Amit Kumar Agrawal

Mobile healthcare infrastructure for home and small clinic

In this paper, the authors describe the software infrastructure built to provide healthcare solution for home and small clinic scenario using mobile healthcare devices. The paper discusses how the data from the devices have been stored into a generic Internet of Things (IoT) backend and then retrieved using web services to provide an interactive portal which provides patient data in a unified manner. The novelty of the approach lies in the use of a mobile device as a gateway to reach the backend IoT platform, which supports application deployment and then applying it to the health-care vertical. The mobile device has hence been used as a pervasive healthcare gateway that collects data from the medical devices using either Bluetooth or Wi-Fi and uploads the data to the back-end server in desired format. Hence the mobile application acts as the edge platform. Several connected healthcare devices like ECG, pulse oxygen meter, body fat analyzer and blood pressure monitor have been used in conjunction with the mobile gateway.

UbiHeld: ubiquitous healthcare monitoring system for elderly and chronic patients

Once the persons identity is established, the most important aspects of ubiquitous healthcare monitoring of elderly and chronic patients are location, activity, physiological and psychological parameters. Since smartphones have become the most pervasive computing platform today, it is only a logical extension to use the same in healthcare domain for bringing ubiquity. Besides smartphone, skeleton based activity detection and localization using depth sensor like Kinect make ubiquitous monitoring effective without compromising privacy to a large extent. Finally sensing mental condition is made possible by analysis of the subjects social network feed. This paper presents an end-to-end healthcare monitoring system code named UbiHeld (Ubiquitous Healthcare for Elderly) using the techniques mentioned above and an IoT (Internet of Things) based back-end platform.

SmartEvacTrak: A people counting and coarse-level localization solution for efficient evacuation of large buildings

Efficient evacuation of large buildings which seat more than twenty thousand people is a major challenge for building management authorities. Although many enterprise buildings have access control to count those entering various areas, such controls need to be switched off at times of emergency to avoid people getting trapped at a particular location. Hence, counting people during evacuation remains largely manual and error prone. Also, last known location based on access control from business management systems can be erroneous due to issues like tail gating. Accurate people counters like Radio Frequency Identification (RFID) often prove costly in large deployments. In this paper we present “SmartEvacTrak”, a low-cost evacuation system which can count people entering and exiting with over 98% accuracy and can also localize people at a coarse level with around 97% accuracy, from results obtained from an evacuation experiment performed on a set of 350+ people randomly exiting a building.

Using social network graphs for search space reduction in internet of things

In this paper, we explore reduction of search space in sensor data analytics using social network graph theory. Human centric social network allow graphical connect of individuals either based on familiarity or common intent. This facilitates people centric applications as they now operate on a much smaller data set. Extending this analogy to sensor networks, if sensors can be associated with meaningful social groups, it will reduce sensor data analytics and processing overhead for an application by a huge order. In this paper we explore how in Internet-of-Things sensors can be assigned to human beings who in turn are connected in social networks. Effectively in this way, sensors become part of a social network that results in a reduced data set for sensor data analytics.

A methodology for GPS-based waterlogging prediction and smart route generation

This paper describes a system for predicting water logging prone areas in multiple routes. The approach is based on the theory that water tends to accumulate in low-lying areas and hence a route which contains more and bigger basins is more likely to behave worse on a rainy day. Using this basic principle, algorithms are formulated and applied to identify and quantify water logging zones on a route. To prove the effectiveness of the proposed system, the derived confidence scores for multiple routes given by the system are compared to judgements given by human commuters. A view of all possible routes along with quantified estimates of waterlogging confidence scores are rendered in Google map.

Automatic alert generation from train to the people at unmanned level crossings using principles of iot

This paper presents a novel concept of an application to improve safety at unmanned level crossings without the burden of a huge infrastructure cost to the railways governing body. The application will keep track of all running train locations and the unmanned level crossings on its way. The application needs to have access to location Appplication Programming Interface(API) provided by Telecom Service Providers. With usage of these APIs the application will be able to keep track of all the people near unmanned level crossings. All service providers have Bulk Short Message Service(SMS) APIs using which the application will be able to send alert messages to people in vicinity of unmanned level crossings. It will make road users aware of the train approaching the level crossing and reduce the risk exposure. The same principle of Internet of Things (IoT) can be extended here for developing a query service by users about approaching trains in a particular level crossing.