Multidimensional Householder based high-speed QR decomposition architecture for MIMO receivers
Iput Heri Kurniawan, Ji Hwan Yoon, Jongsun Park
Abstract
Iput Heri Kurniawan, Ji Hwan Yoon, Jongsun Park
Abstract
Conventional QR decomposition (QRD) hardware with a large size of channel matrix suffers from very low throughput and large latencies. This paper presents a high speed multi-dimensional (M-D) coordinate rotation digital computer (CORDIC) based QRD architecture. The novel high speed M-D architecture is enabled by exploiting multiple annihilations in a single CORDIC operation and removing data dependencies between two CORDIC operations (evaluation and application CORDIC) in Householder-based QRD process. The proposed QRD architecture can compute 4×4 complex R matrix for every 8 clock cycles. Our QRD hardware for 4×4 channel matrix was implemented using Samsung 0.13μm CMOS process, and the experimental results show that the proposed architecture achieves 4.74x speed-up compared to the conventional hybrid M-D based QRD.
OpenAlex reports 17 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Conventional QR decomposition (QRD) hardware with a large size of channel matrix suffers from very low throughput and large latencies. This paper presents a high speed multi-dimensional (M-D) coordinate rotation digital computer (CORDIC) based QRD architecture. The novel high speed M-D architecture is enabled by exploiting multiple annihilations in a single CORDIC operation and removing data dependencies between two CORDIC operations (evaluation and application CORDIC) in Householder-based QRD process. The proposed QRD architecture can compute 4×4 complex R matrix for every 8 clock cycles. Our QRD hardware for 4×4 channel matrix was implemented using Samsung 0.13μm CMOS process, and the experimental results show that the proposed architecture achieves 4.74x speed-up compared to the conventional hybrid M-D based QRD.
Key concepts: CORDIC, QR decomposition, Computer science, Process (computing), Matrix decomposition, Matrix (chemical analysis), Throughput, Parallel computing