Pervasive Computing and Object Self-Synchronization

Recent and Current Research

Self-Synchronization of Mobile Objects

Principal Investigator(s): Vijay Kumar, Ph.D.
Partner(s): Ph.D. student Amol Khedkar is working on this project, and at present is developing the prototype design.
Goal: This project as initiated by Vijay Kumar, Ph.D. at AFRL, Rome, NY in the summer of 2009 under AFRL summer research faculty fellowship program. Funding for this project was obtained from Air Force Research Laboratory (AFRL), Rome, NY. The project is developing a scheme under which mobile objects such as vehicles, unmanned airspace vehicles (UAV), etc., will synchronize the use of common resources such as road intersections, sky, etc., mutually. Thus, our scheme will not require the help of a third party (traffic lights, traffic signs, etc.) in deciding which vehicle or set of vehicles will enter the intersection and which set will wait. We call this discipline as self-synchronization. There are only two ways to manage the sharing activity: (a) with the help of a third party and (b) mutually. In the first scheme, a third party tracks the movement of objects wish to share a common resource and enforces the laws of sharing. We restate the situation as follows: the third party enforces synchronization through intelligent tracking of objects. The second scheme eliminates the need of the third party and objects wish to synchronization track each other and achieve self-synchronization. This scheme is highly-cost effective, fair, reliable and self-oriented. The design and simulation of our scheme are already complete. Funding to build a prototype has been requested from AFRL and funding from NSF is being requested.

Object Discovery, Identification, and Binding (DIB)

Principal Investigator(s): Vijay Kumar, Ph.D.
Funded by: Funding to develop DIB system has been requested from AFRL and also from AFOSR.
Goal: This project is initiated by Vijay Kumar at AFRL, Rome, NY in the summer of 2010 under AFRL summer research faculty fellowship program. Funding for this project is being requested from AFRL and AFOSR (Air Force Office of Scientific Research), Washington DC. The objective of this project is to design, develop and implement a scheme for discovering a set of desired objects (mobile or stationary), identify a specific object from the set of discovered objects, and bind them to a set of sensors for capturing their state in real time. The DIB system when completed will become a part of a system called N-CET developed by AFRL researchers. Object discovery, identification and binding is frequently deployed in real life. Police track suspicious vehicles, air towers track airplanes, some UAVs are deployed to track terrorist cells, etc.

Management of Malicious Transactions

Principal Investigator(s): Vijay Kumar, Ph.D.
Partner(s): Praveen Rao, Ph.D.; Ph.D. student John Blakeney is developing the management system.
Funded by: Funding for this project is being requested from the National Science Foundation (NSF)
Goal: The objective of this project is to protect financial institutions from insiders’ threat. It is relatively harder to secure institutions, especially financial, from insiders’ than from outsiders’ threats. A bank manager has total access to customers’ account and he or she is free to execute all types of transactions on their accounts. However, within the privilege he or she is “not supposed to do” some actions. For example, the manager can initiate a fund transfer transaction to transfer fund from a customer’s account to his or her own account but he/she is “not supposed to” execute this malicious transaction. A malicious transaction management system is being developed in SCE.

Immediate Identification and Notification System for Loosely Connected Disaster Area

Principal Investigator(s): Cory Beard, Ph.D.
Partner(s): Vijay Kumar, Ph.D., Yugyung Lee, Ph.D., Praveen R. Rao, Ph.D., and Daniel Leon-Salas, Ph.D.
Funded by: Funding for this project is being requested from the National Science Foundation (NSF)
Goal: The objective of this project is to efficiently and intelligently deploy database discipline, sensor technology, and wireless and mobile communication technology to minimize the damaging effect of earthquakes. Recently, they have generated unimaginable devastation to many parts of the world such as Haiti, Chile, Mexico, etc. The recent occurrences of earthquakes motivated us to engage into a serious study of their effects and the deployment of innovative approaches based on the state of the art technology to minimize the amount of damages they cause, especially on human lives.

Secured Patient Health Care System

Principal Investigator(s): Vijay Kumar, Ph.D.
Partner(s): Ph.D. student Debargh Acharya is developing this system.
Goal: The objective of this project is to secure the patient health care system. The system that we are developing aims to permit only authorized access to patient database. This has become an important requirement since the federal government has mandated the storage of data digitally. A system to provide high power security and trust through secured communication, trust through multi-stage verification, and data access through strong authentication has become a necessity.

Adaptive Query Processing in P2P

Principal Investigator(s): Vijay Kumar, Ph.D.
Partner(s): Ph.D. student Francisco Landeras (Sprint employee) is developing this system. Francisco defended his Ph.D. dissertation in August 2010.
Goal: The objective of this project is to develop an adaptive query processing system for peer-to-peer architecture. In P2P systems a large number of peers join and leave the network in a random fashion, where queries are answered by the peers present at that time. Since participating peers present different levels of dynamicity, autonomy, heterogeneity, data distribution and replication, the environments is highly unpredictable and changeable. Although there have been significant research activities in adaptive query processing (AQP), they are specialized and isolated techniques that work on their specific context and they typically are not easily extendable. To address these limitations; a new P2P adaptive query processing architecture based on an adaptation loop model that integrates the merits of proactive and reactive methods to support a variety of plan-based adaptive query techniques is being developed.