Skip to main content

2006 | Buch

ΣΔ A/D CONVERSION FOR SIGNAL CONDITIONING

verfasst von: Kathleen Philips, Arthur H. M. van Roermund

Verlag: Springer Netherlands

Buchreihe : The International Series in Engineering and Computer Science

insite
SUCHEN

Über dieses Buch

1.1 Background Moore’s Law predicts a decrease by a factor of two in the feature size of CMOS te- nology every three years and has been valid for years. It implies a doubling of the - eration speed and a four times higher transistor count per unit of area, every three years. The combination leads to an eight times higher processing capability per unit of area. This on-going miniaturization allows the integration of complex electronic systems with millions of transistors (Very-Large-Scale-Integration) and enables the integration of el- tronic systems. An electronic system A generic picture of an integrated electronic system is shown in ?g. 1.1. The heart of the system is the signal processing core. This core supports a wide variety of functions, such as customization and programmability of multiple applications, channel coding, the de?nition of the user interface, etc. These functions are enabled by DSP, a controller CPU and various blocks of memory. In advanced ICs these blocks provide (almost) all signal processing and usually dominate in the overall power and area consumption of integrated systems. The huge data rates involved, require high-speed busses for communication between these blocks. A power-management unit fuels the system by providing the - propriate supply voltages and currents.

Inhaltsverzeichnis

Frontmatter
Introduction
Abstract
Moore’s Law predicts a decrease by a factor of two in the feature size of CMOS technology every three years and has been valid for years. It implies a doubling of the operation speed and a four times higher transistor count per unit of area, every three years. The combination leads to an eight times higher processing capability per unit of area. This on-going miniaturization allows the integration of complex electronic systems with millions of transistors (Very-Large-Scale-Integration) and enables the integration of electronic systems.
Kathleen Philips, Arthur H. M. van Roermund
The signal conditioning channel
Abstract
A generic communication channel is represented in fig. 2.1. A message encoded in a signal is transported through a channel to a destination. Various noise sources add to the wanted signal but should not corrupt the message. The capacity C of a communication channel has been calculated by Shannon as a function of the bandwidth (BW) and the signal/noise ratio (SNR) of the channel. In case of a Gaussian signal source and Gaussian noise sources the following well-known formula results:
Kathleen Philips, Arthur H. M. van Roermund
ΣΔ A/D conversion
Abstract
A first patent related to delta modulation was filed in 1948 [21]. In 1952, the delta modulator was first published by de Jager and Greefkes at the Philips Research Laboratories (fig. 3.1, [22]). The invention was inspired by the operation of the human brain: physiological signals are translated into a series of electrical pulses in the nerve system as a means for data transmission to the brain. In a similar way, pulse density modulation was used for robust data transmission in telephony. In the single-bit, delta-modulated code all bits are of equal weight and any bit-flip causes only a small error. In a multi-level PCM code though a bit-flip of the Most-Significant-Bit results in a major error. The paper by de Jager in 1952 [22], on delta modulators was the first in a massive series of delta-sigma and sigma-delta papers. In 1960, the delta-sigma modulator was patented by Cutler [23]. Inose et al. [24] proposed to shift the loop filter in the forward path of the modulator in 1962. In the 1980’s sigma-delta conversion became popular in both the A/D and the D/A part of audio channels. In addition, instrumentation applications widely adopted ΣΔ converters. By that time, a lot of theoretical work had been published, a.o by J. Candy.
Kathleen Philips, Arthur H. M. van Roermund
Power consumption in channel building blocks
Abstract
In literature, many articles on power consumption versus performance are found. Calculations on fundamental limits for the power consumption of a single transistor or a basic cell can be found in a.o. the work of Vittoz [55] (calculating the power consumption per pole as a function of dynamic range for analog) and the work of Meindl [56] (giving the power consumption for a single binary transition).
Kathleen Philips, Arthur H. M. van Roermund
Full-analog and full-digital conditioning channels
Abstract
The operation of the channels (in this and the next chapter) is clarified with the example input signal of fig. 5.1.a. It consists of a small wanted channel, a slightly stronger adjacent channel and two much stronger far-off interferer channels that could cause intermodulation components in the wanted channel. At the output of each conditioning channel, only the digitized wanted channel is available with a specified SINAD and sample rate (fig. 5.1.b). As a starting point, a channel with full-analog conditioning is discussed. Further on it serves as a benchmark for the power/performance ratio of digitized channels.
Kathleen Philips, Arthur H. M. van Roermund
Conditioning ΣΔ ADCs
Abstract
In chapter 5 channels with full-analog and full-digital signal conditioning have been presented (fig. 6.1.a and b respectively). Here, the signal conditioning is integrated into the ΣΔ ADC resulting in a “conditioning ΣΔ ADC” without preceding analog filters or variable gain stages (fig. 6.1.c). First, the basic operation of the conditioning ΣΔ ADC is clarified. Next, a universal model of a ΣΔ ADC is introduced. This model is used for a generic analysis of the interferer immunity of continuous-time ΣΔ ADCs. At the end of this section, a basic power/performance analysis is presented.
Kathleen Philips, Arthur H. M. van Roermund
Digitization of the inter-die interface
Abstract
In case of a multi-die solution, the choice for an analog or a digital interface can be determined by a number of factors, a.o.:
Kathleen Philips, Arthur H. M. van Roermund
Highly analog and highly digital channels for FM/AM radio
Abstract
Although FM and AM radios have been around for decades, there is a clear need for further integration and digitization of the system. This is especially true for a car radio system. Both cost- and performance-wise this is a very demanding application. Digitization of signal processing and conditioning brings the following advantages a.o. [90]:
Kathleen Philips, Arthur H. M. van Roermund
Conditioning ΣΔ ADCs for Bluetooth
Abstract
Bluetooth [112] is a short-range wireless communication system targeting operation in the 2.4GHz ISM band. In most countries, the channels occupy the band from 2.402 to 2.480GHz. The channel spacing is 1MHz and GFSK modulation is used. The latter implies that all information is contained in the phase of the signal. The Bluetooth standard targets a moderate data-rate (i.e. 1Mbps for the early standard) for use in portable applications as for example headsets, printers, etc. By consequence, cost and power drain are the main challenges on the IC solution.
Kathleen Philips, Arthur H. M. van Roermund
General conclusions
Abstract
Signal conditioning channels are being digitized in a quest for flexibility and to benefit from technology scaling. While in the signal processing arena digitization has lead to a power saving per computation, this benefit is not self-evident for the conditioning channel. In this context, this book aims at improving the power/performance relation of the conditioning channel by balancing analog and digital signal conditioning.
Kathleen Philips, Arthur H. M. van Roermund
Backmatter
Metadaten
Titel
ΣΔ A/D CONVERSION FOR SIGNAL CONDITIONING
verfasst von
Kathleen Philips
Arthur H. M. van Roermund
Copyright-Jahr
2006
Verlag
Springer Netherlands
Electronic ISBN
978-1-4020-4680-3
Print ISBN
978-1-4020-4679-7
DOI
https://doi.org/10.1007/1-4020-4680-4

Neuer Inhalt