Robust Locally Optimum Detection in Auto-Regressive Noise
Abstract
This report represents the culmination of a study focusing on more extensive characterizations of a robust autoregressive locally optimum (ARLO) detector for spread spectrum signals operating in a variety of noise environments. The signal is modeled as corrupted by both Gaussian and Non-Gaussian noise as representative of many interference environments. The normal linear correlator is sub-optimum, as is a fixed non-linearity. The ARLO algorithm, in contrast, uses an adaptive nonlinearity that utilizes probability density function (pdf) estimation techniques to alleviate the need for a priori knowledge of the channel noise characteristics. The ARLO algorithm uses pdf estimation techniques for Independent Identically Distributed (IID) noise samples as the input sequence to the AR model is lID. The goal and results of this effort described herein is to further characterize performance improvements using a more efficient C (rather than MatLab) program.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 2003
- Accession Number
- ADA418500
Entities
People
- Donald R. Ucci
- Michael R. Banys
Organizations
- Illinois Institute of Technology