Evaluating smart meters’ power usage in the home
Fri 30 Oct 2015
New research examines the link throughput between smart meters and aggregators as secondary users with respect to power and outage constraints.
The aim of Maximizing the Link Throughput between Smart-meters and Aggregators as Secondary Users under Power and Outage Constraints was to evaluate the communication link from smart meters to aggregators as secondary (or unlicensed) users that transmit data over the primary (licensed) channel.
The report establishes that cognitive radios are a solution for effective use of frequency spectrum. The radio nodes should understand their environment to establish the wireless network with two key aims in mind: Reliable communication and efficient usage of the radio spectrum.
With that in mind, Spectrum sharing is a cognitive radio concept that allows secondary users to transmit information without disturbing primary users over the same frequency band. Although challenging, the concept of spectrum sharing has become more widespread in a new cellular system and various sensor applications.
With that in mind, the paper concentrates on spectrum sharing in the deployment of part of the communication network in modern electric power grids (or smart grids). The proposed scenario is based on four assumptions: The positions of the meters and aggregators are fixed, so highly directional antennae are used; Secondary users transmit with less power than primary users; Meta transmissions are co-ordinated so as to avoid packet collisions; and the strength of the secondary links is guaranteed by an outage constraint. Taking all these assumptions on board, the secondary users’ interference can be neglected.
However, uncertainty is caused by the mobile users of the primary network, whose positions and traffic uncertainty are not known. To combat this, the mobile users’ spatial distribution is modelled as a Poisson Point process (a stochastic process that counts the number of events in a given time interval).
A closed-form solution is then obtained for the most achievable throughput with respect to a reference secondary link subject to transmit power/outage constraints.
The paper looked at the results of this concept. To reconstruct the signal, an aggregator was required. For building the signal, a ‘Reference Energy Disaggregation Data Set’ database was used – a 15 minute average power demand over the space of a day. Data from the smart meter is transmitted to the aggregator every 15 minutes which reconstructs the signal. A notable result of this was that the power demand signal had a burst nature with floor level and few peaks. This was a result of the living habits of the house occupants, such as cooking and showering.
Due to the burst nature of the power demand signal transmitted in the smart meter-aggregator link, the outage events didn’t have that large an effect in the signal reconstruction as compared to the perfect transmission.
It was also found that reasonably high outage constraints improved link throughput. While the power constraint is required by the secondary link to not interfere in primary users, the paper said that the outage constraint is set to guarantee a minimum strength at the secondary link.
The paper presented best and worst cases for RMSD (Root-mean-square-deviation) among 53 households, and found that most households performed better than expected. This strengthened the case that signal reconstruction is weakly affected by regular outage events. Allowing for more frequent outage events was found to actually improve link throughput.
This is a result of the opposing effects of the SIR (signal to interference ratio) requirement on the system performance. The paper explained that lower outage constraints resulted in higher SIR constraints. This in turn led to higher spectral efficiency while lowering success probability. If information is relayed to the average power demand, then signal reconstruction can be possible – even with relatively loose outage constraints.
The report found that for the signal of a 15-minute sampling interval of the power demand of a household, using event-based sampling offered a stronger communication link compared with time-based sampling, owing to less redundant data.