Model-based clustering is the most widely used statistical clustering method, in which heterogeneous data are divided into homogeneous groups using inference based on mixture models. The presence of measurement error in the data can reduce the quality of clustering and, for example, cause overfitting and produce spurious clusters. To solve this problem, model-based clustering assuming a normal distribution for measurement errors has been introduced. However, too large or too small (outlier) values ​​of measurement errors cause poor performance of existing clustering methods. To tackle this problem {and build a stable model against the presence of outlier measurement errors in the data}, in this article, a symmetric $alpha$-stable distribution is proposed as a replacement for the normal distribution for measurement errors, and the model parameters are estimated using the EM algorithm and numerical methods. Through simulation and real data analysis, the new model is compared with the MCLUST-based model, considering cases with and without measurement errors, and the performance of the proposed model for data clustering in the presence of various outlier measurement errors is shown.

In synthetic aperture radar (SAR) imagery, ship-sea contrast can be significantly improved when the polarimetric information is used, compared with information from a single channel SAR. The constant false alarm rate (CFAR) detection algorithm based on the alpha-stable (AS) distribution model is a descent method for ship detection in sea. The most important step in this method is the parameter estimation of the AS probability density function for which the method of log-cumulants (MoLC) parameter estimation is proposed and utilized in this study. Evaluation results of the proposed method on two L- and C-band polarimetric SAR (PolSAR) datasets which are acquired by AIRSAR and RADARSAT-2 sensors confirm the accuracy of the method. Experimental results show that the best results of detection maps are achieved when the polarimetric span and entropy data are used where a high detection accuracy in the latter cases are along with a low number of false alarms.

The most widely used model in small area estimation is the area level or the Fay-Herriot model. In this model, it is typically assumed that both the area level random effects (model errors) and the sampling errors have a Gaussian distribution. However, considerable variations in error components (model errors and sampling errors) can cause poor performance in small area estimation. In this paper, to overcome this problem, the symmetric ,-stable distribution is used to deal with outliers in the error components. The model parameters are estimated with the empirical Bayes method. The performance of the proposed model is investigated in different simulation scenarios and compared with the existing classic and robust empirical Bayes methods. The proposed model can improve estimation results, in particular when both error components are normal or have heavy-tailed distribution.

One of the obvious reasons for most disorders in network service provisioning is network path congestion. Congestion avoidance in today's networks is too costly and sometimes impossible. With the introduction of SDN, centralizing the equipment's control plane has become possible. This paper presents an enhanced method named ESV-DBRA to avoid congestion in multi-tenant SDN networks. At first, ESV-DBRA monitors the traffic load and delay of all network paths for each tenant individually. Then, by merging the parameters obtained from the monitoring, the Service Level Agreements (SLA), and a novel proposed cost function, it calculates the cost of the network paths per tenant. As a result, traffic for each tenant is routed through the path/paths at the lowest possible cost from the tenant's perspective. Next, the bandwidth quotas will be calculated and assigned to the tenants over their optimal routes. Afterward, whenever congestion is likely to occur in a path, ESV-DBRA automatically changes the route or bandwidth of the tenants' traffic related to this path to avoid congestion. Related algorithms are also proposed.Eventually, simulations show that the proposed method effectively increases bandwidth utilization by 10.76%.

Different types of contact, including contact between node pairs, any-contact of nodes, and contacts of the entire network, are used to characterize social relations in mobile social networks. Different modes of routing, from the point of view of message delivery semantics, encompass unicasting, multicasting, any-casting, and broadcasting. Studies have shown that using probability distribution functions of contact data, which is mainly assumed to be homogeneous for nodes, improves the performance of these networks. However, there exists an important challenge in studies on distributions. A lot of works apply the distribution of one type of contact to other types. Hence in routing applications, it causes to use of the distribution of one type of contact for any mode of routing. This study provides a complete solution to model each type of homogeneous contact data distribution and to use them in different modes of routing. We propose a routing algorithm that uses this new model. Results show that our solution improves the average latency of comparing methods Epidemic, TCCB, and DR about 3.5-times, 30%, and 45%, respectively. It achieves a delivery rate of about 5% and 6%, and average latency about 6% and 8% better than that of DR and TCCB, respectively.

In this paper, the detector of a moving target in MIMO radar with widely separated antenna in the presence of the a-stable clutter is proposed. Due to the complexity of the unknown parameter estimation in the presence of the target, the Rao test is applied. The a-stable model is a general model that can be applied to different environments with appropriate selection of parameters. The problem with this model is inexistence of a closed form for the probability density function of the distribution. Fortunately, there are models that approximate the distribution by a closed form expression. In the present paper, mixing of Gaussian and Cauchy model is used for expressing clutter. Finally, the performance of the proposed detector is compared with the optimal Gaussian and MiniMax detector by a graph of the probability of detection versus signal-to-clutter ratio. The simulation results, show that the performance of the proposed detector is better than the optimal Gaussian and MiniMax detector.

Monitoring and evaluation of transformers are essential to prevent their insulation failure. In this paper, the use of 2-furan-carboxaldehyde (2-FAL) and methanol (MeOH) concentrations as the main products of paper insulation degradation, and insulation condition markers, has been studied. In order to study the degradation process and production of degradation products, the thermal aging process of the transformer insulation system was implemented in laboratory conditions. The results of laboratory studies show that in the early stages of degradation the amount of MeOH is significant compared to 2-FAL. Also, the estimation of the degree of polymerization (DP) in the early stages of degradation (DP>800) through MeOH concentration and with decreasing DP (DP <800) through 2-FAL concentration is closer to the real value. The results of studies performed on 35 distribution transformers confirm the production of significant amounts of MeOH in the early stages of degradation. Also, the estimated DP values for the studied transformers were obtained through 2-FAL and MeOH concentration. The results show that estimating the amount of DP through MeOH concentration is associated with a probability of error of about 9% compared to estimating DP through 2-FAL concentration.