Decentralized Sequential Hypothesis Testing in Sensor Network
by
Yajun Mei
Georiga Tech

The decentralized sequential hypothesis testing problem is studied in sensor networks, where a set of sensor receive independent observations and send summary messages to the fusion center, which makes a final decision on which of two hypotheses is true. In this talk we propose to tackle this problem by combining some classical ideas from sequential hypothesis testing and sequential experiment design. In particular, in the scenario where the sensors do not have full access to their past observations, a simple but asymptotically optimal sequential tests is developed, in which sensor message functions are what we call "tandem quantizer," where each sensor only uses two different sensor quantizers with at most one switch between these two quantizers.

Date received: March 6, 2007