An Improved GF(2) Matrix Inverter with Linear Time Complexity

Jasinski, Ricardo P.; Pedroni, Volnei A.; Gortan, Antonio; Godoy, Walter

doi:10.1109/reconfig.2010.86

Cited by 8 publications

(10 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…(1). Our implementation builds on earlier matrix inversion approaches in [15] and [16] and balances hardware complexity with the speed of inversion operation, as verified experimentally in Subsection III.A.…”

Section: Matrix Inversionmentioning

confidence: 99%

“…For the ith (i ≠ 1) row, if both entries in the first column are "1," then the ith row is replaced by the XOR of itself with the first row. Row swapping in [15,16] is done either by a search, at the expense of speed, or through the use of auxiliary memory, at the expense of complexity.…”

Section: Matrix Inversionmentioning

confidence: 99%

“…1 has three primary innovations compared to the approaches in [15] and [16]. while the elements of D are typically many bytes in size.…”

Section: Matrix Inversionmentioning

confidence: 99%

“…Implemented with the same matrix sizes, our design is compared to [16,15] and direct implementation in terms of resource usage and time complexity, listed in Tables I and II, respectively. Synthesis reports for Xilinx and Altera devices are obtained from ISE Design Suite v13.1 [25] and Quartus II v13.0 [26], respectively.…”

Section: Was Coded In Verilog and Synthesized To The Xilinx Spartan-3mentioning

confidence: 99%

“…There are three main contributions in this paper. First, we present an efficient matrix inversion algorithm suitable for hardware implementation that balances the time and hardware complexities of previous approaches [15,16] architecture of a Raptor 10 decoder based on the proposed matrix inversion algorithm is presented. A pipelined architecture for the receiver is developed with the aim of maximizing link throughput at the application layer.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Raptor-Coded Free-Space Optical Communications Experiment

Liu

Hranilovic

2016

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

While free-space optical (FSO) links have many inherent advantages, such as high data rates and license-free operation, they are sensitive to atmospheric conditions. A promising solution to improve the reliability of such communication links is to combine FSO and radio frequency (RF) channels using Raptor codes. This paper presents a first in-field demonstration of real-time Raptor-coded FSO communication over a 1.87 km urban link. This work serves as an important initial step toward field-programmable gate array (FPGA)-based implementation of Raptor-coded hybrid FSO/RF links. The hardware architecture of a Raptor 10 decoder based on a novel resource-and time-efficient matrix inversion algorithm is presented and implemented in an FPGA platform. The receiver is implemented in a pipelined architecture to maximize link utilization. The throughput of the Raptorcoded FSO link is measured both experimentally and numerically under real-life weather conditions for a continuous period of 8 h. Experimental results demonstrate that Raptor-coded FSO links are capable of tracking changes in the channel state to maximize link utilization.

show abstract

Section: Matrix Inversionmentioning

confidence: 99%

Section: Matrix Inversionmentioning

confidence: 99%

“…1 has three primary innovations compared to the approaches in [15] and [16]. while the elements of D are typically many bytes in size.…”

Section: Matrix Inversionmentioning

confidence: 99%

Section: Was Coded In Verilog and Synthesized To The Xilinx Spartan-3mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%