在無線感測網路中使用隱蔽式安全技術之資料聚合

Abstract

無線感測網路(Wireless Sensor Networks)最主要的應用為監測目標環境，並將蒐集到的資料傳回基地台，以供使用者作分析或處理。由於無線感測資源的限制，在蒐集資料並同時對資料做處理，可以減少封包的傳送量，此種技術稱為資料聚合。然而在聚合的功能被執行時，攻擊者可能會針對聚合節點(Aggregator)進行攻擊，進而產生錯誤的聚合結果。近年來，許多學者提出安全資料聚合的機制，但在安全性與通訊負擔方面的考量，往往不夠周全。本論文提出三種分組演算法作比較，分別為Top-down分組演算法、Bottom-up分組演算法與以聚合節點演算法，將網路分為多個群組，其中由實驗結果可以看出以聚合節點分組演算法最有效率，與Top-down分組演算法、Bottom-up分組演算法相比，大約可以節省16%~36%的加解密次數，以減少資料傳輸時所耗費的計算時間與電力。此外我們提出一個使用隱蔽式安全技術的金鑰分配機制，其時間複雜度只要 ，使節點產生群組金鑰(Group Key)，保證資料傳輸的安全性。而每個節點所儲存的金鑰只要儲存一個金鑰分割區塊(Key Partition)、一個群組金鑰參數(Group Key Parameter)與雜湊函數即可，由於只使用對稱式加密與雜湊函數，可減輕節點的計算負擔。另外，本機制符合大部分的安全需求，並可以抵擋無線感測網路上一些著名的攻擊，證明了本機制不僅資源消耗少，也可兼具安全性。

The major application of WSN (Wireless Sensor Networks) is to monitor the environment and transmit the sensed data to the base station. Because sensor nodes are resource limitation, we need a technique called data aggregation to collect and process the sensed data in networks. The data aggregation technique can reduce the packet numbers of data transmitted. However, the attacker may attack the aggregator node to cause the wrong aggregation result. Although different secure data aggregation schemes were proposed recently, they are not complete on the considerations of security and communication overhead. This paper proposes a key distribute scheme that bases on an implicit security technology. We propose and compare 3 grouping algorithms, Top-down grouping algorithm, Bottom-up algorithm, and ABG (Aggregator Based Grouping) algorithm. Experiments show that ABG can save about 16%~ 36% computations of encrypted and decrypted data. Each node generates a group key and stores a group parameter to protect the data, and the time complexity of generating key is . Each node stores a group parameter and a hash function. Because of the scheme uses the symmetric encryption and hash function, it can reduce the computation overhead of nodes. The scheme conform most of security requirement, and it can resist some well-know attacks. So the scheme not only consumes less resource, but also safety.