Network slicing is a technology that virtualizes a single infrastructure into multiple logical networks (called slices) where resources or virtualized functions can be flexibly configured by demands of applications to satisfy their quality of service (QoS) requirements. Generally, to provide the guaranteed QoS in applications, resources of slices are isolated. In wired networks, this resource isolation is enabled by allocating dedicated data bandwidths to slices. However, in wireless networks, resource isolation may be challenging because the interference between links affects the actual bandwidths of slices and degrades their QoS. In this paper, we propose a slice management scheme that mitigates the interference imposed on each slice according to their priorities by determining routes of flows with a different routing policy. Traffic flows in the slice with the highest priority are routed into shortest paths. In each lower-priority slice, the routing of traffic flows is conducted while minimizing a weighted summation of interference to other slices. Since higher-priority slices have higher interference weights, they receive lower interference from other slices. As a result, the QoS of slices is differentiated according to their priorities while the interference imposed on slices is reduced. We compared the proposed slice management scheme with a naïve slice management (NSM) method that differentiates QoS among slices by priority queuing. We conducted some simulations and the simulation results show that our proposed management scheme not only differentiates the QoS of slices according to their priorities but also enhances the average throughput and delay performance of slices remarkably compared to that of the NSM method. The simulations were conducted in grid network topologies with 16 and 100 nodes and a random network topology with 200 nodes. Simulation results indicate that the proposed slice management increased the average throughput of slices up to 6%, 13%, and 7% and reduced the average delay of slices up to 14%, 15%, and 11% in comparison with the NSM method.

The next generation of mobile networks is predicted to deliver high data speeds, lower latency, and increase the spectral and energy efficiency of wireless communication systems. Several technologies are being investigated for usage in 5G networks. Massive multiple-input multiple-output (mMIMO) systems are one of the most promising technologies for enabling 5G. Even after the recent advancements and research, numerous challenges still exist for channel estimation for mMIMO systems. In the context of pilot contamination and feedback overhead, this study tracks the most recent developments in research on mMIMO system difficulties. The primary goals of this study are to identify the problems with channel estimation, provide a summary of the cutting-edge solutions suggested in the literature, and then discuss newly emerging open research issues that must be taken into account for the implementation of beyond-5G networks.

This study explored the impact of confusion on concentration and cognition, emphasizing the importance of detecting and preventing confusion from enhancing learning outcomes. By leveraging electroencephalogram (EEG) data, we proposed a novel deep learning model that uses long short-term memory (LSTM) networks to predict confusion levels in online massive open courses (MOOCs). LSTM’s ability to model sequential data such as EEG signals has been harnessed to capture long-term dependencies and temporal dynamics effectively. To enhance pattern detection, we incorporated probabilistic features from machine learning (ML) models. By training them on the same dataset, we utilized their predictions as additional features for the deep learning model. Thereby, the neural network could make more informed decisions and improve its ability to detect and analyze EEG data patterns. Using LSTM and probabilistic features, our model effectively captured temporal dependencies, enabling an accurate online assessment of student perplexity to identify moments of confusion. The integration of attention mechanisms further enhanced the focus on critical EEG features, providing valuable insights into students’ cognitive states during online learning. We evaluated our approach by comparing the deep-learning model trained on the original dataset with that trained on feature-engineered data using K-fold cross-validation. Preliminary testing showed that the proposed DNN + LSTM model, which incorporates probabilistic features and a custom attention layer, achieves high accuracy in identifying moments of confusion among MOOC students. This study advances the EEG data analysis, leading to a better understanding of confusion patterns and supports personalized interventions for online education platforms.

Reconfigurable Intelligent Surface (RIS) can offer a customizable wireless transmission and is regarded as an incredibly crucial enabling technology when used as reflectors for current wireless base stations (BSs) to overcome the blockage challenges of millimeter wave (mmWave) wireless communications systems. RISs are instrumental in driving the progress of sixth-generation (6G) wireless services and beyond, with the goal of achieving gigabit-per-second (Gbps) data rates for networks operating in the mmWave frequency bands. RIS has the significant potential to diminish the effects of signal blockages and unnecessary handovers because precise phase adjustment of RIS elements enhances the scattering environments and creates multiple reflective signal paths. However, the large number of RIS elements can complicate the optimization of base station (BS) and RIS reflector configurations and result in performance degradation. This paper introduces a deep reinforcement learning strategy to dynamically configure the numerous RIS elements for multi-user downlink transmissions. The proposed method employs the twin-delayed deep deterministic policy gradient (TD3) method to solve non-convex optimization issues where the base station receives state information from the RIS, including feedback on user channel states. For real-world systems requiring collaborative control over phase shift and beamforming matrix, the BS determines the optimal actions, which comprise the allocation of transmission power and phase shift adjustments within a Nakagami-m fading environment. The experimental results indicate that the proposed TD3-based solution outperforms other existing benchmarks.

In the UAV network with unstable channel conditions compared to the stationary network, delay-tolerant data transmission from a source node to a destination node is important. To make data be successfully received at the destination within a certain time threshold, the same data can be transmitted through multiple routing paths to increase path diversity. However, an increase in path diversity also increases the power consumption of relay UAVs. In this letter, we propose a Lyapunov-based energy-efficient path diversity method for data transmissions in UAV networks. In the proposed method, the delay robustness according to the path diversity is calculated, and then the trade-off relation between path diversity and power consumption is derived based on Lyapunov drift-plus-penalty approach. By minimizing the derived drift-plus-penalty term, the proposed multi-path transmission method provides delay-tolerant data transmissions with acceptable power consumption in UAV networks.

Machine learning techniques have attracted considerable attention for wireless networks because of their impressive performance in complicated scenarios and usefulness in various applications. However, training with and sharing raw data obtained locally from each wireless node does not guarantee privacy and requires a large communication overhead. To mitigate such issues, federated learning (FL), in which sharing parameters for model updates are shared instead of raw data, has been developed. FL has also been studied using blockchain techniques to efficiently perform learning in distributed wireless systems without having to deploy a centralized server. Although blockchain-based decentralized federated learning (BDFL) is a promising technique for various wireless sensor networks, malicious attacks can still occur, which result in performance degradation or malfunction. In this study, we analyze the impact of a jamming threats from malicious miners to BDFL in wireless networks. In a wireless BDFL system, it is possible for malicious miners with jamming capability to interfere with the collection of model parameters by normal miners, thus preventing the victim miner from generating a global model. By disrupting normal miners participating in BDFL systems, malicious miners with jamming capability can more easily add malicious data to the mainstream. Through various simulations, we evaluated the success probability performance of malicious block insertion and the participation rate of normal miners in a wireless BDFL system.