Sensor-driven computing and communications

At the Olivetti and Oracle Research Laboratory (ORL) we have been investigating personal computing and communications technologies in collaboration with the University of Cambridge for over ten years. In particular, we have implemented leading edge multimedia, sensor based, and wireless computing systems and networks and have conducted wide ranging experiments in how these can be applied to individuals´ daily lives, both in the office and at home. One of the guiding principles of this work has been to relieve the user of as much of the burden of interaction as possible by adding a sense of the real world to computer systems. The systems we have developed can be partitioned into three layers: at the bottom we have the sensors themselves; in the middle we have a layer of abstraction which takes the varied (and possibly voluminous) output of the range of installed sensors and makes this output available to applications in a succinct, uniform, robust and efficient manner using the latest database and distributed system technologies; and at the top we have the applications which use the processed sensor data provided by the middle layer to adapt their behaviour to the user´s environment and preferences. We look at various sensors that we have built over the years, the abstractions through which the sensor information is presented, the applications that were developed to make use of this information, and the important issues that arose from our experiments in each case. We then discuss emerging technologies and what the future might hold for sensor based personalised computing