high performance computing on graphics processing units: hgpu.org

With the advances in wireless communications large scale Wireless Sensor Networks (WSN) are emerging with many applications. These networks are deployed to serve single objective application, with high optimization requirements such as performance enhancement and power saving. Application specific optimization is achieved using formal models and evaluation based simulations of distributed algorithms (DA) controlling such networks. The WSN design problem is of high complexity, and requires robust methodologies, including simulation support. Although we know several works on WSN simulation, these solutions fail to match requirements on a whole set of desirable criteria: scalability, flexibility, concurrent execution and performance. We present a model based approach of WSN application specification separating network organization from behaviors, allowing them to vary independently. Through this approach, network description can be achieved with high level tools independently from the programming syntax. We have specified and simulated number of WSN protocols with varying objectives and semantics using a time-driven execution model incorporated in the proposed meta-simulator. The empirical analysis has revealed flexibility, scalability and performance. The tool flow targets an Occam compiler producing efficient multi-threaded binaries. We expect the final flow of this project to enable sensor code production out of the simulated specification.