Maria Elena Larraga

Adaptive traffic signal control (ATSC) remains a critical challenge for urban mobility. In this direction, deep reinforcement learning (DRL) has been widely investigated for ATSC, showing promising improvements in simulated environments. However, a noticeable gap remains between simulation-based results and practical implementations, due to reward formulations that do not address phase instability. Stochastic variations may trigger premature phase changes (“flickers”), affecting signal behavior and potentially limiting deployment in real scenarios. Although several works have examined delay, queues, and decentralized coordination, stability-focused variables remain comparatively less explored, particularly in single yet complex intersections. This study proposes a decentralized DRL model for ATSC with noise injection (ATSC-DRLNI) applied to a single intersection, introducing a stability-oriented reward function that integrates flickers, queue length, and advantage actor-critic (A2C) learning feedback. The model is evaluated in the Simulation of Urban MObility (SUMO) platform and compared against seven baseline methods, using real traffic data from a Mexican city for calibration and validation. Results suggest that penalizing flickers may contribute to more stable phase transitions, while reductions of up to 40% in queue length were observed in heavy-traffic scenarios. These findings indicate that incorporating stability-related variables into reward functions may help in implementing DRL-based ATSC studies.

Adaptive Traffic Signal Control (ATSC) remains a critical challenge for urban mobility. In this direction, Deep Reinforcement Learning (DRL) has been widely investigated for ATSC, showing promising improvements in simulated environments. However, a noticeable gap remains between simulation-based results and practical implementations, due to reward formulations that do not address phase instability. Stochastic variations may trigger premature phase changes ("flickers”), affecting signal behavior and potentially limiting deployment in real scenarios. Although several works have examined delay, queues, and decentralized coordination, stability-focused variables remain comparatively less explored, particularly in single yet complex intersections. This study proposes a decentralized DRL model for ATSC with Noise Injection (ATSC-DRLNI) applied to a single intersection, introducing a stability-oriented reward function that integrates flickers, queue length, and Advantage Actor-Critic (A2C) learning feedback. The model is evaluated in Simulation of Urban MObility (SUMO) platform and compared against seven baseline methods, using real traffic data from a Mexican city for calibration and validation. Results suggest that penalizing flickers may contribute to more stable phase transitions, while reductions of up to 40\% in queue length were observed in heavy-traffic scenarios. These findings indicate that incorporating stability-related variables into reward functions may help bridge the gap between DRL-based ATSC studies.

Understanding the spread of infectious diseases such as COVID-19 has been a persistent challenge for scientists and policymakers. While numerous studies have researched general aspects of the COVID-19 transmission dynamics, few have specifically examined its spread within public transportation systems in highly interconnected urban environments while incorporating the role played by the degree of effectiveness of time-varying mitigation strategies. This paper presents a novel approach using a discrete, probabilistic, and age-structured model that integrates the concept of compartmental models in epidemiology with cellular automata networks to investigate the role of public transportation and mitigation strategies in COVID-19 transmission in Mexico City districts. Therefore, the transmission of the disease is based on the definition of two distinct forces of infection: one associated with the use of public transport and the other with the occurrence within each district. Furthermore, the model incorporates several key time-varying parameters such as district population, commuting patterns, and contact matrices for different settings. Moreover, a stringency index was constructed based on the Oxford index in order to evaluate the impact of mitigation strategies established by the Mexican government and their compliance. Simulation results successfully reproduce COVID-19 transmission dynamics in Mexico City districts before vaccination efforts. Findings suggest that public transportation significantly contributed to the spread of the disease during the initial phase of the pandemic in Mexico City. These insights provide valuable guidance for public health strategies to mitigate infectious disease transmission in densely populated urban environments.

Wireless sensor networks (WSNs) have increased in popularity since they play a relevant role in many applications, such as environmental monitoring, fire detection, and health care, to mention a few. These applications periodically collect data that is relayed to a single sink employing a many-to-one communication pattern. This pattern requires energy-efficient routing protocols since sensors closer to the sink node deplete their energy faster than those sensors located further away. As a result, several techniques have been proposed to solve this problem. For instance, some pieces of research split the network into concentric coronas to provide more energy resources in areas with heavier traffic. However, these techniques use either a predefined network deployment that is not well-suited for all sensor applications or do not always guarantee homogeneous energy consumption. This paper proposes a simple energy-aware routing method based on the Game of Life cellular automaton, which provides a homogeneous energy depletion while extending the network lifespan by considering factors such as residual energy, number of active neighbors, and a sleep schedule. To this end, a discrete dynamic model that takes into account different behaviors of WSN through a set of rules combined with a variation of the A-star algorithm is proposed. Simulation results show that the proposed model accurately balances the energy consumption rate while expanding the network lifetime compared to most recent research works. Furthermore, the proposed method can be combined with path-planning algorithms to improve energy consumption in sparse WSNs.

Nowadays, methodologies coming from studying physical systems are being applied to the description of a wide variety of complex systems. In particular, one can study thermodynamical methods to describe the overall behavior of many systems, independent of the precise microscopic construction. In this paper, a real Mexican highway is studied as a cellular automata system using available official data released by the Mexican Government. The system studied is the Cuernavaca bypass which was modified in 2016. Official data allows to compare the highway before and after the modifications. As more complex thermodynamic variables such as entropy is difficult to define and measure in discrete traffic models, it is shown how other more simple variables such as the standard deviation can be enough to have a complete analysis of the system. More specifically, it is shown how standard deviation can be seen as a measure of order. Results from the study of the highway show how, taking a minimal measure such as ordering the transit of heavy trucks can reduce up to 32\% the travel time from one end to another. Otherwise, travel times stays practically constant with respect to the original system.

The use of smartphones has become an inherent part of daily human life. It allows users to keep personal information, emails, pictures, social media accounts, and financial data in one place. Consequently, smartphones are an attractive target for malware developers to spread malicious content, aiming at extracting information without the user's knowledge. Therefore, understanding malware propagation characteristics could provide a means to evaluate how they behave in order to plan security solutions accordingly. Bluetooth antennas are a channel for spreading malware through smartphones, where the probability of infection, similar to biological viruses, depends mainly on the attacker's physical proximity. This work presents a model based on cellular automata and epidemiological compartmental models for studying the spatial and temporal propagation of Bluetooth worms in smartphones. The proposed model incorporates the individual characteristics of each device, such as security settings, latency time, operating system, different classes of Bluetooth antennas (range and transfer rate), and different user mobility patterns. Several simulation scenarios are analyzed in order to study the spreading dynamics of Bluetooth-based worms, considering the location where the outbreak begins, and the different types of antennas integrated into the smart devices. Simulation results indicated that the proposed model is appropriate for studying how the users' demographics affect the worm's propagation dynamics in time and space. Moreover, the model permits an analysis of the impact of users' awareness about the risks inherent in using smart devices in Bluetooth networks, based on the acceptance of incoming communication and the effects of recovery and immunity to threats. Finally, the proposed model preserves simplicity and computational efficiency, with the possibility of extending beyond Bluetooth in order to include other transmission media.

Nowadays, methodologies coming from studying physical systems are being applied to the description of a wide variety of complex systems. In particular, one can study thermodynamical methods to describe the overall behavior of many systems, independent of the precise microscopic construction. In this paper, a real Mexican highway is studied as a cellular automata system using available official data released by the Mexican Government. The system studied is the Cuernavaca bypass which was modified in 2016. Official data allows to compare the highway before and after the modifications. As more complex thermodynamic variables such as entropy is difficult to define and measure in discrete traffic models, it is shown how other more simple variables such as the standard deviation can be enough to have a complete analysis of the system. More specifically, it is shown how standard deviation can be seen as a measure of order. Results from the study of the highway show how, taking a minimal measure such as ordering the transit of heavy trucks can reduce up to 30\% the travel time from one end to another. Otherwise, travel times stays practically constant with respect to the original system.