Fig 1 - uploaded by Elias Giacoumidis
Content may be subject to copyright.
Conceptual dendrogram for agglomerative and divisive Hierarchical based clustering [19]. Hierarchical clustering is comprised of six steps [20]: 1. Enter the number of targeted clusters, e.g. four for QPSK. 2. Initiate disjoint cluster having zero level (L(0) = 0) and order (m =0). 3. Identify the least unrelated pair of clusters (r, s) w.r.t. D(r,s) = min{d[i,j]} (1) 4. Increase the order by m=m+1 and the clusters r and s into one cluster, creating a new cluster m. The level of such cluster is formed by L(m) = d[r,s] (2) 

Conceptual dendrogram for agglomerative and divisive Hierarchical based clustering [19]. Hierarchical clustering is comprised of six steps [20]: 1. Enter the number of targeted clusters, e.g. four for QPSK. 2. Initiate disjoint cluster having zero level (L(0) = 0) and order (m =0). 3. Identify the least unrelated pair of clusters (r, s) w.r.t. D(r,s) = min{d[i,j]} (1) 4. Increase the order by m=m+1 and the clusters r and s into one cluster, creating a new cluster m. The level of such cluster is formed by L(m) = d[r,s] (2) 

Source publication
Article
Full-text available
Fiber-induced intra- and inter-channel nonlinearities are experimentally tackled using blind nonlinear equalization (NLE) by unsupervised machine learning based clustering (MLC) in $\sim$ 46-Gb/s single-channel and $\sim$ 20-Gb/s (middle-channel) multi-channel coherent multi-carrier signals (OFDM-based). To that end we introduce, for the first time...

Contexts in source publication

Context 1
... isolate n OFDM symbols into multiple effective groups for each subcarrier [18][19][20]. Due to the statistical structure of agglomerative methods, they most commonly characterized by a two-dimensional (2-D) diagram. This diagram is widely identified as dendrological (from the Greek word 'tree'). An illustration of such dendrogram is depicted in Fig. 1, demonstrating the divisions or fusions made at each successive stage of analysis. Hierarchical clustering harnessing agglomerative processing that harvests a number of symbol partitions (P): Pn, Pn-1, …, P1. Where n corresponds to single symbol-based clusters and l to one group encompassing the total n cases. During each step, this ...
Context 2
... of two clusters are performed. Case-2 was inspired by the well-known fact that nonlinear phase noise is accumulated on outer clusters in 16- QAM. Fig. 9 illustrates the adopted clustering designs in which the grouping centroids (denoted with 'x') are also depicted: light-blue on step 1 and white on step 2 for Case-1; black for single-step Case-2. Fig. 10 shows the performance of these designs for FLC (best MLC) on experimental single- channel 16-QAM CO-OFDM at 2000 km. It is shown that both clustering designs have almost identical performance to the conventional clustering approach; except for Case-2 at very high LOPs where up to ~0.3 dB increase in Q-factor is observed, reaching the ...

Similar publications

Article
Full-text available
With an increasing need for secured water supply, a better understanding of the water consumption behavior is beneficial. This can be achieved through end-use classification, i.e., identifying end-uses such as toilets, showers or dishwashers from water consumption data. Previously, both supervised and unsupervised machine learning (ML) techniques a...

Citations

... The most effective ML techniques in terms of performance but with a computational cost are the supervised artificial neural networks (ANNs) and support F o r P e e r R e v i e w 4 vector machines (SVMs) [40][41][42]. Blind ML-based nonlinearity compensators have also shown great potential with reduce computational cost, using unsupervised ML clustering, such as Fuzzy-logic C-means (FL), K-means, Hierarchical clustering [43], Density-Based Spatial Clustering of Applications with Noise (DBSCAN) [44], affinity propagation (AP) [45] and Gaussian mixture [46]. Both supervised and unsupervised ML algorithms have shown benefits over deterministic algorithms due to the fact they can tackle stochastic nonlinear distortions such as the interplay of polarization mode dispersion (PMD) with nonlinearity, transceiver nonlinearities and electro-optical noise. ...
Article
Full-text available
In this survey, we analyze the newest machine learning (ML) techniques applied in modern optical orthogonal frequency division multiplexing (O-OFDM) systems for access, core networks, and multi-channel transmission. From rudimentary to more advanced approaches, ML is proven to be a gold standard technique for signal quality improvement when low transmitter modulation extinction ratio dominates in coherent O-OFDM, and when stochastic-induced nonlinearities are present such as parametric noise amplification in long-haul transmission and the interplay between polarization-mode dispersion and the Kerr-induced nonlinearity. In addition, ML algorithms can effectively tackle determinist nonlinear distortions in O-OFDM networks, as well as inter-subcarrier nonlinear effects (ie, inter-subcarrier four-wave mixing and cross-phase modulation). In essence, ML techniques could be potentially beneficial for any multi-carrier approach (eg, filter bank modulation). The survey illustrates an extensive ML taxonomy for O-OFDM based networks, covering supervised, reinforcement learning and unsupervised ML categories. The transmission performance of various ML-assisted O-OFDM systems is presented taking into account the ML computational complexity toward real-time implementation. We also highlight the strict operating conditions for such systems under which a ML algorithm should perform classification, regression or clustering. Finally, the survey opens research issues and future directions toward ML implementation in radio-over-fiber (RoF) and 5G new radio (NR) systems.
... In this paper, BP neural network is applied to the multi-channel signal acquisition system to achieve accurate prediction of detection data, overcome the influence of nonlinear effect on data detection and reduce the detection limit of the system [24,25]. The detection limit of the detection instrument is inversely correlated with the sensitivity, and is positively correlated with the standard deviation of the sample detection. ...
Article
Full-text available
A multi-channel optical sensing system for heavy metal concentration detection is presented in this paper. The system utilizes a multi-channel optical path combined with a unique circuit design and BP neural network (BP-ANN) to perform the online analysis of multi-wavelength signals, achieving accurate concentration detection of a heavy metal solution. An array photodiode is used to detect the transmission light intensities at multiple wavelengths through the optical path of the system, which enables the collection of useful spectral information of the solution. The system uses a unique signal acquisition method to effectively improve the efficiency of both signal acquisition and operation. BP-ANN is applied to the online analysis of multi-channel information, which overcomes the influential issue of nonlinear effect on data detection, optimizes the anti-interference ability, and lowers the detection limit of the system. This system eliminates the necessary employment of the expensive and large spectrometers and therefore greatly reduces the instrument cost and occupying space. Additionally, the detection limit of the system is extended lower than that of the conventional spectrophotometer. Compared with the detection limits of heavy metal solution obtained by using a single characteristic light wavelength, the detection limits of Cd2+, Cu2+ and Cr6+ achieved through using multi-channel detection system can be reduced by 42.64%, 38.12%, and 20.62%, respectively, and these detection limits are found as 0.0041mg/L, 0.0091mg/L, and 0.0112mg/L, respectively.
... In the last decade, machine learning (ML) has triggered widespread interest in optical network research, such as nonlinearity compensation [13], nonlinear equalizer design [14][15][16], optical channel modeling [17], and OPM. The signal features for ML-based OPM include asynchronous amplitude histograms (AAH) [18][19][20], asynchronous delay-tap sampling portraits (ADTPs) [21,22], eye diagram [23], constellation diagram [24][25][26], auto-covariance function [27], and tunable optical bandpass filter (OBPF) [28,29], etc. ...
... Simulation setup for OPM (Mux: multiplexer, DeMux: demultiplexer, EDFA: erbium-doped optical fiber amplifier, PSR: polarization splitter-rotator, FTSA: Fourier transform spectrum analyzer, ANN: artificial neural networks)From Eq.(15)(16)(17), we can see that the OSNR at the upper and lower path is different from the true OSNR when the polarizations of fiber and device are mismatched. But the average of the two OSNR values is always identical to the true OSNR. ...
Article
Full-text available
A novel optical performance monitoring (OPM) method based on Fourier transform spectrum analysis (FTSA) is designed for optical signal-to-noise ratio (OSNR) monitoring, modulation format and baud rate recognition in the presence of fiber nonlinearities. The interference intensities, which reflect spectral features of signals, are obtained by exploiting the FTSA consisting of two-stage Mach-Zehnder interferometer (MZI) arrays. Then, the mapping between the OPM parameters and modulated interference intensity (MII) is characterized using neural networks without prior knowledge of the configuration of the communication network. Results show that optical performance parameters are monitored simultaneously. Meanwhile, the accuracy of modulation format and baud rate recognition is 94.8% and most (over 86%) OSNR monitoring errors are less than ±1 dB under complex transmission conditions in presence of frequency offset and delay jitter. Besides, the FTSA can be fabricated on a silicon on insulator (SOI) platform with a large fabrication tolerance, and it has broad working bandwidth to support the full optical communication band. Therefore, the proposed OPM method is capable of integration and miniaturization, which can be ubiquitously applied in network intermediate nodes to support the construction of smart optical networks.
... The literature [23] optimizes the application of deep belief networks to large-scale data, as well as proposes an implementation of deep belief networks based on MapReduce. The literature [24] proposes a pre-training optimization strategy for deep confidence networks based on sparse symmetric encoders. The literature [25] combines deep belief networks with deep convolutional neural networks to optimize the accuracy of graph classification. ...
Article
Full-text available
People’s cognition of objectively existing things is developing in the direction of digitization, and more and more objectively existing things are constantly being presented in the form of data. In order to realize the analysis of constantly changing large-scale data and the automatic extraction of valuable information, this paper analyzes the application of artificial intelligence auxiliary equipment based on machine learning in the cognitive learning process, and proposes a collaborative filtering method based on fusion of global and local parameters. Moreover, this paper studies a cognitive computing model based on context-aware data stream to realize effective analysis of context-aware data and obtain effective cognitive results. In addition, this paper proposes a task scheduling algorithm based on improved queue matching to solve the task scheduling problem in a distributed computing environment. Finally, this paper constructs an improved algorithmic cognitive system architecture, and verifies the performance of the system through experimental research. The research results show that the system constructed in this paper is highly reliable.
... Foreign scholars' research on the General Principles of Commercial Law focuses on commercial behavior and merchant's leading legislative model, and explores which legislative model is more suitable for the regulation of commercial transactions. e main basic system research has reached a high level [8]. ...
Article
Full-text available
At present, a series of economic structural changes created by the network economy have brought challenges to the entire economy and society. Traditional social commerce has also suffered severe tests under the background of network economy and global integration, and the rise and development of network commercial activities lack legal constraints. Based on the Big Data technology, in view of the characteristics of data mining services, this paper expands and changes the traditional model and proposes the Big Data alliance data mining service process model. Moreover, this paper uses intelligent decision-making theory and knowledge reasoning methods to construct a fast-responding, service-reusable, and intelligent service model to realize the scalability of data mining services. In addition, this paper constructs system function modules according to the requirements of Internet commercial judicial data mining and verifies the performance of the system constructed in this paper through experimental analysis. The research results show that the system constructed in this paper has certain practical effects.
... Fast clustering is one of the most widely used methods of clustering analysis. is method has a small amount of calculation and rapid classification and can effectively process data with multiple indicators and large samples [8]. In the digital age, information processing problems of large samples are often involved, and the rapid clustering method shows a good application prospect and application value. ...
Article
Full-text available
At present, there is no method to predict or monitor patients with AMI, and there is no specific treatment method. In order to improve the analysis of clinical influencing factors of acute myocardial infarction, based on the machine learning algorithm, this paper uses the K-means algorithm to carry out multifactor analysis and constructs a hybrid model combined with the ART2 network. Moreover, this paper simulates and analyzes the model training process and builds a system structure model based on the KNN algorithm. After constructing the model system, this paper studies the clinical influencing factors of acute myocardial infarction and combines mathematical statistics and factor analysis to carry out statistical analysis of test results. The research results show that the system model constructed in this paper has a certain effect in the clinical analysis of acute myocardial infarction.
... Recently, machine learning (ML) techniques have been explored for dealing with nonlinear equalization in optical communication systems. K-means [18], fuzzy logic C-means [19], hierarchical clustering [19] and Gaussian mixture models (GMM) [20] have been used for constellation points clustering. K-nearest neighbors (KNN) [21] and support vector machines (SVM) [22] have been used as nonlinear classifiers to classify constellation points. ...
... Recently, machine learning (ML) techniques have been explored for dealing with nonlinear equalization in optical communication systems. K-means [18], fuzzy logic C-means [19], hierarchical clustering [19] and Gaussian mixture models (GMM) [20] have been used for constellation points clustering. K-nearest neighbors (KNN) [21] and support vector machines (SVM) [22] have been used as nonlinear classifiers to classify constellation points. ...
Article
Full-text available
We propose a bi-directional gated recurrent unit neural network based nonlinear equalizer (bi-GRU NLE) for coherent optical communication systems. The performance of bi-GRU NLE has been experimentally demonstrated in a 120 Gb/s 64-quadrature amplitude modulation (64-QAM) coherent optical communication system with a transmission distance of 375 km. Experimental results show that the proposed bi-GRU NLE can significantly mitigate nonlinear distortions. The Q-factors can exceed the hard-decision forward error correction (HD-FEC) limit of 8.52 dB with the aid of bi-GRU NLE, when the launched optical power is in the range of -3 dBm to 3 dBm. In addition, when the launched optical power is in the range of 0 dBm to 2 dBm, the Q-factor performances of the bi-GRU NLE and bi-directional long short-term memory neural network based nonlinear equalizer (bi-LSTM NLE) are similar, while the number of parameters of bi-GRU NLE is about 20.2% less than that of bi-LSTM NLE, the average training time of bi-GRU NLE is shorter than that of bi-LSTM NLE, the number of multiplications required for the bi-GRU NLE to equalize per symbol is about 24.5% less than that for bi-LSTM NLE.
... Machine learning uses detailed methods in statistics, evidence theory, neural networks, fuzzy sets, primitive sets, evolutionary computing, and other fields to complete data summarization, concept description, classification rule extraction, grouping analysis, data analysis, analysis and sequence pattern discovery, and other types of work [18,19]. Compared with data mining, machine learning tends to provide different algorithms, while data mining tends to use these algorithms to solve problems. ...
Article
Full-text available
With the development of the globalization of science and technology, innovation has become an important driving force for regional economic development. As a core element of regional innovation, financial R&D resources have also become a key element to enhance national innovation capabilities and national economic competitiveness. National and regional innovation capabilities have a direct impact. There are also many deep-seated problems behind the world-renowned achievements, such as irrational industrial structure, insufficient independent innovation capabilities, low resource utilization efficiency, and the service quality and efficiency of financial institutions for the transformation of total factor productivity. These problems extremely restrict the efficiency upgrade and further development of our country’s total factor productivity. This study uses the DEA-Malmquist index model to measure the efficiency of fiscal R&D resource allocation in 28 provinces and regions in China in the past 10 years and uses Mapinfo12.0 software to analyze regional differences in the efficiency of fiscal R&D resource allocation in China from a spatial perspective. During the year, the overall R&D resource allocation efficiency of 28 provinces and autonomous regions in China has shown an upward trend. The efficiency of fiscal R&D resource allocation and the concentration of financial factors have had a positive impact on total factor productivity, transform and upgrade factors, increase total factor productivity, and provide empirical evidence for building a strong country.
... Under the impact of fiber nonlinearity, other nonlinear equalizers based on inverse Volterra series transfer functions may also be deployed to partially invert the nonlinear distortion induced by the transmission link. However, the Volterra-based nonlinear compensation (NLC) has shown worse performance than an optimized machine learning and their complexity is also high [10], [11]. Recently, a supervised machine-learningbased technique, namely artificial neural network (ANN), has been proposed and studied for uniform 64-QAM as a predistortion compensator for impairments induced by a low resolution DAC, but ignoring other nonlinear effects [12]. ...
Article
Full-text available
In this paper, we experimentally demonstrate the combined benefit of artificial neural network-based nonlinear-ity compensation and probabilistic shaping for the first time. We demonstrate that the scheme not only compensates for transceiver's nonlinearity, enabling the full benefits of shaping to be achieved, but also the combined effects of transceiver and fiber propagation nonlinearities. The performance of the proposed artificial neural network is demonstrated at 28 Gbaud for both 64-QAM and 256-QAM probabilistically shaped systems and compared to that of uniformly distributed constellations. Our experimental results demonstrate: the expected performance gains for shaping alone; an additional SNR performance gain up to 1 dB in the linear region; an additional mutual information gain of 0.2 bits per channel use in the constellation-entropy limited region. In the presence of coupled transceiver and fiber-induced nonlinearities, an additional mutual information enhancement of ∼0.13 bits/symbol is experimentally observed for a fiber link of up to 500 km with the aid of the proposed artificial neural network.
... Actuellement, la majeure partie des travauxévalue l'apport des techniques de classification pour traiter directement la problématique de détection. Précisément, différents classifieurs basés sur l'algorithme des k-moyennes [273], des SVM [118,120,187,121,119] ou des réseaux de neurones [138,118,108,274,149] ontété proposés. La problématique de compensation des non-linéarités via des techniques de régression est un sujet moins couvert dans la littérature. ...
Thesis
Full-text available
Ce mémoire résume mes activi- tés de recherches menées ces 12 dernières années à l’IRDL puis au Lab-STICC. Dans un premier temps, il présente une synthèse des travaux réalisés au sein de l’IRDL sur la période 2008-2018. Ces travaux couvrent essen- tiellement deux problématiques : le diagnos- tic des machines électriques et la surveillance des réseaux électriques (smart-grid ). Pour résoudre ces problématiques, l’originalité de nos travaux réside dans l’exploitation systématique de la structure signaux au moyen de techniques de traitement du signal paramétriques. Dans un second temps, ce mémoire présente les activités de recherches initiées au laboratoire Lab-STICC depuis mon intégration en 2018. Ces activités concernent la concep- tion d’algorithmes pour la compensation des imperfections dans les chaines de communi- cations optiques cohérentes. Après avoir lister nos travaux en cours portant sur l’utilisation des approches paramétriques pour la compensation des imperfections linéaires au sens large (imperfections du laser, déséquilibre IQ et dispersion chromatique), ce mémoire insiste sur le potentiel des approches mixtes paramétriques / machine learning pour compenser conjointement les imperfections linéaires et non-linéaires de la chaine de communication.