Abstract
This paper presents a generic methodology to optimize constellations based on their geometrical shaping for bit-interleaved coded modulation (BICM) systems. While the method can be applicable to any wireless standard design it has been tailored to two delivery scenarios typical of broadcast systems: 1) robust multimedia delivery and 2) UHDTV quality bitrate services. The design process is based on maximizing the BICM channel capacity for a given power constraint. The major contribution of this paper is a low complexity optimization algorithm for the design of optimal constellation schemes. The proposal consists of a set of initial conditions for a particle swarm optimization algorithm, and afterward, a customized post processing procedure for further improving the constellation alphabet. According to the broadcast application cases, the sizes of the constellations proposed range from 16 to 4096 symbols. The BICM channel capacities and performance of the designed constellations are compared to conventional quadrature amplitude modulation constellations for different application scenarios. The results show a significant improvement in terms of system performance and BICM channel capacities under additive white Gaussian noise and Rayleigh independently and identically distributed channel conditions.
| Original language | English |
|---|---|
| Article number | 7883869 |
| Pages (from-to) | 603-614 |
| Number of pages | 12 |
| Journal | IEEE Transactions on Broadcasting |
| Volume | 63 |
| Issue number | 4 |
| DOIs | |
| Publication status | Published - Dec 2017 |
| Externally published | Yes |
Keywords
- BICM channel capacity
- QAM
- coding and modulation
- constellation design
- non uniform constellation
- signal shaping processing for transmission