*3.6. Study Cases*

This section describes experimental results of study cases implemented by the research groups of the co-authors of this review, developed in Tecnologico de Monterrey, under the Campus City initiative. Figure 10 shows a visual representation of the study cases discussed in this section.

*Health*. In this study, by using EEG electrodes and Bluetooth, brain signals from students were recorded to assess learning outcomes under different modalities [177]. The aim of the work was to propose EEG sensing as a support in education by inferring the state of the brain. The results showed that machine learning models based on the EEG recordings were able to predict with 85% accuracy, the cognitive performance of the students, and it could also be used to identify unwanted conditions, such as mental fatigue, anxiety, and stress under different contexts in the healthcare sector.

*Security*. PiBOT is a multifunctional robot developed to monitor spaces and implement regulations in the context of the COVID-19 pandemic [178]. Such robot integrates video and thermal cameras, LiDAR, ultrasonic, and IR sensors to allow object and people detection, facemask recognition, temperature maps, and distance between persons estimation. It also integrates automated navigation algorithms, teleoperation control modes and cloud connection (IoT) protocols for data transmission. The robot is also able to generate

and send real-time data to a web server about people count, facemask misuse, and safe distance violations.

**Figure 10.** Study cases of smart cities implementations in Tecnologico de Monterrey, (**a**) Health: EEG monitoring for educational services; (**b**) Security: Space monitoring in the context of COVID-19 pandemic; (**c**) Mobility: Urban accessibility analysis in Monterrey City; (**d**) Water: Detection of SARS-CoV-2 RNA in freshwater environments from Monterrey City; (**e**) Waste: GPS monitoring and app for recycling vehicles; (**f**) Energy: Impact of A/C usage on light-duty vehicle's fuel consumption.

*Mobility*. The following study presents the results from the analysis of accessibility to different services (health and education) in the urban environment of the city of Monterrey, Mexico [179]. The software tools used in this study enabled to obtain quantitative representations of the accessibility of the city when using different transport modes (walking and cycling). The results from this work showed low accessibility to medical services, but acceptable accessibility to educational services in the city. The study found that the use of bicycles and other micro mobility vehicles can enhance the accessibility to services in the city.

*Water*. A recent study from one of the co-authors studies the presence of SARS-CoV-2 RNA in different freshwater environments (groundwater and surface water from rivers and dams) in urban settings [180] from the city of Monterrey, Mexico. The detection and quantification of the viral loads in such environments were determined by RT-qPCR in samples acquired from October 2020 to January 2021. The results of the study demonstrated the feasibility of the presence of SARS-CoV-2 in freshwater environments. It also found that viral loads variations in groundwater and surface water over time at the submetropolitan level reflected the reported trends in COVID-19 cases in the city of Monterrey.

*Waste*. A recent collaboration between Tecnologico de Monterrey and industrial partners from Smart City Colombia [181] and SmarTech [182] has started for the development of smart solutions for cities. This collaboration proposes the use of a smart recycling implementations using an urban mechanism for intelligent disposal (MUDI). The MUDI is a GPS monitored vehicle that establishes optimal routes for collection of recycled material, while informing both the recycler and the citizen through an app about the final disposal of said materials.

*Energy*. Using engine sensors, information about the fuel consumption associated to the operation of the A/C in light-duty vehicles was monitored via OBD for a five-month period [131]. Results showed that specific fuel consumption due to A/C usage is higher at lower speeds of the vehicle and it is lower at higher driving speeds; and shows the potential of proposing solutions towards vehicle energy efficiency by analyzing information coming from engine sensors through the OBD port.

#### **4. Challenges and Opportunities**

Advances in the use and implementation of sensors and their application for the development of smart cities will allow residents to access a better quality of life. Even though the use of wireless sensor networks provides valuable data that are used for a better managemen<sup>t</sup> of resources, there are still areas of opportunity for improvement. Although different areas within the smart city face specific challenges, a common opportunity is the development of new sensors and new approaches to the problem of detection, prevention, or anticipation of the dangers which future smart cities can face.

A major challenge in health applications is privacy and the secure transmission of data, a concern for which different studies have been conducted. One example is a decentralized mobile-health system that leverages patients' verifiable attributes in order to run an authentication process and preserve the attribute and identity [183]. Another study designs two schemes that focus on the privacy of medical records. These schemes ensure that highly similar plaintexts can be transformed into distinctly different ciphertexts and resist ciphertext-only attacks. Nevertheless, important performance metrics, such as computation overhead, network connectivity, delay and power consumption, are ignored [184]. Low-cost wireless sensor networks also help to achieve a direct communication between a user's mobile terminals and wearable medical devices while enforcing privacy-preserving strategies [149]. In addition, a study presented a big health application system based on big data and the health internet of things (IoT). This study also proposed the cloud to end fusion big health application system architecture [185]. One last study modifies the design of sensor networks in a way that each one can manipulate four symbols (quaternary) instead of two (binary), resulting in more efficient systems [186].

With respect to mobility, ITS are used in smart cities, having a positive impact on saving resources, such as time and workforce, while reducing the use of fuels and emissions into the atmosphere. ITS image-based mobility applications are simple and inexpensive, but face decreased efficiency during lightning and weather changes [18]. Another challenge is faced by flexible traffic control, as well as collision avoidance systems as high speed detections and data exchange are needed for successful V2V, V2P, V2I, and V2X protocols [18,38,66]. This information exchange process is susceptible to security threats, such as malicious attacks or data leaks [61]. To address these challenges, more reliable sensors and faster data transfer protocols need to be implemented.

In smart security, there is a big effort on detection, prevention, or anticipation of the dangers that citizens and infrastructure of the smart cities can face. Sensors for security present a tendency to improve the sensibility, resolution, and precision of the current sensors. Almost all services in a smart city use digital data and are completely dependent on the security and integrity of that data. Due to this reason, sensors must be hardened with effective security solutions such as cryptography and advanced self-protection techniques.

Regarding smart water monitoring, sensors are used to measure water quantity and quality data continuously and consistently. The obtained data can be processed and visualized in real-time to the end-users, or forecasts can be developed for the water agencies. These technologies allow minimization of the risks associated to poor water quality and deficiencies in water supply. Future sensors need to be improved in cost and energy consumption to withstand long periods of measurements without intervention, in addition to an enhanced robustness, to resist adverse environmental conditions.

Solid waste managemen<sup>t</sup> is crucial in any town or city, but take a new role in the smart city scheme, and it is focused on a more clean, tidy, and healthy environment for living, using sensors and IoT technologies to improve waste managemen<sup>t</sup> [42,43]. Currently there are only sensors that can identify wet, dry, or metal garbage; however, it would be optimal to develop sensors that allow identifying waste in greater detail. For this reason, new sensors oriented to waste segregation need to be developed and implemented. Segregation is a key component in the waste managemen<sup>t</sup> system, as it allows much of what is discarded to be recycled or reused, resulting in a reduction in the amount of garbage that reaches landfills [128,172].

Innovations in energy consumption monitoring in buildings, public lighting systems, and urban spaces using ICTs are an excellent option due to their adaptability. The literature proposes the implementation of virtual sensors by building information modeling (BIM), integrated with IoT devices including qualification tools to develop ecological buildings [134,187]. Intelligent lighting systems with sensors adaptable to weather conditions, hours of use and presence of people or vehicles [20] where the street lamps serve as Wi-Fi connection points, allowing interconnected networks over the entire urban area monitoring the quality of the environment, noise levels, and surveillance, among others. Battery or energy consumption, high volume of data storage and security, life span and replacement, size, cost, installation, and maintenance are the main deficiencies when designing a WSN. Studies have proposed the implementations of energy efficiency surveillance using multimodal sensors [188] and low power hardware systems [189]. The implementation of low-cost sensors and using energy harvesting are the most attractive technologies for buildings' sensors in the future. Artificial intelligence, big data, and machine learning [190,191] become essential due to the vast amount of data gathered and analyzed in the presented applications.
