We present in this paper an investigation on a special class of wireless sensor networks for monitoring critical infrastructures that may extend for hundreds of miles in distances. Such networks are fundamentally different from traditional sensor networks in that the sensor nodes in this class of networks are deployed along narrowly elongated geographical areas and form a chain-type topology. Based on careful analysis of existing sensor network architectures, we first demonstrate the needs to develop new architecture and networking protocols to match the unique topology of chain-type sensor networks. We then propose hierarchical network architecture that consists of clusters of sensor nodes to enable the chain-type sensor networks to be scalable to cover typically long range of infrastructure with tolerable delay in network-wide data collection. To maintain energy efficient operations and maximize the lifetime for such a chain-type sensor network, we devise a smart strategy for the deployment of cluster heads. Protocols for network initialization and seamless operations of the chain-type sensor networks are also developed to match the proposed hierarchical architecture and cluster head deployment strategy. Simulations have been carried out to verify the performance of the hierarchical architecture, the smart node deployment strategy, and the corresponding network initialization and operation protocols.
We present in this paper an adaptive linear neural network architecture called PLSNET. This network is based on partial least-squares (PLS) regression. The architecture is a modular network with stages that are associated with the desired number of PLS factors that are to be retained. PLSNET actually consists of two separate but coupled architectures, PLSNET-C for PLS calibration, and PLSNET-P for prediction (or estimation). We show that PLSNET-C can be trained by supervised learning with three standard Hebbian learning rules that extracts the PLS weight loading vectors, the regression coefficients, and the loading vectors for the univariate output component case (single target values). The PLS information that is extracted by PLSNET-C after training, i.e., three sets of synaptic weights, is used by the PLSNET-P as fixed weights (through the coupling) in its architecture. PLSNET-C can then yield predictions of the output variable given test measurements as its input. Two examples are presented, the first illustrates the typical improved predictive capability of PLSNET compared to classical least-squares, and the second shows how PLSNET can be used for parametric system identification.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.