Self-Sustaining Multi-Sensor LoRa-Based Activity Monitoring for Community Workout Parks

TL;DR

This paper presents a self-sustaining LoRa-based sensor node for monitoring community workout parks, combining energy harvesting and adaptive sampling to achieve long-term, low-power operation and provide valuable usage data for urban planning.

Contribution

It introduces a novel energy-efficient sensor design with adaptive sampling and energy harvesting, enabling sustainable, long-term activity monitoring in urban environments.

Findings

Achieved 2.8 seconds activity detection accuracy

Reduced power consumption to 1.147 mW, enabling 46 days of operation

Demonstrated sustainable operation with energy harvesting in field tests

Abstract

With the rise of the Internet of Things (IoT), more sensors are deployed around us, covering a wide range of applications from industry and agriculture to urban environments such as smart cities. Throughout these applications the sensors collect data of various characteristics and support city planners and decision-makers in their work processes, ultimately maximizing the impact of public funds. This paper introduces the design and implementation of a self-sustaining wireless sensor node designed to continuously monitor the utilization of community street workout parks. The proposed sensor node monitors activity by leveraging acceleration data capturing micro-vibrations that propagate through the steel structures of the workout equipment. This allows us to detect activity duration with an average measured error of only 2.8 seconds. The sensor is optimized with an energy-aware, adaptive sampling and transmission algorithm which, in combination with the Long Range Wide Area Network (LoRaWAN), reduces power consumption to just 1.147 mW in normal operation and as low as 0.712 mW in low-power, standby mode allowing 46 days of battery runtime. In addition, the integrated energy-harvesting circuit was tested in the field. By monitoring the battery voltage for multiple days, it was shown that the sensor is capable of operating sustainably year-round without external power sources. To evaluate the sensor effectiveness, we conducted a week-long field test in Zurich, placing sensors at various street workout parks throughout the city. Analysis of the collected data revealed clear patterns in park usage depending on day and location. This dataset is made publicly available through our online dashboard. Finally, we showcase the potential of IoT for city applications in combination with an accessible data interface for decision-makers.