Search

This paper presents the design of a divide-by-three frequency divider operating at 30 GHz with an injection-switched cross-coupled pair topology. A wider locking range as well as a lower operation voltage can be achieved because of this newly proposed topology. The divider is implemented in a 90-nm standard CMOS process. The total locking range of the divider core is 4.5 GHz with a power consumption of 2.85 mW from a supply voltage of 0.5 V. The total power consumption of the buffers is 2.65 mW from a supply voltage of 1.0 V. The measured output phase noise is −141 dBc/Hz at 1-MHz offset when the input referred signal has a phase noise of −131 dBc/Hz at 1-MHz offset from 30 GHz. The phase-noise difference of 10 dB is close to the theoretical value of 9.5 dB for division-by-three. The total chip size is 0.48 mm2, and the divider core size is only about 0.14 mm2. [ABSTRACT FROM AUTHOR]

In this paper, an equi-section division method utilizing the symmetry property and amplitude approximation of a sinusoidal waveform to design a direct digital frequency synthesizer (DDFS) is proposed. The sinusoidal phase of a onequarter period is divided into equi-sections. The error value between each line segment value and the sinusoidal amplitude value is stored in a read-only memory (ROM) to reconstruct the real sinusoidal waveform. The upper/lower bound of the maximum error value stored in error-compensation ROM will be derived to determine the minimum required memory wordlength relative to the bit number of the equi-sections. In addition, the minimum size of the total ROMs can be computed according to the bit number of the equi-sections. Thus, the equi-section division method is implemented on a field programmable gate array (FPGA) development board. As a result, the total compression ratio of the DDFS using the equisection division method is superior to that of the DDFS using the traditional compression methods. The simulation and experimental results show that the proposed ROM compression method can effectively achieve a better compression ratio and lower complexity, compared with the DDFS using the traditional compression methods, without affecting the spectrum performance. [ABSTRACT FROM AUTHOR]