Conventional time-series forecasting methods typically aim to minimize overall prediction error, without accounting for the varying importance of different forecast ranges in downstream applications. We propose a training methodology that enables forecasting models to adapt their focus to application-specific regions of interest at inference time, without retraining. The approach partitions the prediction space into fine-grained segments during training, which are dynamically reweighted and aggregated to emphasize the target range specified by the application. Unlike prior methods that predefine these ranges, our framework supports flexible, on-demand adjustments. Experiments on standard benchmarks and a newly collected wireless communication dataset demonstrate that our method not only improves forecast accuracy within regions of interest but also yields measurable gains in downstream task performance. These results highlight the potential for closer integration between predictive modeling and decision-making in real-world systems.

In this paper, we study network reliability in relation to a periodic time-dependent utility function that reflects the system’s functional performance. When an anomaly occurs, the system incurs a loss of utility that depends on the anomaly’s timing and duration. We analyze the long-term average utility loss by considering exponential anomalies’ inter-arrival times and general distributions of maintenance duration. We show that the expected utility loss converges in probability to a simple form. We then extend our convergence results to more general distributions of anomalies’ inter-arrival times and to particular families of non-periodic utility functions. To validate our results, we use data gathered from a cellular network consisting of 660 base stations and serving over 20k users. We demonstrate the quasi-periodic nature of users’ traffic and the exponential distribution of the anomalies’ inter-arrival times, allowing us to apply our results and provide reliability scores for the network. We also discuss the convergence speed of the long-term average utility loss, the interplay between the different network’s parameters, and the impact of non-stationarity on our convergence results.

This paper considers an anomaly detection problem in which a detection algorithm assigns anomaly scores to multi-dimensional data points, such as cellular networks' Key Performance Indicators (KPIs). We propose an optimization framework to refine these anomaly scores by leveraging side information in the form of a causality graph between the various features of the data points. The refinement block builds on causality theory and a proposed notion of confidence scores. After motivating our framework, smoothness properties are proved for the ensuing mathematical expressions. Next, equipped with these results, a gradient descent algorithm is proposed, and a proof of its convergence to a stationary point is provided. Our results hold (i) for any causal anomaly detection algorithm and (ii) for any side information in the form of a directed acyclic graph. Numerical results are provided to illustrate the advantage of our proposed framework in dealing with False Positives (FPs) and False Negatives (FNs). Additionally, the effect of the graph's structure on the expected performance advantage and the various trade-offs that take place are analyzed.

The energy consumption of the fifth generation (5G) of cellular technology is concerning for the mobile industry and the entire society. To minimize the environmental footprint and economic costs of 5G, it is necessary to adapt the transmission capabilities of networks to end-users’ quality of service requirements. In this paper, we focus on the carrier shutdown approach that enables a base station (BS) to autonomously switch off during low traffic periods, by transferring its load to neighbouring active BSs. More specifically, we propose a data-driven framework, constructed through real network measurements, which statistically characterizes the user equipment (UE) transfer across neighbouring BSs, when carrier shutdown operates. The implementation of this framework allows the 5G system to determine a poor load distribution due to energy saving mechanisms, prevent drastic reductions in UE performance, and ultimately estimate energy savings when activating carrier shutdown.

We consider an antenna tuning problem that is quite important to ensure a satisfying user experience in wireless networks. We aim to maximize the coverage ratio of a large service region by properly choosing the antenna angles. In order to embrace the true complexity of the practical networks and the radio channels, a system level simulator is required. The optimization algorithm has to be performed based on the stochastic numerical output of the simulator. We proposed a zeroth order continuation method to solve the challenging stochastic black-box non-convex optimization problem. The basic idea is to use two observations of the simulator output to generate a gradient estimator, that can be applied to optimize the smoothed version of the original objective function. We optimize a series of smoothed functions to make the solution progressively closer to the global optimum. The performance guarantee of the proposed algorithm has been investigated under weaker assumptions compared to those of the state-of-art analysis. Although the proposed algorithm can be applied to general problems, we perform simulations considering an ideal network model and present numerical results to corroborate our claim.

We consider a distributed stochastic optimization problem in networks with finite number of nodes. Each node adjusts its action to optimize the global utility of the network, which is defined as the sum of local utilities of all nodes. While Gradient descent method is a common technique to solve such optimization problem, the computation of the gradient may require much information exchange. In this paper, we consider that each node can only have a noisy numerical observation of its local utility, of which the closed-form expression is not available. This assumption is quite realistic, especially when the system is either too complex or constantly changing. Nodes may exchange partially the observation of their local utilities to estimate the global utility at each timeslot. We propose a distributed algorithm based on stochastic perturbation, under the assumption that each node has only part of the local utilities of the other nodes. We use stochastic approximation tools to prove that our algorithm converges almost surely to the optimum, given that the objective function is smooth and strictly concave. The convergence rate is also derived, under the additional assumption of strongly concave objective function. It is shown that the convergence rate scales as O(K-0.5) after a sufficient number of iterations K > K0, which is the optimal rate order in terms of K for our problem. Although the proposed algorithm can be applied to general optimization problems, we perform simulations for a typical power control problem in wireless networks and present numerical results to corroborate our claims.

This paper concentrates on the study of finite time control for nonlinear switched systems. Based on a newly introduced adding a barrier integrator technique, a novel adaptive fuzzy control strategy is proposed for a class of nonlinear switched systems. Compared with the existing adaptive control methods, the proposed method has several distinguishing features. Finite time control: the proposed adaptive control method can solve the exact finite time control problem for the stabilization and some types of tracking issues. Namely, the errors will converge to zero in finite time. For a general tracking problem, the practical finite time control can be achieved. More general systems: the proposed method is suitable for high order nonlinear switched systems with arbitrary switching and unknown control gains. Some strict assumptions on the system dynamics are relaxed. Full state constraints: the proposed method can be utilized to deal with the full state constraints problem. Simple controller structure: the "explosion of the complexity" in the backstepping design is avoided. Singularity free design: the singularity problem is carefully handled during the whole design procedure. Examples are presented to illustrate the effectiveness of the proposed method.

Propagation of faulty data is a critical issue. In case of Delay Tolerant Networks (DTN) in particular, the rare meeting events require that nodes are efficient in propagating only correct information. For that purpose, mechanisms to rapidly identify possible faulty nodes should be developed. Distributed faulty node detection has been addressed in the literature in the context of sensor and vehicular networks, but already proposed solutions suffer from long delays in identifying and isolating nodes producing faulty data. This is unsuitable to DTNs where nodes meet only rarely. This paper proposes a fully distributed and easily implementable approach to allow each DTN node to rapidly identify whether its sensors are producing faulty data. The dynamical behavior of the proposed algorithm is approximated by some continuous-time state equations, whose equilibrium is characterized. The presence of misbehaving nodes, trying to perturb the faulty node detection process, is also taken into account. Detection and false alarm rates are estimated by comparing both theoretical and simulation results. Numerical results assess the effectiveness of the proposed solution and can be used to give guidelines for the algorithm design.

This paper considers a delay tolerant network consisting of nodes equipped with sensors, some of them producing outliers. A distributed faulty node detection (DFD) algorithm, whose aim is to help each node in estimating the status of its sensors, has been proposed recently by the authors. The aim of this paper is to analyze the robustness of the DFD algorithm to the presence of misbehaving nodes performing Byzantine attacks. Two types of attacks are considered and analyzed, each trying to mislead the other nodes in the estimation of the status of their sensors. This provides insights on the way the parameters of the DFD algorithm should be adapted to minimize the impact of misbehaving nodes. Theoretical results are illustrated with simulations considering nodes with random displacements, as well as traces of node inter-contact times from real databases.

This work analyses the impact of channel access issues and packet losses on a distributed defective sensor detection algorithm. A theoretical analysis is performed to characterize the detection performance. Matlab simulation results for the detection algorithm are then provided. Finally, experimental results conducted on a real wireless sensor network involving the IEEE 802.15.4 standard are reported.

Due to their inherent feature of exhibiting frequent disconnections, propagation of faulty data in Delay Tolerant Networks can be a critical aspect to counteract. Indeed the rare meeting events require that nodes are effective and efficient in propagating the correct information. Accordingly mechanisms to rapidly identify possible faulty or misbehaving nodes should be searched. Distributed fault detection has been addressed in the literature in the context of sensor and vehicular networks, but unfortunately these solutions suffer for long delays in identifying and isolating misbehaving nodes. In this paper instead we propose a fully distributed, easily implementable, and fast convergent approach to allow each DTN node to rapidly identify whether its sensors are producing outliers. The behavior of the proposed algorithm is described by some continuous-time state equation, whose equilibrium is characterized. Detection and false alarm rates are estimated by comparing both theoretical and simulation results. Numerical results assess the effectiveness of the proposed solution and can give guidelines in the design of the algorithm.

This paper addresses the information theoretical analysis of data compression achieved by random linear network coding in wireless sensor networks. A sparse network coding matrix is considered with columns having possibly different sparsity factors. For stationary and ergodic sources, necessary and sufficient conditions are provided on the number of required measurements to achieve asymptotically vanishing reconstruction error. To ensure the asymptotically optimal compression ratio, the sparsity factor can be arbitrary close to zero in absence of additive noise. In presence of noise, a sufficient condition on the sparsity of the coding matrix is also proposed.

To guarantee its integrity, a wireless sensor network needs to efficiently detect faulty nodes producing erroneous measurements. This paper proposes a fully distributed fault detection algorithm. A node first collects the measurements of its neighborhood, processes them to decide whether they contain outliers, and broadcasts the result. Then, it decides autonomously about its functioning status. The detection algorithm is proposed in two variants, depending on the proportion of faulty nodes in the network. A theoretical analysis of the probability of error and of the convergence of the algorithm is provided. The tradeoff between false alarm probability and detection probability is characterized using simulation.