Signal Monitoring Device
By integrating multiple beams through an averaging process, the device reduces computational overhead in signal monitoring, addressing inefficiencies in existing systems and optimizing resource usage.
Patent Information
- Application Number
- JP2020030469
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-02-26
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2040-02-26
AI Technical Summary
Existing signal monitoring devices require excessive computational resources due to the large number of beams formed, leading to inefficient FFT and frequency direction normalization calculations, especially when monitoring narrowband signals.
The device incorporates an averaging processing section that integrates multiple beams to create fewer, weighted integrated beams, reducing the number of calculations required for FFT and frequency direction normalization.
This approach significantly reduces computational load by averaging multiple beams into integrated ones, thereby minimizing the overall calculation burden while maintaining effective signal monitoring capabilities.
Smart Images

Figure 0007585609000008 
Figure 0007585609000009 
Figure 0007585609000010
Abstract
Description
[Technical field]
[0001] The present invention relates to a signal monitoring device that monitors a target signal based on outputs from a plurality of sensors. [Background technology]
[0002] Conventionally, as an application using an array 2 composed of a plurality of sensors 2a such as microphones or hydrophones, a signal monitoring device 300 is known that monitors a monitoring target signal arriving from the periphery of the array 2. Here, the monitoring target signal may be, for example, a narrowband signal. In such an application, a beam is created by beamforming based on the output from the plurality of sensors 2a constituting the array 2, and FFT (Fast Fourier Transform) and normalization in the frequency direction are applied to the output of each created beam. This allows the signal monitoring device 300 to detect the monitoring target signal, regardless of the direction from which the monitoring target signal arrives around the array 2. Here, beamforming is known as a method for improving the SNR (Signal Noise Ratio) of the monitoring target signal arriving from the periphery of the array 2.
[0003] Fig. 4 is a block diagram showing a conventional signal monitoring device 300 using an array 2. As shown in Fig. 4, the signal monitoring device 300 includes the array 2, an electronic circuit 3, a beamforming processing unit 4, and a monitoring unit 5. The array 2 is composed of N sensors 2a (2a-1, 2a-2, ..., 2a-N), and narrowband signals arrive at the array 2. The electronic circuit 3 is composed of, for example, an amplifier, a filter, an A / D converter, and the like. The electronic circuit 3 digitizes the output (time waveform) from each sensor 2a constituting the array 2.
[0004] The beamforming processing unit 4 creates multiple beams (beam 1, beam 2, ... beam M) from the digitized data. The monitoring unit 5 monitors the narrowband signal and has an FFT processing unit 6, a frequency direction normalization processing unit 7, and a detection processing unit 8. The FFT processing units 6 (6-1, 6-2, ..., 6-M) perform a fast Fourier transform on the multiple beams. The frequency direction normalization processing units 7 (7-1, 7-2, ..., 7-M) normalize the multiple beams that have been fast Fourier transformed in the frequency direction. The detection processing unit 8 monitors the narrowband signal based on the normalized multiple beams. [Prior art documents] [Non-patent literature]
[0005] [Non-Patent Document 1] 'The effect of improper normalization on the performance of an automated energy detector' J. Acoust. Soc. Am. 78(3), Sep. 1985
[0006] Next, a description will be given of the operation of the conventional signal monitoring device 300. The narrowband signal received at the center of the array 2 is expressed by the following equation (1).
[0007]
number
[0008] In equation (1), t is time, A is the amplitude of the narrowband signal, f is the frequency of the narrowband signal, and φ0 is the initial phase. In this case, for N sensors 2a constituting the array 2, the time waveform of the output from the n-th sensor 2a is expressed by the following equation (2).
[0009]
number
[0010] Here, τ n is the delay time when the n-th sensor 2a receives the narrowband signal. n is expressed by the following equation (3).
[0011]
number
[0012] In equation (3), d is a vector indicating the direction of arrival of a narrowband signal, and r n is the position vector of the n-th sensor 2a, and c is the propagation velocity of the narrowband signal.
[0013] The output (time waveform) from each sensor 2a constituting the array 2 passes through an amplifier and a filter in the electronic circuit 3, is digitized by an A / D converter, and is input to the beamforming processor 4. The beamforming processor 4 applies the process shown in the following equation (4) to the digitized data.
[0014]
number
[0015] where x m [t] is the output of beamforming for the mth direction (discretized time waveform), w n is the weight given to the nth sensor 2a, τ n is the delay time given to the output of the n-th sensor 2a. Here, it is assumed that the beamforming processing unit 4 obtains M beamforming outputs (discretized time waveforms).
[0016] Next, an FFT (Fast Fourier Transform) is applied to the output from each beamforming by the FFT processing unit 6 as shown in the following equation (5).
[0017]
number
[0018] In equation (5), I is the number of samples when the FFT is applied.
[0019] Next, the output from the FFT processor 6 is subjected to normalization in the frequency direction by a frequency direction normalization processor 7. In the normalization in the frequency direction, the processing shown in Non-Patent Document 1 is applied to the FFT results of all beamforming outputs. In this way, the conventional signal monitoring device 300 monitors narrowband signals arriving from all directions around the array 2 by performing FFT and frequency direction normalization on all formed beams. Summary of the Invention [Problem to be solved by the invention]
[0020] However, the signals received by adjacent beams do not change that much, and the difference is very small. For this reason, if FFT and normalization in the frequency direction are performed on all the formed beams as in the conventional method, waste occurs. Therefore, the amount of calculation required for FFT and normalization in the frequency direction increases. In addition, the number of beams formed is generally equal to or greater than the number of sensors 2a, and depending on the application, several hundred or more beams are formed. In this case, the amount of calculation required for FFT and normalization in the frequency direction increases.
[0021] The present invention has been made in light of the above-mentioned problems, and provides a signal monitoring device that reduces the amount of calculation. [Means for solving the problem]
[0022] The signal monitoring device according to the present invention includes a beamforming processing unit that generates beams based on outputs from a plurality of sensors, and a plurality of beams generated by the beamforming processing unit. Time waveform of Average the integrated beam Time waveform ofand a monitoring unit that monitors a monitored signal based on information in the frequency direction of the integrated beam created by the averaging processing unit. Effect of the Invention
[0023] According to the present invention, the averaging processing unit creates an integrated beam by averaging a plurality of beams, and the monitoring unit monitors the monitoring target signal based on the integrated beam. In this way, since the integrated beam is an average of a plurality of beams, the total number of integrated beams is smaller than the total number of beams. Therefore, the amount of calculation can be reduced. [Brief description of the drawings]
[0024] [Figure 1] 1 is a block diagram showing a signal monitoring device according to a first embodiment of the present invention; [Diagram 2] FIG. 11 is a block diagram showing a signal monitoring device according to a second embodiment of the present invention. [Diagram 3] FIG. 11 is a block diagram showing a signal monitoring device according to a third embodiment of the present invention. [Figure 4] FIG. 1 is a block diagram showing a conventional signal monitoring device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0025] Hereinafter, an embodiment of a signal monitoring device according to the present invention will be described with reference to the drawings. Note that the present invention is not limited to the embodiment described below. Also, in the following drawings, including FIG. 1, the size relationships of the components may differ from the actual ones. In the following description, terms indicating directions are used as appropriate to facilitate understanding, but these are for the purpose of explanation and do not limit the present invention. Examples of terms indicating directions include "up", "down", "right", "left", "front" and "rear".
[0026] Embodiment 1 Fig. 1 is a block diagram showing a signal monitoring device 1 according to the present embodiment 1. The signal monitoring device 1 according to the present embodiment 1 differs from the conventional signal monitoring device 300 shown in Fig. 4 in that it includes an averaging processing unit 10. Therefore, a description of the same configuration as the conventional signal monitoring device 300 shown in Fig. 4 will be omitted, and only the differences will be described.
[0027] As shown in Fig. 1, the signal monitoring device 1 includes an array 2, an electronic circuit 3, a beam forming processing unit 4, an averaging processing unit 10, and a monitoring unit 5. The array 2 is composed of N sensors 2a (2a-1, 2a-2, ..., 2a-N), and a narrowband signal that is a monitoring target signal arrives at the array 2. The electronic circuit 3 is composed of, for example, an amplifier, a filter, an A / D converter, etc. The electronic circuit 3 digitizes the output (time waveform) from each sensor 2a that constitutes the array 2. The beam forming processing unit 4 creates multiple beams (beam 1, beam 2, ... beam M) from the digitized data.
[0028] The averaging processor 10 (10-1, 10-2, . . . , 10-M') averages multiple beams created by the beamforming processor 4 to create integrated beams (integrated beam 1, integrated beam 2, . . . , integrated beam M'). Specifically, the averaging processor 10 averages a predetermined number of beams to output one integrated beam (time waveform). The averaging processor 10 averages the m'th integrated beam y m’ [t] is calculated using the following equation (6).
[0029]
number
[0030] In equation (6), x m’,k[t] is the beam (output of beamforming) used when calculating the m-th integrated beam, and K is the number of beams used in the averaging unit 10. Let the total number of integrated beams created by this process be M'. Note that when the averaging unit 10 creates integrated beams, it is also possible to use the original beams repeatedly when calculating a plurality of integrated beams. That is, the averaging unit 10 may be configured to repeatedly average a plurality of beams created by the beamforming unit 4.
[0031] The monitoring unit 5 monitors narrowband signals and includes an FFT processing unit 6, a frequency-direction normalization processing unit 7, and a detection processing unit 8. The FFT processing unit 6 (6-1, 6-2, ···, 6-M') performs fast Fourier transform on a plurality of integrated beams. The frequency-direction normalization processing unit 7 (7-1, 7-2, ···, 7-M') performs normalization in the frequency direction on the plurality of integrated beams that have been fast Fourier transformed. The detection processing unit 8 monitors narrowband signals based on the plurality of normalized integrated beams.
[0032] According to the first embodiment, the averaging unit 10 creates an integrated beam by averaging a plurality of beams, and the monitoring unit 5 monitors the signal to be monitored based on the integrated beam. In this way, since the integrated beam is an average of a plurality of beams, the total number M' of integrated beams is less than the total number M of beams. Therefore, the amount of calculation can be reduced.
[0033] Comparing equation (4) described in the section explaining the conventional signal monitoring device 300 with equation (6) of the first embodiment, the total number of beams in the conventional signal monitoring device 300 is M, and the total number of integrated beams in the signal monitoring device 1 according to the first embodiment is M'. That is, according to the first embodiment, the amount of calculation required for the FFT and the normalization in the frequency direction that make up the monitoring unit 5 can be suppressed to M' / M (M'<M, that is, M' / M is less than 1) times.
[0034] However, if an attempt is made to reduce the amount of calculations by simply reducing the number of beams formed by beamforming to M', there is a risk that non-monitored areas will occur in the monitoring ranges of adjacent beams if the number of beams is reduced too much. In this case, there is a risk that the array 2 will not be suitable for monitoring narrowband signals. Therefore, in the first embodiment, instead of simply reducing the number of beams, the averaging processor 10 creates an integrated beam by averaging multiple beams.
[0035] Embodiment 2 2 is a block diagram showing a signal monitoring device 100 according to the second embodiment. The second embodiment differs from the first embodiment in the function of an averaging processor 110. In the second embodiment, the same reference numerals are used to designate parts common to the first embodiment, and descriptions thereof will be omitted, and differences from the first embodiment will be mainly described.
[0036] As shown in FIG. 2, the averaging processors 110 (110-1, 110-2, . . . , 110-M′) weight and average the multiple beams created by the beamforming processor 4 based on the importance of the multiple beams. The averaging processor 110 averages the m′-th integrated beam y m’ [t] is calculated using the following equation (7).
[0037]
number
[0038] In equation (7), x m’,k [t] is the beam used in calculating the mth integrated beam (beamforming output), K is the number of beams used in the averaging processor 110, and q k is a weighting factor assigned to the beam input to the averaging processor 110. The total number of integrated beams created by this process is M'. When the averaging processor 110 creates an integrated beam, it is possible to use the original beams redundantly in calculating multiple integrated beams.
[0039] According to the second embodiment, when averaging and integrating a plurality of beams, the averaging processing unit 110 assigns weights. That is, the importance of each input beam can be selected. Therefore, the central position for calculating the integrated beam can be controlled, and the degradation of the narrowband signal that may occur due to integration can be suppressed.
[0040] Also, comparing Equation (4) and Equation (7), the total number of beams in the conventional signal monitoring device 300 is M, and the total number of integrated beams in the signal monitoring device 100 according to the second embodiment is M'. Thus, according to the second embodiment, similar to the first embodiment, the computational amount required for the FFT and the normalization in the frequency direction that constitute the monitoring unit 5 can be suppressed to M' / M (M' < M, that is, less than 1) times.
[0041] Embodiment 3. FIG. 3 is a block diagram showing a signal monitoring device 200 according to the third embodiment. The third embodiment is different from the second embodiment in that the signal to be monitored is a non-stationary signal and the monitoring unit 205 has a non-stationary signal analysis processing unit 209. In the third embodiment, parts common to the first and second embodiments are denoted by the same reference numerals and the description thereof is omitted, and the description will be centered on the differences from the first and second embodiments.
[0042] As shown in FIG. 3, the monitoring unit 205 includes a non-stationary signal analysis processing unit 209 (209-1, 209-2, ··· 209-M') and a detection processing unit 8. The non-stationary signal analysis processing unit 209 analyzes a signal whose frequency components change with time. The non-stationary signal analysis processing unit 209 is, for example, an analysis method using a short-time Fourier transform, a wavelet transform, or a Wigner distribution. The detection processing unit 8 monitors the analyzed non-stationary signal.
[0043] Even when the signal to be monitored is a non-stationary signal as in the third embodiment, the computational amount can be reduced in the same manner as in the first and second embodiments.
[0044] The beamforming processing unit 4 generally uses a delay-and-sum beamformer, but is not limited to this. For example, the beamforming processing unit 4 may use an adaptive beamformer such as a generalized sidelobe canceller or an eigenspace beamformer. [Explanation of symbols]
[0045] 1 signal monitoring device, 2 array, 2a sensor, 3 electronic circuit, 4 beamforming processing unit, 5 monitoring unit, 6 FFT processing unit, 7 frequency direction normalization processing unit, 8 detection processing unit, 10 averaging processing unit, 100 signal monitoring device, 110 averaging processing unit, 200 signal monitoring device, 205 monitoring unit, 209 non-stationary signal analysis processing unit, 300 signal monitoring device.
Claims
1. a beamforming processing unit that creates beams based on outputs from the plurality of sensors; an averaging processor that averages the time waveforms of the plurality of beams generated by the beam forming processor to generate a time waveform of an integrated beam; a monitoring unit that monitors a monitoring target signal based on information in a frequency direction of the integrated beam created by the averaging processing unit; A signal monitoring device comprising:
2. a beamforming processing unit that creates beams based on outputs from the plurality of sensors; an averaging processor that averages the multiple beams generated by the beamforming processor in an overlapping manner to generate an integrated beam; a monitoring unit that monitors a monitoring target signal based on the integrated beam created by the averaging processing unit; A signal monitoring device comprising:
3. a beamforming processing unit that creates beams based on outputs from the plurality of sensors; an averaging processor that creates an integrated beam by weighting and averaging the multiple beams based on the importance of the multiple beams created by the beam forming processor; a monitoring unit that monitors a monitoring target signal based on the integrated beam created by the averaging processing unit; A signal monitoring device comprising:
4. The monitored signal is It is a narrowband signal 4. A signal monitoring device according to claim 1, wherein the signal monitoring device is a signal monitoring device for monitoring a signal.
5. The monitored signal is It is a non-stationary signal 4. A signal monitoring device according to claim 1, wherein the signal monitoring device is a signal monitoring device for monitoring a signal.
Citation Information
Patent Citations
Wireless device, radio network node, and methods performed therein for handling communication in a wireless communication network
WO2018203815A1
