TY - GEN
T1 - A novel design of CAVLC decoder with low power consideration
AU - Tsai, Tsung Han
AU - Fang, De Lung
PY - 2007
Y1 - 2007
N2 - This paper proposes a novel architecture and its VLSI design for MPEG-4 AVC/H.264 CAVLC decoding. In order to improve throughput of CAVLC decoder, we propose two new methods, which are called MLD (Multi-Level Decoding) and NZS (Non Zero Skip for run_before decoding). By performing parallel operation on level decoder, MLD can decode two levels in one cycle at most situations, and NZS can produce several run_befores in the same cycle. These two methods have the advantages of low complexity and regularity design. According to the evaluation, our design only needs 137 cycles in average for one macroblock decoding. Moreover, the proposed CAVLC decoder can run at 33.5 MHz to meet the real time requirement for H.264 video decoding on 1920 x 1088 resolution. Compared with the previous designs, it can reduce around 29.1% to 71.5% on operation frequency for the same requirement, but not increase the gate count so much. With an aid on a lower operation frequency, it will be suitable for a low power application.
AB - This paper proposes a novel architecture and its VLSI design for MPEG-4 AVC/H.264 CAVLC decoding. In order to improve throughput of CAVLC decoder, we propose two new methods, which are called MLD (Multi-Level Decoding) and NZS (Non Zero Skip for run_before decoding). By performing parallel operation on level decoder, MLD can decode two levels in one cycle at most situations, and NZS can produce several run_befores in the same cycle. These two methods have the advantages of low complexity and regularity design. According to the evaluation, our design only needs 137 cycles in average for one macroblock decoding. Moreover, the proposed CAVLC decoder can run at 33.5 MHz to meet the real time requirement for H.264 video decoding on 1920 x 1088 resolution. Compared with the previous designs, it can reduce around 29.1% to 71.5% on operation frequency for the same requirement, but not increase the gate count so much. With an aid on a lower operation frequency, it will be suitable for a low power application.
UR - http://www.scopus.com/inward/record.url?scp=51349099491&partnerID=8YFLogxK
U2 - 10.1109/ASSCC.2007.4425764
DO - 10.1109/ASSCC.2007.4425764
M3 - 會議論文篇章
AN - SCOPUS:51349099491
SN - 1424413605
SN - 9781424413607
T3 - 2007 IEEE Asian Solid-State Circuits Conference, A-SSCC
SP - 196
EP - 199
BT - 2007 IEEE Asian Solid-State Circuits Conference, A-SSCC
T2 - 2007 IEEE Asian Solid-State Circuits Conference, A-SSCC
Y2 - 12 November 2007 through 14 November 2007
ER -