Affiliation:
1. Electronics and Telecommunication Department, COEP Technological University, Pune, Maharashtra, INDIA
Abstract
The performance, energy efficiency, power dissipation, and cost of the processors used in Internet of Things edge devices can be improved by introducing approximate computing. In this paper, the design and synthesis of an application-specific instruction set processor for approximate computing is proposed. Approximate computing can also be defined by using Artificial Intelligence-based technique like the Fractal theory and Inconsistent Information System. Application Specific Instruction Set Processor is a new processor design area and is used to design a new instruction set using existing processor configuration available on the EDA tool. From the results, it is observed that significant improvement is achieved in processor performance using the TIE application as compared to without the TIE application which is applied to user-designed instruction.
Publisher
World Scientific and Engineering Academy and Society (WSEAS)
Reference14 articles.
1. Nohl, Achim et al. “Application specific processor design: Architectures, design methods and tools.” 2010 IEEE/ACM International Conference on Computer-Aided Design (ICCAD) (2010), San Jose, CA, pp. 349-352.
2. Q. Xu, T. Mytkowicz and N. S. Kim, "Approximate Computing: A Survey," in IEEE Design & Test, vol. 33, no. 1, pp. 8-22, Feb. 2016, doi: 10.1109/MDAT.2015.2505723.
3. P. V.Bhandarkar, S.S.Limaye, “Modeling with LISA- A Power Full Machine Description Language”, IJCSET, Vol-3, Issue 1, 2013, pp. 1-7.
4. Laura Micconi, Jan Madsen, Paul Pop, “System-level synthesis of multi-ASIP platforms using an uncertainty model”, Integration, the VLSI Journal, Volume 51, 2015, pp. 118-138, ISSN: 0167-9260.
5. S. Ramanathan, V. Visvanathan, S.K. Nandy, “Synthesis of ASIPs for DSP algorithms”, Integration, Vol. 28, Issue 1, 1999, pp. 13-32, ISSN 0167-9260.