Fang Zhou

Dr. Fang Zhou

Nanjing University of Aeronautics and Astronautic

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

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

Representative Publications

View All Publications
16 June 2020 in Electronics
Verified
Journal Article
Open access
PDF file
NWNing Wu
Ning Wu
TJTao Jiang
Tao Jiang
LZLei Zhang
Lei Zhang
Fang ZhouFang Zhou
Fang Zhou
FGFen Ge
Fen Ge
As a typical artificial intelligence algorithm, the convolutional neural network (CNN) is widely used in the Internet of Things (IoT) system. In order to improve the computing ability of an IoT CPU, this paper designs a reconfigurable CNN-accelerated coprocessor based on the RISC-V instruction set. The interconnection structure of the acceleration chain designed by the predecessors is optimized, and the accelerator is connected to the RISC-V CPU core in the form of a coprocessor. The corresponding instruction of the coprocessor is designed and the instruction compiling environment is established. Through the inline assembly in the C language, the coprocessor instructions are called, coprocessor acceleration library functions are established, and common algorithms in the IoT system are implemented on the coprocessor. Finally, resource consumption evaluation and performance analysis of the coprocessor are completed on a Xilinx FPGA. The evaluation results show that the reconfigurable CNN-accelerated coprocessor only consumes 8534 LUTS, accounting for 47.6% of the total SoC system. The number of instruction cycles required to implement functions such as convolution and pooling based on the designed coprocessor instructions is better than using the standard instruction set, and the acceleration ratio of convolution is 6.27 times that of the standard instruction set.

5 May 2019 in Electronics
Verified
Journal Article
Open access
PDF file
FGFen Ge
Fen Ge
NWNing Wu
Ning Wu
HXHao Xiao
Hao Xiao
YZYuanyuan Zhang
Yuanyuan Zhang
Fang ZhouFang Zhou
Fang Zhou
As a classical artificial intelligence algorithm, the convolutional neural network (CNN) algorithm plays an important role in image recognition and classification and is gradually being applied in the Internet of Things (IoT) system. A compact CNN accelerator for the IoT endpoint System-on-Chip (SoC) is proposed in this paper to meet the needs of CNN computations. Based on analysis of the CNN structure, basic functional modules of CNN such as convolution circuit and pooling circuit with a low data bandwidth and a smaller area are designed, and an accelerator is constructed in the form of four acceleration chains. After the acceleration unit design is completed, the Cortex-M3 is used to construct a verification SoC and the designed verification platform is implemented on the FPGA to evaluate the resource consumption and performance analysis of the CNN accelerator. The CNN accelerator achieved a throughput of 6.54 GOPS (giga operations per second) by consuming 4901 LUTs without using any hardware multipliers. The comparison shows that the compact accelerator proposed in this paper makes the CNN computational power of the SoC based on the Cortex-M3 kernel two times higher than the quad-core Cortex-A7 SoC and 67% of the computational power of eight-core Cortex-A53 SoC.

26 May 2021 in Micromachines
Verified
Journal Article
Open access
PDF file
FGFen Ge
Fen Ge
CCChenchen Cui
Chenchen Cui
Fang ZhouFang Zhou
Fang Zhou
NWNing Wu
Ning Wu
More and more attention is being paid to the use of massive parallel computing performed on many-core Networks-on-Chip (NoC) in order to accelerate performance. Simultaneously deploying multiple applications on NoC is one feasible way to achieve this. In this paper, we propose a multi-phase-based multi-application mapping approach for NoC design. Our approach began with a rectangle analysis, which offered several potential regions for application. Then we mapped all tasks of the application into these potential regions using a genetic algorithm, and identified the one which exhibited the strongest performance. When the packeted regions for each application were identified, a B*Tree-based simulated annealing algorithm was used to generate the optimal placement for the multi-application mapping regions. The experiment results show that the proposed approach can achieve a considerable reduction in network power consumption (up to 23.45%) and latency (up to 24.42%) for a given set of applications.

13 July 2019 in Journal of Electrical Engineering & Technology
Verified
Journal Article
Closed access
PDF file
JZJinbao Zhang
Jinbao Zhang
NWNing Wu
Ning Wu
Fang ZhouFang Zhou
Fang Zhou
FGFen Ge
Fen Ge
XZXiaoqiang Zhang
Xiaoqiang Zhang

6 July 2020 in Electronics
Verified
Journal Article
Open access
PDF file
Fang ZhouFang Zhou
Fang Zhou
BZbenjun zhang
benjun zhang
NWNing Wu
Ning Wu
XBXiangli Bu
Xiangli Bu
In order to achieve the purpose of defending against side channel attacks, a compact SM4 circuit was designed based on the mask and random delay technique, and the linear transformation module was designed with random insertion of the pseudo operation method. By analyzing the glitch data generated by the S-box of SM4 with different inputs, the security against glitch attacks was confirmed. Then, the DPA (Differential Power Analysis) was performed on the designed circuit. The key could not be successfully obtained even in the case of 100,000 power curves, so that the safety of SM4 against DPA is verified. Finally, using Synopsys DC (Design Compiler, Mountain View, CA94043DC, USA) to synthesize the designed circuit, the results show that the area of the designed circuit in the SMIC 0.18 process is 82,734 μm2, which is 48% smaller than results reported in other papers.

15 January 2019 in Electronics
Verified
Journal Article
Open access
PDF file
JZJinbao Zhang
Jinbao Zhang
NWNing Wu
Ning Wu
Fang ZhouFang Zhou
Fang Zhou
MYMuhammad Rehan Yahya
Muhammad Rehan Yahya
JLJianhua Li
Jianhua Li
As a family of lightweight block ciphers, SIMON has attracted lots of research attention since its publication in 2013. Recent works show that SIMON is vulnerable to differential fault analysis (DFA) and existing DFAs on SIMON assume the location of induced faults are on the cipher states. In this paper, a novel DFA on SIMON is proposed where the key schedule is selected as the location of induced faults. Firstly, we assume a random one-bit fault is induced in the fourth round key KT−4 to the last. Then, by utilizing the key schedule propagation properties of SIMON, we determine the exact position of induced fault and demonstrate that the proposed DFA can retrieve 4 bits of the last round key KT−1 on average using one-bit fault. Till now this is the largest number of bits that can be cracked as compared to DFAs based on random bit fault model. Furthermore, by reusing the induced fault, we prove that 2 bits of the penultimate round key KT−2 could be retrieved. To the best of our knowledge, the proposed attack is the first one which extracts a key from SIMON based upon DFA on the key schedule. Finally, correctness and validity of our proposed attack is verified through detailed simulation and analysis.