Modern information and communication technologies (ICTs) have entered a mobile and intelligent era. Mixed-signal and radio frequency (MS/RF) integrated circuits (ICs) are critical for emerging ICTs such as wireless local area network (WLAN), passive optical network (PON), satellite communication, wireless access techniques, which are rapidly penetrating all aspects of our lives. Especially in the new 5G system, the same frequency interfering is the major challenge for the same frequency & same time RF transceivers. RF ICs are also widely used in the biotech applications such as implantable devices, wearable medical devices, which have an ultra-low power requirement and cause great challenges. Moreover, circuit is a key component of sensors, which have been widely used in the internet of thing (IoT). The sensor detector chips usually have to process many weak signals at the same time, which requires low power along with a high resolution. The main contributions are included as follows:
1) High performance RF circuits.
Research group in the Institute of Microelectronics have dedicated to the research of high performance RF device models and transceivers. They proposed Physics based scalable models for inductive devices. High sensitivity UHF RFID reader transceiver, ultra-low power transceiver for wearable medical applications and reconfigurable sub 1.0GHz receiver for internet of things (IoT) were also proposed. To improve RFIC/MMIC design efficiency, physics based scalable models for integrated inductor and transformer were developed, which have only one fitting parameter. A new equation of high-frequency resistance by considering skin effect and proximity effect simultaneously, and for substrate capacitive loss, a quasi-linear relationship between capacitive coupling effect and proximity effect were first established. In high sensitivity UHF RFID reader transceiver, a fully on-chip directional coupler was proposed for the first time, which is based on electric and magnetic signals cancellation, and achieves 67dB isolation with chip area of 220mm′220mm. A thermal noise cancellation circuit with simplified phase cancellation also proposed to improve the reader sensitivity. The proposed RFID reader achieves -92dBm sensitivity, around 20dB better than conventional ones. A sub-1.0 GHz reconfigurable low power RF receiver for IoT applications was verified in a 65nm CMOS process. A generic-purpose active-RC filter with highly power-efficient operational amplifier was proposed, and has a best FoM of 0.04fJ. An off-chip pre-filter and LNA merged RF frontend was also proposed, which achieves 2~3dB noise figure with only 3.5mW power consumption. For the wearable medical device applications, an ultra-low power transceiver using human body as antenna was first proposed. Experiments show that human body is an efficient antenna working at 50MHz ~150MHz frequency range. With the help of human body antenna, the wearable medical devices can be realized with smaller forming factor and the power consumption as low as sub 1.0mW. Related works have been published in IEEE JSSC, IEEE TCAS-I, IEEE T-MTT, IEEE EDL, IEEE T-ED and other prestigious journals and conferences in the RFIC research area.
2) Mixed-signal Readout Circuits.
A front-end ASIC chip for silicon PiN radiation detector was fulfilled, which is used in high-energy physics, space physics, radiation protection and medical applications. Several circuit structures have been proposed to improve signal-to-noise ratio, stability, linearity, and insensitivity to PVT variances. Several circuit structures have been proposed to improve signal-to-noise ratio, stability, linearity, and insensitivity to PVT variances. For different applications, several chips are fabricated and on trial. Several ASICs have been developed for detectors with different sizes (288×4 with TDI, 384×288, 640×512, 1024×1024) and sensor types (HgCdTe, InSb, QWIP, micro-bolometer) and used in imaging system. The front-end ASIC chip for silicon PiN radiation detector won the first prize of Science and Technology Progress Award, the Ministry of Education in 2016. In particular, 30,000 chips of single-photon counting type have been provided to users to be applied to personal dosimeter. It is the first domestic chip obtaining high volume applications in radiation detection system.
