| HAL-Inria | Publications, software ... of Inria's scientists |
Conference papers
Nomographic Gossiping for f-Consensus
Abstract : In this paper, we present a novel class of iterative gossip algorithms called nomographic gossiping that partly allow to efficiently achieve a rapid global consensus among nodes/agents in a clustered wireless network with respect to an arbitrary function of the initial states. The algorithms are based on the surprising fact that every real-valued multivariate function has a nomographic representation, which is simply a function of a superposition of a finite number of univariate functions. Since superpositions can be effectively generated via the wireless channel by letting nodes in a cluster transmit simultaneously their pre-processed states to a cluster head, the convergence speed can be significantly increased provided that some connectivity condition between clusters is fulfilled.
Cited literature [17 references]
https://hal.inria.fr/hal-00763402
Contributor : Service Ist Inria Sophia Antipolis-Méditerranée / I3s <>
Submitted on : Monday, December 10, 2012 - 4:59:08 PM
Last modification on : Monday, December 28, 2020 - 10:22:04 AM
Long-term archiving on: : Monday, March 11, 2013 - 12:46:50 PM
Contributor : Service Ist Inria Sophia Antipolis-Méditerranée / I3s <>
Submitted on : Monday, December 10, 2012 - 4:59:08 PM
Last modification on : Monday, December 28, 2020 - 10:22:04 AM
Long-term archiving on: : Monday, March 11, 2013 - 12:46:50 PM
File
p130-goldenbaum.pdf
Explicit agreement for this submission