Page - 28 - in Intelligent Environments 2019 - Workshop Proceedings of the 15th International Conference on Intelligent Environments

In order to quantify sleep performance, the following sleep parameters are usually considered and measured [9] [8]: Sleep Onset Latency (SOL), Wake After Sleep Onset (WASO), Total Sleep Time (TST), Sleep Efficiency (SE), Number of Awakenings (NAWK). Table 1 presents their definitions. Figure 1. Main sleep stages transitions. Table 1. Sleep parameters Sleep parameter Description Sleep Onset Latency (SOL) Time needed to start sleeping Wake After Sleep Onset (WASO) Time awake after sleep onset Total Sleep Time (TST) Time asleep less awake time Sleep Efficiency (SE) Percentage of time in bed spent in sleep Number of Awakenings (NAWK) Number of awakenings during sleep time 2.2. Sleep Monitoring Systems and Environmental Data Our search of related work considering sleep monitoring systems that take into account the effect of environmental factors has resulted in the following three cases. The first example is Lullaby [10], a capture and access system for understanding the sleep environment. The authors of this system use temperature, light, and motion sensors, audio and photos, and an off-the-shelf sleep sensor to acquire a complete recording of a person’s sleep. The system visualizes graphs and parameter recordings, allowing users to find trends and potential causes of sleep disruptions relatively to environmental factors, helping them understand their sleep environment. SleepExplorer [11] is a web-based visualization tool to make sense of correlations between personal sleep data and contextual factors. It imports data from commercial sleep trackers and online diaries and helps users to better understand their sleep and to discover novel relationships between sleep data and contextual factors. The contextual factors collected during the field study were organized under four main categories: physiological factors (weight, body temperature and menstrual cycles), phychological factors (mood, stress, tiredness and dream), behavioral factors (steps, minutes very active, minutes fairly active, minutes lightly active, calories in, calories out, activity calories, coffee, coffee time, alcohol, electronic devices usage, evening light, nap time, nap duration, social activities, exercise time and dinner time) and environmental factors (ambient temperature and ambient humidity). Borazio and Laerhoven [13] present a long-term sleep monitoring system combining wearable and environmental data to automatically detect the user’s sleep at home. The system uses inertial, ambient light, and time data tracked from a wrist-worn sensor, synchronized with night vision camera footage for easy visual inspection. C.Gonçalvesetal. /MultisensorMonitoringSystem toEstablishCorrelations28