Explore the vast range of titles available.

Next-generation wireless communication networks are expected to offer extremely high data rates supported by very low latency and radically new applications, which require a new wireless radio technology paradigm. However, it is crucial to assist the radio link over the fast varying and highly dynamic channel to satisfy the diverse requirements of next-generation wireless networks. Recently, communication via autonomous electric vertical takeoff and landing (eVTOL) has gained momentum, owing to its potential for cost-effective network deployment. It is considered one of the most promising technologies conceived to support smart radio terminals. However, to provide efficient and reliable communications between ground base stations and eVTOLs as well as between eVTOLs and other eVTOLs, realistic eVTOL channel models are indispensable. In this paper, we propose a nonstationary geometry-based stochastic channel model for eVTOL communication links. The proposed eVTOL channel model framework considers time-domain nonstationarity and arbitrary eVTOL trajectory and is sufficiently general to support versatile C bands. One of the critical challenges for eVTOL is the fast vertical takeoff and landing flight patterns affecting the regular propagation communication channel. Moreover, we present a new method for estimating the SNR over the non-stationary fast dynamic time-variant eVTOL channel by utilizing the sliding window adaptive filtering technique. Furthermore, we present an information–theoretic approach to characterize the end-to-end transmission delay over the eVTOL channel and prove that the optimal transmission scheme strongly depends upon the eVTOL link configuration. In addition, to analyze the occurrence of deep fade regions in eVTOL links, we analyze the outage probability, which is an important performance metric for wireless channels operating over dynamic fading channels, and make an important observation that the outage probability increases non-linearly with the eVTOL height. Furthermore, we consider the commercially available eVTOL specifications and data to validate the channel model and analyze the Doppler shift and latency for the exponential acceleration and exponential deceleration velocities profiles during the takeoff and landing operation. This paper provides a new and practical approach for the design, optimization, and performance evaluation of future eVTOL-assisted next-generation wireless communications.

While the number of things is growing accompanied with an explosive increase in wireless traffic, network providers are facing a set of challenges, especially that the massive number of devices are expected to communicate over the current cellular networks. In an attempt to relieve network congestion and increase throughput, we turn toward cell shrinking and offloading, a key technology in future 5G networks. Using this potential solution, we are mainly targeting two important issues: i) enabling cyber-physical systems (CPS) communications over cellular networks to provide CPS with several benefits such as ubiquitous coverage, global connectivity, reliability and security; and ii) offloading a proportion of CPS traffic to small cells, which in tun increases the throughput of macrocells, and frees more network resources to other users. Using stochastic geometry, we present an analysis on CPS offloading rate and achievable throughput when small cells base stations (SCBSs) are powered by solar energy. The solar energy harvesting allows SCBSs to offset the costs of serving CPS devices. Our results show the potential benefits for both macrocells and small cells in terms of minimum achievable throughput when the CPS offloading rate is high.

5G and beyond cellular networks (NextG) will support the continuous execution of resource-expensive edge-assisted deep learning (DL) tasks. To this end, Radio Access Network (RAN) resources will need to be carefully \"sliced\" to satisfy heterogeneous application requirements while minimizing RAN usage. Existing slicing frameworks treat each DL task as equal and inflexibly define the resources to assign to each task, which leads to sub-optimal performance. In this paper, we propose SEM-O-RAN, the first semantic and flexible slicing framework for NextG Open RANs. Our key intuition is that different DL classifiers can tolerate different levels of image compression, due to the semantic nature of the target classes. Therefore, compression can be semantically applied so that the networking load can be minimized. Moreover, flexibility allows SEM-O-RAN to consider multiple edge allocations leading to the same task-related performance, which significantly improves system-wide performance as more tasks can be allocated. First, we mathematically formulate the Semantic Flexible Edge Slicing Problem (SF-ESP), demonstrate that it is NP-hard, and provide an approximation algorithm to solve it efficiently. Then, we evaluate the performance of SEM-O-RAN through extensive numerical analysis with state-of-the-art multi-object detection (YOLOX) and image segmentation (BiSeNet V2), as well as real-world experiments on the Colosseum testbed. Our results show that SEM-O-RAN improves the number of allocated tasks by up to 169% with respect to the state of the art.

Test Time Adaptation (TTA) has emerged as a practical solution to mitigate the performance degradation of Deep Neural Networks (DNNs) in the presence of corruption/ noise affecting inputs. Existing approaches in TTA continuously adapt the DNN, leading to excessive resource consumption and performance degradation due to accumulation of error stemming from lack of supervision. In this work, we propose Domain-Aware Real-Time Dynamic Adaptation (DARDA) to address such issues. Our key approach is to proactively learn latent representations of some corruption types, each one associated with a sub-network state tailored to correctly classify inputs affected by that corruption. After deployment, DARDA adapts the DNN to previously unseen corruptions in an unsupervised fashion by (i) estimating the latent representation of the ongoing corruption; (ii) selecting the sub-network whose associated corruption is the closest in the latent space to the ongoing corruption; and (iii) adapting DNN state, so that its representation matches the ongoing corruption. This way, DARDA is more resource efficient and can swiftly adapt to new distributions caused by different corruptions without requiring a large variety of input data. Through experiments with two popular mobile edge devices - Raspberry Pi and NVIDIA Jetson Nano - we show that DARDA reduces energy consumption and average cache memory footprint respectively by 1.74x and 2.64x with respect to the state of the art, while increasing the performance by 10.4%, 5.7% and 4.4% on CIFAR-10, CIFAR-100 and TinyImagenet.

Distributed deep neural networks (DNNs) have been shown to reduce the computational burden of mobile devices and decrease the end-to-end inference latency in edge computing scenarios. While distributed DNNs have been studied, to the best of our knowledge, the resilience of distributed DNNs to adversarial action remains an open problem. In this paper, we fill the existing research gap by rigorously analyzing the robustness of distributed DNNs against adversarial action. We cast this problem in the context of information theory and rigorously proved that (i) the compressed latent dimension improves the robustness but also affect task-oriented performance; and (ii) the deeper splitting point enhances the robustness but also increases the computational burden. These two trade-offs provide a novel perspective to design robust distributed DNN. To test our theoretical findings, we perform extensive experimental analysis by considering 6 different DNN architectures, 6 different approaches for distributed DNN and 10 different adversarial attacks using the ImageNet-1K dataset.

Deep neural networks (DNN) usually come with a significant computational burden. While approaches such as structured pruning and mobile-specific DNNs have been proposed, they incur drastic accuracy loss. In this paper we leverage the intrinsic redundancy in latent representations to reduce the computational load with limited loss in performance. We show that semantically similar inputs share many filters, especially in the earlier layers. Thus, semantically similar classes can be clustered to create cluster-specific subgraphs. To this end, we propose a new framework called Semantic Inference (SINF). In short, SINF (i) identifies the semantic cluster the object belongs to using a small additional classifier and (ii) executes the subgraph extracted from the base DNN related to that semantic cluster for inference. To extract each cluster-specific subgraph, we propose a new approach named Discriminative Capability Score (DCS) that finds the subgraph with the capability to discriminate among the members of a specific semantic cluster. DCS is independent from SINF and can be applied to any DNN. We benchmark the performance of DCS on the VGG16, VGG19, and ResNet50 DNNs trained on the CIFAR100 dataset against 6 state-of-the-art pruning approaches. Our results show that (i) SINF reduces the inference time of VGG19, VGG16, and ResNet50 respectively by up to 35%, 29% and 15% with only 0.17%, 3.75%, and 6.75% accuracy loss (ii) DCS achieves respectively up to 3.65%, 4.25%, and 2.36% better accuracy with VGG16, VGG19, and ResNet50 with respect to existing discriminative scores (iii) when used as a pruning criterion, DCS achieves up to 8.13% accuracy gain with 5.82% less parameters than the existing state of the art work published at ICLR 2023 (iv) when considering per-cluster accuracy, SINF performs on average 5.73%, 8.38% and 6.36% better than the base VGG16, VGG19, and ResNet50.

As emerging networks such as Open Radio Access Networks (O-RAN) and 5G continue to grow, the demand for various services with different requirements is increasing. Network slicing has emerged as a potential solution to address the different service requirements. However, managing network slices while maintaining quality of services (QoS) in dynamic environments is a challenging task. Utilizing machine learning (ML) approaches for optimal control of dynamic networks can enhance network performance by preventing Service Level Agreement (SLA) violations. This is critical for dependable decision-making and satisfying the needs of emerging networks. Although RL-based control methods are effective for real-time monitoring and controlling network QoS, generalization is necessary to improve decision-making reliability. This paper introduces an innovative attention-based deep RL (ADRL) technique that leverages the O-RAN disaggregated modules and distributed agent cooperation to achieve better performance through effective information extraction and implementing generalization. The proposed method introduces a value-attention network between distributed agents to enable reliable and optimal decision-making. Simulation results demonstrate significant improvements in network performance compared to other DRL baseline methods.

Although recent advances in wireless technologies have enabled high rate communication in the dynamic wireless air channel, these technologies are not effective at communication through metallic enclosures due to Faraday shielding. Existing solutions for wireless communications through metallic enclosures have achieved one way communication through metallic barriers at achievable data rates under 20 Mbps, which is relatively low compared to the rates achieved in the wireless air channel. The first portion of this thesis presents a low-rate ultrasonic throughwall communication system which allows for simultaneous two-way data transmission through metallic barriers. A frequency tracking algorithm is also presented which allows the system to adapt to changing channel conditions. Such a system could enable the wireless configuration of sensors and monitoring of data in hermetically sealed enclosures and other environments where conventional wireless techniques are ineffective. The next portion, and majority of this thesis investigates the use of methods to achieve higher data transmission rates using ultrasonic signalling techniques on frequency selective acoustic-electric channels. The nature of the acoustic-electric channel is discussed and compared to the characteristics of other communication media including the wireless air channel as well as wired channels, highlighting some similarities and notable differences. The use of rate maximization techniques such as bit-loading and power allocation in a multicarrier modulation signalling scheme are investigated. Orthogonal frequency division multiplexing (OFDM) is employed which achieves high spectral efficiency in frequency selective channels. Next, the use of multiple-input multiple-output (MIMO) techniques is explored as a method of further increasing the achievable data transmission rates in ultrasonic channels. Co-channel interference (crosstalk) is significant in the MIMO acoustic-electric channel and, without the use of crosstalk mitigation techniques, the aggregate theoretical capacity of multiple closely-spaced channels will produce marginal capacity performance increases or decreases compared with that of the single channel alone, depending on the average signal-to-noise (SNR) level. Several crosstalk mitigation structures are then investigated including the zero forcing receiver, eigenmode transmission, and the minimum mean-square-error (MMSE) receiver, and their theoretical capacity performances are compared for specific multichannel configurations including a two channel system and seven channel system. With the use of crosstalk mitigation techniques, the aggregate multichannel capacity approximately scales with the number of channels used, with capacity performance exceeding 1 Gbps for the seven-channel configuration at high average SNR levels. Once the theoretical capacity performance results are obtained for multiple channels under various configurations, a similar investigation is conducted to see what rates are achievable using various rate maximization techniques such as bit-loading and power allocation. The throughput performances achieved by employing these rate maximization techniques are then compared with each other and to the multichannel theoretical capacity performances. Results are also presented which show the performance of each crosstalk mitigation technique to be effective under vary degrees of transducer misalignment. The results of this thesis suggest that using a slowly adaptive crosstalk mitigation structure having minimal complexity may be most appropriate in quasi-static MIMO acoustic-electric channel arrays. The MMSE receiver, which is implemented using the least mean square (LMS) algorithm in a decision directed mode, may be used while achieving similar data transmission rates to those achieved by more complex structures such as eigenmode transmission which require feedback.

Channel Autoencoders (CAEs) have shown significant potential in optimizing the physical layer of a wireless communication system for a specific channel through joint end-to-end training. However, the practical implementation of CAEs faces several challenges, particularly in realistic and dynamic scenarios. Channels in communication systems are dynamic and change with time. Still, most proposed CAE designs assume stationary scenarios, meaning they are trained and tested for only one channel realization without regard for the dynamic nature of wireless communication systems. Moreover, conventional CAEs are designed based on the assumption of having access to a large number of pilot signals, which act as training samples in the context of CAEs. However, in real-world applications, it is not feasible for a CAE operating in real-time to acquire large amounts of training samples for each new channel realization. Hence, the CAE has to be deployable in few-shot learning scenarios where only limited training samples are available. Furthermore, most proposed conventional CAEs lack fast adaptability to new channel realizations, which becomes more pronounced when dealing with a limited number of pilots. To address these challenges, this paper proposes the Online Meta Learning channel AE (OML-CAE) framework for few-shot CAE scenarios with dynamic channels. The OML-CAE framework enhances adaptability to varying channel conditions in an online manner, allowing for dynamic adjustments in response to evolving communication scenarios. Moreover, it can adapt to new channel conditions using only a few pilots, drastically increasing pilot efficiency and making the CAE design feasible in realistic scenarios.

Thanks to the ubiquitousness of Wi-Fi access points and devices, Wi-Fi sensing enables transformative applications in remote health care, security, and surveillance. Existing work has explored the usage of machine learning on channel state information (CSI) computed from Wi-Fi packets to classify events of interest. However, most of these algorithms require a significant amount of data collection, as well as extensive computational power for additional CSI feature extraction. Moreover, the majority of these models suffer from poor accuracy when tested in a new/untrained environment. In this paper, we propose ReWiS, a novel framework for robust and environment-independent Wi-Fi sensing. The key innovation of ReWiS is to leverage few-shot learning (FSL) as the inference engine, which (i) reduces the need for extensive data collection and application-specific feature extraction; (ii) can rapidly generalize to new tasks by leveraging only a few new samples. We prototype ReWiS using off-the-shelf Wi-Fi equipment and showcase its performance by considering a compelling use case of human activity recognition. Thus, we perform an extensive data collection campaign in three different propagation environments with two human subjects. We evaluate the impact of each diversity component on the performance and compare ReWiS with a traditional convolutional neural network (CNN) approach. Experimental results show that ReWiS improves the performance by about 40% with respect to existing single-antenna low-resolution approaches. Moreover, when compared to a CNN-based approach, ReWiS shows a 35% more accuracy and less than 10% drop in accuracy when tested in different environments, while the CNN drops by more than 45%.