ResearchissuesinSensornetworksPresentedbyBrajendra.pptVIP

Research issues in wireless Sensor networks Presented by: Brajendra Kumar Singh Course: Wireless Communications Systems 88-563 Winter 2007 Semester Instructor: Dr. Kemal E. Tepe Department of electrical and computer engineering University of Windsor, Windsor, Ontario, Canada Introduction Wireless sensor network: Massively distributed, untethered, and unattended systems to cover spatially distributed phenomena in natural obstructed environments A sensor node is a Single-chip systems with Low-power CPU and less memory, Radio or optical communication MEMs-based on-chip sensors Referred to as “motes” or “smart dust” Mote applications are deeply tied to hardware Each mote runs a single application at a time. Driven by interaction with environment Event arrival and data processing are concurrent activities Reliability It must collect data without human interaction for months at a time. No real recovery mechanism in the field except for automatic reboot. In-network processing Self configuring system OS and Programming TinyOS designed for network embedded systems. the core OS requires 400 bytes of code and data memory, combined. a component-based architecture, a simple event-based concurrency model nesC nesC is an extension of C Whole-program analysis nesC is a static language nesC supports and reflects TinyOS’s design two types of components in nesC: modules and configurations. Simulators: Tossim Emstar Others: Shawn, VisualSense, J-Sim etc. Factors influencing design of WSN Fault tolerance, Scalability, Production costs, Operating environment, Sensor network topology, Hardware constraints, Transmission media, Power consumption Data Management Geographic routing challenges Monitor and Maintenance THE PHYSICAL LAYER OPEN RESEARCH ISSUES Modulation schemes: Simple and low-power modulation schemes needed can be either baseband or passband Solution Strategy: overcome signal propagation effects Hardware design: Tiny, low-power,