Cheng Huang

Dr. Cheng Huang


Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

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

Representative Publications

View All Publications
7 September 2020 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
CHCheng Huang
Cheng Huang
YZYuan Zhuang
Yuan Zhuang
HLHao Liu
Hao Liu
JLJianyu Li
Jianyu Li
WWWei Wang
Wei Wang
Bluetooth Low Energy (BLE) has been significantly contributed to Internet of Things applications due to its advantages, such as low power consumption and scalability. To further expand its applications, BLE has adopted the angle of arrival (AoA) and the angle of departure (AoD) to improve the accuracy in direction finding (DF). The AoA/AoD is calculated using the phase difference (PD) between slots in the received BLE data. The PD performance of BLE receiver (RX) in all directions should be evaluated in detail because it directly determines the accuracy of AoA/AoD. In the traditional test method, the reference PD is converted by an antenna array with switches, which is inaccurate due to the conversion errors and path mismatches. Moreover, because of the mechanical rotation of the turntable, the test processing is very time-consuming. This paper propose a BLE test waveform generation method that directly includes the reference PD. A commonly-used vector signal generator (VSG) is the only required instrument in the evaluation. The compact test setup not only eliminates the requirement of antenna array, switch, turntable, and anechoic chamber, but also eliminates the measurement errors caused by these devices. After creating and downloading all BLE test waveforms with reference PD in the range of -180∘ to 180∘ by user-defined steps, the PD performance of BLE RX can be quickly evaluated by switching the test waveforms using commands or running them in the sequence mode. Furthermore, three enhanced test cases with automatic test procedures are proposed to evaluate the detailed PD accuracy of BLE RX under the low signal to noise ratio (SNR), carrier frequency offset and modulation frequency deviation environments. Experimental results show that the PD error of proposed method is 0.19∘, and it needs 360ms to evaluate the PD accuracy of CC2640R2F in all directions in 1∘ steps.

13 September 2022 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
WSWeiwei Shan
Weiwei Shan
JQJunyi Qian
Junyi Qian
LZLixuan Zhu
Lixuan Zhu
LHLuchi Hua
Luchi Hua
CHCheng Huang
Cheng Huang
HCHao Cai
Hao Cai
As a widely used speech-triggered interface, deep-learning-based keyword spotting (KWS) chips require both ultra-low power and high detection accuracy. We propose a sub-microwatt KWS chip with an acoustic activity detection (AAD) to achieve the above two requirements, including the following techniques: first, an optimized feature extractor circuit using nonoverlapping-framed serial Mel frequency cepstral coefficient (MFCC) to save half of the computations and data storage; second, a zero-cost AAD by using MFCC’s 1st-order output to clock gate neural network (NN) and postprocessing (PP) unit, with 0 miss rate; third, a tunable detection window to adapt to different keyword lengths for better accuracy; and finally, a true form computation method to decrease data transitions and optimized PP. Implemented in a 28-nm CMOS process, this AAD-KWS chip has a 0.4-V supply, an 8-kHz frequency for MFCC, and a 200-kHz frequency for other parts. It consumes $0.36~\mu \text{W}$ in quiet scenarios when AAD is enabled and $0.8~\mu \text{W}$ in normal scenarios, where the MFCC circuit consumes only 170 nW. Its accuracy reaches 97.8% for two keywords in the Google Speech Command Dataset (GSCD).

9 December 2019 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
CHCheng Huang
Cheng Huang
HLHao Liu
Hao Liu
WWWei Wang
Wei Wang
JLJianyu Li
Jianyu Li
Bluetooth low-energy (BLE) beacons have been used in a variety of IoT applications. This article proposes a compact and cost-effective BLE beacon, which only uses a 2.4-GHz Gaussian frequency-shift keying (GFSK) transceiver and an ultralow power 8-bit 16-MHz microcontroller unit (MCU). The minimum requirements and the data processing procedure for commonly used GFSK transceivers to generate the BLE-compatible advertising packages are also provided in this article. Benefited from the compact architecture and the simple generation procedure, the BLE-compatible packages can be directly controlled to broadcast at any given radio frequency (RF) channel or time. Based on the flexible characteristics of our proposed beacon, the multiprotocol and dynamic content advertising are developed in this article. With the help of multiprotocol and dynamic content advertising, the proposed beacons can be used to provide more sophisticated services. Since the broadcast channels and time are freely controlled, the proposed BLE beacons are also very suitable for anti-collision applications with custom-defined channels and time intervals. The experimental results show that the proposed beacon can broadcast BLE-compatible protocol data units (PDUs) at a minimum interval of about 1 ms. The measured average current is about 51 uA when the advertising interval is 800 ms. Different protocols and dynamic content PDUs generated by the proposed beacon can be properly received by BLE-enabled devices, proving that they are fully compatible with the BLE specification. The measured bit rate of the proposed beacon in dynamic content advertising is about 198.4 kb/s.

3 May 2015 in Analog Integrated Circuits and Signal Processing
Verified
Journal Article
Closed access
PDF file
CHCheng Huang
Cheng Huang
CCC. Chen
C. Chen
MZM. J. Zhou
M. J. Zhou
JWJ. H. Wu
J. H. Wu

4 July 2018 in Analog Integrated Circuits and Signal Processing
Verified
Journal Article
Closed access
PDF file
CHCheng Huang
Cheng Huang
JWJianhui Wu
Jianhui Wu
CBChunfeng Bai
Chunfeng Bai

17 March 2017 in Journal of Circuits, Systems and Computers
Verified
Journal Article
Closed access
CHCheng Huang
Cheng Huang
ZLZhilun Lin
Zhilun Lin
JWJianhui Wu
Jianhui Wu
CCChao Chen
Chao Chen