Injuries due to unintentional falls cause high social cost in which several systems have been developed to reduce them. Recently, two trends can be recognized. Firstly, the market is dominated by fall detection systems, which activate an alarm after a fall occurrence, but the focus is moving towards predicting and preventing a fall, as it is the most promising approach to avoid a fall injury. Secondly, personal devices, such as smartphones, are being exploited for implementing fall systems, because they are commonly carried by the user most of the day. This paper reviews various fall prediction and prevention systems, with a particular interest to the ones that can rely on the sensors embedded in a smartphone, i.e., accelerometer and gyroscope. Kinematic features obtained from the data collected from accelerometer and gyroscope have been evaluated in combination with different machine learning algorithms. An experimental analysis compares the evaluated approaches by evaluating their accuracy and ability to predict and prevent a fall. Results show that tilt features in combination with a decision tree algorithm present the best performance.

Falls are critical events for human health due to the associated risk of physical and psychological injuries. Several fall-related systems have been developed in order to reduce injuries. Among them, fall-risk prediction systems are one of the most promising approaches, as they strive to predict a fall before its occurrence. A category of fall-risk prediction systems evaluates balance and muscle strength through some clinical functional assessment tests, while other prediction systems investigate the recognition of abnormal gait patterns to predict a fall in real time. The main contribution of this paper is a nonlinear model of user gait in combination with a threshold-based classification in order to recognize abnormal gait patterns with low complexity and high accuracy. In addition, a dataset with realistic parameters is prepared to simulate abnormal walks and to evaluate fall prediction methods. The accelerometer and gyroscope sensors available in a smartphone have been exploited to create the dataset. The proposed approach has been implemented and compared with the state-of-the-art approaches showing that it is able to predict an abnormal walk with a higher accuracy (93.5%) and a higher efficiency (up to 3.5 faster) than other feasible approaches.

In the vision of the Internet of Things, an object embedded in the physical world is recognizable and becomes smart by communicating data about itself and by accessing aggregate information from other devices. One of the most useful types of information for interactions among objects regards their movement. Mobile devices can infer their position by exploiting an embedded accelerometer. However, the double integration of the acceleration may not guarantee a reliable estimation of the displacement of the device (i.e., the difference in the new location). In fact, noise and measurement errors dramatically affect the assessment. This paper investigates the benefits and drawbacks of the use of the Kalman filter as a correction technique to achieve more precise estimation of displacement. The approach is evaluated with two accelerometers embedded in commercial devices: A smartphone and a sensor platform. The results show that the technique based on the Kalman filter dramatically reduces the percentage error, in comparison to the assessment made by double integration of the acceleration data; in particular, the precision is improved by up to 72%. At the same time, the computational overhead due to the Kalman filter can be assumed to be negligible in almost all application scenarios.

In-memory key-value stores have quickly become a key enabling technology to build high-performance applications that must cope with massively distributed workloads. In-memory key-value stores (also referred to as NoSQL) primarly aim to offer low-latency and high-throughput data access which motivates the rapid adoption of modern network cards such as Remote Direct Memory Access (RDMA). In this paper, we present the fundamental design principles for exploiting RDMAs in modern NoSQL systems. Moreover, we describe a break-down analysis of the state-of-the-art of the RDMA-based in-memory NoSQL systems regarding the indexing, data consistency, and the communication protocol. In addition, we compare traditional in-memory NoSQL with their RDMA-enabled counterparts. Finally, we present a comprehensive analysis and evaluation of the existing systems according to the impact of the number of clients, real-world request distributions, and workload read-write ratios.

Human gait is a dynamic biometrical feature that describes the kinematics of human walking. Gait modeling is studied in order to find a pattern of walking that can be used for diagnosis of walking disorder or abnormal walk detection. Difficulty in walking progressively increases with aging and causes unintentional falls, which is a common incident among elderly people. Fall prediction systems can help to prevent unintentional falls that could cause serious injuries, therefore they can reduce the health service costs. This paper presents an algorithm with polynomial classification model of human gait for real-time fall prediction. This approach enables the user to detect the transition from a normal to an abnormal walking pattern. A dataset based on the state-of-the-art techniques in simulating abnormal walks was created by using an accelerometer embedded in a smartphone, which is recognized to be precise enough for fall avoidance systems. The proposed approach improves state-of-the-art fall prediction approaches, by achieving 99.2% of accuracy in abnormal walk detection.

Internet of Things (IoT) is making a breakthrough for the development of innovative healthcare systems. IoT-based health applications are expected to change the paradigm traditionally followed by physicians for diagnosis, by moving health monitoring