Jihe Wang

Jihe Wang

Auto-generated profile, based on publicly available information

Jihe Wang

School of Computer Science, Northwestern Polytechnical University

Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2017 in Journal of Systems Architecture
Verified
Journal Article
Closed access
JWJihe Wang
Jihe Wang
MQMeikang QIU
Meikang QIU
BGBing Guo
Bing Guo

1 August 2015 in Future Generation Computer Systems
Verified
Journal Article
Closed access
JWJihe Wang
Jihe Wang
MQMeikang Qiu
Meikang Qiu
BGBing Guo
Bing Guo

25 May 2022 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
JWJihe Wang
Jihe Wang
HCHao Chen
Hao Chen
DWDanghui Wang
Danghui Wang
KMKuizhi Mei
Kuizhi Mei
SZShengbing Zhang
Shengbing Zhang
XFXiaoya Fan
Xiaoya Fan
Stochastic computing (SC) has become a promising approximate computing solution by its negligible resource occupancy and ultralow energy consumption. As a potential replacement of accurate multiplication, SC can dramatically mitigate the problematic power consumption by DNNs. However, current SC-multipliers illustrate an extremely imbalanced accuracy across product space, i.e., neglectable noise with large products but significant noise for small ones, which is discordant to the distribution of products by the sparse matrix in neural computing. In this article, we present a heterogeneous SC-multiplier that heuristically performs three divergent approximating multiplication, including “set-to-0,” “look-up-table,” and “low-discrepancy-SC,” for appropriate precision-provision in the whole space of products. Due to those popular DNN models cannot achieve consensus on the boundaries of above operations, a training-involved method is proposed to determine the settings with limited overhead. In this way, those models successively learn the SC-operation characters and exhibit a definitely improvement on network precision. The experiment shows that, for single multiplication, the product noise can be restrained by 36.86% on average, and for multiplication in multiple network models, the accuracy improvement reaches to 5.5% on average. Furthermore, a group of proposed logic-reduction techniques can improve the energy efficiency by 65% in the system-level evaluation.

19 June 2009 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Closed access
JZJinxiu Zhang
Jinxiu Zhang
XCXibin Cao
Xibin Cao
JWJihe Wang
Jihe Wang
XLXiaohui Lin
Xiaohui Lin
A mission requirements-based configuration method of interferometric synthetic-aperture radar (InSAR) formation to design dual-satellite formation by mean orbital elements is derived. The configuration and initial orbital elements of two coordinated running satellites InSAR formation are designed when system parameters, such as the resolution of slant distance, latitude domain of ground coverage area, and the side-looking angle of the synthetic-aperture radar (SAR) beam center, are given.

1 December 2010
Verified
Conference Paper
Closed access
QLQi Li
Qi Li
BGBing Guo
Bing Guo
YSYan Shen
Yan Shen
JWJiHe Wang
JiHe Wang
YWYuanSheng Wu
YuanSheng Wu
YLYunben Liu
Yunben Liu
Nowadays as low carbon economy is greatly advocated worldwide, the electricity consumption caused by a huge number of embedded computer systems is gaining more and more attention. Different instruction set, software algorithm and high-level software architecture can significantly affect the system energy consumption. In this paper, we first analyze the relations between software power consumption and some software characteristics on algorithm level. Through measuring three algorithm complexity characteristics, i.e., time complexity, space complexity and input scale, we propose an embedded software power model based on algorithm complexity. Then, we design and train a back propagation neural network to fit the power model accurately based on a sample training function set and more than 400 software power data. Simulation results show that the error between the estimation values of this model and the real measured values is below 10 percent, and this model can effectively estimate the power consumption of software in an early stage of software design.

16 April 2020 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
CMChenlin Ma
Chenlin Ma
ZSZhaoyan Shen
Zhaoyan Shen
JWJihe Wang
Jihe Wang
YWYi Wang
Yi Wang
RCRenhai Chen
Renhai Chen
ZSZili Shao
Zili Shao
Shingled Magnetic Recording (SMR) Disks are adopted as a high-density, non-volatile media that significantly precedes conventional disks in both the storage capacity and cost. However, inefficient read-modify-writes (RMWs) greatly challenge the management of SMR disks. This paper for the first time presents an approach called Tiler to manage SMR disks by dividing the physical space into small autonomous regions (ARs). Each AR can manage its space allocation, address mapping and cleaning independently. By managing these ARs in a log-structured way, RMWs can be avoided; besides, ARs can also help update data when the adjacent tracks contain no valid data. Tiler is capable of partitioning a large-scale cleaning into self-contained-small-scale cleaning and thus, the data that need to be relocated are limited inside independent ARs, which further minimizes the performance overhead. Our experimental results show that Tiler can shorten the overall system response time by 50.21% and reduce the cleaning time by 90.24% on average.