EEWeb Pulse - Issue 25, 2011

8/3/2019 EEWeb Pulse - Issue 25, 2011

1/24

Electrical Engineering Commun

EEWeb

PULSE

EEWeb.c

Issue December 20, 2

Abolfazl RaziWiseNet Lab

Electrical Engineering Commun

8/3/2019 EEWeb Pulse - Issue 25, 2011

2/24

Contact Us For Advertising Opportunities

[email protected]

www.eeweb.com/advertising

Electrical Engineering CommunityEEWeb

Digi-Key is an authorized distributor for all supplier partners. New products added daily. 2011 Digi-Key Corporation, 701 Brooks Ave. South, Thief River Falls, MN 56701, USA

www.digikey.com/techxchange

Its all about

connections.

The user-to-user forum is for everyone, fromdesign engineers to hobbyists, to discusstechnology, products, designs and more.

Join the discussions that match your interestor offer your expertise to others.

Join the discussion now at:

discussions

hobbyists

engineers

industry experts

resourceslinks

technical documentswhite papers

reference designs

application notes

community

power

microcontroller

lighting

wireless

sensor

students

8/3/2019 EEWeb Pulse - Issue 25, 2011

3/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 3

TABLE OF C ONTENTS

Abolfazl Razi 4WiseNet Lab

Multiple Access for Passive Sensors 8BY ABOLFAZL RAZI

Featured Products 11

State Machine Coding Styles 12BY RAY SALEMI

Filter Selection and Design: The 17Gateway to System PerformanceBY TAMARA SCHMITZ WITH INTERSIL

Return to Zero Comic

Learn about the roles and importance of passive sensors in the electronics industry.

Interview with Abolfazl Razi - PhD Student at the University of Maine

An examination of state machines and the coding styles used to implement them.

Properly select, model and integrate elements of filter circuit design to maximizeperformance.

23

8/3/2019 EEWeb Pulse - Issue 25, 2011

4/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 4

INTERVIEW

WiseNet Lab

How did you get into

electronics/engineeringand when did you start?

I have always been interested inmathematics because to me, itis the only science completelybased on solid proofsone follows from another.Most physical phenomena,hypotheses and theories,

are not understandable and

provable without the magic ofmathematical equations. WhenI was in high school, I got thebook Schaums Mathematical Handbook of Formulas and

Tables as a gift. I studied it andgot familiar with new topics likederivatives and integrals. Thisbook made me so eager to learn

math theory and become moreknowledgeable about thesethings. I also was very interestedin assembling electronic kits likeradio and led blinkers. Whilestudying in high school, I realizedhow fantastically an electricalphenomenon can be describedby mathematical equations.

All of this led me into the fieldof electrical engineering, so Ichose it as my academic major,and as time passes I love it more.

What are your favoritehardware tools that youuse?

I have used different hardwarebased on different projectsrequirements. This spans a

wide category from a tiny smartcard reader to a huge mobileswitching center. Recently I havebeen working with a Xilinx Virtex-4 FPGA board to design a newinterrogator system for passivesensors. It is a very multi-purposed and powerful boardand at the same time is easily

Abolfazl Razi - PhD Student at the University of Maine, ECE Department

AdolfazlRazi

8/3/2019 EEWeb Pulse - Issue 25, 2011

5/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 5

INTERVIEW

programmable by MATLABDSP tools. I am also very luckyto have access to a powerfuland expensive SPIRENT SR5500

wireless channel simulator inour lab that helps me simulatedifferent cellular systems likeGSM, WiMAX and LTE. I canexamine the performance of

various transmission techniquesand coding schemes on wirelessradio channels. Sometimes itcompares the performance ofthe codes developed in MATLABunder idealistic assumptions

against what is really happeningin the wireless channels. Thisinstrument makes a goodconnection between the theorystuff we learned in university andthe practical world of industry.

What are your favoritesoftware tools that you use?

Maybe Id better tell you mysecond favorite software, since I

think the most popular softwareamong all communicationengineers is definitely MATLAB,and I am not an exemption. Iuse MATLAB mostly to simulatecommunication systems. Ialso use different software indifferent projects. I prefer the VBfamily for general programmingbecause of its capabilities andsimplicity. I recommend it for

anyone who wishes to enter theprogramming world. I developeda professional software built tocontrol a companys call centerusing VB6. I also developeda SIMCARD test suite under

VB.NET. I am a C++ fan too,and have developed a charging

entity of a PABX switching center.I have used other softwarelike TEMS Investigation,GemXplore, and Mentum planet

v4.5 when I was working oncellular networks. Recently Ihave used 4nec2 software forarray antenna modeling, which isa very powerful software to easilymodel any wire-based antennas.I like to try different software andcant imagine how hard it wouldbe to research without it.

My first trick isasking somebodywho may know

the answer beforeendlessly fighting a

problem. This way wecan spend our time

more efficiently.

What is the hardest/trickiestbug you have ever fxed?

I believe every problem lookseasy just after it is solved! One

very challenging problem Ifaced goes back to three yearsago when I was working for amobile operator. We realizedthat some customers were beingpushed out of the network andcouldnt make phone calls. Twotechnical groups had workedon the problem without anyprogress before a team of twoother experts and I got into theproject. We started very time-

consuming tests and traced callsand mobile activities in the regionat three different levels. One testinvolved checking the signaling

links between a radio basetransceiver station (BTS) andthe network, which was called

A-Interface using a networkanalyzer. Another test wasmonitoring Air-interface betweenthe BTS and mobile stations (MS)

with TEMS Investigation toolbox,and I was checking the interfacebetween the mobile equipment(ME) and the SIMCARD using

PC/SC debugging tools. At lastour teamwork paid off and wefound the signaling problemthat was due to a wrong errorcode that MSs were receivingfrom an adjacent local mobilenetwork. The code was NoPLMN Access while it had to bePLMN prohibited. This wrongsignaling message caused theFPLMN file in the SIMCARD to

be filled with a wrong value andprevented MSs from makingcalls. I always remember thissuccessful mission that solvedan annoying problem for manycustomers in that area.

What is on your bookshelf?

Most of my books are Ebooks inmy computer. On my bookshelf,I have a few reference books in

my research area. Two of themthat I most frequently need toread, are Tomas Covers bookon Information Theory, andCostellos on Coding Theory. Irecently borrowed a book fromthe library titled Surface AcousticWave Devices for Mobile and

Wireless Communications

8/3/2019 EEWeb Pulse - Issue 25, 2011

6/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 6

INTERVIEW

written by Colin K. Campbell.This book has been very usefulfor me. My experience saysthat reading the right book

that teaches the fundamentalsis always a necessary stepwhen getting into a new area ofresearch.

Do you have any tricks upyour sleeve?

My first trick is asking somebodywho may know the answer beforeendlessly fighting a problem.This way we can spend our time

more efficiently. The second pointI always remind myself is thatthe computers are developed to

work fine and almost error-free.So, I very often try to shift mymath or engineering problemsinto the computer world bymeans of simulation. Sometimes,I compare my results on a pieceof paper with the simulationresults step by step. It is a time-

consuming process and needspatience, but usually it works andhelps me to localize the issues inmy problems.

What has been your favoriteproject?

My recent favorite project isdealing with interference effect inpassive wireless sensor networks(PWSN). Passive sensors are

widely used to sense variousparameters like temperature,humidity, and stress, wherebattery-run active sensors donot operate well due to harshconditions. The passive sensors

we are using are reflective delaylines implemented on Surface

Acoustic Wave (SAW) devices.

Can you tell us more aboutthe Surface Acoustic Wave

(SAW) devices?These devices are composedof an antenna, an interdigitaltransducer (IDT), and a bunchof reflectors, all printed on apiezoelectric substrate like YZ-LiNbO3. To utilize these devices,an electromagnetic wave isradiated by an interrogatorsystem. The electromagnetic

wave is received by the

antenna and is converted toan acoustic wave by the IDT,and is propagated through thedevice surface. The wave ispartially reflected back by thereflectors and is radiated backto the interrogator system bythe same antenna. The reflected

wave carries information aboutthe sensing parameterin ourcase temperatureas well as

the sensors identity defined bythe reflector patterns.

In this project we try to developan efficient algorithm to extractthis information by analyzingthe reflected wave. The goal isto achieve higher accuracy inthe range of operation and toincrease the number of sensors ina single-interrogator system. The

difficulty of employing traditionalsignal processing techniques isthat these devices are passivein nature and all the intelligenceshould be moved to theinterrogator side. Also, the signallevel is very low compared to theactive battery-powered sensors,and that makes the analysis even

more challenging. I think we arestill in the beginning, and there isstill much to be done.

Do you have any note-worthy engineeringexperiences?

I remember a fun trick I learnedfrom one of my professors whenI was an undergraduate. InMicrowave circuit design lab,sometimes the high frequencycircuits like oscillators and RFfilters were not working properlyeven after comprehensive

checks and careful analysis.Our professor taught us a veryeasy trick to use as a last step

when every regular test lookeddisappointing. He said to justput your finger on the back of theelectric board and move it slowlyand check the output signalcarefully. When you obtain thedesired signal, just solder a 20 to100 pico Farad capacitor where

you have put your finger on theboard. It was very surprisingto us to see that this trick was

working fine for some circuits.

What are you currentlyworking on?

Currently I am working on de- veloping interference reductionalgorithms for passive wirelesssensor networks. Our projectsare financially supported byNASA.

8/3/2019 EEWeb Pulse - Issue 25, 2011

7/24

Tri-color High Brightness SMT LEDsfrom Avago Technologies

Avagos PLCC-6 SMT LEDs are high

brightness, high reliability, high

perormance, IPX6 compliant and

are water and dust proo. They are

designed with a separate heat path or

each LED die, enabling it to be driven at

higher current. They deliver super wide

viewing angle at 120 together with

the built in refector pushing up the

intensity o the light output.

Applications

Indoorandoutdoorfullcolordisplay

LEDadvertisementpanels

Decorativelighting

FeaturesWater-resistance(IPX6*)perIEC60529:2001

VerysmallPLCC6packagedimensions

3.4x2.8x1.8mm

In-lineRGBdiesconguration

AvailableinWhiteSurface,Black-Surface

andFullBlack-Body

Wideoperatingtemperaturerange:

-40to+110

Avago TechnologiesTri-color High Brightness

PLCC6 SMT LEDs

gives you a reliable,

long life product for

ease of design in full

color interior and

exterior signs

www.avagoresponsecenter.com/led

AvagoTechnologiesLEDLightingSolutions

One LED. Infnite colors.Worlds frst waterproo package

For more inormation or to

request a sample please go to:

8/3/2019 EEWeb Pulse - Issue 25, 2011

8/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 8

PROJECT

Why are sensors

so important?

The role of sensors in our dailylife is more important than wemay think. We are surrounded bydifferent sensors helping us to do

our tasks. Almost every buildingis equipped with fire and smokesensors, as well as surveillancecamera systems. TVs come withinfrared sensors. Any typicalcar is equipped with varioussensors like speedometer, fuel,temperature, oil pressure andcharging gauges, door lock andbrake indicators, and so on. Therole of sensors in industry is evenmore essential. Every electrical

appliance utilizes differenttypes of sensors. An airplaneflight goes smoothly and stayssafe largely due to thousandsof sensors. Any modern factorytoday is totally controlled bythe advanced sensing systems.Health monitoring systems along

with robotic technology help

physicians on disease diagnosis,treatment, and performingsurgical operations. So it is kindof hard to imagine the world

without sensors.

Wireless Sensor

NetworksThe use of sensors is growing fastand finds new applications in the

field of habitat monitoring, trafficcontrol, energy distribution, andhealthcare. A system includinga number of sensors spread outin an area that collect data andtransmit it to a central stationto be processed is called a

Wireless Sensor Network(WSN). In past decades, WSNis intensively investigated from

M Aultiple ccessfor

assiveP S

ensorsBy Abolfazl Razi

Figure 1: Showing how we are surrounded by sensors.

8/3/2019 EEWeb Pulse - Issue 25, 2011

9/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 9

PROJECT

different aspects includingsensor fabrication, sensorutilization, data communication,data processing, applicationdevelopment, and market study.Our main focus in this project ison the communication part.

The trend in WSN technologyis to build sensors as small andcost efficient as possible. In thenext-generation sensor systems,a sheer number of very tinysensors are spread out in an areato collect different data types

with very low resolution and high

accuracy. One of the challengingproblems in this field is how todevelop efficient communicationprotocols to carry this hugeamount of information betweensensors and the control unit in areliable fashion.

Passive sensors

Traditional sensors are of activetype. Active, in sensor technology,means that the sensors arebattery-run electrical devices.They include a measurement

Figure 2: An example of passive sensor: a Binary modulated reflective delay lineSAW Device.

block, a data processing unitthat converts the measurementto electrical signals, a radiotransceiver that transmits thesignals to the destination, and acontrol unit that controls differentparts. Even though these sensorsare very effective and widelyused, in some applicationsthey are not applicable due toharsh conditions like extremelyhigh temperature. For instance,inside a jet engine where thetemperature can go as high as1,000 degrees, no active sensorcan survive because of battery

explosion. In these applicationsbattery-free passive sensorsare the only possible choice.The WSN composed of passivesensors is called PWSN.

BPSK Modulated

Reflective Delay Line

The technology we have used toimplement a passive sensor isbased on Surface Acoustic Wave

(SAW) devices. SAW devices arebattery-free, long lasting, ruggedsensors that operate based

on the fact that acoustic wave velocity on the device dependson the measured parameter liketemperature, stress, and so on.These devices are implementedin various types including (i)Resonators, (ii) Delay lines,and (iii) Reflective delay lines.In this project, we are workingon temperature sensing usingbinary coded reflective delaylines that can be individuallyaccessed by DS-CDMA codesin a multiple access fashion.

As shown in Figure 2, each device

is composed of an antenna, anInter Digital Transducer (IDT),and two series of reflectors withmirrored pattern, all printedon a temperature-sensitivesubstrate such as YZ-LiNbO3.

An ElectroMagnetic (EM)radio wave is radiated by aninterrogator system. The EM

wave is received by the antennain the middle of the device andis converted to an Acoustic Wave(AW) by IDT. The AW travelsthrough the device surface inboth directions and is partiallyreflected back by specificallyshaped reflectors at two endsof the device. The AW then isconverted back to an EM waveand radiated by the antenna.This reflected signal is sensedby a very sensitive interrogatorsystem and is convolved with

the original signal. The resultingcurve, as illustrated in Figure3, includes two peaks andthe distance between peaksreveals information about themeasured parameter, in our casetemperature. These devicesand interrogator system aredeveloped at the University of

0 0 1 0 1...0 1 0 01 ...

d2

d1

8/3/2019 EEWeb Pulse - Issue 25, 2011

10/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 10

PROJECT

Maine as a joint project in LASTand WiseNet labs.

Multiple Access and

Interference Problem

In some applications, theinterrogation technique is

based on a point-to-pointcommunication, where a singlesensor is placed close to thedata source and an interrogatorsystem communicates with thesensor to read the measurement.In some applications, due tothe existence of many datasources, or the need for highermeasurement accuracy andresolution, more sensors are

required to be put in a cluster. Inthis case, when we are targetinga particular sensor, other sensorsalso respond to the interrogationsignal that causes an undesiredinterference effect and makes itdifficult to read the measurementof a specific sensor. So weneed to implement a Multiple

Figure 2: A sample response signal of the sensors.

Access (MA) technique forthe communication system.To do so in active WSN cases,different schemes such asTDMA, FDMA, and CDMA areused at sensor side. However,in a PWSN, since the passivesensors cannot perform signalprocessing, it is much morechallenging. In fact, we need torealize the MA technique on thesensor fabrication or push all theintelligence to the interrogatorside. In the first phase of ourproject, we use a combinationalmethod. Orthogonal codes aredeveloped on the sensors byusing different reflector patterns

on the device. In the interrogatorside, BPSK modulated DS-CDMA signals with GOLDcodes are employed. Thisenables us to have a simple MAsystem with a few sensors. Now

we are thinking of extendingour technique and going furtherby employing new techniques

to remove interference amongsensors and improve the systemperformance. One idea is toemploy a rotational directionalantenna structure based on arrayantennas in the interrogatorsystem that enables us to targeta specific sensor and reducethe interference level. Also, weare working on implementingan intelligent multi-stageinterference removal technique,in which we first try to detect thepatterns of interfering signalsand then remove them from thereceived signals by advancedsignal processing methods. Thisphase of the project is still in

the beginning steps and we arecurrently working on it.

8/3/2019 EEWeb Pulse - Issue 25, 2011

11/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 11

FEATURED PRODU CTS

Scheme-it Online Schematic Tool

Electronic components distributor Digi-Key Corporation, recognized by

design engineers as having the industrys broadest selection of electronic

components available for immediate shipment, and Aspen Labs, creator

of EEWeb, have produced the industrys first online schematic drawingtool, Scheme-it. Digi-Key is excited to launch Scheme-it, the industrys

only fully-online schematic tool, said Tony Harris, chief marketing

officer, Digi-Key Corporation. With Scheme-it, engineers are able to

easily create the blueprint for any design while gaining access to Digi-Keys expansive product listing in one online

experience. For more information, please click here.

Analog System Monitor and Controller

LMP92018 is a complete analog monitoring and control circuit which

integrates an eight channel 10-bit Analog-to-Digital Converter (ADC),

four 10-bit Digital-to-Analog Converters (DACs), an internal reference,

an internal temperature sensor, a12-bit GPIO port, and a 10MHz SPI

interface. The eight channels of the ADC can be used to monitor rail

voltages, current sense amplifier outputs, health monitors or sensors

while the four DACs can be used to control PA (Power Amplifier) bias points, control actuators, potentiometers, etc.

Both the ADC and DACs can use either the internal 2.5V reference or an external reference independently allowing

for flexibility in system design. For more information, please click here.

other portable electronic devices and will enable even more sleek products in the future. In addition to the space-

saving benefits, STs newest iNEMO module enhances user experience and motion-sensing realism through its

advanced design and real-time sensor data fusion capabilities. The module design ensures precise alignment of the

two sensors reference axes and superior thermal and mechanical stability, while STs iNEMO Engine data-fusion

software employs sophisticated prediction and filtering algorithms to automatically correct measurement distortionsand inaccuracies. For more information, please click here.

Small iNEMO Inertial Module

STMicroelectronics, a global semiconductor leader serving customers

across the spectrum of electronics applications and the leading supplier

of MEMS (Micro-Electro-Mechanical Systems) for consumer and

portable applications,1 announced a new inertial module that integrates

three-axis sensing of linear and angular motion in a miniature 3 5.5

1mm package. Reducing size by almost 20% over devices currently in

production, STs newest iNEMO module with six degrees of freedom

delivers advanced motion-sensing capabilities in todays space-

constrained consumer applications, such as smart phones, tablets and

8/3/2019 EEWeb Pulse - Issue 25, 2011

12/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 12

State

MachinesCoding Styles

Ray SalemiVerification Consultant

State machines are so common that there are tools

devoted to creating them by drawing circles and arts.

There are simulators that will recognize your statemachine and animate it to help you debug it. There are

even synthesis tools that will add error correcting logic

to your state machine so that it can recover from the

single event upsets that can happen at high altitude or in

electrically noisy environments.

You can take advantage of these tools by programming

state machines using a commonly accepted coding

style. Last month we began our discussion of this coding

style when we saw how to create named states and a

designated state register. This month well examine

various coding styles that we can use to create the next-state and output generating logic.

We are venturing into the world of combinatorial coding

style, a subject that is so controversial that adult engineers

have nearly come to blows when strong drink is present

at a design review meeting. (What, You dont have strong

drink present at your design review meetings?) This

article examines three basic coding styles you can use

to implement a state machine. We will use this simple

state machine as our example: the life of a dog.

Figure 1

sleeping

barking

happy

eating

tired

mouth

8/3/2019 EEWeb Pulse - Issue 25, 2011

13/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 13

TECHNICAL ARTICLE

The dog leads a simple life. By default he sleeps, but

if you feed him hes happy and then starts eating. After

hes done eating he goes back to being happy unless he

gets tired, you feed him again, or another dog walks in

the street. If there is another dog, he starts barking, if you

feed him he eats, and if hes tired he goes back to sleep.

The dog state machine has three inputs: tired, food, and

other_dog.

The dog state machine has two outputs: mouth and tail.

The mouth operates when the dog is barking or eating,

the tail operates when the dog is happy (product

marketing might argue that the tail could also operate

when eating, but were not implementing that approach).

Were going to code up our dog several ways.

Were going to see that HDL gives you control over

how your final code implements a state machine. There

are two steps to the process. First well choose a state

machine architecture, then well code it up. The key

is to realize that there are my possible state machine

architectures and that they all deliver different benefits

and challenges.

All state machine architectures have three or four pieces:

1. Combinatorial logic to determine the next state.

2. Combinatorial logic to determine the output basedon the state.

3. A state register.

4. Clocked output signals (optional).

The simplest way to implement a state machine is to

create a process for the first three pieces:

Figure 2

We have a cloud of logic implementing the next state

based upon the inputs and current state, a current state

register, and a cloud of logic implementing the output.

Here is the code that implements this simple state

machine:

Figure 3

You can code this kind of state machine with three

separate processes. This has the advantage of clarity.

Each process does one thing. Notice that the output logic

and the next-state logic have different sensitivity lists.

The output is only sensitive to the current state, while

the next-state logic is sensitive to the state and the input

signals. This is the definition of a Moore state machine.

We can also create a Mealy state machine if the output

logic is dependent upon the inputs as well as the state,as we see here:

Figure 4

Lets use a Mealy state machine to create a dog who

barks whenever there is another dog outside, regardless

of our dogs current state.

The rest of the state machine code is the same as in

Figure 2. The only difference is that we set the mouth

signal to be equal to the other_dog signal before

Next Stage

Logic

Output

Logic

State

Regis

ter

38

3940

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

always @(current_state, food, other_dog, tired)

begin : next_stage_block_proc

case (current_state)

sleeping:

if (food)

next_state = happy;

else

next_state = sleeping;

eating: next_state = happy;

happy:

if (other_dog)

next_state = barking;

else if (food)

next_state = eating;

else if (tired)

next_state = sleeping

else

next_state = happy;

barking: next_state = happy;

default:

next_state = sleeping;

endcase

end // Next State Block

71

7273

74

75

76

77

78

79

80

81

82

83

always @(current_state

)begin : output_block_proc

// Combined Actions

mouth = 0;

tail = 0;

case (current_state)

eating: mouth = 1;

happy: tail = 1;

barking: mouth = 1;

endcase

end // Output Block

89

90

91

92

93

94

95

96

always @(

posedge clk,

negedge rst

)

if (!rst)

current_state

8/3/2019 EEWeb Pulse - Issue 25, 2011

14/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 14

TECHNICAL ARTICLE

checking the state. Now another dog will get the mouth

going regardless of what our dog is doing (this is a fairly

accurate simulation of my dog).

Now that weve seen how we can separate the next-state

logic from the output logic, I have to tell you that I dont

like this coding style because it uses two case statements

to respond to the same information. This means that if I

add a state to my state machine, I have to add it to both

the next-state and the output case statements. I have a

hard enough time getting things right once, I hate having

to get things right twice. We can get around the problem

with this architecture:

Figure 5

Figure 6

See Figure 7 for the code.

Now I have one process that is sensitive to the state

and the input signals. The case statement is a little

more complex because I need to describe the outputs

response to the input signals in each state, and once

again I can wind up duplicating conditions if different

states respond to the input differently.

That said, I have one more pet peeve that we may want

to address. If the outputs of this state machine directly

drive the outputs of a module or component, then I have

a problem with unclocked outputs. Unclocked outputs

create design problems:

1. Synthesis tools cannot do timing analysis across

module boundaries. This means that if you have

10ns between clocks, then you can run into trouble if

generating the combinatorial outputs from one block

that takes 6ns while processing the combinatorialinputs of another block that takes 5ns. Both blocks

think they are meeting timing, but together they are

not.

2. You can get glitches on the output signals if they are

settling after youve changed your state.

3. Your synthesis tool will miss optimizing opportunities

because it cannot see a complete combinatorial

circuit that crosses module boundaries.

Figure 7

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

//-----------------------------------------------// Output Block for machine csm

//-----------------------------------------------

always @(

current_state, other_dog

)

begin : output_block_proc

mouth = other_dog;

tail = 0;

case (current_state)

eating: mouth = 1;

happy: tail = 1;

barking: mouth = 1;

endcase

end // Output Block

Next-State

& Output

LogicState

Register

38

39

40

41

42

43

4445

46

47

48

49

50

51

52

53

54

55

56

57

58

59

6061

62

63

64

65

66

67

68

69

70

71

72

73

74

always @ (current_state, food, other_dog, tired)

begin : next_state_block_proc

case (current_state)

sleeping: begin

mouth = 0;

tail = 0;if (food)

next_state = happy;

else

next_state = sleeping;

end

eating: begin

mouth = 1;

next_state = happy;

end

happy: begin

mouth = 0;

tail = 0;

if (other_dog)

next_state = barking;else if (food)

next_state = eating;

else if (tired)

next_state = sleeping;

else

next_state = happy;

end // case: happy

barking: begin

mouth = 1;

tail = 0;

next_state = happy;

end

endcase

8/3/2019 EEWeb Pulse - Issue 25, 2011

15/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 15

TECHNICAL ARTICLE

You can avoid all these problems by clocking your output

signals. This creates a circuit where your output signals

appear one clock cycle after the state that generates

them, but you avoid the problems of unclocked outputs.

This design looks like this:

See Figure 9 for the code.

Figure 8

Now we have one process that implements the entire

state machine. This process is sensitive to the clock and

the reset. On every clock edge the state machine outputs

the signals associated with the current state and moves

the state machine to the next state. Notice that mouth andtail are registers now, so they get set on line 51 along with

the current state. This state machine will deliver clean

edges on every clock.

Summary

Weve now examined several ways to code a state

machine. Design tools, simulators, and synthesis tools

will all recognize and generate state machines that you

code using any of these techniques. This of course raises

the question, Which is best?

The answer is, Whichever best matches your

application. If you have a complex output circuit, you

may choose to separate it from the next-state logic.

You may find that youd rather duplicate the state case

statement rather than duplicate the logic that responds

to inputs. You may not be able to afford the clock cycle

it takes to create clocked outputs. You may have coding

styles at work that limit your approach.

Figure 9

There is no single optimum style. Instead of slavishly

using one style because it was the first one we learned,

the best approach is to understand all our options and

make a conscious decision.

Next-State& Output

Logic

StateRegister

OutputFlops

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

6162

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

always @(posedge clk,

negedge rst

)

begin : clocked_block_proc

if (!rst) begin

mouth

8/3/2019 EEWeb Pulse - Issue 25, 2011

16/24

Single, Low Voltage Digitally Controlled Potentiometer

(XDCP)

ISL23315The ISL23315 is a volatile, low voltage, low noise, low power, I2C

Bus, 256 Taps, single digitally controlled potentiometer (DCP),

which integrates DCP core, wiper switches and control logic on

a monolithic CMOS integrated circuit.

The digitally controlled potentiometer is implemented with a

combination of resistor elements and CMOS switches. The

position of the wipers are controlled by the user through the

I2C bus interface. The potentiometer has an associated

volatile Wiper Register (WR) that can be directly written to and

read by the user. The contents of the WR controls the position

of the wiper. When powered on, the ISL23315s wiper will

always commence at mid-scale (128 tap position).

The low voltage, low power consumption, and small packageof the ISL23315 make it an ideal choice for use in battery

operated equipment. In addition, the ISL23315 has a VLOGIC

pin allowing down to 1.2V bus operation, independent from the

VCC value. This allows for low logic levels to be connected

directly to the ISL23315 without passing through a voltage

level shifter.

The DCP can be used as a three-terminal potentiometer or as a

two-terminal variable resistor in a wide variety of applications

including control, parameter adjustments, and signal processing.

Features 256 resistor taps

I2C serial interface

- No additional level translator for low bus supply

- Two address pins allow up to four devices per bus

Power supply

- VCC = 1.7V to 5.5V analog power supply

- VLOGIC = 1.2V to 5.5V I2C bus/logic power supply

Wiper resistance: 70 typical @ VCC = 3.3V

Shutdown Mode - forces the DCP into an end-to-end open

circuit and RW is shorted to RL internally

Power-on preset to mid-scale (128 tap position)

Shutdown and standby current

8/3/2019 EEWeb Pulse - Issue 25, 2011

17/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 17

All too often, designers spend so much time focusingon the specification and selection of their complex,

higher-cost devices such as processors, FPGAs, and

ADCs, that they dont fully take into account the major

performance impacts, as well as cost and power issues,

that are driven by the selection and integration of so-

called low-end components such as filters.

In fact, the specification and design of filter circuitry

can often be among the most important factors that

determine overall system success. For instance, if the

filtering design in a relatively low-bandwidth application

fails to eliminate high-bandwidth noise at the front-end of

the signal chain, the system has to burn extra power just

to propagate that unwanted noise down the signal chain.

Or, when processing signals of a certain bandwidth such

as in a channel-based radio receiver, an inadequate

filtering design can lead to unnecessary processing of

signals from adjacent channels. This leads to a waste of

power and a potential degradation of the target signal.

Proper filter design also prevents the signal path fromcontributing noise to other parts of the system, such as

keeping high-frequency signals from coupling back into

the power circuit and causing instability or fluctuations.

In essence, bypass capacitors are actually the simplest

category of analog filters and are widely used in all types

of designs.

Below is a brief overview of the role of filters, the types

of filters, and differences between digital and analog

approaches and passive versus active filtering designs.

Then, there are some specific design tips for simulating

and selecting analog filters for an example application

to drive high performance analog-digital converters

(ADCs).

The Role of Filters

Filters are electronic signal processing circuits that are

aimed at removing unwanted frequency components

from the signal, enhancing wanted signals, or both.

Filter Selection

& Design:The Gateway to

System Performance

Tamara SchmitzSenior Principal Applications Engineer

And Global Training Coordinator

8/3/2019 EEWeb Pulse - Issue 25, 2011

18/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 18

TECHNICAL ARTICLE

Electronic filters generally can be grouped into the

following categories:

High-pass filters attenuation of frequencies below

the cut-off points.

Low-pass filters attenuation of frequencies abovethe cut-off points.

Band-pass filters attenuation of frequencies both

above and below the target frequency band.

Notch filters attenuation of certain frequencies

while allowing all others to pass.

The frequency response of a filter is typically

characterized in terms of Bode Plots, which express

amplitude gain in decibels and phase in radians or

degrees. The bandwidth, f, of a filter is defined as

energy versus frequency. The Q factor of the filter isdefinedbyf0/fwheref0isthecenterfrequencyofa

bandpassfilter andfis thebandwidth.See Figure2

for example curves. The higher the Q, the narrower and

sharper the peak will be. In essence, the Q-factor is a

measure of quality for a particular resonance point on

the frequency spectrum.

Depending on the specific application requirements,

there are a number of different filter configurations that

can be considered to provide alternatives with regard to

optimizing the Q factor, the filters stability and other key

characteristics:

Chebyshev filter slight peaking/ripple in the

passband before the corner; Q>0.7071 for 2nd-

order filters.

Butterworth filter flattest amplitude response;

Q=0.7071 for 2nd-order filters.

LinkwitzRiley filter desirable properties for audio

crossover applications; Q = 0.5 (critically damped).

Bessel filter best time-delay, best overshoot

response; Q=0.577 for 2nd-order filters. Paynter or transitional Thompson-Butterworth or

compromise filter faster fall-off than Bessel;

Q=0.639 for 2nd-order filters.

Elliptic filter or Cauer filter add a notch (or zero)

just outside the passband, to give a much greater

slope in this region than the combination of order

and damping factor without the notch.

Digital Filters

Digital filters sample discrete time signals and need

to operate on digital inputs in order to provide digital

outputs. Digital filtering is a more complex approach

than analog and is typically used in conjunction withFPGAs or microcontrollers, where there is already a

significant amount of things like programming and

digital gates. Because the signals in these systems need

to be digitized to go into the FPGA or microprocessor

anyway, it is not a major cost to the system to digitize the

signals first and then conduct filtering within the digital

processing chain.

A digital filter usually consists of an analog-to-digital

converter to sample the input signal, followed by a

microprocessor and some peripheral components such

as memory to store data and filter coefficients etc. In atypical digital filtering application, software running on

a digital signal processor (DSP) reads input samples

from an A/D converter, performs the mathematical

manipulations dictated by theory for the required

filter type, and outputs the result via a D/A converter.

In mathematical terms, filtering is in essence the

multiplication of the signal spectrum by the frequency

domain impulse response of the filter. For an ideal

lowpass filter, the pass band part of the signal spectrum

is multiplied by one and the stopband part of the signal

by zero.

In addition to adding complexity, digital filters generally

consume more power than analog approaches. Also,

because digital filters use a sampling process and

discrete-time processing, there is an inherent latency

factor, which must be taken into account with regard to

overall design objectives. Digital filter designs can be

simplified by limiting the number of bits used, depending

on the accuracy requirements of the application.

Digital filters are most often used in systems that require

complex signal management and/or a high degree ofprecision. Since these designs digitize the signal before

they perform operations on it, they offer a great degree

of flexibility for complex mathematical functions. This

makes it very easy to change the coefficient in order to

tune the filter for the desired output, such as in a Finite

Impulse Response (FIR) filter design.

8/3/2019 EEWeb Pulse - Issue 25, 2011

19/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 19

TECHNICAL ARTICLE

Analog Filters

Analog filters are used to process continuous-time

signals and work directly on analog inputs, typically

providing a near instantaneous response from input to

output. In the ideal case, an analog filter would provide100 percent transmission within the specified frequency

passband and 100 percent attenuation outside of that

passband. However, in the real world, analog filters can

only approximate this ideal performance, with some

attenuation in the passband and less than 100 percent

attenuation over the stopband frequency range.

Figure 1: Comparison of Digital and Analog Filtering Approaches.

Analog is generally the preferred solution when the

designs filtering requirements are relatively simple

and especially if minimizing system cost and/or power

consumption is an important factor.

Passive vs. Active

If a very precise filter is required, designers may build it

from scratch using all passive components. However the

cost, space and complexity can often become prohibitive.

So, where possible, most designs lean toward an activefilter approach using a combination of op amps and

passives. This section provides a very brief overview of

the differences and the tradeoffs between passive and

active filter approaches.

By definition, a passive filter is made only from passive

elements. It does not require an external power source

beyond the signal itself. Since most filters are linear,

passive filters are typically composed of just the four

basic linear elements: resistors, capacitors, inductors,

and transformers. More complex passive filters may

involve nonlinear elements, or more complex linear

elements, such as transmission lines.

A passive filter has several advantages over an active

filter:

Guaranteed stability.

Passive filters scale better to large signals (tens of

amperes, hundreds of volts).

No power consumption.

May be less expensive in discrete designs (unless

large coils are required).

Passive filters are commonly used in applicationsinvolving higher voltage and current levels, such as

speaker crossover designs, filters in power distribution

networks, and power supplies. Passive filters are

uncommon in monolithic integrated circuit designs,

where active devices are inexpensive compared to

resistors and capacitors, and inductors are prohibitively

expensive.

Active filters are implemented using a combination

of passive and active (amplifying) components, and

require an outside power source. Operational amplifiers

are frequently used in active filter designs. These canhave high Q factor, and can achieve resonance without

the use of inductors. However, their upper frequency

limit is limited by the bandwidth of the amplifiers used.

Active filters have three main advantages over passive

filters:

Inductors can be avoided. Passive filters cannot

obtain a high Q without inductors but they are large

and expensive at low frequencies, have significant

internal resistance, and may pick up surrounding

electromagnetic signals.

The shape of the response, the Q factor, and the

tuned frequency can be set simply by varying

resistors and one parameter can often be adjusted

without affecting the others.

The amplifier powering the filter can also be used

to buffer the filter from the electronic components it

drives or is fed from, thereby eliminating variations

that could affect the shape of the frequency response.

Filter Type Characteristics Application Area

Digital Filters Digital Inputs & Digitaloutputs

High complexity Greater flexibility

FPGA & micro-processor based

systems High-precision,

complex filtering

requirements

Analog Filters Analog inputs & analogoutputs

Lower cost Lower power

Analog-onlysignal chains

Front-end filteringin mixed-signal

design

8/3/2019 EEWeb Pulse - Issue 25, 2011

20/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 20

TECHNICAL ARTICLE

Design Example: Modeling an Active

Filter for High-performance ADCs

When creating any mixed-signal system, the first thing

that the designer typically chooses is the ADC because

the system needs a certain number of bits, and a specifiedspeed and sampling frequency. Once the ADC has been

selected, everything else in the supporting circuitry is

aimed at driving the ADC to the optimal specifications,

without degrading performance. It is critical that the op

amps and other components in the filter circuit provide

high-speed signal processing without adding to the noise

floor in order for the ADC to run as close as possible to

its maximum specified dynamic range or resolution.

To model the specifics for implementing an active filter,

engineers need simulation tools that can take the target

filter shape and deliver a set of op amps and externalpassives that accurately achieve the desired filter

response. There are a number of such tools available

online but its important to keep in mind the key

differences between modeling with ideal specifications

versus modeling for real-world production. Most tools

estimate amplifier frequency response with a single-pole

approximation instead of taking into account the second

and third order effects present in commercial amps.

When filters were designed for low frequencies and the

amplifiers had abundant bandwidth compared to the

application, this was a reasonable estimation. This is no

longer true and a better simulation tool is crucial. One

such comprehensive simulation tool can be accessed at

web.transim.com/iSimFilter.

It is also imperative that the model simulation goes

beyond specifying just the op amps and also models

the various resistors and capacitors needed to complete

an optimal filter circuit for the specific application. For

example, Figure 2 shows a simulation from the iSimFilter

tool that models noise gain in dB for various resistor

levels. The Sallen-Key filter (SKF) used in this example is

an electronic filter topology used to implement second-

order active filters, (The alpha values listed represent

the ratio of feedback resistor to gain resistors).

In our specific design example for driving a high-speed

ADC, such as the 500MSPS ISLA112P50 12-bit and the

ISLA214P50 14-bit ADCs, the recommended op amp

solution is the ISL55210, which operates at very low

power (115mW) and has negligible noise with respect to

the ADC. Therefore, adding gain and filtering has vir tually

no impact on the system SFDR. The ISL55210 features

very high slew rates, low noise, ultra-low distortion and

Figure 2: iSimFilter Modeling of Sallen Key Filter Noise for Various Resistor Levels

Noi

seGain(dB)

a = 0.17

a = 0.4

a = 0.85

a = 1.1

a = 2.15

a = 3.1

SKF Noise Gain VS Resistor Ratio

Frequency (Hz)

1.00E+04 1.00E+041.00E+041.00E+04

8/3/2019 EEWeb Pulse - Issue 25, 2011

21/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 21

TECHNICAL ARTICLE

provides a 4GHz gain bandwidth with input noise of only

0.85nV/Hz and consistent performance over a wide

temperature and gain range. The op amp suppresses

even-order harmonic distortion, which is usually caused

by asymmetrical or unbalanced signal paths, andsupports gains greater than two with minimal bandwidth

or SFDR degradation.

Figure 3 illustrates a typical filter design using the

ISL55210 and passives support circuitry to drive a 12-bit

ISLA112P50 ADC.

Summary

Filtering may seem to be a simple design issue and

therefore it doesnt always get the attention that it

deserves. But filter circuits actually are the sentinels

at the gateway to the signal path and are critical forachieving full performance in the higher-cost devices

that form the heart of the overall system.

Proper selection, modeling and integration of the op

amps, passives and other elements of the filtering

circuitry can make or break the success of the overall

design. Even the most advanced ADCs, processors

and other high-end devices become a waste of money if

Figure 3: Filter design with ISL55210 and support circuitry.

they cant be driven as close as possible to their optimal

specification levels.

About the Author

Tamara Schmitz is a Senior Principal ApplicationsEngineer and Global Technical Training Coordinator

at Intersil Corporation, where she has been employed

since 2007. Tamara holds a BSEE and MSEE in electrical

engineering and a PhD in RF CMOS Circuit Design

from Stanford University. From 1997 until 2002 she was a

lecturer in electrical engineering at Stanford; from 2002

until 2007, she served as assistant professor of electrical

engineering at San Jose State University.

+3.3V

ISL55210

ISLA112P50

35mA(115mW)495

495

ADT4-1WT

100

1:2

40.2

210

210

33nH0.1F

0.1F

0.1F

Vdif

Vdif

Vi

Vi

20pF

20pF33nH

100

50

40.2

500kHz 180MHz SPAN

20log ( -------- ) = 17.3dB gain

105MHz SINGLE TONE

180mVpp for -1dBFS

500MSPSCLK

12 Bit

8/3/2019 EEWeb Pulse - Issue 25, 2011

22/24

BeStarACOUSTICS & SENSORS

Teamwork TechnologyInvention Listen Hear

PRODUCTS

Speakers

Buzzers

Piezo Elements

Back-up Alarms

Horns

Sirens/Bells

Beacons

Microphones

Sensors

INDUSTRIES

Automotive

Durables

Medical

Industrial

Mobile

Fire / Safety

Security

Consumer

Leisure

QS9000 TS/ ISO16949 ISO14001 ISO13485 ISO9001

bestartech.com | [email protected] | 520.439.9204

Preferred acoustic componentsupplier to OEMs worldwide

EEWebElectrical Engineering Community

Contact Us For Advertising Opportunities

[email protected]

www.eeweb.com/advertising

8/3/2019 EEWeb Pulse - Issue 25, 2011

23/24EEWeb |Electrical Engineering Community Visit www.eeweb.com 23

RETURN TO ZERO

8/3/2019 EEWeb Pulse - Issue 25, 2011

24/24

RETURN TO ZERO

EEWebElectrical Engineering Community

Join Today