Nikola Bogdanovic

Distributed Incremental-Based LMS for Node-Specific Adaptive Parameter Estimation

We introduce an adaptive distributed technique that is suitable for parameter estimation in a network where nodes have different but overlapping interests. At each node, the parameters to be estimated can be of local interest, global interest to the whole network and common interest to a subset of nodes. To estimate each set of local, common and global parameters, a least mean squares (LMS) algorithm is implemented under an incremental mode of cooperation. Coupled with the estimation of the different sets of parameters, the implementation of each LMS algorithm is only undertaken by the nodes of the network interested in a specific set of local, common or global parameters. Besides obtaining the conditions under which the proposed strategy converges in the mean to the solution of a centralized unit that processes all the observations, a spatial-temporal energy conservation relation is provided to evaluate the steady-state performance at each node across the network. Finally, the theoretical results are validated through generic computer simulations as well as simulation results in the context of cooperative spectrum sensing in cognitive radio networks.

Distributed Diffusion-Based LMS for Node-Specific Adaptive Parameter Estimation

A distributed adaptive algorithm is proposed to solve a node-specific parameter estimation problem where the nodes are interested in estimating parameters that can be of local interest, common interest to a subset of nodes and global interest to the whole network. To address the different node-specific parameter estimation problems, this novel algorithm relies on a diffusion-based implementation of different, yet coupled Least Mean Squares (LMS) algorithms, each associated with the estimation of a specific set of local, common or global parameters. The study of convergence in the mean sense reveals that the proposed algorithm is asymptotically unbiased. Moreover, a spatial-temporal energy conservation relation is provided to evaluate the steady-state performance at each node in the mean-square sense. Finally, the theoretical results and the effectiveness of the proposed technique are validated through computer simulations in the context of cooperative spectrum sensing in Cognitive Radio networks.

Distributed diffusion-based LMS for node-specific parameter estimation over adaptive networks

A distributed adaptive algorithm is proposed to solve a node-specific parameter estimation problem where nodes are interested in estimating parameters of local interest and parameters of global interest to the whole network. To address the different node-specific parameter estimation problems, this novel algorithm relies on a diffusion-based implementation of different Least Mean Squares (LMS) algorithms, each associated with the estimation of a specific set of local or global parameters. Although all the different LMS algorithms are coupled, the diffusion-based implementation of each LMS algorithm is exclusively undertaken by the nodes of the network interested in a specific set of local or global parameters. To illustrate the effectiveness of the proposed technique we provide simulation results in the context of cooperative spectrum sensing in cognitive radio networks.

Distributed incremental-based LMS for node-specific parameter estimation over adaptive networks

We introduce an adaptive distributed technique that is suitable for node-specific parameter estimation in an adaptive network where each node is interested in a set of parameters of local interest as well as a set of network global parameters. The estimation of each set of parameters of local interest is undertaken by a local Least Mean Squares (LMS) algorithm at each node. At the same time and coupled with the previous local estimation processes, an incremental mode of cooperation is implemented at all nodes in order to perform an LMS algorithm which estimates the parameters of global interest. In the steady state, the new distributed technique converges to the MMSE solution of a centralized processor that is able to process all the observations. To illustrate the effectiveness of the proposed technique we provide simulation results in the context of cooperative spectrum sensing in cognitive radio networks.

A prediction-error-based method for data transmission and damage detection in wireless sensor networks for structural health monitoring

In this work, a prediction-error-based method that aims to reduce the amount of data transmission and storage in structural health monitoring wireless sensor networks (WSN) while keeping the damage detection capabilities intact is presented. WSNs offer tremendous promise for accurate and continuous structural monitoring using a dense array of inexpensive sensors and possess many advantages over conventional wired systems, particularly for large civil infrastructures. This paper validates the studied approach with the data collected during a test on a three-story steel frame mounted at the ELSA laboratory of the EU Joint Research Center in Ispra, Italy. The frame is tested with and without steel bracing, which results in non-sway behavior and sway frame responses. The excitation is provided by a shaker mounted on top of the frame. Accelerometer signals are collected by a multi-channel data acquisition system. The proposed method has been tested extensively via recorded experimental data and it offers considerable saving in transmitted energy while at the same time keeping the ability to detect changes or damage in the structure.

Coalitional game theoretic approach to distributed adaptive parameter estimation

In this paper, the parameter estimation problem based on diffusion least mean squares strategies is studied from a coalitional game theoretical perspective. The problem has been modeled as a non-transferable coalitional game and two scenarios have been considered, one where the value function includes only a suitable estimation accuracy criterion and another one in which the cost of coalition formation is taken into account as well. In the former scenario, we first analyze the non-emptiness of the core of the games corresponding to traditional diffusion strategies, and then, we extend the analysis to a recently proposed node-specific parameter estimation setting where the nodes have overlapped but different estimation interests. In the latter scenario, we employ a distributed coalition formation algorithm, based on merge-and-split steps, which converges to a stable coalition structure.

Coalitional games for a distributed signal enhancement application

We consider a scenario in which a number of sensor nodes monitor an area, where several sources are active. Each node has an interest to estimate the signal of a particular source using measurements that, unavoidably, are mixtures of the source signals. Nodes could improve the quality of the signal of interest if they were able to use the signals measured by other nodes, however, in a such a case, communication costs must be properly taken into account. To this end, coalitional game theory is used in our study. In the case where the communication cost is zero, we prove that the cooperation of all nodes is beneficial for all. In contrast, when the communication costs are taken into account, we employ a distributed merge-split coalition formation algorithm to organize the nodes into stable cooperative groups. Simulation results are in accordance with the theoretical findings.

A Coalitional Game Theoretic Outlook on Distributed Adaptive Parameter Estimation

In this paper, the parameter estimation problem based on diffusion least-mean-squares strategies is analyzed from a coalitional game theoretical perspective. Specifically, while selfishly minimizing only their own mean-square costs, the nodes in a network form coalitions that benefit them. Due to its nature, the problem is modeled as a nontransferable game and two scenarios are studied, one where each nodes payoff includes only a suitable estimation accuracy criterion and another one in which a graph-based communication cost is also considered. In the former scenario, we first analyze the nonemptiness of the core of the games corresponding to traditional diffusion strategies, and then, the analysis is extended to a recently proposed node-specific parameter estimation setting where the nodes have overlapped but different estimation interests. In the latter scenario, after formulating a coalitional graph game and providing sufficient conditions for its core nonemptiness, we propose a distributed graph formation algorithm, based on merge-and-split approach, which converges to a stable coalition structure.

Cooperative transmission of measurements in WSN for monitoring applications

Wireless Sensor Networks (WSNs) have recently received great attention from the scientific community, because they hold the key to revolutionize many aspects of the industry and our life. The process of collecting the measurements, acquired by a sensor network into a central sink node, constitutes one of the main challenges in this area of research and is often referred to as the sensor reachback problem. In this work, we extend a recently proposed power and rate allocation algorithm so as to be able to take into account possible cooperation between the nodes in the WSN. The derived power and rate allocation algorithm considers Distributed Source Coding (DSC), in order to reduce the amount of information that must be transmitted to the sink. Under the assumption that there are several very bad channels between nodes and the sink, our method achieves both a lower peak power threshold, as well as reduced total power consumption.