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: Overview of Radar DMTI Processing The SPS-48E radar (Fig. 1) uses a triple conversion receiver. ########### The system is wideband until the second intermediate frequency (IF) conversion, where the individual beams are bandpass filtered and separated. Since three beams are used in the DMTI, there are three coherent oscillator frequencies (one for each beam) in the final conversion of the receiver (final IF is about 1.5 MHz). ################ A single analog-to-digital (A/D) converter is used for each beam. In-phase and quadrature (I/Q) data are developed based on samples that are spaced at multiples of 90° at the IF frequency. The interpolation filter develops the I/Q estimates from A/D samples (see the boxed insert, Intermediate-Frequency Sampling Technique). The I/Q data preserve the amplitude and phase of the IF radar return. The amplitude of the radar return is computed as ( ). I Q 2 2 + The phase of the return is computed as tan21 (Q/I). From pulse to pulse, a phase progression will be seen on moving targets due to Doppler, and no phase progression will be seen on stationary reflectors. It is this phase progression on moving targets that allows such targets to be separated from stationary reflectors (clutter). To remove clutter and pass targets, DMTI filters are employed in each beam independently. A bank of digital filters is used to cover the region between low velocity (small phase shift per pulse) and higher velocity (near 360° phase shift per pulse). Targets moving at speeds such that they present more than a 360° phase progression per pulse are said to be velocity ambiguous, since the radar pulse repetition interval causes aliasing. For example, a phase progression of 400° per pulse appears exactly as a phase progression of 40° per pulse. To avoid velocity blinds (i.e., targets moving at speeds such that their phase progression is 360° per pulse, thus appearing as 0° per pulse), the pulse repetition frequency is jittered on a burst-to-burst basis. This ensures that the phase progression presented by the target will vary on a burst-to-burst basis, and thus the target will not be velocity blinded on all bursts http://www.jhuapl.edu/techdigest/TD/td1803/roul.pdf Description The AN/SPS-48G is a long-range, three-dimensional (3D) Air Search Radar that will be installed on CVN, LHA, LHD, and LPD 17 class ships. The AN/SPS-48G is used to find full volumetric detection data for Ships Self Defense System and the Cooperative Engagement Capability (CEC), Air Intercept Control, Anti-Ship Cruise Missile detection including Low Elevation and High Diver targets, backup aircraft marshalling, and the new Hazardous Weather Detection and Display Capability. http://www.navy.mil/navydata/fact_display.asp?cid=2100&tid=1250&ct=2 AN/SPS-48E - Compared to the C variant, the SPS-48E has twice the radiated power, increased receiver sensitivity, four stage solid-state transmitter, half the components of a -48C and built-in testing for easier diagnostics. Originally developed as part of the New Threat Upgrade (NTU) Program to support the SM-2 Launch On Search (LOS) capability. 1975 under the Guided Missile Frigate Anti-Air Warfare Modernization Program. The AN/SPS-48E included a digital receiver and signal processor that could automatically detect and track very small targets, even when jammed. It was included in the New Threat Upgrade of the 1980s. The deployment of the AN/SPY-1 and the end of the Cold War led to the decommissioning of a large number of such ships, and many of these vessels AN/SPS-48 sets were reused on aircraft carriers and amphibious ships, where it is used to direct targets for air defense systems such as the Sea Sparrow and RIM-116 SAM missiles. Existing sets are being modernized under the ROAR program to AN/SPS-48G standard for better reliability and usability. ################# INTERMEDIATE-FREQUENCY SAMPLING TECHNIQUE To develop in-phase (I) and quadrature (Q) data, the SPS-48E radar uses an intermediate-frequency (IF) sampling technique with an IF bandwidth of approximately 400 kHz, IF center frequency of about 1.5 MHz, and analog-to-digital (A/D) sampling frequency of 6 MHz. There is a precise 4:1 relation between the IF sample frequency and the IF center frequency. If modulation effects across the received pulsewidth are ignored, the echo may be thought of as several cycles of a sine wave. The sine wave is sampled at four times its rate, i.e., every 90°. Therefore, alternate samples will be in quadrature with each other. To account for modulation effects across the pulse, one sample is defined to be “I”; two leading and two trailing samples are combined by the following equation to create the “Q” sample (s): 180° phase shift 90° phase shift Time Q ssss =− − + + 1 16 9 16 9 16 1 16 1234 Q I QI Q I Q s s Is s −− −− 1234 This technique provides accuracy acceptable for the clutter cancellation requirements of the SPS-48E lowelevation-mode DMTI. If higher clutter cancellation is required, a more elaborate finite impulse response filter for both the I and the Q channel is required. The advantage of the current technique is that I/Q data are developed with only a single A/D converter. The two baseband analog channels in a conventional receiver are not required, and aliasing due to channel gain mismatch is avoided. Amplitude modulation effects across the received pulse do, however, cause some degradation.
milstar: Динамический диапазон радара AN/FPQ программы Аполлон более 120 дб Антенна 8.8 метра диаметром C band 5.4-5.9 Ghz 4.8 квт средней мощности,3 мегаватта импульсной мощности промежуточная частота-30 мегагерц, полоса сигнала -1.6 мегагерц Дальность более 60 000 километров при разрешении +- 2 метра http://en.wikipedia.org/wiki/AN/FPQ-6 http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19680003409_1968003409.pdf
milstar: The three radar intermediate-frequency inputs to the A/D board are 20 MHz bandwidth centered at 10 MHz, and are thus sampled with a 40 MHz clock http://www.ll.mit.edu/publications/journal/pdf/vol21_no1/21_1_7_Eshbaugh.pdf FIGURE 16. Single-channel radar channel processing performed by DPCS for a typical stretch waveform.
milstar: an/tpy-2 We assume a module duty factor of 0.2, which is consistent with the module peak and average powers discussed above.[8] With the 1 millisecond pulses length assumed above, this gives fP = 200 Hz. Thus for our baseline case which integrates 20 pulses, the dwell time will be 100 milliseconds. http://mostlymissiledefense.com/2012/09/21/ballistic-missile-defense-radar-range-calculations-for-the-antpy-2-x-band-and-nas-proposed-gbx-radars-september-21-2012/ S/N: For our baseline case, we consider two values of S/N. First a low value of S/N = 20 (which we refer to as the “detection” value) and a higher figure of S/N = 100 (which we refer to as the “discrimination” value).[9] LS: We estimate LS = 8 dB = 6.3.[10] Then for our TPY-2 baseline cases we get: R = 870 km detection (S/N = 20) R = 580 km discrimination (S/N = 100)
milstar: In the linear ASIC design arena there is an even wider range of services — where full turnkey delivery is offered. Internally developed linear designs are sent out for fabrication and packaging, but returned for advanced on-site testing. Raytheon currently tests and delivers over 100K linear ICs per year. http://www.cs.wustl.edu/~schmidt/Tech_030612.pdf
milstar: The SPS-48E radar (Fig. 1) uses a triple conversion receiver. The system is wideband until the second intermediate frequency (IF) conversion, where the individual beams are bandpass filtered and separated. Since three beams are used in the DMTI, there are three coherent oscillator frequencies (one for each beam) in the final conversion of the receiver (final IF is about 1.5 MHz). A single analog-to-digital (A/D) converter is used for each beam. In-phase and quadrature (I/Q) data are developed based on samples that are spaced at multiples of 90° at the IF frequency. The interpolation filter develops the I/Q estimates from A/D samples http://techdigest.jhuapl.edu/TD/td1803/roul.pdf The clutter-to-noise ratio presented to the radar by this land has peaks on the order of 100 dB, thus Figure 7. Beam 1 land clutter data collected approximately 14 nmi off the coast of Point Loma, California, using the normal 3-1-2 beam transmit sequence with a wideband limiter at 36 dB: (a) normal sensitivity time control, (b) constant 30-dB attenuation, (c) constant 50- dB attenuation, (d) constant 70-dB attenuation. Noise level is approximately 24 dB. Clutter-to-noise ratios on the order of 100 dB are seen. 20 10 20 10 Range (nmi) Range (nmi) 20 10 20 10 Range (nmi) Range (nmi) 55 80 60 65 70 75 Bearing (deg) 55 80 60 65 70 75 Bearing (deg) 55 80 60 65 70 75 Bearing (deg) 55 80 60 65 70 75 Bearing (deg) > 60 60 ≥ 52 52 ≥ 44 44 ≥ 36 36 ≥ 28 28 ≥ 20 ≤ 20 Amplitude (dB) (a) (b) (c) (d) confirming that land clutter can definitely present very large clutter echoes, an enormous challenge for radar systems.
milstar: USS Abraham Lincoln receives AN/SPS-48 primary air search radar antenna 0 15 May 2015 Newport News Shipbuilding, along with the US Navy, has successfully reinstalled the AN/SPS-48 primary air search radar antenna on the aircraft carrier, USS Abraham Lincoln (CVN 72). This new development is part of the ongoing refuelling and complex overhaul (RCOH). Combat Systems Department division officer lieutenant Loudon Westgard said: "Installing the radar on time is one of the most important measures taken in the refuelling and complex overhaul process. "This was a major accomplishment, and the shipyard workers and sailors aboard Lincoln should be very proud of the progress they are making." The long-range, three-dimensional air search radar AN/SPS-48 a for 360° coverage. It is also capable of dectecting the height of a target above the surface of the water. http://www.naval-technology.com/news/newsuss-abraham-lincoln-receives-ansps-48-primary-air-search-radar-antenna-4578470
milstar: В группу компаний «Ангстрем» входит расположенный в Зеленограде АО «Ангстрем-Т» – научно-производственный комплекс по производству субмикронных полупроводниковых изделий по технологическим нормам 130–90 нм, с перспективой перехода на производственный уровень 65 нм и нижe http://oborona.gov.ru/news/view/12040 для высокоскоростных аналого-цифровых преобразователей 1-5 gigasample достаточно
milstar: http://www.e2v.com/resources/account/download-datasheet/4368
milstar: BarsMonster 21 февраля 2014 в 00:10 Микрон: Чуть детальнее о производстве 65нм микросхем в России IT-инфраструктура Вчера все отечественные сайты облетела новость о том, что в России Микроном разработана технология производства микросхем по нормам 65нм (или даже «В России выпущены первые 65-нм микросхемы»). Ранее Микрон имел лицензированную у STMicroelectronics технологию 90нм. Попробуем чуть детальнее разобраться, как там обстоят дела. Микрон на этот раз на удивление опубликовал достаточно много информации. На фотографиях — разметка одного тестового транзистора и фотографии сделанные электронным микроскопом. Под катом — посмотрим, как это можно было сделать и сравним с Intel 65nm. Сравнение техпроцессов Микрон опубликовал таблицу с параметрами их техпроцесса. Для сравнения, я добавил техпроцесс Intel 65нм: Технология Микрон 65нм Микрон 90нм Intel 65нм Gate length 45нм На фотографии 54нм 65нм 35-38нм +SiGe stress Gate oxide thickness (electrical) 2.2nm (n) / 2.2nm (p) 2.2nm (n) / 2.2nm (p) 1.2nm SiON Interconnect 9-Cu + 1-Al 7-Cu + 1-Al 8-Cu Metal 1 pitch 0.18µm 0.24µm 0.21µm Inter-level dielectric k = 2.9 k = 2.9 k=2.9 M1 pitch (шаг первого уровня металлизации) вызывает некоторое сомнение — по мере уменьшении шага металла M1 менее 0.2-0.3мкм (для 65нм технологии) быстро падает скорость работы микросхемы из-за увеличения RC-константы, потому Intel и не стал его уменьшать менее 0.21-0.22мкм. Очередное напоминание, что именно межсоединения являются основным тормозом прогресса микроэлектроники. Длина затвора и толщина подзатворного диэлектрика говорит о том, что это LP техпроцесс — с низким потреблением и меньшей скоростью работы. Так что сделать процессор, аналогичный первым Core2Duo на Микроне пока не выйдет, но и для LP техпроцессов есть масса применений. Количество металлов позволяет реализовывать процессоры любой сложности. В погоне за 65нм Как мы помним, разрешение оптической фотолитографии подчиняется критерию Рэлея: На данный момент самая продвинутая установка фотолитографии на Микроне (сканер ASML PAS/1150C) имеет NA=0.75 и работает на длине волны 193нм. Параметр k — множитель используемых «ухищрений», позволяющих улучшить получаемое разрешение. k для фотолитографии без хитростей — 0.4. В случае Микроновских 90нм — k был уже 0.35. Чтобы с тем же сканером получить честные 65нм, k нужно было бы как-то снизить до 0.25 (т.е. добавить достаточно много хитростей). Однако учитывая слова из пресс-релиза («были разработаны специальные алгоритмы внесения оптической коррекции фотолитографии»), обычной, классической топологии тестового транзистора (не используя «одномерные» структуры) и длину затвора на фотографии (54нм) — на данный момент похоже просто на текущем оборудовании без дополнительных хитростей сделали транзисторы с затвором меньшего размера для первых тестов (это резко увеличивает процент брака, но для тестовых транзисторов приемлемо) + отработали новые технологические шаги техпроцесса, отличающиеся от 90нм. Говорят, в Марте 2014 года на Микроне ждут приход нового сканера — и там 65нм получится без дополнительных хитростей, а с хитростями — и более тонкие техпроцессы (45нм, ниже?). Вот тогда, к концу года (а то и в 2015) — и выйдут первые полноценные микросхемы по технологии 65нм. Объем производства ожидается порядка 500 200мм пластин в месяц — это практически гарантирует, что производство получится очень дорогим, и доступным только для государства. Наконец о возможных хитростях 65нм можно было получить и на текущем оборудовании Микрона. Достаточно вспомнить про то, как Интел в 2007-м сделал 45нм техпроцесс на «сухой» фотолитографии используя сканер с апертурой 0.93 (у Микрона напомню 0.75): критические слои экспонировали в 2 захода: в первый заход экспонировали ряд горизонтальных линий (используя dipole illumination, поляризацию — так можно достичь большего разрешения, но только вдоль одной оси). Затем второй экспозицией нарезали линии на кусочки нужной формы. Результат на фотографии. Собственно, аналогичным образом получается разрешение 32нм. Этот подход позволяет получить k=0.21, и для Микроновского сканера это позволило бы получить 55нм техпроцесс. Но безусловно объем работ был бы весьма внушительным. Резюме Говорить о «65нм микросхемах сделанных в России» пока преждевременно — это единичные тестовые транзисторы на существующем оборудовании. Технология LP (бОльшая длина затвора, более толстый подзатворный диэлектрик) — с низким потреблением и меньшей скоростью, ожидать процессоров аналогичных Intel 65nm (первые Core2Duo) не стоит. С новым оборудованием (в первую очередь сканер), которое должно заработать на Микроне в этом году — будут возможна как 65нм технология, так и более тонкие. Из-за очень маленького объема производства (500 пластин в месяц) себестоимость пластины обещает быть довольно высокой, завалить конкурентоспособной гражданской 65нм продукцией рынок не выйдет. Но этого и не требовалось.
milstar: БЦВМ на процессорном модуле с Эльбрус 4С Производительность БЦВМ (при работе с 32 разрядными числами), ГФлопс Не менее 35 Программное обеспечение операционную систему «Эльбрус», драйвера носителя мезонинов Габариты (Д x Г x В), мм не должны превышать размеров 280×230×110 Масса, кг Не более 5,5 Средняя наработка на отказ в полете, ч не менее 9000 http://www.ipmce.ru/custom/path7/path1/path2/
milstar: STRECHING THE DYNAMIC RANGE OF ADCS WHITE PAPER http://www.rfel.com/upload/docs/12013StrechingthedynamicrangeofADCsWP1.0.pdf
milstar: http://www.ti.com/lit/ds/symlink/adc12dj3200.pdf enob 9 bit 997 mhz sfdr -69 dBFS 997 mhz up to 6.4 GSPS in single channel mode Up to 3.2 GSPS in dual channel mode Analog input bandwidth (-3 dB): 8.0 GHz allows direct RF sampling of L-band,S-band, C-band and X-band for frequency agile systems. Power consumption: 3.0 W ################## Time interleaving is achieved internally through 4 active cores. ######### In dual channel mode, two cores are interleaved per channel to increase the sample rate to 2x the core sample rate. In single channel mode, all 4 cores are time interleaved to increase the sample rate to 4x
milstar: http://www.mwrf.com/systems/differences-between-receiver-types-part-1 ##### As a general statement, a properly designed superheterodyne receiver will have far superior sensitivity and immunity to interference when compared to a direct conversion receiver. ############################### For a phased array digital beam-forming the challenge becomes size, power, and cost constraints when many receivers are needed across the array. http://www.mwrf.com/components/receiver-design-considerations-digital-beamforming-phased-arrays Peter Delos is lead RF/RFIC engineer for Lockheed Martin Corp.
milstar: AD9208 0.028 micron CMOS 2*14 bit 196 ball bga 12*12 mm http://www.analog.com/media/en/technical-documentation/data-sheets/AD9208.pdf ENOB 9.6 1800 mhz a in -9dbfs SINAD 59.7 dbfs SFDR -81 dbfs ########### eight-lane operation, with lane rates of up to 16 Gbps/lane. Using default settings, total power per channel at 3 Gbps is 1.65 W. temperature range of -40°C to +85°C, the AD9208 costs $1326. Based on a 28 nm CMOS process, the AD9208 dual 14 bit ADC from Analog Devices enables IF sampling of signals at up to 9 GHz (-3 dB point) When wide signal bandwidths are required, gigasampling high speed data converters from ADI provide the direct RF conversion capabilities to support advanced multiband radios designs. Frequency agility and signal aggregation are necessary for ease of deployment in regions where service providers own fragmented frequency bands and desire a single radio design to cover them all. RF DACs and RF ADCs from ADI support wide signal bandwidths up to 1.5 GHz. Enhanced JESD204B serial lanes support rates up to 15 Gbps, reducing the number of lanes required for data transport. Outstanding linearity performance enables direct signal generation up to 4.2 GHz and direct signal capture up to 6 GHz http://www.analog.com/media/en/news-marketing-collateral/product-highlight/AD9208-AD9172-High-Speed-Converters.pdf
milstar: 4. Studio-quality audio Data converters are also playing a critical role in enabling the high-resolution audio content for ultra-high-quality music playback. They help filter the unwanted noise and provide immunity against high jitter. Secondly, they ensure low power consumption to maximize the battery life of music playback devices like headphones. The CS43130 digital-to-analog converter from Cirrus Logic is a case in point. It consumes 23 milliwatts of power, which, according to the audio chipmaker, is four times less than other high-fidelity DACs available in the market. And it offers up to 32-bit resolution and 384-kHz sampling rate to deliver superior audio quality.
milstar: Benefits of Using 28 nm and Lower Process Converters As converter transistor sizes decrease, the parasitic gate capacitance is reduced, and converters are able to be operated at faster sample rates. These faster rates result in generally wider Nyquist bandwidths, enabling the broad instantaneous bandwidths required for EW systems. Additionally, smaller transistor size results in the ability to fit more transistors onto a given die, resulting in greater converter channel counts per package. Faster sample rates and greater channel counts provide the ability to achieve both the broad instantaneous spectrum sniffing and simultaneous multi-band interrogation required for signals intelligence (SIGINT) systems. Reduced feature sizes could potentially enable the integration of both a DAC and ADC core into a single monolithic package, creating the ability to design full-duplex EW transmitter and receiver systems with a single converter chip. Further, smaller transistor sizes allow for additional circuitry to be added to converter packages, such as numerically-controlled oscillators (NCO) and digital down-converters (DDC), to simplify the large digital signal processing (DSP) tasks required by EW systems. 28 nm processes lead to smaller cross-sections, usually resulting in fewer radiation-induced soft errors compared to similar features in a 65 nm process. The reduced damage from stray ionizing energy creates more robust electronic protect (EP) or electronic support (ES) systems. http://www.microwavejournal.com/articles/29758-its-time-to-convert-our-ew-radio-designs
milstar: российский скоростной ацп конвейерного типа Resolution 14 Bit Sample rate 125 MSPS; Parallel CMOS and LVDS output; Single power supply 1.8V; SNR - 69.9dBFS; INL - 3.0 LSB; 180 nm CMOS process. http://www.milandr.com/ICDCS.php#/
milstar: российский скоростной ацп конвейерного типа Resolution 14 Bit Sample rate 125 MSPS; Parallel CMOS and LVDS output; Single power supply 1.8V; SNR - 69.9dBFS; INL - 3.0 LSB; 180 nm CMOS process. http://www.milandr.com/ICDCS.php#/ https://www.milandr.ru/upload/smi/konveyernyy_atsp.pdf В статье представлен первый конвейерный аналого-цифровой преобра- зователь (АЦП) 5101н В025 в разрабатываемой линейке АЦП компании «миландр». Первый быстродействующий 14 - разряд - ный АЦП в линейке преобразователей ком - пании «Миландр» К5101НВ025, выполнен- ный по технологии 0,18 мкм, достигает со- отношения сигнал/шум 64 дБ и диапазона, свободного от гармоник, 75 дБ при частоте выборки 75 Мвыб./c.
milstar: Динамический диапазон радара AN/FPQ программы Аполлон более 120 дб Антенна 8.8 метра диаметром C band 5.4-5.9 Ghz 4.8 квт средней мощности,3 мегаватта импульсной мощности промежуточная частота-30 мегагерц, полоса сигнала -1.6 мегагерц Дальность более 60 000 километров при разрешении +- 2 метра http://en.wikipedia.org/wiki/AN/FPQ-6 http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19680003409_1968003409.pdf ########## 37 metr Dish Lincoln laboratory radar The three radar intermediate-frequency inputs to the A/D board are 20 MHz bandwidth centered at 10 MHz, and are thus sampled with a 40 MHz clock http://www.ll.mit.edu/publications/journal/pdf/vol21_no1/21_1_7_Eshbaugh.pdf FIGURE 16. Single-channel radar channel processing performed by DPCS for a typical stretch waveform. --- The SPS-48E radar (Fig. 1) uses a triple conversion receiver. The system is wideband until the second intermediate frequency (IF) conversion, where the individual beams are bandpass filtered and separated. Since three beams are used in the DMTI, there are three coherent oscillator frequencies (one for each beam) in the final conversion of the receiver (final IF is about 1.5 MHz). A single analog-to-digital (A/D) converter is used for each beam. In-phase and quadrature (I/Q) data are developed based on samples that are spaced at multiples of 90° at the IF frequency. The interpolation filter develops the I/Q estimates from A/D samples http://techdigest.jhuapl.edu/TD/td1803/roul.pdf The AN/SPS-48G is a long-range, three-dimensional (3D), air search radar that is progressively being installed on CVN, LHA, LHD and LPD classes of ships, replacing the AN/SPS-48E. The program of record is to backfit the existing AN/SPS-48E population with the AN/SPS-48G variant from 2011 through 2021, and to keep this system operational through the year 2050. As of the end of 2016, the AN/SPS-48G is already installed or in the process of installation aboard CVNs 68-72, CVNs 74-76, LHDs 1-3, LHD 7, LHA 7 and LPDs 26-27. The AN/SPS-48G is used to provide full volumetric detection data for the Ship Self Defense System (SSDS) via the Cooperative Engagement Capability (CEC) or the SYS-2 tracker; Air Intercept Control; Anti-Ship Cruise Missile detection including low elevation and high diver targets; backup aircraft marshalling; and the new Hazardous Weather Detection and Display Capability. http://www.navy.mil/navydata/fact_display.asp?cid=2100&tid=1250&ct=2
полная версия страницы