JP6058441B2 - Signal detection apparatus, signal detection method, and signal detection program - Google Patents

Description

  Embodiments described herein relate generally to a signal detection apparatus that receives an incoming wave and detects a signal component included in the received incoming wave, and a signal detection method and a signal detection program used in the apparatus.

  A received signal in wireless communication is degraded by interference waves, multipath fading, external noise, and the like. These deterioration factors exhibit various changes in time, frequency, and space. In this way, radio waves travel through complicated and diverse paths. A path through which the radio wave is transmitted is called a radio wave propagation path.

  In general, the signal quality of the signal received from the transmission signal that has passed through the radio wave propagation path is degraded. When spectrum sensing processing is performed on this received signal, it is possible to make a mistake in determining the presence or absence of the signal, for example, to improve the miss probability and increase the error detection probability. Also, multiple detection increases if the detection threshold is set low to reduce the missed probability. In particular, for a received signal having a low CNR (Carrier to Noise Ratio), the influence on the determination error is extremely large, so that the performance of the signal detection apparatus is significantly degraded.

  As described above, in order to solve the problem at the time of signal detection, diversity technology has been used in wireless communication. With this diversity technique, it is possible to reduce fluctuations in the received signal due to fading.

  By the way, diversity technology in wireless communication is premised on signal quality improvement upon demodulation and decoding. Therefore, the point of interest in the diversity-processed signal is a portion where the signal exists. That is, there is no interest or little interest in the noise portion where no signal exists.

Abdulsattar A.M. et al., “A New Multiple Antennas Method based Energy Detector for Cognitive Radio over Fading Channels,” IJCA, vol.52, no.5, Aug. 2012.

  In conventional signal detection, attention is paid to diversity processing in a portion where a signal is present, and thus the power level of a noise section may be high. Since signal detection is performed based on the difference in power level between the signal and noise, there is a problem that the signal detection performance deteriorates if the power level in the noise section is large.

  Accordingly, an object is to provide a signal detection device, a signal detection method, and a signal detection program capable of improving signal detection performance by paying attention to a noise portion where no signal exists.

  According to the embodiment, the signal detection apparatus includes a band division unit, a calculation unit, a selection unit, and a detection unit. The band dividing unit divides a plurality of reception signals received by a plurality of antennas for each first frequency width. The calculation unit accumulates, for each of the plurality of reception signals, the power at which the divided signal divided for each first frequency width is received during the first reception period while sliding the first reception period. The selection unit selects a reception signal having the maximum accumulated power from the plurality of reception signals. The detection unit generates power received in a second reception period set based on a signal component that is assumed to be included in the received signal, as a result of the second reception for the selected received signal. The reference power is obtained by accumulating while sliding the period, and the divided signal for the selected reception signal is received in the third reception period set based on the signal component assumed to be included in the reception signal. The target power is acquired by accumulating the power to be performed while sliding the third reception period, and the signal component included in the received signal is detected based on the reference power and the target power.

It is a block diagram which shows the function structure of a radar apparatus provided with the signal detection apparatus which concerns on embodiment. It is a figure which shows the example of distribution of the reception power of the signal received with the antenna part of FIG. It is a figure which shows the example of the sliding process by the calculation part shown in FIG. It is a figure which shows the example of the relationship between the 1st receiving period in the calculation part shown in FIG. 1, and the 2nd and 3rd receiving periods in a detection part. It is a figure which shows the example of the relationship between the 2nd and 3rd receiving period in the detection part shown in FIG. 1, and the signal component contained in an incoming wave. It is a flowchart which shows the operation | movement in the signal detection apparatus shown in FIG. It is a figure which shows the simulation result about the signal detection apparatus shown in FIG. It is a figure which shows the example of the power level of the received signal received by the signal detection apparatus shown in FIG. 1, and the power level of the received signal received by the conventional apparatus.

  Hereinafter, embodiments will be described with reference to the drawings.

  FIG. 1 is a block diagram illustrating a functional configuration of a radar apparatus including a signal detection apparatus according to the present embodiment. The radar apparatus shown in FIG. 1 includes antenna units 10-1 to 10-N, reception processing units 20-1 to 20-N, and a signal detection device 30.

  The antenna units 10-1 to 10-N receive radio waves that are incoming waves and output received signals to the reception processing units 20-1 to 20-N. The received signal includes a signal component and a noise component transmitted from a master station (not shown).

  The reception processing units 20-1 to 20-N frequency-convert the reception signals received by the antenna units 10-1 to 10-N into signals in an IF (Intermediate Frequency) band or a baseband band. In addition, the reception processing units 20-1 to 20-N perform analog-digital conversion of the frequency-converted signal into a digital signal. The reception processing units 20-1 to 20 -N output digital signals to the signal detection device 30.

  The signal detection device 30 includes, for example, a CPU (Central Processing Unit), and a storage area for programs and data for the CPU to execute processing such as ROM (Read Only Memory) and RAM (Random Access Memory). The signal detection device 30 realizes the functions of the band division units 31-1 to 31 -N, the calculation units 32-1 to 32 -N, the selection unit 33, and the detection unit 34 by causing the CPU to execute an application program. .

  The band dividing unit 31-1 divides the digital signal output from the reception processing unit 20-1 into a band having a predetermined width. The band dividing unit 31-1 outputs the divided signal after the band division to the calculating unit 32-1.

The reason for dividing the band will be described first. Here, in order to improve the prospect of explanation, it is assumed that a plurality of signals having the same average power level and independent from each other are received through the Rayleigh fading environment at different frequencies. In this assumption, the environment in which a plurality of signals are received corresponds to a case where the frequency width is wide, and the environment in which a small number of signals are received corresponds to a case where the frequency width is narrow. The received power is assumed to follow a normalized complex Gaussian distribution. The received power in such a state can be approximated by a gamma distribution, and the x-square distribution at this time is given by the following equation.

Here, x is the received power, and n is the number of signals. A graph of this x-square distribution is shown in FIG. As shown in FIG. 2, when the number of signals is 1, it follows that the amplitude fluctuation range is wide in order to follow the exponential distribution. On the other hand, it can be seen that the amplitude fluctuation range becomes narrower as the number of signals exceeds 2 and increases. Thus, it can be seen that when the number of signals to be received is large, that is, when the bandwidth is wide, the amplitude fluctuation width due to Rayleigh fading becomes small. This indicates that when the bandwidth is wide, the effect of fading suppression tends to be low. Therefore, in order to improve the signal detection performance, it is necessary to divide the bandwidth up to a bandwidth that can be expected to increase the effect of fading suppression.

  Next, an example of a bandwidth determination method will be described. Generally, the fading becomes deeper as the frequency width becomes smaller. On the other hand, when applying the algorithm, generally, the operation cost increases as the frequency width is reduced. For this reason, setting a time width uniformly for all signals is not good from the balance between effect and cost. Therefore, the frequency width is determined as follows.

First, let Δt and Δf be the time / frequency resolution of the spectrogram to be finally used. The spectrogram at this time is expressed as <P> (m · Δf, n · Δt) (where <P> is a bold letter P). Here, signal processing based on the correlation matrix is performed as an example of array antenna signal processing. The correlation matrix <R> using the spectrogram (where <R> is the bold letter R) is

It is expressed. When <P> (m · Δf, n · Δt) is converted using the maximum eigenvector <E> of the correlation matrix <R> (where <E> is the bold letter E),

It is expressed. The spectrogram <P ′> (m · Δf, n · Δt) calculated here (where <P ′> is a bold P ′) is a spectrogram in which the SNR is maximized. The band dividing unit 31-1 extracts a portion exceeding a specified power level from the spectrogram <P ′> (m · Δf, n · Δt). The extracted part is a desired signal included in the radio wave. The band dividing unit 31-1 sets the bandwidth in which the signal having the lowest frequency band among the extracted signals can be detected as the bandwidth for band division. Note that the band dividing unit 31-1 may determine the bandwidth in the band dividing unit 31-1 in consideration of the minimum reception sensitivity and cost performance.

  Since the operations of the band dividing units 31-2 to 31-N are the same as those of the band dividing unit 31-1, the description thereof is omitted here.

The calculation unit 32-1 receives the divided signal output from the band dividing unit 31-1. The calculation unit 32-1 accumulates the power of the divided signals received during the predetermined first reception period while sliding the first reception period. FIG. 3 shows a diagram in which bins, which are one band of the divided signals output from the band dividing unit 31-1, are arranged in time series. The sliding power P _{n, i} is

Is required. Here, n indicates the number of antennas, i indicates a sample number, and j indicates a signal number for accumulating power. As illustrated in FIG. 3, the calculation unit 32-1 accumulates the signals to be accumulated while sliding the signals one by one in time series. Here, the predetermined first reception period T ₁ is, for example, a period in which the BT ₁ product, which is the product of the bandwidth B and the first reception period T ₁ , is about 4. First reception period _{T 1,} when the frequency band is 100Hz is 4 / 100s. For example, j indicates the number of signals received during this 4/100 s. The calculation unit 32-1 accumulates the power of each received divided signal and outputs the accumulated power value for each divided band to the selection unit 33. Since the operations of the calculation units 32-2 to 32-N are the same as those of the calculation unit 32-1, the description thereof is omitted here.

The selector 33 receives the power value accumulated for each divided band output from the calculators 32-1 to 32-N. The selection unit 33 compares the power values output from the calculation units 32-1 to 32-N for each divided band. The selection unit 33 selects the antenna having the maximum power for each divided band as follows.

The selection unit 33 outputs information about the antenna number that is the maximum power obtained by Expression (5) to the detection unit 34.

The detection unit 34 receives the division signal divided by the band division units 31-1 to 31 -N and the information output from the selection unit 33. The detection unit 34 receives the divided signal of the system that matches the antenna number notified from the selection unit 33 in the second reception period. The detection unit 34 accumulates the power of the divided signals received in the second reception period while sliding the second reception period. The power value at this time is called a reference power value P _Ref ,

It is expressed. In addition, the detection unit 34 receives the divided signal of the system that matches the antenna number notified from the selection unit 33 in the third reception period. The detection unit 34 accumulates the power of the divided signals received in the third reception period while sliding the third reception period. The power at this time is called a target power value P _Obj ,

It is expressed. Here, the second and third reception periods T ₂ and T ₃ have the same time width. For example, the first reception period T ₁ > the second and third reception periods T ₂ , T ₃ , The BT product has a time width satisfying less than 4 and satisfying T _Obj > second and third reception periods T ₂ and T ₃ . T _Obj refers to the cycle of the signal having the shortest cycle among signals that are assumed to be received by the radar apparatus. A second reception period _{T 2,} and the third reception period _{T 3,} is in the time series relationship displaced by _{T Obj} / 2, x in the formula _(7), the signal contained in the _{T Obj} / 2 Indicates a number. Relationship between power P _{n, i} received in the first reception period T ₁ , power P _Ref received in the _second reception period T ₂ , and power P _Obj received in the _third reception period T ₃ For example, as shown in FIG. Further, as shown in FIG. 5, the signal component is not included in the incoming wave received in the second reception period, while the signal component is included in the incoming wave received in the third reception period. Become.

The detection unit 34 calculates the calculated reference power value and the target power value,

Compare like this. The detection unit 34 determines that a signal component has appeared when the comparison result exceeds a preset threshold value, and determines that there is no signal component when the comparison result is less than the threshold value. Note that the detection unit 34 can also determine the disappearance of the signal by reversing the relationship between the denominator and the numerator in Expression (8).

  Next, the operation of the signal detection device 30 configured as described above will be described with reference to the flowchart shown in FIG.

  First, the signal detection device 30 divides the digital signal output from the reception processing unit 20-1 into a band having a predetermined width by the band dividing units 31-1 to 31-N (step S61). The signal detection apparatus 30 accumulates the power of the divided signals for each band while sliding the reception period in the calculation units 32-1 to 32-N (step S62).

  In the signal detection device 30, the selection unit 33 selects, for each band, the antenna number from which the maximum power value is obtained among the power values calculated by the calculation units 32-1 to 32-N (step S63). .

In the signal detection device 30, the detection unit 34 calculates the reference power value P _Ref and the target power value P _Obj from the divided signal of the selected antenna number (step S64). The signal detection device 30 determines whether or not a signal component is generated from the calculated reference power value P _Ref and the target power value P _Obj (step S65).

  FIG. 7 is a diagram illustrating a simulation result indicating the effectiveness of the spatio-temporal signal detection device 30 according to the present embodiment. This simulation evaluates the detection performance in an uncorrelated Rayleigh fading environment. In FIG. 7, the solid line indicates the detection result of the signal component by the signal detection apparatus 30 according to the present embodiment, and the broken line indicates the detection result by the conventional detection method. Here, the conventional detection method refers to a method of accumulating power without dividing a received signal into bands and detecting a signal component based on a received signal from an antenna having the largest accumulated power value. Moreover, the dashed-dotted line in FIG. 7 shows the detection system using the omni system by a non-directional single antenna. From the results shown in FIG. 7, it can be seen that the detection method according to the present embodiment is greatly improved over the conventional detection method. Note that the conventional detection method has a lower detection probability in an environment where the SNR is lower than the omni method. However, it is possible to confirm the effect of diversity synthesis by comparing the overall detection performance (diversity thioder, etc.).

  As described above, in the above-described embodiment, the band dividing units 31-1 to 31-N divide the received signal up to a bandwidth that can be expected to increase the effect of fading suppression. Calculation units 32-1 to 32-N accumulate power values while sliding the first reception period. The selection unit 33 selects a system having the maximum power value from a plurality of reception systems, and notifies the detection unit 34 of the selected system and the like. The detection unit 34 acquires the reference power value by accumulating the power of the divided signal of the selected system in the second reception period, and acquires the target power value by accumulating in the third reception period. To do. The detection unit 34 detects the signal component included in the incoming wave by comparing the accumulated power values.

  By the way, the technique described in Non-Patent Document 1 introduces the concept of two-stage power detection. Here, in the first power detection, fading suppression by SLC (Square Law Combining) is performed. In fading suppression by SLC, noise during a period in which no signal component is included is not suppressed. In the second power detection, noise suppression is performed by taking an arithmetic average with respect to the output of the SLC. In the first power detection, noise in a period in which no signal component is included is not suppressed, and thus the effect of noise suppression in the second power detection is small. On the other hand, in the signal detection device 30 according to the present embodiment, the band division units 31-1 to 31-N perform band division, and the calculation units 32-1 to 32-N perform sliding calculation, thereby performing space-time processing. Therefore, it is possible to suppress noise during a period in which no signal component is included. FIG. 8 presents a diagram illustrating an example of the power level of the received signal. In FIG. 8, the solid line indicates the power level when a method corresponding to this embodiment, that is, the larger received signal received by two antennas in a predetermined time window section such as the signal detection window width is selected. Indicates. The broken line indicates the power level when the larger one of the received signals received by the two antennas in the conventional method is selected in units of samples. In FIG. 8, signal components are included after time 50. According to FIG. 8, the power level of the noise component when no signal component is included is smaller in the method corresponding to this embodiment than in the conventional method. Thereby, as compared with the conventional method, the detection threshold can be set deeper in the method corresponding to the present embodiment. That is, it is possible to provide a stable signal detection by reducing erroneous detection.

  As described above, according to the signal detection device 30 according to the present embodiment, a noise portion in which no signal exists in the incoming wave is suppressed. Further, signal detection performance is improved by detecting a signal by comparing a reference power value that is a power value in a region including only a noise component and a target power value that is a power value in a region including a signal component. It will be.

  Therefore, according to the signal detection apparatus according to the present embodiment, it is possible to improve the signal detection performance by paying attention to a noise portion where no signal exists.

  In the present embodiment, the calculation units 32-1 to 32 -N and the detection unit 34 perform processing on the divided signals divided in the predetermined frequency band by the band dividing units 31-1 to 31 -N. Was described as an example. However, the present invention is not limited to this. For example, the band of the divided signal processed by the detection unit 34 may be smaller than the band of the divided signal processed by the calculation units 32-1 to 32-N.

  For example, the band division units 31-1 to 31 -N divide the band by a band Δf / 2 that is half of the assumed frequency band Δf. The calculation units 32-1 to 32-N perform a sliding process on the frequency band Δf. At this time, for example, two signals are included in the sliding width by the calculation units 32-1 to 32-N. The selector 33 selects an antenna number based on the accumulated power calculated by the calculators 32-1 to 32-N. The detection unit 34 acquires the reference power and the target power by performing the sliding process in the reception period for the frequency band Δf / 2. The detection unit 34 detects the generation of a signal based on the acquired reference power and target power.

  Thereby, it is possible to improve the signal detection capability while suppressing the calculation cost.

  Although the embodiment of the present invention has been described, this embodiment is presented as an example and is not intended to limit the scope of the invention. This embodiment can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. This embodiment and its modifications are included in the scope of the present invention and the gist thereof, and are also included in the invention described in the claims and the equivalent scope thereof.

  10-1 to 10-N ... antenna unit, 20-1 to 20-N ... reception processing unit, 30 ... signal detection device, 31-1 to 31-N ... band division unit, 32-1 to 32-N ... calculation Part 33 ... selection part 34 ... detection part

Claims (6)

A band dividing unit that divides a plurality of reception signals received by a plurality of antennas for each first frequency width;
A calculation unit that accumulates power received by each of the plurality of received signals for each of the plurality of reception signals while sliding the first reception period, in which the divided signal divided for each first frequency width is received during the first reception period;
A selection unit that selects a reception signal having a maximum power value accumulated in the calculation unit from the plurality of reception signals;
The power received during the second reception period set based on the signal component that is assumed to be included in the reception signal is the divided signal for the selected reception signal is slid through the second reception period. While accumulating a reference power to obtain the power received during the third reception period in which the divided signal for the selected received signal is set based on the signal component assumed to be included in the received signal. A detection unit that acquires the target power by accumulating the third reception period while sliding, and detects a signal component included in the reception signal based on the reference power and the target power,
The signal detection device in which the second and third reception periods are shorter than the first reception period. The calculation unit slides the first reception period at an interval corresponding to a second frequency width wider than the first frequency width,
The signal detection device according to claim 1, wherein the detection unit slides the second and third reception periods at intervals corresponding to the first frequency width. Dividing a plurality of reception signals received by a plurality of antennas for each first frequency width;
The power received by the divided signal divided for each first frequency width during the first reception period is accumulated for each of the plurality of reception signals while sliding the first reception period,
Wherein the plurality of received signals, power value accumulated by the calculation unit selects the maximum of the received signal,
The power received during the second reception period set based on the signal component that is assumed to be included in the reception signal is the divided signal for the selected reception signal is slid through the second reception period. While accumulating the reference power,
The power received in the third reception period set based on the signal component that is assumed to be included in the reception signal is the divided signal for the selected reception signal is slid in the third reception period. While accumulating the target power,
A signal detection method for detecting a signal component included in the received signal based on the reference power and the target power. Sliding the first reception period at an interval corresponding to a second frequency width wider than the first frequency width;
The signal detection method according to claim 3, wherein the second and third reception periods are slid at intervals corresponding to the first frequency width. Band division processing for dividing a plurality of received signals received by a plurality of antennas for each first frequency width;
A calculation process for accumulating power received by the divided signal divided for each first frequency width during the first reception period for each of the plurality of reception signals while sliding the first reception period;
A selection process for selecting a received signal having a maximum power value accumulated in the calculation unit from the plurality of received signals;
The power received during the second reception period set based on the signal component that is assumed to be included in the reception signal is the divided signal for the selected reception signal is slid through the second reception period. A first acquisition process for accumulating and acquiring a reference power,
The power received in the third reception period set based on the signal component that is assumed to be included in the reception signal is the divided signal for the selected reception signal is slid in the third reception period. A second acquisition process for accumulating and acquiring the target power,
A signal detection program for causing a computer of a signal detection device to execute a detection process for detecting a signal component included in the received signal based on the reference power and the target power. In the calculation process, the first reception period is slid at an interval corresponding to a second frequency width wider than the first frequency width,
6. The signal detection program according to claim 5, wherein, in the first and second acquisition processes, the second and third reception periods are slid at intervals corresponding to the first frequency width.

Images