Národní úložiště šedé literatury Nalezeno 5 záznamů.  Hledání trvalo 0.01 vteřin. 
Logarithmic number system and floating-point arithmetics an FPGA
Pohl, Zdeněk
An introduction to a logarithmic number system (LNS) is presented. Range and procision of this arithmetic is briefly discussed. We show that the LNS arithmetics is suitable for a FPGA implementation. A case study will compare parameters of our LNS arithmetic library to a conventional floating-point arithmetic.
Utilization of the HSLA toolbox for the FPGA prototyping
Pohl, Zdeněk ; Líčko, M.
An innovative design-flow for DSP algorithms usign high-speed logarithmic arithmetic (HSLA) toolbox is introduced on a demo RLS lattice application. Utilization of toolbox provides effective design time shortening and it makes designer work easier. Firstly, scripts in Matlab are written and checked. From working scripts, design is decomposed in Simulink to cycle-exact simulation and rewritten in Celoxica DK1 tool. Finally, hardware is targeted from DK1 and results are compared with simulations.
Recursive Finite Interval Constant Modulus Algorithm for blind equalization
Heřmánek, Antonín ; Regalia, P.
In the paper, we reexamine the Finite Interval Constant Modulus Algorithm (FI-CMA) proposed by P.Regalia and we present our modifications leading to its recursive form. The proposed modifications are based on recursive updates of QR decomposition. The presented results shows the computation savings and ability to use Recursive FI-CMA for tracking time variant channels.
Floating-Point-Like Arithmetic for FPGA
Matoušek, Rudolf ; Líčko, Miroslav ; Heřmánek, Antonín ; Softley, C.
In recent years we have investigated the use of a logarithmic number representation as an alternative to floating-point. Efficient techniques have been developed to facilitate arithmetic comparable to single precision floating-point in the logarithmic domain.

Chcete být upozorněni, pokud se objeví nové záznamy odpovídající tomuto dotazu?
Přihlásit se k odběru RSS.