Dynamic reconfigurable architecture for adaptive DCT implementation

In this paper, we propose a reconfigurable architecture for discrete cosine transform (DCT) computation. The objective of the paper is to integrate the DCT computation in a complete embedded system based on ARM processors. Based on dynamic partial reconfigurable FPGAs, different versions of DCT computation are used to give adaptability and flexibility to the architecture. These adaptability responses to different service requirements at run time, such as image quality levels, and system performance. We also explore an efficient management of the reconfigurable area by adjusting the size of the reconfigurable region to the different variable sized hardware module related to the adaptable DCT IP core. The ZedBoard development kit based on the Xilinx Zynq-7000 was used in the study. The results of implementation offer a number of benefits such as optimized hardware resources utilization, efficiently handled the reconfigurable area and reduced reconfiguration time.

[1]  Daniel Llamocca,et al.  A framework for self-reconfigurable DCTs based on multiobjective optimization of the Power-Performance-Accuracy space , 2012, 7th International Workshop on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC).

[2]  S. Rajaram,et al.  FPGA implementation of fully parallel distributed arithmetic based DCT architecture , 2015, 2015 International Conference on Innovations in Information, Embedded and Communication Systems (ICIIECS).

[3]  Abdellatif Mtibaa,et al.  Efficient relocation of variable-sized hardware tasks for FPGA-based adaptive systems , 2014, 2014 26th International Conference on Microelectronics (ICM).

[4]  Serge Weber,et al.  A Novel Framework for the Design of Adaptable Reconfigurable Partitions for the Placement of Variable-sized IP Cores , 2014, IEEE Embedded Systems Letters.

[5]  Khaled Benkrid,et al.  A run-time reconfigurable system for adaptive high performance efficient computing , 2014, CARN.

[6]  Jooheung Lee,et al.  A Self-Reconfigurable Platform for Scalable DCT Computation Using Compressed Partial Bitstreams and BlockRAM Prefetching , 2009, 2009 IEEE Computer Society Annual Symposium on VLSI.

[7]  Earl E. Swartzlander,et al.  DCT Implementation with Distributed Arithmetic , 2001, IEEE Trans. Computers.

[8]  Ayman Alfalou,et al.  A Generalized Algorithm and Reconfigurable Architecture for Efficient and Scalable Orthogonal Approximation of DCT , 2015, IEEE Transactions on Circuits and Systems I: Regular Papers.