Blind equalization and Automatic Modulation Classification based on pdf fitting

Fki, Souhaila; Bey, Abdeldjalil Aïssa El; Chonavel, Thierry

doi:10.1109/icassp.2015.7178519

Cited by 8 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The computations are very similar to the ones related to the estimation of the carrier frequency, carrier phase and timing epoch derived in [7]. The next result follows from (22) by recalling that symbols d { } k are assumed to be i.i.d. with zero mean and variance σ d…”

Section: Cramer-rao Bound For the Roll-off Estimation Problemmentioning

confidence: 66%

“…Linear and nonlinear digital modulation classification has received a lot of attention in the literature. Several classification rules based on the maximum likelihood method [29,9,20,8,11,4,15] or on appropriate features [22,10,24,5,13,21] extracted from the received communication signal have been investigated. The robustness of the resulting classifiers to synchronization errors or channel impairments has also been studied [2,30].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Blind roll-off estimation for digital transmissions

2017

View full text Add to dashboard Cite

A B S T R A C TThis paper addresses the problem of estimating the roll-off factor of a received communication signal. We study two new statistical estimation methods that determine the roll-off factor by minimizing the difference between the theoretical and empirical power or power spectral density of the received signal. Another interesting contribution is the derivation of the roll-off Cramér-Rao bound which provides a reference in terms of estimation variance. Simulation results conducted on synthetic data allow the performance of the proposed methods to be evaluated. They are compared to a recent technique based on the amplitude fluctuations of the power spectral density associated with the received communication signal. The estimation methods are shown to be robust to channel impairments (including white Gaussian noise and synchronization errors). The proposed strategies are finally tested on real signals with known ground truth showing their possible application to digital communication problems.

show abstract

Section: Cramer-rao Bound For the Roll-off Estimation Problemmentioning

confidence: 66%

Section: Introductionmentioning

confidence: 99%

Blind roll-off estimation for digital transmissions

2017

View full text Add to dashboard Cite

show abstract

“…The main objective of blind equalization is to remove the ISI of the received signal, for example, in wireless communications systems, ISI occurs frequently because of limited bandwidth and multipath propagation [45]. The BCMA method is used to perform equalization in the AMC2N method.…”

Section: Blind Equalizationmentioning

confidence: 99%

AMC2N: Automatic Modulation Classification Using Feature Clustering-Based Two-Lane Capsule Networks

et al. 2021

View full text Add to dashboard Cite

The automatic modulation classification (AMC) of a detected signal has gained considerable prominence in recent years owing to its numerous facilities. Numerous studies have focused on feature-based AMC. However, improving accuracy under low signal-to-noise ratio (SNR) rates is a serious issue in AMC. Moreover, research on the enhancement of AMC performance under low and high SNR rates is limited. Motivated by these issues, this study proposes AMC using a feature clustering-based two-lane capsule network (AMC2N). In the AMC2N, accuracy of the MC process is improved by designing a new two-layer capsule network (TL-CapsNet), and classification time is reduced by introducing a new feature clustering approach in the TL-CapsNet. Firstly, the AMC2N executes blind equalization, sampling, and quantization in trilevel preprocessing. Blind equalization is executed using a binary constant modulus algorithm to avoid intersymbol interference. To extract features from the preprocessed signal and classify signals accurately, the AMC2N employs the TL-CapsNet, in which individual lanes are incorporated to process the real and imaginary parts of the signal. In addition, it is robust to SNR variations, that is, low and high SNR rates. The TL-CapsNet extracts features from the real and imaginary parts of the given signal, which are then clustered based on feature similarity. For feature extraction and clustering, the dynamic routing procedure of the TL-CapsNet is adopted. Finally, classification is performed in the SoftMax layer of the TL-CapsNet. This study proves that the AMC2N outperforms existing methods, particularly, convolutional neural network(CNN), Robust-CNN (R-CNN), curriculum learning(CL), and Local Binary Pattern (LBP), in terms of accuracy, precision, recall, F-score, and computation time. All metrics are validated in two scenarios, and the proposed method shows promising results in both.

show abstract

“…We can cite the multi-branch architecture [3], it uses several branches each linked to a constellation and the one that gives the smallest error matches the transmitted constellation. Other algorithms use the cost function relative to the quadrature phase shift keying (QPSK) constellation to equalize any transmit constellation belong-ing to the phase-shift keying modulation (PSK) or the QAM modulation [4] [5]. As a result, the output of the equalizer is constituted of symbols belonging to the transmitted constellation and compressed in the unit radius.…”

Section: Introductionmentioning

confidence: 99%

“…Blind equalizers based on PDF like the stochastic quadratic distance (SQD) and multi modulus SQD (MSQD-ℓ p ) outperform equalizers based on high order statistical properties (HOSP) like constant modulus algorithm (CMA) and multi modulus algorithm (MMA) [7]. In [5] and for SISO communication systems, generic equalizer MSQDℓ 2gen based on MSQD-ℓ 2 criterion outperforms generic equalizer CMA gen based on CMA. Therfore, in [6] and for SISO communication systems, the multi-criteria generic equalizer MC-MSQDℓ 2gen outeprforms the MSQD-ℓ 2gen .…”

Section: Introductionmentioning

confidence: 99%

Multi-user generic channel equalization and sources separation for MIMO communication systems

Farhati,

Fki,

Aïssa-El-Bey

et al. 2023

2023 IEEE Tenth International Conference on Communications and Networking (ComNet)

Self Cite

View full text Add to dashboard Cite

In this work, we study blind equalization techniques to mitigate inter-symbol interference (ISI), and blind sources separation (BSS) in multiple inputs multiple outputs (MIMO) communication systems. Mainly, we are interested in the multi-user generic blind equalizer (MU-GBE). A MU-GBE has no prior information about the transmission channels and the used constellations. To solve this challenge, a joint MU-GBE, based on a new multi-criteria cost function and automatic modulation classification (AMC) is proposed. The new multi-users multi-criteria cost function is based on the probability density fitting (PDF) and the Knearest neighbors (KNN) algorithm is considered for the AMC stage. This approach is implemented in its linear and nonlinear versions. For the nonlinear case, we use a neural network. Numerical results, in terms of mean square error (MSE) and symbol error rate (SER) with quadrature amplitude modulation (QAM) signals show that our multi users multi-criteria GBE is effective in alleviating the ISI.

show abstract

Blind equalization and Automatic Modulation Classification based on pdf fitting

Cited by 8 publications

References 11 publications

Blind roll-off estimation for digital transmissions

Blind roll-off estimation for digital transmissions

AMC2N: Automatic Modulation Classification Using Feature Clustering-Based Two-Lane Capsule Networks

Multi-user generic channel equalization and sources separation for MIMO communication systems

Contact Info

Product

Resources

About