Optical reconfigurable intelligent reflecting surface (ORIS), as a new type of programmable optical communication equipment, can reconstruct the optical channel environment and expand the application scenarios of optical wireless communications (OWC), which have attracted widespread attention. However, with the increasing demand for OWC scale in B5G and even 6G scenarios, the coverage range of a single ORIS (below 300 m) can no longer meet communication needs. Cascaded ORISs’ technology has become an inevitable trend in the development of ORIS, which has not been deeply studied in existing work. To address the above challenges, in this paper we set up cascaded ORISs in a typical long-distance free space optcis (FSO) scenario and conduct a detailed analysis of their channels and performance. This system employs cascaded ORISs to extend ORIS coverage and expand the overall scale of the FSO system, thereby enabling the FSO link to evolve from a traditional point-to-point configuration to a broader, spatially distributed communication layout. The physical model of multiple ORISs’ cascades and the effect on their communication performance are analyzed in detail. Meanwhile, the closed-form expressions of the bit error rate (BER) and outage probability of the multi-hop cascaded ORISs-assisted FSO system are derived and verified by simulations. Based on the theoretical results and simulation results, the influence of each parameter on the system performance is carefully analyzed, which provides guidance for the design of the actual system.

The vision of 6G communication demands autonomous and resilient networking in environments without fixed infrastructure. Yet most multi-agent reinforcement learning (MARL) approaches focus on isolated stages—exploration, relay formation, or access—under static deployments and centralized control, limiting adaptability. We propose the AI Agent Access (A3) Network, a unified, embodied intelligence-driven framework that transforms multi-agent networking into a dynamic, decentralized, and end-to-end system. Unlike prior schemes, the A3Network integrates exploration, target user access, and backhaul maintenance within a single learning process, while supporting on-demand agent addition during runtime. Its decentralized policies ensure that even a single agent can operate independently with limited observations, while coordinated agents achieve scalable, communication-optimized coverage. By embedding link-level communication metrics into actor–critic learning, the A3Network couples topology formation with robust decision-making. Numerical simulations demonstrate that the A3Network not only balances exploration and communication efficiency but also delivers system-level adaptability absent in existing MARL frameworks, offering a new paradigm for 6G multi-agent networks.

Wireless power transfer enhanced mobile edge computing (WPT-MEC) has emerged as a key technology to support low-latency and energy-efficient computation in wireless networks. With increasing network density, multi-base-station architectures emerge where wireless devices (WDs) offload tasks to distributed base stations (BSs), creating challenges in maintaining quality-of-service fairness during complex resource coordination in multi-BS WPT-MEC networks. To address these challenges, we investigate a non-orthogonal multiple access (NOMA)-enhanced WPT-MEC network comprising multiple WDs and BSs with finite computational capacities. For ensuring fairness, we formulate a max-min problem to maximize the minimum task computation amount by jointly optimizing offloading decisions, NOMA decoding orders, offloading powers and time resource allocation, which results in a challenging mixed integer, sequence and nonlinear programming (MISNLP). To tackle this problem, we propose a two-stage distributed multi-agent algorithm. In the first stage, each BS agent generates offloading preferences based on partial observations, guiding WDs' offloading decisions. In the second stage, given these offloading decisions, we develop an efficient convex-based algorithm to solve the per-BS resource allocation subproblem, jointly optimizing NOMA decoding order, offloading powers and time resource allocation. For effective training, we leverage off-policy training and the centralized training with decentralized execution (CTDE) paradigm with two key innovations: (1) a convex-based critic that evaluates the joint action without bias, and (2) a counterfactual baseline that isolates individual agent credit assignment. The proposed C3MA algorithm achieves six times faster convergence and at least 20% performance improvement when serving more than 20 WDs, compared with existing multi-agent schemes, while maintaining a near-optimal Jain's fairness index of 0.97. Moreover, it sustains an ultra-low execution delay below 5 milliseconds even with 40 WDs, confirming its efficiency and scalability.

Optical reconfigurable intelligent reflecting surface (ORIS), as a new type of programmable optical communication device, can deflect, split, and shape incident light, thereby intelligently reconstructing the optical wireless communication (OWC) environment. The application of ORIS has effectively expanded the application scenarios of OWC and improved system performance and robustness, which has attracted widespread attention. However, the existing ORIS is limited by its basic structure, and has problems such as low stability and poor signal quality in long-distance transmission. Aiming at the pain points of the existing ORIS, this paper proposes an optical reconfigurable distributed sources and reflecting surface (ORDSS) for distributed OWC. By arranging distributed sources in partial areas of the metasurface, users’ performance optimization is achieved based on power compensation and beam adaptive optimization algorithms. In this work, we performed physical modeling of an ORDSS-assisted OWC system, incorporating factors such as the characteristics and spatial distribution of ORDSS sources and reflection units, surface jitter of the ORDSS, pointing errors, and atmospheric attenuation. We derived the expression for the receiving performance of the ORDSS-assisted OWC system and compared it with that of conventional ORIS-assisted OWC systems to evaluate the performance gains brought by ORDSS. Building on the physical model and performance analysis, we investigated an adaptive power control algorithm for the distributed sources in ORDSS. In addition, we explored power compensation mechanisms based on ORDSS distributed sources to enhance system performance in mobile communication scenarios.

Angle of arrival (AOA) estimation with tilted photodiodes (PDs) is fast, accurate and low-cost for visible light positioning (VLP) systems. However, the most popular component assembly technology, surface-mount technology (SMT), is not able to orient the PDs towards their desired tilting angles, making it challenging to manufacture the AOA estimators massively and efficiently. In this work, we propose a novel AOA estimator based on an array of horizontally placed PDs with different lenses, where the incident light angles can be estimated by the ratios of the received optical signal intensities. Such an AOA estimator can be manufactured with the SMT lines in factories and we derive closed-form expressions for the involved AOA estimation algorithm. The proposed scheme provides an SMTcompatible solution for PD-based AOA estimation, and lays theoretical foundations for the error analysis.

The rise of smart applications in wireless devices increasingly relies on mobile edge computing (MEC), where longterm system energy efficiency holds crucial significance for both green computing and application vendors. This paper focuses on long-term energy efficiency in a wireless power transferenabled MEC system. This system faces the challenges of timevarying channel states and stochastic task arrivals. We first formulate this problem to simultaneously optimize offloading, power transfer duration, and energy consumption, while ensuring device queue stability. We then introduce a novel algorithm based on Lyapunov-guided deep reinforcement learning, referred to as LyCNN-DRL. This approach efficiently handles the mixed integer non-linear programming problem by transforming it into a deterministic per-slot problem for online optimization, without needing prior knowledge of future conditions. Specifically, we tackle the problem by dividing it into resource allocation and binary offloading components, applying a convolutional neural network model for near-optimal offloading decisions, and obtaining the optimal solution for resource allocation. Simulation results show that LyCNN-DRL outperforms baseline algorithms, stabilizing MEC network task queues. Furthermore, we quantitatively derive the trade-off between energy efficiency and queue length, represented as$[O(1/V), O(V)]$with the variable$V$.

The heading direction of an unmanned aerial vehicle (UAV) is determined by the yaw angle sensors, which are traditionally known as the inertial measurement units (IMU). IMUs have the problem of error accumulation when estimating the yaw angle, and thus magnetometers and data fusion algorithms can be embedded to counter the accumulated error. However, when the magnetometer is absent or fails in a strong electromagnetic interference environment, the estimated yaw angle will drift, and this could cause UAV collision or crash especially when the speed is high. This paper proposes to use photodiode arrays as angle of arrival (AOA) estimators to aid yaw angle estimation. By fusing the IMU data with the AOA data using the maximum likelihood estimation (MLE), the accumulated yaw angle error can be corrected. Simulations and experiments verify that the yaw angle samples obtained by the MLE are more accurate for a long period of time.

ABSTRACT The robustness against timing offsets of filter bank multi‐carrier (FBMC) is appealing for grant‐free massive access scenarios that mainly adopt asynchronous transmissions. In this work, we propose a compressed sensing based algorithm for joint active user detection as well as timing offset and channel estimation in uplink communication under the combination of FBMC and grant‐free massive access systems, which is critical for subsequent decoding or other processes at receiver. The channel estimation part is based on generalized approximate massage passing (GAMP). The active user detection and timing offset estimation are based on loopy belief propagation (LBP) rules, where the expressions of message passing and belief distributions are derived. Besides, since the receiver may have no prior knowledge about some parameters such as noise variance and activity probability, we introduce the expectation maximization (EM) approach into the proposed algorithm. Moreover, we develop a preamble design method to improve the detection and estimation performance. Simulation results show that the proposed EM‐LBP‐GAMP algorithm can achieve satisfying performance in terms of missed activity detection probability, timing offset estimation error and normalized mean square error of channel estimation.

Light-emitting diode (LED) based visible light positioning (VLP) has been widely studied to provide high-accuracy and low-cost localization services. Existing single-LED VLP systems suffer from severe energy attenuation when the receiver is far from the LED. Even though multiple LEDs could be deployed, the fusion of their positioning information remains an open problem. In this paper, we propose a multi-LED indoor VLP system with LED signal selection or combining. The receiver contains several differently-oriented photodiodes (PDs) working cooperatively as an angle-of-arrival estimator, and each LED individually provides the receiver with a localization result. With these single-LED positioning results, we select the results with the minimum localization error, or combine the results according to their distributions. Closed-form positioning error expressions are derived for the proposed methods. Moreover, increasing the LED number is shown to reduce the positioning error, but the reduction is negligible when the LED number is sufficiently large. Finally, an experimental testbench is built to verify the feasibility of the proposed methods, and a millimeter-level accuracy has been achieved, which is better than that of the classic trilateration method.

Mobile-edge computing (MEC) and wireless power transfer (WPT) are pivotal for enhancing computational power and battery life in 5G/6G networks. However, their performance declines in remote or disaster-stricken areas due to the lack of access points and energy sources. This paper proposes a wirelessly powered unmanned aerial vehicle enabled MEC (UAV-MEC) system to address this issue, focusing on nodes with ignorable computing capabilities and randomly arriving, size-varying tasks. We aim to maximize the long-term average computation rate under constraints such as UAV coverage, time resources, energy, and task causality, formulating a non-convex problem with dynamic states and complex actions. To solve this problem, we introduce an exploration-enhanced deep reinforcement learning (EDRL) algorithm with a bi-layered structure: the main problem determines the UAV’s flying actions, while the sub-problem allocates time resources given these actions. EDRL employs a deep neural network to analyze real-time UAV positions and task demands, determining optimal flight paths. Upon path determination, an efficient algorithm utilizing bisection and golden section search methods allocates WPT and computational offloading durations. Simulations reveal that EDRL achieves an execution latency of just 11.5 ms in thirty-node networks, outperforming baseline DRL algorithms and predetermined trajectory schemes by 20% and 25% in long-term computation rates, respectively. These results highlight EDRL’s effectiveness and low computational complexity, making it a robust solution for challenging environments.

Non-orthogonal multiple access (NOMA) emerges as a superior technology for enhancing spectral efficiency, reducing latency, and improving connectivity compared to orthogonal multiple access. In NOMA networks, successive interference cancellation (SIC) plays a crucial role in decoding user signals sequentially. The challenge lies in the joint optimization of SIC ordering and power allocation, a task complicated by the factorial nature of ordering combinations. This study introduces an innovative solution, the Attention-based SIC Ordering and Power Allocation (ASOPA) framework, targeting an uplink NOMA network with dynamic SIC ordering. ASOPA aims to maximize weighted proportional fairness by employing deep reinforcement learning, strategically decomposing the problem into two manageable subproblems: SIC ordering optimization and optimal power allocation. We use an attention-based neural network to process real-time channel gains and user weights, determining the SIC decoding order for each user. A baseline network, serving as a mimic model, aids in the reinforcement learning process. Once the SIC ordering is established, the power allocation subproblem transforms into a convex optimization problem, enabling efficient calculation of optimal transmit power for all users. Extensive simulations validate ASOPA’s efficacy, demonstrating a performance closely paralleling the exhaustive method, with over 97% confidence in normalized network utility. Compared to the current state-of-the-art implementation, i.e., Tabu search, ASOPA achieves over 97.5% network utility of Tabu search. Furthermore, ASOPA has two orders of magnitude less execution latency than Tabu search when$N=10$and even three orders magnitude less execution latency less than Tabu search when$N=20$. Notably, ASOPA maintains a low execution latency of approximately 50 milliseconds in a ten-user NOMA network, aligning with static SIC ordering algorithms. Furthermore, ASOPA demonstrates superior performance over baseline algorithms besides Tabu search in various NOMA network configurations, including scenarios with imperfect channel state information, multiple base stations, and multiple-antenna setups. These results underscore the robustness and effectiveness of ASOPA, demonstrating its ability to ability to achieve good performance across various NOMA network environments.

Unmanned Aerial Vehicle (UAV)-enabled wirelessly powered Mobile Edge Computing (MEC) is emerging as a powerful technology for boosting computational capability and energy supplementation in Internet of Things (IoT). This work addresses the long-term task completion ratio maximization problem in UAV-enabled wirelessly powered MEC systems. Besides the large number of optimization parameters, the environment can only be partially observed as the UAVs cannot cover the whole network area. Then, it is very challenging to obtain good solutions due to the lack of global information. We introduce a novel distributed Multi-Agent Deep Reinforcement Learning (MADRL) framework for optimizing UAVs’ actions and resource allocation, considering the constraints of tasks that vary in size, arrival times, and required computation completion time. To decouple the complicated parameters, we divide the problem into two manageable subproblems—UAVs’ action decision and resource allocation under a given UAV’s action. We employ a distributed Deep Reinforcement Learning (DRL) scheme for the former subproblem to cope with the partially observable nature. By revealing some important properties of the later subproblem, we design an efficient two-stage optimal algorithm to minimize the total consumed energy of nodes while maximizing the task-completing number. Extensive simulations validate the effectiveness of the proposed framework, achieving over a 50% improvement in task completion ratio compared to baseline schemes in some scenarios.

This paper addresses long-term energy efficiency in a wireless power transfer-enabled mobile-edge computing (MEC) system, facing challenges from time-varying channels and stochastic task arrivals. We formulate the problem to optimize offloading, power transfer duration, and energy consumption while ensuring queue stability. We propose a novel Lyapunov-guided deep reinforcement learning (LyCNN-DRL) algorithm to efficiently solve the long-term mixed integer non-linear programming problem without prior knowledge of future conditions. The approach decomposes the problem into resource allocation and binary offloading components, using a convolutional neural network for near-optimal offloading decisions and the Lagrange dual function for optimal resource allocation. Extensive simulations show that LyCNN-DRL outperforms benchmark algorithms in energy efficiency and latency, achieving over 97% of the optimal utility while reducing execution latency to approximately 50 milliseconds in ten-WD networks. Additionally, we derive the trade-off between energy efficiency and queue length as [O(1/V),O(V)], where V is the Lyapunov control parameter.

Deep neural network (DNN) has emerged as a superior technology in mobile applications. However, due to the computation-intensive nature of DNN’s execution, it is challenging for mobile devices with limited computational power and energy to deploy them and meet real-time requirements. This paper proposes an adaptive edge-device collaborative (AEC) framework for DNN inference optimization. Different from previous works, our approach dynamically optimizes model segmentation and feature compression based on the current environmental state, ensuring that accuracy requirements are met. Specifically, the problem is divided into two subproblems: segmentation and feature compression. A policy-based deep reinforcement learning (DRL) model determines the optimal segmentation layer, while an attention-based compression algorithm maximizes feature compression. We conducted the experiments in an edge computing scenario consisting of an NVIDIA Jetson Nano and an edge server equipped with RTX 3090. Experiments show that the scheme significantly optimizes DNN inference in time-varying network environments. The proposed adaptive dynamic segmentation outperforms the local scheme, edge-only scheme, and the static split computing scheme, which can effectively reduce the latency and energy consumption while satisfying the accuracy requirements with varying network environments. Based on the measured results, the inference performance of AEC for different network environments is improved by more than 30% on average compared to all the baseline algorithms in this paper.

In this paper, we propose an innovative camera-screen communication system designed for real-time communication between 60Hz commercial screens and smartphone cameras. Firstly, we employ Color Modulation to encode information into a video, achieving flicker-free transmission imperceptible to human eyes. Subsequently, we utilize a consecutive characteristics of this modulation method and employ the Random Sample Consensus algorithm for the detection of Regions of Interest without screen detection.In addition, we apply least-squares method to detect the refresh stripe imposed on image samples, which is a severe interference in camera-screen communications. Finally, inter-frame differencing is used for decoding. Our experimental results highlight the robustness and utility of the proposed system, even in complex environments. Specifically, the system achieves up to 80% accuracy at a range of 3 meters.

Higher positioning rate means safer automatic driving, more authentic virtual reality experience, more stable drones, and more efficient robotics in intelligent factories. To achieve this goal, low-complexity algorithm and low-latency hardware are indispensable. In this paper, we propose a novel 3D visible light positioning algorithm based on angle of arrival (AOA). We apply two light-emitting diodes as transmitters and a photodiode array as a receiver. The algorithm can express, in closed form, the 3D coordinates and the azimuth of the receiver with respect to the AOAs, enabling the algorithm to have the lowest complexity. Furthermore, We derive the asymptotic probability distribution functions of the estimated coordinates, making it possible to evaluate the error pattern at different positions. Finally, we design and implement the positioning system with a field programmable gate array and achieve an ultra high positioning speed of over $\mathbf{8 k}$ samples per second with the average positioning error less than $\mathbf{6 m}$.

Massive access is recognized as one of the main use cases of future wireless networks. The characteristic of sporadic transmission in massive access makes the processes of channel estimation (CE) and active user detection (AUD) essential prerequisites for successful data decoding. In this article, we study joint CE and AUD in massive access system with cell-free structure. Specifically, we first establish the expectation maximization approximate message passing (EM-AMP) framework tailored for cell-free structure as a benchmark. Then, we investigate three new methods that exploit coarse user location information in different ways, namely, the variance bounding method, the variance fusion method, and the proposed EM on location method, where the last two methods can also generate finer location estimation as byproduct to CE and AUD at the cost of higher complexity. For single user scenario, we theoretically prove the optimality of the proposed method in channel variance estimation, validating the foundation of the proposed method. For multiuser scenario, we conduct various simulations to compare the performance of different methods. Our findings illustrate that harnessing coarse user location information yields substantial enhancements in CE and AUD performance. Moreover, the proposed method exhibits superior localization accuracy compared to the variance fusion method, all while maintaining comparable complexity, making it a good candidate for applications with both communication and sensing requirements.

Multipath causes interferences to the optical receivers, and reduces the accuracy of visible light positioning (VLP). The existing works mainly focused on the simulation analysis of the signal responses incurred by the scattering of the indoor surfaces, or the positioning error caused by the multipath in the VLP systems. However, such works can only provide insights for how the multipath propagation affects the VLP systems, but cannot separate and exploit the multipath components. In this paper, we design a mode-mixed photodiode (PD) array to receive the overlapping signals of two light sources. The light sources can be light-emitting diodes (LEDs) or reflectors. Based on the diversity of the PDs, we propose an algorithm to separate the signals from the two light sources, which can estimate the directions and the powers of the light sources. Based on the analytical and simulation results, we show that the two light sources can be separated at the receiver without consuming extra time or frequency resources. Moreover, we can distinguish the line-of-sight component even though the received signal has been distorted by interferences from another light source or a reflecting wall, and improve the localization accuracy of the PD array.

Visible light positioning (VLP) is endowed with high accuracy in indoor scenarios. However, the positioning algorithms require plenty of beacon light-emitting diode (LED) coordinates stored in databases, which are expensive to obtain by manual measurements. To circumvent such laborious efforts, we propose a two-step automatic scheme for simultaneous VLP and LED database construction. Specifically, in the first step, a receiver with a photodiode (PD) array samples the optical signals from few benchmark LEDs to locate itself. In the second step, the receiver estimates the unknown beacon LED coordinates through its own locations and the beacon LED signals. For the proposed two-step scheme, we derive closed-form error expressions for the beacon LED coordinates to evaluate the benchmark LEDs’ arrangement and the sampling places. Simulation results agree with the analytical error expressions and reveal that the proposed scheme can achieve centimeter-level accuracy with reasonable transmit powers. Experimental results from the hardware platform verify the feasibility of the scheme. The proposed scheme can circumvent laborious manual measurements and allow the LED database to “grow” while the receivers wander and more receivers enter.

Reconfigurable intelligent surface (RIS)-assisted spatial index modulation (SIM) is a promising candidate for future wireless communication systems due to its high energy and spectrum efficiencies. When only one transmit antenna and one receive antenna are active in the RIS-assisted SIM, we develop a phase-alignment algorithm based on the signature constellation, which utilizes the signature constellation technology to achieve a good bit error rate (BER) performance with a low complexity. Furthermore, in the general scenario, we propose a novel optimization algorithm for RIS reflection coefficients, called maximum Euclidean distance-based low BER (MED-LBER) algorithm. The proposed algorithm aims to maximize the Euclidean distance between different received signals to improve the system performance. The achievable rate, BER performances, and computational complexities of the proposed algorithms are analyzed in detail. The impacts of imperfect channels on the performance of the proposed algorithms are studied. Extensive simulations are carried out to compare the performance of the proposed algorithms with existing algorithms, and demonstrate their superior BER performance.

Light-emitting diodes (LEDs) are widely used in visible light communication (VLC) systems, but their bandwidth is generally incomparable to the laser diodes (LDs). Though LDs are exploited in free space optical systems to provide high data rate, such systems generally fail for mobile users due to the challenges in acquisition, tracking and pointing. Recently, various centimeter-level visible light positioning (VLP) systems have been realized, making it possible to realize VLP-assisted communication systems. In this work, we design a novel VLP-assisted indoor laser communication system, where an LED provides the positioning signal for the mobile user and a narrow-beam laser carries high-speed data stream. The estimated position is updated regularly to continuously orient the laser boresight to the user. The error of the VLP system and the outage probability of the VLP-assisted laser communication system are both derived in closed forms. According to the analytical and simulation results, the outage probability is related to the changes of the system parameters, such as the transmission power of the laser and the user position. Moreover, the outage probability can be minimized by optimizing the laser’s beam divergence angle. An experimental platform is built to verify the feasibility of our proposed system.

Energy efficiency (EE) is currently one of the primary concerns in wireless communication systems. In future 6G systems, achieving high energy-efficient communication with T-bit transmission rate requirements will heavily depend on resource and power parameters. This paper proposes a novel energy efficiency model that incorporates wireless resources and power for a multi-user multiple-input-multiple-output (MIMO) wireless communication system. Firstly, we analyze the equivalency between frequency and spatial dimensions and introduce a two-dimensional resource domain. Additionally, we propose an energy efficiency model incorporating the wireless resource and power. Using the model, we analyze the interaction between the two parameters aiming at the maximize energy efficiency and present an energy-efficient algorithm. Furthermore, We propose two methods for judging high energy efficiency communication of each user based on the model. Finally, our numerical results illustrate the performance of energy efficiency for each user and the advantages and disadvantages of each measurement method in the multi-user system.

With the continuous improvement of the communication frequency band, optical wireless communication (OWC) has attracted extensive attention. Due to the short wavelength and high directivity of optical signals, OWC has high requirements for precise positioning and beam alignment, which additionally increases the positioning and computing burden of the base station. In order to solve this pain point problem, this paper proposes an optical integrated sensing and communication system based on combination of optical intelligent reflecting surface (OIRS) and photodiode (PD) array. This system enables the integration of sensing and communication functionalities with a single transmission, a single device, and ultimately a single network infrastructure, which saves a lot of resources and improves system performance. Based on the OWC channel model and the OIRS physical model, we deduce the probability distribution function (PDF) of channel fading and the closed-form expression of system performance, which shows the influence of various parameters on performance. Simulations have been performed to verify the accuracy of the derived results. Based on theoretical results and simulation results, we make suggestions for system design.

This work studies the unsourced random access (URA) problem via random dictionary learning over Rayleigh block-fading channels with multiple receive antennas. We propose a novel pilot-free URA equipped with a concatenated URA coding structure enabled by on-off patterns. The inherent collision problem in URA is mitigated with slotted encoding structure. Other problems of state-of-the-arts such as low code rate and inefficient transmission are well-tackled. Distinctively, activity detection is not only accomplished in the original pilot-free manner but also enhanced with softer verdict. Facing the potential false alarm (FA) error, a joint channel pruning and dictionary learning procedure maintains FA under a low level. The learning process updates the noise variance of pseudo noise with explicit models to obtain channel estimation. Subsequently, for the first time, this work realizes joint on-off pattern detection and data restoration for on-off division multiple access (ODMA) under multiple-input and multiple-output (MIMO). Also, a successive interference cancellation (SIC) structure is adopted as an iterative approach to produce desirable performance. Moreover, we analyze the approximation of error ceiling and the performance lower-bound determined by feasible deterministic parameters. Finally, numerical simulations with fair comparisons validate the viability of the proposed URA scheme.

Massive machine-type communications (mMTC) or massive access is a critical scenario in the fifth generation (5G) and the future cellular network. With the surging density of devices from millions to billions, unique pilot allocation becomes inapplicable in the user ID-incorporated grant-free random access protocol. Unsourced random access (URA) manifests itself by focusing only on unwrapping the received signals via a common codebook. In this paper, we propose a URA protocol for a massive access cellular system with multi-user single input multiple output (MIMO). The proposed scheme encompasses a codebook enabling construction of sparse transmission frame, a receiver equipped with dictionary learning and error-correcting codes and a collision resolution strategy for the collided codeword. Discrepant to the existing schemes with necessary overhead for preamble signals, no overhead or pre-defined pilot sequences are needed in the proposed scheme, which is favorable for energy-efficient transmission and latency reduction. Numerical results verify the viability of the proposed scheme in practical massive access scenario.

In this paper, a novel random scattering transceiver design for unsourced random access (URA) is investigated. Distinctive from the state-of-the-art such as coded compressed sensing (CCS) kind and interleaving division multiple access (IDMA) kind, the data is split into two portions and the back-and-forth interleaving/de-interleaving for soft information update is dismantled. A portion of the data is encoded by compressed sensing (CS) and the rest is encoded by convolutional code LDPC (CC-LDPC). Based on the principle of probabilistic data association (PDA), a receiver with a turbo PDA multi-user detector (MUD) is designed by the random scattering transmission. Meanwhile, the proposed turbo PDA detector can also cooperate with the CC-LDPC decoder. A sliding window decoding structure is embedded in the proposed receiver. Moreover, given that permeable collision in URA can undermine the system performance, this paper treats collision as interference and elaborates different slot-wise MUD signal models combining the feature of random scattering, based on which a collision resolution procedure is proposed. Empirically, the proposed scheme shows faster system performance convergence and more accurate MUD performance than the IDMA kind. Numerical results with fair comparisons validate the viability of the proposed scheme.

The coordinates of beacon light-emitting diodes (LEDs) are assumed as known prior information in most visible light positioning (VLP) systems. However, such an assumption may be unpractical because the structures of the buildings change unpredictably due to the material aging or the changes of temperature and humidity. Therefore, a periodical update of the LEDs’ coordinates is necessary for accurate VLP. When the number of beacon LEDs increases, the manual measurement becomes too time-consuming and laborious. In this paper, we propose a novel LED coordinates estimator exploiting arbitrary number of angle-of-arrival (AOA) estimators. Each AOA estimator contains several differently oriented photodiodes (PDs) to detect the incident light direction. Considering the PDs’ thermal noises, a closed-form error expression is derived for the proposed LED coordinates estimator. Moreover, AOA estimator selection methods are proposed to detect the AOA estimators of a low signal-to-noise ratio (SNR) and improve the positioning accuracy of the LED coordinates. Analytical and simulation results show centimeter-level positioning accuracy, and verify that the selection methods can reduce the positioning error.

Massive multiple input multiple output (MIMO) antennas can generate narrow and high-gain beams to overcome the severe propagation loss and accurately locate the users. However, even with MIMO, the space resources provided by the basestations (BSs) are still limited. This motivates the study of the collaboration of multiple BSs with massive MIMO antennas, which has the potential to efficiently exploit the space resources and reduce the inter-user interference. In this paper, we study the joint beam allocation of multiple BSs based on the positioning information. We assume that several users are randomly located in a circular area, and one or two BSs are available to serve the users. If two users are in the same beam of a BS, the signals from the BS to these users would interference with each other. To avoid the inter-user interference, we propose the beam allocation scheme for the users based on their position information. Further, we derive the closed-form probability of no inter-user interference for one or two BSs. Simulation and analytical results show that the joint beam allocation of two BSs can reduce the inter-user interferences compared with just exploiting one BS.

Visible light positioning is a promising indoor positioning technique for its high accuracy and immunity to radio-frequency interferences. However, it is expensive and time-consuming to manually measure the coordinates of the massive beacon light-emitting diodes (LEDs). We propose a two-step scheme to locate the LEDs automatically. Firstly, the mobile receiver estimates its own coordinates based on the signals of few LEDs with known coordinates; secondly, the receiver samples the signals of the other unknown LEDs, and estimates their coordinates. The signals are all sampled by a photodiode (PD) array, which outputs the angles of arrival of the LED signals. Explicit positioning error expression is derived as a function of the noise terms and the locations of sampling. Simulation results show that the positioning scheme for the LEDs has centimeter-level accuracy with practical transmit powers. The proposed method can construct the database of the massive LED coordinates based on few beacon LEDs, significantly reducing the cost of manual measurements. Both analytical and simulation results show that the accuracy can be improved by properly choosing the transmit powers and the sampling places.

Existing visible light positioning systems suffer from high computational complexity or cannot output rotation estimation results, making it difficult to support indoor navigation. This paper introduces an indoor positioning system with two beacon light-emitting diodes (LEDs) and a photodiode array at the receiver. The photodiode array can estimate the angles of arrival of the light signals from the beacon LEDs, and the user coordinates can be expressed as closed-form functions of the LED coordinates and the measured light directional vectors. We also carry out asymptotic error analysis for the positioning algorithm, and the analytical results reveals important insights for the system design. Simulation results show that the system can achieve centimeter-level accuracy and low average rotation estimation error.

Abstract This paper considers the wireless powered mobile edge computing combining wireless power transmission (WPT) and mobile edge computing, where the hybrid access point (HAP) uses multiple radio beams to charge multiple wireless devices (WDs) and WD adopts the binary offloading mode to offload computation workload to HAP via FDMA. It is aimed to maximize the sum computation rate (SCR) of WDs by jointly optimizing the binary offloading decision, transmit power of each radio beam, and WPT duration. Due to the strong coupling between the offloading decision and other optimization variables, the SCR maximization is formulated as a mixed integer nonlinear programming problem. To address this challenging problem, an online DRL‐based decoupling optimization algorithm is proposed. Specifically, the original problem is first split into a top‐problem of optimizing binary offloading decision and a sub‐problem of optimizing transmit powers and WPT duration under given offloading decision. Then a self‐learning DRL framework is designed to output the near‐optimal offloading decisions. Finally, for the sub‐problem, based on the Lagrangian dual theory, an efficient approach to fast obtain the closed‐form expression of the optimal solution is proposed. The simulation results show that the proposed DRL‐based algorithm can achieve the near‐maximal SCR with low computational complexity.

The large scale reflector array of programmable metasurfaces is capable of increasing the power efficiency of backscatter communications via passive beamforming and thus has the potential to revolutionize the low-data-rate nature of backscatter communications. In this paper, we propose to design the power-efficient higher-order constellation and reflection pattern under the amplitude constraint brought by backscatter communications. For the constellation design, we adopt the amplitude and phase-shift keying (APSK) constellation and optimize the parameters of APSK such as ring number, ring radius, and inter-ring phase difference. Specifically, we derive closed-form solutions to the optimal ring radius and inter-ring phase difference for an arbitrary modulation order in the decomposed subproblems. For the reflection pattern design, we propose to optimize the passive beamforming vector by solving a multi-objective optimization problem that maximizes reflection power and guarantees beam homogenization within the interested angle range. To solve the problem, we propose a constant-modulus power iteration method, which is proven to be monotonically increasing, to maximize the objective function in each iteration. Numerical results show that the proposed APSK constellation design and reflection pattern design outperform the existing modulation and beam pattern designs in programmable metasurface enabled backscatter communications.

Visible light positioning (VLP) systems provide higher accuracy and lower costs compared to their radio-frequency counterparts. Angle-of-arrival (AOA) based VLP systems are stable and accurate. However, existing AOA VLP systems with a photodiodes (PD) array usually require an unrotated receiver, densely placed LEDs, or auxiliary devices such as accelerometers and gyroscopes. In this paper, we develop an approach to exploit LED clusters and a PD array to locate the receiver. The new approach increases the angle-difference-of-arrival (ADOA) positioning accuracy and circumvents the rotation estimation in AOA positioning. More importantly, the method only requires the solution of linear equation sets, which is computationally efficient. Simulations show that the proposed positioning method can achieve an average mean error below 0.1 m within a 3 m × 3 m area.

Abstract Mobile edge computing is the network technology for providing computing resources of edge computing server to Internet of Things (IoT) applications. Additionally, wireless power transfer (WPT) technology can provide stable energy supplying to IIoT nodes and to overcome the limited node lifetime problem faced when using batteries. In this paper, the wireless powered mobile edge computing network is considered where the edge computing server transfers RF energy to IIoT nodes which use harvested energy to offload partial computation workload based on OFDMA and also conduct local computation. The aim is to maximise the weighted sum computation rate by jointly optimising the WPT duration and the amount of energy used for offloading at each node for each time frame. This paper proposes an offloading approach based on deep reinforcement learning which is able to quickly obtain the near‐optimal offloading solutions. Specifically, the original offloading problem is decomposed into the sub‐problem of optimising the energy allocated for offloading under a given WPT duration and the top‐problem of optimising the WPT duration. Simulation results demonstrate that the proposed algorithm can achieve the near‐optimal weighted sum computation rate with very low complexity, which is tailored for the practical dynamic‐channel environment.

Multi-access edge computing (MEC) has been one promising solution to reduce the computation delay of wireless devices. Due to the high spectrum efficiency of non-orthogonal multiple access (NOMA), this paper studies the single-user multi-edge-server MEC system based on downlink NOMA, aiming to minimize task computation delay by jointly optimizing the NOMA-based transmission duration (TD) and workload offloading allocation (WOA) among edge computing servers. This task computation delay minimization (CDM) problem is formulated as a nonconvex optimization problem. To solve the CDM problem efficiently, we decompose it into the sub-problem of determining the optimal WOA with a given TD and the top-problem of optimizing the TD. For the sub-problem, we first derive its some important properties and then design an efficient channel quality ranking based algorithm to obtain the optimal WOA. We solve the top-problem for the static-channel and dynamic-channel scenarios, respectively. For the static-channel scenario, we design an optimal algorithm which only apply once the golden section search method to obtain the optimal TD of first task and directly obtain the optimal offloading solution for any consequently arrived task with different workloads. For the dynamic-channel scenario where the channel qualities from the wireless device to the edge-computing servers are varying, it is critical to quickly determine the current task’s offloading solution under the current channel state and task workload, which is very challenging for the traditional optimization methods. In order to conquer this challenge, we propose the deep reinforcement learning (DRL) based algorithm, which can obtain the near-optimal offloading solution instantly after enough learning. Finally, we validate through simulations the advantages of NOMA over frequency division multiple access (FDMA).

Traditional visible light positioning (VLP) systems estimate receivers’ coordinates based on the assumption that light-emitting diode (LED) coordinates are known accurately, which is not always practical. Even with accurate laser range finders, the structural changes of the buildings due to temperature, humidity, and material aging might affect the LED coordinates unpredictably. Besides, when the number of LEDs increases, it is time-consuming to build a database of LED coordinates. In this paper, we propose a novel system exploiting two optical angle-of-arrival (AOA) estimators to localize the LEDs, and the AOA estimators have fixed relative positions. Each AOA estimator has four photodiodes (PDs) towards different orientations to estimate the incident light direction. Considering the additive noises imposed on the PDs, we derive a closed-form error expression of the LED coordinate estimator. Analytical and simulation results show that the two AOA estimators should be set symmetrically with respect to the room center to reduce the error of the LED coordinates database. Moreover, an experimental platform is built with two fixed AOA estimators and one moving LED in order to verify the approach. The measured average localization error is 4.46 cm and the time of one localization is about 23.4 ms.

Lognormal-Rician random variables (RV) can model large-scale and small-scale fading, line-of-sight and non-line-of-sight channels in wireless communications. However, the versatile stochastic tool is mathematically intractable for correlated cases. In this work, we derive closed-form expressions for the outage probabilities of diversity receptions over dual-branch lognormal-Rician fading channels with arbitrary correlation at high signal-to-noise ratio (SNR). The analytical results show that the lognormal RV and the Rician RV, respectively, contribute an exponential factor and a fractional factor to the outage probability expressions. Some important insights are revealed based on the elegant expressions. For example, negative correlation between the lognormal RVs can suppress the outage probability, and the phase of the complex correlation coefficient between the accompanying Gaussian RVs in the Rician channels can significantly influence the performance.

Abstract Intelligent reflecting surface (IRS) is a new concept originating from metamaterials, which can achieve beamforming through controllable passive reflecting. This device makes it possible to engineer the wireless communication environment, and has drawn increasing attention. However, the associated channel models in current literature are mainly borrowed from conventional wireless channel models directly, omitting the unique features of IRS. In this paper, a geometry‐based stochastic channel model for IRS‐assisted wireless communication system is employed. The model has certain accuracy and low computational complexity. In particular, it captures the correlations of subchannels associated with different IRS elements, which is typically not considered in current works. Based on this channel model and the derived channel spatial correlation functions (CFs), an iterative reflection coefficients configuration method is proposed exploiting statistical channel state information to maximise the ergodic channel capacity. The impacts of the IRS spatial positions as well as the number of the IRS elements on the ergodic channel capacity is investigated through simulations. It is found that to obtain a larger ergodic channel capacity, the IRS should be placed in the vicinity of either the transmitter side or the receiver side, which is a useful guideline for practical deployment.

Abstract Low earth orbit constellation satellite communication has the characteristics of low propagation delay, low path loss, low launch cost and wide range of applications. Due to the low orbital altitude and the short orbital period of low earth orbit satellites, the relative position between satellites and gateway stations changes fast. As a result, the links between gateway stations and satellites need to be switched continuously. Based on the minimum handover frequency algorithm, this paper considers the balance of satellite workloads, and proposes a load balanced satellite handover strategy. In the proposed handover strategy, a joint optimization algorithm is employed, and the power allocation of the satellite is optimized to improve the system capacity. In the multi‐satellite connection model, an adaptive power allocation algorithm is proposed to guarantee the service quality of the system. Simulation results demonstrate the efficiency of the proposed satellite handover strategy.

Free space optical communication has been applied in many scenarios because of its security, low cost and high rates. In such scenarios, a tracking system is necessary to ensure an acceptable signal power. Free space optical links were considered unable to support fast moving nodes, such as unmanned aerial vehicles, cars or pedestrians, in optical mobile communication because existing tracking schemes fail to track the nodes accurately and rapidly. In this paper, we propose a novel tracking system exploiting multiple beacon lasers. Each beacon laser's power is measured to estimate the orientation of the target. Unlike existing schemes which drive servo motors multiple times based on consecutive measurements and feedback, our scheme can directly estimate the next optimal targeting shift for the servo motors based on a single measurement, allowing the tracking system to converge much faster. Closed-form outage probability expression is derived for the optical mobile communication system with ideal tracking, where pointing error and moving statistics are considered. To maintain sufficient average power and reduce the outage probability, the recommended beam divergence of the main laser is expressed in closed form as a function of the target's statistics of random shift, providing insights to the system design.

Traditional asymptotic analysis techniques fail for lognormal random variables (RV) because a lognormal probability density function cannot be quantified by Taylor series at the origin, and the moment-generating function of a lognormal RV does not yield a unified form. In this work, we derive two closed-form asymptotic expressions for the multivariate cumulative distribution function (CDF) of L correlated lognormal RVs. Among these two asymptotic expressions, one has an elegant form but low converging speed, while the other has a fast converging speed but involves Gaussian Q-functions. Furthermore, we derive asymptotically tight upper and lower bounds for the multivariate CDF to evaluate the error of the asymptotic approximation. The new analytical results are shown to be effective in evaluating the outage probability of selection combining (SC) over correlated lognormal fading channels when Monte Carlo simulation is expensive. More importantly, the elegant asymptotic expressions show that little outage performance improvement can be introduced to the SC system by adding more antennas for some cases.

It is challenging to quantify the left tail behavior of the cumulative distribution function of a sum of lognormal random variables, especially when they are correlated and non-identically distributed. The mathematical intractability hampers the analysis on diversity receptions over lognormal fading channels. In this paper, a novel asymptotic technique is exploited to study non-identically distributed dual-branch lognormal fading channels with correlation. Closed-form asymptotic outage probability expressions are derived for maximum-ratio combining (MRC), equal-gain combining (EGC), and selection combining (SC). Three counterintuitive facts are revealed: 1) one of the two channels contributes no asymptotic performance gain to the diversity reception systems under certain conditions, implying that a link can be discarded without causing asymptotic performance loss; 2) SC outperforms EGC under certain conditions; and 3) the outage probability of SC is arbitrarily close to that of MRC under certain conditions. The new result reveals insights into the long-standing problem of asymptotic analysis for correlated lognormal fading channels, paving the way for analysis on more general channel models. This paper also provides an efficient approach to evaluate the non-identically distributed lognormal fading channels at a high signal-to-noise ratio. Practically, the analytical results prevent some redundant units that contribute no performance gain.

Exact outage probability analysis for equal-gain combining (EGC) over lognormal fading channels results in multi-fold integrals, and traditional asymptotic analysis techniques cannot work for lognormal fading channels due to the channels' infinite diversity order. In this work, we derive a closed-form asymptotic outage probability expression for dual-branch EGC over correlated lognormal fading channels with different statistics. The new result reveals insights into the long-standing problem of asymptotic analysis for correlated lognormal fading channels, paving the way for analysis on more general channel models. The result also provides an efficient way to evaluate the performance of dual-branch EGC over correlated and nonidentically distributed lognormal channels at high signal-to-noise ratio.

Existing indoor positioning methods for visible-light communication systems require large database, powerful signal processing units, additional sensors, such as gyroscopes, or the receiver placed toward a certain angle. These assumptions limit the applications of the indoor positioning systems in low-cost scenarios. In this paper, we propose a novel positioning framework based on the angle differences of arrival (ADOA) in 3-D coordinate systems, which can be used in receivers with image sensors or photodiodes. The proposed ADOA positioning does not require the receiver to be placed toward a certain angle and no additional sensor is required. Two positioning algorithms are proposed: one is based on the method of exhaustion (MEX), and the other is based on the least squares method (LSM). The MEX algorithm is analytically proved to be the optimal, while the LSM algorithm has much lower complexity. Both the upper and lower bounds are derived for the average discrepancy between the exact position and the estimated position. These performance bounds can facilitate the design of a light-emitting diode array. Experimental results show that the MEX algorithm can achieve an average error of 3.20 cm with a time cost of 0.36 s, and the LSM algorithm can achieve an average error of 14.66 cm with a time cost of 0.001 s.

It is challenging to derive closed-form performance expressions for composite channels considering both the large-scale lognormal fading and small-scale fading, thus most existing analyses on the composite channels are based on numerical approaches such as the Gauss-Hermite quadrature. In this paper, we develop a simple framework to achieve closed-form asymptotic expressions for the outage probabilities and the error rates of the composite lognormal-X fading channels, where the “X” can be Rayleigh, Rician, Nakagami-m or any other fading channels that have finite diversity orders. It is proved that the lognormal fading contributes an exponential factor to the asymptotic expressions, and the exponential factor is related to the diversity order of the X channel and the lognormal parameters. The new analytical tool is shown to be versatile in evaluating the performance of radio-frequency communications suffering both the large-scale and small-scale fading, including diversity reception systems, relaying systems, and distributed antenna systems. It is also shown to be useful in evaluating free-space optical communication systems with pointing error, and in deriving asymptotic signal-to-noise ratio gaps between different modulation formats over lognormal fading channels. The elegant asymptotic expressions reveal insights into the composite fading channels and can be a criterion for various system designs.

Prior asymptotic performance analyses are based on the series expansion of the moment-generating function (MGF) or the probability density function (PDF) of channel coefficients. However, these techniques fail for lognormal fading channels because the Taylor series of the PDF of a lognormal random variable is zero at the origin and the MGF does not have an explicit form. Although lognormal fading model has been widely applied in wireless communications and free-space optical communications, few analytical tools are available to provide elegant performance expressions for correlated lognormal channels. In this paper, we propose a novel framework to analyze the asymptotic outage probabilities of selection combining (SC), equal-gain combining (EGC), and maximum-ratio combining (MRC) over equally correlated lognormal fading channels. Based on these closed-form results, we show: 1) the outage probability of EGC or MRC becomes an infinitely small quantity compared to that of SC at high signal-to-noise ratio (SNR); 2) channel correlation can cause an infinite performance loss at high SNR; and 3) negatively correlated lognormal channels can outperform the independent lognormal channels. The analyses reveal insights into the long-standing problem of asymptotic performance analyses over correlated lognormal channels, and circumvent the time-consuming Monte Carlo simulation and numerical integration.

The Beckmann distribution has a wide range of applications in radio-frequency communications, free-space optical (FSO) communications, and underwater wireless optical communications (UWOC). However, the cumulative distribution function (cdf) of the Beckmann random variable (RV) does not have a closed-form expression, which makes it challenging to derive analytical solutions for the outage probability of systems involving Beckmann RVs. In this paper, we study the generalized Beckmann distribution, which includes the Beckmann, Rayleigh, Rician, Nakagami-m, Hoyt, κ-μ, η-μ, single-sided Gaussian, and the Beaulieu-Xie distributions as special cases. Three approaches are proposed to estimate the cdf of the generalized Beckmann distribution, including closed-form upper and lower cdf bounds, single-fold integration based on the closed-form characteristic function, and a left-tail cdf approximation. We compare the three approaches in terms of the ranges of applications and the computation time complexity. Based on the new cdf estimation techniques, one can efficiently evaluate the outage probabilities of pointing-error-limited FSO systems, UWOC systems, and maximum-ratio combining over arbitrarily correlated generalized Beckmann channels.

Prior visible-light communication (VLC) positioning systems require the receiver to be placed towards a certain angle, or extra sensors like gyroscopes must be equipped to estimate the angle of the receiver. In this work, we propose a positioning scheme using cameras based on angle difference of arrival (ADOA). ADOA-based positioning does not need the receiver to know its tilting angle, and the position estimation is modeled as a three-variate optimization problem in a finite search region, which is tractable in most smart devices. Our experimental and numerical results show that the ADOA-based VLC positioning scheme achieves accuracy of centimeters and the average cost time is as low as 0.2 seconds.

In this paper, an indoor visible light communication (VLC) system with the input-dependent noise is considered. In the system, the main noise is caused by Gaussian noise, however, with a noise variance depending on the current input signal strength. In the presence of the input-dependent noise, the theoretical expression of the bit error rate (BER) for the VLC using on-off keying is derived. Based on the derived BER, an optimization problem is formulated to improve the BER performance by tilting the receiver plane. The proposed optimization problem is proven to be a convex optimization problem, which can be efficiently solved by using the specialized solver such as the CVX toolbox for MATLAB. To verify the accuracy of the derived expression of the BER, all theoretical results are thoroughly confirmed by using the Monte-Carlo simulations. Moreover, simulation results show that the larger the variance of the input-dependent noise is, the worse the BER performance becomes. Additionally, the BER performance can be dramatically improved by tilting the receiver plane properly.

Co-time co-frequency full-duplex (CCFD) communication is challenging for radio frequency communications due to the overwhelming self- interference. For visible light communications, it was reported that two light-emitting diodes could construct a half-duplex communication link, but no CCFD communication using two diodes have been realized. In this work, we introduce an on-chip CCFD system using a pair of micrometer-scale InGaN/GaN multiple quantum-well diodes, which can detect and emit light at the same time. Maximum-likelihood estimator is derived and applied to extract the useful signal from the received mixed signal. The proposed CCFD system provides a promising way to decrease the size of on-chip optical communication systems and portable optical communication devices, and it can also reduce the number of optical devices in optical fiber communications and wireless optical communications without introducing much extra cost.

We propose and fabricate a monolithic optical interconnect on a GaN-on-silicon platform using a wafer-level technique. Because the InGaN/GaN multiple-quantum-well diodes (MQWDs) can achieve light emission and detection simultaneously, the emitter and collector sharing identical MQW structure are produced using the same process. Suspended waveguides interconnect the emitter with the collector to form in-plane light coupling. Monolithic optical interconnect chip integrates the emitter, waveguide, base, and collector into a multi-component system with a common base. Output states superposition and 1×2 in-plane light communication are experimentally demonstrated. The proposed monolithic optical interconnect opens a promising way toward the diverse applications from in-plane visible light communication to light-induced artificial synaptic devices, intelligent display, on-chip imaging, and optical sensing.

Prior work made a comparison between the diversity orders of adaptive decode-and-forward (ADF) and improved adaptive decode-and-forward (IADF) over free- space optical (FSO) Gamma-Gamma fading channels. However, the diversity orders of FSO lognormal channels are infinite, and thus the prior method of comparison becomes invalid for FSO lognormal fading channels. In this work, we compare the performance of ADF and IADF over lognormal fading channels based on asymptotic relative diversity order. We analytically prove that the frame error rate of IADF is much smaller than that of ADF when the signal-to-noise ratio is sufficiently large.

Prior results on exact error rates and outage probabilities of diversity receptions over arbitrarily correlated fading channels involve intractable nested integrals, which makes the performance analyses challenging to carry out. In this paper, we derive asymptotic expressions and asymptotically tight bounds for error rates and outage probabilities of diversity receptions (including maximal-ratio combining, selection combining, and equal-gain combining) over arbitrarily correlated Beckmann fading channels. These new asymptotic performance expressions reveal insights into the diversity systems over arbitrarily correlated Beckmann fading channels, and the accuracy of the asymptotic performance expressions is explicitly quantified by the performance bounds. Using these analytical tools, one can assess the performance of the diversity reception systems over arbitrarily correlated Beckmann fading channels without resorting to time-consuming Monte Carlo simulation or numerical multifold integration.

Prior performance analyses of diversity receptions over arbitrarily correlated Nakagami-m fading channels use power correlation matrix to define correlation among branches. However, the power correlation matrix cannot uniquely determine the joint probability density function of the branch amplitudes as well as the system performance. In this paper, we derive asymptotically tight and closed-form upper and lower bounds for the error rates and the outage probabilities of maximum-ratio combining, equal-gain combining, and selection combining over arbitrarily correlated Nakagami-m fading channels. The correlation between the Nakagami-m random variables is defined through the correlation matrix of the accompanying Gaussian random variables, which can uniquely determine the joint probability density function of the Nakagami-m variables. Using the analytical results and Monte Carlo simulation, we show that the performance of diversity reception systems over Nakagami-m fading channels cannot be uniquely determined by the power correlation matrix of the branches. Furthermore, we also study the factors that determine the asymptotic performance of the diversity receptions.

Exact error rate expressions for equal-gain combining over arbitrarily correlated Rician channels involve intractable integrals. Thus, asymptotic error rate expression was derived to provide accurate approximation in large signal-to-noise ratio region. However, the existing asymptotic analysis cannot evaluate the accuracy of the asymptotic error rate expression for equal-gain combining. In this work, we derive closed-form and asymptotically tight upper and lower bounds for the error rate of equal-gain combining over Rician fading channels with arbitrary correlation. These bounds can be used to evaluate the accuracy of the asymptotic error rate expression for equal-gain combining at high signal-to-noise ratio without using the time-consuming Monte Carlo simulation.

Prior work on exact error rate of diversity receptions over arbitrarily correlated Rician channels involves intractable integrals. Although asymptotic error rates provide closed-form approximations which are accurate at high signal-to-noise ratio, it is still unknown at what signal-to-noise ratio the approximation becomes accurate. In this paper, asymptotically tight closed-form upper and lower bounds are derived for maximal-ratio combining and selection combining over arbitrarily correlated Rician channels. Using these bounds, we can show at what signal-to-noise ratio that the asymptotic error rates are accurate without resorting to time-consuming Monte Carlo simulation.

Optimal relay placement is studied for free-space optical multihop communication with link obstacles and infeasible regions. An optimal relay placement scheme is proposed to achieve the lowest outage probability, enable the links to bypass obstacles of various geometric shapes, and place the relay nodes in specified available regions. When the number of relay nodes is large, the searching space can grow exponentially, and thus, a grouping optimization technique is proposed to reduce the searching time. We numerically demonstrate that the grouping optimization can provide suboptimal solutions close to the optimal solutions, but the average searching time linearly grows with the number of relay nodes. Two useful theorems are presented to reveal insights into the optimal relay locations. Simulation results show that our proposed optimization framework can effectively provide desirable solution to the problem of optimal relay nodes placement.

We study the performance of multi-branch hierarchical combining diversity systems. Closed-form expressions are derived for asymptotic error rate and asymptotic outage probability of hierarchical selection-combining maximal-ratio combining (SC-MRC) and selection-combining equal-gain combining over Rician channels with arbitrary correlation. Comparison is made between hybrid-selection maximal-ratio combining (HS/MRC) and hierarchical SC-MRC, and it is shown that SC-MRC suffers at most 0.6 dB signal-to-noise ratio (SNR) loss with respect to HS/MRC for practical number of antennas. However, SC-MRC incurs much smaller insertion loss. Numerical results show that the analytical solutions can provide accurate estimation of error rate and outage probability in large SNR region.

Optimal relay placement is studied for free-space optical multi-hop communication when obstacles and infeasible regions exist. We propose an optimization framework to find the optimal relay locations in the specified available regions and make laser links bypass obstacles of various shapes. A grouping optimization technique is proposed to reduce the optimization time when the number of relays is large, and we numerically demonstrate that this technique can provide solutions close to the optimal solutions, and its average searching time grows linearly with the number of relays. Simulation results show that our proposed optimization framework can effectively provide good solution to the problem of optimal relay placement in limited available regions, and enable the laser links to bypass obstacles in order to achieve minimum end-to-end outage probability.