Research Initiation: Optimum Signal Detection in Asynchro- nous Gaussian Multiple-Access Channels

研究启动：异步高斯多址信道中的最优信号检测

• 批准号: 8504752
• 负责人: Sergio Verdu
• 金额: $ 6万
• 依托单位: Princeton University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1985
• 资助国家: 美国
• 起止时间: 1985-05-01 至 1988-04-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8504752&HistoricalAwards=false
• 关键词: Research; Initiation; Optimum; Signal; Detection

The objective of this research project is to develop analytical tools for the study of the error probability and computational complexity of decentralized multi-user detectors and the performance limits achievable by coding. To approach the analysis of the error probability of these signal detection problems, the techniques of optimum sequence detection in colored non-Gaussian noise will be applied. Computational complexity will be studied to determine whether polynomial-time rules exist with near-minimum probability of error. The primary objectives of the analysis of performance limits are to determine whether existing results that assume symbol-synchronization among users holds in the completely asynchronous case, and to analyze the dependence of system performance on signal energies.

本研究项目的目标是开发分析工具，用于研究分散的多用户检测器的错误概率和计算复杂性以及通过编码可以实现的性能限制。为了分析这些信号检测问题的误差概率，本文将应用有色非高斯噪声中最优序列检测技术。计算复杂度将被研究，以确定多项式时间规则是否以接近最小的误差概率存在。性能限制分析的主要目标是确定假设用户之间符号同步的现有结果在完全异步情况下是否成立，并分析系统性能对信号能量的依赖性。