Home About Faculty and Research Industrial Participants Members Only Resources
CRITICAL MIXED-SIGNAL CIRCUIT BLOCKS FOR HIGHLY INTEGRATED
WIRELESS TRANSCEIVERS

PROJECT OBJECTIVE:
The proposed research consists of three parts: 1) Wideband VCO Modulation in Fractional-N PLLs, 2) Combined DAC Noise Shaping and Cancellation for High-Bandwidth DS ADCs, and 3) Direct Conversion Receiver and VCO Improvements for 3G Mobile Systems. The three parts are closely related in that they each address critical mixed-signal circuit implementation problems in modern wireless communication systems.

GENERAL DESCRIPTION:
Wireless communication systems invariably process analog signals because they interface to the physical world which is inherently analog. For example, a cellular telephone transmits and receives signals through its antenna in the form of analog waveforms regardless of whether it is an "analog" or "digital" telephone. However, in many situations digital circuits, wherein information is represented as binary digits (bits) instead of current or voltage waveforms, offer numerous advantages over analog circuits such as reduced cost, greater programmability, virtually no variability over temperature changes and time, and amenability to security encryption and automatic error correction. Therefore, the bulk of the signal processing and control operations in most modern wireless communication systems is done by digital circuitry. Often, the only significant analog circuitry is that associated with the generation and reception of radio-frequency (RF) signals, and the conversion of analog signals to and from digital form. Nevertheless, while these analog functions represent only a small portion of the total signal processing performed by a given communication system, they often represent much of the overall cost of the system. The primary reason for this is that conventional techniques for implementing the analog functions tend to be incompatible with the limitations imposed by the integrated circuit technology used for digital circuitry, so they are usually implemented with more expensive, but "analog-friendly" integrated circuit technologies and off-chip components.

The proposed research involves the invention, development, analysis, and integrated circuit implementation of critical mixed-signal (combined analog and digital) wireless communication system blocks. The emphasis of the research is on the development of digital signal processing techniques to mitigate the effects of non-ideal analog circuit behavior in mixed-signal integrated circuits. The resulting circuits will tend to blur the traditionally sharp analog-digital dividing lines in communication systems in order to reduce the precision requirements of the analog circuitry and thereby reduce the overall system cost.

TECHNICAL ABSTRACT:
The proposed research consists of three parts as outlined below. The three parts will be conducted by different combinations of principal investigators, but are closely related in that they each address critical mixed-signal circuit implementation problems in modern wireless communication systems.

Part 1 of the proposed research will develop an enhancement to the conventional delta-sigma (DS) fractional-N PLL architecture that largely circumvents the tradeoff between phase noise and bandwidth yet is not sensitive to component mismatches. The enhancement allows the PLL bandwidth to be widened without incurring the usual phase noise penalty resulting from insufficient filtering of the DS quantization noise. Unlike other solutions to this problem, the proposed technique is not sensitive to component mismatches, and does not require an extended-range phase detector or other non-standard PLL components. As a proof-of-concept, a Bluetooth wireless LAN compliant CMOS transmitter will be implemented based on the technique.

Part 2 of the proposed research will modify a DAC noise cancellation technique originally developed by one of the PIs for pipelined ADCs to use with mismatch-shaping DACs in DS ADCs. By combining DAC noise cancellation and mismatch-shaping DACs, it is anticipated that DAC component mismatches can be removed as the limiting factor in high dynamic range DS ADCs with oversampling ratios below 10. As a proof of concept, a CMOS delta-sigma ADC with an oversampling ratio of 8, and a target dynamic range of 90dB over a signal bandwidth of 4MHz will be developed.

Part 3 of the proposed research consists of two tasks. The first task is to develop an improved VCO, suitable for wide dynamic range applications. This circuit will utilize innovative techniques to minimize the noise contributions from the active components and maximize the signal swing at the output of the circuit, resulting in a significant improvement in phase noise performance. The second task is to develop a fully balanced sub-harmonic downconversion mixer topology with a substantial improvement in second-order distortion. This will mitigate the well-known problems of time varying dc offsets in direct conversion receivers. A background calibration scheme will be developed that senses second order distortion components and modifies the mixer bias to adaptively null them out, even with large and rapid variations in the received signal strength. A proof-of-concept demonstration incorporating both circuit approaches will be developed in an advanced BiCMOS technology.

PARTICIPATING FACULTY:
The following CWC faculty are participating in this research project: Ian Galton(lead PI), Larry Larson.
 
 
 
 
 
 
Home | Search | Site Map | Contact Us

Send questions, comments, and suggestions about this website to: info@cwc.ucsd.edu

Copyright © 1995 - 2006 Center for Wireless Communications. All Rights Reserved.