SNR Wall Effect Alleviation by Generalized Detector Employed in Cognitive Radio Networks

The most commonly used spectrum sensing techniques in cognitive radio (CR) networks, such as the energy detector (ED), matched filter (MF), and others, suffer from the noise uncertainty and signal-to-noise ratio (SNR) wall phenomenon. These detectors cannot achieve the required signal detection performance regardless of the sensing time. In this paper, we explore a signal processing scheme, namely, the generalized detector (GD) constructed based on the generalized approach to signal processing (GASP) in noise, in spectrum sensing of CR network based on antenna array with the purpose to alleviate the SNR wall problem and improve the signal detection robustness under the low SNR. The simulation results confirm our theoretical issues and effectiveness of GD implementation in CR networks based on antenna array.


Introduction
The main aim of the cognitive radio (CR) network is to improve the spectrum utilization efficiency, by introducing an opportunistic use of unemployed frequency band by the primary user (PU) (see Figure 1). The spectrum sensing is needed to define the frequency holes that could be allocated for OPEN ACCESS the secondary user (SU). The spectrum sensors search continuously an availability of frequency holes and assign them to SU without causing harmful interference to the PU. Fundamental limitations in practice are involved in spectrum sensing process [1][2][3]. The sensitivity to noise power uncertainty, for example, variations in the noise variance as a function of real time, is one of the most common and serious problems among the well-known spectrum sensors such as the energy detector (ED), matched filter (MF), and even the cyclostationary detector under some conditions at the low signal-to-noise ratio (SNR) [4]. The impact of noise power uncertainty is quantified by SNR wall location, i.e., if the SNR value is less than the SNR wall, the PU signal detector will fail to achieve the desired performance and maintain a robustness against power noise uncertainty independently of how long the sensing time is [3][4][5]. Both theoretical and experimental analysis confirmed the SNR wall phenomenon existence under the noise power uncertainty conditions. This phenomenon negatively effects the receiver operation characteristic (ROC). Other uncertainties also can be considered as SNR wall generators, for example, the noise power estimation error, assumptions made under the white and stationary noise, fading process, shadowing, non-ideal filters, non-precise analog-to-digital (A/D) converters, quantization noise, aliasing effect caused by imperfect front-end filters, and interference between the PU and SU. An alternative presentation for the SNR wall is given by the number of samples N as a function of SNR, the probability of false alarm FA P and probability of miss miss P , i.e., ( , , ) FA miss N f SNR P P = . The PU signal detector should minimize the number of samples N required to achieve the desired detection performance. The lowest SNR satisfying the probability of false alarm FA P and the probability of miss miss P constraints is called the detector sensitivity [3]. In general, the ideal ED does not have the SNR wall, but owing to the noise power uncertainty the ED suffers from the SNR wall phenomenon making the ED non-robust under the low SNR [6,7]. In many published papers, the ED spectrum sensing performance is investigated under the noise uncertainty conditions. Different solutions are presented in the form of the dynamic detection threshold [8], log-normal approximation of the noise uncertainty [9], falling the SNR wall using the cross-correlation [10], improving the noise power estimation using the maximum likelihood (ML) estimator [6], SNR estimation based on the pseudo bit error rate (BER) for the modified ED [11], and algebraic spike detection method introduced in [12,13]. In fact, the best non-coherent detector is non-robust as the ED under the noise power uncertainty. In the coherent detector case, the SNR wall is pushed back only to a limited value and for a large channel coherence time c K → ∞ . In the MF case, the SNR wall location is proportional to 1/ c K [3], and in the case of feature detector, the SNR wall value is less in comparison with the ED one and scales only as 1/ c K with the relevant channel coherence time [3].
An interesting new four-level hypothesis blind detector for spectrum sensing in CR systems is presented in [14]. The proposed detector in [14] reduces the negative effects on the CR system performance, which are forming under the in-phase and quadrature-phase (I/Q) imbalance, based on the orthogonal frequency division multiplexing (OFDM) multiple access scheme, and presents a promising solution for any noise power uncertainties or SNR wall problem that could be caused by this I/Q imbalance. Cooperative spectrum sensing, in the course of which the multiple sensors are involved in cooperative spectrum sensing, demonstrates an effective approach to improve the spectrum sensing performance under several problems such as the noise power uncertainty, multipath fading, shadowing, and receiver uncertainties issues. The cooperative spectrum sensing can also solve the critical energy efficiency issue as shown in [15] where the energy efficient cooperative spectrum sensing is proposed and the optimal scheduling of active time for each spectrum sensor helps to extend the network lifetime. Selective grouping based on the cooperative sensing is discussed in [16] where during the sensing time each sensors group senses different radio channels while sensors in the same group perform the joint detection by the targeted channel. This process assures obtaining the more robust and efficient sensing performance comparing with the individual spectrum sensor case under the noise power uncertainty.
To mitigate the negative effects of noise power uncertainty at the low SNR, an implementation of the generalized detector (GD), which is constructed based on the generalized approach to signal processing (GASP) in noise, for the spectrum sensing in CR networks based on antenna array is proposed. The GD represents a combination of the correlation detector, which is optimal in the Neyman-Pearson (NP) criterion sense when there is a priori information about the PU signal parameters, and ED, which is optimal in the NP criterion sense if there is no any a priori information about the PU signal parameters that are random [17][18][19]. The GD likelihood ratio test, based on which we can make a decision about the PU signal presence or absence in the process incoming at the SU input, demonstrates a definition of the jointly sufficient statistics of the mean and variance of the likelihood ratio and does not require any information about the PU signal and its parameters [17], ([18], Chapter 3). As was discussed in detail in ( [18], Chapter 7, pp. 685-692), the main function of GD energy detector (GD ED) is to detect the PU signal and the main function of the GD correlation detector is to define the detected PU signal parameters and make a decision: the detected signal is the expected PU signal with the required parameters or not.
Note that the conventional correlation detector makes a decision about the PU signal presence or absence in the incoming process based on definition of the mean only of the process incoming at the SU input. The conventional ED defines the decision statistics with respect to PU signal presence or absence at the SU input based on determination of the variance only of the process incoming at the SU input. Definition of the jointly sufficient statistics of the mean and variance based of the incoming process at the SU input allows us to make more accurate decision in favor of the PU signal presence or absence and obtain more information about the PU signal parameters under GD employment in CR networks in comparison with the conventional MF, ED, correlation receiver and so on.
A great difference between the GD ED and conventional ED is a presence of the additional linear system (the additional bandpass filter at the GD input) considered as the secondary data or reference noise source. The PU signal bandwidth is mismatched with the additional linear system bandwidth. The PU signal bandwidth is matched only with another linear system bandwidth at the GD front-end. Thus, the GD has two input linear systems, namely, the preliminary filter (PF) and the additional filter (AF). The last is considered as the reference noise source ( [18], Chapter 3), [19]. The GD PF central frequency is detuned relatively to the GD AF central frequency to ensure, firstly, the PU signal passing only through GD PF and, secondly, the independence and uncorrelatedness between the stochastic processes at the GD PF and AF outputs. Thus, it is possible to obtain the PU signal plus noise at the GD PF output in the case of "a yes" PU signal at the GD input and only the noise in the opposite case. Consequently, only the noise is obtained at the GD AF output for both cases of "a yes" and "a no" PU signal at the GD input, in other words, under the hypotheses 1 H and 0 H . The case when there is the PU signal generated by another source with the frequency content within the limits of the GD AF bandwidth, and considered as the additional interference, is discussed in [20]. The GD employment in wireless communications [21,22], radar sensor systems [20,23], and CR networks for spectrum sensing [24] allows us to improve the signal detection performance of these systems in comparison with implementation of widely used conventional detectors. This work differs from the previously published paper [24] by introducing a new advantage of GD employing in CR network systems based on antenna array, which is the SNR wall problem alleviation under the noise power uncertainty. Additionally, the GD optimal detection threshold is defined based on the minimal probability of error criterion under the noise power uncertainly at the low SNR condition. Intuitive approach to reduce the noise power uncertainty at run time by employing the GD in CR network is to define the noise power at the GD AF output, i.e., the another narrow band closed to the PU signal frequency band, with the purpose to calibrate the noise power in the PU signal frequency band. Even if we believe that the noise power forming at the GD PF and AF outputs is not the same, the noise calibration error can be much lower than the noise power uncertainty itself. The noise power calibration in real time improves the immunity against the SNR wall phenomenon [3]. In this paper, we investigate the GD noise power calibration effects on the SNR wall problem in coarse spectrum sensing for CR network systems based on antenna array and we define the GD sample complexity under the noise power uncertainty. The complementary receiver operating characteristic (ROC) and sample complexity of the ED, MF, and GD are compared under the same initial conditions for different uncertainty parameters. The real scenario of simulation demonstrates that the GD is able to alleviate the SNR wall problem and achieve the low probability of error in comparison with the conventional ED.
The reminder of this paper is organized as follows. Section 2 presents the system model and the GD test statistics. Section 3 delivers the GD signal detection performance under the noise power uncertainty. The real scenario simulation results are discussed in Section 4. The concluding remarks are presented in Section 5.

System Model
The spectrum sensor has an antenna array with the number of elements equal to M and each antenna array element receives N samples during the sensing time. The spectrum sensing problem can be modeled as the conventional binary hypothesis test:  uncorrelated between each other. The same channel model is widely used in [25][26][27]. In general, the ED does not require channel state information (CSI) for spectrum sensing [28] and the GD shares this property with ED because the ED is a constituent of the GD. It is well known that information about the CSI allows us to obtain better spectrum sensing performance in comparison with unknown CSI case. The knowledge about CSI can be more useful and effective in the cooperative spectrum sensing case. Under the low SNR and noise power uncertainty conditions, we can claim that we have imperfect CSI [29]. When the noise power estimation is applied, we have partial knowledge about the CSI. In this paper, we assume that the coarse spectrum sensing is performed without knowledge about the CSI. Owing to its simplicity, the exponential matrix model is widely used to describe the spatial correlation between the adjacent antenna array elements [30]. The components of the M M × antenna array element correlation matrix C can be presented in the following form: where ρ is the coefficient of spatial correlation between the adjacent antenna array elements (0 1 ≤ ρ ≤ , the real values). Applying the results presented in [30], the coefficient of spatial correlation ρ can be given as where Λ is the angular spread, an important propagation parameter defining a distribution of multipath power of radio waves coming in at the receiver input from a number of azimuthal directions with respect to the horizon; λ is the wavelength; and d is the distance between two adjacent antenna array elements (the antenna array element spacing). The correlation matrix of antenna array elements C given by Equation (2) is the symmetric Toeplitz matrix [25].
We define the 1 NM × signal vector Z that collects all the observed signal samples during the sensing time using the following form: where 0 M is the M M × zero matrix.

GD Statistics
The GD has been constructed based on the (GASP) in noise discussed in detail in [17][18][19]. The GD is considered as a linear combination of the correlation detector, which is optimal in the Neyman-Pearson criterion sense under detection of signals with a priori known parameters, and the ED, which is optimal in the Neyman-Pearson criterion sense under detection of signals with a priori unknown or random parameters. The main functioning principle of GD is a complete matching between the model signal generated by the local oscillator in GD and the information signal, in particular, the PU signal at the GD input by whole range of parameters. In this case, the noise component of the GD correlation detector caused by interaction between the model signal generated by the local oscillator in GD and the input noise and the random component of the GD ED caused by interaction between the incoming information signal (the PU signal) and input noise are cancelled in the statistical sense. This GD feature allows us to obtain the better detection performance in comparison with other classical receivers or detectors.
The specific feature of GASP is introduction of the additional noise source that does not carry any information about the incoming signal with the purpose to improve a qualitative signal detection performance. This additional noise can be considered as the reference noise without any information about the PU signal [17]. The jointly sufficient statistics of the mean and variance of the likelihood ratio is obtained in the case of GASP implementation, while the classical and modern signal processing theories can deliver only a sufficient statistics of the mean or variance of the likelihood ratio. Thus, the implementation of GASP allows us to obtain more information about the input process or received information signal (the PU signal). Owing to this fact, an implementation of receivers constructed based on the GASP basis allows us to improve the spectrum sensing performance of CR wireless networks in comparison with employment of other conventional receivers at the sensing node. The GD flowchart is presented in Figure 2. As we can see from Figure 2, the GD consists of three channels: • The GD correlation channel-the PF, multipliers 1 and 2, model signal generator MSG; • The GD ED channel-the PF, AF, multipliers 3 and 4, summator 1; • The GD compensation channel-the summators 2 and 3 and accumulator Σ.
As follows from Figure 2 To describe the GD flowchart we consider the discrete-time processes without loss of any generality. Evidently, the cancelation in the statistical sense between the GD correlation channel noise component h m , respectively. For simplicity of analysis, we assume that these filters have the same amplitude-frequency characteristics or impulse responses by shape. Moreover, the GD AF central frequency is detuned with respect to the GD PF central frequency on such a value that the information signal (the PU signal) cannot pass through the GD AF. Thus, the PU signal and noise can appear at the GD PF output and the only noise is appeared at the GD AF output (see Figure 3). If a value of detuning between the GD AF and PF central frequencies is more than 4 or 5 s f Δ , where s f Δ is the PU signal bandwidth, the processes at the GD AF and PF outputs can be considered as the uncorrelated and independent processes and, in practice, under this condition, the coefficient of correlation between GD PF and AF output processes is not more than 0.05 that was confirmed experimentally [31,32].
In the present paper, we consider the spectrum sensing problem of a single radio channel where the GD AF bandwidth is always idle and cannot be used by the SU because it is out of the useful spectrum of the PU network. There is a need to note that in a general case, the GD AF portion of the spectrum may be occupied by the PU signals from other networks and can be not absolutely unoccupied. In this case, the PU signals from other networks can be considered as interferences or interfering signals. Investigation and study of GD under this case is discussed in [20].
The processes at the GD AF and PF outputs present the input stochastic samples from two independent frequency-time regions. If the noise [ ] w k at the GD PF and AF inputs is Gaussian, the noise at the GD PF and AF outputs is Gaussian, too, because the GD PF and AF are the linear systems, and we believe that these linear systems do not change the statistical parameters of the input process. We use this assumption for simplicity of theoretical analysis. Thus, the GD AF can be considered as a reference noise source with a priori knowledge a "no" signal (the reference noise sample). Detailed discussion of the GD AF and PF can be found in [18,19]. The noise at the GD PF and AF outputs can be presented in the following form: Under the hypothesis 1 , H the signal at the GD PF output can be defined is the observed noise at the GD PF output and   [17,18] is extended to the case of antenna array employment when an adoption of multiple antennas and antenna arrays is effective to mitigate the negative attenuation and fading effects [20,24]. The GD decision statistics can be presented in the following form: where 1 1 is the stochastic process vector at the GD PF output and GD THR is the GD detection threshold. We can rewrite Equation (11) in the vector form: where is the 1 M × vector of the random process at the GD PF output with elements defined as is the 1 M × vector of the process at the MSG output with the elements defined as is the 1 M × vector of the random process at the AF output with the elements defined as and GD THR is the GD detection threshold. According to GASP and GD structure shown in Figure 2 and the main GD functioning condition (8), the GD test statistics takes the following form under the hypotheses 1 H and 0 H , respectively: The term We use representation Equation (22) in the following discussion, for example, in Section 3.

Moment Generation Function of the GD Partial Test Statistics
is required. The MGF of the GD partial test statistics ( ) X GD k T is presented as: Derivation of Equation (24)

GD Spectrum Sensing and Sample Complexity
The spectrum sensor should minimize the number of samples  ( 1 ) is the Gaussian Q -function.
In the noise power uncertainty case, the noise power or variance at the GD PF and AF outputs can be determined only within the limits of a definite range [3] (see Figure 4 is the uncertainty parameter; ε is the parameter used to define the amount of non-probabilistic uncertainty in the noise power. In the case of noise power uncertainty, Equations (29) and (30) can be written in the following form: is the SNR at the GD input. Based on Equations (34) and (35) As 1 SNR << , substituting Equation (37) in Equation (35), the GD sample complexity can be defined as Here we assume that 2 w w − η σ ∈ ρ σ ρσ . As follows from Equation (38), the sample complexity GD N is inversely proportional to the squared SNR.
We can notice that there is no additional term involving the SNR in the denominator of Equation (38) which leads to the noise power uncertainty calibration, i.e., the SNR wall alleviation. This is caused by the complete compensation in the ideal case between the noise component Following the above-mentioned procedure we can obtain the sample complexity for ED, which can be determined in the following form: As follows from (40), we can define the ED SNR wall in the following form [3] The relation between the probability of miss ED miss P and probability of false alarm ED FA P can be defined as In the MF case, the effective SNR is provided by the coherent processing gain. Thus, the MF sample complexity is given by [3] where c K is the coherence time of the radio channel, i.e., the time interval, within the limits of which the channel impulse response is not varied; θ is a fraction of the total power that is allocated to the known pilot tone. This concept covers many practical wireless communication systems employing the pilot tones and training known sequences for synchronization and timing acquisition. The MF SNR wall can be presented in the following form [3]: The ED has the better sample complexity performance at the high SNR in comparison with the MF because the ED uses the total average PU signal power for detection while the MF uses only a fraction of the total PU signal power. In the case of MF possessing the pilot tone detection scheme, the SNR wall phenomenon is a consequence of time-selectivity of the channel fading process and the signal power is increased with the factor c K owing to increasing the coherent processing gain. This is the reason why we see that the MF is sensitive to the channel coherence time c K . Thus, the effective SNR of the coherently combined signal according to [1,2]  where μ is the amplitude coefficient of proportionality. Under the condition given by Equation (46), the MGF of the GD partial decision statistics Based on Equation (48) In the case of noise power uncertainty and under the condition given by Equation (46), Equation (49) allows us to define the probability of false alarm GD FA P and the probability of miss GD miss P for GD using the following form: ; σ σ σ 2ρσ Under delivering (51) we ignore the term 2 2 (2 1) SNR μ − since CR networks operate at very low SNR values, i.e., 1 SNR << . Defining the threshold GD THR in terms of the probability of false alarm GD FA P based on (50) and substituting it in (51), we obtain At the low SNR values, we can apply the following approximation 2 2 ( 1) 1 1 SNR μ − + ≈ and determine the GD sample complexity using the following form: At μ 1 = we obtain the sample complexity GD N given by Equation (38).

The GD Optimal Threshold
As a matter of fact, the ED and GD ignore the PU signal characteristics and rely only on the PU signal energy. Thus, the ED and GD optimal threshold should be proportional to the nominal noise power at the SU input. In practice, the noise power is unknown and should be estimated by the GD noise power estimator (NPE in Figure 2). As a result, both the ED and GD detection thresholds can be defined based on the total error rate minimization [34][35][36]. In the case of the additive white Gaussian noise (AWGN) channel, the GD optimal threshold can be defined using the minimal probability of error in the following form: where GD er P is the probability of error given by σ ; σ σ σ 1 ρ As we can see from (56), in the ideal case, i.e., when there is no noise power uncertainty 1 ρ = and 1 β = or 2 2 2 ξ η σ = σ = σ , the optimal detection threshold is determined as In practice, in the GD case, there is no need to define or know a priori the value of ρ since the noise power is estimated in the real time using the NPE (see Figure 2). We consider the optimal threshold under the noise power uncertainty for the theoretical analysis presented in this paper.
Since the estimated noise power is differed from the real noise power, the noise power uncertainty is an unavoidable problem in practice [2,3,37]. As discussed in [6,38], in the ED case, the SNR wall phenomenon is caused by insufficient refinement of the noise power estimation while the observation time is increased and the noise power estimation approach can avoid the SNR wall problem if the noise power estimate is consistent within the limits of the observation interval. Finally, we cannot rely on the noise power estimation to solve the SNR wall problem. The results presented in [6] are applicable for ED under the use of the noise power estimation and can be applied to GD implementation in CR networks.

Simulation and Discussion
The sample complexity and existence of the SNR wall for the ED and MF and non-existence of the SNR wall for the GD are verified by the real scenario simulation that is performed using MATLAB in accordance with the parameters presented in the IEEE 802.22 standards, i.e., the standard for wireless regional area network WRAN using white spaces in the TV broadcast bands such as the digital video broadcasting-terrestrial DVB-T. The simulation parameters are presented in Table 1.  , the GD presents the best sample complexity performance in comparison with the ED and MF under conditions of the noise power uncertainty. The GD overcomes a negative impact of the noise power uncertainty. Thus, the GD can detect the PU signal at any arbitrary low SNR increasing the number N of samples. In other words, there is no SNR wall. In the case of ED, when there is no noise power uncertainty, i.e., 1 ρ = , there is no SNR wall and the PU signal can be detected at any low SNR by increasing the sensing time or the number N of samples. If there is the noise power uncertainty, there is the SNR wall for the ED and its location depends on the value of ε and, consequently, the uncertainty parameter . ρ Small values of ε , the least uncertainty case, are preferred because, in this case, there is a decreasing in SNR wall.  σ ≠ σ , is presented in Figure 6 at 1 ε = dB, 2 M = and several values of μ and β . As we can see from Figure 6, the best GD sample complexity efficiency is obtained at 1 μ = or mod  = σ . Additionally, we can see that there is no SNR wall in the GD case, but the GD sample complexity efficiency decreases at 1 μ ≠ and 1 β ≠ . In this case, more samples are needed at the same SNR value to achieve the required probability of false alarm. The complementary receiver operating characteristic (ROC) curves, which are widely used in practice, for example in [39][40][41], for the ED and GD are presented in Figure 7 with and without the noise power uncertainty at 6 M = and 20 N = . In a general, for both detectors the noise power uncertainty leads to the complementary ROC curves shifting away from the (0,0) origin. As shown in Figure 7, the GD demonstrates the better sensing performance in comparison with the ED and the sensing performance degradation rate of GD is less under the noise power uncertainty conditions. In the GD case, under the low SNR or if the SNR is above the  Figure 6). In Figure 8, a comparison between the ED and GD performance in terms of the probability of error er P as a function of the sample number N, the analogous performance is discussed in [36], is shown at 0.1 ε = dB, 10 SNR = − dB, and 13 SNR = − dB. The GD demonstrates the better sensing performance in comparison with the ED one. For example, at 2 10 N = the probability of error er P is equal to 0.3126 in the GD case and 0.5346 in the ED case. At 13 SNR = − dB that corresponds to the ED wall SNR when 0.1 ε = dB, we can see that the probability of error er P in the ED case fails to be robust and is distinctly differed owing to the SNR wall phenomenon. In this case, an increasing in the number of samples N is not effective to improve the probability of error er P performance for ED. At the same time, the GD has the same normal behavior meaning that the probability of error er P performance for GD is improved with increasing in the number of samples N.  Comparison between the probability of error er P for the ED and GD as a function of the normalized optimal detection threshold, where NM is the normalization factor, is presented in Figure 9. The probability of error er P is evaluated for both detectors in two cases: there is the noise power uncertainty and there is no noise power uncertainty at the 5 SNR = − dB, 2 M = , 100 N = , 0.1 ε = and 1dB. As shown in Figure 9, the GD can achieve the lower probability of error er P in comparison with the ED for both cases. For example, if there is no noise power uncertainty the minimal probability of error er P is equal to 0.13 in the GD case and 0.25 in the ED case. If there is the noise power uncertainty with 0.1 ε = dB, the lowest probability of error er P for the GD is equal to 0.24 and 0.33 for the ED. In a general case, the noise power uncertainty affects negatively on the ED and GD probability of error er P . Thus, we can make the following conclusion: increasing in the noise power uncertainty leads to increasing in the probability of error er P .  σ ≠ σ at the GD PF and AF outputs on the probability of error er P as a function of the normalized detection threshold given by Equation (56) dB. We can notice that the β value effects GD performance. For example, at 0.9 β = the probability of error GD er P is approximately equal to 0.26 and at 0.5 β = the probability of error GD er P is equal to 0.32.
As follows form the theoretical analysis and simulation results, the GD implementation allows us to improve the spectrum sensing accuracy that is defined by the probability of false alarm FA P and the probability of detection D P . Additionally, the GD allows us to alleviate the SNR wall problem by calibrating the noise power uncertainly increasing the number of samples. Thus, the GD employment allows us to improve the signal detection and signal processing performance. The main GD can be applicable in many practical systems, such as the adaptive and spectrum efficient communication systems, CR network systems, and carrier sense multiple access based on wireless networks. In terms of complexity, the GD implementation can be more complicated in comparison with some conventional detectors, for example, the ED. The complexity of GD implementation in practice is caused by the following problems: (1) the inequality between the noise power or variances at the GD PF and AF outputs (discussed in this paper); (2) the problem of matching by parameters between the model signal and the incoming PU signal parameters, for example, by the amplitude or energy (discussed in this paper); (3) the interfering signals within the frequency content of the GD AF, i.e., the GD AF bandwidth (discussed in [20]).

Conclusions
The actual spectrum sensing performance of the well-known detectors employed in CR networks based on antenna array, such as the ED and MF deviates from the theoretical results owing to the noise power uncertainty and SNR wall phenomenon. This phenomenon has a negative impact on the spectrum sensing performance and on the receiver operation characteristic (ROC) when increasing in the sensing time has no any compensating effects. In this paper, we demonstrate that under implementation of the GD in CR networks based on antenna array there is no GD wall SNR in the case of noise power uncertainty that is confirmed by the real scenario simulation. The GD can calibrate the noise power uncertainty problem by the compensation channel (see Figure 2) using the reference noise forming at the GD AF output. The GD is able to detect the PU signal at any low SNR value with increasing in the number of samples that is still not the ideal solution under fast spectrum sensing. Thus, the GD implementation in CR networks based on antenna array allows us to reduce some negative effects caused by the noise power uncertainty and improve the PU signal detection performance and robustness. The probability of error er P as a function of the normalized optimal detection threshold is evaluated for the ED and GD both under presence and absence of the noise power uncertainty. The GD demonstrates the better probability of error er P performance in comparison with the ED in both cases. Finally, as is demonstrated by the simulation results, with an increase in the noise power uncertainty, the probability of error er P increases as well. We say that any random variable x has a chi-square distribution with υ degree of freedom if its probability density function (pdf) is presented as where c is a constant given by [42] 0 where the values a and b are the arbitrary constants, we can represent Equation (B4) using the following form: