Browsing by Author "Chen, Ying-Liang"

Now showing 1 - 4 of 4

CDC42 Negatively Regulates Testis-Specific SEPT12 Polymerization
(2018-09-01)
Huang, Chia-Yen
;
Wang, Ya-Yun
;
Chen, Ying-Liang
;
Chen, Mei-Feng
;
Chiang, Han-Sun
Septin (SEPT) genes encode well-preserved polymerizing GTP-binding cytoskeletal proteins. The cellular functions of SEPTs consist of mitosis, cytoskeletal remodeling, cell polarity, and vesicle trafficking through interactions with various types of cytoskeletons. We discovered that mutated SEPTIN12 in different codons resulted in teratozoospermia or oligozoospermia. In mouse models with a defective Septin12 allele, sperm morphology was abnormal, sperm count decreased, and sperms were immotile. However, the regulators of SEPT12 are completely unknown. Some studies have indicated that CDC42 negatively regulates the polymerization of SEPT2/ 6/ 7 complexes in mammalian cell lines. In this study, we investigated whether CDC42 modulates SEPT12 polymerization and is involved in the terminal differentiation of male germ cells. First, through scanning electron microscopy analysis, we determined that the loss of Septin12 caused defective sperm heads. This indicated that Septin12 is critical for the formation of sperm heads. Second, CDC42 and SEPT12 were similarly localized in the perinuclear regions of the manchette at the head of elongating spermatids, neck region of elongated spermatids, and midpiece of mature spermatozoa. Third, wild-type CDC42 and CDC42Q61L (a constitutive-acting-mutant) substantially repressed SEPT12 polymerization, but CDC42T17N (a dominant-negative-acting mutant) did not, as evident through ectopic expression analysis. We concluded that CDC42 negatively regulates SEPT12 polymerization and is involved in terminal structure formation of sperm heads.
Cost-Efficiency FFT Using Hardware-Reduction and Dynamic Current Scaling Approaches
(2014)
Chen, Ying-Liang
;
Hsu, Terng-Yin
;
資訊工程學系
;
Department of Computer Science
In this paper, we disclosure a cost-efficiency FFT via hardware-reduction and dynamic current scaling (DCS) schemes in a low-power OFDM modem. These reductions are realized by the pipelined data path modifying and the hardware sharing at two stages. For low power, the operating current is scalable to reduce power consumption, namely DCS. All our works are synthesized and simulated by the TSMC 65nm CMOS technology. It can perform high efficient computing power at FFTs/Energy.
Error-Tolerant and Energy-Efficient FFT with Dynamic Current and Frequency Scaling (DCFS)
(2015)
Chen, Ying-Liang
;
Hsu, Terng-Yin
;
資訊工程學系
;
Department of Computer Science
This work utilizes the dynamic current scaling (DCS) to control the pMOS array for outputting the sufficient operating current. It makes the FFT modules work under the lower power-consummation operating mode. Furthermore, to guarantee the DCS current is enough for the whole system. We need the online error detection. In this work, we just apply the adaptive sequential elements to determine the critical path delay and whether the timing constraint is met or not. When the error occurred, the system will perform the re-computing process to increase clock rate to recover errors via dynamic frequency scaling (DFS). All this work was synthesized and simulated with TSMC 65nm technology.
Low-Complexity Beam Steering Algorithm Based on Binary Search Strategy for Receive Beamform
(2012)
Lin, You-Hsien
;
Tsai, Tsung-Yu
;
Chen, Ying-Liang
;
Ko, Ta
;
Chien, Chun-Che
The complexity of beamforming, such as complexity complex weighting and combining process, are extremely important for implementation. This work presents a low complexity time-domain statistic noise suppression and beam steering algorithm based on binary search strategy to estimate the optimal beam vector at the receivers. The beam steering algorithm based on binary search strategy could achieve closely (about 99%) optimal SNR gain within only two binary search. It is desirable choice to archive the suitable trade-off between the performance and the processing overhead. Hence, it could obtain a near-optimal performance with considerably low complexity and high convergence speed. In addition, this algorithm is well-matched to different specifications.