West Virginia University
Optimization of bit interleaved coded modulation using genetic algorithms
Abstract
dc:description.abstractModern wireless communication systems must be optimized with respect to both bandwidth efficiency and energy efficiency. A common approach to achieve these goals is to use multi-level modulation such as quadrature-amplitude modulation (QAM) for bandwidth efficiency and an error-control code for energy efficiency. In benign additive white Gaussian noise (AWGN) channels, Ungerboeck proposed trellis-coded modulation (TCM), which combines modulation and coding into a joint operation. However, in fading channels, it is important to maximize diversity. As shown by Zehavi, diversity is maximized by performing coding and modulation separately and interleaving bits that are passed from the encoder to the modulator. Such systems are termed "BICM" for bit-interleaved coded modulation. Later, Li and Ritcey proposed a method for improving the performance of BICM systems by iteratively passing information between the demodulator and decoder. Such systems are termed "BICM-ID", for BICM with Iterative Decoding. The bit error rate (BER) curve of a typical BICM-ID system is characterized by a steeply sloping "waterfall" region followed by an error floor with a gradual slope.;This thesis is focused on optimizing BICM-ID systems in the error floor region. The problem of minimizing the error bound is formulated as an instance of the Quadratic Assignment Problem (QAP) and solved using a genetic algorithm. First, an optimization is performed by fixing the modulation and varying the bit-to-symbol mapping. This approach provides the lowest possible error floor for a BICM-ID system using standard QAM and phase-shift keying (PSK) modulations. Next, the optimization is performed by varying not only the bit-to-symbol mapping, but also the location of the signal points within the two-dimensional constellation. This provides an error floor that is lower than that achieved with the best QAM and PSK systems, although at the cost of a delayed waterfall region.
Degree
thesis:*- Name thesis:degree_name
- MS
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Lane Department of Computer Science and Electrical Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Doppalapudi, Raghu Chaitanya
Subjects
dc:subject × 1Identifiers
dc:identifier.*- Identifier
- https://researchrepository.wvu.edu/etd/2668