Marko Beko

Non terrestrial networks (NTNs), particularly low Earth orbit (LEO) satellite systems, play a vital role in supporting future mission critical applications such as disaster relief. Recent advances in artificial intelligence (AI)-native communications enable LEO satellites to act as intelligent edge nodes capable of on board learning and task oriented inference. However, the limited link budget, coupled with severe path loss and fading, significantly constrains reliable downlink transmission. This paper proposes a deep joint source-channel coding (DJSCC)-based downlink scheme for AI-native LEO networks, optimized for goal-oriented visual inference. In the DJSCC approach, only semantically meaningful features are extracted and transmitted, whereas conventional separate source-channel coding (SSCC) transmits the original image data. To evaluate information freshness and visual event detection performance, this work introduces the age of misclassified information (AoMI) metric and a threshold based AoI analysis that measures the proportion of users meeting application specific timeliness requirements. Simulation results show that the proposed DJSCC scheme provides higher inference accuracy, lower average AoMI, and greater threshold compliance than the conventional SSCC baseline, enabling semantic communication in AI native LEO satellite networks for 6G and beyond.

This letter investigates a two-tier Uncrewed Aerial Vehicle (UAV) assisted Cellular Internet of Things (IoT) architecture inspired by a practical LoRa/NB-IoT deployment, where the fronthaul consists of IoT devices accessing a UAV-mounted LoRa gateway via slotted ALOHA with multi-packet reception (MPR), and the backhaul relays decoded packets to a terrestrial base station (BS) over a coded NB-IoT link. We develop an analytical framework that jointly models fronthaul contention with MPR and coded short-packet transmission over a realistic backhaul channel. The results reveal system configurations that maximize end-to-end throughput while maintaining efficient backhaul utilization across varying UAV–BS distances.

This paper evaluates the performance of a low-complexity receiver for Non-Orthogonal Multiple Access (NOMA) systems, eliminating the need to perform the Successive Interference Cancellation (SIC) operation common in these systems. This research advances the state of the art in four main aspects: It presents a comparison between the Bit Error Rate (BER) performance of a system using a low-complexity decoder and a system with an SIC decoder; it uses a power allocation strategy to optimize BER performance, followed by a sensitivity analysis to evaluate the impact of small variations in power allocations; a detailed analysis of the low-complexity coding technique applied to Rayleigh and Nakagami channels; and the study of the impact of varying Nakagami fading parameters on BER performance for a low-complexity system. The results show that low-complexity decoding under consideration enables a 90% reduction in the memory needed to simulate the system.

Most of the signals widely employed in wireless communications can have significant envelope fluctuations that make them very prone to nonlinear (NL) effects, leading to significant performance degradation when conventional receivers (designed for ideal linear conditions) are utilized. However, if optimum (ML) receivers are employed, (NL) effects do not necessarily lead to performance degradation, and can actually outperform the corresponding linear systems. This paper presents a general framework for studying the impact of (NL) effects on a wide class of block transmission techniques with blockwise pre-processing where the transmitted signals have significant envelope fluctuations. This class includes many of the widely employed transmission techniques like (OFDM), (MIMO), Single Carier-Frequency Domain Equalization (SC-FDE) Code Division Multiple Access (CDMA). Our approach provides accurate bounds on the achievable performance of optimum receivers, and enables the design of iterative receivers able to approach that optimum performance with complexity much lower than the corresponding optimum ML receivers.

This work addresses the problem of autonomous target navigation in indoor environments through wireless sensing. To accomplish accurate navigation, it proposes a novel yet simple localization algorithm based on basic geometry and Weighted Central Mass (WCM) by extracting range measurements from received wireless signals. To avoid obstacle collision in the considered indoor environments, the work proposes a new obstacle detection scheme that is based on wireless sensing, where abrupt signal fluctuations throughout the target's movement are exploited to detect and avoid obstructions. Therefore, integrating the two proposed solutions allows for partially autonomous target navigation in indoor environments without resorting to expensive and complex hardware, such as LiDARs or cameras. The proposed solutions are validated through both simulation and experimental test beds, that corroborate their effectiveness, both in terms of navigation and obstacle detection accuracy.

This work proposes a novel approach to reinforce localization security in wireless networks in the presence of malicious nodes that are able to manipulate (spoof) radio measurements. It substitutes the original measurement model by another one containing an auxiliary variance dilation parameter that disguises corrupted radio links into ones with large noise variances. This allows for relaxing the non-convex maximum likelihood estimator (MLE) into a semidefinite programming (SDP) problem by applying convex-concave programming (CCP) procedure. The proposed SDP solution simultaneously outputs target location and attacker detection estimates, eliminating the need for further application of sophisticated detectors. Numerical results corroborate excellent performance of the proposed method in terms of localization accuracy and show that its detection rates are highly competitive with the state of the art.

This work proposes a novel approach to reinforce localization security in wireless networks in the presence of malicious nodes that are able to manipulate (spoof) radio measurements. It substitutes the original measurement model by another one containing an auxiliary variance dilation parameter that disguises corrupted radio links into ones with large noise variances. This allows for relaxing the non-convex maximum likelihood estimator (MLE) into a semidefinite programming (SDP) problem by applying convex-concave programming (CCP) procedure. The proposed SDP solution simultaneously outputs target location and attacker detection estimates, eliminating the need for further application of sophisticated detectors. Numerical results corroborate excellent performance of the proposed method in terms of localization accuracy and show that its detection rates are highly competitive with the state of the art.