The Case for a Signal-Oriented Data Stream Management System

Sensors capable of sensing phenomena at high data rates—on the order of tens to hundreds of thousands of samples per second—are useful in many industrial, civil engineering, scientific, networking, and medical applications. In these applications, high-rate streams of data produced by sensors must be processed and analyzed using a combination of both event-stream and signal-processing operations. This paper motivates the need for a data management and continuous query processing architecture that integrates these two different desired classes of functions into a single, unified software system. The key goals of such a system include: the ability to treat a sequence of samples that constitute a “signal segment” as a basic data type; ease of writing arbitrary event-stream and signal-processing functions; the ability to process several million samples per second on conventional PC hardware; and the ability to distribute application code across both PCs and sensor nodes.

• Type: Conference
• Year: 2007
• Where: CIDR
• Authors: Lewis Girod, Yuan Mei, Ryan Newton, Stanislav Rost, Arvind Thiagarajan, Hari Balakrishnan, Samuel Madden
• Keywords: Algorithms, Basic Data Type, CIDR 2007, Continuous Query Processing, Signal-processing Operations