and AM/PM characteristics of the amplifier unit 207.

The distortion compensation signal Se is shown as follows:

Here, Ae(t) indicates the amplitude of the distortion compensation signal Se, and φe(t) indicates the phase of the distortion compensation signal Se.

According to the flowchart shown in FIG. 10, the deterioration of the amplifier 207 in which feedback distortion compensation is performed by the feedback amplifier 110c can be determined, and the AM/AM characteristics and AM/PM characteristics of the amplifier 207 can be obtained.

<AM/PM characteristic measurement method 3 (predistortion compensation)>
Fig. 14 is a diagram showing an example of a configuration including a distortion compensator. In Fig. 14, a predistortion distortion compensator 131 is provided as the distortion compensator.

The predistortion distortion compensation unit 131 is configured with an adaptive digital predistortion (ADPD). The predistortion distortion compensation unit 131 converts the input digital complex signal into a distortion-compensated digital complex signal having the inverse distortion of the amplifier by digital signal processing, and outputs the signal.

The digital-to-analog conversion unit 132 converts the distortion-compensated digital complex signal output from the predistortion distortion compensation unit 131 into a distortion-compensated analog complex signal and outputs it. The orthogonal modulation unit 104b performs orthogonal modulation on the distortion-compensated analog complex signal output from the digital-to-analog conversion unit 132, and outputs it to the frequency conversion unit (IF → RF) 105 as a distortion-compensated signal Se.

Here, the predistortion distortion compensation unit 131 uses a method such as polynomial approximation or a look-up table (LUT) for digital signal processing.

The orthogonal demodulation unit 112 orthogonally demodulates the IF (intermediate frequency) signal output from the frequency conversion unit (RF → IF) 111, i.e., the feedback signal, and outputs it as a demodulated complex signal (IQ signal). The analog-to-digital conversion unit 138 converts the analog demodulated complex signal (IQ signal) output from the orthogonal demodulation unit 112 into a digital demodulated complex signal and outputs it.

On the other hand, the delay unit 141 delays the digital complex signal (input digital complex signal) input to the predistortion distortion compensation unit 131 to align the timing of the input digital complex signal and the digital demodulation complex signal. The input/output comparison unit 139 outputs the difference between the input digital complex signal and the digital demodulation complex signal, the timing of which is aligned, as a digital error complex signal Se to the predistortion distortion compensation unit 131.

The adaptive operation unit 140 adapts (i.e., changes) the parameters of the digital signal processing of the predistortion distortion compensation unit 131 so that the digital error complex signal Se approaches zero.

The amplitude and phase measurement unit 113 measures the amplitude and phase of the input digital complex signal and the digital error complex signal Se, which are aligned in time. The digital error complex signal Se (distortion compensation signal Se) is expressed as follows:

Here, Ae(t) denotes the amplitude of the digital error complex signal, and φe(t) denotes the phase of the digital error complex signal.

According to the flowchart shown in FIG. 10, the predistortion distortion compensation unit 131 can determine the deterioration of the amplifier unit 107 for which predistortion distortion compensation is performed, and can also obtain the AM/AM characteristics and AM/PM characteristics of the amplifier unit 107.

Here, the initial values of the digital signal processing parameters of the predistortion distortion compensation unit 131 are determined in accordance with the AM/AM characteristics and AM/PM characteristics of the amplifier unit 107 at the time of manufacture so that the digital error complex signal Se approaches 0. Even if the characteristics of the amplifier unit 107 change due to aging or the like, the adaptive operation unit 140 causes the predistortion distortion compensation performed by the predistortion distortion compensation unit 131 to follow them.

In other words, changes in the digital signal processing parameters (coefficients of the polynomial approximation equation, conversion coefficients of the LUT) of the predistortion distortion compensation unit 131 represent changes in the AM/AM characteristics and AM/PM characteristics of the amplifier unit 107.

Therefore, steps ST101 and steps ST103 to ST107 in the flowchart shown in FIG. 10 may be omitted, and the parameters of the digital signal processing of the predistortion distortion compensation unit 131 (coefficients of the polynomial approximation equation, conversion coefficients of the LUT) may be set to AM/AM characteristics and AM/PM characteristics.

The functions of analyzing, recording, or storing the above-mentioned AM/AM characteristics and AM/PM characteristics may be performed by a configuration provided in the wireless device 101, or may be performed by a configuration provided in another device connected via a network, etc. Furthermore, some of the above-mentioned functions may be performed by a configuration provided in the wireless device 101, and other functions may be performed by a configuration provided in another device connected via a network, etc.

In addition, the voltage measured by the voltage measuring unit 115, the current measured by the current measuring unit 117, the transmission power output by the transmission power measuring unit 120, and the modulation accuracy shown in FIG. 2 may be measured and recorded simultaneously.

<Identifying deteriorating components>
Next, a method of identifying a degraded component using a measurement value other than the measurement value input from amplitude and phase measurement section 113 will be described.

First, the memory unit 22 stores the transmission power measurement value measured by the transmission power measurement unit 120 shown in FIG. 2, the voltage measurement value measured by the voltage measurement unit 115 shown in FIG. 2, and the current measurement value measured by the current measurement unit 117 shown in FIG. 2.

If the determination unit 32 determines that the amplifier unit 107 is degraded as shown in step ST108 in FIG. 10, it further identifies the degraded components according to the following criteria:

First, if there is a decrease in output (measured transmission power) and a change in the measured current, the determination unit 32 identifies the power transistor 107a in the amplifier unit 107 as a degraded component.

Next, if the output (transmission power measurement value) and the current measurement value do not change and the error amplitude increases in proportion to the magnitude of the input amplitude, the determination unit 32 identifies the local oscillator unit 106 as a component that has deteriorated due to a decrease in the output level.

Next, if the output (transmission power measurement value) and the current measurement value do not change and the error phase is large regardless of the magnitude of the input amplitude, the determination unit 32 identifies the power supply unit 114 as a component that has deteriorated due to increased ripple or noise interference from the ground.

<About AM/AM measurement, AM/PM measurement, and frequency characteristic measurement using test signals>
In the above, it is assumed that user data during operation is used to obtain the AM/AM characteristics and AM/PM characteristics of the amplifier section 107. However, if there is a time when user data does not need to be transmitted, a test signal may be generated and measured instead of the user data.

The test signal is a triangular wave whose amplitude changes linearly and periodically from a minimum value to a maximum value, or a signal generated by switching the frequency of that wave at an even slower cycle. Using the test signal makes it easier to perform AM/AM and AM/PM measurements. It also makes it easier to measure frequency characteristics.

The operation of the control unit 20 in the above case is described below.

Figure 15 is a flowchart showing an example of the operation of the control unit 20 in this embodiment.

First, in step ST121, the control unit 20 determines whether or not communication is unnecessary based on the wireless communication standard, etc. Then, if communication is unnecessary, that is, if it corresponds to "YES" branching from step ST121 shown in an example in FIG. 15, the process proceeds to step ST122 shown in an example in FIG. 15. On the other hand, if communication is unnecessary (i.e., communication is necessary), that is, if it corresponds to "NO" branching from step ST121 shown in an example in FIG. 15, the process proceeds to step ST112 shown in an example in FIG. 15.

If it is determined in step ST121 that communication is not required, the control unit 20 measures the amplitude and phase shift as follows.

In step ST122, the control unit 20 causes the data generation unit 102 to generate and output a test signal instead of user data.

Next, in step ST123, the control unit 20 instructs the high-frequency transmission unit 108 to stop outputting to the antenna 109.

Next, in step ST124, the control unit 20 causes the amplitude and phase measurement unit 113 to measure the amplitude and phase.

Next, in step ST125, the control unit 20 records the amplitude and phase measured by the amplitude and phase measurement unit 113 as data in the memory unit 22.

On the other hand, if it is determined in step ST121 that communication is not necessary (is necessary), the control unit 20 returns the wireless device 101 to basic operation as follows.

In step ST112, the control unit 20 instructs the data generation unit 102 to select and output the user data.

Next, in step ST113, the control unit 20 causes the high-frequency transmission unit 108 to output to the antenna 109.

Next, in step ST104, the control unit 20 instructs the amplitude and phase measurement unit 113 to stop measuring the amplitude and phase.

Next, in step ST115, the control unit 20 stops recording the amplitude and phase data.

<When measurement timing is specified in a communication standard>
Next, an operation in the case where the measurement timing is defined in the communication standard will be described.

Figure 16 is a timing chart showing an example of the operation of the wireless device 101 when the communication standard specifies a time during which no communication is performed.

In the communication standard P01, the communication permission P02 indicates that a period of time during which communication is prohibited is set. For example, the period of time during which communication is prohibited is set as one communication pause frame every N frames for the wireless frames P03 determined by the communication standard.

The operation of the wireless device 101 in accordance with the communication standard that specifies this non-communication time will be explained with reference to Figures 2 and 16.

The data generating unit 102 selects and outputs user data when communication is permitted. The data generating unit 102 also generates and outputs a test signal when communication is prohibited.

The high-frequency transmission unit 108 outputs the amplified high-frequency signal to the antenna 109 when communication is permitted. The high-frequency transmission unit 108 also stops outputting the signal to the antenna 109 when communication is prohibited.

The amplitude and phase measurement unit 113 stops operating at the timing when communication is permitted. In addition, the amplitude and phase measurement unit 113 measures the amplitude and phase from the digital demodulated signal (demodulated complex signal) output by the orthogonal demodulation unit 112 at the timing when communication is prohibited.

When the amplitude and phase measurement unit 113 completes the measurement of the amplitude and phase, the data accumulation unit 16 adds and accumulates the measured amplitude and phase values as data. The amount of data accumulated by the data accumulation unit 16 will be huge, equivalent to the operation period of the wireless device 101. For this reason, the data accumulation unit 16 may compress the amount of data by performing processing such as averaging on multiple pieces of data.

Second Embodiment
A deterioration determination system according to the present embodiment will be described. In the following description, components similar to those described in the above embodiment are denoted by the same reference numerals, and detailed description thereof will be omitted as appropriate.

<Configuration of the Deterioration Judgment System>
In the first embodiment, a method was shown in which the amplitude and phase are measured at the input and output of the amplifier 107 and the AM/AM characteristics and the AM/PM characteristics are output using these values in order to measure the characteristics of the amplifier 107. On the other hand, in the present embodiment, a method will be described in which the output signal of the amplifier 107 is converted from a time domain signal to a frequency domain signal to obtain the spectrum characteristics.

The configuration of the communication system is, for example, similar to that shown in Figure 1.

Figure 17 shows an example of the configuration of fixed or mobile wireless equipment for this embodiment.

The wireless device 101A shown in FIG. 17 includes at least a control unit 20A, a data generation unit 102, a digital modulation unit 103, a quadrature modulation unit 104b, a frequency conversion unit (IF to RF) 105, a local oscillation unit 106, an amplifier unit 107, a high-frequency transmission unit 108, an antenna 109, a waveform detection unit 110, a feedback amplifier unit 110a, a power supply unit 114, a voltage measurement unit 115, a current sensor 116, a current measurement unit 117, a temperature sensor 118, a power detection unit 119, a transmission power measurement unit 120, an analog-to-digital conversion unit 138, a recording unit 202, a fast Fourier transform (FFT) unit 203, and a spectrum analysis unit 204. The control unit 20A includes a determination unit 26.

The orthogonal modulation unit 104b converts the digitally modulated signal output from the digital modulation unit 103 into a two-dimensional complex signal (IQ signal) that differs in phase by 90 degrees (i.e., is orthogonal) while compensating for distortion. The frequency conversion unit (IF → RF) 105 converts the two-dimensional complex signal (IQ signal) output from the orthogonal modulation unit 104b to a carrier frequency by mixing it with a local signal input from the local oscillation unit 106, and outputs a high-frequency signal.

In this embodiment, instead of the amplitude and phase measurement unit 113 in FIG. 2, a spectrum analysis unit 204 for measuring the spectrum is connected to the output side of the amplifier unit 107. The operation of measuring the spectrum will be described below with reference to FIG. 17.

A portion of the output of the amplifier 107 is extracted as a feedback signal from the waveform detector 110. The analog-to-digital converter 138 then converts the feedback signal output from the feedback amplifier 110a into a digital signal.

Then, the recording unit 202 accumulates and records the digital signal output from the analog-to-digital conversion unit 138 in a recording device such as a memory for a certain period of time.

Then, the FFT unit 203 converts the digital signal recorded in the recording unit 202 for a certain period of time into a digital signal in the frequency domain by fast Fourier transform.

Furthermore, the spectrum analysis unit 204 extracts the spectrum characteristics of the amplifier unit 107 from the frequency domain digital signal output from the FFT unit 203.

Figures 18 and 19 are conceptual diagrams showing the concept of converting a time domain signal into a frequency domain signal by FFT. In Figure 18, the vertical axis shows amplitude (power) and the horizontal axis shows time. In Figure 19, the vertical axis shows amplitude (power) and the horizontal axis shows frequency.

The signal output from the amplifier 107 is a signal having a time domain waveform as shown in FIG. 18. On the other hand, the signal output from the FFT unit 203 is a signal having a frequency domain waveform as shown in FIG. 19.

It is known that strict band limitations are imposed on digitally modulated signals in digital communication systems. The above-mentioned frequency domain signal has the characteristic that the signal power is concentrated in the in-band region, as shown in Figure 19. Specifically, the frequency domain signal has adjacent channel regions above and below the frequency of the in-band region, and an out-of-band region outside of that.

The Radio Law prescribes standards for output power (i.e., amplitude) in each frequency domain, such as the in-band domain, adjacent channel domain, or out-of-band domain. For example, if there is an unwanted emission component with output power (i.e., amplitude) that exceeds the standard in the out-of-band frequency domain, it is a violation of the Radio Law.

<Spectrum characteristics measurement method 1 (without distortion compensation)>
The spectrum analysis unit 204 analyzes the spectrum characteristics and extracts, as a feature, the margin amount of the analysis result with respect to the Radio Law standard in each frequency domain, such as the in-band domain, the adjacent channel domain, or the out-of-band domain. In this case, if a characteristic frequency is known in advance in the target bandwidth, the feature amount may be extracted by focusing on that frequency. The determination unit 26 determines that the amplifier 107 has deteriorated when the margin amount for each frequency of the analysis result falls below the reference value for the corresponding frequency (i.e., when the margin amount has decreased beyond a threshold value determined for each frequency).

<Relationship between deterioration of component characteristics and measured values>
The relationship between the deterioration of the power amplification characteristic of the wireless device 101 and the deterioration of the characteristics of the components in the wireless device 101 is as shown in FIG.

The spectrum measurement method can detect "deterioration of spurious characteristics (increase in unwanted emissions)" among the output characteristics shown in Figure 3.

In addition, it is known that the changes in the AM/AM characteristics and the AM/PM characteristics in FIG. 3 appear in the spectrum in the adjacent channel region. Therefore, the measurement method in this embodiment can be replaced with the measurement method in the first embodiment.

The characteristics of the amplifier 107 change depending on the temperature. Therefore, it is important to determine whether the change in characteristics is due to temperature or due to deterioration over time.

The determination unit 26 shown in FIG. 17 can improve the accuracy of determining deterioration of the amplifier unit 107 by determining the spectrum characteristics of the power components in the adjacent channel region or the power components in the out-of-band region while removing the effects of the temperature measured at the same time, as in the first embodiment.

<Spectrum characteristics measurement method 2 (with feedback distortion compensation)>
FIG. 20 is a diagram showing an example of the configuration of fixed or movable wireless equipment when performing feedback type distortion compensation.

The wireless device 101B shown in FIG. 20 includes at least a control unit 20A, a data generation unit 102, a digital modulation unit 103, a feedback-type distortion compensation and orthogonal modulation unit 104, a frequency conversion unit (IF to RF) 105, a local oscillation unit 106, an amplifier unit 107, a high-frequency transmission unit 108, an antenna 109, a waveform detection unit 110, and a feedback amplifier unit 112 that amplifies the output from the waveform detection unit 110. 0b, a frequency conversion unit (RF → IF) 111 to which the output from the feedback amplifier 110b is input, an orthogonal demodulation unit 112, a power supply unit 114, a voltage measurement unit 115, a current sensor 116, a current measurement unit 117, a temperature sensor 118, a power detection unit 119, a transmission power measurement unit 120, an analog-to-digital conversion unit 201, a recording unit 202, an FFT unit 203, and a spectrum analysis unit 204. The control unit 20A includes a determination unit 26.

In this embodiment, instead of the feedback signal output from the amplifier 107, the distortion compensation signal Se output from the distortion compensation and orthogonal modulation unit 104 is FFT transformed.

The analog-to-digital conversion unit 201 converts the distortion compensation signal Se output by the distortion compensation and quadrature modulation unit 104 into a digital signal.

Then, the recording unit 202 accumulates and records the digital signal output from the analog-to-digital conversion unit 201 in a recording device such as a memory for a certain period of time.

Then, the FFT unit 203 converts the digital signal recorded in the recording unit 202 for a certain period of time into a digital signal in the frequency domain by fast Fourier transform.

Furthermore, the spectrum analysis unit 204 extracts the spectrum characteristics of the amplifier unit 107 from the frequency domain digital signal output from the FFT unit 203.

As described in the first embodiment, in feedback distortion compensation, the distortion compensation signal is a signal that has the inverse characteristics of the distortion characteristics of the amplifier 107. Therefore, the spectrum analysis of the distortion compensation signal is equivalent to the spectrum analysis of the output of the amplifier 107 in the absence of feedback.

<Spectral characteristics measurement method 3 (with predistortion compensation)>
Fig. 21 is a diagram showing an example of a configuration including a distortion compensation and quadrature modulation unit. In Fig. 21, a predistortion distortion compensation unit 131 is provided. The predistortion distortion compensation unit 131 is configured by adaptive digital predistortion. The predistortion distortion compensation unit 131 converts an input digital complex signal into a distortion-compensated digital complex signal having an inverse distortion of the distortion of the amplifier by digital signal processing, and outputs the signal.

In the configuration shown in FIG. 21, instead of the feedback signal that is the output of the amplifier 107, the error signal output by the input/output comparator 139 is FFT transformed.

The digital-to-analog conversion unit 132 converts the distortion-compensated digital complex signal output from the predistortion distortion compensation unit 131 into a distortion-compensated analog complex signal and outputs it. The orthogonal modulation unit 133 performs orthogonal modulation on the distortion-compensated analog complex signal output from the digital-to-analog conversion unit 132, and outputs it to the frequency conversion unit (IF → RF) 105 as a distortion-compensated signal Se.

The orthogonal demodulation unit 112 orthogonally demodulates the IF (intermediate frequency) signal output from the frequency conversion unit (RF → IF) 111, i.e., the feedback signal, and outputs it as a demodulated complex signal (IQ signal). The analog-to-digital conversion unit 138 converts the analog demodulated complex signal (IQ signal) output from the orthogonal demodulation unit 112 into a digital demodulated complex signal and outputs it.

The FFT unit 203 converts the digital error signal recorded in the recording unit 202 for a certain period of time into a digital signal in the frequency domain by fast Fourier transform.

Furthermore, the spectrum analysis unit 204 extracts the spectrum characteristics of the amplifier unit 107 from the frequency domain digital signal output from the FFT unit 203.

As explained in the first embodiment, in the predistortion distortion compensation, the predistortion parameters (such as the coefficients of the Look Up Table) are determined so that the error signal becomes zero, and the error signal becomes a signal that represents the change in distortion of the amplifier 107.

Therefore, the spectrum analysis of the error signal is equivalent to the spectrum analysis of the distortion output of the amplifier unit 107.

<Identifying deteriorating components>
We describe how measurements other than spectral characteristics can be used to identify degrading components.

First, the memory unit 22 records the transmission power measurement value measured by the transmission power measurement unit 120 shown in FIG. 17, the voltage measurement value measured by the voltage measurement unit 115 shown in FIG. 17, and the current measurement value measured by the current measurement unit 117 shown in FIG. 17 together with the spectrum characteristics.

If the determination unit 26 determines that the amplifier unit 107 is degraded as shown in step ST108 in FIG. 10, it further identifies the degraded components according to the following criteria:

First, if there is a decrease in output (measured transmission power) and a change in the measured current, the determination unit 26 identifies the power transistor (amplification unit 107) as a degraded component.

Next, if the output (transmission power measurement value) and the current measurement value do not change and the overall power of the adjacent channel region increases, the determination unit 26 identifies the local oscillator unit 106 as a component that has deteriorated due to a decrease in the output level.

Next, if the output (transmission power measurement) and current measurement do not change and the unwanted emission components of a particular frequency are large, the determination unit 26 identifies the power supply unit 114 as a component that has deteriorated due to increased ripple or noise interference from the ground.

The recording unit 202, the FFT unit 203, the spectrum analysis unit 204, and the determination unit 26 may be provided in the wireless device 101B, or may be provided in another device connected via a network or the like.

In addition, the voltage measured by the voltage measuring unit 115, the current measured by the current measuring unit 117, and the transmission power output by the transmission power measuring unit 120 shown in FIG. 2 or FIG. 20 may be measured and recorded simultaneously.

<Spectrum characteristic analysis and frequency characteristic measurement using test signals>
In the above, it is assumed that user data in operation is used when analyzing the spectrum characteristics of the amplifier section 107. However, if there is a time when user data does not need to be transmitted, a test signal may be generated and measured instead of the user data.

The test signal can be a sine wave, a signal that combines two sine waves detuned at any frequency, or a signal that is generated by switching the frequency of these at an even slower cycle. Using a test signal makes it easier to analyze spectrum characteristics. It also makes it easier to measure frequency characteristics.

For example, a sine wave has no fluctuations in the signal envelope, making it suitable for finding unwanted emission components. Also, it is known that a signal that is the combination of two sine waves detuned at an arbitrary frequency will have distortion components at frequencies that are an integer multiple of the detuning frequency width away from the frequency of the input signal, making it suitable for measuring the amount of distortion at a pinpoint frequency.

Regarding the timing of using the test signal, as in the first embodiment, the wireless communication standard provides a time period during which periodic communication is not required, and the test signal may be generated during that time period to perform measurements, as in the case of the timing chart shown in Figure 16.

Third Embodiment
The deterioration determination system, learning device, and inference device according to the present embodiment will be described. In the following description, components similar to those described in the above embodiment are illustrated with the same reference numerals, and detailed description thereof will be omitted as appropriate.

<Configuration of the Deterioration Judgment System>
In the first and second embodiments, the amplifier 107 has been described as having a known criterion for determining deterioration of the components. On the other hand, in the present embodiment, a method for improving the accuracy of determination (inference) by learning a determination method using a learning device and an inference device will be described. It is obvious that the present embodiment can be combined with the first or second embodiment.

The configuration of the communication system is, for example, similar to that shown in Figure 1.

<About the learning phase>
22 is a diagram conceptually illustrating an example of the configuration of a learning device that performs machine learning on wireless equipment. As illustrated in the example of FIG. 22, a learning device 300 includes a data acquisition unit 302 and a model generation unit 304.

The data acquisition unit 302 acquires the above-mentioned time-series characteristic data, measurement data (data such as AM/AM characteristics, AM/PM characteristics, spectrum characteristics, temperature measurements, current measurements, voltage measurements, or transmission power measurements), and manufacturing data of the wireless equipment (manufacturing lot, parts lot, manufacturing factory, inspection data, etc.) as data B11. Of the data B11, the time-series characteristic data and measurement data (data such as AM/AM characteristics, AM/PM characteristics, spectrum characteristics, temperature measurements, current measurements, voltage measurements, or transmission power measurements) can be acquired, for example, from the data accumulation unit 16 described below. Of the data B11, the manufacturing data of the wireless equipment (manufacturing lot, parts lot, manufacturing factory, inspection data, etc.) can be acquired, for example, from the manufacturing information management server 310 described below. The data acquisition unit 302 also acquires the component in which the failure occurred and the time from the start of operation of the device until the failure occurred as data B12 (correct answer data). Here, manufacturing conditions such as manufacturing lot, parts lot, and manufacturing factory generally affect the manufacturing results (for example, the rate of deterioration). The component in which the failure occurred and the time from the start of operation of the device until the failure occurred are also referred to as the failure judgment result.

Fig. 23 is a diagram showing an example of a method for acquiring training data. As shown in the example in Fig. 23, data B11 including time-series characteristic data and measurement data, and data B12 (correct answer data) including the component in which a failure occurred and the time from the start of operation of the device until the failure occurred are accumulated in the data accumulation unit 16. The data acquisition unit 302 can acquire the data accumulated in the data accumulation unit 16.

Here, data B11 may be acquired from wireless device 101 by applying stress stronger than that in the actual usage environment to accelerate the deterioration of the components, or may be acquired by adding the failure determination result of data B12 to data B11 of wireless device 101 that has failed in the actual usage environment. Furthermore, in addition to data acquired from an actual wireless device 101, data to which various variations have been simulated by simulation or the like may be added.

The model generation unit 304 learns data C11 indicating a degraded component and data C12 indicating the remaining life of the degraded component based on training data created by combining data B11 and data B12 output from the data acquisition unit 302. That is, based on data B11 including time-series characteristic data and measurement data measured by the wireless device 101 and data B12 (correct answer data) including a failure determination result, a learned model is generated that infers data C12 indicating a degraded component and data C12 indicating the remaining life of the degraded component. For example, data C12 indicating the remaining life can be inferred based on the difference between data B12 and the operating time of the device.

Here, the training data is data that associates data B11, data B12, data C11, and data C12 with each other.

It is known that when the amplifier unit 107 of the wireless device 101 from which data is acquired fails due to aging, its characteristics gradually deteriorate from a certain point in time, and when a certain threshold is exceeded, the deterioration of the characteristics accelerates rapidly (i.e., it fails). This is because the amplifier unit 107 is a component of the wireless device 101 that operates with a relatively large amount of power.

There is a relationship between changes in characteristics and deterioration of components, as shown in Figure 3, so it is possible to infer which components are deteriorating based on a combination of changes in characteristics.

For example, if the current consumption of the amplifier unit 107 and the AM/AM and AM/PM characteristics change simultaneously, it can be estimated that there is a high possibility of degradation of the power transistor.

It is clear from Figure 3 that there is a correlation between various characteristic data (or measurement data). Therefore, by using more data to learn the correlation between various characteristic data, it is possible to improve the accuracy of identifying deteriorating components. It is also clear that if deteriorating components can be identified with high accuracy, the accuracy of estimating their remaining lifespan based on changes in characteristics will also improve.

The learning device (or inference device) is used to learn data C11 indicating degraded components of the wireless device 101 and data C12 indicating the remaining life time of the degraded components, but the learning device (or inference device) may be a device separate from the wireless device 101, for example, connected to the wireless device 101 via a network. Alternatively, the learning device (or inference device) may be built into the wireless device 101. Furthermore, the learning device (or inference device) may exist on a cloud server.

The model generation unit 304 learns the data C11 and C12, for example, according to a neural network model, by so-called supervised learning. Here, supervised learning refers to a method of providing a learning device with a combination of input data and result (label) data, learning the features present in the training data, and inferring the result data from the input data.

A neural network is composed of an input layer consisting of multiple neurons, an intermediate layer (hidden layer) consisting of multiple neurons, and an output layer consisting of multiple neurons. The intermediate layer may be one layer or two or more layers.

Figure 24 is a diagram showing an example of a neural network. In a three-layer neural network such as the one shown in Figure 24, when multiple pieces of input data are input to the input layer (X1-X3), the data are multiplied by weight W1 (w11-w16) and input to the intermediate layer (Y1-Y2), and the result is further multiplied by weight W2 (w21-w26) and output from the output layer (Z1-Z3). This output result changes depending on the value of weight W1 and the value of weight W2.

In this embodiment, the neural network learns data C11 and data C12 by so-called supervised learning according to training data created by combining data B11 and data B12 (correct answer data) acquired by the data acquisition unit 302.

In other words, the neural network learns by inputting data B11, which includes time-series characteristic data and measurement data, into the input layer and adjusting weights W1 and W2 so that the result output from the output layer approaches data B12 (correct answer data), which includes the component where the failure occurred and the time from the start of operation of the device until the failure occurred.

The model generation unit 304 generates a trained model by performing the above-mentioned machine learning, and outputs the trained model.

The trained model storage unit 306 stores the trained model output from the model generation unit 304.

Next, the process by which the learning device learns will be described with reference to FIG. 25. FIG. 25 is a flowchart showing the learning process of the learning device.

First, in step ST130, the data acquisition unit 302 acquires data B11 including time-series characteristic data and measurement data, and data B12 (correct answer data) including the component in which the failure occurred and the time from the start of operation of the device until the failure occurred.

In the above, data B11 and data B12 are acquired simultaneously, but as long as data B11 and data B12 are associated with each other and input to the learning device 300, data B11 and data B12 may be acquired at different times.

Next, in step ST131, the model generation unit 304 learns data C11 indicating degraded components of the wireless device 101 and data C12 indicating the remaining life time of the degraded components by so-called supervised learning according to the training data created by combining data B11 and data B12 acquired by the data acquisition unit 302, and generates a learned model.

Next, in step ST132, the trained model storage unit 306 stores the trained model generated by the model generation unit 304.

<About the utilization phase>
26 is a diagram conceptually illustrating an example of the configuration of an inference device related to wireless equipment. As illustrated in the example of FIG. 26, an inference device 400 includes a data acquisition unit 302, an inference unit 308, and a notification unit 312.

The data acquisition unit 302 acquires data B10, which includes various time-series characteristic data and measurement data related to the wireless device 101, from the data storage unit 16.

The inference unit 308 infers data C11 indicating degraded components of the wireless device 101 and data C12 indicating the remaining life time of the degraded components, which are obtained using the learned model stored in the learned model storage unit 306.

In other words, the inference unit 308 can input data B10 acquired by the data acquisition unit 302 to the learned model obtained from the learned model storage unit 306, and output data C11 and data C12 inferred from data B10.

In the present embodiment, the inference unit 308 has been described as outputting the deteriorated data C11 and data C12 using a trained model trained based on data obtained from the wireless device 101 (and further data stored in the data storage unit 16). However, as shown in FIG. 26, the inference unit 308 may output the data C11 and data C12 using a trained model trained based on various manufacturing data acquired from external devices (e.g., the manufacturing information management server 310) such as other wireless equipment connected via a network.

Next, the process by which the inference device obtains data C11 and data C12 will be described with reference to FIG. 27. FIG. 27 is a flowchart showing the inference process of the inference device.

First, in step ST140, the data acquisition unit 302 acquires data B10 including various time-series characteristic data and measurement data.

Next, in step ST141, the inference unit 308 inputs data B10 into the learned model stored in the learned model storage unit, and as a result, obtains (infers) data C11 indicating the degraded components of the wireless device 101 and data C12 indicating the remaining life time of the degraded components.

Next, in step ST142, the notification unit 312 outputs and notifies the maintenance personnel of the wireless device 101 or the maintenance management server of the data C11 and data C12 obtained in step ST141.

Next, in step ST143, the maintenance technician or maintenance management server of the wireless device 101 uses data C11 and data C12 to determine whether or not the wireless device 101 needs to be replaced or the component identified as degraded needs to be repaired.

By performing the above operations, preventive maintenance can be performed on parts or components that show signs of failure (i.e., parts or components with a short remaining life) at an appropriate timing before the wireless device 101 fails or begins to operate abnormally. During preventive maintenance, it is desirable to maintain the state of each component so that the remaining life is sufficiently longer than the time required for repair or replacement.

This makes it possible to reduce the number of regular inspections of the performance of the wireless device 101 (i.e., periodic inspections), and to reduce the number of spare parts required to ensure the reliability and availability of the communication system 1. As a result, the running costs of the communication system that uses the wireless device 101 can be reduced.

In this embodiment, a case has been described in which supervised learning is applied to the learning algorithm used by the model generation unit 304, but the learning algorithm is not limited to this. As for the learning algorithm, in addition to supervised learning, reinforcement learning, unsupervised learning, semi-supervised learning, etc. can also be applied.

The model generation unit 304 may learn the data C11 and the data C12 according to training data created for multiple wireless devices 101. The model generation unit 304 may acquire training data from multiple wireless devices 101 used in the same area, or may learn the data C11 and the data C12 using training data collected from multiple wireless devices 101 operating independently in different areas.

It is also possible to add or remove a wireless device 101 that collects training data during learning. Furthermore, a learning device 300 that has learned data C11 and data C12 for a certain wireless device 101 may be applied to another wireless device 101, and data C11 and C12 for the other wireless device 101 may be re-learned and updated.

The learning algorithm used in the model generation unit 304 may be deep learning, which learns to extract the features themselves, or machine learning may be performed according to other known methods, such as genetic programming, functional logic programming, or support vector machines.

<Effects of the above-described embodiment>
Next, examples of effects produced by the above-described embodiments are shown. In the following description, the effects are described based on the specific configurations shown as examples in the above-described embodiments, but they may be replaced with other specific configurations shown as examples in the present specification as long as the same effects are produced. In other words, for convenience, only one of the corresponding specific configurations may be described as a representative below, but the representatively described specific configuration may be replaced with another corresponding specific configuration.

Furthermore, the replacement may be made across multiple embodiments. In other words, configurations shown as examples in different embodiments may be combined to produce the same effect.

According to the embodiment described above, the degradation determination system includes an amplifier 107 (or an amplifier 207) and a determination unit 32. The amplifier 107 amplifies and outputs an input radio signal. The determination unit 32 determines that the amplifier 107 is degraded when the radio signal output from the amplifier 107 exceeds an amplitude threshold and a phase threshold that are determined for each amplitude of the corresponding radio signal input to the amplifier 107.

With this configuration, deterioration of the amplifier unit 107 can be determined based on changes in the amplitude and phase of the radio signal output from the amplifier unit 107, so that failure of the amplifier unit 107 can be predicted and preventive maintenance can be performed appropriately. This makes it possible to reduce the number of regular inspections of the performance of the wireless device 101, and also to reduce the number of spare parts required to ensure the reliability and availability of the communication system 1. As a result, the running costs of the communication system 1 that uses the wireless device 101 can be reduced.

The same effect can be achieved even if other configurations, examples of which are shown in this specification, are added to the above configuration, i.e., other configurations in this specification that are not mentioned as the above configuration are added.

According to the embodiment described above, the deterioration determination system includes a sorting unit 24. The sorting unit 24 acquires the amplitude and phase of the corresponding radio signal output from the amplifier 107 in the order of the amplitude of the radio signal input to the amplifier 107. The sorting unit 24 then acquires the respective relationships with the amplitude of the radio signal input to the amplifier 107 (for example, input amplitude vs. error amplitude (AM/AM characteristics), input amplitude vs. error phase (AM/PM characteristics)). According to this configuration, a table of input amplitude vs. error amplitude (AM/AM characteristics) and a table of input amplitude vs. error phase (AM/PM characteristics) can be generated and used to determine deterioration of the amplifier 107.

According to the embodiment described above, the deterioration determination system includes a distortion compensation unit 104a. The distortion compensation unit 104a compensates for the distortion of the amplifier unit 107 by feeding back the radio signal output from the amplifier unit 107. Here, the radio signal input to the amplifier unit 107 is the radio signal input via the distortion compensation unit 104a. The determination unit 32 determines that the amplifier unit 107 is deteriorated when the radio signal output from the distortion compensation unit 104a exceeds an amplitude threshold value and a phase threshold value determined for each amplitude of the corresponding radio signal input to the distortion compensation unit 104a. According to this configuration, deterioration of the amplifier unit 107 can be determined based on changes in the amplitude and phase of the radio signal output from the distortion compensation unit 104a, so that a failure of the amplifier unit 107 can be predicted and preventive maintenance can be appropriately performed.

Also, according to the embodiment described above, the deterioration judgment system includes a distortion compensation unit. Here, the distortion compensation unit corresponds to, for example, the predistortion distortion compensation unit 131. The predistortion distortion compensation unit 131 compensates for the distortion of the amplifier unit 107 by a predistortion method. Here, the radio signal input to the amplifier unit 107 is a radio signal input via the predistortion distortion compensation unit 131. The deterioration judgment system also includes a delay unit 141 and an input/output comparison unit 139. The delay unit 141 delays the radio signal input to the predistortion distortion compensation unit 131 to align the timing with the radio signal output from the amplifier unit 107. The input/output comparison unit 139 outputs a difference signal (digital error complex signal Se) based on the difference between the radio signal input to the predistortion distortion compensation unit 131 and the radio signal output from the amplifier unit 107 to the predistortion distortion compensation unit 131. The determination unit 32 then determines that the amplifier unit 107 is degraded when the differential signal exceeds an amplitude threshold value and a phase threshold value that are determined for each amplitude of the corresponding wireless signal input to the predistortion distortion compensation unit 131. With this configuration, it is possible to determine the degradation of the amplifier unit 107 based on changes in the amplitude and phase of the differential signal output from the input/output comparison unit 139, and therefore it is possible to predict failure of the amplifier unit 107 and perform appropriate preventive maintenance.

Furthermore, according to the embodiment described above, the degradation determination system includes an adaptive operation unit 140. The adaptive operation unit 140 adapts the parameters of the digital signal processing in the predistortion distortion compensation unit 131 based on the differential signal. Then, the determination unit 32 determines that the amplifier unit 107 has deteriorated based on the parameters of the digital signal processing in the predistortion distortion compensation unit 131. With this configuration, degradation of the amplifier unit 107 can be determined based on the parameters of the digital signal processing adapted by the adaptive operation unit 140, so that failure of the amplifier unit 107 can be predicted and preventive maintenance can be appropriately performed.

Furthermore, according to the embodiment described above, the degradation determination system includes an amplitude and phase measurement unit 113 and a temperature sensor 118. The amplitude and phase measurement unit 113 measures the amplitude and phase of the wireless signal. The temperature sensor 118 measures the temperature of the amplifier 107. The measured values of the amplitude and phase of the wireless signal output from the amplifier 107 are then classified based on the temperature of the amplifier 107. With this configuration, the effect of temperature on the amplitude and phase of the wireless signal can be suppressed, thereby improving the determination accuracy.

Furthermore, according to the embodiment described above, the degradation determination system includes a transmission power measurement unit 120, a voltage measurement unit 115, and a current measurement unit 117. The transmission power measurement unit 120 measures the transmission power, which is the power of the radio signal output by the amplifier unit 107. The voltage measurement unit 115 measures the bias voltage for supplying power to the amplifier unit 107. The current measurement unit 117 measures the supply current, which is the current for supplying power to the amplifier unit 107. Then, when the determination unit 32 determines that the amplifier unit 107 is deteriorated, and further when the transmission power and the supply current are changed, it determines the deterioration of the power transistor 107a in the amplifier unit 107, and when the transmission power and the supply current are not changed and the distortion compensation signal (error signal) output from the distortion compensation unit 104a in proportion to the amplitude of the input signal input to the distortion compensation unit 104a exceeds the amplitude threshold, it determines the output level of the local oscillation unit 106 to be lowered, and when the transmission power and the supply current are not changed and the distortion compensation signal (error signal) output from the distortion compensation unit 104a exceeds the phase threshold regardless of the amplitude of the input signal input to the distortion compensation unit 104a, it determines the increase of the ripple of the power supply unit 114 or the introduction of noise from the ground. Here, the local oscillation unit 106 oscillates a local signal used for frequency conversion of the radio signal input to the amplifier unit 107. Also, the power supply unit 114 supplies a bias voltage to the amplifier unit 107. With this configuration, it is possible to identify components that are degraded based on the transmission power, bias voltage, and supply current, narrowing down the components that need repair or replacement and enabling efficient maintenance.

Furthermore, according to the embodiment described above, the degradation determination system includes a data generating unit 102. The data generating unit 102 generates a test signal for measuring the characteristics of the amplifier unit 107. The data generating unit 102 then switches from user data to the test signal at a timing when communication is not required, and inputs a wireless signal to the amplifier unit 107. According to such a configuration, the use of the test signal makes it easier to perform AM/AM measurement and AM/PM measurement. Furthermore, it also makes it easier to measure frequency characteristics.

Furthermore, according to the embodiment described above, the test signal is a triangular wave whose amplitude changes linearly and periodically from a minimum value to a maximum value, or a signal generated by switching the frequency of the triangular wave at an even slower period. With such a configuration, the use of the test signal makes it easy to perform AM/AM measurement and AM/PM measurement. Furthermore, it also makes it easy to measure frequency characteristics.

Furthermore, according to the embodiment described above, the data generating unit 102 switches from user data to a test signal at a timing when communication is not required as defined by the wireless communication standard, and inputs the wireless signal to the amplifying unit 107. With such a configuration, the use of a test signal while complying with the wireless communication standard makes it easier to perform AM/AM measurement and AM/PM measurement. Furthermore, it also makes it easier to measure frequency characteristics.

According to the embodiment described above, the degradation determination system includes an amplifier 107, an analog-digital converter 138 (or an analog-digital converter 201), an FFT unit 203, a spectrum analyzer 204, and a determination unit 26. The amplifier 107 amplifies and outputs the input radio signal. The analog-digital converter 138 performs analog-digital conversion on the radio signal output from the amplifier 107. The FFT unit 203 performs fast Fourier transform processing on the radio signal that has been analog-digitally converted by the analog-digital converter 138. The spectrum analyzer 204 analyzes the spectrum characteristics of the radio signal that has been fast Fourier transform processed by the FFT unit 203. The determination unit 26 determines that the amplifier 107 is degraded when the spectrum characteristics exceed a threshold value determined for each frequency of the corresponding radio signal input to the amplifier 107.

With this configuration, deterioration of the amplifier unit 107 can be determined based on the spectrum characteristics obtained from the radio signal output from the amplifier unit 107, so that failure of the amplifier unit 107 can be predicted and preventive maintenance can be performed appropriately. This makes it possible to reduce the number of regular inspections of the performance of the wireless device 101, and also to reduce the number of spare parts required to ensure the reliability and availability of the communication system 1. As a result, the running costs of the communication system 1 that uses the wireless device 101 can be reduced.

According to the embodiment described above, the deterioration determination system includes a distortion compensation unit. Here, the distortion compensation unit corresponds to, for example, the distortion compensation and orthogonal modulation unit 104. The distortion compensation and orthogonal modulation unit 104 compensates for the distortion of the amplifier unit 107 by feeding back the radio signal output from the amplifier unit 107. Here, the radio signal input to the amplifier unit 107 is the radio signal input via the distortion compensation and orthogonal modulation unit 104. Then, the analog-digital conversion unit 201 performs analog-digital conversion on the radio signal output from the distortion compensation and orthogonal modulation unit 104. According to this configuration, deterioration of the amplifier unit 107 can be determined based on the spectrum characteristics obtained from the radio signal output from the distortion compensation and orthogonal modulation unit 104, so that a failure of the amplifier unit 107 can be predicted and preventive maintenance can be appropriately performed.

According to the embodiment described above, the degradation determination system includes a predistortion distortion compensation unit 131. The predistortion distortion compensation unit 131 compensates for the distortion of the amplifier unit 107 by a predistortion method. Here, the radio signal input to the amplifier unit 107 is a radio signal input via the predistortion distortion compensation unit 131. The degradation determination system also includes a delay unit 141 and an input/output comparison unit 139. The delay unit 141 delays the radio signal input to the predistortion distortion compensation unit 131 to align the timing with the radio signal output from the amplifier unit 107. The input/output comparison unit 139 outputs a differential signal based on the difference between the radio signal input to the predistortion distortion compensation unit 131 and the radio signal output from the amplifier unit 107 to the predistortion distortion compensation unit 131. Then, the FFT unit 203 performs fast Fourier transform processing on the differential signal. Furthermore, the spectrum analysis unit 204 analyzes the spectrum characteristics of the difference signal that has been subjected to the fast Fourier transform process by the FFT unit 203. With this configuration, it is possible to determine the deterioration of the amplifier unit 107 based on the spectrum characteristics obtained from the difference signal output from the input/output comparison unit 139, so that it is possible to predict failure of the amplifier unit 107 and perform appropriate preventive maintenance.

Furthermore, according to the embodiment described above, the degradation determination system includes an amplitude and phase measurement unit 113 and a temperature sensor 118. The amplitude and phase measurement unit 113 measures the amplitude and phase of the wireless signal. The temperature sensor 118 measures the temperature of the amplifier 107. The measured values of the amplitude and phase of the wireless signal output from the amplifier 107 are classified based on the temperature of the amplifier 107. With this configuration, the effect of temperature on the amplitude and phase of the wireless signal can be suppressed, thereby improving the determination accuracy.

According to the embodiment described above, the degradation determination system includes a transmission power measurement unit 120, a voltage measurement unit 115, and a current measurement unit 117. The transmission power measurement unit 120 measures the transmission power, which is the power of the radio signal output by the amplifier unit 107. The voltage measurement unit 115 measures a bias voltage for supplying power to the amplifier unit 107. The current measurement unit 117 measures the supply current, which is the current for supplying power to the amplifier unit 107. Then, when the determination unit 26 determines that the amplifier unit 107 is degraded, and further, when the transmission power and the supply current are changed, it identifies the degradation of the power transistor, when the transmission power and the supply current do not change and the total power of the adjacent channel region increases, it identifies a decrease in the output level of the local oscillation unit 106, when the transmission power and the supply current do not change and the unnecessary emission component of a specific frequency is large, it identifies an increase in the ripple of the power supply unit 114 or noise mixing from the ground. Here, the local oscillation unit 106 oscillates a local signal used for frequency conversion of the radio signal input to the amplifier unit 107. In addition, the power supply unit 114 supplies a bias voltage to the amplifier unit 107. With this configuration, it is possible to identify components that are degraded based on the transmission power, bias voltage, and supply current, narrowing down the components that need repair or replacement and enabling efficient maintenance.

Furthermore, according to the embodiment described above, the degradation determination system includes a data generating unit 102. The data generating unit 102 generates a test signal for measuring the characteristics of the amplifier unit 107. The data generating unit 102 then switches from user data to the test signal at a timing when communication is not required, and inputs a wireless signal to the amplifier unit 107. According to such a configuration, the use of the test signal makes it easier to perform AM/AM measurement and AM/PM measurement. Furthermore, it also makes it easier to measure frequency characteristics.

Furthermore, according to the embodiment described above, the test signal is a sine wave, or a signal that is a combination of two sine waves detuned at any frequency, or a signal that is generated by switching the frequency of these at an even slower period. With such a configuration, spurious components or third-order distortion can be easily extracted from the spectrum characteristics of the amplifier unit 107.

Furthermore, according to the embodiment described above, the data generating unit 102 switches from user data to a test signal at a timing when communication is not required as defined by the wireless communication standard, and inputs the wireless signal to the amplifying unit 107. With such a configuration, the use of a test signal while complying with the wireless communication standard makes it easier to perform AM/AM measurement and AM/PM measurement. Furthermore, it also makes it easier to measure frequency characteristics.

According to the embodiment described above, the learning device includes a data storage unit 16, a manufacturing information management server 310, and a model generation unit 304. The data storage unit 16 stores, as time-series data, at least the amplitude/amplitude characteristics and the amplitude/phase characteristics (which may further include spectrum characteristics, temperature, channel frequency, transmission power measurement value, bias voltage measurement value, and current measurement value) obtained at regular intervals from the amplifier unit 107 provided in any of the above-mentioned deterioration judgment systems. The manufacturing information management server 310 stores the manufacturing lot, parts lot, manufacturing plant, and inspection data of the wireless equipment as manufacturing data. The model generation unit 304 generates a trained model that infers the deteriorated components and the remaining life time of the deteriorated components based on the time-series data stored in the data storage unit 16, the manufacturing data, data indicating the components in which a failure has occurred, and data indicating the time until the failure occurs. According to this configuration, the accuracy of the judgment (inference) can be improved by learning a judgment method based on the time-series data stored in the data storage unit 16, the manufacturing data, and the failure judgment result.

According to the embodiment described above, the learning device includes a data storage unit 16, a manufacturing information management server 310, and a model generation unit 304. The data storage unit 16 stores, as time-series data, at least the amplitude/amplitude characteristics and the amplitude/phase characteristics (which may further include spectrum characteristics, temperature, channel frequency, transmission power measurement value, bias voltage measurement value, and current measurement value) obtained at regular intervals from the amplifier unit 107 provided in the wireless equipment. The manufacturing information management server 310 stores the manufacturing lot, parts lot, manufacturing plant, and inspection data of the wireless equipment as manufacturing data. The model generation unit 304 generates a trained model that infers the deteriorated components and the remaining life time of the deteriorated components based on the time-series data stored in the data storage unit 16, the manufacturing data, data indicating the components in which a failure has occurred, and data indicating the time until the failure occurs.

With this configuration, the accuracy of judgment (inference) can be improved by learning a judgment method based on the time-series data, manufacturing data, and fault judgment results stored in the data storage unit 16.

The same effect can be achieved even if other configurations, examples of which are shown in this specification, are added to the above configuration as appropriate, i.e., other configurations in this specification that are not mentioned as the above configuration are added as appropriate.

Furthermore, according to the embodiment described above, the inference device includes an inference unit 308 and a notification unit 312. The inference unit 308 inputs time-series data stored in the data storage unit 16 into a trained model generated in any one of the above learning devices, and infers the degraded components and remaining life time. The notification unit 312 issues a notification based on the inferred degraded components and remaining life time. With this configuration, by using the trained model to obtain the degraded components and their remaining life time, it is possible to predict failure of the amplifier unit 107 with high accuracy and perform appropriate preventive maintenance.

<Modifications of the above-described embodiments>
In the embodiments described above, the dimensions, shapes, relative positional relationships, and implementation conditions of each component may be described, but these are merely examples in all respects and are not limiting.

Therefore, countless variations and equivalents not shown are contemplated within the scope of the technology disclosed in this specification. For example, this includes modifying, adding, or omitting at least one component, and even extracting at least one component from at least one embodiment and combining it with a component from another embodiment.

In addition, unless a contradiction arises, when it is stated in the above-described embodiments that "one" component is provided, it is understood that "one or more" of that component may be provided.

Furthermore, each component in the embodiments described above is a conceptual unit, and the scope of the technology disclosed in this specification includes cases where one component is made up of multiple structures, where one component corresponds to a part of a structure, and even where multiple components are provided in one structure.

Furthermore, each component in the embodiments described above includes structures having other structures or shapes as long as they perform the same function.

Furthermore, the descriptions in this specification are incorporated by reference for all purposes related to this technology, and none of them are admitted to be prior art.

1 communication system, 12 on-board radio device, 14 ground radio device, 16 data storage unit, 18 data analysis unit, 20, 20A control unit, 22 memory unit, 24 sorting unit, 26, 32 judgment unit, 101, 101A, 101B radio device, 102 data generation unit, 103 digital modulation unit, 104 distortion compensation and quadrature modulation unit, 104a distortion compensation unit, 104b, 133 quadrature modulation unit, 105 frequency conversion unit (IF → RF), 105a frequency mixer, 105b filter, 106 local oscillation unit, 107, 207 amplifier unit, 107a power transistor, 107b matching circuit, 108 high frequency transmission unit, 109 antenna, 110 waveform detection unit, 110a, 110b, 110c feedback amplifier unit, 111 Frequency conversion unit (RF to IF), 112 Quadrature demodulation unit, 113 Amplitude and phase measurement unit, 114 Power supply unit, 115 Voltage measurement unit, 116 Current sensor, 117 Current measurement unit, 118 Temperature sensor, 119 Power detection unit, 120 Transmission power measurement unit, 125 Ideal amplifier, 131 Predistortion distortion compensation unit, 132 Digital-analog conversion unit, 138, 201 Analog-digital conversion unit, 139 Input/output comparison unit, 140 Adaptive operation unit, 141 Delay unit, 202 Recording unit, 203 FFT unit, 204 Spectrum analysis unit, 300 Learning device, 302 Data acquisition unit, 304 Model generation unit, 306 Learned model storage unit, 308 Inference unit, 310 Manufacturing information management server, 312 Notification unit, 400 Inference device, 1001, 1002, 1003, 1004, 1005, 1006 graph.

Claims (18)

a wireless device having an amplifier for amplifying and outputting an input wireless signal;
a determination unit for determining that the amplifier unit is deteriorated when the radio signal output from the amplifier unit exceeds an amplitude threshold and a phase threshold that are determined for each amplitude of the corresponding radio signal input to the amplifier unit,
Deterioration assessment system. The deterioration determination system according to claim 1,
The wireless device further includes a sorting unit for acquiring the amplitude and phase of the corresponding wireless signal output from the amplifier in order of the amplitude of the wireless signal input to the amplifier, and for acquiring a relationship between the amplitude and the phase of the corresponding wireless signal input to the amplifier.
Deterioration assessment system. A wireless device in the deterioration determination system according to claim 1,
a sorting unit for acquiring the amplitude and phase of the corresponding radio signal output from the amplifier in order of the amplitude of the radio signal input to the amplifier, and further acquiring a relationship between the amplitude and the phase of the corresponding radio signal input to the amplifier,
Wireless device. A wireless device according to claim 3,
a distortion compensation unit for compensating for distortion of the amplifier unit by feeding back the radio signal output from the amplifier unit,
The radio signal input to the amplifier is the radio signal input via the distortion compensation unit.
Wireless device. A wireless device according to claim 3,
Further comprising a distortion compensation unit for compensating for distortion of the amplifier unit by a predistortion method,
the radio signal input to the amplifier unit is the radio signal input via the distortion compensation unit,
an input/output comparison unit for outputting to the distortion compensation unit a differential signal based on a difference between the radio signal input to the distortion compensation unit and the radio signal output from the amplification unit,
Wireless device. A wireless device according to claim 5,
Further comprising an adaptive operation unit for adapting a parameter of digital signal processing in the distortion compensation unit based on the difference signal.
Wireless device. A wireless device according to any one of claims 3 to 5,
an amplitude and phase measurement unit for measuring the amplitude and phase of the radio signal;
Further comprising a temperature sensor for measuring a temperature of the amplification unit;
The measured values of the amplitude and phase of the radio signal output from the amplifier are classified based on the temperature of the amplifier.
Wireless device. A wireless device according to any one of claims 3 to 7,
a transmission power measurement unit for measuring a transmission power, which is the power of the radio signal output by the amplifier unit;
A current measuring unit is further provided for measuring a supply current which is a current for supplying power to the amplifier unit.
Wireless device. A wireless device according to claim 8,
a voltage measuring unit for measuring a bias voltage for supplying power to the amplifier unit,
Wireless device. A wireless device according to any one of claims 3 to 9,
a data generating unit for generating a test signal for measuring a characteristic of the amplifier unit;
the data generating unit switches from user data to the test signal at a timing when communication is not required, and inputs the wireless signal to the amplifying unit.
Wireless device. A wireless device according to claim 10,
The test signal is a triangular wave whose amplitude changes linearly and periodically from a minimum value to a maximum value, or a signal generated by switching the frequency of the triangular wave at an even slower period.
Wireless device. A wireless device according to claim 10 or 11,
the data generating unit switches from user data to the test signal at a timing when communication is not required and is specified by a wireless communication standard, and inputs the wireless signal to the amplifying unit.
Wireless device. a data storage unit for storing, as time-series data, at least the amplitude/amplitude characteristic and the amplitude/phase characteristic obtained at regular intervals from the amplifier unit included in the wireless device according to any one of claims 3 to 12;
a manufacturing information management server for storing manufacturing lots, parts lots, manufacturing plants, and inspection data of the wireless devices as manufacturing data;
and a model generation unit for generating a trained model for inferring a deteriorated component and a remaining life time of the deteriorated component based on the time-series data stored in the data storage unit, the manufacturing data, data indicating a component in which a failure has occurred, and data indicating a time until the failure occurs.
Learning device. an inference unit for inferring the deteriorated components and the remaining life time by inputting the time-series data stored in the data storage unit into the trained model generated by the learning device according to claim 13;
a notification unit for notifying the user based on the inferred deteriorated component and the remaining life time,
Inference device. determining that a radio signal output from an amplifier unit for amplifying and outputting an input radio signal exceeds an amplitude threshold and a phase threshold that are determined for each amplitude of the corresponding radio signal input to the amplifier unit, the amplifier unit being deteriorated;
Judging device. The determination device according to claim 15 ,
and when it is determined that the amplifier unit is deteriorated, and further when a transmission power, which is the power of the wireless signal output by the amplifier unit, and a supply current, which is a current for supplying power to the amplifier unit, are changed , the deterioration of a power transistor is identified.
Judging device. The determination device according to claim 15 or 16 ,
and when it is determined that the amplifier unit has deteriorated, if the transmission power, which is the power of the radio signal output by the amplifier unit, and the supply current, which is the current for supplying power to the amplifier unit, do not change, and the radio signal output from the amplifier unit exceeds the amplitude threshold value in proportion to the amplitude of the radio signal input to the amplifier unit, a decrease in the output level of a local oscillator that oscillates a local signal used for frequency conversion of the radio signal input to the amplifier unit is identified.
Judging device. The determination device according to any one of claims 15 to 17 ,
When it is determined that the amplifier unit is deteriorated, if the transmission power, which is the power of the wireless signal output by the amplifier unit, and the supply current, which is the current for supplying power to the amplifier unit, do not change and the wireless signal output from the amplifier unit exceeds the phase threshold value regardless of the amplitude of the wireless signal input to the amplifier unit, an increase in ripple in a power supply unit that supplies a bias voltage for supplying power to the amplifier unit or noise interference from the ground is identified.
Judging device.

Images