Design of a Delay-Optimized Multiplier for MAC Architectures
i-manager's Journal on Electronics Engineering · 20.09.2026 · 16 (4) · 1
DOI: 10.26634/jele.16.4.11252 ↗Kurz gefasst
Fachartikel in der Zeitschrift i-manager's Journal on Electronics Engineering, online erschienen am 15.09.2026. Verfasst von Yatindra Gaurav und 5 weiteren Autor:innen. Die Arbeit nennt 14 Literaturangaben.
Abstract (Originalsprache)
Multiply–Accumulate (MAC) architectures are widely used in digital signal processing, image processing, and high-performance computing applications where multiplication speed significantly affects overall system performance. This paper presents the design of a delay-optimized multiplier for MAC architectures using modified Half Adder (HA) and Full Adder (FA) structures. In the proposed design, conventional XOR-based logic is replaced with OR, AND, NAND, and NOT gate combinations to simplify carry generation and reduce propagation delay. Furthermore, a modified Full Adder-based 4-2 compressor is used in the partial product reduction stage to reduce the number of addition levels and improve delay performance. The modified HA is used to construct a modified FA, which is implemented in array multiplier and compressor-based array multiplier architectures. The compressor structure is implemented using the modified FA to enhance partial product reduction efficiency. Initially, 4×4 multiplier structures are designed and later extended to 8×8 architectures using hierarchical instantiation and shift-add operations. The proposed multipliers are integrated within an 8-bit MAC unit and compared with the NAND-based FA array multiplier from existing literature. The comparison includes array multiplier based on modified FA and HA and array multiplier based on proposed compressor architectures. The designs are implemented and verified using Xilinx Vivado 2018.2. Performance evaluation based on LUT utilization and total delay shows that the array multiplier based on proposed compressor achieves improved delay performance compared to the reference NAND-based architecture while maintaining comparable hardware utilization. Simulation results verify correct operation of multiplication, accumulation, reset, and enable functionalities within the MAC unit.
Zitieren
APA
Gaurav, Y., Mishra, A., Chauhan, A., Soni, M., Gupta, S. K., & Joshi, A. (2026). Design of a Delay-Optimized Multiplier for MAC Architectures. i-manager's Journal on Electronics Engineering, 16(4), 1. https://doi.org/10.26634/jele.16.4.11252
Vancouver
Gaurav Y, Mishra A, Chauhan A, Soni M, Gupta SK, Joshi A. Design of a Delay-Optimized Multiplier for MAC Architectures. i-manager's Journal on Electronics Engineering. 2026;16(4):1. doi:10.26634/jele.16.4.11252
BibTeX
@article{Gaurav2026,
author = {Gaurav, Yatindra and Mishra, Akanksha and Chauhan, Ayushi and Soni, Muskan and Gupta, Sanjiv Kumar and Joshi, Anant},
title = {Design of a Delay-Optimized Multiplier for MAC Architectures},
journal = {i-manager's Journal on Electronics Engineering},
year = {2026},
volume = {16},
number = {4},
pages = {1},
doi = {10.26634/jele.16.4.11252}
}
EUVIMED