Demand side management (DSM) makes it possible to adjust the load experienced by the power grid while reducing the consumers' bill. Game-theoretic DSM is an appealing decentralized approach for collaboratively scheduling the usage of domestic electrical appliances within a set of households while meeting the users' preferences about the usage time. The drawback of distributed DSM protocols is that they require each user to communicate his/her own energy consumption patterns, which may leak sensitive information regarding private habits. This paper proposes a distributed privacy-friendly DSM system that preserves users' privacy by integrating data aggregation and perturbation techniques: users decide their schedule according to aggregated consumption measurements perturbed by means of additive white Gaussian noise. We evaluate the noise power and the number of users required to achieve a given privacy level, quantified by means of the increase of the information entropy of the aggregated energy consumption pattern. The performance of our proposed DSM system is compared to the one of a benchmark system that does not support privacy preservation in terms of total bill, peak demand, and convergence time. Results show that privacy can be improved at the cost of increasing the peak demand and the number of game iterations, whereas the total bill is only marginally incremented.
Enabling Privacy in a Distributed Game-Theoretical Scheduling System for Domestic Appliances
ROTTONDI, CRISTINA EMMA MARGHERITA;BARBATO, ANTIMO;VERTICALE, GIACOMO
2016-01-01
Abstract
Demand side management (DSM) makes it possible to adjust the load experienced by the power grid while reducing the consumers' bill. Game-theoretic DSM is an appealing decentralized approach for collaboratively scheduling the usage of domestic electrical appliances within a set of households while meeting the users' preferences about the usage time. The drawback of distributed DSM protocols is that they require each user to communicate his/her own energy consumption patterns, which may leak sensitive information regarding private habits. This paper proposes a distributed privacy-friendly DSM system that preserves users' privacy by integrating data aggregation and perturbation techniques: users decide their schedule according to aggregated consumption measurements perturbed by means of additive white Gaussian noise. We evaluate the noise power and the number of users required to achieve a given privacy level, quantified by means of the increase of the information entropy of the aggregated energy consumption pattern. The performance of our proposed DSM system is compared to the one of a benchmark system that does not support privacy preservation in terms of total bill, peak demand, and convergence time. Results show that privacy can be improved at the cost of increasing the peak demand and the number of game iterations, whereas the total bill is only marginally incremented.File | Dimensione | Formato | |
---|---|---|---|
main.pdf
accesso aperto
Descrizione: manuscript
:
Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione
2.21 MB
Formato
Adobe PDF
|
2.21 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.