Contents I. Introduction
The great amount of low-budget sensors available on the market allows producers to design and develop a large number of different acquisition systems. These platforms are able to collect detailed information about our health status and they are usually paired with mobile applications to provide a near real-time feedback to the patient. The data collected and analyzed by artificial intelligence algorithms can be exploited to predict possible adverse events before they occur, to periodically monitor overall health status or to evaluate the effectiveness of a given drug therapy [1]. In order to provide relevant clinical information and to give a reliable output, these platforms usually need to collect synchronously multiple biological parameters and to process together several different time-series based on a common time base. To fulfill this requirement, most recent platforms are multi-node systems and they are based on a wired architecture, in order to ensure the synchronization among different sensors' acquisitions [2]. This wired architecture can be exploited only in products designed to be used in a controlled environment, such as a research or an hospital lab. This is an important limitation that prevents the development of a fully wearable multi-node platform that can be exploited also during users' daily life. Indeed users may not only be clinical patients with medical needs, but also healthy subjects such as athletes interested in how their bodies respond to a specific stimulus. In this case the final goal is the physiological values' monitoring during fitness activities in order to improve physical performance in a competitive sports scenario (e.g. athletes may need to check specific physiological parameters in order to evaluate the most suitable training plan before a competition) [3]. A possible application example may be a platform to evaluate the relationship between a muscular activation and the corresponding articular movement during a workout. In this case, a synchronized sampling of both the electromyographic signal (EMG) and the limb inertial sensor signal is needed, as well as a wireless system that can be used directly on the field during the activities. Another example may be a platform to collect both the electrocardiographic signal (ECG) and a peripheral plethysmographic signal (PPG) of an athlete to monitor his cardiovascular system status during an endurance race. Also in this case, collected signals must have a common time base to be synchronized and processed without compromising athlete's comfort.
