This paper delves into addressing road safety concerns by exploring cost-effective solutions for detecting sound events (SED) specifically designed for driving scenarios. While advanced technologies such as deep learning show promise in enhancing road safety, their practical implementation often involves expensive sensors and hardware. Given that distractions are a significant contributor to accidents, it is crucial to have effective detection and mitigation measures in place. This study focuses on auditory distractions and employs SED with affordable edge devices to identify and timestamp relevant audio events, offering valuable insights into the driving environment. We assess the performance of state-of-the-art deep learning models on various edge devices, including the 2023 DCASE baseline with convolutional recurrent neural networks (CRNN) and a customized YOLO vision model for audio spectrograms. Our analysis spans different hardware options, ranging from single board computers (SBCs) to desktop equipment, providing guidance on selecting cost-effective hardware for in-vehicle SED applications. The research aims to contribute to the development of affordable SED solutions in the context of driving safety, with the ultimate goal of advancing road safety initiatives globally.

Room acoustical parameters have been widely used to describe sound perception in indoor environments, such as concert halls, conference rooms, etc. Many of them have been standardized and often have a high computational demand. With the increasing presence of deep learning approaches in automatic monitoring systems, wireless acoustic sensor networks (WASNs) offer great potential to facilitate the estimation of such parameters. In this scenario, convolutional neural networks (CNNs) offer significant reductions in the computational requirements for in-node parameter predictions, enabling the so-called Artificial Intelligence-Internet of Things (AI-IoT). In this article, we describe the design and analysis of a CNN trained to predict simultaneously a set of common room acoustical parameters directly from speech signals, without the need for specific impulse response measurements. The results show that the proposed CNN-based prediction of room acoustical parameters and speech intelligibility achieves a relative error rate of less than a 5.5%, accompanied by a computational speedup factor close to 250 with respect to the conventional signal processing approach.

Industry 4.0 is currently viewed as the future era in which a new generation of wireless communications will enable the widespread connectivity between machines, objects, and users, promising significant advancements in industrial automation. Industrial scenarios are highly dynamic environments that require the ability to adapt to changes in operating conditions. Reliable wireless communications can be challenging in industrial environments, where the communications signal is often blocked due to the presence of industrial machinery, with possible intermittent and/or unpredictable blockages. This paper analyzes the wireless coverage provided by a low-cost IoT system in an industrial environment. A measurement campaign in different points of an industrial warehouse were carried out, showing substantial differences in terms of coverage between a totally empty area and an area with machinery and workers. The results show that the IoT system is able to provide enough coverage around the whole area, even with mobility or blockages.

The environmental psycho-acoustic annoyance is an important metric in the Smart City with Next Generation technologies. The use of such technologies can help to rapidly deploy large quantity of elements, which can dynamically be used for different applications. Also, the psycho-acoustic annoyance is usually based on the Zwicker’s model, but this model does not consider tonality of sounds to weight the subjective nuisance produced. In this work, we show a on-going work for the implementation of the nodes and EDGE/Fog to determine a modified version of the Zwicker’s model, to consider sound tonality (based on Aures’ method). This implementation has been designed to consider two options for offloading, one with sampling and computation on the EDGE and another with sampling in 5G-nodes and computation on the EDGE.

The proliferation of commercial low-cost Small Board Computers (SBC) devices have allowed the deployment of many Wireless Sensor Networks (WSN) focused on different applications, mainly based on monitoring issues. These networks are characterized by a set of these SBCs devices working in a collaborative way where each device is sensing, processing and later sending out the data to the sink. These devices are equipped with power supply, a processing unit and communications capabilities (in particular WiFi), making themselves very interesting to fit in many topologies. However, their performance in terms of communications basically depends on the environment and usually heuristic techniques are used with any guarantee if they will succeed or not. To model the different scenarios there are theoretically several approaches, such as ray tracing technique, two-ray, as well as semi-empirical models, but it should be interesting to analyze them also from a experimental point of view. In this paper we carry out a thorough performance analysis, in particular using Raspberry Pi3 (RPi3) devices, in order to understand their behavior in terms of Quality of Services (throughput, delay, losses and jitter (delay variation) for different types of communications: indoor and outdoor environments, with and without multipath reflections in order to establish precisely the amount of signal required depending on the required througput. In the outdoor environments, we analyze the QoS over the links between two nodes in a deployment.

Lightnings are violent natural phenomena and can generate many expenditures, specially when they strike in urban areas. The identification of the concrete geographic area where they strike is of critical importance for emergency services in order to enhance their effectiveness by doing an intensive coverage of the affected area. To achieve this aim, this paper proposes a design, prototype and validation of a distributed network of Internet of Things (IoT) devices to enable detection and location of lightning strikes. The IoT devices are empowered with lightning detection capabilities and are synchronized with the other devices in the sensor network. All of them cooperate within a network that is able to locate different events thanks to a trilateration algorithm implemented in a big data environment. The designed low cost lightning detection system is based on the AS3935 sensor. This alone device has a limited range of effective detection, but when it is embedded in a IoT mesh network, the accuracy and performance is increased up to good levels, in the order of kilometres. A fully operational IoT network has been deployed and a functional validation and empirical measurements are provided.

Soundscape standard (ISO 12913) is mainly oriented to describe the psychoacoustic annoyance (PA) due to the perceived sound in different environments. The evaluation of this annoyance is commonly based on the Zwicker and Fastl model that defines several components related to this subjective annoyance, such as loudness, sharpness, roughness, and fluctuation strength. But due to their complexity, these components are difficult to be calculated on small board computers (SBCs) in real time in order to enable a wireless acoustic sensor network for PA monitoring. In this article, we describe the necessary procedures to implement the complete psychoacoustic model by Zwicker and Fastl in a precise and efficient way, keeping the standards and criteria defined for each model, adding hints to improve its performance. We evaluate their execution time on different platforms and languages, including SBCs, in particular on Raspberry Pi family. Thus, this article shows a feasible implementation of this monitoring approach on low-cost platforms as well as a visualization tool.

The evaluation of QoS in mobile multipath environments is an interesting topic in modern communications. The use of Raspberry Pi as a low cost microcomputer board developed in the United Kingdom by the Raspberry Pi Foundation can be helpful in onsite deployments. The Pi3 model has integrated WiFi and Bluetooth connectivity and also 4 USB ports to expand the device with all kinds of peripheral. Under the premise of offering the power of a computer at low prices, this device is one of the most affordable ways to have a particularly efficient and dynamic hardware accessible at a low price to develop a Wireless Sensor Netwok (WSN). The use of the Raspberry Pi3 permits the development of a powerful and hight efficient mote in the development of a WSN, where data is highly processed in each mote and transmissions between motes and the sink or base station (BS) can be keep at minimum. We develop an experimental setup to analyse the quality of transmission between these devices in order to a further modelization of the channel. We program the Raspberry Pi3 to evaluate the quality of the transmission between two devices within a controlled multipath environment, particularly within an anechoic chamber.