154092 research outputs found
Sort by
Machine Learning Predictions for Traffic Equilibria in Road Renovation Scheduling
Get PDFAccurately estimating the impact of road maintenance schedules on traffic conditions is important because maintenance operations can substantially worsen congestion if not carefully planned. Reliable estimates allow planners to avoid excessive delays during periods of roadwork. Since the exact increase in congestion is difficult to predict analytically, traffic simulations are commonly used to assess the redistribution of the flow of traffic. However, when applied to long-term maintenance planning involving many overlapping projects and scheduling alternatives, these simulations must be run thousands of times, resulting in a significant computational burden. This paper investigates the use of machine learning-based surrogate models to predict network-wide congestion caused by simultaneous road renovations. We frame the problem as a supervised learning task, using one-hot encodings, engineered traffic features, and heuristic approximations. A range of linear, ensemble-based, probabilistic, and neural regression models is evaluated under an online learning framework in which data progressively becomes available. The experimental results show that the Costliest Subset Heuristic provides a reasonable approximation when limited training data is available, and that most regression models fail to outperform it, with the exception of XGBoost, which achieves substantially better accuracy. In overall performance, XGBoost significantly outperforms alternatives in a range of metrics, most strikingly Mean Absolute Percentage Error (MAPE) and Pinball loss, where it achieves a MAPE of 11% and outperforms the next-best model by 20% and 38% respectively. This modeling approach has the potential to reduce the computational burden of large-scale traffic assignment problems in maintenance planning
Quantification of the relation between continuous glucose monitoring observation period and the estimation error in assessing long-term glucose regulation
Get PDFIntroduction The integration of continuous glucose monitoring (CGM) into clinical practice has rapidly emerged in the last decade, changing the evaluation of long-term glucose regulation in patients with diabetes. When using CGM-derived metrics to evaluate long-term glucose regulation, it is essential to determine the minimal observation period necessary for a reliable estimate. The approach of this study was to calculate mean absolute errors (MAEs) for varying window lengths, with the goal of demonstrating how the CGM observation period influences the accuracy of the estimation of 90-day glycemic control. Research design and methods CGM data were collected from the DIABASE cohort (ZGT hospital, The Netherlands). Trailing aggregates (TAs) were calculated for four CGM-derived metrics: time in range (TIR), time below range (TBR), glucose management indicator (GMI) and glycemic variability (GV). Arbitrary MAEs for each patient were compared between the TAs of window lengths from 1 to 89 days and a reference TA of 90 days, which is assumed to reflect long-term glycemic regulation. Results Using 14 days of CGM data resulted in 65% of subjects having their TIR estimation being below a MAE threshold of 5%. In order to have 90% of the subjects below a TIR MAE threshold of 5%, the observation period needs to be 29 days. Conclusions Although there is currently no consensus on what is an acceptable MAE, this study provides insight into how MAEs of CGM-derived metrics change according to the used observation period within a population and may thus be helpful for clinical decision-making.</p
Albuminuria Responses to Dapagliflozin in Patients With Type 2 Diabetes:A Crossover Trial
Get PDFImportance: Dapagliflozin reduces the urine albumin-to-creatinine ratio (UACR) and estimated glomerular filtration rate (eGFR) decline at a population level, but individuals show a large variation in responses. The n-of-1 trial design allows for direct assessment of treatment effects within an individual, and digital technologies and remote study assessments can reduce clinic visits, ease participant burden, and improve trial efficiency. Objective: To assess individual UACR responses to dapagliflozin treatment in a decentralized clinical trial and the feasibility of remote data collection. Design, Setting, and Participants: This decentralized, randomized, double-blind, placebo-controlled crossover trial using an n-of-1 approach was conducted using data from the Dutch primary and secondary health care systems between May 2021 and September 2022. Participants included adults with type 2 diabetes, a UACR greater than 20 mg/g, and an eGFR greater than 30 mL/min/1.73 m2. Statistical analyses were performed between June and August 2023. Interventions: Participants were assigned to two 1-week treatment periods with dapagliflozin, 10 mg/d, and two 1-week treatment periods with placebo in random order, with 1-week washout periods in between. Main Outcomes and Measures: The primary outcome was the difference in the change in UACR from start to end of treatment between dapagliflozin and placebo in the per-protocol population. A post hoc exploratory analysis assessed the feasibility of remote data collection, including the proportion of urine and capillary blood samples successfully delivered to the central laboratory. Results: In total, 20 participants (mean [SD] age, 64.9 [8.7] years; 17 [85.0%] male) with a mean (SD) eGFR of 70.2 (20.3) mL/min/1.73 m2 and a median UACR of 94.7 (IQR, 29.8-242.6) mg/g were included in the study. They experienced a relative change in UACR with dapagliflozin compared with placebo of -15.1% (95% CI, -28.2% to -3.3%; P =.01). UACR changes showed considerable variation during both dapagliflozin and placebo treatment (first treatment period: median, -12.8% [range, -56.3% to 36.2%] and 2.9% [range, -86.7% to 35.1%], respectively). UACR changes correlated significantly between the first and second dapagliflozin exposure (r = 0.50; P =.03), with no correlation observed between the placebo exposure periods (r = 0.09; P =.69). With regard to remote data collection, 811 of 816 urine samples (99.4%) and 433 of 440 capillary blood samples (98.4%) were successfully delivered to the central laboratory. Conclusions and Relevance: In this crossover trial, individual UACR responses to dapagliflozin reflected a pharmacological response. Remote data collection proved to be reliable, supporting its use in future studies and clinical practice for monitoring individual dapagliflozin responses.</p
GNSS Coexistence Experiments Using a Vibrating Intrinsic Reverberation Chamber
No full textThere is no standardized way of performing radiated global navigation satellite system (GNSS) coexistence tests. ETSI 303413 describes a radiated measurement technique of testing GNSS receivers with embedded antenna in an anechoic chamber but without clear levels, and only for a very narrow frequency range. Furthermore, a wideband noisy amplifier is needed, and thus the requisite for a very specific band-stop filter. This article discusses a method to perform a radiated coexistence measurement on a GNSS antenna and receiver using a reverberation chamber, and specifically a vibrating intrinsic reverberation chamber (VIRC), and for a much wider frequency band. The VIRC is increasing the field strength levels and therefore no noisy amplifier is needed. Based on the test method indicated by ETSI 303413 requiring an anechoic chamber and a GNSS signal generator, a new test setup is proposed here. The test setup is composed–amongst others—of a VIRC and a GNSS repeater. The application of the method in a reverberant electromagnetic environment, i.e., VIRC, introduces the need of defining an alternative way to evaluate the signal reception degradation of the equipment under test. Such an evaluation method is proposed in this article based on empirical test results on a commercial off the shelf GNSS receiver.</p
Examining the applicability of hard data protection law on demographically identifiable information (DII):the case of humanitarian UAV/drone images in Malawi
Get PDFResearch carried out in Northern Malawi concluded that protecting and ensuring responsible use of drone-collected data in disaster-risk areas should be left entirely to the humanitarian organisations collecting these data, with little participation expected of the affected residents, mostly because they will always be coerced into giving up their data in exchange for assistance. One way to guarantee this protection is by applying national rules of data protection law to these drone data processes in the country. However, aerial drone data (e.g. high-resolution images) of a community would usually be demographically identifiable information (DII) which, unlike personally identifiable information (PII), is not substantively regulated by contemporary hard data protection law, i.e. the 2023 Malawi Data Protection Bill (MDPB) which leaves drone DII without any binding regulatory framework and hence less legal protection.Faced with this regulatory obstacle, this paper sets out to propose and evaluate methods through which the data processing principles and rights provided by the MDPB could nevertheless be applied to regulate drone DII collected and processed by humanitarian organisations in Malawi. First, it sought to establish the feasibility of their application among the humanitarian community: to this end, 20 semi-structured interviews were conducted with humanitarian officials operating in the country, with results showing that these officials largely believed the MDPB principles and rights could effectively govern their drone data processes. The paper then proposes some regulatory modifications or ‘nudges’ which, if adopted by Malawian regulators, could probe humanitarian organisations towards applying the MDPB principles and rights to their drone DII processes, and examines how these principles and rights could be reflected in concrete, drone-related internal policies adopted by these organisations
Movement-Robust mmWave VR via Dual-Beam Reception and Predictive Beam Transition
Get PDFThe abundant bandwidth in the mmWave band supports high data rates and low latency communication, making it ideal for delivering realistic and seamless virtual reality experiences. However, a key challenge lies in adapting the mmWave beams to the highly dynamic user movements, which often cause beam misalignment, resulting in signal degradation and potential outages. Additionally, maintaining uninterrupted signal reception during beam re-alignment due to head rotation requires low-overhead and timely beam transitions to prevent signal drops caused by delayed switching. This paper addresses these challenges with a joint solution at both the access point (AP) and head-mounted display (HMD) ends. Specifically, the proposed solution integrates coordinated multi-point networks with dual-beam reception at the HMD to enhance diversity, improve channel gain, and mitigate outages caused by user movement. Evaluation using real HMD movement datasets demonstrates that dual-beam reception within a coordinated multi-AP setup achieves up to a 22.8% improvement in reliability by reducing outage rates compared to single-beam reception. Experimental validation further highlights the effectiveness of combining widely distributed APs with a locally distributed subarray configuration on the HMD, improving angular coverage during head rotations. Furthermore, our predictive beam transition approach anticipates the future beam during user movements, preventing received signal degradation from delayed transitions while reducing overhead by 43.8% compared to exhaustive periodic beam searches.</p
Time-Sensitive Importance Splitting
Get PDFState-of-the-art methods for rare event simulation of non-Markovian models face practical or theoretical limits if observing the event of interest requires prior knowledge or information on the timed behavior of the system. In this paper, we attack both limits by extending importance splitting with a time-sensitive importance function. To this end, we perform backwards reachability search from the target states, considering information about the lower and upper bounds of the active timers in order to steer the generation of paths towards the rare event. We have developed a prototype implementation of the approach for input/output stochastic automata within the Modest Toolset. Preliminary experiments show the potential of the approach in estimating rare event probabilities for an example from reliability engineering
Convergence and Running Time of Time-dependent Ant Colony Algorithms
Get PDFAnt Colony Optimization (ACO) is a well-known method inspired by the foraging behavior of ants and is extensively used to solve combinatorial optimization problems. In this paper, we first consider a general framework based on the concept of a construction graph - a graph associated with an instance of the optimization problem under study, where feasible solutions are represented by walks. We analyze the running time of this ACO variant, known as the Graph-based Ant System with time-dependent evaporation rate (GBAS/tdev), and prove that the algorithm's solution converges to the optimal solution of the problem with probability 1 for a slightly stronger evaporation rate function than was previously known. We then consider two time-dependent adaptations of Attiratanasunthron and Fakcharoenphol's -ANT algorithm: -ANT with time-dependent evaporation rate (-ANT/tdev) and -ANT with time-dependent lower pheromone bound (-ANT/tdlb). We analyze both variants on the single destination shortest path problem (SDSP). Our results show that -ANT/tdev has a super-polynomial time lower bound on the SDSP. In contrast, we show that -ANT/tdlb achieves a polynomial time upper bound on this problem
A Systematic Study on the Design of Odd-Sized Highly Nonlinear Boolean Functions via Evolutionary Algorithms
Get PDFThis paper focuses on the problem of evolving Boolean functions of odd sizes with high nonlinearity, a property of cryptographic relevance. Despite its simple formulation, this problem turns out to be remarkably difficult. We perform a systematic evaluation by considering three solution encodings and four problem instances, analyzing how well different types of evolutionary algorithms behave in finding a maximally nonlinear Boolean function. Our results show that genetic programming generally outperforms other evolutionary algorithms, although it falls short of the best-known results achieved by ad-hoc heuristics. Interestingly, by adding local search and restricting the space to rotation symmetric Boolean functions, we show that a genetic algorithm with the bitstring encoding manages to evolve a -variable Boolean function with nonlinearity 241