A SYSTEMATIC SURVEY ON IOT-BASED DEVICES FOR ENHANCING WOMEN’S SAFETY

Abstract

Since many women face varied safety concerns, such as harassment, rape, molestation, and domestic abuse, for a variety of societal or cultural causes, women's safety has been emphasised as one of the main priorities of any community. The Internet of Things, or IoT, is emerging as a potential technology to help with daily issues and assist in managing a variety of matters. The community has developed a number of IoT-based tools to assist women in addressing possible safety risks. This study offers a comprehensive overview of the literature on IoT devices for women's safety, including wearable technology, sensors, and machine learning algorithms, as well as the key characteristics these devices provide. The evaluation process involves a thorough analysis and synthesis of the research publications published in reputable research venues between 2016 and 2022. The findings showed that several sensor types are employed to record the condition of women experiencing safety concerns, with pressure and pulse-rate sensors being the most often utilised sensors in these devices. Additionally, the devices employed a variety of technologies, such as the Raspberry Pi, worldwide positioning system (GPS), and global system for mobile communication (GSM), to send the notifications. Additionally, a number of machine learning methods, including logistic regression, hidden Markov, and decision trees, are employed to detect possible ladies who may be in danger and assist in averting the unfavourable circumstances for them in advance. It was determined that in order to successfully address the issue, systems that emphasise auto-activation of alarm creation with less human participation and greater accuracies must be created, even though much research has been produced in the underlying topic. This study proposes a taxonomy of various wearables, sensors, features, and approaches utilised in IoT-based women's safety devices in addition to analysing the literature. Additionally, the shortcomings and difficulties related to IoT devices and their suitability for women's safety have been brought to light. Furthermore, this paper suggests an architectural model that highlights key elements required to create IoT-based safety devices for women. Finally, this study highlights the utilisation of sensor combinations to get various input data types that may result in more accurate and precise threat detection.