Unformatted text preview: de decoding. Reduced State Sequence Estimation (RSSE) approaches, which have recently be proposed for large-state problems, can also be eﬀective. Kevin Buckley - 2007 11 2.5 Symbol-by-Symbol MAP
This Subsection is an extension of the discussion in Subsection 5.1.5 of the Course Text. In Sections 2.2 and 2.3 of the Course, we described symbol detection and sequence estimation, respectively. The basic point of Section 2.2 is that if, in our sampled matched ﬁlter output observation, there is no symbol memory, optimum processing is symbol-by-symbol decoupled. We can optimally estimate one symbol at a time using only the observation over that symbol time. In Section 2.3 we saw that if there is symbol memory in our observation, MLSE requires joint processing of all the data to optimally estimate the symbol sequence. In understanding the signiﬁcance of this current Section, the key point is that MLSE does optimum sequence estimation. What if we want optimum symbol estimation when our observation has symbol memory? This is the question we address below. In this Section we discuss what is known in t...
View Full Document
This document was uploaded on 10/12/2009.
- Spring '09