Automated warehouses involve robots that move across the floor, avoiding obstacles while remaining connected via an access point (AP) to a central controller that instructs the robots. The complex propagation environment and presence of metallic surfaces results in spotty coverage, which changes over time as the location of stored products and machinery changes. Thus, maintaining an assured connectivity to APs while performing navigation is a challenge, although it is needed to relay local sensor data from the robots to the controller and receive directions from the latter.$\rm{DARWIN}$, involves creating a digital twin of the warehouse for training the robots by jointly optimizing the navigation and avoiding wireless dead-spots.$\rm{DARWIN}$has three key capabilities: First, it captures the features of both physical and RF environments in the digital world. Second, it allows real-time updating of the digital twin if significant disparity is detected compared to the physical environment. Finally, it includes a reinforcement learning algorithm that jointly optimizes navigation and network resource management, while accounting for handover and outage. We validate$\rm{DARWIN}$on an emulation environment consisting of Robot Operating System and Gazebo platforms along with real-world RF measurements. Results reveal that$\rm{DARWIN}$reduces the number of steps by 43% compared to choosing the closest AP, while detecting environmental changes with maximum 96% accuracy to maintain a high-fidelity digital twin.

Deep learning (DL) based resource allocation (RA) has recently gained significant attention due to its performance efficiency. However, most related studies assume an ideal case where the number of users and their utility demands, e.g., data rate constraints, are fixed, and the designed DL-based RA scheme exploits a policy trained only for these fixed parameters. Consequently, computationally complex policy retraining is required whenever these parameters change. In this paper, we introduce a DL-based resource allocator (ALCOR) that allows users to adjust their utility demands freely, such as based on their application layer requirements. ALCOR employs deep neural networks (DNNs) as the policy in a time-sharing problem. The underlying optimization algorithm iteratively optimizes the on-off status of users to satisfy their utility demands in expectation. The policy performs unconstrained RA (URA)–—RA without considering user utility demands–—among active users to maximize the sum utility (SU) at each time instant. Depending on the chosen URA scheme, ALCOR can perform RA in either a centralized or distributed scenario. The derived convergence analyses provide theoretical guarantees for ALCOR’s convergence, and numerical experiments corroborate its effectiveness compared to meta-learning and reinforcement learning approaches.

Robotic factory floors will revolutionize the future of manufacturing and the service industry by automating tasks. However, to fully supplement human effort, these robots will need low-latency, reliable connectivity throughout the work zone through links established by wireless access points (APs). This will allow the robot to assuredly respond to programming directives that rely on the real-time relaying of robot-generated sensor data to the Mobile Edge Computing (MEC) server. In this paper, we propose L-NORM, a multi-AP and multi-robot coordination framework, as a multi-tiered solution for such autonomous edge networks. First, multi-robot motion planning through reinforcement learning occurs at the MEC, using as input multi-modal robot sensor data. Second, multi-AP resource orchestration is performed using another reinforcement learning-based method that maps a subset of available APs to each robot toward meeting their sensor data delivery requirements. Furthermore, we suggest diversity combination of uplink channels with the 802.11ax scheduled access mode that will (i) support high reliability of multi-robot uplink sensor packets and (ii) enable multi-AP coordination, for optimized resource utilization. Through extensive simulation studies, we show that the probability of robot deviation to remain within 0.5 m from its optimal path, is 19% more in L-NORM compared to classical 802.11ax based edge network solution, considering$\sim$1 MB of sensor data per robot.

Industrial Internet of Things (IIoT) use cases have stringent reliability and latency requirements to enable real-time wireless control systems, which are supported by the 5G ultra-reliable low-latency communications (URLLC). However, extremely high quality-of-service (QoS) requirements in 5G URLLC causes huge radio resource consumption and low spectral efficiency, thus limiting network capacity in terms of the number of supported devices. Industrial control applications typically incorporate redundancy in their design and may not always require extreme QoS to achieve the expected control performance. Therefore, we propose both communication-control co-design and dynamic QoS to address the capacity issue for robotic manipulation use cases in 5G-based IIoT. We have developed an advanced co-simulation framework that includes a network simulator, physics simulator, and compute emulator, for realistic performance evaluation of the proposed methods. Through simulations, we show significant improvements in network capacity (i.e., the number of supported URLLC devices), and about 2x gain for the robotic manipulation use case.

Mobile and autonomous robots are among the most critical technology applications requiring wireless connectivity with deterministic performance, including bounded latency with high reliability, even under congested network conditions. Emerging Wireless Time-Sensitive Networking (WTSN) capabilities over Wi-Fi and 5G can enable time synchronization and bounded low latency through time-aware scheduling mechanisms. Such Wireless TSN capabilities have been demonstrated in several industrial/robotic use cases, but under static conditions. Mobility introduces new challenges due to the roaming events and associated network outages owing to signaling between client devices and network infrastructure during these events. In this paper, we show how we can take advantage of the TSN redundancy capability (as defined in the IEEE 802.1CB standard) to eliminate outages or delays due to events like roaming and interference in a mobile robot use case enabled by Wi-Fi 6 TSN. We demonstrate the roaming performance with no delay impact on the applications though simulations of a mobile robot in a factory scenario and experimental results with a mobile robot connected via multiple Wi-Fi 6 radios in a warehouse environment.

We consider the problem of power control in wireless networks, consisting of multiple transmitter-receiver pairs communicating with each other over a single shared wireless medium. To achieve both a high total rate and a level of fairness across users, we formulate a policy optimization problem with constraints on the minimum per-user rate across network configurations with an adaptive slack parameter. To apply unsupervised learning algorithms in the dual domain, we parameterize the power control policy, slack variable, and dual parameters using graph neural networks (GNNs), which leverage the network topology to create a scalable and network-invariant processing architecture. We use a primal- dual algorithm to learn the optimal GNN parameters and demonstrate via numerical simulations the resulting GNNs' success in achieving the right balance between sum- and 5thpercentile rates throughout a range of network configurations.

We consider the problem of resource allocation in large scale wireless networks. When contextualizing wireless network structures as graphs, we can model the limits of very large wireless systems as manifolds. To solve the problem in the machine learning framework, we propose the use of Manifold Neural Networks (MNNs) as a policy parametrization. In this work, we prove the stability of MNN resource allocation policies under the absolute perturbations to the Laplace-Beltrami operator of the manifold, representing system noise and dynamics present in wireless systems. These results establish the use of MNNs in achieving stable and transferable allocation policies for large scale wireless networks. We verify our results in numerical simulations that show superior performance.

We consider the problem of controlling a series of industrial systems, such as industrial robotics, in a factory environment over a shared wireless channel leveraging edge computing capabilities. The wireless control system model supports the offloading of computational intensive functions, such as perception workloads, to an edge server. However, wireless communications is prone to packet loss and latency and can lead to instability or task failure if the link is not kept sufficiently reliable. Because maintaining high reliability and low latency at all times prohibits scalability due to resource limitations, we propose a communication-control co-design paradigm that varies the network quality of service (QoS) and resulting control actions to the dynamic needs of each plant. We further propose a modular learning framework to solve the complex learning task without knowledge of plant or communication models in a series of learning steps and demonstrate its effectiveness in learning resource-efficient co-design policies in a robotic conveyor belt task.

This paper investigates the general problem of resource allocation for mitigating channel fading effects in Free Space Optical (FSO) communications. The resource allocation problem is modeled as the constrained stochastic optimization framework, which covers a variety of FSO scenarios involving power adaptation, relay selection and their joint allocation. Under this framework, we propose two algorithms that solve FSO resource allocation problems. We first present the Stochastic Dual Gradient (SDG) algorithm that is shown to solve the problem exactly by exploiting the strong duality but whose implementation necessarily requires explicit and accurate system models. As an alternative we present the Primal-Dual Deep Learning (PDDL) algorithm based on the SDG algorithm, which parameterizes the resource allocation policy with Deep Neural Networks (DNNs) and optimizes the latter via a primal-dual method. The parameterized resource allocation problem incurs only a small loss of optimality due to the strong representational power of DNNs, and can be moreover implemented without knowledge of system models. A wide set of numerical experiments are performed to corroborate the proposed algorithms in FSO resource allocation problems. We demonstrate their superior performance and computational efficiency compared to the baseline methods in both continuous power allocation and binary relay selection settings.

We consider optimal resource allocation problems under asynchronous wireless network setting. Without explicit model knowledge, we design an unsupervised learning method based on Aggregation Graph Neural Networks (Agg-GNNs). Depending on the localized aggregated information structure on each network node, the method can be learned globally and asynchronously while implemented locally. We capture the asynchrony by modeling the activation pattern as a characteristic of each node and train a policy-based power allocation method. We also propose a permutation invariance property which indicates the transferability of the trained Agg-GNN. We finally verify our strategy by numerical simulations compared with baseline methods.

This paper investigates the optimal resource allocation in free space optical (FSO) fronthaul networks. The optimal allocation maximizes an average weighted sum-capacity subject to power limitation and data congestion constraints. Both adaptive power assignment and node selection are considered based on the instantaneous channel state information (CSI) of the links. By parameterizing the resource allocation policy, we formulate the problem as an unsupervised statistical learning problem. We consider the graph neural network (GNN) for the policy parameterization to exploit the FSO network structure with small-scale training parameters. The GNN is shown to retain the permutation equivariance that matches with the permutation equivariance of resource allocation policy in networks. The primal-dual learning algorithm is developed to train the GNN in a model-free manner, where the knowledge of system models is not required. Numerical simulations present the strong performance of the GNN relative to a baseline policy with equal power assignment and random node selection.

We consider the problem of finding optimal power allocations subject to system constraints in ad-hoc wireless networks. The resulting optimization problem has the form of a constrained learning problem, motivating the use of a function parameterization such as a neural network. While such a policy can be trained with a primal-dual learning method, traditional neural network architectures are unsuitable for execution in wireless networks, as such networks change frequently in practice, rendering the learned neural network ineffective. To learn a transferable policy that can generalize to varying and growing networks, we propose the use of so-called random edge graph neural networks (REGNNs). Such REG-NNs are shown to exhibit an essential permutation invariance property for the power allocation problem that suggest transference capabilities. In numerical simulations, we validate this suggestion by showing how REGNNs trained on a single ad-hoc network outperform baselines in new randomly drawn networks of growing size.

Optimal resource allocation in real-world wireless systems is rather challenging, not only due to the unavailability of accurate statistical channel models, but also because expressions of maximal or achievable information rates are most often unknown, or not adequately precise. Under a modular stochastic functional optimization framework, we propose a new zeroth-order stochastic primal-dual algorithm for completely data-driven, model-free and gradient-free learning of optimal resource allocation policies for ergodic network optimization. Our contribution relies on Gaussian smoothing of the corresponding constrained policy search problem, and on the representation power of universal policy parameterizations, such as Deep Neural Networks (DNNs). Indeed, our simulations demonstrate that DNN-based policies produced by the proposed primal-dual method attain near-ideal performance, based exclusively on limited channel probing, completely bypassing the need for gradient computations, and at the absence of channel or information rate models.

We consider the problem of allocating 5G radio resources over wireless communication links to control a series of independent low-latency wireless control systems common in industrial settings. Each control system sends state information to the base station to compute control signals under tight latency requirements. Such latency requirements can be met by restricting the uplink traffic to a single subframe in each 5G frame, thus ensuring a millisecond latency bound while leaving the remaining subframes available for scheduling overhead and coexisting broadband traffic. A linear assignment problem can be formulated to minimize the expected number of packet drops, but this alone is not sufficient to achieve good performance. We propose an optimal scheduling with respect to a control operation cost that allocates resources based on current control system needs. The resulting control-aware scheduling method is tested in simulation experiments that show drastically improved performance in 5G settings relative to control-agnostic scheduling under the proposed time-sliced frame structure.

Radio on Free Space Optics (RoFSO), as a universal platform for heterogeneous wireless services, is able to transmit multiple radio frequency signals at high rates in free space optical networks. This paper investigates the optimal design of power allocation for Wavelength Division Multiplexing (WDM) transmission in RoFSO systems. The proposed problem is a weighted total capacity maximization problem with two constraints of total power limitation and eye safety concern. The model-based Stochastic Dual Gradient algorithm is presented first, which solves the problem exactly by exploiting the null duality gap. The model-free Primal-Dual Deep Learning algorithm is then developed to learn and optimize the power allocation policy with Deep Neural Network (DNN) parametrization, which can be utilized without any knowledge of system models. Numerical simulations are performed to exhibit significant performance of our algorithms compared to the average equal power allocation.

We consider the problem of allocating radio resources over wireless communication links to control a series of independent low-latency wireless control systems common in industrial settings. Supporting wireless control in time sensitive settings requires fast data rates over wireless links, which comes at the cost of reliability. It is challenging to meet both latency and reliability requirements with an equal or arbitrary allocation of resources. We thus propose a novel control-aware approach to the low-latency scheduling problem in which we incorporate control and channel state information in allocating bandwidth and data rates across the wireless links. Control systems that are in desirable states are given modest requirements on error rates, while systems in undesirable states are given more priority. We derive control-aware packet error rate targets for each system to satisfy stability goals and make scheduling decisions to meet such targets while reducing total transmission time. The resulting control-aware based method is tested in simulation experiments that demonstrate its effectiveness in meeting control-based goals under tight latency constraints relative to control-agnostic scheduling.

We consider the problem of finding optimal resource allocations subject to system constraints in a generic class of problems in wireless communications. These problems are inherently challenging due to functional optimization and potential non-convexities. However, these problems can be observed to take the form of a regression problem, although one in which the statistical loss function appears as a constraint. This motivates the use of machine learning model parameterizations. To apply gradient-based solution algorithms that do not require model knowledge, we convert the constrained optimization problem to an unconstrained one using Lagrangian duality. Despite the non-convexity in the problem, we formally show that the sub-optimality of the dual domain problem is small when the learning parameterization is sufficiently dense. We then present a primal-dual learning algorithm that looks for solutions to the dual problem using model-free gradient estimates. In a numerical simulation, we demonstrate the near-optimality of the proposed model-free algorithm using a neural network parametrization for a capacity maximization problem.

We consider the problem of allocating radio resources over wireless communication links to control a series of independent wireless control systems. Low-latency transmissions are necessary in enabling time-sensitive control systems with high sampling rates to operate over wireless links. Enabling low-latency through fast data rates comes at the cost of reliability in the form of higher packet error rates due to channel noise. However, the impact of such communication link errors on the control system performance depends dynamically on the control system state. We propose a novel control-aware communication design to the low-latency resource allocation problem. In our proposed method, we incorporate both control and channel state information in scheduling transmissions across time slots, frequency bands, and data rates using the next-generation Wi-Fi scheduling architecture. Control systems that are closer to instability or further from a desired range in a given control cycle are given higher packet delivery rate targets to meet. Rather than a simple priority ranking, we derive precise adaptive packet error rate targets for each system needed to satisfy control-specific performance requirements. We use these adaptive rate targets to make scheduling decisions that reduce total transmission time. The resulting control-aware low-latency scheduling (CALLS) method is tested in numerous simulation experiments that demonstrate its effectiveness in meeting control-based goals under tight latency constraints relative to control-agnostic scheduling.

Most second order methods are inapplicable to large scale empirical risk minimization (ERM) problems because both, the number of samples N and number of parameters p are large. Large N makes it costly to evaluate Hessians and large p makes it costly to invert Hessians. This paper propose a novel adaptive sample size second-order method, which reduces the cost of computing the Hessian by solving a sequence of ERM problems corresponding to a subset of samples and lowers the cost of computing the Hessian inverse using a truncated eigenvalue decomposition. Although the sample size is grown at a geometric rate, it is shown that it is sufficient to run a single iteration in each growth stage to track the optimal classifier to within its statistical accuracy. This results in convergence to the optimal classifier associated with the whole set in a number of iterations that scales with $\log(N)$. The use of a truncated eigenvalue decomposition result in the cost of each iteration being of order $p^2$. Theoretical performance gains manifest in practical implementations.

This paper considers the resource allocation problem in wireless systems over an unknown time-varying non-stationary channel. The goal is to maximize a utility function, such as a capacity function, over a set of wireless nodes while satisfying a set of resource constraints. To bypass the need for a model for channel distribution as it varies over time, samples of the channel are taken at every time epoch to estimate the channel. The resulting stochastic optimization problem is converted in its Lagrange dual problem, where the resulting stochastic optimization problem can viewed equivalently as minimizing a certain empirical risk measure, a well-studied problem in machine learning. The second order Newton's method is used to quickly learn statistically approximated optimal resource allocation policies over the sampled dual function as the channel evolves over time epochs. The quadratic convergence rate of Newton is used to establish, under certain conditions on the sampling size and rate of channel variation, an instantaneous learning and tracking of these policies. Numerical simulations demonstrate the effectiveness of the learning algorithm on a low-dimensional wireless capacity maximization problem.

We present an incremental Broyden-Fletcher-Goldfarb-Shanno (BFGS) method as a quasi-Newton algorithm with a cyclically iterative update scheme for solving large-scale optimization problems. The proposed incremental quasi-Newton (IQN) algorithm reduces computational cost relative to traditional quasi-Newton methods by restricting the update to a single function per iteration and relative to incremental second-order methods by removing the need to compute the inverse of the Hessian. A local superlinear convergence rate is established and a strong improvement is shown over first order methods numerically for a set of common large-scale optimization problems.

We present an authorship attribution method based on relational data between function words. These are content independent words that help define grammatical relationships. As relational structures we use normalized word adjacency networks. We interpret these networks as Markov chains and compare them using entropy measures. We illustrate the accuracy of the method developed through a series of numerical experiments including comparisons with frequency based methods. We show that accuracy increases when combining relational and frequency based data, indicating that both sources of information encode different aspects of authorial styles.