- Email: [email protected]
Adaptive fuzzy assisted detector under impulsive noise for DVB-T systems
Optik 127 (2016) 5196–5199
Contents lists available at ScienceDirect
Optik journal homepage: www.elsevier.de/ijleo
Adaptive fuzzy assisted detector under impulsive noise for DVB-T systems Ahmet Rizaner a,∗ , Ali Hakan Ulusoy a , Hasan Amca b a b
Department of Information Technology, Eastern Mediterranean University, via Mersin 10, Magosa, North Cyprus, Turkey Department of Electrical and Electronic Engineering, Eastern Mediterranean University, via Mersin 10, Magosa, North Cyprus, Turkey
a r t i c l e
i n f o
Article history: Received 22 January 2016 Accepted 29 February 2016 Keywords: DVB-T Adaptive fuzzy median filter Impulsive noise
a b s t r a c t The main novel contribution of this paper is the improvement achieved on the Digital Video BroadcastingTerrestrial (DVB-T) system by the addition of a fuzzy median filter to reduce the effect of impulsive components. The proposed detector includes the fuzzy median filter and can be adaptively adjusted according to the noise level and type of the noise. The system is studied under Gaussian and impulsive noise. Simulation results show that fuzzy assisted detector reduces the effect of impulsive noise and improves the performance of the system significantly while also working efficiently in Gaussian noise. © 2016 Elsevier GmbH. All rights reserved.
1. Introduction Digital Video Broadcasting (DVB) turns to be a suite of internationally accepted open standards for digital television after switching from the analogue to digital television broadcasting. The first generation terrestrial DVB (DVB-T) standard that is based on Orthogonal Frequency Division Multiplexing (OFDM) introduced an advance in digital TV transmission with well-structured transmission. A large number of countries, including many European Union countries, has already deployed or planning to deploy DVBT. Low-complexity and low-cost receiver design has been the main focus in order to enhance its popularity and enables its spreading around the globe [1]. High frequency efficiency of OFDM also played an important role in its adaptation and proposed it as the next generation communication and broadcasting standard. As the number of constellations in a DVB-T system increased, the data rate of OFDM systems could be increased significantly. However, in such OFDM systems as DVB-T, impulsive noise has been observed to effect system performance significantly [2,3]. Methods such as time domain processing before FFT transformation [4], OFDM with silent subcarriers and Reed–Solomon (RS) coding [5] have been successfully applied. Traditional methods of multicarrier signal detection in impulsive noise environments are using signal processing in time domain before demodulating the OFDM signals. Suppressing the effects of impulsive noise using these traditional methods in
∗ Corresponding author. Tel.: +90 392 630 2480; fax: +90 392 365 1574. E-mail address: [email protected] (A. Rizaner). http://dx.doi.org/10.1016/j.ijleo.2016.02.079 0030-4026/© 2016 Elsevier GmbH. All rights reserved.
OFDM systems have been shown to provide a degree of protection, which is far away from theoretical bounds [4]. The RS coder in a complex field encodes the OFDM signal in such a way that, zero sub-carriers are placed in predefined consecutive positions, exploiting some properties of the OFDM signal. This method has been employed for impulsive noise suppression in [5]. However, it has been concluded that conventional error correction decoders cannot eliminate the effects of impulsive interference and Gaussian noise simultaneously. More effective iterative methods based on frequency-domain equalization in conjunction with impulsive noise estimation have been proposed by other researchers in order to combat impairments due to impulsive noise [6,7]. Other methdos such as time-domain approach with a pre-processing mean filter and RS coding [8] and compensation method suitable for DVB-T systems in the frequency-domain after OFDM demodulation [9] have been also proposed for the suppression of impulsive noise. In [10], a fuzzy assisted decorrelating detector has been proposed that uses a fuzzy median filter to eliminate the effect of impulsive noise in multipath CDMA systems so that detector also works efficiently in impulsive channels. In this paper, an adaptive detector is proposed based on the principle of fuzzy logic that is used to reduce the effect of impulsive components of noise to improve the DVB-T system performance. An optimal threshold value for the fuzzy median filter is searched for each value of Signal-to-Noise Ratio (SNR) and used to adjust to minimize Bit Error Rate (BER). Although the proposed detector improves the performance under the impulsive noise, it performs worse in Gaussian noise since the fuzzy median filter designed for impulsive noise also removes useful signal content along with noise. To solve this problem, the detector is modified
A. Rizaner et al. / Optik 127 (2016) 5196–5199
to apply the fuzzy median filter only in the presence of impulsive noise so that the proposed detector is equally suitable for Gaussian noise. The rest of the paper is organized as follows. Section 2 presents a short overview of the DVB-T system model. In Section 3, we discuss the addition of the fuzzy media filter to the DVB-T system. Section 4 presents the simulation results and their discussions. Finally, in Section 5 we conclude with a final overview of the obtained results.
5197
vector, rg , respectively. The filter is then applied to the elementwise multiplication of received vector. The elements of received signal vector after the application of fuzzy filter can be expressed as: rF (k) = rg (k) ◦ F(k)
(4)
where ◦ is the element-wise multiplication operator. 4. Simulation results
2. System model DVB-T employs OFDM as the transmission scheme. After performing RS coding, outer interleaving, inner coding (i.e. punctured convolutional code), and inner interleaving, inverse FFT is performed on the modulated data symbols before transmission. The data symbols are then transmitted block by block at a fixed rate after inserting the cyclic prefixes as Guard Intervals (GI) between two consecutive blocks [1]. After GI is removed, the received baseband DVB-T signal plus the impulsive noise, v(k), can be expressed as rg (k) =
1 NFFT
Nu −1
X(n) exp
n=0
2jkn NFFT
+ v(k)
(1)
where k = 1 . . . NFFT , NFFT is the FFT length and X(n) are the modulated data symbols of Nu subcarriers. In this paper, two-term Gaussian-mixture model, which is widely used, is adopted for the ambient noise. This noise model can be defined by the following probability density function:
f = (1 − ε)v 0, 2 + εv 0, 2
(2)
with 0 ≤ ε ≤ 1 and ≥ 1. Here, ε and represent the probability that impulses occur and the variance factor of the impulsive component, respectively. The nominal Gaussian background noise with variance 2 is represented by the term v 0, 2 . The impulsive noise component is modeled with a larger variance, 2 , Gaussian density as v 0, 2 . The Gaussian-mixture model (2) which is an approximation to the more fundamental Middleton Class-A noise model [11] has been extensively used to model impulsive noise arising in mobile radio channels.
Extended simulation results are conducted to present the performance of adaptive fuzzy assisted detector for the DVB-T system both in Gaussian and impulsive channels. The key parameters of the simulated DVB-T system are given in Table 1. The other DVBT parameters are used as defined in [1]. The first simulations are performed to investigate the effect of vx in (3) on the performance of fuzzy assisted DVB-T system and determine an optimal value of this parameter. BER results are presented in Fig. 2 at different SNR values for the impulsive channel. It is observed from the results that the optimum value of vx is between 0.02 and 0.06 for the given SNR range. The optimal choices of vx values minimizing the BER at the given SNR values are presented in Table 2 together with the minimum BER values achieved. It is also concluded from the results of Fig. 2 and Table 2 that there is a linear relation between the optimal vx and SNR as vx = 0.005 SNR + 0.03. Hence, the fuzzy filter is adaptively set to vx according to this linear relation. In Fig. 3, the fuzzy filter is applied to the DVB-T system at different SNR values for Gaussian channel (ε = 0). The results show that fuzzy assisted Table 1 Simulation parameters. Parameters
Specification
NFFT (DVB-T mode) Number of active subcarriers GI fraction Outer coding Inner coding Signal constellation
2048 (2 K) 1705 1/4 RS (204, 188, t = 8) Convolutional code with code rate 3/4 QPSK
3. Proposed adaptive fuzzy assisted detector The effect of the impulsive noise should be suppressed at the DVB-T receiver by reducing the effects of extreme amplitudes. Hence, an adaptive filter that is based on fuzzy logic and rank ordering is added to the receiver as shown in Fig. 1. The received signal vector rg = rg (1) rg (2). . .rg (NFFT ) is fed into the fuzzifier to calculate membership degrees by using the following Gaussian membership functions:
2 −
rg (k) − g F(k) = exp 2vx g
(3)
where vx is the constant that is used to adaptively set the range of fuzzifier applied, g and g are the median and standard deviation of the absolute value of the rank-ordered version of the received Fig. 2. BER against vx under impulsive channel (ε = 0.2, = 100). Table 2 Optimum vx values at different SNR.
Fig. 1. Block diagram of the DVB-T receiver.
SNR
0 dB
2 dB
4 dB
6 dB
Optimum vx Minimum BER
0.03 7.99 × 10−5
0.04 7.59 × 10−5
0.05 1.45 × 10−5
0.06 3.99 × 10−6
5198
A. Rizaner et al. / Optik 127 (2016) 5196–5199
Fig. 3. BER against SNR under Gaussian channel (ε = 0).
Fig. 6. BER against SNR in impulsive channel (ε = 0.2, = 100).
Fig. 4. BER against threshold in Gaussian (ε = 0) and impulsive channels (ε = 0.2, = 100).
DVB-T performs worse under Gaussian noise since the fuzzy filter is designed for impulsive noise. To solve this problem, receiver structure is modified to apply the fuzzy filter only in the presence of impulsive noise. To understand if the noise is impulsive or not,
Fig. 5. Time-domain waveform plot of the signals before
the mean and the standard deviation of the absolute value of the incoming signal are calculated. The ratio of the difference between the mean and the standard deviation to standard deviation is compared to a predefined threshold. If this is less than or equal to the threshold, the channel is assumed to be impulsive and fuzzy filter is used. Fig. 4 presents the performance of the fuzzy assisted DVB-T at different threshold (Th) values under Gaussian and impulsive (ε = 0.2, = 100) noise. The results clearly show that applying a fuzzy filter with Th = 0.74 makes fuzzy assisted DVB-T system performing superior under the impulsive noise while working efficiently in Gaussian noise. Hence fuzzy assisted DVB-T can be used in both Gaussian and impulsive channels with a proper selection of the threshold. A time-domain waveform plot is presented in Fig. 5 to illustrate how the impulsive noise is by the
suppressed use of fuzzy filter showing the signals before rg (k) and after rF (k) the filter at 0 and 4 dB SNR. Fig. 6 presents the performance of the fuzzy assisted DVB-T system at Th = 0.72 and Th = 0.74 values with respect to SNR. The results show that the performance of fuzzy assisted DVB-T system under impulsive channel significantly outperforms the performance of the traditional DVB-T system at all SNRs since the effect of the impulsive noise is reduced successfully by the fuzzy filter. Finally, the effect of using the fuzzy filter on a sample picture is presented in impulse channel from another perspective in Figs. 7 and 8. As it is clearly shown in Fig. 7, the impulsive noise totally corrupts the picture for all SNR values when the traditional DVB-T system is used without the fuzzy filter. However by
rg (k) and after rF (k) the filter in impulsive channel (ε = 0.2, = 100).
A. Rizaner et al. / Optik 127 (2016) 5196–5199
5199
the use fuzzy assisted DVB-T system, the picture is received with acceptable BER at all SNR values as shown in Fig. 8. 5. Conclusion In this paper, an adaptive fuzzy assisted DVB-T system is designed to work efficiently under impulsive noise. Simulation results show the proposed fuzzy assisted DVB-T system provides a superior performance under impulsive noise and equally works under Gaussian noise with a proper threshold selection for the noise identification. References
Fig. 7. Received picture with traditional DVB-T system in impulsive channel (ε = 0.2, = 100).
Fig. 8. Received picture with fuzzy assisted DVB-T system in impulsive channel (ε = 0.2, = 100, Th = 0.74).
[1] DVB-T standards published, not a journal paper. ETSI EN 300 744 V1.6.1 (200901). Digital Video Broadcasting (DVB); Framing structure, channel coding and modulation for digital terrestrial television. ETSI, 01/2009. [Online] Available at: http://www.dvb-h.org/technology.htm. [2] Y. Wu, Performance comparison of ATSC 8-VSB and DVB-T COFDM transmission systems for digital television terrestrial broadcasting, IEEE Trans. Consum. Electron. 45 (August (3)) (1999) 916–924. [3] J. Ferrer-Coll, B. Simane, J. Chilo, P. Stenumgaard, Detection and suppression of impulsive noise in OFDM receiver, Wireless Pers. Commun. 85 (December (4)) (2015) 2245–2259. [4] T.C. Giles, On the Design of HF Radio Modems, University of Adelaide, Adelaide, December 1995 (Ph.D. dissertation). [5] J.K. Wolf, Redundancy, the discrete Fourier transform, and impulse noise cancellation, IEEE Trans. Commun. 31 (3) (1983) 458–461. [6] J. Haring, A.J. Han Vinck, Iterative decoding of codes over complex numbers for impulsive noise channels, IEEE Trans. Inf. Theory 49 (March (5)) (2003) 1251–1261. [7] S.V. Zhidkov, Impulsive noise suppression in OFDM based communication systems, IEEE Trans. Consum. Electron. 49 (November (4)) (2003) 944–948. [8] J. Jia, J. Meng, A dual protection scheme for impulsive noise suppression in OFDM systems, AEU Int. J. Electron. Commun. 68 (1) (2014) 51–58. [9] A. Bansal, A.K. Kohli, Suppression of impulsive noise in OFDM system using imperfect channel state information, Optik—Int. J. Light Electron Opt. 127 (February (4)) (2016) 2111–2115. [10] A.H. Ulusoy, A. Rizaner, Adaptive path selective fuzzy decorrelating detector under impulsive noise for multipath fading CDMA systems, IEEE Commun. Lett. 12 (4) (2008) 228–230. [11] X. Wang, H.V. Poor, Robust multiuser detection in non-Gaussian channels, IEEE Trans. Signal Process. 47 (2) (1999) 289–305.
Contents lists available at ScienceDirect
Optik journal homepage: www.elsevier.de/ijleo
Adaptive fuzzy assisted detector under impulsive noise for DVB-T systems Ahmet Rizaner a,∗ , Ali Hakan Ulusoy a , Hasan Amca b a b
Department of Information Technology, Eastern Mediterranean University, via Mersin 10, Magosa, North Cyprus, Turkey Department of Electrical and Electronic Engineering, Eastern Mediterranean University, via Mersin 10, Magosa, North Cyprus, Turkey
a r t i c l e
i n f o
Article history: Received 22 January 2016 Accepted 29 February 2016 Keywords: DVB-T Adaptive fuzzy median filter Impulsive noise
a b s t r a c t The main novel contribution of this paper is the improvement achieved on the Digital Video BroadcastingTerrestrial (DVB-T) system by the addition of a fuzzy median filter to reduce the effect of impulsive components. The proposed detector includes the fuzzy median filter and can be adaptively adjusted according to the noise level and type of the noise. The system is studied under Gaussian and impulsive noise. Simulation results show that fuzzy assisted detector reduces the effect of impulsive noise and improves the performance of the system significantly while also working efficiently in Gaussian noise. © 2016 Elsevier GmbH. All rights reserved.
1. Introduction Digital Video Broadcasting (DVB) turns to be a suite of internationally accepted open standards for digital television after switching from the analogue to digital television broadcasting. The first generation terrestrial DVB (DVB-T) standard that is based on Orthogonal Frequency Division Multiplexing (OFDM) introduced an advance in digital TV transmission with well-structured transmission. A large number of countries, including many European Union countries, has already deployed or planning to deploy DVBT. Low-complexity and low-cost receiver design has been the main focus in order to enhance its popularity and enables its spreading around the globe [1]. High frequency efficiency of OFDM also played an important role in its adaptation and proposed it as the next generation communication and broadcasting standard. As the number of constellations in a DVB-T system increased, the data rate of OFDM systems could be increased significantly. However, in such OFDM systems as DVB-T, impulsive noise has been observed to effect system performance significantly [2,3]. Methods such as time domain processing before FFT transformation [4], OFDM with silent subcarriers and Reed–Solomon (RS) coding [5] have been successfully applied. Traditional methods of multicarrier signal detection in impulsive noise environments are using signal processing in time domain before demodulating the OFDM signals. Suppressing the effects of impulsive noise using these traditional methods in
∗ Corresponding author. Tel.: +90 392 630 2480; fax: +90 392 365 1574. E-mail address: [email protected] (A. Rizaner). http://dx.doi.org/10.1016/j.ijleo.2016.02.079 0030-4026/© 2016 Elsevier GmbH. All rights reserved.
OFDM systems have been shown to provide a degree of protection, which is far away from theoretical bounds [4]. The RS coder in a complex field encodes the OFDM signal in such a way that, zero sub-carriers are placed in predefined consecutive positions, exploiting some properties of the OFDM signal. This method has been employed for impulsive noise suppression in [5]. However, it has been concluded that conventional error correction decoders cannot eliminate the effects of impulsive interference and Gaussian noise simultaneously. More effective iterative methods based on frequency-domain equalization in conjunction with impulsive noise estimation have been proposed by other researchers in order to combat impairments due to impulsive noise [6,7]. Other methdos such as time-domain approach with a pre-processing mean filter and RS coding [8] and compensation method suitable for DVB-T systems in the frequency-domain after OFDM demodulation [9] have been also proposed for the suppression of impulsive noise. In [10], a fuzzy assisted decorrelating detector has been proposed that uses a fuzzy median filter to eliminate the effect of impulsive noise in multipath CDMA systems so that detector also works efficiently in impulsive channels. In this paper, an adaptive detector is proposed based on the principle of fuzzy logic that is used to reduce the effect of impulsive components of noise to improve the DVB-T system performance. An optimal threshold value for the fuzzy median filter is searched for each value of Signal-to-Noise Ratio (SNR) and used to adjust to minimize Bit Error Rate (BER). Although the proposed detector improves the performance under the impulsive noise, it performs worse in Gaussian noise since the fuzzy median filter designed for impulsive noise also removes useful signal content along with noise. To solve this problem, the detector is modified
A. Rizaner et al. / Optik 127 (2016) 5196–5199
to apply the fuzzy median filter only in the presence of impulsive noise so that the proposed detector is equally suitable for Gaussian noise. The rest of the paper is organized as follows. Section 2 presents a short overview of the DVB-T system model. In Section 3, we discuss the addition of the fuzzy media filter to the DVB-T system. Section 4 presents the simulation results and their discussions. Finally, in Section 5 we conclude with a final overview of the obtained results.
5197
vector, rg , respectively. The filter is then applied to the elementwise multiplication of received vector. The elements of received signal vector after the application of fuzzy filter can be expressed as: rF (k) = rg (k) ◦ F(k)
(4)
where ◦ is the element-wise multiplication operator. 4. Simulation results
2. System model DVB-T employs OFDM as the transmission scheme. After performing RS coding, outer interleaving, inner coding (i.e. punctured convolutional code), and inner interleaving, inverse FFT is performed on the modulated data symbols before transmission. The data symbols are then transmitted block by block at a fixed rate after inserting the cyclic prefixes as Guard Intervals (GI) between two consecutive blocks [1]. After GI is removed, the received baseband DVB-T signal plus the impulsive noise, v(k), can be expressed as rg (k) =
1 NFFT
Nu −1
X(n) exp
n=0
2jkn NFFT
+ v(k)
(1)
where k = 1 . . . NFFT , NFFT is the FFT length and X(n) are the modulated data symbols of Nu subcarriers. In this paper, two-term Gaussian-mixture model, which is widely used, is adopted for the ambient noise. This noise model can be defined by the following probability density function:
f = (1 − ε)v 0, 2 + εv 0, 2
(2)
with 0 ≤ ε ≤ 1 and ≥ 1. Here, ε and represent the probability that impulses occur and the variance factor of the impulsive component, respectively. The nominal Gaussian background noise with variance 2 is represented by the term v 0, 2 . The impulsive noise component is modeled with a larger variance, 2 , Gaussian density as v 0, 2 . The Gaussian-mixture model (2) which is an approximation to the more fundamental Middleton Class-A noise model [11] has been extensively used to model impulsive noise arising in mobile radio channels.
Extended simulation results are conducted to present the performance of adaptive fuzzy assisted detector for the DVB-T system both in Gaussian and impulsive channels. The key parameters of the simulated DVB-T system are given in Table 1. The other DVBT parameters are used as defined in [1]. The first simulations are performed to investigate the effect of vx in (3) on the performance of fuzzy assisted DVB-T system and determine an optimal value of this parameter. BER results are presented in Fig. 2 at different SNR values for the impulsive channel. It is observed from the results that the optimum value of vx is between 0.02 and 0.06 for the given SNR range. The optimal choices of vx values minimizing the BER at the given SNR values are presented in Table 2 together with the minimum BER values achieved. It is also concluded from the results of Fig. 2 and Table 2 that there is a linear relation between the optimal vx and SNR as vx = 0.005 SNR + 0.03. Hence, the fuzzy filter is adaptively set to vx according to this linear relation. In Fig. 3, the fuzzy filter is applied to the DVB-T system at different SNR values for Gaussian channel (ε = 0). The results show that fuzzy assisted Table 1 Simulation parameters. Parameters
Specification
NFFT (DVB-T mode) Number of active subcarriers GI fraction Outer coding Inner coding Signal constellation
2048 (2 K) 1705 1/4 RS (204, 188, t = 8) Convolutional code with code rate 3/4 QPSK
3. Proposed adaptive fuzzy assisted detector The effect of the impulsive noise should be suppressed at the DVB-T receiver by reducing the effects of extreme amplitudes. Hence, an adaptive filter that is based on fuzzy logic and rank ordering is added to the receiver as shown in Fig. 1. The received signal vector rg = rg (1) rg (2). . .rg (NFFT ) is fed into the fuzzifier to calculate membership degrees by using the following Gaussian membership functions:
2 −
rg (k) − g F(k) = exp 2vx g
(3)
where vx is the constant that is used to adaptively set the range of fuzzifier applied, g and g are the median and standard deviation of the absolute value of the rank-ordered version of the received Fig. 2. BER against vx under impulsive channel (ε = 0.2, = 100). Table 2 Optimum vx values at different SNR.
Fig. 1. Block diagram of the DVB-T receiver.
SNR
0 dB
2 dB
4 dB
6 dB
Optimum vx Minimum BER
0.03 7.99 × 10−5
0.04 7.59 × 10−5
0.05 1.45 × 10−5
0.06 3.99 × 10−6
5198
A. Rizaner et al. / Optik 127 (2016) 5196–5199
Fig. 3. BER against SNR under Gaussian channel (ε = 0).
Fig. 6. BER against SNR in impulsive channel (ε = 0.2, = 100).
Fig. 4. BER against threshold in Gaussian (ε = 0) and impulsive channels (ε = 0.2, = 100).
DVB-T performs worse under Gaussian noise since the fuzzy filter is designed for impulsive noise. To solve this problem, receiver structure is modified to apply the fuzzy filter only in the presence of impulsive noise. To understand if the noise is impulsive or not,
Fig. 5. Time-domain waveform plot of the signals before
the mean and the standard deviation of the absolute value of the incoming signal are calculated. The ratio of the difference between the mean and the standard deviation to standard deviation is compared to a predefined threshold. If this is less than or equal to the threshold, the channel is assumed to be impulsive and fuzzy filter is used. Fig. 4 presents the performance of the fuzzy assisted DVB-T at different threshold (Th) values under Gaussian and impulsive (ε = 0.2, = 100) noise. The results clearly show that applying a fuzzy filter with Th = 0.74 makes fuzzy assisted DVB-T system performing superior under the impulsive noise while working efficiently in Gaussian noise. Hence fuzzy assisted DVB-T can be used in both Gaussian and impulsive channels with a proper selection of the threshold. A time-domain waveform plot is presented in Fig. 5 to illustrate how the impulsive noise is by the
suppressed use of fuzzy filter showing the signals before rg (k) and after rF (k) the filter at 0 and 4 dB SNR. Fig. 6 presents the performance of the fuzzy assisted DVB-T system at Th = 0.72 and Th = 0.74 values with respect to SNR. The results show that the performance of fuzzy assisted DVB-T system under impulsive channel significantly outperforms the performance of the traditional DVB-T system at all SNRs since the effect of the impulsive noise is reduced successfully by the fuzzy filter. Finally, the effect of using the fuzzy filter on a sample picture is presented in impulse channel from another perspective in Figs. 7 and 8. As it is clearly shown in Fig. 7, the impulsive noise totally corrupts the picture for all SNR values when the traditional DVB-T system is used without the fuzzy filter. However by
rg (k) and after rF (k) the filter in impulsive channel (ε = 0.2, = 100).
A. Rizaner et al. / Optik 127 (2016) 5196–5199
5199
the use fuzzy assisted DVB-T system, the picture is received with acceptable BER at all SNR values as shown in Fig. 8. 5. Conclusion In this paper, an adaptive fuzzy assisted DVB-T system is designed to work efficiently under impulsive noise. Simulation results show the proposed fuzzy assisted DVB-T system provides a superior performance under impulsive noise and equally works under Gaussian noise with a proper threshold selection for the noise identification. References
Fig. 7. Received picture with traditional DVB-T system in impulsive channel (ε = 0.2, = 100).
Fig. 8. Received picture with fuzzy assisted DVB-T system in impulsive channel (ε = 0.2, = 100, Th = 0.74).
[1] DVB-T standards published, not a journal paper. ETSI EN 300 744 V1.6.1 (200901). Digital Video Broadcasting (DVB); Framing structure, channel coding and modulation for digital terrestrial television. ETSI, 01/2009. [Online] Available at: http://www.dvb-h.org/technology.htm. [2] Y. Wu, Performance comparison of ATSC 8-VSB and DVB-T COFDM transmission systems for digital television terrestrial broadcasting, IEEE Trans. Consum. Electron. 45 (August (3)) (1999) 916–924. [3] J. Ferrer-Coll, B. Simane, J. Chilo, P. Stenumgaard, Detection and suppression of impulsive noise in OFDM receiver, Wireless Pers. Commun. 85 (December (4)) (2015) 2245–2259. [4] T.C. Giles, On the Design of HF Radio Modems, University of Adelaide, Adelaide, December 1995 (Ph.D. dissertation). [5] J.K. Wolf, Redundancy, the discrete Fourier transform, and impulse noise cancellation, IEEE Trans. Commun. 31 (3) (1983) 458–461. [6] J. Haring, A.J. Han Vinck, Iterative decoding of codes over complex numbers for impulsive noise channels, IEEE Trans. Inf. Theory 49 (March (5)) (2003) 1251–1261. [7] S.V. Zhidkov, Impulsive noise suppression in OFDM based communication systems, IEEE Trans. Consum. Electron. 49 (November (4)) (2003) 944–948. [8] J. Jia, J. Meng, A dual protection scheme for impulsive noise suppression in OFDM systems, AEU Int. J. Electron. Commun. 68 (1) (2014) 51–58. [9] A. Bansal, A.K. Kohli, Suppression of impulsive noise in OFDM system using imperfect channel state information, Optik—Int. J. Light Electron Opt. 127 (February (4)) (2016) 2111–2115. [10] A.H. Ulusoy, A. Rizaner, Adaptive path selective fuzzy decorrelating detector under impulsive noise for multipath fading CDMA systems, IEEE Commun. Lett. 12 (4) (2008) 228–230. [11] X. Wang, H.V. Poor, Robust multiuser detection in non-Gaussian channels, IEEE Trans. Signal Process. 47 (2) (1999) 289–305.