Semantic communication (SemCom) has been recently deemed a promising next-generation wireless technique to enable efficient spectrum savings and information exchanges, thus naturally introducing a novel and practical network paradigm where cellular and device-to-device (D2D) SemCom approaches coexist. Nevertheless, the involved wireless resource management becomes complicated and challenging due to the unique semantic performance measurements and energy-consuming semantic coding mechanism. To this end, this paper jointly investigates power control and spectrum reuse problems for energy-efficient D2D SemCom cellular networks. Concretely, we first model the user preference-aware semantic triplet transmission and leverage a novel metric of semantic value to identify the semantic information importance conveyed in SemCom. Then, we define the additional power consumption from semantic encoding in conjunction with basic power amplifier dissipation to derive the overall system energy efficiency (semantic-value/Joule). Next, we formulate an energy efficiency maximization problem for joint power and spectrum allocation subject to several SemCom-related and practical constraints. Afterward, we propose an optimal resource management solution by employing the fractional-to-subtractive problem transformation and decomposition while developing a three-stage method with theoretical analysis of its optimality guarantee and computational complexity. Numerical results demonstrate the adequate performance superiority of our proposed solution compared with different benchmarks.

Massive multiple input multiple output (MIMO) systems are integral to next-generation wireless technologies due to their ability to meet the growing demands of throughput and support a plethora of applications. An efficient operation of massive MIMO requires accurate channel state information (CSI). In a frequency division duplex (FDD) MIMO system, the base station can rely on CSI feedback that user equipment (UE) estimates from downlink CSI from orthogonal pilot sequences. Recently, artificial intelligence (AI), i.e., deep learning approaches, have been introduced to compress and reconstruct CSI matrices at UE and the base station, respectively. However, these existing approaches still rely on channel estimation at the UE side, which introduces additional errors in the autoencoder design. To address these issues, we propose to implement the autoencoder that processes the pilot sequences directly to avoid excessive processing errors. Moreover, a higher compression can be achieved due to the lower error. Evaluation results demonstrate that the proposed scheme can significantly reduce the communication overhead by using a higher compression ratio while maintaining high CSI reconstruction performance in addition to lower bit error rates compared to the existing deep learning approach. © 2024 IEEE.

Federated learning (FL) has emerged as a distributed machine learning (ML) technique that can protect local data privacy for participating clients and improve system efficiency. Instead of sharing raw data, FL exchanges intermediate learning parameters, such as gradients, among clients. This article presents an efficient wireless communication approach tailored for FL parameter transmission, especially for Internet of Things (IoT) devices, to facilitate model aggregation. Our study considers practical wireless channels that can lead to random bit errors, substantially affecting FL performance. Motivated by empirical gradient value distribution, we introduce a novel received bit masking method that confines received gradient values within prescribed limits. Moreover, given the intrinsic error resilience of ML gradients, our approach enables the delivery of approximate gradient values with errors without resorting to extensive error correction coding or retransmission. This strategy reduces computational overhead at both the transmitter and the receiver and minimizes communication latency. Consequently, our scheme is particularly well-suited for resource-constrained IoT devices. Our simulations demonstrate that our proposed scheme can effectively mitigate random bit errors in FL performance, achieving similar learning objectives but with the 50% air time required by existing methods involving error correction and retransmission. © 2014 IEEE.

The accurate modeling of indoor radio propagation is crucial for localization, monitoring, and device coordination, yet remains a formidable challenge, due to the complex nature of indoor environments where radio can propagate along hundreds of paths. These paths are resulted from the room layout, furniture, appliances and even small objects like a glass cup. They are also influenced by the object material and surface roughness. Advanced machine learning (ML) techniques have the potential to take such non-linear and hard-to-model factors into consideration. However, extensive and fine-grained datasets are urgently required. This paper presents WiSegRT11https://github.com/SunLab-UGA/WiSegRT, an open-source dataset for indoor radio propagation modeling. Generated by a differentiable ray tracer within the segmented 3-dimensional (3D) indoor environments, WiSegRT provides site-specific channel impulse responses for each grid point relative to the given transmitter location. We expect WiSegRT to support a wide-range of applications, such as ML-based channel prediction, accurate indoor localization, radio-based object detection, wireless digital twin, and more. © 2024 IEEE.

With the rising user demands to support multiple applications, the next-generation wireless communications exploit coexistence of licensed and unlicensed spectrum to much improve the spectrum efficiency. However, spectrum sharing is a crucial yet currently challenging element in enabling the coexistence of diverse technologies and functionalities in next-generation wireless network systems. While cognitive radio (CR) is one such prominent technology to identify the unused portions of the spectrum, it usually requires extra process or separate hardware for sensing. In this work we propose a novel approach to integrate spectrum sensing with the existing channel estimation process. In particular, we assume that channel estimation is implemented constantly for licensed spectrum usage for user equipment (UE). Based on channel reciprocity, adjacent unlicensed spectrum will be sensed for occupancy and also estimated if unoccupied simultaneously. The integration of spectrum sensing and channel estimation offers multiple benefits such as accurate resource allocation, reduced latency and processing load, faster decision making, and adaptable to real-time scenarios. Moreover, the proposed scheme can be applicable to most communication systems, e.g., the fourth-generation mobile network and onwards without extra hardware implementation. Evaluation results based on the open-source dataset is included to demonstrate the proposed concept and scheme. © 2024 IEEE.

As the deployment of 5G technology matures, the anticipation for 6G is growing, which promises to deliver faster and more reliable wireless connections via cutting-edge radio technologies. A pivot to these radio technologies is the effective management of large-scale antenna arrays, which aims to construct valid spatial streams to maximize system throughput. Traditional management methods predominantly rely on user feedback to adapt to dynamic wireless channels. However, a more promising approach lies in the prediction of spatial channel state information (spatial-CSI), which is a channel characterization that consists of all robust line-of-sight (LoS) and non-line-of-sight (NLoS) paths between the transmitter (Tx) and receiver (Rx), with three-dimensional (3D) trajectory, attenuation, phase shift, delay, and polarization of each path. Recent advances in hardware and neural networks make it possible to predict such spatial-CSI using precise environmental information, and further explores the possibility of holographic communication, which implies complete control over every aspect of the radio waves. This paper presents a preliminary exploration of using generative artificial intelligence (AI) to accurately model the environment particularly for radio simulations and identify valid paths within it for real-time spatial-CSI prediction. Our validation project demonstrates promising results, highlighting the potential of this approach in driving forward the evolution of 6G wireless communication technologies. IEEE

This paper presents an approximate wireless communication scheme for federated learning (FL) model aggregation in the uplink transmission. We consider a realistic channel that reveals bit errors during FL model exchange in wireless networks. Our study demonstrates that random bit errors during model transmission can significantly affect FL performance. To overcome this challenge, we propose an approximate communication scheme based on the mathematical and statistical proof that machine learning (ML) model gradients are bounded under certain constraints. This bound enables us to introduce a novel encoding scheme for float-to-binary representation of gradient values and their QAM constellation mapping. Besides, since FL gradients are error-resilient, the proposed scheme simply delivers gradients with errors when the channel quality is satisfactory, eliminating extensive error-correcting codes and/or retransmission. The direct benefits include less overhead and lower latency. The proposed scheme is well-suited for resource-constrained devices in wireless networks. Through simulations, we show that the proposed scheme is effective in reducing the impact of bit errors on FL performance and saves at least half the time than transmission with error correction and retransmission to achieve the same learning performance. In addition, we investigated the effectiveness of bit protection mechanisms in high-order modulation when gray coding is employed and found that this approach considerably enhances learning performance. © 2023 Owner/Author.

Mobile edge computing (MEC) has emerged as a crucial paradigm for enhancing the capabilities in machine-type communication, which relies on sufficient spectrum resources to efficiently process the offloading tasks. However, the rise in the number of IoT devices and the expansion of machine-type communication traffic can cause spectrum shortages. Non-orthogonal multiple access (NOMA) allows multiple users to share the same bandwidth simultaneously, which can be applied to improve spectrum efficiency. Furthermore, the growing connectivity of devices and transmission of sensitive data in IoT networks give rise to significant concerns regarding both security threats and energy efficiency. In this paper, a NOMA-enabled MEC system is studied. In the presence of an eavesdropper, the secrecy rate is further adopted to measure the security performance of offloading. We aim to maximize secrecy computation efficiency in a NOMA-enabled MEC system. The simulation results demonstrate the improvement of secrecy computation efficiency by applying NOMA in mobile edge computing systems. © 2023 IEEE.

Multiple-input multiple-outputs (MIMO) systems are integral to the implementation of the current fifth-generation (5G) and beyond wireless networks. Accurate channel state information (CSI) is imperative to a MIMO system for its optimal performance. In this work, we develop an end-to-end software evaluation platform for the channel estimation process in a MIMO system. With this platform, different channel estimation and reconstruction processes, as well as precoding methods can be implemented and evaluated. Channel reconstruction error and transmission bit error rate are chosen metrics in the current implementation. Direct channel estimation with the least square method, and CSI feedback methods with compressive sensing and deep-learning approaches are tested to demonstrate the evaluation platform. © 2023 IEEE.

Automated Internet of Things (IoT) devices generate a considerable amount of data continuously. However, an IoT network can be vulnerable to botnet attacks, where a group of IoT devices can be infected by malware and form a botnet. Recently, Artificial Intelligence (AI) algorithms have been introduced to detect and resist such botnet attacks in IoT networks. However, most of the existing Deep Learning-based algorithms are designed and implemented in a centralized manner. Therefore, these approaches can be sub-optimal in detecting zero-day botnet attacks against a group of IoT devices. Besides, a centralized AI approach requires sharing of data traces from the IoT devices for training purposes, which jeopardizes user privacy. To tackle these issues in this paper, we propose a federated learning based framework for a zero-day botnet attack detection model, where a new aggregation algorithm for the IoT devices is developed so that a better model aggregation can be achieved without compromising user privacy. Evaluations are conducted on an open dataset, i.e., the N-BaIoT. The evaluation results demonstrate that the proposed learning framework with the new aggregation algorithm outperforms the existing baseline aggregation algorithms in federated learning for zero-day botnet attack detection in IoT networks. © 2023 IEEE.