Designs and Applications of Bit Interleaved Coded Modulation Systems for Noncoherent Communications
Date Issued
2015
Date
2015
Author(s)
Tsai, Hua-Lung
Abstract
Bit-interleaved coded modulation (BICM), is a coded modulation system constructed by serially concatenating a channel encoder, an interleaver and a symbol mapper. BICM has the structure of the serial concatenated turbo code which can be decoded in an iterative way using message passing. The impressive performance of BICM-ID implicitly assumes perfect synchronization, i.e., the carrier phase, frequency offset can be recovered accurately before data detection. To overcome the unknown carrier phase for BICM-ID systems is yet a very challenging task since the receiver usually operates at extremely low SNR values. In this thesis, we study BICM system for noncoherent communication without pilots which can overcome the unknown carrier phase. For the first design, we consider a noncoherent communication system that can be used in the communication environment in which the frequency estimation is only roughly obtained and the phase synchronization is not employed. The transmitter side has the serial turbo coding structure composed of a convolutional encoder followed by an interleaver, an MPSK signal mapper and a differential modulator. At the receiver side an expanded trellis is employed to represent not only the possible code paths but also the paths resultant from the possible phase variations due to the frequency offset and the phase noise. For the second design, we study a new two/three-level differentially encoded 16/64QAM (TD16/64QAM) design which is suitable for tackling the CFO problem. This new TD16/64QAM is superior to the previously known TD16/64QAM in the bit-interleaved coded modulation (BICM) system corrupted by CFO. For overcoming the carrier frequency problem, we also address the problem of carrier frequency offset estimation for the BICM using TD QAM, in which the symbol phase can be roughly estimated by the $4th$ power method without pilots and can be more correctly estimated by the Taylor series algorithm.
Subjects
BICM-ID
Differential modulation
noncoherent communications
phase ambiguity
4th power algorithm
frequency offset estimation
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-104-D95942013-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):09a1172e1f531dfbb5705c53a7f5a505