Annals of Telecommunications最新文献

Border Gateway Protocol (BGP) is increasingly becoming a multipurpose protocol. However, it keeps suffering from security issues such as bogus announcements for malicious goals. Some of these security breaches are especially critical for distributed intrusion detection systems that use BGP as the underlay network for interchanging alarms. In this sense, assessing the confidence level of detection alarms transported via BGP messages is critical to prevent internal attacks. Most of the proposals addressing the confidence level of detection alarms rely on complex and time-consuming mechanisms that can also be a potential target for further attacks. In this paper, we propose an out-of-band system based on machine learning to infer the confidence level of BGP messages, using just the mandatory fields of the header. Tests using two different data sets, (i) from the indirect effects of a widespread worm attack and (ii) using up-to-date data from the IPTraf Project, show promising results, considering well-known performance metrics, such as recall, accuracy, receiver operating characteristics (ROC), and f1-score.

 In the present work, a modified Costas loop is presented with the help of mathematical modeling and numerical simulation. The voltage-controlled oscillator output phase along with frequency is controlled using the input control voltage. The modified loop is tested as frequency demodulator circuit where the improvement in sideband attenuation is clearly visible using an additional phase control arrangement. Numerical simulation result leads to a similar conclusion when the ratio of third harmonic to first harmonic and the ratio of first sideband attenuation to carrier are obtained for different proportions of the phase control. Noise bandwidth and lock range of the modified loop are investigated with special emphasis on the dependence of these parameters on the phase modulator gain. Lock range of the loop is evaluated analytically. An excellent demodulation capability of the loop has been reported in the presence of the additional phase control. Analytical results coupled with numerical findings presented are in good agreement.

The growing cybersecurity threats make it essential to use high-quality data to train machine learning (ML) models for network traffic analysis, without noisy or missing data. By selecting the most relevant features for cyber-attack detection, it is possible to improve both the robustness and computational efficiency of the models used in a cybersecurity system. This work presents a feature selection and consensus process that combines multiple methods and applies them to several network datasets. Two different feature sets were selected and were used to train multiple ML models with regular and adversarial training. Finally, an adversarial evasion robustness benchmark was performed to analyze the reliability of the different feature sets and their impact on the susceptibility of the models to adversarial examples. By using an improved dataset with more data diversity, selecting the best time-related features and a more specific feature set, and performing adversarial training, the ML models were able to achieve a better adversarially robust generalization. The robustness of the models was significantly improved without their generalization to regular traffic flows being affected, without increases of false alarms, and without requiring too many computational resources, which enables a reliable detection of suspicious activity and perturbed traffic flows in enterprise computer networks.

With the increasing sophistication and heterogeneity of network infrastructures, the need for Virtual Network Embedding (VNE) is becoming more critical than ever. VNE consists of mapping virtual networks on top of the physical infrastructure to optimize network resource use and improve overall network performance. Considered as one of the most important bricks of network slicing, it has been proven to be an NP-hard problem with no exact solution. Several heuristics and meta-heuristics were proposed to solve it. As heuristics do not provide satisfactory solutions, meta-heuristics allow a good exploration of the solutions’ space, though they require testing several solutions, which is generally unfeasible in a real world environment. Other methods relying on deep reinforcement learning (DRL) and combined with heuristics yield better performance without revealing issues such as sticking at local minima or poor space exploration limits. Nevertheless, these algorithms present varied performances according to the employed approach and the problem to be treated, resulting in robustness problems. To overcome these limits, we propose a robust placement approach based on the Algorithm Selection paradigm. The main idea is to dynamically select the best algorithm from a set of learning strategies regarding reward and sample efficiency at each time step. The proposed strategy acts as a meta-algorithm that brings more robustness to the network since it dynamically selects the best solution for a specific scenario. We propose two selection algorithms. First, we consider an offline selection in which the placement strategies are updated outside the selection period. In the second algorithm, the different agents are updated simultaneously with the selection process, which we call an online selection. Both solutions proved their efficiency and managed to dynamically select the best algorithm regarding acceptance ratio of the deployed services. Besides, the proposed solutions succeed in commuting to the best placement strategy depending on the strategies’ strengths while outperforming all standalone algorithms.

In this study, we examine the performance of higher-order quadrature amplitude modulation (QAM) schemes in a two-way multiple-relay network. This network employs three-phase analog network coding and an opportunistic relay selection algorithm while dealing with imperfect channel state information (CSI) and nonlinear power amplifiers (NLPA). Specifically, we derive lower-bound expressions for general-order rectangular QAM, hexagonal QAM, and cross QAM schemes. We assess performance over Nakagami-m fading channels with integer-valued fading parameters that are independently and non-identically distributed. Our analysis focuses on variable-gain amplify-and-forward relaying combined with maximal ratio combining receivers. To calculate closed-form average symbol error rate (ASER) expressions, we utilize a well-established approach based on cumulative distribution functions. We validate the accuracy of our derived expressions by comparing them to results obtained through Monte Carlo simulations. Furthermore, we investigate how fading parameters, the number of relay nodes, imperfect CSI, and NLPA affect the network’s performance.