A New Remote Monitor and Control System Based on Sigfox IoT Network

TL;DR

The paper tackles the challenge of maintaining remote monitoring and control of sensitive laboratory instrumentation during power outages and network disruptions. It proposes a battery-powered platform built around the Arduino MKR FOX 1200 with SigFox UNB for bidirectional communication, supplemented by Ethernet as a normal-path and SigFox for emergencies. Key contributions include a modular I2C/1-Wire sensor network, a novel AC power line monitor delivering power-loss detection in around 2 ms, a compact 8-byte SigFox payload framework for multi-sensor data, and an Android/Firebase-based end-user interface for real-time alerts and remote commands. The system demonstrates robust operation across normal and emergency modes, with quantified latencies and a practical, expandable architecture for laboratory and industrial remote monitoring scenarios.

Abstract

We describe a new, low-cost system designed to provide multi-sensor remote condition monitoring of modern scientific laboratories, as well as to allow users to perform actions from remote locations in case of detection of specified events. The system is battery operated and does not require the presence of a Local Area Network (LAN) or WiFi (which are typically not available in case of, e.g. power losses), as it exploits the growing infrastructure of Internet of Things (IoT) Low Power Wide Area Networks (LPWAN). In particular our system exploits the new SigFox ultra-narrow-bandwidth (UNB) infrastructure, and provides for a bidirectional link between the instrumentation and the remote user even in case of power line outages, which are among the most critical situations that a scientific laboratory can withstand. The system can detect the occurrence of predefined events in very short times, and either autonomously react with a series of predefined actions, also allowing a remote user to timely perform additional actions on the system through an user-friendly smartphone application or via a browser interface. The system also embeds a novel power-loss detection architecture, which detects power line failures in less than 2 ms. We provide a full characterization of the prototype, including reaction times, connection latencies, sensors sensitivity, and power consumption.

Figures (5)

• Figure 1: Prototypal board. a: A photo of the final assembled prototypical board. b: The board conceptual schematics showing the main components of the power supply unit and of the sensing and communication unit. The sketch also shows a possible arrangement of the sensor network consisting of two sensors connected in "star" configuration to the I$^2$C bus and four thermometer probes connected in a "daisy chain" configuration to the 1-wire bus. See text for details.
• Figure 2: Sensor data acquisition. a - b: Example of sensor data log during a simulated power outage event. Panel a reports the log of a 1-Wire DS18B20 thermometer probe monitoring the laboratory ambient temperature while panel b reports the log of the main AC power line RMS value. The dashed and gray shadowed areas mark the time intervals during which the board was battery powered, and those in which SigFox communication was employed, respectively. At approximately 10:30 the board automatically switches to battery supply in response to a power outage. Later (approx. 11:15), a failure of the Ethernet connectivity forces the board to switch to SigFox network to communicate sensor data. Sensor data update rate to the database is (1 min)$^{-1}$ during normal operating conditions and (5 min)$^{-1}$ when SigFox communication is employed. c - d: Typical board reaction time to a main AC line power outage. The blue line reports the sinusoidal AC RMS value in entrance to the board while the discontinuity of the red dashed line marks the time at which the board detect the power outage. In this example the detection time is 2.15 ms.
• Figure 3: AC power line monitoring schematics: an AC-AC converter reduces the main electricity line voltage to a lower voltage which is fed into a precision rectifier in order to generate a positive-only periodic signal. This signal is sampled at regular times exploiting a dedicated ADC of the Arduino board in order to provide the current RMS voltage of the main line and detect eventual power outages.
• Figure 4: Platform information flow. Double head arrows indicate link bidirectional communication capability. Laboratory sensor data are transferred from the board to an online database via internet or SigFox connection, depending respectively whether the system is operating in normal or emergency conditions. Data can be accessed by the end-user via an Android application which also provides instantaneous visualization of critical events exploiting the Google Firebase push notification service. The links bidirectionality allows the user to set remote commands to be executed by the board.
• Figure 5: Examples of Android application screenshots showing the list of sensors connected to the board (a) and the time series plot of two temperature probes (b).