Production Control in a Failure-Prone Manufacturing System: Diffusion Approximation and Asymptotic Optimality

Cell-free massive multiple-input multiple-output (MIMO) is a novel beyond 5G (B5G) and 6G paradigm that, through the use of a common central processing unit (CPU), coordinates a large number of distributed access points (APs) to coherently serve mobile stations (MSs) on the same time/frequency resource. By exploiting the characteristics of new less-congested millimeter wave (mmWave) frequency bands, these networks can improve the overall system spectral and energy efficiencies by using low-complexity hybrid precoders/decoders. For this purpose, the system must be correctly dimensioned to provide the required quality of service (QoS) to MSs under different traffic load conditions. However, only heavy traffic load conditions are usually taken into account when analysing these networks and, thus, many APs might be underutilized during low traffic load periods, leading to an inefficient use of resources and waste of energy. Aiming at the implementation of energy-efficient AP switch on/off strategies, several approaches have been proposed in the literature that only consider rather unrealistic uniform spatial traffic distribution in the whole coverage area. Unlike prior works, this paper proposes energy efficient AP sleep-mode techniques for cell-free mmWave massive MIMO networks that are able to capture the inhomogeneous nature of spatial traffic distribution in realistic wireless networks. The proposed framework considers, analyzes and compares different AP switch ON-OFF (ASO) strategies that, based on the use of goodness-of-fit (GoF) tests, are specifically designed to dynamically turn on/off APs to adapt to both the number and the statistical distribution of MSs in the network. Numerical results show that the use of properly designed GoF-based ASO strategies under a non-uniform spatial traffic distribution can serve to considerably improve the achievable energy efficiency.

A singular stochastic control problem with state constraints in two-dimensions is studied. We show that the value function is $C^1$ and its directional derivatives are the value functions of certain optimal stopping problems. Guided by the optimal stopping problem, we then introduce the associated no-action region and the free boundary and show that, under appropriate conditions, an optimally controlled process is a Brownian motion in the no-action region with reflection at the free boundary. This proves a conjecture of Martins, Shreve and Soner [SIAM J. Control Optim. 34 (1996) 2133--2171] on the form of an optimal control for this class of singular control problems. An important issue in our analysis is that the running cost is Lipschitz but not $C^1$. This lack of smoothness is one of the key obstacles in establishing regularity of the free boundary and of the value function. We show that the free boundary is Lipschitz and that the value function is $C^2$ in the interior of the no-action region. We then use a verification argument applied to a suitable $C^2$ approximation of the value function to establish optimality of the conjectured control.

We propose the Lagrangian relaxation power and rate control (LRPRC) algorithm, which achieves relatively higher average system throughput, but with substantially lower power consumption and comparable outage probability when compared to two previously published schemes in S. L. Kim, et al. (Sept. 1999). We find that these are achieved by lowering the data rates of users with unfavorable channel conditions, instead of driving their powers to the maximum as in S. L. Kim, et al. (Sept. 1999). We conclude that if power efficiency is a major concern, the LRPRC is a relatively strong alternative among the three schemes, for light to heavy traffic load conditions.

Personal Communication Networks (PCNs) have become an important filed of activity and they are expected to play a fundamental role in next-generation telecommunication systems. In particular, the emerging personal communication services, such as portable computing, paging, personal e-mail, etc. have led to the need of an efficient integration of voice and data traffic in a same PCN by means of a suitable multiple-access scheme. This paper considers stability and optimization issues for the Packet Reservation Multiple Access (PRMA) protocol used for integrating voice and burst data traffic in a PCN. An important result here devised is that the optimized PRMA protocol permits to be achieved good performance even under heavy traffic load conditions.

The combination of user-centric network densification and distributed massive multiple-input multiple-output (MIMO) operation has recently brought along a new paradigm in the wireless communications arena, referred to as cell-free massive MIMO networking. In these networks, a large number of distributed access points (APs), coordinated by a central processing unit (CPU), cooperate to coherently serve a large number of mobile stations (MSs) in the same time/frequency resource. Similar to what has been traditionally done with conventional cellular networks, cell-free massive MIMO networks will be dimensioned to provide the required quality of service (QoS) to MSs under heavy traffic load conditions, and thus they might be underutilized during low traffic load periods, leading to an inefficient use of both spectral and energy resources. Aiming at the implementation of green cell-free massive MIMO networks, this paper proposes and analyzes the performance of different AP switch ON/OFF (ASO) strategies designed to dynamically turn ON/OFF some of the APs based on the number and/or location of the active MSs in the network. The proposed framework considers line-of-sight (LOS) and non-line-of-sight (NLOS) links between APs and MSs, the use of different antenna array architectures at the access points (APs), suitably characterized by array-dependent spatial correlation matrices, and specific power consumption models for APs, MSs and fronthaul links between the APs and the CPU. Numerical results show that the use of properly designed ASO strategies in cell-free massive MIMO networks clearly improve the achievable energy efficiency. Moreover, they also reveal the existing trade-offs among the achievable energy efficiency, the available network-state information, and the hardware configuration (i.e., number of APs, number of transmit antennas per AP, and number of MSs).

We consider a single-server queueing system with a doubly stochastic Poisson arrival flow under heavy traffic conditions. We prove the convergence of the limiting stationary or periodic distribution to the exponential distribution. In a scheme of series, we consider the C-convergence of the waiting time process to a diffusion process with constant coefficients and reflection at the zero boundary. Examples of computation of the diffusion coefficient in terms of characteristics of the arrival flow and service time are given.

This study aimed to assess operational and safety issues related to rural crossing mechanisms: a “traditional” diamond interchange design with roundabout intersections at the ramp termini, and a less conventional design incorporating local roads, called an extended diamond interchange. Various levels of traffic demand were considered through computer simulation. The primary safety metrics were the speed differences between an approaching link and a circulating link, and the speed differences between consecutive time windows on an approaching link. In both light and heavy traffic conditions, the extended diamond interchange with roundabouts including local roads seems to be safer, by these metrics, than the typical design because the longer ramp to roundabouts makes traffic flow more stable. For the operational performance, total delay time was surveyed. In heavy traffic conditions, the extended design is stronger than the typical design thanks to more space for vehicle queues.

Vehicular ad hoc networks (VANETs) enable higher safety, enhanced mobility management, and new infotainment services. Currently, the foreseen standard at the medium access control (MAC) layer for VANETs is IEEE 802.11p, which is based on carrier sense multiple access with collision avoidance (CSMA/CA). However, under heavy traffic conditions, CSMA/CA suffers from a high collision probability, particularly in the presence of hidden terminals. Furthermore, the adoption of the request-to-send/clear-to-send (RTS/CTS) mechanism is not effective when a high data rate is required. If high-throughput services are addressed, a new MAC protocol should, thus, be designed. To this aim, in this paper, we propose a new protocol, which is denoted as the orthogonal frequency-division multiple-access (OFDMA)-based MAC protocol for VANETs (OBV), and we compare it with other MAC protocols taken as benchmarks. To verify the feasibility and the performance of the proposed algorithm, we first propose an analytical model in a simplified scenario. Then, we develop exhaustive simulations in realistic scenarios, considering both urban and highway environments. Results show that OBV outperforms all reference protocols, even doubling their throughput under heavy-load network conditions.

Understanding the in-cabin microenvironment of vehicles is important for assessing human exposure to ultrafine particles (UFPs) of vehicular origin. Filtration through the cabin filter is one of the processes that determine the ratio of in-cabin to on-roadway (I/O) UFP concentrations. In this study, two filter test systems were used to measure the particle filtration efficiencies of fine, ultrafine, and coarse particles. Two types of particles (diesel exhaust UFPs and Arizona test particles) were used to represent the particle types expected in the on-roadway environment. The most penetrating particle size was around 300 nm with filtration efficiency lower than 20%. As the filter face velocity increased from 0.1 to 0.5 m s−1, the filtration efficiency decreased by 10–20%. For vehicles that were frequently driven under heavy traffic conditions (65,000–72,000 vehicles day−1) the pressure drop across the cabin filter increased up to 45 Pa within 10 months. It took 20 months to achieve the same pressure drop under moderate traffic conditions (10,000–24,000 vehicles day−1) and 30 months under light conditions (700–2,000 vehicles day−1). When the vehicle ventilation fan was on and the recirculation was off, it took approximately 10 months under heavy traffic conditions for UFP I/O ratios to increase by 40%. Explicit relationships between UFP I/O ratios and filter usage under various conditions were derived to facilitate cabin filter change decisions based on individual preferences.

SummaryA simulation‐based optimization is a decision‐making tool that helps in identifying an optimal solution or a design for a system. An optimal solution and design are more meaningful if they enhance a smart system with sensing, computing, and monitoring capabilities with improved efficiency. In situations where testing the physical prototype is difficult, a computer‐based simulation and its optimization processes are helpful in providing low‐cost, speedy and lesser time‐ and resource‐consuming solutions. In this work, a comparative analysis of the proposed heuristic simulation‐optimization method for improving quality‐of‐service (QoS) is performed with generalized integrated optimization (a simulation approach based on genetic algorithms with evolutionary simulated annealing strategies having simplex search). In the proposed approach, feature‐based local (group) and global (network) formation processes are integrated with Internet of Things (IoT) based solutions for finding the optimum performance. Further, the simulated annealing method is applied for finding local and global optimum values supporting minimum traffic conditions. A small‐scale network of 50 to 100 nodes shows that genetic simulation optimization with multicriteria and multidimensional features performs better as compared to other simulation‐optimization approaches. Further, a minimum of 3.4% and a maximum of 16.2% improvement is observed in faster route identification for small‐scale IoT networks with simulation‐optimization constraints integrated model as compared to the traditional method. The proposed approach improves the critical infrastructure monitoring performance as compared to the generalized simulation‐optimization process in complex transportation scenarios with heavy traffic conditions. The communicational and computational‐cost complexities are least for the proposed approach.

Closed Circuit Television (CCTV) systems are widely used in monitoring and security surveillance applications to assess status and implement measures necessary to address problems and concerns. CCTV nowadays are visible in roads for monitoring and analyzing traffic behavior and conditions. Multiple cameras are utilized to capture the different angles of the road. This is useful in improving traffic management systems, determination of road traffic density, accident reviews, and in some advanced applications, contactless apprehension. Small to medium scale community areas like industrial parks, villages, hospitals, and even academic campuses require traffic monitoring systems. In this study, a network of IP-CCTV cameras was designed to capture vehicular movement, density, and road condition in a campus setting. The network is composed of eight IP-CCTV cameras processed by four Raspberry Pi computer boards in a 2:1 camera-to-computer ratio. A graphical user interface displays the video feed of the cameras, time customizable traffic report, and the road map visualization and notifications. All computer boards can send and receive data and can create visual traffic maps displayed in the user interface. Color-coding is used in the road segments to indicate light, moderate, or heavy traffic conditions. The vehicle detection accuracy of the system is 93% while its status notification accuracy is 84%. In a campus-based application, especially those with medical and health research institutes, this model suffices its requirements in monitoring, analyzing, basis for emergency rerouting, and improvements in traffic management.

Effect of audio-visual interaction on soundscape in the urban residential context: A virtual reality experiment

The field of autonomous driving has seen increasing proposed use of machine learning methodologies. However, there are still challenges in applying such methods since autonomous driving involves complex and dynamic interactions with the environment. Supervised learning algorithms such as imitation learning can work in environments represented in the training data set, however, it is impractical or cost-prohibitive to collect data for all possible environments. Reinforcement learning methods can train the agent through trial and error, but it may still fail in new environments. To overcome these shortcomings, we thus propose a meta reinforcement learning (MRL) method to improve the agent's generalization capabilities to new environments. The method is applied to automated lane-changing maneuvers at different traffic congestion levels. Specifically, we train the model at light to moderate traffic conditions under a reinforcement learning framework, and then test it at heavy traffic conditions that are never encountered during training. For performance evaluation, we use both collision rate and success rate of the lane-change maneuvers to quantify the safety and effectiveness of the proposed model. A pretrained model is established as a benchmark, which uses the same network structure and training tasks as our proposed model for fair comparison. Simulation results show that the proposed method achieves an overall success rate up to 20% higher than the benchmark model when it is generalized to the new environment of heavy traffic density. The collision rate is also reduced by up to 18% compared with the benchmark model. Furthermore, the proposed model shows more stable and efficient adaptation capabilities and can achieve 100% successful rate and 0% collision rate with only a few steps of gradient update adaptation.

In mobile ad hoc networks (MANETs), the estimation of packet end-to-end delay depends on that of one-hop packet delay. In this paper, we conduct an analysis of the one-hop packet delay in MANETs, where the medium access control (MAC) layer uses the IEEE 802.11 distributed coordination function (DCF) to share the medium. In the MANET, each node runs the IEEE 802.11 DCF and a routing protocol. It is assumed that all nodes are one-hop neighbors, and that any pair of nodes can send data over the wireless channel with a fixed data rate. The light traffic condition is used, i.e., each node generates packets at the network layer according to a Poisson process. By modeling each wireless node as an M/M/1 queueing system, we derive the mean one-hop packet delay analytically under the light traffic condition. Simulation analysis is carried out to verify the derived results. Results show that the mean one-hop packet delay increases with either the network size or the packet generation rate in networks subject to the light traffic condition. The mean one-hop packet delay derived in this paper is analytical and exact for networks under the light traffic condition. Results that can be found in the literature are usually based on the heavy traffic condition, and they tend to overestimate by a large amount the mean one-hop delay for networks with light traffic.

Performance analysis of broadcast star network with collision-avoidance switch