Форум » Дискуссии » Operazionnie ysiliteli ,ZAP/AZP & (продолжение) » Ответить
Operazionnie ysiliteli ,ZAP/AZP & (продолжение)
milstar: 1941: First (vacuum tube) op-amp An op-amp, defined as a general-purpose, DC-coupled, high gain, inverting feedback amplifier, is first found in US Patent 2,401,779 "Summing Amplifier" filed by Karl D. Swartzel Jr. of Bell labs in 1941. This design used three vacuum tubes to achieve a gain of 90dB and operated on voltage rails of ±350V. ###################################################### It had a single inverting input rather than differential inverting and non-inverting inputs, as are common in today's op-amps. Throughout World War II, Swartzel's design proved its value by being liberally used in the M9 artillery director designed at Bell Labs. ######################################################################### This artillery director worked with the SCR584 radar system to achieve extraordinary hit rates (near 90%) that ####################################################################### would not have been possible otherwise.[3] ########################### http://en.wikipedia.org/wiki/Operational_amplifier
milstar: SAR: Range Resolution http://www.eecis.udel.edu/~xma/ELEG467_667/SAR%20FINAL.pdf
milstar: http://www.afc-ingenieros.com/uploads/Afc/InfoTecn/RTSA/pdf/17%20Advanced%20Radar%20Analysis%2037W23378-0.pdf
milstar: Radar Digital Signal Processing James J. Alter Jeffrey O. Coleman Naval Research Laboratory http://www.mhprofessional.com/downloads/products/0071485473/SkolnikCh25.pdf
milstar: Stretch Processing ,str 9 ################### http://www.eecis.udel.edu/~xma/ELEG467_667/SAR%20FINAL.pdf – “De-ramp” received signal by mixing with a longer reference signal of the same chirp rate • Target returns are “beat frequencies” • A/D rate directly related to range window extent and low pass filter • Reduced sampling requirements eases downstream processing – The frontend hardware must support the additional RF bandwidth required for the “de-ramp reference” ---------------------- str.64 Recall: – Most SARs use Stretch Processing to deramp the transmitted pulse • Eases hardware and processing constraints • Also allows us to analyze SAR data as a projection of a 2 dimensional scene into a one dimensional function – Similar to Tomography!! – Let’s take a look at the stretch waveform and its output
milstar: Synchronous Detector http://www.radartutorial.eu/10.processing/sp06.en.html http://spiedigitallibrary.org/proceedings/resource/2/psisdg/2747/1/98_1?isAuthorized=no The combination of IF sampling and stretch processing is advantageous because A/D samplers are now able to operate at adequately short sample- and-hold aperture times, for use at IF frequencies, with a good number of bits resolution, and stretch processing can use narrow IF bandwidths. Therefore, high range resolution can be achieved at a lower cost than with quadrature channels at baseband and dual A/D's. Added benefits are the elimination of I-Q imbalance effects, A/D DC offset effects, and the need for calibration of these effects. Some A/D saturation can also be tolerated.
milstar: Understanding State of the Art in ADCs © 2008 Brad Brannon, Analog Devices, Inc. What http://www.converter-radio.com/understandingSOA.pdf
milstar: What are the Hardest Limitations? A number of limitations exist in data converters. As mentioned above, there is a natural conflict between SNR and SFDR. These are improved by advances in architecture and optimizations in process. While improvements have continually occurred over time, the process is at least partially reset each time a new semiconductor process is employed (see above). Because of this, improvements often tend to be two steps forward and one step backwards. Cost pressures push decisions to switch to newer processes whereas remaining on older processes would indicate that generational improvements in performance are possible but not necessarily a cost reduction http://www.converter-radio.com/understandingSOA.pdf
milstar: While there is a clear need for fullscale performance in the range of -110 to -120 dBFS, consistent performance of only -95 to -100 dBFS is available in the baseband region today. At the same time great attention has been paid to increasing SFDR performance in the mid and high IF regions with a result that usable IF frequencies as high as 450 MHz are not uncommon for some applications. How do we get there? Because of the physics of ADCs, there will always exist a natural opposition between spurious performance and SNR. Therefore, designers are constantly forced to compromise one specification for another. Therefore, it is key for Understanding State of the Art in ADCs Page 16 of 20 http://www.converter-radio.com/understandingSOA.htm 1/9/2012 customers to give feedback and guidance to ADC manufacturers. Without clear direction, ADC manufactures will move in the direction they are most drawn to which may not be the direction most needed by the customer. Clearly however, a focus needs to be placed on fullscale and near fullscale spurious performance. This will come in time but the issue is continually compounded by process changes that require lower power supplies
milstar: http://www.converter-radio.com/IIC2003.pdf Software Defined Radio, the new reality Brad Brannon & David Buchanan Analog Devices, Inc. Greensboro NC, USA
milstar: http://www.emrsdtc.com/conferences/2007/downloads/pdf/conference_papers/A023.pdf Dynamic Range Enhancements in Radar Sensors B. J. Harker, Z. Dobrosavljevic, E. P. Craney, S. Miles, R. A. Belcher* and J. Chambers* Roke Manor Research Ltd, Roke Manor, Romsey, Hampshire, SO51 0ZN, UK *Cardiff University, Centre of DSP, Wales, CF24 0YF, UK http://www.roke.co.uk, brett.harker@roke.co.uk © 2007 Roke Manor Research Ltd
milstar: re: Understanding State of the Art in ADCs © 2008 Brad Brannon, Analog Devices, Inc. What http://www.converter-radio.com/understandingSOA.pdf From:"Brannon, Brad" Analog Device To milstar Date:Sun, 15 Jan 2012 6:54 PM (15 hours 20 minutes ago) Show Raw MessageShow full header Sir Thank you for your inquiry and the references. I looked through them. They seem interesting. As for the converters, while I am not aware of any commercial technology right now, this technology seems feasible. Here is what I see today based largely on public domain information similar to those references you provided. SNR in 20 MHz BW with a sample rate of about 125 MSPS can approach the mid-80s. this seems consistent with what you have asked. As for SFDR, this is the difficult spec to hit. While these numbers are easily achievable for baseband (audio) converters, maintaining this kind of performance into the IF bands is very difficult. As some of your papers point out there are numerous digital techniques to linearize converters. We are certainly familiar with a number of these. Unfortunately as is apparent, this comes at the penalty of higher power. I would say that right now to achieve consistent performance above 100 dB at 70 MHz would be less than 5 watts. Perhaps 2-3 is possible. One key setback right now is customers are not willing to add the price of the current digital logic or dissipate the extra power to achieve this. However, I will say that as converters move to 65 nm and smaller geometries both cost and power will continue to fall enabling these kinds of features to be standard on many standard converter products. Stay tuned over the next 2 years or so. It probably will happen. As for the 20-24 bit question, that is more difficult to say at this point. I would guess this is 5-10 years before commercial products are available. Also, I'm not sure about the noise numbers. These may well be below what is possible from a thermal noise point of view. Regards, Brad ################## To: Brannon, Brad Subject: Fwd-2: 18 bit ADC, Fin 70 mhz ,20 mhz, 80-100 msps ,SFDR 100+ dBFS ,SNR 80+db by input -1 dbFS , Is is possible today ?/P.S. http://highfrequencyelectronics.com/Archives/May08/HFE0508_Cannata.pdf http://www.ll.mit.edu/HPEC/agendas/proc09/Day2/S4_1405_Song_presentation.pdf re:http://www.converter-radio.com/understandingSOA.pdf Dear Sir Author of this e-mail have some questions about ADC .EXcuse the author ,if questions are not relevant 1.Is it possible to develop today (from technology ,not hte market demand point of view ) 18 bit ADC, Fin 70 mhz ,BW= 20 mhz ,80-100 msps ,SFDR 100+ dBFS ,SNR 80+db by input -1 dbFS If powerd dissispated limited a. until 5 watt b until 50 watt 2. AN/ FPQ -6 C band Appolo radar http://en.wikipedia.org/wiki/AN/FPQ-6 have had Dynamic range in excess of 120 db , IF 30 mhz and BW until 2 mhz Is it possible to develop 20-24 bit ADC converter with atleast 10 msps Fin 30 mhz , SFDR 110-120 dbFS ,SNR 90 -100 dbfs Relevant comment would appreciated ##################################
milstar: From:"Brannon, Brad" Analog Device [Add] To: milstar Date:Mon, 16 Jan 2012 4:59 AM (5 hours 17 minutes ago) Show Raw MessageShow full header Sir Without an NDA in place, I am not at liberty to discuss much more. ############################################# NDA ? error - NSA ? (prim awtora postinga) I am sorry for the inconvenience. I will forward your requests to our converter marketing. Brad -----Original Message----- From milstar To: Brannon, Brad Subject: re3 -18 bit ADC, Fin 70 MHz ,20 MHz, 80-100 msps ,SFDR 100+ dBFS ,SNR 80+db by input -1 dbFS To Analog Device Mr. Brad Brannon Copy to GMR (DARPA NLEQ) Gil Raz Dear Sir Thank you for your answer You wrote 1 ... SNR in 20 MHz BW with a sample rate of about 125 MSPS can approach the mid-80s. this seems consistent with what you have asked. 2...I would say that right now to achieve consistent performance above 100 dB at 70 MHz would be less than 5 watts. Perhaps 2-3 is possible Witch kind of result be matched ,from technology perspective (without limitation of market demand ,expense and another questions ) ? if power consumated of one core ADC limited untill 50 watt and ADC can be cryogen cooled/example 1 kg belov) ? On 70 mhz ? For 20 mhz IF ,Input -1 db DBFS 60-100 msps 18-20 bit ? ,SNR more as 8x dbfs ? ,SFDR more as 1xx dbFS ? for ADC only ################################################################## 120 dbfs with NLEQ from DARPA ? Is it realistic ? For DARPA project ? EXcuse the author if questions are not relevant 3. As you knew -most broadly used PAW 70 mhz and 140 mhz http://www.triquint.com/products/types/filters/if/ http://www.triquint.com/products/types/filters/if/standard-70mhz-family.cfm http://www.triquint.com/products/types/filters/if/standard-140mhz-family.cfm 4. From M.Skolnik NRL Bandwitch for different radar mode For most of the function Fin/BW 70/200 mhz or 140/40 mhz For SAR/ISAR -strech processing Table 5.1 shows the range of parameters that can be observed as a function of radar mode. 5.12 MULTIFUNCTIONAL RADAR SYSTEMS FOR FIGHTER AIRCRAFT 1.Real beam map 0.5 -10 mgz 2.Doppler beam sharp 5-25 mgz 3. SAR 10 -500 mgz 4.A-S range 1-50 mgz 5.PVU 1-10 mgz 6.TF/TA 3-15 mgz 7.Sea surface search 0.2 -500 mgz 8.Inverse SAR 5-100 mgz 9. GMTI 0.5-15 mgz 10.Fixed target track 1-50 mgz 11.GMTT 0.5 -15 mgz 12.Sea Surface track 0.2-10 mgz 13.Hi power Jam 1-100 mgz 14.CAl/A.G.C 1-500 mgz 15A-S data link 0.5-250 mgz 6. Exist some Radar ,where can be accepted cryogen cooling of LNA ,Mixer ,IF and ADC a. All with average(!) power 1 megawatt -SBX, Don-2N, Cobra Dane and 250+ square metr appertura b. Some of radar as russian A-2500/S-300v4 have electrostatic amplifier ( can hold miltiple 100 watt on LNA input ) example of miniature cryogen cooling http://callisto-space.com/en/products/miniature-cryogenic-low-noise-amplifiers.html
milstar: definizii oversampling and undersampling ot razrabotchika http://www.edn.com/archives/1996/110796/23df_07.htm?text=brannon
milstar: smotri gl. 2.13 M.Skolnik ,Radar izdanie 2008 goda http://www.scribd.com/doc/49249408/0071485473-Radar-Handbook3rd 2.13 Dynamic Range and A/D convesion cosiderations
milstar: smotri M.Skolnik ,Radar izdanie 2008 goda 4.3 Dynamic Range and Stability requirement 4.24 ADC in Doppler radar http://www.scribd.com/doc/49249408/0071485473-Radar-Handbook3rd ...The most stressing dynamic range requirement is due to main beam clutter ,when searching for small low flying target ------------------------------------------- example 4.15 C/N 53 db na wisote RLS 1000 fut/300 metrow trebuetsja 12 bit dlja 63 db ------------------- 12 bit ENOB -eto 16 bitnie AZP ######################### 12 bit wisokoskorostnie AZP/ 1-3.6 GSPS / e2v,TI,National eto 9 bit ENOB
milstar: Statja Analog Device -averaging 32 ADC + 15 db in SNR http://www.analog.com/library/analogDialogue/archives/39-05/multichannel.pdf Lutschie 16 bit ADC segodnja dajut po 80 -81 db SNR na 70 mhz pri 80 -125 msps (TI ADS5481-83, LLTC LT2215-17,) i li 84 db na 10 mhz(TI) pri cene 40-70 $
milstar: http://www.ti.com/lit/ds/symlink/ads5482.pdf ADS5482 na 30 mhz(IF programmi Appolo AN/FPQ -6 http://en.wikipedia.org/wiki/AN/FPQ-6) SNR 80.7 dBFS SFDR -98 dbc SINAD-79.3 dbc pri 105 msps na 70 mhz SNR -80.1 dBFS SFDR -91 dbc /silno padaet) SINA-78.2 dbc ENOB na 30 mhz -12.88 bit ################### Konkurent Low noise LTC221 7105 msps http://cds.linear.com/docs/Datasheet/2217f.pdf 30 mhz SNR -81.1 dBFS SFDR -95 dbc ( pri 25°C) SINAD -81.1 70 mhz SNR -80.4 dBFS SFDR -92 dbc(25°C) SINAD -80
milstar: Esche odin maloschum AD9265 http://www.analog.com/en/analog-to-digital-converters/ad-converters/ad9265/products/product.html Wse 3 stojat menee 100$ za stuku T.e. wpolne mozno ispolzowat averaging 32 stuki na 70 mhz IF s 20 mhz polosoj teoreticheski SNR 80 + 15 dbc (is stati AD wische) Prakticheski werojatno meschne TI poluchilo + 4.8 db dlja 4 ADC experimentalno No s xoroschim chansom SFDR budet 93 + 10 db ot NLEQ/DARPA SNR bolee 90 dbc a okno pri strech processing w 5 raz wische chem y MMW /Lincoln po publichnim dannim /nawernjaka yze modernizirowan/= 20 mhz
milstar: ypominaetsja averaging 100 ADC s ylutschenime SNR 20 db High speed analog to digital converter basics http://www.ti.com/lit/an/slaa510/slaa510.pdf
milstar: http://www.mikron.sitronics.ru/products/micron/technology/ Технологии В «СИТРОНИКС Микроэлектроника» для разработки и производства микросхем используются следующие технологические процессы: БиКМОП SiGe • Проектная норма – 250 нм (2011 год) • Проектная норма – 180 нм (2012 год) • СВЧ и УВЧ до 10ГГц (АФАР, конвертеры и синтезаторы для радиолокационных систем и спутниковой связи) --------------------------- Dostatochno dlaj razrabotki Rossijskogo 16 bit AZP s 250 msps na Fin 300 mhz ,SFDR 90-95 dbFS ,SNR 73-76 dbc
полная версия страницы