無線感測網路中基於資料聚集方法下的安全比較系統

Abstract

無線感測網路在最近幾年快速的發展，廣泛的應用在民生和軍事上，因此安全性的議題也就漸漸被重視。由於感測器是擁有有限的電力，計算能力以及儲存空間，因此要在無線感測網路上建立一套有效率且安全的協定並不容易。關於改善效率，在無線感測網路上存在一個解決方法，稱之為—Data Aggregation(資料聚集)，此方法可以有效的降低資源的消耗以及通訊頻寬。不過由於資料聚集的方式必須是要能在中間的節點 (Aggregator)負責處理資料的整合和計算，因此如何在資料聚集的方法下提供end-to-end的資料保密性就變得是一項挑戰。在近幾年間，有一些研究在解決上述的問題，不過大多數都是集中在解決某一種資料聚集函數像是”加總”。 然而，在感測網路的應用當中也有需要做中位數或是尋找最大最小值等需要做比較運算的資料聚集函數。因此在這篇論文中，我們利用了輕量化的加密演算法且能提供在密文上做資料計算的方法Privacy Homomorphism來達成我們的目標。此外，透過運用在密碼上解決比大小問題的機制我們提出了一套在無線感測網路架構上安全比較的方法，此方法可以做到有效的降低資源的消耗且能在密文上做比較的運算來找出最大最小值。

In recent years, wireless sensor networks (WSNs) continue to grow rapidly and have widely use in both military and civilian. Therefore, security issues of WSNs are more important. Since the sensor nodes have limited power, computation, and storage, it is not easy to establish efficient and secure protocol for WSNs. In terms of efficiency, a countermeasure -- Data Aggregation has been proposed for reducing the resource consumption and communication bandwidth. Since data aggregation needs to compute and aggregate data at the intermediate nodes (Aggregators), how to provide end-to-end privacy with data aggregation in WSNs becomes a challenge. In recent years, some research is interesting in above mentioned secure aggregation for WSNs, but most of them focus on a certain aggregation function like “SUM”. However, a large class of sensor network applications such as median computation or finding maximum/minimum, rely on comparison operations. In this paper, we use a light-weight encryption scheme called Privacy Homomorphism that supports operations over ciphertext for our purpose. Additionally, we propose a secure comparison protocol that achieves the energy benefits and support secure comparison operations over the encrypted values for WSNs by some techniques used to solve “Grater than” problem in cryptographic.