In this paper, we propose Federated Deep Learning (FDL) for intrusion detection in heterogeneous networks. Local Deep Neural Network (DNN) models are used to learn the hierarchical representations of the private network traffic data in multiple edge nodes. A dedicated central server receives the parameters of the local DNN models from the edge nodes, and it aggregates them to produce an FDL model using the Fed+ fusion algorithm. Simulation results show that the FDL model achieved an accuracy of 99.27 ± 0.79%, a precision of 97.03 ± 4.22%, a recall of 98.06 ± 1.72%, an F1 score of 97.50 ± 2.55%, and a False Positive Rate (FPR) of 2.40 ± 2.47%. The classification performance and the generalisation ability of the FDL model are better than those of the local DNN models. The Fed+ algorithm outperformed two state-of-the-art fusion algorithms, namely federated averaging (FedAvg) and Coordinate Median (CM). Therefore, the DNN-Fed+ model is preferable for intrusion detection in heterogeneous wireless networks.

Intelligent reflecting surface (IRS)-aided millimeter-wave (mmWave) multiple-input single-output (MISO) is considered one of the promising techniques in next-generation wireless communication. However, existing beamforming methods for IRS-aided mm Wave MISO systems require high computational power, so it cannot be widely used. In this paper, we combine an unsupervised learning-based fast beamforming method with IRS-aided MISO systems, to significantly reduce the computational complexity of this system. Specifically, a new beamforming design method is proposed by adopting the feature fusion means in unsupervised learning. By designing a specific loss function, the beamforming can be obtained to make the spectrum more efficient, and the complexity is lower than that of the existing algorithms. Simulation results show that the proposed beamforming method can effectively reduce the computational complexity while obtaining relatively good performance results.

Malware traffic classification (MTC) is a key technology for solving anomaly detection and intrusion detection problems. And hence it plays an important role in the field of network security. Traditional MTC methods based on port, payload and statistic depend on the manual-designed features, which have low accuracy. Recently, deep learning methods have attracted significant attention due to their high accuracy in terms of classification. However, in practical application scenarios, deep learning methods require a large amount of labeled samples for training, while the available labeled samples for training are very rare. Furthermore, the preparation of a large amount of labeled samples requires a lot of labor costs. To solve these problems, this paper proposes two methods based on semi-supervised learning (SSL) and transfer learning (TL), respectively. Our proposed methods use a large amount of unlabeled data collected in the Internet traffic, which can greatly improve the accuracy classification with few labeled samples. Through experiments, we obtained the best method to improve the accuracy of few labeled samples in different situations. Experiment results show that our proposed methods can satisfy the requirement of MTC in the case of few labeled samples.

Accurate downlink channel state information (CSI) is one of the essential requirements for harnessing the potential advantages of frequency-division duplexing (FDD) massive multi-input multi-output (MIMO) systems. The current state-of-art in this vibrant research area include the use of deep learning to compress and feedback downlink CSI at the user equipments (UEs). These approaches focus mainly on achieving CSI feedback with high reconstruction performance and low complexity, but at the expense of inflexible compression rate (CR). High training overheads and limited storage capacity requirements are some of the challenges associated with the design of dynamic CR, which instantaneously adapt to propagation environment. This paper applies transfer learning (TL) to develop a multi-rate CSI compression and recovery neural network (TL-MRNet) with reduced training overheads. Simulation results are presented to validate the superiority of the proposed TL-MRNet over traditional methods in terms of normalized mean square error and cosine similarity.

There is not a single country in the world that is so rich that it can remove all level crossings or provide their denivelation in order to absolutely avoid the possibility of accidents at the intersections of railways and road traffic. In the Republic of Serbia alone, the largest number of accidents occur at passive crossings, which make up three-quarters of the total number of crossings. Therefore, it is necessary to constantly find solutions to the problem of priorities when choosing level crossings where it is necessary to raise the level of security, primarily by analyzing the risk and reliability at all level crossings. This paper presents a model that enables this. The calculation of the maximal risk of a level crossing is achieved under the conditions of generating the maximum entropy in the virtual operating mode. The basis of the model is a heterogeneous queuing system. Maximum entropy is based on the mandatory application of an exponential distribution. The system is Markovian and is solved by a standard analytical concept. The basic input parameters for the calculation of the maximal risk are the geometric characteristics of the level crossing and the intensities and structure of the flows of road and railway vehicles. The real risk is based on statistical records of accidents and flow intensities. The exact reliability of the level crossing is calculated from the ratio of real and maximal risk, which enables their further comparison in order to raise the level of safety, and that is the basic idea of this paper.

BACKGROUND Acute ischemic stroke (AIS) frequently results in the development of cognitive impairment, which quite often persists. The pathophysiological mechanisms involved in the development of cognitive impairment are only partially elucidated. The aim of this study was to evaluate the correlation between interleukin 6 (IL-6) and tumor necrosis factor-alpha (TNF-α) serum levels with cognitive impairment in AIS patients. SUBJECTS AND METHODS This hospital-based case-control study was performed during December 2014 - May 2018. A total number of 130 randomly selected patients were prospectively recruited from the Department of Neurology, Clinical Center University of Sarajevo. The study examined 100 first-ever AIS patients, while 30 comprised the non-stroke control group of patients with discogenic lumbosacral radiculopathy. All participants were evaluated using the Mini-Mental State Examination, the Montreal Cognitive Assessment, the Frontal Assessment Battery, and the Addenbrooke's Cognitive Examination-Revised. Cognitive testing and laboratory analyses were performed within the first three days of admission in all patients while AIS patients were reassessed on the 15thday of hospitalization. RESULTS Female stroke patients with cognitive impairment had significantly higher baseline levels of IL-6 (p<0.017), and TNF-α (p<0.017) than those without cognitive impairment. In the control measurement, a significant difference in IL-6 levels (p=0.037) in male and TNF-α levels (p=0.042) in female stroke patients with cognitive impairment was observed. CONCLUSIONS These findings indicate that pro-inflammatory cytokines are probably implicated in the pathogenesis of cognitive decline in AIS patients.