To achieve reliable transmission in visible light communications (VLC), this paper investigates a new delayed asymmetrically-clipped-optical orthogonal-frequency-division-multiplexing (D-ACO-OFDM) system with protographbased low-density parity-check (P-LDPC) codes. Specifically, we first incorporate the delayed bit-interleaved coded modulation (DBICM) structure into ACO-OFDM systems. Subsequently, based on the reliability differences among labeling bits, we propose a novel constrained delay allocation (CDA) scheme, which effectively optimizes the allocation of delayed time slots for labeling bits, thereby accelerating the update of soft information. Furthermore, a bit-level adaptive mapping is integrated into the proposed delay allocation framework, establishing an enhanced CDA (E-CDA) scheme. We also construct a family of partitioned-protograph-based LDPC (PP-LDPC) codes tailored for the E-CDA-aided D-ACO-OFDM system. In addition, a modified weighted density evolution (MWDE) algorithm is employed to analyze the convergence performance of the proposed PP-LDPC-coded D-ACO-OFDM system with the E-CDA scheme. Numerical results illustrate that our proposed design achieves noticeable performance gains compared with the existing counterparts and demonstrate that it is a competent scheme to meet the high-reliability requirements for VLC applications.
To tackle the challenges of high spectrum demand from massive Internet of Things (IoT) communications, this paper proposes a uncrewed aerial vehicle-assisted cognitive radio (CR) non-orthogonal multiple access (NOMA) transmission scheme. Specifically, a secondary aerial IoT base station concurrently acts as a relay to assist the transmission of primary network with favorable aerial-ground channels, and in return, IoT devices of the secondary network can utilize the primary network’s spectrum resources for data transmission. To reduce the signaling overhead of the secondary network, we adopt a hybrid long-short-packet transmission strategy, with primary users using long-packet transmission and secondary IoT devices utilizing short-packet transmission. Inspired by the notion of CR-NOMA, we propose two power allocation algorithms to respectively avoid affecting the quality-of-service (QoS) and age-of-information (AoI) of the primary network. For rigorous performance evaluation, we derive the theoretical expressions of outage probability, block error rate, and AoI performance for both power allocation algorithms. Moreover, for analytical tractability, we derive the high signal-to-noise-ratio (SNR) approximation expressions for the peak AoI (PAoI) of secondary users (SUs), by which the transmission blocklength of the SUs is optimized to minimize the PAoI performances. Finally, we present simulation results to validate the correctness of the theoretical expressions. The results further demonstrate that the SUs of the AoI-based approach achieves superior performance in the low SNR region, whereas the QoS-based approach is more favorable in the moderate SNR region.
This paper investigates the diversity-multiplexing tradeoff (DMT) for two-user hybrid automatic repeat request with incremental redundancy (HARQ-IR) assisted non-orthogonal multiple access (NOMA) systems employing variable-power (VP) transmission. Since the exact outage probability of the VP-HARQ-IR-NOMA system is analytically intractable, we first derive the diversity order, revealing it to be a piecewise function of the first decoded user’s transmission rate. Building upon this, we derive the asymptotic outage probabilities for both users in the high signal-to-noise ratio (SNR) regime to characterize the diversity gain. Furthermore, to provide deeper insights into performance, we derive the effective multiplexing gain using the DMT analytical framework, highlighting the gains of the proposed system over traditional orthogonal multiple access (OMA) schemes. The DMT analysis also confirms a degradation in diversity performance compared to OMA-based counterparts. Finally, numerical results are provided to validate the analytical derivations.
Unmanned aerial vehicle (UAV)-assisted cooperative perception enables connected and autonomous vehicles (CAVs) to exploit bird’s-eye views (BEVs) to alleviate occlusions and extend perception coverage in complex traffic environments. However, hybrid heterogeneity across platforms and sensors, together with resource constraints and potentially unreliable nodes, causes cross-view feature mismatch and latency-induced misalignment, which require joint consideration of communication and perception. To address these challenges, we propose a latency-aware hybrid-heterogeneous UAV-assisted cooperative perception framework (LHU-CP) with adaptive BEV grid size and bandwidth control. LHU-CP maps hybrid-heterogeneous observations into a unified BEV space and captures the coupling between communication and perception through a co-design architecture. Moreover, a latency-aware joint BEV-bandwidth adaptation (LJBA) module jointly optimizes BEV grid size and communication bandwidth to mitigate latency-induced misalignment under resource constraints. In addition, a reputation-based feature gating (RFG) module jointly considers detection and regression consistency to suppress perception- and communication-faulty nodes. Finally, experiments on an extended dataset demonstrate that LHU-CP improves AP@0.5 by 2.08% and AP@0.7 by 3.79% over the best-performing baselines while maintaining real-time efficiency and robustness in dynamic environments.
This paper investigates a pinching-antenna system (PASS)-assisted coordinated multi-point (CoMP) transmission scheme with non-orthogonal multiple access (NOMA). In particular, two access points (APs) each having a single waveguide serve multiple users via CoMP-NOMA. Users are grouped into CoMP-NOMA clusters, each comprising one CoMP user to be jointly served by both APs and two non-CoMP users to be individually served by each AP. Both single-cluster and multiple-cluster cases are considered. 1) For the single-cluster scenario, a system sum-rate maximization problem is formulated to jointly optimize pinching antenna (PA) positions and power allocation. To address this problem, a two-stage algorithm is developed. PA positions are optimized to regulate phase differences between adjacent PAs, so that signals at intended users are coherently aligned, while the signal at the non-intended user is out of phase. Subsequently, transmit power is optimized using successive convex approximation (SCA). 2) For the multi-cluster scenario, the corresponding system sum-rate maximization problem is formulated to jointly optimize user clustering, power allocation and PA positions. A block coordinate descent (BCD)-based algorithm is then developed, in which the user clustering subproblem is solved using matching theory, the power allocation subproblem is handled via SCA, and the PA position optimization subproblem is addressed using gradient-driven SCA. Simulation results demonstrate that: (i) PASS achieves significant performance gains compared with conventional antenna systems; and (ii) by optimizing the positions of PAs on each AP, we can simultaneously achieve phase alignment for its associated non-CoMP user and the CoMP user, while signals received by the other AP’s non-CoMP user remain in phase reversal. This design not only facilitates CoMP transmission, but also enables efficient NOMA-based interference management.
