Abstract
In our previous work, general principles to develop efficient architectures for matrix-vector arithmetics and video/image processing were proposed based on high-radix (4,8, or 16) multiplier extensions. In this work, we propose a radix-8 multiplier design and its extension to Multifunctional Architecture for Video and Image Processing (MAVIP). MAVIP may operate either as a programmable unit with DSP-specific operations such as multiplication, multiply-accumulate, parallel addition or as one or another HWA such as matrix-vector multiplier, FIR filter, or sum-of-absolute-difference accelerator. Simulations indicate that being a small device, MAVIP has competitive performance in video coding.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Seidel, P.M., McFearin, L.D., Matula, D.W.: Secondary radix recodings for higher radix multipliers. IEEE Trans. Comput. 54, 111–123 (2005)
Al-Twaijry, H.A., Flynn, M.J.: Technology scaling effects on multipliers. IEEE Trans. Computers 47, 1201–1215 (1998)
Shah, S., Al-Khalili, A.J., Al-Khalili, D.: Secondary radix recodings for higher radix multipliers. In: 12th Int. Conf. On Microelectronics, ICM 2000, pp. 75–80 (2000)
Sam, H., Gupta, A.: Generalized multibit recoding of two’s complement binary numbers and its proof with applications in multiplier implementations. IEEE Trans. Computers 39, 1006–1015 (1990)
Conway, C., Swartzlander Jr., E.: Product select multiplier. In: IEEE 28th Asilomar Conf. On Signals, Systems, and Computers, pp. 1388–1392 (1994)
Schwarz, E., Averill III, R., Sigal, L.: A radix-8 cmos s/390 multiplier. In: 13th IEEE Symposium on Computer Arithmetic, pp. 2–9 (1997)
Williams, T.: US patent No 4,965,762. Mixed Size Radix Recoded Multiplier (1990)
Cherkauer, B., Friedman, E.: A hybrid radix-4/radix-8 low power signed multiplier architecture. IEEE Trans. Circuits and Systems–II, Analog and Digital Signal Processing 44, 656–659 (1997)
Chen, H.Y., Gai, W.X.: US patent No 5,828,590. Multiplier based on a variable radix multiplier coding (1998)
Guevorkian, D., Liuha, P., Launiainen, A., Lappalainen, V.: A family of accelerators for matrix-vector arithmetics based on high-radix multiplier structures. In: Pimentel, A.D., Vassiliadis, S. (eds.) SAMOS 2004. LNCS, vol. 3133, pp. 118–127. Springer, Heidelberg (2004)
Guevorkian, D., Launiainen, A., Lappalainen, V., Liuha, P., Punkka, K.: A method for designing high-radix multiplier based processing units for multimedia applications. IEEE Trans. Circuits and Sustems for Video Technology (2005) (to appear)
Guevorkian, D., Launiainen, A., Liuha, P., Lappalainen, V.: Architectures for the sum of absolute differences operation. In: IEEE Workshop on Signal Processing Systems (SIPS 2002), pp. 57–62 (2002)
Texas Instruments (2001), http://www-s.ti.com/sc/psheets/srst143/srst143.pdf
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Guevorkian, D., Liuha, P., Launiainen, A., Punkka, K., Lappalainen, V. (2005). A Radix-8 Multiplier Design and Its Extension for Efficient Implementation of Imaging Algorithms. In: Hämäläinen, T.D., Pimentel, A.D., Takala, J., Vassiliadis, S. (eds) Embedded Computer Systems: Architectures, Modeling, and Simulation. SAMOS 2005. Lecture Notes in Computer Science, vol 3553. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11512622_35
Download citation
DOI: https://doi.org/10.1007/11512622_35
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-26969-4
Online ISBN: 978-3-540-31664-0
eBook Packages: Computer ScienceComputer Science (R0)