On Disturbed Sequential Hypothesis Testing
Abstract
Sequential hypothesis testing procedures are known to provide a significant advantage over fixed-sample-size (FSS) hypothesis testing in terms of the average test length. The problem of distributed FSS detection has been treated extensively in the literature recently. On the other hand, the problem of distributed sequential detection has received limited attention. This report deals with some distributed sequential detection procedures and their analysis. The performance of our testing schemes is determined in terms of average sample number (ASN) and operating characteristic (OC) functions. Most of the attention is given to parametric schemes in that a centralized testing scheme based on quantized multi-sensor data and two distributed schemes and without fusion and another with fusion are considered. The case of memoryless grouped data sequential (MLGDS) testing procedure is considered in detail. One distributed nonparametric sequential hypothesis testing procedure is also investigated. Numerical results are presented to demonstrate the superiority of distributed multisensor procedures over the classical single sensor schemes.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1991
- Accession Number
- ADA238691
Entities
People
- Mohammad M. Al-ibrahim
- Pramod Varshney
Organizations
- Syracuse University