Journals
  Publication Years
  Keywords
Search within results Open Search
Please wait a minute...
For Selected: Toggle Thumbnails
Estimating Nitrogen and Phosphorus Pollution Load in Bali Lake Basin of Jiangxi Province Based on SWAT Model
CHEN Yan, ZHAO Yanxin, ZHAO Yue, WANG Dong, BAI Hui, GUO Huaicheng
Acta Scientiarum Naturalium Universitatis Pekinensis    2019, 55 (6): 1112-1118.   DOI: 10.13209/j.0479-8023.2019.086
Abstract1096)   HTML    PDF(pc) (1883KB)(286)       Save
The simulation of nitrogen (N) and phosphorus (P) pollution load structure in Bali Lake basin was performed by SWAT model using the spatial and attribute database acquired through social survey and statistical data. The model was applied and validated based on the data of meteorology, runoff and water quality from 2010 to 2014. The spatial-temporal distribution of N and P in the basin was also studied. The results showed that the urban life was the most important source of N and P load, following by agriculture and livestock breeding pollution source, and the least was industry pollution source. The annual TN and TP loads were mainly concentrated in flood season (from May to September) which took up nearly 55% of the whole year load. As the amount of precipitation increased, TN and TP loads increased during flood period, which was mainly due to the increasing of non-point rural living water and agricultural fertilization pollution source with scouring action of the heavy rainfall during flood season. The spatial distribution of N and P was centered on the Jiujiang county, Xunyang and Lushan district as the urban agglomeration area affected by urban sewage pollution.
Related Articles | Metrics | Comments0
Study on Continuous Speech Recognition Based on Bottleneck Features for Lhasa-Tibetan Dialect
ZHOU Nan, ZHAO Yue, LI Yaoqiang, XU Xiaona, CAIWANG Lamu, WU Licheng
Acta Scientiarum Naturalium Universitatis Pekinensis    2018, 54 (2): 249-254.   DOI: 10.13209/j.0479-8023.2017.154
Abstract906)   HTML1)    PDF(pc) (695KB)(209)       Save

The bottleneck features extracted from deep neural network not only have long term contextdependence and compact representation of speech signal, but also can replace the traditional MFCC features for GMM-HMM acoustic modeling. The authors apply bottleneck features and their concatenated features with MFCC into Lhasa-Tibetan continuous speech recognition. The experiments in Lhasa-Tibetan continuous speech recognition show that the concatenated features of bottleneck features and MFCC achieve better performance than the posterior features of deep neural network and mono-bottleneck features.

Related Articles | Metrics | Comments0
Design and Implementation of 16-bit RISC MCU for Medical Electronics Applications
WANG Teng,XIE Zheng,ZHAO Yueming,WANG Xin’an,HU Ziyi,ZHANG Xu
Acta Scientiarum Naturalium Universitatis Pekinensis   
Abstract621)      PDF(pc) (3633KB)(359)       Save
By analyzing the application demand of the medical electronics, an MCU architecture named PKU-DSPII with application specific instruction set is proposed. Multilevel storage structures with direct memory access controller (DMAC) and abundant system control modules and peripheral interfaces are integrated in the MCU, which supports three booting modes including self-updating mode. PKU-DSPII is implemented with CSMC 0.18μm technology and LQFP package while the die area is 3.2 mm×3.2 mm and the measured power consumption is 96.9 mW at working frequency of 100 MHz.
Related Articles | Metrics | Comments0
An Effective Parallel Processing Architecture for Deblocking Filter in H.264
ZHAO Yuexi,JIANG Anping
Acta Scientiarum Naturalium Universitatis Pekinensis   
Abstract635)            Save
An efficient parallel processing method for deblocking filter design in H.264 video coding standard is presented. In order to reduce the memory reference and make the intermediate data reused as soon as possible, an advanced filtering order is taken, and so read/write operation on external memory is executed in parallel with filtering computation. Furthermore, preloading operation is used to reduce complexity of memory structure. As a result, the processing cycles of the proposed architecture with single-port memory architecture is reduced by 9.6%-74.4% compared with the advanced architecture of previous proposals.
Related Articles | Metrics | Comments0