DAFT-Spread Affine Frequency Division Multiple Access for Downlink Transmission

TL;DR

This work introduces DAFT-s-AFDMA, a DAFT-spread downlink access scheme for AFDM designed to alleviate high PAPR that plagues conventional AFDM and O-AFDMA. It analyzes two chirp-subcarrier allocation strategies—interleaved and localized—through system modeling, PAPR analysis, and MMSE-based reception. The results show that both strategies can outperform O-AFDMA in PAPR, with the interleaved approach preserving BER while the localized approach can incur diversity loss, making the interleaved method more favorable for practical downlink deployments. Overall, DAFT-s-AFDMA demonstrates a viable path to lower PAPR in high-mobility scenarios, enabling more efficient power amplifier usage and broader deployment; future work includes low-complexity detection and uplink adaptations.

Abstract

Affine frequency division multiplexing (AFDM) and orthogonal AFDM access (O-AFDMA) are promising techniques based on chirp signals, which are able to suppress the performance deterioration caused by Doppler shifts in high-mobility scenarios. However, the high peak-to-average power ratio (PAPR) in AFDM or O-AFDMA is still a crucial problem, which severely limits their practical applications. In this paper, we propose a discrete affine Fourier transform (DAFT)-spread AFDMA scheme based on the properties of the AFDM systems, named DAFT-s-AFDMA to significantly reduce the PAPR by resorting to the DAFT. We formulate the transmitted time-domain signals of the proposed DAFT-s-AFDMA schemes with localized and interleaved chirp subcarrier allocation strategies. Accordingly, we derive the guidelines for setting the DAFT parameters, revealing the insights of PAPR reduction. Finally, simulation results of PAPR comparison in terms of the complementary cumulative distribution function (CCDF) show that the proposed DAFT-s-AFDMA schemes with localized and interleaved strategies can both attain better PAPR performances than the conventional O-AFDMA scheme.

Figures (6)

• Figure 1: Transceiver structure of the proposed DAFT-s-AFDMA scheme.
• Figure 2: An example of the proposed DAFT-s-AFDMA schemes with the transmitted symbols in the DAF domain for the $k$-th user, where $N=16$, $M=4$ and $K=4$.
• Figure 3: An example of the proposed DAFT-s-AFDMA schemes with the transmitted symbols in the time domain for the $k$-th user, where $N=16$, $M=4$ and $K=4$.
• Figure 4: Comparison of the BER performance of the proposed DAFT-s-AFDMA schemes with interleaved and localized strategies, and the conventional O-AFDMA scheme, where $N=1024$ and $[K, P,{\alpha_{\rm max}},{l_{\rm max}}]=[4, 3, 1, 1]$ or [8, 6, 4, 2].
• Figure 5: The effect of DAFT parameter $\lambda_2$ and $\lambda'_2$ on the BER performance of the proposed DAFT-s-AFDMA schemes with interleaved and localized strategies, where $N=1024$, and $[K, P,{\alpha_{\rm max}},{l_{\rm max}}]$=[8, 6, 4, 2].