Evaluation of Coarse-Grained Reconfigurable Array for a Dual Mode OTFS-OFDM Modulator

被引:0
作者
Hassan, Zohaib [1 ]
Hussain, Waqar [2 ]
Ometov, Aleksandr [1 ]
Lohan, Elena Simona [1 ]
Nurmi, Jari [1 ]
机构
[1] Tampere Univ, Fac ITC, Wireless Res Ctr, Tampere, Finland
[2] Nordic Semicond ASA, Trondheim, Norway
来源
2024 IEEE WORKSHOP ON SIGNAL PROCESSING SYSTEMS, SIPS | 2024年
关键词
CGRA; Reconfigurable Computing; 6G; Gearbox; PHY; OFDM; OTFS; 2D FFT;
D O I
10.1109/SIPS62058.2024.00040
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Orthogonal Time Frequency Space (OTFS) multiplexing, a prospective waveform for 6G offers high reliability under high mobility compared to Orthogonal Frequency Division Multiplexing (OFDM), but it is characterized by relatively high computational complexity. To address this challenge, a dual-mode framework has been proposed in the literature that dynamically switches between OFDM and OTFS as needed. However, processing such a computationally intensive workload is not feasible on general-purpose processors and therefore must be accelerated through an Application-Specific Integrated Circuit (ASIC) or a reconfigurable hardware device. A typical form of a reconfigurable hardware is a Coarse-Grained Reconfigurable Array (CGRA). The concept of CGRA has existed for a few decades, but it is still gaining traction due to its massively parallel processing capabilities and run-time reconfigurability. This paper proposes a CGRA-based modulator to meet the run-time switching requirements for a dual-mode OFDM-OTFS system. The CGRA is synthesized on a Stratix-IV FPGA device for prototyping purposes. Operating at a frequency of 181.0 MHz, the CGRA delivers a performance of 11.58 GOPS with a dynamic power dissipation of 308 mW. The proposed architecture is evaluated for multiple OTFS scales ( 8 x 4, 8 x 8, 16 x 16) and demonstrates that the proposed solution offers promising results in terms of both performance and scalability.
引用
收藏
页码:183 / 188
页数:6
相关论文
共 23 条
[1]   An ML-assisted OTFS vs. OFDM adaptable modem [J].
Ahmed, I. Zakir ;
Sadjadpour, Hamid .
2023 IEEE FUTURE NETWORKS WORLD FORUM, FNWF, 2024,
[2]   AN ALGORITHM FOR MACHINE CALCULATION OF COMPLEX FOURIER SERIES [J].
COOLEY, JW ;
TUKEY, JW .
MATHEMATICS OF COMPUTATION, 1965, 19 (90) :297-&
[3]   A Survey of PAPR Techniques Based on Machine Learning [J].
da Silva, Bianca S. de C. ;
Souto, Victoria D. P. ;
Souza, Richard D. ;
Mendes, Luciano L. .
SENSORS, 2024, 24 (06)
[4]   Low Complexity Implementation of OTFS Transmitter using Fully Parallel and Pipelined Hardware Architecture [J].
Dora, Sai Kumar ;
Mishra, Himanshu B. B. ;
Sahoo, Manodipan .
JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY, 2023, 95 (08) :955-964
[5]  
Fettweis Gerhard, 2021, IEEE BITS the Information Theory Magazine, V1, P71, DOI 10.1109/MBITS.2021.3118662
[6]   OTFS vs. OFDM in the Presence of Sparsity: A Fair Comparison [J].
Gaudio, Lorenzo ;
Colavolpe, Giulio ;
Caire, Giuseppe .
IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2022, 21 (06) :4410-4423
[7]  
Hadani R, 2017, IEEE WCNC
[8]   Coarse-grained reconfigurable architectures for radio baseband processing: A survey [J].
Hassan, Zohaib ;
Ometov, Aleksandr ;
Lohan, Elena Simona ;
Nurmi, Jari .
JOURNAL OF SYSTEMS ARCHITECTURE, 2024, 154
[9]  
Hoffmann M., 2023, IEEE Access
[10]   Designing Fast Fourier Transform Accelerators for Orthogonal Frequency-Division Multiplexing Systems [J].
Hussain, Waqar ;
Garzia, Fabio ;
Ahonen, Tapani ;
Nurmi, Jari .
JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY, 2012, 69 (02) :161-171