Bullet Environmental Measurement OS for a Tiny CRF-STACK Used in Wireless Network

• Main Authors: Vasanth Iyer, G. Rammurthy, M. B. Srinivas
• Format: Article
• Language: English
• Published: IFSA Publishing, S.L. 2008-04-01
• Series: Sensors & Transducers
• Subjects: State machine stack; Sensor networks; Routing protocols; Fault recognition algorithms
• Online Access: http://www.sensorsportal.com/HTML/DIGEST/march_08/Special_Issue_Vol_90/P_SI_27.pdf

To respond to the new development needs of sensor networks and their unique deployment, there are many available processors and varying target footprints for deploying sensor networks. The most common needs are supporting many different types of wireless radios and an environmental measuring operating system which allows interfacing to 8, 16 bit target microcontrollers. The constraints of the sensor nodes (that form the network) however, are that they have limited processing power, limited wireless range, limited memory, low data transmission rates and low cost packaging. The design of the software stack being proposed in this paper is based on the previous work by the authors who focused on implementation of Control Radio Flooding (CRF) protocol to self organizing sensor networks. The proposed stack, which is fully power-aware, is referred to as CRF-STACK. It integrates the hierarchical space partitioning tree with a data transaction model that allows seamless exchanges between data collecting sensors and its parent nodes in the hierarchy and could be compatible with emerging IEEE standards. This heart of model is a scalable real-time OS which allows a programming interface to develop sensor application and underlying radio communication. Through routing simulations we demonstrate that the energy-aware reusability of resources in sensor networks has a QUOTE O(logN) complexity where N is the total number of nodes. In sensor network the processing power for a given operation is typically measured as a collaborative processing of a group of nodes which is always higher than the individual sensors capabilities, the residual energy remaining of the sensor network at the end of the simulation is a good measure of the collaborative factor (lesser the residual value the better the collaborative factor). We show by simulation the optimal values for reusability to attain max lifetime converges without sensor faults for N <= 20% for CRF and LEACH-E. Even though LEACH-S results peak at routing N=5% it has a complex lifetime with errors with a residual energy after faults at 27.9%. To accommodate the lifetime with faults in the case of LEACH we extend this static sensor network model with a fault-recognition algorithm making the real-time values measured from sensors more fault resistant and reliable throughout its maximum life-time.