Ransom’s notes on signal integrity and more. . .
Jitter, noise, BER analysis Articles Vital information on performing compliance and diagnostic tests for Gen4 enterprise receivers with Bit Error Rate Testers: Overcoming Receiver Test Challenges in Gen4 I/O Applications, Application Note for Tektronix, 2017. Intersymbol interference occurs when symbols overlap with each other: It’s Not Jitter, It’s Noise Tips from the pros on how to compare circuit simulations to measurements: How do I know if my simulation correlates to reality? Is the PRBS31 a good fit for high speed serial testing? Long test patterns cause mistakes no matter what equipment you use: Part 0 Test patterns are supposed to stress you out Part 1 PRBS31: Slower, costlier, worse Part 2 Test Pattern Redemption! The idea behind receiver tolerance testing is to submit the receiver to “the worst case but compliant” stress: Deterministic Jitter for High Speed Serial Receiver Tolerance Testing Receiver Tolerance Testing – with crosstalk! The Rules of Jitter Analysis - published in the online trade magazine, AnalogZone for Keysight Technologies. Analyzing Jitter at High Data Rates - published in IEEE Optical Communications. Conference papers, white papers & application notes (sponsored where noted) Advanced Serdes Debug with a BERT: Virtual Probing with Precision Stress and Error Location Analysis, learn simple strategies to pinpoint bit errors to the exact bit position and timing with powerful Error Location Analysis and a BERT, Application Note for Tektronix, 2017. Jitter analysis on Oscilloscopes and BERTs, White paper published and distributed by Anritsu Corp., 2014. Crossing the Digital-Analog Divide, a white paper for Keysight Technologies, 2012. Introduction to Precision Analysis of High Speed Serial Systems and Components, a webinar and Tech Note for Picosecond Pulse Labs, 2011. Signal Integrity Analysis at Extreme Data Rates - focusing on 100 Gb Ethernet to see where we're headed and what we need to anticipate, a White Paper written for Anritsu, 2010. Analysis of Random Noise and the Effect of Band-Limited Noise on Stressed-Eye Receiver Tolerance Tests - with Marcus Mueller, DesignCon 2009. Characterization of Gaussian Noise Sources - with Bob Muro, NoiseCom, DesignCon 2008. The Future of Multi-Clock Systems - with Roman Boroditsky and Juan Gomez of NEL, DesignCon2008; also available as a White Paper for NEL Frequency Controls, Inc., 2007. Using Clock Jitter Analysis to Reduce BER in Serial Data Applications - an Application Note written without a byline for Agilent Technologies, 2006. Measurement of Total Jitter at Low Probability Levels Using the Optimized BERT Scan Method - with Marcus Mueller and Russ McHugh, this paper presented a new technique invented by Ransom that reduced test time by a factor of ten with improved accuracy and won a DesignCon paper award. Jitter Analysis: The dual-Dirac model, RJ/DJ, and Q-scale - a Keysight white paper Ransom wrote that is the most frequently referenced paper on dual-Dirac and Q-scale by standards groups and other authors. Keysight white papers describing the techniques and results of an exhaustive comparison of jitter analysis techniques, a project that Ransom conceived and led, which solved the long-standing problem of how jitter ought to be measured: Precision Jitter Transmitter documents the most accurate jitter transmitter ever made. Comparison of Different Jitter Analysis Techniques with a Precision Jitter Transmitter - provides results of the study. Three part knowledge series sponsored by Tektronix (2008): Answering Gen-2 and Gen-3 Serial Challenges Jitter 360 - Six-part series on jitter analysis sponsored by Tektronix, 2005-2006 (download the whole series here): The Meaning of Total Jitter What the Dual-Dirac Model is and What it is Not All About the Acronyms: RJ, DJ, DDJ, ISI, DCD, ... Jitter Analysis in Systems with Crosstalk Clock Recovery in Serial-Data Systems Reference Clock Jitter and Data Jitter
© Ransom Stephens 2012 Training & Seminars Technical Marketing Ransomís Notes
Physics research with concepts that you can use A bluffers guide to evaluating scientific results: How to determine the accuracy of scientific results reported by journalists: Part 1: Systematic Bias Rules of thumb for estimating the significance of research results: Part 2: Rules of Thumb Irreproducible results have come under attack but sometimes that's just how nature rolls: Part 3: Why Some Results are Irreproducible It's time for educating technology to make a sharp turn from problem solving to simulating systems. It will be expensive but will have the greatest RoI of any investment in history: Simulated Education: STEM must change! The ultimate signal-to-noise problem: 1.) When neutrinos pass through matter, they rarely leave a trace, but when they do, it comes in the form of a ghostly ultra-violet cone of light: Measure of neutrinos, nature's most elusive particles 2.) Evidence for dark matter has been building for decades—it’s about time it showed up: Is dark matter about to be discovered? 3.) How physicists try to detect dark matter is a good example of how the humans roll: Measuring dark matter: The ultimate signal-to-noise problem Quarks follow a whole different set of rules than electrons. Electrons repel and the farther you separate them, the less force they feel—these behaviors are probably synonymous with your concept of “force,” but not after you read this article: The quirks of quarks Atomic wave functions have finally been measured and it’s now time to put the Bohr Model of the atom to rest, once and for all: Quantum wave functions come alive! May the Bohr Model rest in peace By solving a 10th grade science homework problem you can decide for yourself whether or not human beings are causing global warming/climate change by burning fossil fuels: Global warming/climate change: an easy calculation Properties of the calm before a storm can be measured and used to predict the transition from tranquility to disaster: Measuring the transition to disaster Can a butterfly flapping its wings in Japan cause a tornado in Silicon Valley? Ransom explains the roots of chaos with a simple carpentry metaphor: Measure of chaos: When uncertainty runs amok Gravity waves (1) Gravitational radiation must exist, but through a dozen years of searching, they have yet to be seen: Measuring gravitational radiation (2) We see electromagnetic waves, why not gravity waves? In search of gravity waves Flicker shows up in electric circuits, music, stars, brain waves, and the stock market and seems to indicate a significant relationship between distant events—but no one knows what it is: Noise in the time dimension: the strange case of flicker Discovery of the Higgs Boson: Pt 1, Herding cats on the Franco-Swiss border, Pt 2, What they detect Pt 3, What they actually measure Pt 4, Identifying the stuff in the detectors Pt 5, ATLAS and CMS - the biggest test devices on earth Pt 6, ATLAS and CMS - catching muons and neutrinos Pt 7, What a Higgs Boson looks like Pt 8, Is the particle observed at CERN really the HIggs Boson? Pt 9, What makes King Carl XVI Gustaf think it’s the Higgs Boson?
PAM4, crosstalk, ISI, equalization, channel operating margin Articles <video> Test, Measurement, and Analysis of PAM4 Signals (62 minutes) an introduction to the nuances of PAM4 signaling, techniques for evaluating if PAM4 is necessary for a given system, and test techniques for evaluating PAM4 signals and the performance of PAM4 receivers, emphasizing what can be leveraged from NRZ and what is new for PAM4 (to replay the seminar click on “register for this event.” The most complete and lucid description of Channel Operating Margin you will ever find, a White Paper for Anritsu: Measuring Channel Operating Margin Problems we’ve dealt with for a long time become more complicated in the move from NRZ to PAM4, but skew and compression are two all new problems: The biggest change in PAM4: How do the eyes line up? Understand how to use S-parameters in PAM4 systems, how to think of channel response and crosstalk in terms of S-parameters, and where to look for problems that might be caused by inaccurate measurements of S-parameters, a White Paper for Anritsu: PAM4 Demands Accurate S-parameters, 2016 A little out of date (tests have evolved), but probably still the most complete description of PAM4 signal and system analysis you can find in one place: PAM4 Signaling in High Speed Serial Technology: Test, Analysis, and Debug, an Application Note for Tektronix, 2016 Transmitter FFE tricks the channel into doing its own equalizing: Transmitter FFE makes the channel do the work Keep your eyes open for subtlety because PAM4 signal quality is measured by SER, not BER: Don’t confuse PAM4 SER and BER PAM4 measurements continue to evolve. Here are two ways to measure symbol level, separation, and nonlinearity, and which is best: PAM4: Symbol levels and voltage compression measurements Does COM set us up for failure in SerDes measurements? Can we avoid COMmitting the same mistakes we made with jitter? Eyes closed at high data rates, one way to analyze them is to remove the biggest cause of eye closure and see what’s left over: Analyzing closed eyes: CRJ and CDJ How emerging high speed specifications provide designers flexibility in how they balance loss and crosstalk: The tradeoff between crosstalk and loss Equalization rescues BER, but optimizing your Eq design at both the transmitter and receiver can be tricky: Optimize equalization for FFE, CTLE, DFE, and crosstalk FEC (Forward Error Correction) spends bandwidth to buy BER (bit error ratio), but it has to work just right or the whole thing falls apart: Why FEC plays nice with DFE At what cost did we recover almost 20% bandwidth in switching from 8B/10B encoding to 128B/130B? Trading signal complexity for bandwidth Channel Operating Margin provides a more flexible way to characterize channels and receivers in the presence of FEC, but what you get in flexibility you pay for in complexity: 100G and things to COM Every year, PAM4 threatens to replace non-return to zero signaling. It’s going to happen when multimoding kicks in: When will PAM4 take over from NRZ? Measuring ISI at high data rates is impossible Crosstalk problems are back, feature article for Test & Measurement World, July, 2012. Conference papers, white papers & application notes (Two+ year delay, sponsored where noted) Physical Layer Tests of 100 Gb/s Communications Systems, Tektronix Application Note, 2012, fully revised and re-released in 2015. A new Technique for Stessed Receiver Tolerance Testing Using Crest Factor Emulation, with John Calvin, Winner of the 2010 DesignCon Best Paper Award. Anticipating, Characterizing, and Avoiding Problems with Crosstalk  Equalization: The Correction and Analysis of Degraded Signals - a Keysight white paper also presented at DesignCon 2008 as Opening Closed Eyes: Analysis and Equalization of High Data Rate Signals on Buses and Backplanes.
Standards Updates Changes in going to Gen 4: adaptive equalization and link training, the impact of forward error correction (FEC) on compliance testing, debugging protocol handshaking and physical layer issues and new trends in channel performance evaluation, Understanding the Transition to Gen4 Enterprise & Datacenter I/O Standards, White Paper for Tektronix, 2017. Quick update on what’s hot in PCIe from the 2015 PCI-SIG Developer’s Conference: PCIe Update: Power, IoT, storage, OCuLink, simulation, and equalization Silicon photonics is sort of like fusion energy, always right around the corner. But this time, it's for sure: Silicon photonics: Will the hare finally catch the tortoise? With 100 GbE now developed and in deployment, we can now peek ahead to what's next: Four 400G quandaries to look forward to Logo testing lays down the law on not just compliance tests but how those tests must be performed. So if you’re a test engineer, why go to work? If you’re a tech company, why innovate? The Man has decided and you must conform! Are standards technological socialism? The big news at the PCIe-SIG Developers Conference was the announcement of a new specification for mobile applications: M-PCIe, here are the major details: M-PCIe – Goin’ mobile! How the OIF-CEI very short reach 25+ Gb/s conductor technology plugs a big hole in 100G communications: The very short reach of 25 Gb/s PCIe is twenty years old and PCIe 4.0 supports 16 GT/s: PCI Express update: the next, next, next generation Top ten reasons to love (and hate) 100 Gigabit Ethernet Peace at the plugfest
Tools and concepts for innovation and insight Is One Person’s Realith Anything Like Anyone Else’s? The funny thing about reality is that you can only get so close to it. Our senses compose an interface between our brains and the universe, a reality interface. 5 keys to unlocking your innate creativity The three timescales of thought dictate the time it takes to hit the brakes or to run from a tiger, but it takes longer to have an idea and much longer to learn how to do things quickly: The speed of thought Answer resolution combines the intense focus needed to find nuance in the very similar with the intense defocus needed to find relationships among the very dissimilar: Improve your answer resolution and unleash innovation The biggest obstacle to innovation is our own internal prejudice systems and the "not invented here" mentality. Fortunately, we can get around them: Reduce your prejudice to innovation Success in the 21st century is built on innovation but every industry builds TLA walls that stifle creativity: You can't think out of a box built of TLAs Moore's Law comes from the way people work and we can derive Moore’s-like laws for almost any developing technology or human endeavor: Moore's Law extends to cover human progress Do you go for sales leads? Networking? Solutions to your problems? Swag? Free beer? No, you’re there to get lucky, here’s how: Getting lucky at trade shows Diversifying the tech workforce means more than political appeasement; diversity sets the stage for innovation: The high tech diversity problem’s bottom line If you can possibly finagle this opportunity, do it right now!: Measure career advancement by blowing your mind Steve Weir shares the strategy that has made him a highly successful EE consultant for over 13 years: 10 tips on being a consultant Is it impossible in principle to create accurate economic models? Mismeasure of econoimcs Why do we pretend that small sheets of paper and cheap metal disks have inordinate value? It’s like the Emperor’s money is made of the same stuff as his new clothes, right? Well, sort of: Measuring money—the beer standard It's time for educating technology to make a sharp turn from problem solving to system simulation. It will be expensive but have the greatest RoI of any investment in history: Simulated Education: STEM must change From bee hives to consciousness, emergence seems like magic, but if you apply serious computing power, you can make sense of it: Measure of Emergence Are electronic health care records another inane federal boondoggle or are they a health care cure-all? The answer has more to do with Netflix than you might think: Can electronic health records save your world? The value of creativity far outweighs any other trait. Measures of creativity are elusive but the four key ingredients: Measuring and generating creativity Philosophies, codes of ethics, and religion all assume that we act with free will. Newton replaced free will with determinism; Heisenberg brought it back—or did he? The measure of free will and red herrings In a new eco-techno-science fiction thriller, three Si Valley geeks develop VR technology based on the relationship between the senses and the mind: The science and technology of Sensory Deception UC Berkeley neuroscientists posted a YouTube video that demonstrates rudimentary mind reading—We explain what they did: Decoding the brain using fMRI and YouTube Sure, the BRAIN initiative will cure diseases, but it will also open a Pandora’s box of philosophical issues from accurate lie detection to the existence of free will, but the technology begins with improving brain scan resolution: Resolving to map the brain: resolution, resolution, resolution If the position and function of every neuron is known, then a person’s thoughts and reactions can be predicted. Or not: Measure of the mind: connectome And you thought EDN was above proposing a diet for the holidaze: Measure of holiday calories - the thermodynamic diet