KR101132927B1 - Entropy based sensing system using phase information - Google Patents

Abstract

A signal detection system for first detecting a user signal based on entropy using phase information is disclosed. A signal detection system using phase entropy includes a phase calculator configured to calculate a phase value of a received signal received at a receiver; An entropy calculator configured to calculate a phase entropy using a phase value of the received signal; And a determination unit that first determines whether a user exists by comparing phase entropy and a threshold value.

Description

Entropy-based signal detection system using phase information {ENTROPY BASED SENSING SYSTEM USING PHASE INFORMATION}

Embodiments of the present invention relate to a signal detection system for first detecting a user signal based on entropy using phase information.

With the rapid development of wireless communication systems and the emergence of various services, the demand for wireless resources is increasing day by day. Thus, to provide bandwidth, a new spectrum allocation policy is required. The Federal Communications Commission (FCC) has approved an unlicensed approach to frequency bands that have been temporarily unused as a condition of minimizing interference in order to use radio resources efficiently. The concept of Cognitive Radio (CR) has been proposed as a method for efficiently using radio resources. Wireless recognition technology refers to a technology in which a frequency band is allocated but can be used efficiently by detecting an empty frequency band without actually using it.

In order for a radio-aware user to communicate using frequency, the user must first find an empty spectrum that is not occupied by the user. This process is called spectrum sensing and is one of the core technologies for constructing a radio-cognitive system. . However, there are many technical difficulties in spectrum sensing. In particular, the method of improving spectrum sensing performance at low signal to noise ratio (SNR) is an important issue.

In information theory, Shannon entropy is a number that indicates the amount of information included in a signal. If there is a user first in the received signal, the entropy will be lower than in the case of noise only. By measuring the phase entropy of the received signal, it is possible to efficiently detect the user first.

In phase shift keying (PSK) modulation, all transmitted information is modulated by phase. However, the existing sensing method using entropy uses the power entropy of experimental signal samples. Calculating power is like calculating the energy of a sample. Therefore, in the conventional method, phase information is lost while calculating energy. As phase information is lost, it is not possible to obtain enough information of the signal, thus degrading the detection performance.

An entropy-based signal detection system is provided that provides a novel method of signal detection based on the advantage that phase is not as sensitive to signal power as noise and that entropy represents the amount of information in a received signal.

First, an entropy-based signal detection system is provided that improves the complexity of calculation as well as accurate detection performance in detecting a user.

A phase calculator for calculating a phase value of the received signal received at the receiver; An entropy calculator configured to calculate a phase entropy using a phase value of the received signal; And a signal detection system using phase entropy is provided that includes a determination unit for first determining whether the user exists by comparing the phase entropy and the threshold value.

According to one side, it is possible to first determine whether a user is present for a received signal using a phase shift keying (PSK) modulation scheme.

According to another aspect, a matching filter may be used at the receiving end, and the phase calculator may calculate the phase value of the received signal using the output value of the matching filter.

According to another aspect, the entropy calculator may calculate phase entropy using a histogram approach that represents a phase distribution of a received signal as a histogram.

According to another aspect, the entropy calculator may calculate phase entropy by matching a phase value with an amplitude of a received signal, and may first determine whether a user exists for a signal using a PAM method.

Since both amplitude and phase of the received signal are used to calculate entropy, more information can be obtained and thus the detection accuracy can be greatly increased.

Because both the amplitude and phase of the received signal are used, frequency information can be used to detect the user first, and cooperative sensing is possible for each performance improvement.

By protecting the phase signal of the received signal, very accurate detection performance can be provided, and the histogram approach can be used to calculate entropy, thereby improving the complexity of the calculation.

1 shows a complex signal diagram of an 8-PSK transmission signal.
2 shows a complex signal diagram of an 8-PSK received signal.
3 shows a phase distribution histogram of a received signal.
4 is a block diagram showing an internal configuration of a signal detection system using phase entropy according to an embodiment of the present invention.
5 illustrates an entropy distribution graph for a conventional signal detection algorithm.
6 illustrates a graph of phase entropy distribution of a signal detection algorithm according to an embodiment of the present invention.
7 illustrates a graph of ROC comparison of a signal detection algorithm according to an embodiment of the present invention.
8 illustrates an entropy distribution graph of a conventional signal detection algorithm in a PAM signal.
9 shows a graph of phase entropy distribution of a signal detection algorithm according to an embodiment of the present invention in a PAM signal.
10 illustrates a graph of ROC comparison of a signal detection algorithm according to an embodiment of the present invention in a PAM signal.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

An embodiment of the present invention relates to a wireless cognitive communication system based on a radio recognition technology, and more particularly, to a signal detection system that determines whether a user is present based on entropy of a signal received at a receiving end.

In the present specification, the radio recognition technology refers to a technology in which a frequency band is allocated but can be used by efficiently detecting an empty frequency band without actually using the same. In addition, unlike a preferred user who pays a fee and uses a licensed frequency band, the CR user is a user who uses a band not used by the first user and can wirelessly communicate with the wireless recognition technology base station. A terminal, ie, a user device and a device, may refer to a device that enables access to and use of a transmission path of a public communication service provider. In addition, entropy means a numerical value representing the amount of information included in a signal. If the received signal contains only noise, entropy is maximized, and entropy is reduced if it contains modulated information.

The signal detection system using phase entropy according to an embodiment may first detect a user's signal through a phase value of a received signal and an entropy measurement equation using a histogram approach.

The entropy measurement equation using phase information in the signal detection system using phase entropy according to an embodiment will be described in detail as follows.

1 shows a complex signal diagram of an 8-PSK transmission signal. FIG. 1 shows a preferred user's signal using 8-PSK (phase shift keying) modulation with the assumption that the signal is transmitted on a flat fading channel and subject to additional white Gaussian noise.

  Referring to FIG. 1, it can be seen that all PSK waveforms have the same energy. Equations 1, 2, and 3 define a complex signal vector of the received signal and the received signal.

는 채널 페이딩 이득이고,

은 변조된 신호이다. 평면 페이딩 채널에서, 채널은 데이터 심볼 기간에 비해 매우 천천히 변화한다. 따라서, g_i는 관찰 구간 동안 변하지 않는다.

와

는 무선채널에 의한 진폭과 위상을 나타내며 진폭

는 Rayleigh 또는 Rician분포이고 위상

는 구간

에서 uniformly 또는 non-uniformly분포이다.

은 M-PSK 변조된 신호이고 이것의 복소수 신호 벡터는 수학식 2와 같다.

Is the channel fading gain,

Is the modulated signal. In a flat fading channel, the channel changes very slowly compared to the data symbol period. Thus, g _i does not change during the observation interval.

Wow

Represents amplitude and phase by radio channel and amplitude

Is the Rayleigh or Rician distribution

Is the interval

In a uniformly or non-uniformly distribution.

Is an M-PSK modulated signal and its complex signal vector is given by Equation (2).

E _h is the passband pulse energy and M is the modulation level. The received signal may be represented as in Equation 3 using Equations 1 and 2.

2 shows a complex signal diagram of an 8-PSK received signal. Figure 2 shows a received signal of a preferred user using 8-PSK modulation.

The phase of the signal received through FIG. 2 may be calculated using Equations 4 and 5.

3 shows a phase distribution histogram of a received signal. FIG. 3 is a histogram illustrating the phase distribution of the 8-PSK signal described with reference to FIG. 2.

For the histogram approach, phase entropy can be defined as in Equation 6. It can be found in FIG. 3 that the distribution is regular. This shows that by measuring entropy, you can first distinguish your signal from noise.

Phase entropy is inversely related to signal-to-noise ratio. Thus it is maximized if the received signal is only noise. Therefore, the above two assumptions are shown in equation (7).

)과 오경보 확률(

)은 같이 높아진다. 이런 점에서 만약 CR이 오경보 확률을 신뢰하면 스펙트럼의 낭비가 발생한다. 그러나, 최적의 임계값은 이미 나와 있다.Using Equation 7, first, the user can be detected. When selecting a threshold, there is a balance between false alarm and miss detection.

) And false alarm probability (

) Is raised together. In this respect, if CR trusts the probability of false alarms, spectrum waste occurs. However, the optimal threshold is already shown.

According to an embodiment, a signal detection algorithm in a signal detection system using phase entropy is PAM (pulse). It can be extended to amplitude modulation) signals. In PAM signals, information is modulated according to amplitude. Therefore, the histogram approach is used to calculate the phase entropy after matching the amplitude of the received signal to the phase as shown in Table 1.

4 is a block diagram showing an internal configuration of a signal detection system using phase entropy according to an embodiment of the present invention.

As shown in FIG. 4, a signal detection system using phase entropy may include a phase calculator 420, an entropy calculator 430, and a determiner 440.

In the present embodiment, first, assuming that a user knows a waveform of a user signal wirelessly, the matching filter 410 may be used at a receiving end of the signal detection system. The phase calculator 420 calculates a phase value of the received signal using the output value of the matched filter 410 (Equations 4 and 5). In addition, the entropy calculator 430 may calculate phase entropy using the phase values of the received signal (Equation 6). Finally, the determination unit 440 may first determine whether the user exists by comparing the phase entropy calculation result and the threshold value (Equation 7).

Conventional signal detection algorithms using entropy use the power entropy of experimental signal samples. Calculating the power is the same as calculating the energy of the sample and has the following disadvantages. Equations 8 and 9 show disadvantages of the conventional method.

는 반송파 주파수를 나타낸다. 그리고

와

는 수학식 9에 나타나 있다.Considering any received signal consisting of sinusoids as shown above

Denotes the carrier frequency. And

Wow

Is shown in equation (9).

이므로

안에 포함된 위상 정보는 에너지를 계산하는 동안 손실된다. 그러나, 수학식 5와 수학식 6을 사용하면 엔트로피 측정 공식을 수학식 10처럼 다시 표현할 수 있다.

Because of

The phase information contained within is lost while calculating energy. However, using Equation 5 and Equation 6, the entropy measurement formula can be expressed as Equation 10.

In the above equation, phase information is preserved in the calculation process. Therefore, the signal detection algorithm using the phase entropy of the present embodiment can protect the phase information.

5 shows the entropy distribution of the conventional method. When the SNR is low, it is impossible to distinguish the PSK signal from the noise, so the user's signal cannot be detected first. In contrast, the method of the present invention in FIG. 6 can first detect the user's signal even when the SNR is low. The result is that phase is less sensitive to power than noise interference, so more information is available.

를 고정시키고 성능을 측정하면 제안된 방법의 검출확률

는 항상 기존의 방법보다 높게 나온다. 검출확률

를 고정시키면 제안된 방법은 기존의 방법보다 낮은 오경보 확률을 가진다. 이러한 결과는 제안된 방법이 기존의 방법과 비교해서 더 정확하고 신뢰적인 검출을 수행한다는 것을 보여준다.Another performance analysis was drawn for Receiver Operating Characteristic (ROC). Figure 7 shows the detection performance of the method of the present invention and the existing method at SNR = -4dB. In these two ROC curves, the slope with the larger slope shows better performance. False alarm probability

The probability of detection of the proposed method

Is always higher than the conventional method. Detection probability

If we fix, the proposed method has lower false alarm probability than the existing method. These results show that the proposed method performs more accurate and reliable detection than the conventional method.

As a result of the performance test on the PAM signal in FIGS. 8 and 9, it can be seen that the phase entropy deteriorates faster than the general entropy. This rapid reduction is an advantage for detecting a user first from the received signal. Also, re-simulating the ROC curve with SNR = -4dB, the proposed algorithm shows much better performance.

As described above, embodiments of the present invention can protect the phase signal of the received signal and provide very accurate detection performance. In addition, it shows good detection performance even at low SNR. If the existing sensing algorithm can use only amplitude information, the signal detection algorithm of the present invention can be used for phase modulated information.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above embodiments, and those skilled in the art to which the present invention pertains various modifications and variations from such descriptions. This is possible.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the claims below but also by the equivalents of the claims.

420: phase calculator
430: entropy calculation unit
440: judgment

Claims (5)

A phase calculator for calculating a phase value of the received signal received at the receiver;
An entropy calculator configured to calculate a phase entropy by using a phase value of the received signal; And
Determination unit for determining the presence of the user first by comparing the phase entropy and the threshold value
Including,
The entropy calculation unit,
Calculating the phase entropy using a histogram approach representing the phase distribution of the received signal as a histogram
Signal detection system using phase entropy characterized in that. The method of claim 1,
The received signal is
With signals using PSK (phase shift keying) modulation
Signal detection system using phase entropy characterized in that. The method of claim 1,
The signal detection system uses a matched filter at the receiving end,
The phase calculation unit,
Calculating a phase value of the received signal using the output value of the matched filter
Signal detection system using phase entropy characterized in that. delete The method of claim 1,
When the received signal is a signal using a pulse amplitude modulation (PAM) scheme,
The entropy calculation unit,
Calculating the phase entropy by matching the phase value to an amplitude of the received signal.
Signal detection system using phase entropy characterized in that.

Images