Browse Prior Art Database

VIRTUAL BANDWIDTH DIGITAL RECEIVER CONVERSION SIGNAL PROCESSOR

IP.com Disclosure Number: IPCOM000008427D
Original Publication Date: 1997-Dec-01
Included in the Prior Art Database: 2002-Jun-13
Document File: 4 page(s) / 163K

Publishing Venue

Motorola

Related People

Brian T. Kelley: AUTHOR

Abstract

Abstract-Due to advances in AID technology, direct digital IF conversion receivers are being used in place of more traditional superheterodyne receiver platforms. Direct conversion of analog IF signals to digital requires an oversampled bandpass A/D followed by digital signal processing (DSP) hard- ware. A significant limitation upon the use of over- sampled A/D technology is the limited bandwidth over which the desired S/N ratio is maintained. In this paper we describe a novel method that increases the usable frequency bands (bandwidth) over which an oversampled A/D based receiver achieves its theoretical maximum SIN measure.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 45% of the total text.

Page 1 of 4

m MOTOROLA Technical Developments

VIRTUAL BANDWIDTH DIGITAL RECEIVER CONVERSION SIGNAL PROCESSOR

by Brian T. Kelley

  Abstract-Due to advances in AID technology, direct digital IF conversion receivers are being used in place of more traditional superheterodyne receiver platforms. Direct conversion of analog IF signals to digital requires an oversampled bandpass A/D followed by digital signal processing (DSP) hard- ware. A significant limitation upon the use of over- sampled A/D technology is the limited bandwidth over which the desired S/N ratio is maintained. In this paper we describe a novel method that increases the usable frequency bands (bandwidth) over which an oversampled A/D based receiver achieves its theoretical maximum SIN measure.

  The objective of a digital receiver device is to extract a desired IF received signal from amongst a litany of unwanted out-of-band noise. The combina- tion of bandpass A/D" and digital signal processing hardware to perform signal recovery have some- times been referred to as a conversion signal processor (CSP)'.

  Outside of the narrow signal band, modulation noise increases rapidly in oversampled sigma-delta A/D converters. Thus, the constrained signal band- width over which the signal-to-noise (9%') ratio of

the converter maintains its peak fidelity represents a practical limitation of conventional CSP technology. Methods of increasing the usable sigma-delta band- width include (I) increasing the sampling rate (fs) of the oversampled A/D, (2) increasing the number of quantization levels in the comparator within the sigma delta, and (3) increasing the order of the

sigma delta loop determining the modulation noise transfer function. However, for a fixed topology and sampling rate, signals whose bandwidth exceed the specifications of the A/D cannot be recovered with the same bit-accuracy as their narrow band counter parts, Furthermore, conventional CSPs are inflexi- ble in so far as only being able to "tune" to a single frequency'.

  This method of Virtual Bandwidth Frequency Readjustment (VBFR) solves the limited bandwidth and inflexible tuning problem by allowing the CSP design to readjust or tune itself to various control- lable frequencies. By doing so, this new receiver maintains its high fidelity reception over's wider sequence of "virtual" frequency bands which may or may not be contiguous. The system can therefore be used to (1) increase the S/N ratio by tuning to various extremely high fidelity "sweet spots", (2) by utilizing the A/D over a wider virtual bandwidth, or (3) perform digitally adjustable tuning of the sig- nal directly within the CSP interface.

  Figure I illustrates how two different IFS transferred from the sigma-delta A/D (neglecting modulation noise) can be mixed to baseband while simultaneously reducing the sampling rate. Figure la illustrates two possible IF frequencies either at frequency fl or t2. In Figure lb, one of the two frequencies are decimated by the appropriate...