具隱私性之智慧電表系統

Abstract

智慧電網將成為新一代的電力網系統。智慧電網結合了智慧與自動化到傳統電網系統中，藉此提高能源運用的效率，電力系統的可靠性與整體供電品質。 智慧電網需要佈建智慧電表，建立智慧電表基礎建設，與提供智慧電網應用。其中在智慧電表基礎建設中，每個用電戶的用電量將以微小的時間刻度為單位來測量並記錄測量的數據，現行的智慧電表約5~60分鐘記錄一次用電量，未來將進化到以秒為單位進行記錄，許多智慧電網應用都需要用到這些詳細的用電資訊，例如用電量監測，自動化電費帳單系統，以及動態電價。但是，這些詳細的讀值可以透露用電戶每天的生活型態，這嚴重威脅到了用電戶的個人隱私。各種不同的智慧電網應用對電表讀值的時間刻度有不同的需求，我們想要研究這其中的差異並且定義出合理且正式的隱私權需求，更進一步的，我們希望能夠提出同時能夠保障使用者隱私權又能夠達到智慧電網應用功能的智慧電表系統。因為智慧電表只有有限的計算與儲存資源，所以需要大量運算或儲存空間的設計不適合智慧電表系統。我們的目標是使用成本較低的運算方式來設計一個具使用者隱私性的智慧電表系統。

The emergence of smart grids has established a trend towards building our next generation of power grid systems. Smart grids integrate intelligence and automation into the conventional power grid system to increase energy efficiency and improve system reliability and quality. One essential technology of smart grids is advanced metering infrastructure. Advanced metering infrastructure provides fine-grained meter readings within a very short period of time per household. Current smart meters record electricity usage every 5 to 60 minutes. The next generation of smart meters will upgrade a time unit to seconds. Fine-grained meter readings enable various applications in advanced metering infrastructure, such as load monitoring, automatic billing, and dynamic pricing. However, those meter readings threaten personal privacy by implying a sketch of daily activities of households. Actually, the granularity of meter readings often exceeds the need of some underlying applications. We consider that different applications demand different granularities and study the privacy issue in a smart metering system. We would like to study and define privacy model for smart grid applications and to support smart grid applications in a privacy preserving manner. Computation-expensive solutions may not suitable in our scenario due to the limited resources of a smart meter. Lightweight solutions to privacy preserving smart metering systems are our main goals.