A radio frequency （RF） receiver frontend for single-carrier ultra-wideband （SC-UWB） is presented. The front end employs direct-conversion architecture, and consists of a differential low noise amplifier （LNA）, a...A radio frequency （RF） receiver frontend for single-carrier ultra-wideband （SC-UWB） is presented. The front end employs direct-conversion architecture, and consists of a differential low noise amplifier （LNA）, a quadrature mixer, and two intermediate frequency （IF） amplifiers. The proposed LNA employs source inductively degenerated topology. First, the expression of input impedance matching bandwidth in terms of gate-source ca- pacitance, resonant frequency and target Sll is given. Then, a noise figure optimization strategy under gain and power constraints is proposed, with consideration of the integrated gate inductor, the bond-wire inductance, and its variation. The LNA utilizes two stages with different resonant frequencies to acquire flat gain over the 7.1-8.1 GHz frequency band, and has two gain modes to obtain a higher receiver dynamic range. The mixer uses a double bal- anced Gilbert structure. The front end is fabricated in a TSMC 0.18-μm RF CMOS process and occupies an area of 1.43 mm2. In high and low gain modes, the measured maximum conversion gain are 42 dB and 22 dB, input 1 dB compression points are -40 dBm and -20 dBm, and Sll is better than -18 dB and -14.5 dB. The 3 dB IF bandwidth is more than 500 MHz. The double sideband noise figure is 4.7 dB in high gain mode. The total power consumption is 65 mW from a 1.8 V supply.展开更多
基金Project supported by the National Science and Technology Major Projects of China （Nos. 2011ZX03004-001-02, 2010ZX03007-001-03）.
文摘A radio frequency （RF） receiver frontend for single-carrier ultra-wideband （SC-UWB） is presented. The front end employs direct-conversion architecture, and consists of a differential low noise amplifier （LNA）, a quadrature mixer, and two intermediate frequency （IF） amplifiers. The proposed LNA employs source inductively degenerated topology. First, the expression of input impedance matching bandwidth in terms of gate-source ca- pacitance, resonant frequency and target Sll is given. Then, a noise figure optimization strategy under gain and power constraints is proposed, with consideration of the integrated gate inductor, the bond-wire inductance, and its variation. The LNA utilizes two stages with different resonant frequencies to acquire flat gain over the 7.1-8.1 GHz frequency band, and has two gain modes to obtain a higher receiver dynamic range. The mixer uses a double bal- anced Gilbert structure. The front end is fabricated in a TSMC 0.18-μm RF CMOS process and occupies an area of 1.43 mm2. In high and low gain modes, the measured maximum conversion gain are 42 dB and 22 dB, input 1 dB compression points are -40 dBm and -20 dBm, and Sll is better than -18 dB and -14.5 dB. The 3 dB IF bandwidth is more than 500 MHz. The double sideband noise figure is 4.7 dB in high gain mode. The total power consumption is 65 mW from a 1.8 V supply.
