In this paper, we introduce a novel user association (UA) and resource block group (RBG) allocation (RA) method utilizing multi-agent reinforcement learning (MARL) for a multi-user multiple-input multiple-output (MU-MIMO) downlink system. Unlike traditional MARL radio resource management (RRM) approaches, which utilize user equipment (UE) as learning agents, base stations (BS) are deployed as agents for practical consideration. However, this will significantly enlarge the action space and bring about action constraint violation problems. We employ dual-actor neural networks to separate UA and RBG allocation actions, thereby effectively reducing the joint action space and accelerating exploration. In addition, a Q-value-rank-based action projection algorithm is proposed to address the cross-agent coupling constraints. The simulation results demonstrate that the proposed MARL framework with action projection outperforms other baselines in terms of RRM performance.

Optimizing antenna parameters like azimuth, down-tilt, and power is crucial for coverage and capacity optimization (CCO) in next-generation wireless networks. However, traditional expert knowledge-based methods struggle to maintain optimal results when faced with changing environments. To address this, we propose a guided deep reinforcement learning (DRL) algorithm that learns a policy to dynamically adjust antenna parameters based on the evolving environment. Our approach employs proximal policy optimization-based DRL and integrates a problem-specific pretraining process using zero-order gradient descent. The pretrain policy serves as a guiding policy, enabling the agent to explore and discover high-reward regions, thus accel-erating the learning process. The performance of our solution is validated by numerical experiments conducted on a 5G simulation platform with real-world topological properties. The results show that our approach achieves significantly faster convergence and outperforms baseline methods in terms of CCO performance.

The Terahertz band, spanning from 0.1 THz to 10 THz, is envisioned as a key technology to realize ultrahigh data rates in the 6 G and beyond mobile communication systems, due to its abundant bandwidth resource. However, to realize $\mathbf{T H z}$ communications, one substantial step is to fully understand the THz channels, which relies on extensive channel measurements. In this paper, using a correlation-based time domain channel sounder, measurement campaigns are conducted in a laboratory at 140 GHz and 220 GHz. In the data postprocessing procedures, the time drift of clock signals is corrected using a linear interpolation/extrapolation method. Based on the measured results, the main objects that provide significant oncescattering clusters are found, based on which the scattering losses are calculated and analyzed. Furthermore, the channel characteristics, including path loss, shadow fading, K-factor, etc. are calculated and compared to 3GPP standard values. The propagation analysis and channel characteristics are helpful to study channel modeling and guide system design for $\mathbf{T H z}$ communications.

The Terahertz band, ranging from 0.1 THz to 10 THz, is envisioned as a key technology to realize ultra-high-speed communications in 6G and beyond wireless networks, attributed to its abundant bandwidth resource. Channel measurements and link-level analysis are still missing in a typical use case of THz picocells. In this paper, using a correlation-based time domain channel sounder, channel measurement campaigns are conducted on a university street at 220 GHz. Based on the measurement results, a full portrait of channel characteristics, including path loss, shadow fading, K-factor, delay and angular spreads, as well as cluster parameters, is calculated and analyzed. Comparison with existing 3GPP standard models shows weak multipath effects and strong sparsity in the THz picocell scenario. Moreover, small-scale fading is evaluated and fitted, where a Rician distribution among other competitors shows great fitting performance. Furthermore, considering realistic THz communication links, the ergodic capacity and outage probability are analyzed. Results have shown that more than 150 Gbps channel capacity can be achieved and reliable communication links can be guaranteed for coverage up to 40 m in the THz picocell. The results and analysis in this work offer guidance for effective system design for future THz picocell communications.

This paper presents an exhaustive comparison of channel measurements and appropriate channel statistics at 28 GHz, 38 GHz, and 130 GHz in the light of massive measurements conducted in an aircraft cabin environment. A total of 84 transmitter-receiver (Tx-Rx) positions are measured, covering both line-of-sight (LoS) and non-line-of-sight (NLoS) cases, with Tx-Rx distances ranging from 1 m to 10 m. The close-in and floating-intercept path loss models are presented for the cabin environment, and root-mean-square (RMS) delay spread (DS) and angular spread (AS) are compared and analyzed. The results indicate that the path loss exponent (PLE) is smaller than the free space PLE in the LoS aisle case over all measured frequencies. Moreover, the RMS DS and AS decrease as frequency increases in LoS and NLoS cases. This work can be applied to the design and optimization of the wireless communication system within the aircraft cabin.

This paper presents a comparative investigation of channel measurements and corresponding channel characteristics at 28 GHz, 38 GHz and 132 GHz in two different industrial environments, including a micro drilling-milling area and a large milling area. To depict the channel characteristics in industrial environments accurately, the close-in free space path-loss model and the alpha-beta-gamma model are used to examine path loss. Additionally, we analyze and compare the root mean square (RMS) delay spread (DS) and angular spread (AS) in different areas. The findings indicate the path loss exponent (PLE) has a dependency on the frequency, and all PLEs are less than the free space value 2.0 for the LoS case. Furthermore, RMS DS and AS exhibit distinct characteristics across various environments. This research provides valuable insights for design and optimization of industrial environmental systems.

Terahertz (THz) communication technology holds significant potential for applications in the industrial internet of things (IIoT). Accurately characterizing the THz channel is critical for designing and optimizing communication systems in IIoT scenarios. However, the significantly higher frequencies in the THz band impede the effective utilization of channel models designed for microwave or millimeter-wave frequency bands. To overcome this challenge, extensive measurement campaigns are necessary to thoroughly investigate the characteristics of THz channels in indoor IIoT scenarios. This paper presents a measurement-based channel characterization in indoor IIoT scenarios at a frequency range of 215–225 GHz. We first present VNA-based channel measurement campaigns in micro drilling-milling and large milling areas. The measured data are further processed to obtain the channel impulse response. The key prop-agation channel parameters, e.g., path loss, power delay angle profile, delay spread, and angular spread, are calculated and analyzed in the line-of-sight case. Results demonstrate a favorable spatio-temporal consistency in multipath signal propagation and the physical spatial environment. Furthermore, significant correlations are observed between the channel characteristics and scatterer distribution within IIoT scenarios. The findings of this paper will make substantial contributions to the design and development of THz communication systems in IIoT scenarios.

Terahertz (THz) communications are envisioned as a promising technology for sixth-generation (6G) and beyond systems, owing to its unprecedented multi-gigahertz (GHz) bandwidth. In this paper, channel measurement campaigns in indoor scenarios at low-THz frequencies, i.e., 201-209 GHz, are reported. Four different communication scenarios including 90 transmitter-receiver pairs are measured in two channel measurement campaigns of a meeting room and an office room, respectively. The two measurement campaigns contains four scenarios, namely, a meeting room, cubicle area, hallway and non-line-of-sight (NLoS) case. The propagation of multi-path components (MPCs) in the four scenarios is characterized by the power-delay-angular profiles. Based on them, the temporal and spatial consistency for varying receiver locations in the complex hallway and NLoS scenarios are verified. To characterize, the large-scale best-direction and omni-directional path losses in indoor scenarios are separately analyzed and modeled by the close-in (CI) model. Furthermore, the small-scale channel parameters, e.g., the number of clusters, delay spread, angular spread, and cluster time-of-arrival are analyzed and modeled by proper distributions. As a general framework, a ray-tracing-statistical hybrid model is proposed for wireless propagation at 201-209 GHz, although, admittedly, the measurement results and analysis reveal that the channel characteristics in various indoor scenarios exhibit noticeable differences that need tailored parameter settings.

This paper presents the large-scale fading characteristics of Terahertz (THz) channel in indoor hotspot scenarios. A series of channel measurements at 219–224 GHz are conducted in a classroom and a hallway. In order to investigate the large-scale fading characteristics, the omnidirectional and best directional path loss are separately analyzed by close-in and floating-intercept models, and the Rician$K$-factor, root mean square delay spread and angular spread are analyzed to estimate the multi-path component richness, time and angle dispersion in various indoor scenarios. Further, these values are compared with what has been given in the Third Generation Partnership Project 38.901 in frequency bands lower than 100 GHz. In light of the results, the office area in hallway scenario shows the most severe path loss, and derives the highest time and angle dispersion. Besides, the measurement results enrich the datasets of THz channel propagation, which is helpful for the design and optimization of THz communication systems for sixth-generation.

Terahertz (THz) communication is considered as one of the potential candidate technologies in the sixth generation (6G) wireless systems. This paper introduces channel measure-ments in two sub-THz bands, i.e., 140- and 220-GHz bands, in an outdoor street canyon environment for both line-of-sight (LoS) and non-line-of-sight (NLoS) scenarios with a frequency-domain vector network analyzer (VNA)-based sounder. Based on the measurement results, we computed and analyzed the statistical features of wireless propagation channels, including path loss, root mean square (RMS) delay spreads (DSs), azimuth spreads of arrival (ASA), and elevation spreads of arrival (ESA). Moreover, we observe the birth and death of clusters over a straight trajectory under the NLoS condition. A high-resolution param-eter estimation algorithm, i.e., the space-alternating generalized expectation-maximization (SAGE) algorithm, was employed to eliminate the effects of antenna patterns and the density-based spatial clustering of applications with noise (DBSCAN) algorithm was used to clustered the multipath components (MPCs). The obtained statistical properties of measured channels and obser-vations on channel evolution can be employed in THz channel modeling and system design.

Routing is critical for maximizing the performance of Dragonfly networks. It decides how the packets are forwarded from their source nodes to their destination nodes. Considering the traffic pattern varies over time in real networks, adaptive routing is desirable. Existing adaptive routing algorithms employ local information to make dynamic routing decisions, which have shown significant limitations since the local information typically fails to reflect the global network condition. In this paper, inspired by the hierarchical topology of Dragonfly, we develop a hierarchical Reinforcement Learning (RL) algorithm named Q-hierarchical for Dragonfly networks. Q-hierarchical learns to adapt from data without the need to model the traffic pattern. It simplifies the complexity of traditional RL-based routing by routing in a hierarchical manner, i.e., inter-group routing and intra-group routing. Hence it can be applied on a large network. We also develop a fast and effective learning strategy for the hierarchical RL. The performance of Q-hierarchical is evaluated through comprehensive tests on two Dragonfly topologies. The results show that our approach provides comparable performance under uniform random traffic pattern and outperforms some routing algorithms in terms of averaging packet delay and affordable load under adversarial traffic pattern.

Owning abundant bandwidth resource, the Tera-hertz (0.1-10 THz) band is a promising spectrum to support sixth-generation (6G) and beyond communications. As the foundation of channel study in the spectrum, channel measurement is ongoing in covering representative 6G communication scenarios and promising THz frequency bands. In this paper, a wideband channel measurement in an L-shaped university campus street is conducted at 306–321 GHz and 356–371 GHz. In particular, ten line-of-sight (LoS) and eight non-line-of-sight (NLoS) points are measured at the two frequency bands, respectively. In total, 6480 channel impulse responses (CIRs) are obtained from the measurement, based on which multi-path propagation in the L-shaped roadway in the THz band is elaborated to identify major scatterers of walls, vehicles, etc. in the environment and their impact on multi-path components (MPCs). Furthermore, outdoor THz channel characteristics in the two frequency bands are analyzed, including path losses, shadow fading, cluster pa-rameters, delay spread and angular spread. In contrast with the counterparts in the similar outdoor scenario at lower frequencies, the results verify the sparsity of MPCs at THz frequencies and indicate smaller power spreads in both temporal and spatial domains in the THz band.

We consider a generalized collision channel model for general multi-user communication systems, an extension of Massey and Mathys’ collision channel without feedback for multiple access communications. In our model, there are multiple transmitters and receivers sharing the same communication channel. The transmitters are not synchronized and arbitrary time offsets between transmitters and receivers are assumed. A "collision" occurs if two or more packets from different transmitters partially or completely overlap at a receiver. Our model includes the original collision channel as a special case.This paper focuses on reliable throughputs that are approachable for arbitrary time offsets. We consider both slot-synchronized and non-synchronized cases and characterize their reliable throughput regions for the generalized collision channel model. These two regions are proven to coincide. Moreover, it is shown that the protocol sequences constructed for multiple access communication remain "throughput optimal" in the generalized collision channel model. We also identify the protocol sequences that can approach the outer boundary of the reliable throughput region.

To leverage the computing power in an edge network, one can divide a machine learning task into several subtasks and assign the subtasks to several computing devices to complete. Under master-worker architecture, the master divides and distributes the data to several workers. In each iteration, the master asks the workers to compute some function of the local data stored in the workers. For example, in gradient-based learning, this function can be the partial gradient function. Since the workers have different computing resources, the speed of the distributed learning is hindered by some workers with long latency, called the stragglers. Gradient coding solves the problem of stragglers by allowing the master to recover the desired feedback information in the presence of s stragglers. If the total number of stragglers is n, the master can just wait for the n−s fastest workers. In this paper we consider the problem of data allocation so that the gradient vector can be approximated obtained by the master node with small error. A block repetition scheme is proved to be the optimal data allocation scheme if we want to minimize the average recovery error.

In this paper, we propose a scatterer-based hybrid channel model framework for integrated sensing and communication (ISAC). The basic idea of the scatterer-based ISAC channel model is to first generate scatterers in the propagation environment and then use simplified ray-tracing to generate both communication and sensing channels. The proposed model is hybrid in the aspect of scatterers i.e., the scatterers consists deterministic part and stochastic part. The deterministic part of the scatterers is from the input environment geometry while the stochastic part is constructed from the stochastic MPCs generated by 3GPP channel standardization framework. A method to couple the stochastic scatterers with the input environment geometry is proposed to avoid unrealistic ray propagation. To support the proposed ISAC channel model in the application of THz communications, we conduct RCS measurement campaigns with common building materials for developing deterministic propagation models at 140 GHz. Also, we conduct communication and sensing channel measurement campaigns in both indoor and outdoor scenarios for extracting fast-fading parameters for 3GPP channel standardization framework. The simulation results show that the proposed model has the capability of characterizing ISAC channel properties and validate with the measurement results.

The Terahertz (0.1-10 THz) band has been envisioned as one of the promising spectrum bands to support ultra-broadband sixth-generation (6G) and beyond communications. In this paper, a wideband channel measurement campaign in a 500-square-meter indoor lobby at 306-321 GHz is presented. The measurement system consists of a vector network analyzer (VNA)-based channel sounder, and a directional antenna equipped at the receiver to resolve multi-path components (MPCs) in the angular domain. In particular, 21 positions and 3780 channel impulse responses (CIRs) are measured in the lobby, including the line-of-sight (LoS), non-line-of-sight (NLoS) and obstructed-line-of-sight (OLoS) cases. The multi-path characteristics are summarized as follows. First, the main scatterers in the lobby include the glass, the pillar, and the LED screen. Second, best direction and omnidirectional path losses are analyzed. Compared with the close-in path loss model, the optimal path loss offset in the alpha-beta path loss model exceeds 86 dB in the LoS case, and accordingly, the exponent decreases to 1.57 and below. Third, more than 10 clusters are observed in OLoS and NLoS cases, compared to 2.17 clusters on average in the LoS case. Fourth, the average power dispersion of MPCs is smaller in both temporal and angular domains in the LoS case, compared with the NLoS and OLoS counterparts. Finally, in contrast to hallway scenarios measured in previous works at the same frequency band, the lobby which is larger in dimension and square in shape, features larger path losses and smaller delay and angular spreads.

The TeraHertz (THz) band, spanning the spectrum from 0.1 THz to 10 THz, is envisioned as a key technology in the next generation mobile communication systems. However, the much higher frequencies in the THz band prevents the effective utilization of channel models dedicated for microwave or millimeter-wave frequency bands. To address this problem, numerous measurement campaigns are needed to fully investigate the characteristics of the THz channels. In this paper, a measurement campaign is conducted in an indoor corridor scenario at 306-321 GHz with a frequency-domain Vector Network Analyzer (VNA)-based sounder. The measured data are further processed to obtain the channel impulse response (CIR) and power-delay-angle profile (PDAP), based on which the multipath components (MPCs) are further extracted. Furthermore, the MPCs are clustered using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm and the clusters are matched with propagation paths by considering the real geometry. Moreover, the channel characteristics, including the path loss, delay spread, angular spreads, and cluster parameters are calculated and analyzed. The resulting numerology is helpful to guide system design for THz communications.

The TeraHertz (THz) band (0.1-10 THz), which supports terabit-per-second (Tbps) data rates, has been envisioned as one of the promising spectrum bands for ultra-broadband sixth-generation (6G) communications. In this paper, an angular-resolvable ultra-wideband channel measurement campaign in an indoor L-shaped hallway at 306-321 GHz is presented, by using a frequency-domain vector network analyzer (VNA)-based channel sounder. In particular, four points in the line-of-sight (LoS) case and nine points in the non-line-of-sight (NLoS) case are measured, with a directional antenna equipped at the receiver (Rx) side to resolve multi-path components (MPCs) in the angular domain. The multi-path propagation in the L-shaped hallway in the THz band is elaborated in terms of power delay angular profiles (PDAPs), based on the multi-path component distance (MCD)-based Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. Indoor THz channel characteristics are analyzed in depth. Specifically, the path loss exponents (PLEs) of the LoS case are 1.7222 for the best direction and 1.3910 for the omni-directional path losses. Moreover, in the LoS case, the average logarithmic values of root-mean-square (RMS) delay spread (DS) and azimuth spread of angle (ASA) are -7.7464 and 1.52, while the NLoS case yields larger average RMS DS and ASA of -7.5501 and 1.68, respectively. Besides, the cluster delay difference follows an exponential distribution, for which the average values of cluster delay difference are 37.15 ns in the LoS and 22.59 ns in the NLoS case.

Towards the sixth generation (6G) wireless communications, multiple access is a potential technology to achieve thousand times of traffic capacity enhancement comparing with 5G. This paper investigates the feasible signal-to-interference-plus-noise-ratio (SINR) region for the multi-cell downlink and uplink non-orthogonal multiple access (NOMA) systems. Within a cell, the signals of different users are multiplexed on a channel with different power levels and successive interference cancellation is applied at the receivers to decode the signals. For the downlink, based on Perron-Frobenius Theory, we first derive a necessary and sufficient condition for an SINR vector to be feasible under a fixed decoding order. Then, a closed-form expression for the feasible SINR region is given by the union of the SINR regions under all possible decoding orders. Following a similar idea, the expression of the feasible SINR region for the uplink is also derived. Furthermore, the duality between the multi-cell downlink and uplink NOMA systems is explored. It is proved that the two systems have same feasible SINR region under dual channels. Finally, we propose an efficient algorithm to approximate the SINR region boundary. Simulation results are provided to validate the efficiency of the algorithm and compare the performance with orthogonal multiple access scheme.

Packet routing is a fundamental problem in communication networks that decides how the packets are directed from their source nodes to their destination nodes through some intermediate nodes. With the increasing complexity of network topology and highly dynamic traffic demand, conventional model-based and rule-based routing schemes show significant limitations, due to the simplified and unrealistic model assumptions, and lack of flexibility and adaption. Adding intelligence to the network control is becoming a trend and the key to achieving high-efficiency network operation. In this paper, we develop a model-free and data-driven routing strategy by leveraging reinforcement learning (RL), where routers interact with the network and learn from the experience to make some good routing configurations for the future. Considering the graph nature of the network topology, we design a multi-agent RL framework in combination with graph attention network (GAT), tailored to the routing problem. Three deployment paradigms, centralized, federated, and cooperated learning, are explored respectively. Simulation results demonstrate that our algorithm outperforms some existing benchmark algorithms in terms of packet transmission delay and affordable load.

In this work, we develop practical user scheduling algorithms for downlink bursty traffic with emphasis on user fairness. In contrast to the conventional scheduling algorithms that either equally divides the transmission time slots among users or maximizing some ratios without practical physical interpretations, we propose to use the 5%-tile user data rate (5TUDR) as the metric to evaluate user fairness. Since it is difficult to directly optimize 5TUDR, we first cast the problem into the stochastic game framework and subsequently propose a multi-agent reinforcement learning (MARL)-based algorithm to perform distributed optimization on the resource block group (RBG) allocation. Furthermore, each MARL agent is designed to take information measured by network counters from multiple network layers (e.g. Channel Quality Indicator, Buffer size) as the input states while the RBG allocation as action with a carefully designed reward function developed to maximize 5TUDR. Extensive simulation is performed to show that the proposed MARL-based scheduler can achieve fair scheduling while maintaining good average network throughput as compared to conventional schedulers.

TeraHertz (THz) communications are envisioned as a promising technology, owing to its unprecedented multi-GHz bandwidth. One fundamental challenge when moving to new spectrum is to understand the science of radio propagation and develop an accurate channel model. In this paper, a wideband channel measurement campaign between 130 GHz and 143 GHz is conducted in an office room. The channel measurement campaign in the office room scenario consists of three cases: 1) LoS case in the office area 2) Line-of-Sight (LoS) case in the hallway and 3) NLoS case. Directional antennas are utilized and rotated for resolving the multi-path components (MPCs) in the angular domain. With careful system calibration that eliminates system errors and antenna effects, realistic power delay angular profiles (PDAP) are developed. In light of the measurement results, channel parameters and insights in the THz indoor channel are comprehensively analyzed, including the path loss, delay spread, angular, and correlations among the channel parameters. Furthermore, the propagation of NLoS multipath components (MPCs) in the office room scenario and its impact on the THz channel are discussed and compared with the channel measurement in the meeting at 140 GHz. Our analysis shows that the MPCs that reflect from the partitions dominate in NLoS propagation in office room.

Unmanned aerial vehicles (UAVs) facilitate information collection greatly in the Internet-of-Things (IoT) systems due to their superior flexibility and mobility. On the other hand, nonorthogonal multiple access (NOMA) is regarded as a promising technology to provide high spectral efficiency and support massive connectivity in fifth-generation networks. The integration of NOMA into UAV-assisted wireless networks shows great potential, but how to determine the user grouping and power allocation in NOMA according to the high mobility of UAV is challenging. In this article, we propose a general NOMA-enabled UAV-assisted data collection (NUDC) protocol to maximize the sum rate of a wireless sensor network (WSN), where the location of UAV, sensor grouping, and power control are jointly considered. Moreover, a joint signal-to-interference ratio (SIR) hypergraph-based grouping and power control (SHG-PC) NOMA scheme is provided to obtain the appropriate sensor grouping and the optimal power control solutions efficiently, in which the hypergraph and the greedy coloring algorithm are exploited to find out the optimized group relationships. Extensive simulation results demonstrate the efficiency of our proposed protocol.

Directional antennas and beamforming techniques that bring promising transmission gain to the wireless links are widely incorporated in the system-level analysis of wireless networks. In many existing studies, the beamforming gain model to calculate the beamforming gain considers beamforming pattern and aligned as well as misaligned cases. However, the channel properties, e.g., the$K$factor and the spatial distribution of multi-paths, are neglected, which could significantly influence the beamforming gain. In this paper, a general beamforming gain model is appropriately defined, while the traditional beamforming gain model is proved to be only a special case in the proposed general model by considering an oversimplified channel with no angular spread. In light of this, expressions of the received signal amplitude and the beamforming gain are rigorously derived for narrowband fading and wideband statistical mmWave channels, respectively. Thorough comparison between the proposed beamforming gain model and the traditional beamforming gain is provided, which demonstrates and validates that the traditional model incorrectly captures the beamforming gain and thereby, leads to inaccurate system-level network analysis. To this end, the effectiveness and importance of the proposed general beamforming gain model are revealed, particularly for millimeter-wave wideband systems.

This paper studies distributed power control for the uplink of multi-cell non-orthogonal multiple access (NOMA) systems. Within a cell, the transmissions of different users are modeled as a Gaussian multiple access channel, treating inter-cell interference as additive Gaussian noise. By analyzing the geometry of the feasible power region, using successive interference cancellation at each base station is proved to be optimal in minimizing the total transmission power of all users under rate constraints. The decoding order at each base station, however, remains to be determined. If the decoding order is decided without base station cooperation, the overall control algorithm is fully distributed but is suboptimal in general. To achieve optimality, a partially distributed algorithm is designed, which requires base stations to exchange control messages. Given any feasible instance, the algorithm is proved to converge to the optimal solution. The performance of the fully distributed and partially distributed power control algorithms is compared by computer simulations. The fully distributed algorithm is nearly optimal in terms of outage probability. When a high data rate is required, the partially distributed algorithm is able to reduce the total power consumption by about 20%.

TeraHertz (THz) communications are envisioned as a promising technology, owing to its unprecedented multi-GHz bandwidth. One fundamental challenge when moving to new spectrum is to understand the science of radio propagation and develop an accurate channel model. In this paper, a wideband channel measurement campaign between 130 GHz and 143 GHz is investigated in a typical meeting room. Directional antennas are utilized and rotated for resolving the multi-path components (MPCs) in the angular domain. With careful system calibration that eliminates system errors and antenna effects, a realistic power delay profile is developed. Furthermore, a combined MPC clustering and matching procedure with ray-tracing techniques is proposed to investigate the cluster behavior and wave propagation of THz signals. In light of the measurement results, physical parameters and insights in the THz indoor channel are comprehensively analyzed, including the line-of-sight path loss, power distributions, temporal and spatial features, and correlations among THz multi-path characteristics. Finally, a hybrid channel model that combines ray-tracing and statistical methods is developed for THz indoor communications. Numerical results demonstrate that the proposed hybrid channel model shows good agreement with the measurement and outperforms the conventional statistical and geometric-based stochastic channel model in terms of the temporal-spatial characteristics.

The Terahertz (THz) band is envisioned as a promising technique to support bandwidth-hungry and secure applications. Although the significant path loss and strong directivity make THz communications secure naturally, the information security is still imperfect at near regions along the beam propagation path. In this paper, a novel distance-adaptive absorption peak modulation (DA-APM) is developed for THz covert communications, by exploring the unique spectrum features of frequency-dependent molecular absorption. Although high-attenuation molecular absorption is unfavored for communications, the main principle to enhance covertness or equivalently, minimize the eavesdroppable distance, is dynamically modulating signals under the molecular absorption peaks in the THz spectrum, where the eavesdroppable distance is defined as the threshold distance within which an eavesdropper can wiretap the transmission. Furthermore, an optimization framework is proposed to minimize the eavesdroppable distance, to which the sub-optimal solutions are derived for the multi-wideband waveform by controlling carrier frequencies, power allocation, and rate distribution on each sub-band. Extensive numerical results show that the THz-spectrum-based DA-APM approach can reduce the eavesdroppable distance by 60% compared with random spectrum selection methods, which significantly reduce the insecure area and enhance the covertness of THz wireless transmission.

Non-orthogonal multiple access (NOMA) is regarded as a promising technology to provide high spectral efficiency and support massive connectivity in 5G systems. In most existing NOMA user grouping approaches, users are grouped into disjoint groups, which may lead to a waste of power resources within each NOMA group. Motivated by this, in this paper we propose a novel generalized user grouping (GuG) concept for NOMA from an overlapping perspective, which allows each user to participate in multiple groups but subject to individual maximum power constraint. In order to achieve effective GuG and maximize the system sum rate, we formulate a joint power control and GuG optimization problem. Then, we address this problem by exploiting the overlapping coalition formation (OCF) game framework, and we further propose an OCF-based algorithm in which each user can be self-organized into a desirable overlapping coalition structure. Simulation results verify the efficiency of GuG in NOMA systems and show that our proposed OCF-based GuG NOMA scheme achieves significant performance gains in terms of system sum rate.

Non-orthogonal multiple access (NOMA) provides high spectral efficiency and supports massive connectivity in 5G systems. Traditionally, NOMA user grouping is non-overlapping, leading to a waste of power resources within each NOMA group. Motivated by this, we propose a novel generalized user grouping (GuG) concept for NOMA from an overlapping perspective, which allows each user to participate in multiple user groups but subject to individual maximum power constraint. We formulate a joint power control and GuG optimization problem, and then provide a machine learning-based GuG scheme to obtain the optimized feasible GuG and the optimal power control solutions efficiently. Simulation results show significant performance gains in terms of system sum rate.

How to predict the wireless network level performance such as the network capacity, the average user data rate, and the 5%-tile user data rate is a million-dollar question. In the literature, some pioneering works have been proposed by exploiting either the information theoretic techniques on the physical layer (PHY) information or the Markov chain techniques on the multiple access control (MAC) layer information. However, since these mathematical model-driven approaches usually focus on a small part of the network structure, they cannot characterize the whole network performance. In this paper, we propose to utilize a data-driven machine learning approach to tackle this problem. More specifically, both PHY and MAC information is fed into a deep neural network (DNN) specifically designed for network-level performance prediction. Simulation results show that the network level performance can be accurately predicted at the cost of higher computational complexity.

The collision channel without feedback (CCw/oFB) introduced by Massey and Mathys, depicts a scenario where multiple users share a communication channel but have arbitrary time offsets, and can never learn these time offsets due to the lack of feedback. This paper considers an extension of the CCw/oFB, which allows the receiver to use successive interference cancellation (SIC) to cancel the interference caused by those collided packets whose contents have been known by the receiver. We derive the zero-packet-loss throughput regions of this model for both the unsynchronized and slot-synchronized cases. Given an arbitrary number of users and a packet alphabet of arbitrary size, it is shown that these two regions coincide, and the outer boundary of this common region is the set of all points with only nonnegative components that add up to one. It is further shown that all points on this outer boundary with only rational components can be achieved without packet loss in the slot-synchronized case. The constructive proofs are based on a joint design of protocol sequences, identification/location algorithm and erasure correcting codes. These findings indicate that the negative impact of the lack of time synchronization on the throughput performance can be removed by the help of SIC.

In this paper, the V2I channel in a typical urban scenario is analyzed by the ray tracing technique at 110 GHz. The channel statistics, including the path loss, time-of-arrival (ToA), and direction-of-arrival (DoA) are thoroughly characterized. To capture the channel non- stationarity, a continuous-time birth and death (B-D) process is utilized to model the dynamic behavior of multiple path components (MPCs). On the basis of the channel statistics from the ray tracing combining the B-D process, a geometric-based stochastic time- varying model (GBSTM) is developed for the low-THz urban V2I scenario, which considers the spherical wave propagation and specular MPCs. The closed-form expressions for the critical statistical quantities of non-stationary channels, e.g., the four- dimensional (4D) time-frequency correlation function (TF-CF) and time- frequency dependent power delay profile (TF-PDP) are derived. The analysis reveals that spherical wave propagation will cause non- linear temporal-spatial correlation and extra non- stationarity of the channel in the time domain.

Covert communication, aiming at concealing the existence of data transmission from an eavesdropper, is attracting increasing concerns for communication security. With the trend of moving to higher carrier frequencies, Terahertz (THz) band communication, i.e., wavelength at 0.03-3mm, is envisioned as a promising technique to support bandwidth-hungry applications, as well as improve physical layer security due to its naturally strong directivity and high path loss. Narrow-beam transmission widely adopted in THz communications can effectively prevent from eavesdropping outside the beam sector. However, the challenge still remains when an eavesdropper resides inside the beam sector. To ensure the covertness under such condition, in this paper, a distance-adaptive absorption peak hopping (DA-APH) modulation scheme is proposed for THz covert communications, which takes advantages of distance- and frequency-selective spectral windows, and the frequency-hopping mechanism over the THz spectrum. In particular, the pulse waveform model with polarization is developed to guarantee the reliability of transmission and covertness from eavesdropping. Furthermore, a distance-adaptive carrier frequency selection scheme is designed to choose optimal hopping frequencies at molecular absorption peaks in the THz band.

Wireless networks-on-chip (WiNoC) communications are envisioned as a promising technology to support the interconnection of hundreds of cores in the chip multi-processor design. To meet the future demand for Tera-bit-per-second (Tbps) ultra-fast links in the WiNoC, the millimeter wave (mmWave) and Terahertz (THz) bands with ultra-broad spectrum resource are proposed for WiNoC communications. In this paper, a hybrid WiNoC architecture and the stratified chip design are described, in which the flip-chip package and heat sink are considered. The electromagnetic fields in the WiNoC stratified medium are theoretically analyzed and verified with full-wave simulation. Based on the developed channel model, the WiNoC propagation is characterized by analyzing the path loss, the channel capacity and the reliability. Furthermore, the impact and guideline of the chip design on the WiNoC wave propagation are extensively evaluated and investigated. In particular, the wave propagation performance in THz WiNoC channel can be improved, by decreasing the underfill thickness, proper choice of the silicon thickness and inserting a bottom layer below the silicon substrate.

In this paper, we investigate the feasible signal-to-interference-plus-noise-power-ratio (SINR) region of a multi-cell downlink non-orthogonal multiple access (NOMA) system with successive interference cancellation (SIC) technique. Based on the Perron-Frobenius Theory, we first derive a necessary and sufficient condition for an SINR vector to be feasible under a fixed SIC decoding order. The feasible SINR region of the system is then given by the union of the feasible SINR regions under all possible decoding orders. Next, we derive a necessary condition for an optimal SIC decoding order in the sense that it can achieve the SINR region boundary. Based on this condition, we further propose an efficient algorithm that can approximate the SINR region boundary. Simulation results are provided to validate the efficiency of the proposed algorithm and compare the performance of NOMA and orthogonal multiple access.

This paper considers a wireless communication network consisting of multiple interfering multicast sessions. Different from a unicast system where each transmitter has only one receiver, in a multicast system, each transmitter has multiple receivers and broadcasts a common message to all of them. It is a well-known result for wireless unicast systems that the feasibility of a signal-to-interference-plus-noise power ratio (SINR) without power constraint is decided by the spectral radius of a nonnegative matrix. We generalize this result and obtain necessary and sufficient conditions for the feasibility of an SINR in a wireless multicast system with and without power constraint. The feasible SINR region and its geometric properties are studied. Besides, an iterative algorithm is proposed, which can efficiently check the feasibility condition and compute the boundary points of the feasible SINR region.

In this paper, we aim to effectively utilize in-network buffering to schedule and route packets with low communication overhead, small delay and throughput optimality in fat-tree networks. While nearly zero in-network queueing can be guaranteed by performing precise time allocation and path assignment at network endpoints as in Fastpass, there is a high communication overhead, and buffer occupancy at the endpoints can become bottlenecks. By spreading scheduling functionalities to different network layers in a fat-tree network, the complexity can be decreased significantly at the cost of moderate buffer occupancy at intermediate switches. Inspired by this observation, we propose a simple dynamic pod scheduling (DPS) scheme, which performs scheduling at the granularity of pod units, each of which is paired to at most one other pod unit to transmit packets at each slot. By doing so, less information is required to arrange the packet transfers and inter-pod traffic will experience less downlink contentions. Through extensive evaluations, we find that DPS outperforms Fastpass in terms of delay while still guaranteeing throughput-optimality.

Non-orthogonal multiple access (NOMA) is an effective means to improve the spectral efficiency of a wireless communication system. When applied to cellular networks, cell edge users may suffer from low bit rate, or the associated base stations may need to use excessively high power to serve those users. In order to alleviate the problem, this paper considers the integration of NOMA with coordinated transmission techniques. A two-cell system is considered, in which there are two users near their associated base stations and a cell edge user served by both base stations. It is assumed that each user has a data rate requirement, and the system objective is to minimize the total transmit power. With a formal problem formulation, the feasibility of the problem is characterized by using Helly's theorem. When the problem is feasible, we design both centralized and distributed algorithms to solve it. Numerical results show that NOMA can significantly outperform an orthogonal multiple access scheme in terms of power consumption and outage probability.

Protocol sequences are periodic zero-one sequences for the scheduling of packet transmissions in a time-slotted channel. A special class of protocol sequences, called shift-invariant sequences, plays a key role in achieving the information-theoretic capacity of the collision channel without feedback. This class of shift-invariant protocol sequences has the property that the pairwise Hamming crosscorrelation functions are invariant to relative delay offsets. However, the common period of shift-invariant sequences grows exponentially as a function of the number of supported users. In this paper, we consider a family of protocol sequences, whose period increases roughly as a quadratic function of the number of the users, and show that it is close to shift-invariant by establishing a bound on the pairwise Hamming crosscorrelation. The construction is based on the Chinese remainder theorem (CRT), and hence the constructed sequences are called CRT sequences. Applications to collision channel allowing successive interference cancellation at the receiver are discussed.

The collision channel without feedback (CCw/oFB) model depicts a scenario in which multiple users share a communication channel with random relative time offsets among their clocks. This paper considers an extension of this model, which allows the receiver to use successive interference cancellation (SIC) to iteratively cancel the interference caused by those collided packets that have been decoded by the receiver. We derive the zero-error capacity region of this channel in the slot-synchronous case, and present a zero-error capacity achieving scheme by joint protocol sequences and channel coding design. It is shown that the negative impact on the zero-error capacity due to a lack of time synchronization can be removed by SIC.

Data centers play an important role in today's Internet development. Research to find scalable architecture and efficient routing algorithms for data center networks has gained popularity. The fat-tree architecture, which is essentially a folded version of a Clos network, has proved to be readily implementable and is scalable. In this paper, we investigate routing on a fat-tree network by deriving its global packing number and by presenting explicit algorithms for the construction of optimal, load-balanced routing solutions. Consider an optical network that employs wavelength division multiplexing in which every user node sets up a connection with every other user node. The global packing number is basically the number of wavelengths required by the network to support such a traffic load, under the restriction that each source-to-destination connection is assigned a wavelength that remains constant in the network. In mathematical terms, consider a bidirectional, simple graph, G and let N ⊆ V(G) be a set of nodes. A path system P of G with respect to N consists of |N|(|N| -1) directed paths, one path to connect each of the source-destination node pairs in N. The global packing number of a path system P, denoted by Φ(G, N, P), is the minimum integer k to guarantee the existence of a mapping φ : P → (1, 2, ..., k), such that φ(P) ≠ φ(P̅) if P and P̅ have common arc(s). The global packing number of (G, N), denoted by Φ(G, N), is defined to be the minimum Φ(G, N, P) among all possible path systems ?. In additional to wavelength division optical networks, this number also carries significance for networks employing time division multiple access. In this paper, we compute by explicit route construction the global packing number of (Tn, N), where Tndenotes the topology of the n-ary fat-tree network, and N is considered to be the set of all edge switches or the set of all supported hosts. We show that the constructed routes are load-balanced and require minimal link capacity at all network links.

This paper investigates the power control problem for the downlink of a multi-cell non-orthogonal multiple access system. The problem, called P-OPT, aims to minimize the total transmit power of all the base stations subject to the data rate requirements of the users. The feasibility and optimality properties of P-OPT are characterized through a related optimization problem, called Q-OPT, which is constituted by some relevant power control subproblems. First, we characterize the feasibility of Q-OPT and prove the uniqueness of its optimal solution. Next, we prove that the feasibility of P-OPT can be characterized by the Perron-Frobenius eigenvalues of the matrices arising from the power control subproblems. Subsequently, the relationship between the optimal solutions to P-OPT and that to Q-OPT is presented, which motivates us to obtain the optimal solution to P-OPT through solving the corresponding Q-OPT. Furthermore, a distributed algorithm to solve Q-OPT is designed, and the underlying iteration is shown to be a standard interference function. According to Yates's power control framework, the algorithm always converges to the optimal solution if exists. Numerical results validate the convergence of the distributed algorithm and quantify the improvement of our proposed method over fractional transmit power control and orthogonal multiple access schemes in terms of power consumption and outage probability.

This paper investigates the power control problem for the downlink of a non-orthogonal multiple access (NOMA) system with two cells. The problem, called p-Opt, aims to minimizes the total transmit power of the base stations subject to the data rate requirements of the users. The feasibility and optimality properties of p-Opt is first characterized. It is proved that the feasible power region of p-Opt can be represented by the feasible regions of four power control subproblems that constitute a related optimization problem called q-Opt. Furthermore, the optimal solution to p-Opt can be obtained by solving the corresponding instance of q-Opt. A distributed algorithm to solve q-Opt is designed and the underlying iteration is shown to be a standard interference function. According to Yates's power control framework, the algorithm always converges to the optimal solution if exists. Numerical results validate the convergence of the distributed algorithm and quantify the improvement of NOMA over its orthogonal multiple access counterparts in terms of power consumption and outage probability.

Slotted ALOHA can benefit from physical-layer network coding (PNC) by decoding one or multiple linear combinations of the packets simultaneously transmitted in a timeslot, forming a system of linear equations. Different systems of linear equations are recovered in different timeslots. A message decoder then recovers the original packets of all the users by jointly solving multiple systems of linear equations obtained over different timeslots. We propose the batched BP decoding algorithm that combines belief propagation (BP) and local Gaussian elimination. Compared with pure Gaussian elimination decoding, our algorithm reduces the decoding complexity from cubic to linear function of the number of users. Compared with the ordinary BP decoding algorithm for low-density generator-matrix codes, our algorithm has better performance and the same order of computational complexity. We analyze the performance of the batched BP decoding algorithm by generalizing the tree-based approach and provide an approach to optimize the system performance.

We propose a low-complexity digital fountain approach for network-coded multiple access (NCMA), where each source node encodes its input packets using a fountain code. In NCMA, both physical-layer network coding and multiuser decoding are employed in the physical layer of the sink node, so that the output of the physical layer is the coupling of the fountain codes employed at the source nodes. We demonstrate that a belief propagation (BP) decoding algorithm can effectively decode the coupled fountain codes to recover the input packets of all source nodes. Our approach significantly reduces the decoding complexity compared with the previous NCMA schemes based on Reed-Solomon codes and random linear codes, and hence has the potential to increase throughput and decrease delay in computation-limited NCMA systems.

This paper derives respective precise bit error probability (BEP) expressions for a two-user binary phase shift keying (BPSK) system in Rayleigh, Nakagami and Rician fading channels. Our expressions allow for different symbol rate and symbol timing asynchronism between the desired user and interfering user. We provide some theoretical results concerning the BEP performance with respect to the fading severity. Comprehensive simulation study and comparison of the BEP performance between the Gaussian and non-Gaussian interference models are also provided. The results show that the Gaussian interference model has limitation in predicting the exact BEP performance in fading channels. It fails in accurately tracking the variation of the BEP with respect to the signal-to-noise ratio (SNR), signal-to-interference ratio (SIR), symbol rate ratio and fading severity.

In some real-time networked control applications, information between sensors and controllers is exchanged over a shared wireless channel. One key issue is to manage multiple access to the shared medium to accomplish different control tasks. In the paper, a protocol sequence based media access control (MAC) design is presented for networked control systems (NCSs). It is challenging to design an optimal or suboptimal controller since the sensor packets and control packets could be lost in unreliable wireless networks. An ad hoc and efficient control policy is presented. Numerical results illustrate that the cost performance of the protocol sequence based NCS is much better than that of the π-persistent random access based NCS.

This paper investigates a wireless communication system where the mutual user interference is not assumed to be a Gaussian process. We derive an exact expression for the average bit error probability (BEP) for such a system and study the non-Gaussian interference model through two types of power control problems. We analyze the situation under which the system can be asymptotically error-free, the behavior of users' BEP when scaling up a fixed power setting by a uniform scalar and the effect of varying symbol rate on the system performance. Our work shows that the non-Gaussian model has significantly different performance characteristics from the traditional Gaussian interference model. Simulations also show that the Gaussian model is generally pessimistic in comparison with the non-Gaussian model.

This paper investigates the power control problem involving a small number of active users whereby the standard Gaussian interference noise assumption does not hold. The model also allows for different user transmission rates. We analyze the situation under which the system can be asymptotically error-free. Subsequently, we formulate a power control optimization problem and propose an iterative descent algorithm for solution. We prove that under suitable conditions, the power control optimization problem has a unique solution which is achieved by the proposed algorithm. Simulations are carried out and compared with the power control results under the classical Gaussian assumption.