Performance evaluation of AMR simultaneous polling strategies in a PRIME PLC network

PLC is gaining prominence as a solution for Smart
Grid developments. However, several aspects still require further
research and analysis. Among the different solutions, PRIME
(PoweRline Intelligent Metering Evolution) standard is one of the
most popular and extended implementations.
This paper analyses the performance of an Auto Meter Reading
(AMR) process for a PRIME PLC network regarding two
aspects. Firstly, the impact caused by the polling strategy utilized.
In particular, the number of connections that the Master Node
manages simultaneously is studied. Secondly, the influence of the
positions of the Switch Nodes in the network is examined. The
time required to read all the meters in the network is utilized as
performance metric.
In order to replicate the performance of a PRIME network, a
co-simulation framework that combines Matlab and OMNeT++
has been used. This architecture allows to take into account both
physical phenomena and control and application message
management. Simulated topologies represent a general European
low-voltage network. Results highlight the importance of
selecting an adequate number of simultaneous connections in
order to optimize the use of the channel, as well as the impact of
the logical structure of the network.