Water Distribution Networks (WDNs) are complex, dynamic systems critical to modern society but increasingly difficult to manage due to urbanization, fluctuating demands, and resource constraints. To address these challenges, Smart Water Distribution Networks (SWDNs) utilize Internet of Things (IoT) devices and protocols like Long Range Wide Area Network (LoRaWAN) for real-time monitoring and analysis, enabling smarter and more efficient water management. This demo presents SWIM (Smart Water Interaction & Monitoring), an innovative application designed to modernize SWDNs. SWIM integrates Digital Twins (DTs), established simulation tools like EPANET, and Machine Learning (ML) to provide predictive analytics, anomaly detection, and real-time control. By employing neural networks, SWIM achieves high-accuracy hydraulic predictions with minimal input data. Built on IoTs and Low Power Wide Area Networks (LPWANs), SWIM delivers scalable, efficient, and user-friendly solutions. It aligns with the principles of Industry 5.0, demonstrating the potential to revolutionize water distribution networks and ensure their sustainability in the face of modern challenges.

This survey explores the convergence of Internet of Things (IoT) technologies with Water Distribution Systems (WDSs), focusing on large-scale deployments and the role of edge computing (EC). Effective water management increasingly relies on IoT monitoring, resulting in massive deployments and the generation of Big Data. While previous research has examined these topics individually, this work integrates them into a comprehensive analysis. We systematically reviewed 255 studies on IoT in WDS, identifying key challenges such as interoperability, scalability, energy efficiency, network coverage, and reliability. We also examined technologies like LPWAN and the growing use of EC for real-time data processing. In large-scale WDS scenarios, where vast amounts of data are generated, we highlighted the importance of technologies like NB-IoT, SigFox, and LoRaWAN due to their low power consumption and wide coverage. Based on our findings, we provide guidelines for sustainable, large-scale IoT deployment in WDS, emphasizing the need for edge data processing to reduce cloud dependency, improve scalability, and enable smarter cities and digital twins.

The digitalization of Water Distribution Systems (WDSs) is becoming a key objective in modern society. The increasing complexity of contemporary WDSs, driven by urbanization, fluctuating consumer demand, and limited resources, makes their management particularly challenging, especially in large-scale scenarios. This paper proposes the SWI-FEED framework designed to facilitate the widespread deployment of the Internet of Things (IoT) for enhanced monitoring and optimization of WDSs. The framework aims to investigate the utilization of massive IoT in monitoring and optimizing WDSs in different contexts, with a particular focus on four use cases such as optimal node activation, IoT gateways deployment, distributed leakage detection and water demand disaggregation. SWI-FEED has been tested with predefined network models available in the Open Water Analytics community public repository. Specifically, the four use cases are evaluated using a large network consisting of 4,419 sensor nodes, 3 tanks and 5,066 pipes. Overall, this comprehensive framework provides a holistic approach to address possible challenges of a WDS and optimize the efficiency of large-scale IoT deployments. It reduces the energy consumption of IoT devices within the WDS while enhancing leak detection and localization capabilities in real-world water networks. Our adopted theoretical methodology is based on graph theory, which allows IoT gateways to be strategically positioned to maximize network coverage and minimize infrastructure redundancy. This makes it possible to significantly reduce the number of gateways required and, consequently, the overall system energy consumption.

This manuscript introduces a novel approach to integrate Large Language Models (LLMs) into wireless network testbeds for automated experimentation and optimization. We propose a framework that leverages LLMs to define system configurations using a structured format called Blueprint and analyse network behaviour through user-prompted queries. Our methodology combines few-shot prompting and prompt engineering to enable automated analysis of network configurations, enhancing decision-making and experimentation efficiency. We validated our approach in both simulated and real-world wireless environments and demonstrated its efficacy in streamlining experiment processes and extracting actionable insights. • Novel LLM-based framework for system configurations using standard Blueprints format. • Domain-agnostic few-shot prompting for automated wireless network configurations. • Real-world validation in Wireless network environments (e.g. LoRaWAN and Wi-Fi). • Practical LLM-driven anomaly detection to enforce network corrective actions.

According to IoT Analytics, NB-IoT, LoRaWAN, and Sigfox are today the most popular technologies for low-power wide-area networks (86% of the market), both in terms of end-user adoption as well as ecosystem support. While NB-IoT utilizes licensed bands, LoRaWAN and Sigfox both employ the sub-GHz ISM bands, potentially interfering with each other.In this paper, we present a thorough coexistence study between LoRaWAN and Sigfox, in realistic urban scenarios, with different duty cycles and traffic conditions. The choice of such scenarios and simulation parameters are supported by an in-depth lit-erature review, and simulations are based on the SEAMCAT simulator. The results offer new insights on the coexistence of LoRaWAN and Sigfox for emerging IoT applications. Finally, as interference mitigation strategy, we analyze the performance obtained applying protection distance mechanisms.

Modern standards for IoT communications support fast deployment, large coverage in the order of kilometers, and physical layer adaptations to increase link robustness under time-varying propagation and interference conditions. A possible use of such IoT technologies is in case of emergency scenarios where first responders (FRs) arrive after a disastrous event. Indeed, an important challenge for emergency management is the need to (re)establish real-time communication capabilities and to offer integrated decision making facilities based on information gathered by FRs acting on the crisis site. In this paper, we present a system architecture based on LoRaWAN technology for connecting emergency operators in real-time and reliably communicating environmental information, audio streams/messages, and vital signs received from the first responders' sensors. In particular, based on LoRa modulation parameters, we propose an adaptation algorithm which adjusts user's voice messages and the resolution of the data flows to keep alive communications also when link quality is critically low, thus avoiding delay and saturation problems. Opportunistically, audio signals can be processed locally by the first responder's equipment with a speech-to-text conversion, thus significantly reducing traffic requirements. We demonstrate that the adaptation scheme can be performed real-time, even on a per-packet basis. Thanks this innovative system, FRs can communicate from the crisis site in an efficient and cost-effective way.

In this paper we show that inter-technology interference can be recognized by commodity WiFi devices by monitoring the statistics of receiver errors. Indeed, while for WiFi standard frames the error probability varies during the frame reception in different frame fields (PHY, MAC headers, payloads) protected with heterogeneous coding, errors may appear randomly at any point during the time the demodulator is trying to receive an exogenous interfering signal. We thus detect and identify cross-technology interference on off-the-shelf WiFi cards by monitoring the sequence of receiver errors (bad PLCP, bad PCS, invalid headers, etc.) and develop an Artificial Neural Network (ANN) to recognize the source of interference. The result is quite impressive, reaching an average accuracy of almost 99% in recognizing ZigBee, Microwave and LTE (in unlicensed spectrum) interference.

Most modern standards for wireless communications support physical layer adaptations, in terms of dynamic selection of channel central frequency, transmission power, modulation format, etc., in order to increase link robustness under time-varying propagation and interference conditions. In this demo, we demonstrate that another powerful solution for extending physical layer flexibility in OFDM-based technologies is the dynamic adaptation of the channel width. Although some standards already define the possibility of utilizing multiple channel widths (e.g. 20MHz, 10MHz, 5MHz for IEEE 802.11a standards), such an utilization is limited to a static configuration of a value defined during the network set-up. Conversely, we demonstrate that channel width adaptations can be performed in real-time during network operation, even on a per-packet basis. To this purpose, we propose an innovative and efficient receiver design, which allows the transmitter to take decisions about the channel width without explicitly informing the receiver.

In this paper, we present Overgrid, a fully distributed peer-to-peer (P2P) architecture designed to automatically control and implement distributed demand response (DR) schemes in a community of smart buildings with energy generation and storage capabilities. As overlay networks in communications establish logical links between peers regardless of the physical topology of the network, the Overgrid is able to apply some power balance criteria to its system of buildings, as they belong to a virtual microgrid, regardless of their physical location. We exploit an innovative distributed algorithm, called flow updating, for monitoring the power consumption of the buildings and the number of nodes in the network, proving its applicability in an Overgrid scenario with realistic power profiles and networks of up to 10 000 buildings. To quantify the energy balance capability of Overgrid, we first study the energy characteristics of several types of buildings in our university campus and in an industrial site to accurately provide some reference buildings models. Then, we classify the amount of “flexible” energy consumption, i.e., the quota that could be potentially exploited for DR programs. Finally, we validate Overgrid emulating a real P2P network of smart buildings behaving according to our reference models. The experimental results prove the feasibility of our approach.

In this paper we describe how a novel backoff mechanism called Moderated Backoff (MB), recently proposed as a standard extension for 802.11 networks, has been prototyped and experimentally validated on a commercial 802.11 card before being ratified. Indeed, for performance reasons, the time critical operations of MAC protocols, such as the backoff mechanism, are implemented into the card hardware/firmware and cannot be arbitrarily changed by third parties or by manufacturers only for experimental reasons. Our validation has been possible thanks to the availability of the so called Wireless MAC Processor (WMP), a prototype of a novel wireless card architecture in which MAC protocols can be programmed by using proper abstractions and a state-machine formal language, which enable easy modifications of legacy operations. Experimental results are in agreement with simulations and prove the effectiveness of Moderated Backoff, as well as the potentialities of the WMP platform.

Recent years have witnessed the increasing adoption of heterogeneous wireless networks working in unlicensed ISM bands, thus creating serious problems of spectrum overcrowding. Although ZigBee, Bluetooth and WiFi networks have been natively designed for working in presence of interference, it has been observed that several performance impairments may occur because of heterogeneous sensitivity to detect or react to the presence of other technologies. In this paper we focus on the WiFi capability to detect interfering ZigBee links. Despite of the narrowband transmissions performed by ZigBee, in emerging scenarios ZigBee interference can have a significant impact on WiFi performance. Therefore, interference detection is essential for improving coexistence strategies in heterogeneous networks. In our work we show how such a detection can be performed on commodity cards working on time and frequency domain and also analysing data in the error domain. Errors are monitored and classified into error patterns observed in the network in terms of occurrence probability and temporal clustering of different error events. Through statistical analysis we are able to detect the presence of ZigBee transmissions measuring the errors raised by the WiFi card.

We introduce MAClets, software programs uploaded and executed on-demand over wireless cards, and devised to change the card's real-time medium access control operation. MAClets permit seamless reconfiguration of the MAC stack, so as to adapt it to mutated context and spectrum conditions and perform tailored performance optimizations hardly accountable by an once-for-all protocol stack design. Following traditional active networking principles, MAClets can be directly conveyed within data packets and executed on hard-coded devices acting as virtual MAC machines. Indeed, rather than executing a pre-defined protocol, we envision a new architecture for wireless cards based on a protocol interpreter (enabling code portability) and a powerful API. Experiments involving the distribution of MAClets within data packets, and their execution over commodity WLAN cards, show the flexibility and viability of the proposed concept.