A Survey on Design and Performance of Higher-Order QAM Constellations

Praveen Kumar Singya, Parvez Shaik, Nagendra Kumar, Vimal Bhatia, Mohamed‐Slim AlouiniarXiv 2020 · DOI · arXiv

Abstract

As the research on beyond 5G heats up, we survey and explore power and bandwidth efficient modulation schemes in details. In the existing publications and in various communication standards, initially square quadrature amplitude modulation (SQAM) constellations (even power of 2) were considered. However, only the square constellations are not efficient for varying channel conditions and rate requirements, and hence, odd power of 2 constellations were introduced. For odd power of 2 constellations, rectangular QAM (RQAM) is commonly used. However, RQAM is not a good choice due to its lower power efficiency, and a modified cross QAM (XQAM) constellation is preferred as it provides improved power efficiency over RQAM due to its energy efficient two dimensional (2D) structure. The increasing demand for high data-rates has further encouraged the research towards more compact 2D constellations which lead to hexagonal lattice structure based constellations, referred to as hexagonal QAM (HQAM). In this work, various QAM constellations are discussed and detailed study of star QAM, XQAM, and HQAM constellations is presented. Generation, peak and average energies, peak-to-average-power ratio, symbol-error-rate, decision boundaries, bit mapping, Gray code penalty, and bit-error-rate of star QAM, XQAM, and HQAM constellations with different constellation orders are presented. Finally, a comparative study of various QAM constellations is presented which justifies the supremacy of HQAM over other QAM constellations. With this, it can be claimed that the use of the HQAM in various wireless communication systems and standards can further improve the performance targeted for beyond 5G wireless communication systems.

Introduction: The Need for Efficiency Beyond 5G

As we push toward "Beyond 5G" (B5G) and 6G, the demand for high-speed multimedia applications is skyrocketing. To meet this, we need modulation schemes that are both bandwidth-efficient (squeezing more bits into limited spectrum) and power-efficient (preserving battery life in mobile devices).

Standard Square QAM (SQAM) is the industry workhorse but is only efficient for even powers of two (16-QAM, 64-QAM). When a system needs an odd number of bits (like 5 bits for 32-QAM), it typically falls back on Rectangular QAM (RQAM), which is power-hungry and inefficient. Furthermore, standard QAM struggles in mobile "fading" environments, often requiring complex Automatic Gain Control (AGC) and carrier recovery to avoid errors. This paper explores how non-standard geometric shapes—circles, crosses, and hexagons—can solve these efficiency and robustness gaps.

Advanced Constellation Families

The authors conduct a detailed analytical and comparative study of three primary "advanced" constellation families, utilizing mathematical channel models (AWGN, Rayleigh, Nakagami-m) to simulate real-world wireless environments.

Star QAM

Star QAM arranges points in concentric circles. This allows for differential detection, which is much more robust in fading channels because it separates amplitude and phase information. This eliminates the need for complex AGC and carrier recovery, which are prone to "false locks" in high-speed mobile environments.

Evolution of Star QAM constellations from BPSK to 64-Star QAM.

Cross QAM (XQAM)

Designed for odd-bit requirements (e.g., 32, 128 points), XQAM takes a rectangular grid and moves the outer corner points to the axes. This cross shape reduces peak and average power. For 32-point constellations, XQAM achieves a PAPR of 1.70, significantly lower than RQAM’s 2.23.

Hexagonal QAM (HQAM)

HQAM is based on a hexagonal lattice (like a honeycomb), which is the densest possible way to pack points in a 2D plane. Because hexagonal regions are more compact than rectangular ones, HQAM provides roughly a 0.6 dB gain over standard QAM for the same power.

16-HQAM constellation showing the hexagonal lattice structure and decision boundaries.

Key Insights and Performance Results

  • The Gray Code Penalty: A major trade-off in bit mapping is the Gray code penalty. In SQAM, every point has 4 neighbors, making "perfect" Gray coding easy. In HQAM, points have 6 neighbors, making perfect Gray coding impossible. This slightly offsets its power gains.
  • Power Efficiency: HQAM consistently requires the lowest average energy ($E_s/d^2$) to maintain a specific distance between symbols compared to SQAM, RQAM, and XQAM.
  • Irregular HQAM: The authors present an "Irregular HQAM" structure that outperforms regular hexagonal grids by shifting points to be even more compact, further reducing average power requirements.
  • Standards Mapping: The study provides a comprehensive reference mapping different QAM orders across global communication standards like IEEE 802.11 (Wi-Fi), DVB (Digital Video Broadcasting), and 3GPP LTE/5G.

Implementation Challenges and Future Directions

While the benefits are clear, implementation remains a challenge. Detection complexity for HQAM is still computationally more intensive than the simple thresholding used for Square QAM. Furthermore, most current RF hardware is optimized for I/Q (rectangular) coordinates, requiring more sophisticated digital signal processing (DSP) at the receiver.

The authors suggest that the next major leap will come from combining Geometric Shaping (like HQAM) with Probabilistic Shaping to finally reach the theoretical Shannon limit of communication channels.

Summary updated

Abstract

As the research on beyond 5G heats up, we survey and explore power and bandwidth efficient modulation schemes in details. In the existing publications and in various communication standards, initially square quadrature amplitude modulation (SQAM) constellations (even power of 2) were considered. However, only the square constellations are not efficient for varying channel conditions and rate requirements, and hence, odd power of 2 constellations were introduced. For odd power of 2 constellations, rectangular QAM (RQAM) is commonly used. However, RQAM is not a good choice due to its lower power efficiency, and a modified cross QAM (XQAM) constellation is preferred as it provides improved power efficiency over RQAM due to its energy efficient two dimensional (2D) structure. The increasing demand for high data-rates has further encouraged the research towards more compact 2D constellations which lead to hexagonal lattice structure based constellations, referred to as hexagonal QAM (HQAM). In this work, various QAM constellations are discussed and detailed study of star QAM, XQAM, and HQAM constellations is presented. Generation, peak and average energies, peak-to-average-power ratio, symbol-error-rate, decision boundaries, bit mapping, Gray code penalty, and bit-error-rate of star QAM, XQAM, and HQAM constellations with different constellation orders are presented. Finally, a comparative study of various QAM constellations is presented which justifies the supremacy of HQAM over other QAM constellations. With this, it can be claimed that the use of the HQAM in various wireless communication systems and standards can further improve the performance targeted for beyond 5G wireless communication systems.
BibTeX
@misc{singya2020survey,
  author = {Praveen Kumar Singya and Parvez Shaik and Nagendra Kumar and Vimal Bhatia and Mohamed‐Slim Alouini},
  title = {{A Survey on Design and Performance of Higher-Order QAM Constellations}},
  year = {2020},
  eprint = {2004.14708},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2004.14708},
}
Figure