Process and temperature insensitive CMOS oscillator using current sinker compensation

Abstract

frequencies which are 200, 500, 800, and 1000 MHz. The patterns are called divisive as they are consists of main lobes as shown in Figure 7. The energy distributed is not equal which is right lobe is smaller in magnitude compared to the left lobes because some energy released has been absorbed by the resistor. CONCLUSIONBased on Refs. 3, 5, 8, and 9, the physical dimensions of the planar monopole antenna presented are considered small in size and broadband. This antenna is fed from the bottom of the patch. After a lot of parameter optimization routine, the compact size with optimum performance is obtained. By integrating 50 dual bevel angle with a single 100 X resistor, the more current flow to the ground and the radiating element has become more magnetic. Thus, the proposed antenna achieved a very wideband impedance bandwidth of approximately more than 200%. A prototype antenna was designed, fabricate and the performance measure is presented. The antenna has the ability to operate for the VHF and partial of UHF spectrum. Besides the targeted white spaces (broadcasting spectrum) is also catered by this antenna. This gives it more credit as a great choice for the CR applications in the future.ABSTRACT: This article presents a simple complementary metaloxide-semiconductor (CMOS) ring oscillator using a voltage-controlled delay and RS-latch, including a compensation circuit for the process and temperature variations. The compensation circuit, added to the original bias circuit, is a simple current sinker which is referenced through a current mirror circuit. The proposed oscillator was designed and implemented using a 0.13 lm CMOS process. The oscillator exhibited a significantly improved frequency variation of 64.25% for a wide temperature range of from À40 to 80 C. Without the compensation circuit, the variation would have been 613.39% from the center frequency of 2.33 MHz. The oscillator also showed a low sensitivity of 0.084% to process variation, according to a Monte-Carlo simulation with 1000 iterations.