Cost Equation for the SKA

Size: px
Start display at page:

Download "Cost Equation for the SKA"

Transcription

1 Cost Equation for the SKA Sander Weinreb, Caltech/JPL, Larry D Addario, NRAO and SETI Institute, ldaddario@astro.berkeley.edu July 21, General form of spreadsheet 2. Performance and cost parameters 3. Results: Cost vs antenna diameter and other parameters 4. Front-end model (Modified July 21, 2001) 5. Signal processing model Appendix A - Notes on Signal Processing Costs for SKA (Added July 21,2001) Appendix B - Cost Dependence When Field of View is Maintained (Added July 21, 2001 (Full Excel Spreadsheet is available as a file on SKA web sites: SKAcostEq1.xls) CostEQ2.doc Created on 7/21/2001 9:35 AM Page 1 of 18

2 SKA Cost Equation Spread Sheet Rows are for input or computed parameters such as cost coefficients or performance parameters Columns are for varying antenna diameters Sheets are for varying major parameters such as cryogenic temperature or era of electronics costs. Output of the program is the total cost and cost of each subsystem. The Aeff/Tsys is constrained at 20,000 by computing the number of antennas needed for each station. Parameters There are 14 performance parameters and 19 cost parameters listed on the next page Models Details are in the equations in the spreadsheet. Antenna cost is 0.1*D^3 K$. CostEQ2.doc Created on 7/21/2001 9:35 AM Page 2 of 18

3 SKA Cost Equation Input Parameters Units: K$US(2001), meters, GHz Param Array Performance Parameters Default Value Param Array Cost Parameters Default Value Ns Number of stations in array 100 Cso Fixed cost per station, land, civil, 300 M Specified Figure of Merit, M = A/Tsys 20,000 Ka Antenna cost coefficient, 0.10 B Processed total continuum bandwidth 4.0 X Antenna cost exponent 3.0 Nl Number of spectral line channels 16,000 Ccl Cooling cost per antenna 20.0 D Physical diameter of element (meters) 10.0 Cfd Average dual-polariz feed cost 2.0 Ef Aperature efficiency 0.70 Cln Average LNA + mixer cost 0.8 Tant Antenna noise temperature, Tant = Clo LO cost 3.0 Kln Lna noise coefficient dependant upon 0.40 Cifo Fixed IF cost per polarization 1.0 Tphy Physical temperature of LNA 15 Kif Dual IF cost per GHz of bandwidt 0.2 F Frequency for system temperature spe 10 a1 _Digitization coefficient 2.00 Nbn Number of frequency bands 3 e _Digitization exponent 2.00 Kch _Number of separately digitized chann 4 a2 _Digitization constant 0.50 Le _Average distance, element to station 0.50 d _Tracking coefficient (per GHz) 0.72 Nbeam_Beams per station 4 f _Tracking constant 0.10 Kproc _Processing cost coefficient (per 0.48 Kmem _Memory cost (per word) 2.10E-04 kct _Corner turner cost coefficient (p 1.00E-03 Kcor _Correlator cost coef (per baselin Kchip Price per large FPGA chip 0.14 Color code: Purple: correction of error since previous version. Blue: change to value or formula since previous version. Red: variable whose value changes across this sheet Green: variable whose value is different from previous sheet Bold type: input parameter; normal type means value is calculated. CostEQ2.doc Created on 7/21/2001 9:35 AM Page 3 of 18

4 SKA Cost vs Antenna Diameter for 3 Cooling Temperatures Aeff/Tsys = 20,000, Aeff=360,000, Tsys=18K, BW=4GHz, Antenna Cost = 0.1D^3 K$, 2001 Electronics Cost = $54K per Element $2,500,000 25,000 Cost, $K (US2001) $2,000,000 $1,500,000 $1,000,000 $500,000 20,000 15,000 10,000 5,000 Number of Antennas $ Element Diameter, Meters 15K Cryogenics 60K Cryogenics 180K Cryogenics # Antennas, 15K # Antennas, 60K # Antennas, 180K CostEQ2.doc Created on 7/21/2001 9:35 AM Page 4 of 18

5 SKA Cost Breakdown by Subsystem vs Antenna Diameter Aeff/Tsys = 20,000, Aeff=360,000, Tsys=18K, BW=4GHz, 15K Cryogenics Antenna Cost = 0.1D^3 K$, 2001 Electronics Cost = $54K per Element 2,000,000 Fixed Costs Civil Station 1,500,000 Signal Transmission Electronics Central Processing Antenna Total Cost, $K 1,000,000 Antenna 500,000 Electronics 0 Signal Transmission Central Processing Fixed Costs Civil Station Antenna Diameter, Meters CostEQ2.doc Created on 7/21/2001 9:35 AM Page 5 of 18

6 SKA Cost Breakdown by Subsystem vs Antenna Diameter Aeff/Tsys = 20,000, Aeff=360,000, Tsys=18K, BW=4GHz, 15K Cryogenics Antenna Cost = 0.1D^3 K$, 2010Electronics Cost = $15K per Element 2,000,000 Fixed Costs Civil Station 1,500,000 Signal Transmission Electronics Central Processing Antenna Total Cost, $K 1,000, ,000 Antenna 0 Electronics Signal Transmission Central Processing Fixed Costs Civil Station Antenna Diameter, Meters CostEQ2.doc Created on 7/21/2001 9:35 AM Page 6 of 18

7 SKA Cost vs Antenna Diameter Compares Current and Projected (2010) Electronics Costs All for 15K cryogenics, 4 GHz BW, 100 Stations, A/T = 20,000 Antenna Cost = 0.1D^3 $K $2,000,000 $100,000 Total Cost, $K (US2001) $1,500,000 $1,000,000 $500,000 $10,000 $1,000 $100 Antenna Unit Cost, $K $0 $ Element Diameter, Meters $54K Electronics $15K Electronics Antenna Cost CostEQ2.doc Created on 7/21/2001 9:35 AM Page 7 of 18

8 SKA Cost vs Processed Bandwidth for Two Antenna Diameters All for 15K Cryogenics $2,000,000 Cost, $K (US2001) $1,500,000 $1,000,000 $500,000 $ Processed Bandwidth, GHz 6551 x 10m Antenna 2912 x 15m Antenna CostEQ2.doc Created on 7/21/2001 9:35 AM Page 8 of 18

9 SKA Cost vs Number of Stations for Two Processed Bandwidths All for 15K Cryogenics, 10m antennas. A/T = 20,000 $2,500, Cost, $K (US2001) $2,000,000 $1,500,000 $1,000,000 $500, Antennas per Station $ Number of Stations 0 1 GHz Bandwidth 4 GHz Bandwidth Antennas per Station CostEQ2.doc Created on 7/21/2001 9:35 AM Page 9 of 18

10 Front-End Cost and Performance Model Cooler Type Physical Temp Cooler Cost, K$ 10 GHz 50 GHz F GHz None *F+16 Peltier *F+10 Pulse or Klemenko *F+4 Gifford-McMann *F+1 Total Front-End Cost, FE$ FE$ = Cooler$ + 3(Bands)*2(Polarizations)* LNA$ LNA$ = $0.8K*(Fmax/10)^.3 i.e. FE$ = $11.9K for Pulse Cooling with Fmax = 20 GHz Total Electronics Cost, EL$, per Antenna EL$ = FE$+ 3*Feed$+LO$+IF$+Fixed$ Feed$ = $1K, LO$ = $5K, IF$ = 2*( *BW), Fixed$ = $1K i.e. EL$ = * * = $29.9K Total System Noise, Tsys Tsys = Tlna + [10 + 4* (F/10)] F>1 GHz, Not Including Antenna Losses CostEQ2.doc Created on 7/21/2001 9:35 AM Page 10 of 18

11 Signal Processing Cost Model EACH ANTENNA 1. Digitization C d = [k d (B/N ch ) e + c d ] N ch (exponent e is approximately 2) Cost per channel increases faster than bandwidth => more channels Presently expensive -- total for array is more than the correlator! 2. Tracking Primarily delay and phase tracking of sources Per antenna per beam -- multi beaming is probably needed to mitigate small FOV of station beam. All-digital implementation assumed EACH STATION 1. Summation to form phased-array, per beam Multi-beaming probably needed due to small station beam 2. Transmission to central processing facility available bandwidth shared among beams. CENTRAL PROCESSING (CORRELATION) 1. Per antenna section filter bank for L spectral channels cost grows as B log(l) if minimum-memory FFTs are used 2. Interconnection section can dominate cost for large N natural architecture grows as B N 3 clever architecture grows as B N log(n) 3. Cross correlation section grows as B L N 2 (note L dependence, even for FX architecture CostEQ2.doc Created on 7/21/2001 9:35 AM Page 11 of 18

12 Appendix A - Notes on Signal Processing Costs for SKA L. D Addario 2001-Jul-04; revised Jul-07, Jul-12 ANTENNA-ELEMENT SIGNAL PROCESSING COSTS 1. Digitization In present technology, this is an expensive element of the overall signal processing. We estimate about 10k$ per element for 4 GHz bandwidth, which becomes 45M$ for 4500 elements (100 stations of 45 elements each). This is about equal to the cost of the central signal processing (correlator) and 3% of the total cost of the array, based on our current estimates for one reasonable case. The cost of digitization rises faster than the bandwidth of each separately-digitized channel, and appears to go approximately as the square over the range 10 to 1000 MHz for presently available commercial devices. Therefore, it is advantageous to break up the total bandwidth prior to digitization into as many channels as practical, provided that performance does not suffer. In our cost estimates, we have usually used at least 4 channels (to allow for 2 polarizations at 2 independent sky frequencies) and kept the channel bandwidth at 1 GHz or less. 2. Tracking For each signal channel of each antenna element, it is necessary to provide for time-varying adjustment of the delay and phase so as to track a desired direction on the sky. In our cost estimates, we assume that this is accomplished digitally. The cost of computation and control for this purpose appears to be negligible (<1% of the total in all cases studied). The elements are clustered into N s stations and the signals are combined so that each station is a phased array. This minimizes the cost of the central processing, which varies as approximately N s 2 log N s. However, it also reduces the field of view on the sky from the size of the element beam to that of the phase array (station) beam. To mitigate this effect, the phase and delay tracking can be duplicated at each element so as to produce several simultaneous station beams. If the bandwidth per beam is fixed, then the total bandwidth increases along with the cost of signal transmission and central processing. Whether the increased sky coverage is worth the extra cost depends on the scientific goals. 3. Signal Transmission. The bandwidth B needs to be transmitted several times: Element to element's center; element center to control center; correlator F section to "corner turner"; and corner turner to correlator X section. In all cases we use the same coefficients for transmitter/receiver cost and installed fiber cost. Average fiber length is taken to be 0.5 km from elements within a station, 100 km from each station to the center, and zero (<<1 km) within the correlator. Sharing of parts of the transmission routes among stations could reduce this, but the number of stations has to exceed 350 before the transmission cost exceeds 5% of the total; at 100 stations it is around 1.5% 12

13 CORRELATOR COST DEPENDENCIES We assume that it's important to cross-correlate the signals from all stations, over the full bandwidth of each. The cost of this is proportional to the total bandwidth, regardless of whether that bandwidth represents a single channel on the sky or several independent channels (beams). There is a weak dependence on the number of channels, but this is insignificant if that number is small compared with the final number of subchannels into which the total bandwidth will be analyzed. The central processing separates naturally into a per-station part and a per-baseline part whose costs are proportional to B N and BN 2, respectively. Each cost term has additional factors that depend on the architecture and on the spectral resolution, as well as a coefficient that depends on the implementation technology. For the large-n case, the baseline term dominates the station term. For sufficiently large N, the baseline electronics will not fit into a single chip, nor a single circuit board or small set of boards, and therefore it must be partitioned into some number K of separate pacakages, each of which accomplishes 1/K of the processing. There are several ways to do the partitioning [1], and this determines the top-level architecture of the system. We concentrate here on the so-called ``FX'' architecture, where the bandwidth B from each antenna is partitioned by frequency into segments of bandwidth B/K and each element of the baseline part processes one such segment for all antennas. The available technology determines how much processing can be fit into one package, and thus how many packages K are required and how small each bandwidth segment B/K must be. We set K = t B N 2 /2, where t is a technology-dependent coefficient that gets smaller as circuitry becomes denser and/or faster. The segmentation is done in the station part, which we assume to consist of one package per antenna. Therefore, C=N K interconnections are needed between different packages in order to get the signals from the station part to the baseline part. Substituting, we see that C = t B N 3 /2. Because of this N 3 dependence, for sufficiently large N the cost can easily be dominated by the interconnections, even if t is small. This situation can be substantially improved by imposing an additional device between the station part and the baseline part whose purpose is to re-order the data; it is called a corner turner'' [2]. Then the number of interconnections is reduced from N K to N+K, where N are needed from the station part to the corner turner, and K from the corner turner to the baseline part. Unfortunately, the corner turner itself is potentially a very large machine whose obvious implementation involves N K internal interconnections and data storage proportional to NK, hence its size grows as N 3. But a much more efficient implementation exists [2][3] that reduces it to a size proportional to N 2 log N, in a manner similar to the FFT's reduction of the number of operations in a discrete Fourier transform of size L from L 2 to L log L. In addition, a new architecture for the baseline part [4] allows it to be organized so that not all K segments need to be connected directly to the corner turner. This allows the corner turner size to be grow only as N log N. 13

14 It may be desired to analyze the bandwidth into many more than K subchannels, say L>>K, for spectroscopy. Then each of the K segments processes L/K subchannels. It is often suggested that the FX architecture allows L to be increased arbitrarily at negligible additional cost. In the station part, filter banks based on the FFT can be used so that the cost increases only as log(l); then, unless L is extreme, the cost of the station part remains small compared to the interconnections and baseline part. The interconnection cost is independent of L for a given B and N. In the baseline part, the computation rate (for cross multiplication and accumulation) is also independent of L, since the data rate of each subchannel goes as 1/L. However, L separate accumulators are needed per baseline, so storage requirements increase as L. Depending on the cost of storage compared with computation, this may be important. Usually it is not possible to take advantage of the low cost per bit of large RAM because efficient use of the computational hardware requires many multiply-add operations to be done in parallel at high speed, leading to many short memories rather than few large ones, and requiring them to be integrated on the same chip with the computational elements. This can cause the accumulator storage to be the dominant cost, so that the overall cost is nearly proportional to L. The dependence is then nearly the same as in the XF architecture, where the costs of both computation and storage are proportional to L. Indeed, this would make the absolute cost of the XF architecture smaller because the individual multipliers and accumulators are considerably less complex. All of this is summarized in the chart on the last page of this document. REFERENCES [1] L. D'Addario, ``Correlators: General design considerations.'' ATA Memo No. 24, 2001-Mar- 13. [2] L. Urry, ``A corner turner architecture.'' ATA Memo No. 14, 2000-Nov-17. [3] L. D'Addario, ``Generalization of the memoryless corner turner to the non-square case.'' ATA Memo No. 22, 2001-Mar-16. (All of the above are temporarily available via [4] L. Urry, this conference. JUSTIFICATIONS OF COST COEFFICIENTS a1,e a2 digitizer cost parameters, fit to prices of commercial ADC chips digitizer cost constant: guess at costs of downconversion, anti-alias filtering, circuit board, $500. This can't be too far off. 14

15 d,e tracking coef, const: One XV300E should do 2 channels at 100 MHz, costs $144; based on preliminary design for ATA. So 722$/GHz. Constant term, $100, is just a guess to cover packaging and power. Kproc Xilinx XV300E, 3072slices/1929perFFT1024inPlace -> 1.5 radix4 butterflys=3 radix2, $144/3= $48 MHz -> 480 $/butterfly/ghz Kmem Xilinx XV300E: 144$/(2mem)*(32b/word)/(131072b block RAM)/100MHz = k$/word/ghz where half of chip cost is allocated to memory Kcorr Xilinx XV300E: 3072slices/51perMAC = 60 MACs, 100 MHz each, $144/chip 144$/(2comp)/60MAC/100MHz =.0120 k$/mac/ghz where half of chip cost is allocated to computation Kinfr Total chip cost/$144 -> chip count. 16 chips/board, $160 for board, rack, power. Check: Total central processing for B=0.2, Ns=350, Nl=2048 is 5.6M$ -- ATA case B=16, Ns= 64, Nl=8192 is 58.7M$ -- ALMA case The ATA result is about 4x higher than the independent cost estimate (hardware only) made for the proposal. That proposal used future FPGA prices that are indeed about 4x lower per unit performance than the present prices used here; with this correction, the agreement is excellent. The ALMA result is also about 4x higher than its presently projected actual cost, but its architecture is quite different than assumed here. 15

16 Correlator Cost Model: FX Architecture Filtering Interconnection Correlation Computation α B N log L β B N log N αb N 2 / 2 Storage χ B N L (control only) δ B L N 2 / 2 Where: B is the total processed bandwidth N is the number of stations L is the number of spectral channels Greek symbols are cost coefficients, dependent on technology. Correlation does not include cross-polarized products. 16

17 Appendix B - Cost Dependence When Field of View is Maintained LRD, 2001-Jul-12 The analyses presented at the conference considered the cost dependence on various parameters, including antenna size, front end cooling temperature, and electronics costs, while keeping the number of stations constant. This required varying the number of antennas per station in order to have enough collecting area to meet the A e /T sys requirement. Formally, the number of phased array beams was kept constant at 4. But the cost would be nearly unaffected if the number of beams were larger, provided that the processed bandwidth B is shared among all the beams. Here we consider relaxing that restriction. If the antennas are clustered into stations at N s each, and if they are close-packed within a station, then each station beam is about 1/ N s the size of the primary antenna (element) beam in solid angle. So if we still want to image the whole primary beam, we will need N s phased-array beams. I put this into the spreadsheet to look at the dependence on N s. Rather than sharing B among the beams, I fixed the bandwidth/beam b, making the total bandwidth B = 2 b N s (2 for polarization). Naturally this can lead to very large B and high cost for transmission and central processing, so I considered values of b only from 0.1 to 0.5 GHz. The result is given in the accompanying figure. It can be seen that a shallow minimum in the total cost occurs at N s = 200. Above this, the slow increase in total cost is caused by rapid increase in central processing (correlation) cost, which reaches 33% of the total at 400 antennas per station and 500 MHz bandwidth. Below the minimum, the correlator cost continues to decrease (in spite of the larger number of beams required), but the total cost is strongly affected by the processing required at each station to form many beams, including both digitization (assumed separate for each beam) and tracking. 17

18 SKA Cost vs Number of Stations for Beam-filled case, 0.1 and 0.5 GHz/beam All for 15K Cryogenics, 12m antennas, 4550 total, A/T = 20,000 $3,500, Cost, $K (US2001) $3,000,000 $2,500,000 $2,000,000 $1,500,000 $1,000,000 $500, Ants/stn == beams $ Number of Stations 0.1 GHz/beam 0.5 GHz/beam Antennas per Station 18

19 SKA Cost Estimates July 21, 2001 File: skacosteq1.xls Units K$US(2001), meters, GHz Antenna Diameter, Meters 10,0 Parameter Array Parmeters Value C Total array cost, fixed costs, elements, processing Cat Total antenna cost, Ne*Ns*Ca Celt Total electronics cost, Ne*Ns*Ce Cpt Total processing cost includiing station sum Ctr Signal transmission costs Csot Total civil costs at stations, Ns*Cso Co Fixed cost per array - design + central civil Cs Total station cost, elements + combining electronics Cm Total element cost, antenna + receivers + processing 160 Cp Central Signal processing costs Ns Number of stations in array 100 Ne Number of elements per station 69 Deq Equivalent single-antenna diameter of station 83 N Total number of elements, N = Ns*Ne A Effective area of array, A=N*Ae M Specified Figure of Merit, M = A/Tsys M Computed Figure of Merit, M = A/Tsys Tsys System noise temperature at frequency, F 19 B Processed total continuum bandwidth 4,0 Nl Number of spectral line channels R Antenna/Electronics ratio, Ca/Ce 1,68 Ropt Minimum cost ratio, Ropt = 1 / (X / 2-1) 2,0 Antenna Parameters D Physical diameter of element (meters) 10,0 Ap Physical area of element, Ap = 0.785*D^2 79 Ef Aperature efficiency 0,70 Ae Effective area of element, Ae=Ap*Ef 55 Tant Antenna noise temperature, Tant = 10 +4*(F/10) 14 Cs Cost per station, Cs =Cso + Ne*(Ca+Ce) Cso Fixed cost per station, land, civil, bunker, cables 300 Ca Cost per antenna, Ca = Ka*D^X 100 Ka Antenna cost coefficient, 0,10 X Antenna cost exponent 3,0 Csm Antenna cost per square meter physical area,k$/m^2 1,274 Receiver Parameters Tsys Tsys = Tln +Tant 19 Tln Tln = Kln*F+K1 5 Kln Lna noise coefficient dependant upon cooler 0,40 Tphy Physical temperature of LNA 15 F Frequency for system temperature specification 10 Ce Electronics cost per antenna, Goal 59,7 Ce Ce = Ccl +Nbn*(Cfd +2* Cln) +Clo+Cif +Cpe 59,7 Ccl Cooling cost per antenna 20,0 Nbn Number of frequency bands 3 Cfd Average dual-polariz feed cost 2,0 Cln Average LNA + mixer cost 0,8 Clo LO cost 3,0 Cif Dual IF cost, Cif =2* (Cifo+Kif*B/2) 3,6 Cifo Fixed IF cost per polarization 1,0 Kif Dual IF cost per GHz of bandwidth 0,2 Signal Transmission Parameters Ctr Cost of transmission, Ctr =Ns*(Clk + Ls*Cfb) Clk Fiber transceiver cost, Clk = Klk*B 8,00 Klk Tranceiver cost per GHz 2,00 Ls Average station to control center distance, km 100 Cfb Fiber installed cost, per fiber per km 2 Signal Processing Parameters (FX architecture assumed for correlator)(bandwidth shared among beams) Cdig Digitization Cdig = (a1*(b/kch)^e+a2) * Kch 10,00 a1 _Digitization coefficient 2,00 e _Digitization exponent 2,00 a2 _Digitization constant 0,50 Kch _Number of separately digitized channels 4 Ctre Transmission, el to stn ctr (Klk*B + Le*Cfb) 9,00 Le _Average distance, element to station center 0,50 Ctrack Tracking, elements Ctrack = (d*b + f*nbeams) 3,28 d _Tracking coefficient (per GHz) 0,72 f _Tracking constant 0,10 Nbeams _Beams per station 4 Cpe Element processing cost = Cdig+Ctre+Ctrack 22,28 Csum Beam summation cost, c*b*(ne-1) 2,726 c _Summation coefficient 0,010 Cps Station processing cost = Cpe*Ne + Csum Cfilt Filtering cost Cfilt = (Kproc+Kmem)*log2(Nl)*B*Ns 2.682,4 Kproc _Processing cost coefficient (per GHz) 0,48 Kmem _Memory cost (per word) 1,76E-04 Cci Correlator interconnections Cci=Cct + 2*Klk*Ns*B Cct Cost of corner turner Cct = kct*b*ns *log2(ns) 2,7 kct _Corner turner cost coefficient (per connection) 1,00E-03 Ccor Cross correlation cost = (Kcor+Kmem*Nl) *B*Ns*Ns/ Kcor _Correlator cost coef (per baseline-ghz) 0,012 Kchip Price per large FPGA chip 0,144 Nchip Large chip count = (Cfilt+Cci+Ccor)/Kchip Cinfr Correlator infrastructure, $10/large chip Cpc Central processing cost = Cfilt+Cci+Ccor +Cinfr Cp Total signal processing cost = Cps*Ns+Cpc

20 Color code: Purple: correction of error since previous version. Blue: change to value or formula since previous version. Red: variable whose value changes across this sheet Green: variable whose value is different from previous sheet Bold type: input parameter; normal type means value is calculated.

21 SKA Cost Equation Input Parameters Units: K$US(2001), meters, GHz Param Array Performance Parameters Default Value Param Array Cost Parameters Default Value Ns Number of stations in array 100 Cso Fixed cost per station, land, civil 300 M Specified Figure of Merit, M = A/Tsys Ka Antenna cost coefficient, 0,10 B Processed total continuum bandwidth 4,0 X Antenna cost exponent 3,0 Nl Number of spectral line channels Ccl Cooling cost per antenna 20,0 D Physical diameter of element (meters 10,0 Cfd Average dual-polariz feed cost 2,0 Ef Aperature efficiency 0,70 Cln Average LNA + mixer cost 0,8 Tant Antenna noise temperature, Tant = Clo LO cost 3,0 Kln Lna noise coefficient dependant upon 0,40 Cifo Fixed IF cost per polarization 1,0 Tphy Physical temperature of LNA 15 Kif Dual IF cost per GHz of bandwidt 0,2 F Frequency for system temperature sp 10 a1 _Digitization coefficient 2,00 Nbn Number of frequency bands 3 e _Digitization exponent 2,00 Kch _Number of separately digitized chan 4 a2 _Digitization constant 0,50 Le _Average distance, element to station 0,50 d _Tracking coefficient (per GHz) 0,72 Nbeam_Beams per station 4 f _Tracking constant 0,10 Kproc _Processing cost coefficient (pe 0,48 Kmem _Memory cost (per word) 2,10E-04 kct _Corner turner cost coefficient (p 1,00E-03 Kcor _Correlator cost coef (per baseli 0,024 Kchip Price per large FPGA chip 0,14 Color code: Klk Fiber Tranceiver per GHz 2 Purple: correction of error since previous version. Blue: change to value or formula since previous version. Red: variable whose value changes across this sheet Green: variable whose value is different from previous sheet Bold type: input parameter; normal type means value is calculated.

22 SKA Cost Estimates July 21, 2001 File: skacosteq1.xls Units K$US(2001), meters, GHz Antenna Diameter, Meters 5,0 8,0 10,0 12,0 15,0 20,0 30,0 Parameter Array Parmeters Value Value Value Value Value Value Value C Total array cost, fixed costs, elements, processing Cat Total antenna cost, Ne*Ns*Ca Celt Total electronics cost, Ne*Ns*Ce Cpt Total processing cost includiing station sum Ctr Signal transmission costs Csot Total civil costs at stations, Ns*Cso Co Fixed cost per array - design + central civil Cs Total station cost, elements + combining electronics Cm Total element cost, antenna + receivers + processing Cp Central Signal processing costs Ns Number of stations in array Ne Number of elements per station Deq Equivalent single-antenna diameter of station N Total number of elements, N = Ns*Ne A Effective area of array, A=N*Ae M Specified Figure of Merit, M = A/Tsys M Computed Figure of Merit, M = A/Tsys Tsys System noise temperature at frequency, F B Processed total continuum bandwidth 4,0 4,0 4,0 4,0 4,0 4,0 4,0 Nl Number of spectral line channels R Antenna/Electronics ratio, Ca/Ce 0,21 0,86 1,68 2,90 5,66 13,40 45,24 Ropt Minimum cost ratio, Ropt = 1 / (X / 2-1) 2,0 2,0 2,0 2,0 2,0 2,0 2,0 Antenna Parameters D Physical diameter of element (meters) 5,0 8,0 10,0 12,0 15,0 20,0 30,0 Ap Physical area of element, Ap = 0.785*D^ Ef Aperature efficiency 0,70 0,70 0,70 0,70 0,70 0,70 0,70 Ae Effective area of element, Ae=Ap*Ef Tant Antenna noise temperature, Tant = 10 +4*(F/10) Cs Cost per station, Cs =Cso + Ne*(Ca+Ce) Cso Fixed cost per station, land, civil, bunker, cables Ca Cost per antenna, Ca = Ka*D^X Ka Antenna cost coefficient, 0,10 0,10 0,10 0,10 0,10 0,10 0,10 X Antenna cost exponent 3,0 3,0 3,0 3,0 3,0 3,0 3,0 Csm Antenna cost per square meter physical area,k$/m^2 0,637 1,019 1,274 1,529 1,911 2,548 3,822 Receiver Parameters Tsys Tsys = Tln +Tant Tln Tln = Kln*F Kln Lna noise coefficient dependant upon cooler 0,40 0,40 0,40 0,40 0,40 0,40 0,40 Tphy Physical temperature of LNA F Frequency for system temperature specification Ce Electronics cost per antenna, Goal 59,7 59,7 59,7 59,7 59,7 59,7 59,7 Ce Ce = Ccl +Nbn*(Cfd +2* Cln) +Clo+Cif +Cpe 59,7 59,7 59,7 59,7 59,7 59,7 59,7 Ccl Cooling cost per antenna 20,0 20,0 20,0 20,0 20,0 20,0 20,0 Nbn Number of frequency bands Cfd Average dual-polariz feed cost 2,0 2,0 2,0 2,0 2,0 2,0 2,0 Cln Average LNA + mixer cost 0,8 0,8 0,8 0,8 0,8 0,8 0,8 Clo LO cost 3,0 3,0 3,0 3,0 3,0 3,0 3,0 Cif Dual IF cost, Cif =2* (Cifo+Kif*B/2) 3,6 3,6 3,6 3,6 3,6 3,6 3,6 Cifo Fixed IF cost per polarization 1,0 1,0 1,0 1,0 1,0 1,0 1,0 Kif Dual IF cost per GHz of bandwidth 0,2 0,2 0,2 0,2 0,2 0,2 0,2 Signal Transmission Parameters Ctr Cost of transmission, Ctr =2*Ns*(Ne*Clk + Ls*Cfb) Clk Fiber transceiver cost, Clk = Klk*B 8,00 8,00 8,00 8,00 8,00 8,00 8,00 Klk Tranceiver cost per GHz 2,00 2,00 2,00 2,00 2,00 2,00 2,00 Ls Average station to control center distance, km Cfb Fiber installed cost, per fiber per km Signal Processing Parameters (FX architecture assumed for correlator)(bandwidth shared among beams) Cdig Digitization Cdig = (a1*(b/kch)^e+a2) * Kch 10,00 10,00 10,00 10,00 10,00 10,00 10,00 a1 _Digitization coefficient 2,00 2,00 2,00 2,00 2,00 2,00 2,00 e _Digitization exponent 2,00 2,00 2,00 2,00 2,00 2,00 2,00 a2 _Digitization constant 0,50 0,50 0,50 0,50 0,50 0,50 0,50 Kch _Number of separately digitized channels Ctre Transmission, el to stn ctr (Klk*B + Le*Cfb) 9,00 9,00 9,00 9,00 9,00 9,00 9,00 Le _Average distance, element to station center 0,50 0,50 0,50 0,50 0,50 0,50 0,50 Ctrack Tracking, elements Ctrack = (d*b + f*nbeams) 3,28 3,28 3,28 3,28 3,28 3,28 3,28 d _Tracking coefficient (per GHz) 0,72 0,72 0,72 0,72 0,72 0,72 0,72 f _Tracking constant 0,10 0,10 0,10 0,10 0,10 0,10 0,10 Nbeams _Beams per station Cpe Element processing cost = Cdig+Ctre+Ctrack 22,28 22,28 22,28 22,28 22,28 22,28 22,28 Csum Beam summation cost, c*b*(ne-1) 11,025 4,282 2,726 1,881 1,189 0,652 0,267 c _Summation coefficient 0,010 0,010 0,010 0,010 0,010 0,010 0,010 Cps Station processing cost = Cpe*Ne + Csum Cfilt Filtering cost Cfilt = (Kproc+Kmem)*log2(Nl)*B*Ns 2.682, , , , , , ,4 Kproc _Processing cost coefficient (per GHz) 0,48 0,48 0,48 0,48 0,48 0,48 0,48 Kmem _Memory cost (per word) 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 Cci Correlator interconnections Cci=Cct + 2*Klk*Ns*B Cct Cost of corner turner Cct = kct*b*ns ^2*log2(Ns) 2,7 2,7 2,7 2,7 2,7 2,7 2,7 kct _Corner turner cost coefficient (per connection) 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 Ccor Cross correlation cost = (Kcor+Kmem*Nl) *B*Ns*Ns/ Kcor _Correlator cost coef (per baseline-ghz) 0,012 0,012 0,012 0,012 0,012 0,012 0,012 Kchip Price per large FPGA chip 0,144 0,144 0,144 0,144 0,144 0,144 0,144 Nchip Large chip count = (Cfilt+Cci+Ccor)/Kchip Cinfr Correlator infrastructure, $10/large chip Cpc Central processing cost = Cfilt+Cci+Ccor +Cinfr Cp Total signal processing cost = Cps*Ns+Cpc

23 Color code: Purple: correction of error since previous version. Blue: change to value or formula since previous version. Red: variable whose value changes across this sheet Green: variable whose value is different from previous sheet Bold type: input parameter; normal type means value is calculated.

24 SKA Cost Breakdown by Subsystem vs Antenna Diameter Aeff/Tsys = 20,000, Aeff=360,000, Tsys=18K, BW=4GHz, 15K Cryogenics Antenna Cost = 0.1D^3 K$, 2001 Electronics Cost = $54K per Element Fixed Costs Signal Transmission Electronics Civil Station Central Processing Antenna Total Cost, $K Antenna Electronics 0 Signal Transmission Central Processing Civil Station Fixed Costs Antenna Diameter, Meters

25 SKA Cost Estimates July 21, 2001 File: skacosteq1.xls Units K$US(2001), meters, GHz Antenna Diameter, Meters 5,0 8,0 10,0 12,0 15,0 20,0 30,0 Parameter Array Parmeters Value Value Value Value Value Value Value C Total array cost, fixed costs, elements, processing Cat Total antenna cost, Ne*Ns*Ca Celt Total electronics cost, Ne*Ns*Ce Cpt Total processing cost includiing station sum Ctr Signal transmission costs Csot Total civil costs at stations, Ns*Cso Co Fixed cost per array - design + central civil Cs Total station cost, elements + combining electronics Cm Total element cost, antenna + receivers + processing Cp Central Signal processing costs Ns Number of stations in array Ne Number of elements per station Deq Equivalent single-antenna diameter of station N Total number of elements, N = Ns*Ne A Effective area of array, A=N*Ae M Specified Figure of Merit, M = A/Tsys M Computed Figure of Merit, M = A/Tsys Tsys System noise temperature at frequency, F B Processed total continuum bandwidth 4,0 4,0 4,0 4,0 4,0 4,0 4,0 Nl Number of spectral line channels R Antenna/Electronics ratio, Ca/Ce 0,27 1,12 2,19 3,78 7,39 17,51 59,11 Ropt Minimum cost ratio, Ropt = 1 / (X / 2-1) 2,0 2,0 2,0 2,0 2,0 2,0 2,0 Antenna Parameters D Physical diameter of element (meters) 5,0 8,0 10,0 12,0 15,0 20,0 30,0 Ap Physical area of element, Ap = 0.785*D^ Ef Aperature efficiency 0,70 0,70 0,70 0,70 0,70 0,70 0,70 Ae Effective area of element, Ae=Ap*Ef Tant Antenna noise temperature, Tant = 10 +4*(F/10) Cs Cost per station, Cs =Cso + Ne*(Ca+Ce) Cso Fixed cost per station, land, civil, bunker, cables Ca Cost per antenna, Ca = Ka*D^X Ka Antenna cost coefficient, 0,10 0,10 0,10 0,10 0,10 0,10 0,10 X Antenna cost exponent 3,0 3,0 3,0 3,0 3,0 3,0 3,0 Csm Antenna cost per square meter physical area,k$/m^2 0,637 1,019 1,274 1,529 1,911 2,548 3,822 Receiver Parameters Tsys Tsys = Tln +Tant Tln Tln = Kln*F Kln Lna noise coefficient dependant upon cooler 1,00 1,00 1,00 1,00 1,00 1,00 1,00 Tphy Physical temperature of LNA F Frequency for system temperature specification Ce Electronics cost per antenna 45,7 45,7 45,7 45,7 45,7 45,7 45,7 Ce Ce = Ccl +Nbn*(Cfd +2* Cln) +Clo+Cif +Cpe 45,7 45,7 45,7 45,7 45,7 45,7 45,7 Ccl Cooling cost per antenna 6,0 6,0 6,0 6,0 6,0 6,0 6,0 Nbn Number of frequency bands Cfd Average dual-polariz feed cost 2,0 2,0 2,0 2,0 2,0 2,0 2,0 Cln Average LNA + mixer cost 0,8 0,8 0,8 0,8 0,8 0,8 0,8 Clo LO cost 3,0 3,0 3,0 3,0 3,0 3,0 3,0 Cif Dual IF cost, Cif =2* (Cifo+Kif*B/2) 3,6 3,6 3,6 3,6 3,6 3,6 3,6 Cifo Fixed IF cost per polarization 1,0 1,0 1,0 1,0 1,0 1,0 1,0 Kif Dual IF cost per GHz of bandwidth 0,2 0,2 0,2 0,2 0,2 0,2 0,2 Signal Transmission Parameters Ctr Cost of transmission, Ctr =2*Ns*(Ne*Clk + Ls*Cfb) Clk Fiber transceiver cost, Clk = Klk*B 8,00 8,00 8,00 8,00 8,00 8,00 8,00 Klk Tranceiver cost per GHz 2,00 2,00 2,00 2,00 2,00 2,00 2,00 Ls Average station to control center distance, km Cfb Fiber installed cost, per fiber per km Signal Processing Parameters (FX architecture assumed for correlator)(bandwidth shared among beams) Cdig Digitization Cdig = (a1*(b/kch)^e+a2) * Kch 10,00 10,00 10,00 10,00 10,00 10,00 10,00 a1 _Digitization coefficient 2,00 2,00 2,00 2,00 2,00 2,00 2,00 e _Digitization exponent 2,00 2,00 2,00 2,00 2,00 2,00 2,00 a2 _Digitization constant 0,50 0,50 0,50 0,50 0,50 0,50 0,50 Kch _Number of separately digitized channels Ctre Transmission, el to stn ctr (Klk*B + Le*Cfb) 9,00 9,00 9,00 9,00 9,00 9,00 9,00 Le _Average distance, element to station center 0,50 0,50 0,50 0,50 0,50 0,50 0,50 Ctrack Tracking, elements Ctrack = (d*b + f*nbeams) 3,28 3,28 3,28 3,28 3,28 3,28 3,28 d _Tracking coefficient (per GHz) 0,72 0,72 0,72 0,72 0,72 0,72 0,72 f _Tracking constant 0,10 0,10 0,10 0,10 0,10 0,10 0,10 Nbeams _Beams per station Cpe Element processing cost = Cdig+Ctre+Ctrack 22,28 22,28 22,28 22,28 22,28 22,28 22,28 Csum Beam summation cost, c*b*(ne-1) 16,266 6,329 4,036 2,791 1,772 0,979 0,413 c _Summation coefficient 0,010 0,010 0,010 0,010 0,010 0,010 0,010 Cps Station processing cost = Cpe*Ne + Csum Cfilt Filtering cost Cfilt = (Kproc+Kmem)*log2(Nl)*B*Ns 2.682, , , , , , ,4 Kproc _Processing cost coefficient (per GHz) 0,48 0,48 0,48 0,48 0,48 0,48 0,48 Kmem _Memory cost (per word) 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 Cci Correlator interconnections Cci=Cct + 2*Klk*Ns*B Cct Cost of corner turner Cct = kct*b*ns ^2*log2(Ns) 2,7 2,7 2,7 2,7 2,7 2,7 2,7 kct _Corner turner cost coefficient (per connection) 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 Ccor Cross correlation cost = (Kcor+Kmem*Nl) *B*Ns*Ns/ Kcor _Correlator cost coef (per baseline-ghz) 0,012 0,012 0,012 0,012 0,012 0,012 0,012 Kchip Price per large FPGA chip 0,144 0,144 0,144 0,144 0,144 0,144 0,144 Nchip Large chip count = (Cfilt+Cci+Ccor)/Kchip Cinfr Correlator infrastructure, $10/large chip Cpc Central processing cost = Cfilt+Cci+Ccor +Cinfr Cp Total signal processing cost = Cps*Ns+Cpc

26 Color code: Purple: correction of error since previous version. Blue: change to value or formula since previous version. Red: variable whose value changes across this sheet Green: variable whose value is different from previous sheet Bold type: input parameter; normal type means value is calculated.

27 SKA Cost Estimates July 21, 2001 File: skacosteq1.xls Units K$US(2001), meters, GHz Antenna Diameter, Meters 5,0 8,0 10,0 12,0 15,0 20,0 30,0 Parameter Array Parmeters Value Value Value Value Value Value Value C Total array cost, fixed costs, elements, processing Cat Total antenna cost, Ne*Ns*Ca Celt Total electronics cost, Ne*Ns*Ce Cpt Total processing cost includiing station sum Ctr Signal transmission costs Csot Total civil costs at stations, Ns*Cso Co Fixed cost per array - design + central civil Cs Total station cost, elements + combining electronics Cm Total element cost, antenna + receivers + processing Cp Central Signal processing costs Ns Number of stations in array Ne Number of elements per station Deq Equivalent single-antenna diameter of station N Total number of elements, N = Ns*Ne A Effective area of array, A=N*Ae M Specified Figure of Merit, M = A/Tsys M Computed Figure of Merit, M = A/Tsys Tsys System noise temperature at frequency, F B Processed total continuum bandwidth 4,0 4,0 4,0 4,0 4,0 4,0 4,0 Nl Number of spectral line channels R Antenna/Electronics ratio, Ca/Ce 0,29 1,20 2,34 4,05 7,91 18,74 63,26 Ropt Minimum cost ratio, Ropt = 1 / (X / 2-1) 2,0 2,0 2,0 2,0 2,0 2,0 2,0 Antenna Parameters D Physical diameter of element (meters) 5,0 8,0 10,0 12,0 15,0 20,0 30,0 Ap Physical area of element, Ap = 0.785*D^ Ef Aperature efficiency 0,70 0,70 0,70 0,70 0,70 0,70 0,70 Ae Effective area of element, Ae=Ap*Ef Tant Antenna noise temperature, Tant = 10 +4*(F/10) Cs Cost per station, Cs =Cso + Ne*(Ca+Ce) Cso Fixed cost per station, land, civil, bunker, cables Ca Cost per antenna, Ca = Ka*D^X Ka Antenna cost coefficient, 0,10 0,10 0,10 0,10 0,10 0,10 0,10 X Antenna cost exponent 3,0 3,0 3,0 3,0 3,0 3,0 3,0 Csm Antenna cost per square meter physical area,k$/m^2 0,637 1,019 1,274 1,529 1,911 2,548 3,822 Receiver Parameters Tsys Tsys = Tln +Tant Tln Tln = Kln*F Kln Lna noise coefficient dependant upon cooler 2,40 2,40 2,40 2,40 2,40 2,40 2,40 Tphy Physical temperature of LNA F Frequency for system temperature specification Ce Electronics cost per antenna, Goal 42,7 42,7 42,7 42,7 42,7 42,7 42,7 Ce Ce = Ccl +Nbn*(Cfd +2* Cln) +Clo+Cif +Cpe 42,7 42,7 42,7 42,7 42,7 42,7 42,7 Ccl Cooling cost per antenna 3,0 3,0 3,0 3,0 3,0 3,0 3,0 Nbn Number of frequency bands Cfd Average dual-polariz feed cost 2,0 2,0 2,0 2,0 2,0 2,0 2,0 Cln Average LNA + mixer cost 0,8 0,8 0,8 0,8 0,8 0,8 0,8 Clo LO cost 3,0 3,0 3,0 3,0 3,0 3,0 3,0 Cif Dual IF cost, Cif =2* (Cifo+Kif*B/2) 3,6 3,6 3,6 3,6 3,6 3,6 3,6 Cifo Fixed IF cost per polarization 1,0 1,0 1,0 1,0 1,0 1,0 1,0 Kif Dual IF cost per GHz of bandwidth 0,2 0,2 0,2 0,2 0,2 0,2 0,2 Signal Transmission Parameters Ctr Cost of transmission, Ctr =2*Ns*(Ne*Clk + Ls*Cfb) Clk Fiber transceiver cost, Clk = Klk*B 8,00 8,00 8,00 8,00 8,00 8,00 8,00 Klk Tranceiver cost per GHz 2,00 2,00 2,00 2,00 2,00 2,00 2,00 Ls Average station to control center distance, km Cfb Fiber installed cost, per fiber per km Signal Processing Parameters (FX architecture assumed for correlator)(bandwidth shared among beams) Cdig Digitization Cdig = (a1*(b/kch)^e+a2) * Kch 10,00 10,00 10,00 10,00 10,00 10,00 10,00 a1 _Digitization coefficient 2,00 2,00 2,00 2,00 2,00 2,00 2,00 e _Digitization exponent 2,00 2,00 2,00 2,00 2,00 2,00 2,00 a2 _Digitization constant 0,50 0,50 0,50 0,50 0,50 0,50 0,50 Kch _Number of separately digitized channels Ctre Transmission, el to stn ctr (Klk*B + Le*Cfb) 9,00 9,00 9,00 9,00 9,00 9,00 9,00 Le _Average distance, element to station center 0,50 0,50 0,50 0,50 0,50 0,50 0,50 Ctrack Tracking, elements Ctrack = (d*b + f*nbeams) 3,28 3,28 3,28 3,28 3,28 3,28 3,28 d _Tracking coefficient (per GHz) 0,72 0,72 0,72 0,72 0,72 0,72 0,72 f _Tracking constant 0,10 0,10 0,10 0,10 0,10 0,10 0,10 Nbeams _Beams per station Cpe Element processing cost = Cdig+Ctre+Ctrack 22,28 22,28 22,28 22,28 22,28 22,28 22,28 Csum Beam summation cost, c*b*(ne-1) 27,913 10,879 6,948 4,813 3,066 1,707 0,736 c _Summation coefficient 0,010 0,010 0,010 0,010 0,010 0,010 0,010 Cps Station processing cost = Cpe*Ne + Csum Cfilt Filtering cost Cfilt = (Kproc+Kmem)*log2(Nl)*B*Ns 2.682, , , , , , ,4 Kproc _Processing cost coefficient (per GHz) 0,48 0,48 0,48 0,48 0,48 0,48 0,48 Kmem _Memory cost (per word) 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 Cci Correlator interconnections Cci=Cct + 2*Klk*Ns*B Cct Cost of corner turner Cct = kct*b*ns ^2*log2(Ns) 2,7 2,7 2,7 2,7 2,7 2,7 2,7 kct _Corner turner cost coefficient (per connection) 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 Ccor Cross correlation cost = (Kcor+Kmem*Nl) *B*Ns*Ns/ Kcor _Correlator cost coef (per baseline-ghz) 0,012 0,012 0,012 0,012 0,012 0,012 0,012 Kchip Price per large FPGA chip 0,144 0,144 0,144 0,144 0,144 0,144 0,144 Nchip Large chip count = (Cfilt+Cci+Ccor)/Kchip Cinfr Correlator infrastructure, $10/large chip Cpc Central processing cost = Cfilt+Cci+Ccor +Cinfr Cp Total signal processing cost = Cps*Ns+Cpc

28 Color code: Purple: correction of error since previous version. Blue: change to value or formula since previous version. Red: variable whose value changes across this sheet Green: variable whose value is different from previous sheet Bold type: input parameter; normal type means value is calculated.

29 SKA Cost vs Antenna Diameter for 3 Cooling Temperatures Aeff/Tsys = 20,000, Aeff=360,000, Tsys=18K, BW=4GHz, Antenna Cost = 0.1D^3 K$, 2001 Electronics Cost = $54K per Element $ Cost, $K (US2001) $ $ $ $ $ Number of Antennas $0 5,0 10,0 15,0 20,0 25,0 30,0 Element Diameter, Meters 15K Cryogenics 60K Cryogenics 180K Cryogenics # Antennas, 15K # Antennas, 60K # Antennas, 180K 0

30 SKA Cost Estimates July 21, 2001 File: skacosteq1.xls Units K$US(2001), meters, GHz Antenna Diameter, Meters 10,0 10,0 10,0 10,0 15,0 15,0 15,0 15,0 Parameter Array Parmeters Value Value Value Value Value Value Value Value C Total array cost, fixed costs, elements, processing Cat Total antenna cost, Ne*Ns*Ca Celt Total electronics cost, Ne*Ns*Ce Cpt Total processing cost includiing station sum Ctr Signal transmission costs Csot Total civil costs at stations, Ns*Cso Co Fixed cost per array - design + central civil Cs Total station cost, elements + combining electronics Cm Total element cost, antenna + receivers + processing Cp Central Signal processing costs Ns Number of stations in array Ne Number of elements per station Deq Equivalent single-antenna diameter of station N Total number of elements, N = Ns*Ne A Effective area of array, A=N*Ae M Specified Figure of Merit, M = A/Tsys M Computed Figure of Merit, M = A/Tsys Tsys System noise temperature at frequency, F B Processed total continuum bandwidth 1,0 2,0 4,0 8,0 1,0 2,0 4,0 8,0 Nl Number of spectral line channels R Antenna/Electronics ratio, Ca/Ce 2,34 2,11 1,68 1,04 7,88 7,11 5,66 3,51 Ropt Minimum cost ratio, Ropt = 1 / (X / 2-1) 2,0 2,0 2,0 2,0 2,0 2,0 2,0 2,0 Antenna Parameters D Physical diameter of element (meters) 10,0 10,0 10,0 10,0 15,0 15,0 15,0 15,0 Ap Physical area of element, Ap = 0.785*D^ Ef Aperature efficiency 0,70 0,70 0,70 0,70 0,70 0,70 0,70 0,70 Ae Effective area of element, Ae=Ap*Ef Tant Antenna noise temperature, Tant = 10 +4*(F/10) Cs Cost per station, Cs =Cso + Ne*(Ca+Ce) Cso Fixed cost per station, land, civil, bunker, cables Ca Cost per antenna, Ca = Ka*D^X Ka Antenna cost coefficient, 0,10 0,10 0,10 0,10 0,10 0,10 0,10 0,10 X Antenna cost exponent 3,0 3,0 3,0 3,0 3,0 3,0 3,0 3,0 Csm Antenna cost per square meter physical area,k$/m^2 1,274 1,274 1,274 1,274 1,911 1,911 1,911 1,911 Receiver Parameters Tsys Tsys = Tln +Tant Tln Tln = Kln*F Kln Lna noise coefficient dependant upon cooler 0,40 0,40 0,40 0,40 0,40 0,40 0,40 0,40 Tphy Physical temperature of LNA F Frequency for system temperature specification Ce Electronics cost per antenna, Goal 42,8 47,4 59,7 96,2 42,8 47,4 59,7 96,2 Ce Ce = Ccl +Nbn*(Cfd +2* Cln) +Clo+Cif +Cpe 42,8 47,4 59,7 96,2 42,8 47,4 59,7 96,2 Ccl Cooling cost per antenna 20,0 20,0 20,0 20,0 20,0 20,0 20,0 20,0 Nbn Number of frequency bands Cfd Average dual-polariz feed cost 2,0 2,0 2,0 2,0 2,0 2,0 2,0 2,0 Cln Average LNA + mixer cost 0,8 0,8 0,8 0,8 0,8 0,8 0,8 0,8 Clo LO cost 3,0 3,0 3,0 3,0 3,0 3,0 3,0 3,0 Cif Dual IF cost, Cif =2* (Cifo+Kif*B/2) 2,4 2,8 3,6 5,2 2,4 2,8 3,6 5,2 Cifo Fixed IF cost per polarization 1,0 1,0 1,0 1,0 1,0 1,0 1,0 1,0 Kif Dual IF cost per GHz of bandwidth 0,2 0,2 0,2 0,2 0,2 0,2 0,2 0,2 Signal Transmission Parameters Ctr Cost of transmission, Ctr =2*Ns*(Ne*Clk + Ls*Cfb) Clk Fiber transceiver cost, Clk = Klk*B 2,00 4,00 8,00 16,00 2,00 4,00 8,00 16,00 Klk Tranceiver cost per GHz 2,00 2,00 2,00 2,00 2,00 2,00 2,00 2,00 Ls Average station to control center distance, km Cfb Fiber installed cost, per fiber per km Signal Processing Parameters (FX architecture assumed for correlator)(bandwidth shared among beams) Cdig Digitization Cdig = (a1*(b/kch)^e+a2) * Kch 2,50 4,00 10,00 34,00 2,50 4,00 10,00 34,00 a1 _Digitization coefficient 2,00 2,00 2,00 2,00 2,00 2,00 2,00 2,00 e _Digitization exponent 2,00 2,00 2,00 2,00 2,00 2,00 2,00 2,00 a2 _Digitization constant 0,50 0,50 0,50 0,50 0,50 0,50 0,50 0,50 Kch _Number of separately digitized channels Ctre Transmission, el to stn ctr (Klk*B + Le*Cfb) 3,00 5,00 9,00 17,00 3,00 5,00 9,00 17,00 Le _Average distance, element to station center 0,50 0,50 0,50 0,50 0,50 0,50 0,50 0,50 Ctrack Tracking, elements Ctrack = (d*b + f*nbeams) 1,12 1,84 3,28 6,16 1,12 1,84 3,28 6,16 d _Tracking coefficient (per GHz) 0,72 0,72 0,72 0,72 0,72 0,72 0,72 0,72 f _Tracking constant 0,10 0,10 0,10 0,10 0,10 0,10 0,10 0,10 Nbeams _Beams per station Cpe Element processing cost = Cdig+Ctre+Ctrack 6,62 10,84 22,28 57,16 6,62 10,84 22,28 57,16 Csum Beam summation cost, c*b*(ne-1) 0,682 1,363 2,726 5,452 0,297 0,595 1,189 2,379 c _Summation coefficient 0,010 0,010 0,010 0,010 0,010 0,010 0,010 0,010 Cps Station processing cost = Cpe*Ne + Csum Cfilt Filtering cost Cfilt = (Kproc+Kmem)*log2(Nl)*B*Ns 670, , , ,8 670, , , ,8 Kproc _Processing cost coefficient (per GHz) 0,48 0,48 0,48 0,48 0,48 0,48 0,48 0,48 Kmem _Memory cost (per word) 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 1,76E-04 Cci Correlator interconnections Cci=Cct + 2*Klk*Ns*B Cct Cost of corner turner Cct = kct*b*ns ^2*log2(Ns) 0,7 1,3 2,7 5,3 0,7 1,3 2,7 5,3 kct _Corner turner cost coefficient (per connection) 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 1,00E-03 Ccor Cross correlation cost = (Kcor+Kmem*Nl) *B*Ns*Ns/ Kcor _Correlator cost coef (per baseline-ghz) 0,012 0,012 0,012 0,012 0,012 0,012 0,012 0,012 Kchip Price per large FPGA chip 0,144 0,144 0,144 0,144 0,144 0,144 0,144 0,144 Nchip Large chip count = (Cfilt+Cci+Ccor)/Kchip Cinfr Correlator infrastructure, $10/large chip Cpc Central processing cost = Cfilt+Cci+Ccor +Cinfr Cp Total signal processing cost = Cps*Ns+Cpc

NMA Antenna and Receiver Concepts

NMA Antenna and Receiver Concepts EVLA Planning Workshop NRAO, Socorro, NM August 23, 2001 NMA Antenna and Receiver Concepts 1. Station Cost Equation 2. Hydroformed Antennas 3. Wideband Receivers Sander Weinreb, Caltech/JPL sweinreb@caltech.edu

More information

Memo 65 SKA Signal processing costs

Memo 65 SKA Signal processing costs Memo 65 SKA Signal processing costs John Bunton, CSIRO ICT Centre 12/08/05 www.skatelescope.org/pages/page_memos.htm Introduction The delay in the building of the SKA has a significant impact on the signal

More information

SKA technology: RF systems & signal processing. Mike Jones University of Oxford

SKA technology: RF systems & signal processing. Mike Jones University of Oxford SKA technology: RF systems & signal processing Mike Jones University of Oxford SKA RF processing Dish receivers Cryogenics RF electronics Fast sampling Antenna processing AA receivers RF gain chain Sampling/antenna

More information

MWA Antenna Description as Supplied by Reeve

MWA Antenna Description as Supplied by Reeve MWA Antenna Description as Supplied by Reeve Basic characteristics: Antennas are shipped broken down and require a few minutes to assemble in the field Each antenna is a dual assembly shaped like a bat

More information

SKA station cost comparison

SKA station cost comparison SKA station cost comparison John D. Bunton, CSIRO Telecommunications and Industrial Physics 4 August 2003 Introduction Current SKA white papers and updates present cost in a variety of ways which makes

More information

A Multi-Fielding SKA Covering the Range 100 MHz 22 GHz. Peter Hall and Aaron Chippendale, CSIRO ATNF 24 November 2003

A Multi-Fielding SKA Covering the Range 100 MHz 22 GHz. Peter Hall and Aaron Chippendale, CSIRO ATNF 24 November 2003 A Multi-Fielding SKA Covering the Range 100 MHz 22 GHz Peter Hall and Aaron Chippendale, CSIRO ATNF 24 November 2003 1. Background Various analyses, including the recent IEMT report [1], have noted that

More information

November SKA Low Frequency Aperture Array. Andrew Faulkner

November SKA Low Frequency Aperture Array. Andrew Faulkner SKA Phase 1 Implementation Southern Africa Australia SKA 1 -mid 250 15m dia. Dishes 0.4-3GHz SKA 1 -low 256,000 antennas Aperture Array Stations 50 350/650MHz SKA 1 -survey 90 15m dia. Dishes 0.7-1.7GHz

More information

Correlator Development at Haystack. Roger Cappallo Haystack-NRAO Technical Mtg

Correlator Development at Haystack. Roger Cappallo Haystack-NRAO Technical Mtg Correlator Development at Haystack Roger Cappallo Haystack-NRAO Technical Mtg. 2006.10.26 History of Correlator Development at Haystack ~1973 Mk I 360 Kb/s x 2 stns. 1981 Mk III 112 Mb/s x 4 stns. 1986

More information

Sideband Smear: Sideband Separation with the ALMA 2SB and DSB Total Power Receivers

Sideband Smear: Sideband Separation with the ALMA 2SB and DSB Total Power Receivers and DSB Total Power Receivers SCI-00.00.00.00-001-A-PLA Version: A 2007-06-11 Prepared By: Organization Date Anthony J. Remijan NRAO A. Wootten T. Hunter J.M. Payne D.T. Emerson P.R. Jewell R.N. Martin

More information

ATA Memo No. 40 Processing Architectures For Complex Gain Tracking. Larry R. D Addario 2001 October 25

ATA Memo No. 40 Processing Architectures For Complex Gain Tracking. Larry R. D Addario 2001 October 25 ATA Memo No. 40 Processing Architectures For Complex Gain Tracking Larry R. D Addario 2001 October 25 1. Introduction In the baseline design of the IF Processor [1], each beam is provided with separate

More information

NRC Herzberg Astronomy & Astrophysics

NRC Herzberg Astronomy & Astrophysics NRC Herzberg Astronomy & Astrophysics SKA Pre-Construction Update Séverin Gaudet, Canadian Astronomy Data Centre David Loop, Director Astronomy Technology June 2016 update SKA Pre-Construction NRC Involvement

More information

Evolution of the Capabilities of the ALMA Array

Evolution of the Capabilities of the ALMA Array Evolution of the Capabilities of the ALMA Array This note provides an outline of how we plan to build up the scientific capabilities of the array from the start of Early Science through to Full Operations.

More information

User Guide for the Calculators Version 0.9

User Guide for the Calculators Version 0.9 User Guide for the Calculators Version 0.9 Last Update: Nov 2 nd 2008 By: Shahin Farahani Copyright 2008, Shahin Farahani. All rights reserved. You may download a copy of this calculator for your personal

More information

The Phased Array Feed Receiver System : Linearity, Cross coupling and Image Rejection

The Phased Array Feed Receiver System : Linearity, Cross coupling and Image Rejection The Phased Array Feed Receiver System : Linearity, Cross coupling and Image Rejection D. Anish Roshi 1,2, Robert Simon 1, Steve White 1, William Shillue 2, Richard J. Fisher 2 1 National Radio Astronomy

More information

March Phased Array Technology. Andrew Faulkner

March Phased Array Technology. Andrew Faulkner Aperture Arrays Michael Kramer Sparse Type of AA selection 1000 Sparse AA-low Sky Brightness Temperature (K) 100 10 T sky A eff Fully sampled AA-mid Becoming sparse Aeff / T sys (m 2 / K) Dense A eff /T

More information

EVLA Memo #166 Comparison of the Performance of the 3-bit and 8-bit Samplers at C (4 8 GHz), X (8 12 GHz) and Ku (12 18 GHz) Bands

EVLA Memo #166 Comparison of the Performance of the 3-bit and 8-bit Samplers at C (4 8 GHz), X (8 12 GHz) and Ku (12 18 GHz) Bands EVLA Memo #166 Comparison of the Performance of the 3-bit and 8-bit Samplers at C (4 8 GHz), X (8 12 GHz) and Ku (12 18 GHz) Bands E. Momjian and R. Perley NRAO March 27, 2013 Abstract We present sensitivity

More information

Very Long Baseline Interferometry

Very Long Baseline Interferometry Very Long Baseline Interferometry Cormac Reynolds, JIVE European Radio Interferometry School, Bonn 12 Sept. 2007 VLBI Arrays EVN (Europe, China, South Africa, Arecibo) VLBA (USA) EVN + VLBA coordinate

More information

ATA memo 73. The ATA Correlator. W.L. Urry, M. Wright, M. Dexter, D. MacMahon. 16 Feb Abstract

ATA memo 73. The ATA Correlator. W.L. Urry, M. Wright, M. Dexter, D. MacMahon. 16 Feb Abstract ATA memo 73 The ATA Correlator W.L. Urry, M. Wright, M. Dexter, D. MacMahon 16 Feb. 2007 Abstract This memo describes the current status of the ATA correlator. The correlator is designed for 352 antennas,

More information

Technology Drivers, SKA Pathfinders P. Dewdney

Technology Drivers, SKA Pathfinders P. Dewdney Technology Drivers, SKA Pathfinders P. Dewdney Dominion Radio Astrophysical Observatory Herzberg Institute of Astrophysics National Research Council Canada National Research Council Canada Conseil national

More information

VLBA Correlator Memo No.-ll

VLBA Correlator Memo No.-ll VLBA Correlator Memo No.-ll (860709) VLBA CORRELATOR HARDWARE TOPICS Ray Escoffier July 9, 1986 I) Introduction This memo will describe the present state of the design of the VLBA correlator. The thoughts

More information

Integrated receivers for mid-band SKA. Suzy Jackson Engineer, Australia Telescope National Facility

Integrated receivers for mid-band SKA. Suzy Jackson Engineer, Australia Telescope National Facility Integrated receivers for mid-band SKA Suzy Jackson Engineer, Australia Telescope National Facility ASKAP/SKA Special Technical Brief 23 rd October, 2009 Talk overview Mid band SKA receiver challenges ASKAP

More information

Receivers for. FFRF Tutorial by Tom Clark, NASA/GSFC & NVI Wettzell, March 19, 2009

Receivers for. FFRF Tutorial by Tom Clark, NASA/GSFC & NVI Wettzell, March 19, 2009 Receivers for VLBI2010 FFRF Tutorial by Tom Clark, NASA/GSFC & NVI Wettzell, March 19, 2009 There is no fundamental difference between the receivers for PRIME FOCUS & CASSEGRAIN Except for: the beamwidth

More information

May AA Communications. Portugal

May AA Communications. Portugal SKA Top-level description A large radio telescope for transformational science Up to 1 million m 2 collecting area Operating from 70 MHz to 10 GHz (4m-3cm) Two or more detector technologies Connected to

More information

Overview of the SKA. P. Dewdney International SKA Project Engineer Nov 9, 2009

Overview of the SKA. P. Dewdney International SKA Project Engineer Nov 9, 2009 Overview of the SKA P. Dewdney International SKA Project Engineer Nov 9, 2009 Outline* 1. SKA Science Drivers. 2. The SKA System. 3. SKA technologies. 4. Trade-off space. 5. Scaling. 6. Data Rates & Data

More information

ngvla Workshop Signal organization for long-distance transfer with wide-band front ends Larry D'Addario 2015 December 9

ngvla Workshop Signal organization for long-distance transfer with wide-band front ends Larry D'Addario 2015 December 9 ngvla Workshop Signal organization for long-distance transfer with wide-band front ends Larry D'Addario 2015 December 9 Outline This is a short talk, giving a high-level, somewhat tutorial view. Illustrates

More information

2-PAD: An Introduction. The 2-PAD Team

2-PAD: An Introduction. The 2-PAD Team 2-PAD: An Introduction The 2-PAD Team Workshop, Jodrell Bank, 10 Presented th November 2009 by 2-PAD: Dr An Georgina Introduction Harris Georgina Harris for the 2-PAD Team 1 2-PAD Objectives Demonstrate

More information

EVLA System Commissioning Results

EVLA System Commissioning Results EVLA System Commissioning Results EVLA Advisory Committee Meeting, March 19-20, 2009 Rick Perley EVLA Project Scientist t 1 Project Requirements EVLA Project Book, Chapter 2, contains the EVLA Project

More information

Focal Plane Array Beamformer for the Expanded GMRT: Initial

Focal Plane Array Beamformer for the Expanded GMRT: Initial Focal Plane Array Beamformer for the Expanded GMRT: Initial Implementation on ROACH Kaushal D. Buch Digital Backend Group, Giant Metrewave Radio Telescope, NCRA-TIFR, Pune, India kdbuch@gmrt.ncra.tifr.res.in

More information

LWA Beamforming Design Concept

LWA Beamforming Design Concept LWA Beamforming Design Concept Steve Ellingson October 3, 27 Contents Introduction 2 2 Integer Sample Period Delay 2 3 Fractional Sample Period Delay 3 4 Summary 9 Bradley Dept. of Electrical & Computer

More information

LOCALIZED LNA COOLING IN VACUUM

LOCALIZED LNA COOLING IN VACUUM Nice, Côte d Azur, France, 27-29 September 2006 LOCALIZED LNA COOLING IN VACUUM Frans Schreuder, Jan Geralt Bij de Vaate ASRON, P.O. Box 2, 7990 AA Dwingeloo, he Netherlands. schreuder@astron.nl ABSRAC

More information

Instrument Requirements and Options for Meeting the Science Opportunities MHz P. Dewdney A. Gray, B. Veidt

Instrument Requirements and Options for Meeting the Science Opportunities MHz P. Dewdney A. Gray, B. Veidt Instrument Requirements and Options for Meeting the Science Opportunities 300-3000 MHz P. Dewdney A. Gray, B. Veidt Dominion Radio Astrophysical Observatory Herzberg Institute of Astrophysics National

More information

"Octave" Project: Application of Superwide-Band Technologies for the RATAN-600 Continuum radiometers

Octave Project: Application of Superwide-Band Technologies for the RATAN-600 Continuum radiometers : Application of Superwide-Band Technologies for the RATAN-600 Continuum radiometers E-mail: marat@sao.ru A.B.Berlin, Saint Petersburg Branch 196140,Saint Petersburg, Russia E-mail: abb_36@mail.ru N.A.Nizhel

More information

Symmetry in the Ka-band Correlation Receiver s Input Circuit and Spectral Baseline Structure NRAO GBT Memo 248 June 7, 2007

Symmetry in the Ka-band Correlation Receiver s Input Circuit and Spectral Baseline Structure NRAO GBT Memo 248 June 7, 2007 Symmetry in the Ka-band Correlation Receiver s Input Circuit and Spectral Baseline Structure NRAO GBT Memo 248 June 7, 2007 A. Harris a,b, S. Zonak a, G. Watts c a University of Maryland; b Visiting Scientist,

More information

ngvla The Next Generation Very Large Array

ngvla The Next Generation Very Large Array Perspective from the Technical Advisory Council Melissa Soriano, Jet Propulsion Laboratory, California Institute of Technology James Lamb, California Institute of Technology ngvla 2017 California Institute

More information

Green Bank Instrumentation circa 2030

Green Bank Instrumentation circa 2030 Green Bank Instrumentation circa 2030 Dan Werthimer and 800 CASPER Collaborators http://casper.berkeley.edu Upcoming Nobel Prizes with Radio Instrumentation Gravitational Wave Detection (pulsar timing)

More information

LOCALIZED LNA COOLING IN VACUUM

LOCALIZED LNA COOLING IN VACUUM Nice, Côte d Azur, France, 27-29 September 2006 LOCALIZED LNA COOLING IN VACUUM Frans Schreuder, Jan Geralt Bij de Vaate ASRON, P.O. Box 2, 7990 AA Dwingeloo, he Netherlands. schreuder@astron.nl ABSRAC

More information

Outline / Wireless Networks and Applications Lecture 3: Physical Layer Signals, Modulation, Multiplexing. Cartoon View 1 A Wave of Energy

Outline / Wireless Networks and Applications Lecture 3: Physical Layer Signals, Modulation, Multiplexing. Cartoon View 1 A Wave of Energy Outline 18-452/18-750 Wireless Networks and Applications Lecture 3: Physical Layer Signals, Modulation, Multiplexing Peter Steenkiste Carnegie Mellon University Spring Semester 2017 http://www.cs.cmu.edu/~prs/wirelesss17/

More information

Towards SKA Multi-beam concepts and technology

Towards SKA Multi-beam concepts and technology Towards SKA Multi-beam concepts and technology SKA meeting Meudon Observatory, 16 June 2009 Philippe Picard Station de Radioastronomie de Nançay philippe.picard@obs-nancay.fr 1 Square Kilometre Array:

More information

EVLA Memo 105. Phase coherence of the EVLA radio telescope

EVLA Memo 105. Phase coherence of the EVLA radio telescope EVLA Memo 105 Phase coherence of the EVLA radio telescope Steven Durand, James Jackson, and Keith Morris National Radio Astronomy Observatory, 1003 Lopezville Road, Socorro, NM, USA 87801 ABSTRACT The

More information

Specifications for the GBT spectrometer

Specifications for the GBT spectrometer GBT memo No. 292 Specifications for the GBT spectrometer Authors: D. Anish Roshi 1, Green Bank Scientific Staff, J. Richard Fisher 2, John Ford 1 Affiliation: 1 NRAO, Green Bank, WV 24944. 2 NRAO, Charlottesville,

More information

ngvla Advanced Cryocoolers For ngvla NATIONAL RADIO ASTRONOMY OBSERVATORY Larry D Addario, Caltech ngvlaworkshop, Socorro, 2017 June 26

ngvla Advanced Cryocoolers For ngvla NATIONAL RADIO ASTRONOMY OBSERVATORY Larry D Addario, Caltech ngvlaworkshop, Socorro, 2017 June 26 NATIONAL RADIO ASTRONOMY OBSERVATORY Advanced Cryocoolers For ngvla Larry D Addario, Caltech ngvlaworkshop, Socorro, 2017 June 26 ngvla Outline How cold do we need to get? Tutorial on cryocoolers (just

More information

HIFAS: Wide-band spectrometer ASIC

HIFAS: Wide-band spectrometer ASIC HIFAS: Wide-band spectrometer ASIC Anders Emrich, Stefan Andersson, Johan Dahlberg, Magnus Hjorth, Omnisys Instruments AB Torgil Kjellberg, Chalmers University Of Technology Microelectronics Presentation

More information

A model for the SKA. Melvyn Wright. Radio Astronomy laboratory, University of California, Berkeley, CA, ABSTRACT

A model for the SKA. Melvyn Wright. Radio Astronomy laboratory, University of California, Berkeley, CA, ABSTRACT SKA memo 16. 21 March 2002 A model for the SKA Melvyn Wright Radio Astronomy laboratory, University of California, Berkeley, CA, 94720 ABSTRACT This memo reviews the strawman design for the SKA telescope.

More information

SKA Correlator Input Data Rate

SKA Correlator Input Data Rate SKA Correlator Input Data Rate John D. Bunton, CSIRO ICT Centre 14 November 003 Introduction One of the major costs in the SKA is data transport from the antenna to the central beamformer or correlator.

More information

2 Gain Variation from the Receiver Output through the IF Path

2 Gain Variation from the Receiver Output through the IF Path EVLA Memo #185 Bandwidth- and Frequency-Dependent Effects in the T34 Total Power Detector Keith Morris September 17, 214 1 Introduction The EVLA Intermediate Frequency (IF) system employs a system of power

More information

Design Considerations for 5G mm-wave Receivers. Stefan Andersson, Lars Sundström, and Sven Mattisson

Design Considerations for 5G mm-wave Receivers. Stefan Andersson, Lars Sundström, and Sven Mattisson Design Considerations for 5G mm-wave Receivers Stefan Andersson, Lars Sundström, and Sven Mattisson Outline Introduction to 5G @ mm-waves mm-wave on-chip frequency generation mm-wave analog front-end design

More information

EVLA Antenna and Array Performance. Rick Perley

EVLA Antenna and Array Performance. Rick Perley EVLA Antenna and Array Performance System Requirements EVLA Project Book, Chapter 2, contains the EVLA system requirements. For most, astronomical tests are necessary to determine if the array meets requirements.

More information

An FPGA-Based Back End for Real Time, Multi-Beam Transient Searches Over a Wide Dispersion Measure Range

An FPGA-Based Back End for Real Time, Multi-Beam Transient Searches Over a Wide Dispersion Measure Range An FPGA-Based Back End for Real Time, Multi-Beam Transient Searches Over a Wide Dispersion Measure Range Larry D'Addario 1, Nathan Clarke 2, Robert Navarro 1, and Joseph Trinh 1 1 Jet Propulsion Laboratory,

More information

Antennas. Greg Taylor. University of New Mexico Spring Astronomy 423 at UNM Radio Astronomy

Antennas. Greg Taylor. University of New Mexico Spring Astronomy 423 at UNM Radio Astronomy Antennas Greg Taylor University of New Mexico Spring 2017 Astronomy 423 at UNM Radio Astronomy Outline 2 Fourier Transforms Interferometer block diagram Antenna fundamentals Types of antennas Antenna performance

More information

A Noise-Temperature Measurement System Using a Cryogenic Attenuator

A Noise-Temperature Measurement System Using a Cryogenic Attenuator TMO Progress Report 42-135 November 15, 1998 A Noise-Temperature Measurement System Using a Cryogenic Attenuator J. E. Fernandez 1 This article describes a method to obtain accurate and repeatable input

More information

Antennas and Propagation

Antennas and Propagation Antennas and Propagation Chapter 5 Introduction An antenna is an electrical conductor or system of conductors Transmission - radiates electromagnetic energy into space Reception - collects electromagnetic

More information

Merging Propagation Physics, Theory and Hardware in Wireless. Ada Poon

Merging Propagation Physics, Theory and Hardware in Wireless. Ada Poon HKUST January 3, 2007 Merging Propagation Physics, Theory and Hardware in Wireless Ada Poon University of Illinois at Urbana-Champaign Outline Multiple-antenna (MIMO) channels Human body wireless channels

More information

Longer baselines and how it impacts the ALMA Central LO

Longer baselines and how it impacts the ALMA Central LO Longer baselines and how it impacts the ALMA Central LO 1 C. Jacques - NRAO October 3-4-5 2017 ALMA LBW Quick overview of current system Getting the data back is not the problem (digital transmission),

More information

Announcements : Wireless Networks Lecture 3: Physical Layer. Bird s Eye View. Outline. Page 1

Announcements : Wireless Networks Lecture 3: Physical Layer. Bird s Eye View. Outline. Page 1 Announcements 18-759: Wireless Networks Lecture 3: Physical Layer Please start to form project teams» Updated project handout is available on the web site Also start to form teams for surveys» Send mail

More information

Phased Array Feeds for the SKA. WP2.2.3 PAFSKA Consortium CSIRO ASTRON DRAO NRAO BYU OdP Nancay Cornell U Manchester

Phased Array Feeds for the SKA. WP2.2.3 PAFSKA Consortium CSIRO ASTRON DRAO NRAO BYU OdP Nancay Cornell U Manchester Phased Array Feeds for the SKA WP2.2.3 PAFSKA Consortium CSIRO ASTRON DRAO NRAO BYU OdP Nancay Cornell U Manchester Dish Array Hierarchy Dish Array L5 Elements PAF Dish Single Pixel Feeds L4 Sub systems

More information

THE KAROO ARRAY TELESCOPE (KAT) & FPA EFFORT IN SOUTH AFRICA

THE KAROO ARRAY TELESCOPE (KAT) & FPA EFFORT IN SOUTH AFRICA THE KAROO ARRAY TELESCOPE (KAT) & FPA EFFORT IN SOUTH AFRICA Dr. Dirk Baker (KAT FPA Sub-system Manager) Prof. Justin Jonas (SKA SA Project Scientist) Ms. Anita Loots (KAT Project Manager) Mr. David de

More information

Detector Systems. Graeme Carrad

Detector Systems. Graeme Carrad Detector Systems Graeme Carrad November 2011 The Basic Structure of a typical Radio Telescope Antenna Receiver Conversion Digitiser Signal Processing / Correlator They are much the same CSIRO. Radiotelescope

More information

Cross Correlators. Jayce Dowell/Greg Taylor. University of New Mexico Spring Astronomy 423 at UNM Radio Astronomy

Cross Correlators. Jayce Dowell/Greg Taylor. University of New Mexico Spring Astronomy 423 at UNM Radio Astronomy Cross Correlators Jayce Dowell/Greg Taylor University of New Mexico Spring 2017 Astronomy 423 at UNM Radio Astronomy Outline 2 Re-cap of interferometry What is a correlator? The correlation function Simple

More information

Future Arrays for Radio Astronomy and Space Communications. Sander Weinreb. Presentation to KNI/MDL Seminar, Aug 3, 2009

Future Arrays for Radio Astronomy and Space Communications. Sander Weinreb. Presentation to KNI/MDL Seminar, Aug 3, 2009 Future Arrays for Radio Astronomy and Space Communications Sander Weinreb Presentation to KNI/MDL Seminar, Aug 3, 2009 Square-Km Array Phased-Array Feeds Large format focal plane imaging IC development

More information

Radio Interferometers Around the World. Amy J. Mioduszewski (NRAO)

Radio Interferometers Around the World. Amy J. Mioduszewski (NRAO) Radio Interferometers Around the World Amy J. Mioduszewski (NRAO) A somewhat biased view of current interferometers Limited to telescopes that exist or are in the process of being built (i.e., I am not

More information

Active Impedance Matched Dual-Polarization Phased Array Feed for the GBT

Active Impedance Matched Dual-Polarization Phased Array Feed for the GBT Active Impedance Matched Dual-Polarization Phased Array Feed for the GBT Karl F. Warnick, David Carter, Taylor Webb, Brian D. Jeffs Department of Electrical and Computer Engineering Brigham Young University,

More information

MAKING TRANSIENT ANTENNA MEASUREMENTS

MAKING TRANSIENT ANTENNA MEASUREMENTS MAKING TRANSIENT ANTENNA MEASUREMENTS Roger Dygert, Steven R. Nichols MI Technologies, 1125 Satellite Boulevard, Suite 100 Suwanee, GA 30024-4629 ABSTRACT In addition to steady state performance, antennas

More information

Integrated receivers for mid-band SKA. Suzy Jackson Engineer, Australia Telescope National Facility

Integrated receivers for mid-band SKA. Suzy Jackson Engineer, Australia Telescope National Facility Integrated receivers for mid-band SKA Suzy Jackson Engineer, Australia Telescope National Facility SKADS FP6 Meeting Chateau de Limelette 4-6 November, 2009 Talk overview Mid band SKA receiver challenges

More information

AVN Training HartRAO 2016

AVN Training HartRAO 2016 AVN Training HartRAO 2016 Microwave 1 Overview Introduction to basic components used in microwave receivers. Performance characteristics of these components. Assembly of components into a complete microwave

More information

Technical Requirements for Fixed Line-of-Sight Radio Systems Operating in the Band GHz

Technical Requirements for Fixed Line-of-Sight Radio Systems Operating in the Band GHz Issue 4 March 2018 Spectrum Management and Telecommunications Standard Radio System Plan Technical Requirements for Fixed Line-of-Sight Radio Systems Operating in the Band 10.7-11.7 GHz Aussi disponible

More information

VIIP: a PCI programmable board.

VIIP: a PCI programmable board. VIIP: a PCI programmable board. G. Bianchi (1), L. Zoni (1), S. Montebugnoli (1) (1) Institute of Radio Astronomy, National Institute for Astrophysics Via Fiorentina 3508/B, 40060 Medicina (BO), Italy.

More information

Antennas. Greg Taylor. University of New Mexico Spring Astronomy 423 at UNM Radio Astronomy

Antennas. Greg Taylor. University of New Mexico Spring Astronomy 423 at UNM Radio Astronomy Antennas Greg Taylor University of New Mexico Spring 2011 Astronomy 423 at UNM Radio Astronomy Radio Window 2 spans a wide range of λ and ν from λ ~ 0.33 mm to ~ 20 m! (ν = 1300 GHz to 15 MHz ) Outline

More information

10. DSP Blocks in Arria GX Devices

10. DSP Blocks in Arria GX Devices 10. SP Blocks in Arria GX evices AGX52010-1.2 Introduction Arria TM GX devices have dedicated digital signal processing (SP) blocks optimized for SP applications requiring high data throughput. These SP

More information

Signals and Systems Lecture 9 Communication Systems Frequency-Division Multiplexing and Frequency Modulation (FM)

Signals and Systems Lecture 9 Communication Systems Frequency-Division Multiplexing and Frequency Modulation (FM) Signals and Systems Lecture 9 Communication Systems Frequency-Division Multiplexing and Frequency Modulation (FM) April 11, 2008 Today s Topics 1. Frequency-division multiplexing 2. Frequency modulation

More information

A Method for Gain over Temperature Measurements Using Two Hot Noise Sources

A Method for Gain over Temperature Measurements Using Two Hot Noise Sources A Method for Gain over Temperature Measurements Using Two Hot Noise Sources Vince Rodriguez and Charles Osborne MI Technologies: Suwanee, 30024 GA, USA vrodriguez@mitechnologies.com Abstract P Gain over

More information

Focal Plane Arrays & SKA

Focal Plane Arrays & SKA Focal Plane Arrays & SKA Peter Hall SKA International Project Engineer www.skatelescope.org Dwingeloo, June 20 2005 Outline Today: SKA and antennas Phased arrays and SKA Hybrid SKA possibilities» A hybrid

More information

Antennas & Propagation. CSG 250 Fall 2007 Rajmohan Rajaraman

Antennas & Propagation. CSG 250 Fall 2007 Rajmohan Rajaraman Antennas & Propagation CSG 250 Fall 2007 Rajmohan Rajaraman Introduction An antenna is an electrical conductor or system of conductors o Transmission - radiates electromagnetic energy into space o Reception

More information

EVLA Memo # 194 EVLA Ka-band Receiver Down Converter Module Harmonics: The Mega-Birdie at MHz

EVLA Memo # 194 EVLA Ka-band Receiver Down Converter Module Harmonics: The Mega-Birdie at MHz EVLA Memo # 194 EVLA Ka-band Receiver Down Converter Module Harmonics: The Mega-Birdie at 29440 MHz R. Selina, E. Momjian, W. Grammer, J. Jackson NRAO February 5, 2016 Abstract Observations carried out

More information

Advanced Digital Receiver

Advanced Digital Receiver Advanced Digital Receiver MI-750 FEATURES Industry leading performance with up to 4 M samples per second 135 db dynamic range and -150 dbm sensitivity Optimized timing for shortest overall test time Wide

More information

N. Pingel, K. Rajwade, D.J. Pisano, D. Lorimer West Virginia University

N. Pingel, K. Rajwade, D.J. Pisano, D. Lorimer West Virginia University Brian D. Jeffs, R. Black, J. Diao, M. Ruzindanna, K. Warnick Brigham Young University R. Prestage, J. Ford, S. White, R. Simon, W. Shillue, A. Roshi, V. Van Tonder NRAO: Green Bank Observatory and Central

More information

Multi-octave radio frequency systems: Developments of antenna technology in radio astronomy and imaging systems

Multi-octave radio frequency systems: Developments of antenna technology in radio astronomy and imaging systems Multi-octave radio frequency systems: Developments of antenna technology in radio astronomy and imaging systems Professor Tony Brown School of Electrical and Electronic Engineering University of Manchester

More information

EE M255, BME M260, NS M206:

EE M255, BME M260, NS M206: EE M255, BME M260, NS M206: NeuroEngineering Lecture Set 6: Neural Recording Prof. Dejan Markovic Agenda Neural Recording EE Model System Components Wireless Tx 6.2 Neural Recording Electrodes sense action

More information

J/K). Nikolova

J/K). Nikolova Lecture 7: ntenna Noise Temperature and System Signal-to-Noise Ratio (Noise temperature. ntenna noise temperature. System noise temperature. Minimum detectable temperature. System signal-to-noise ratio.)

More information

SKA1 low Baseline Design: Lowest Frequency Aspects & EoR Science

SKA1 low Baseline Design: Lowest Frequency Aspects & EoR Science SKA1 low Baseline Design: Lowest Frequency Aspects & EoR Science 1 st science Assessment WS, Jodrell Bank P. Dewdney Mar 27, 2013 Intent of the Baseline Design Basic architecture: 3-telescope, 2-system

More information

Antennas and Receivers in Radio Astronomy

Antennas and Receivers in Radio Astronomy Antennas and Receivers in Radio Astronomy Mark McKinnon Eleventh Synthesis Imaging Workshop Socorro, June 10-17, 2008 Outline 2 Context Types of antennas Antenna fundamentals Reflector antennas Mounts

More information

Why Single Dish? Darrel Emerson NRAO Tucson. NAIC-NRAO School on Single-Dish Radio Astronomy. Green Bank, August 2003.

Why Single Dish? Darrel Emerson NRAO Tucson. NAIC-NRAO School on Single-Dish Radio Astronomy. Green Bank, August 2003. Why Single Dish? Darrel Emerson NRAO Tucson NAIC-NRAO School on Single-Dish Radio Astronomy. Green Bank, August 2003. Why Single Dish? What's the Alternative? Comparisons between Single-Dish, Phased Array

More information

RECOMMENDATION ITU-R S.733-1* (Question ITU-R 42/4 (1990))**

RECOMMENDATION ITU-R S.733-1* (Question ITU-R 42/4 (1990))** Rec. ITU-R S.733-1 1 RECOMMENDATION ITU-R S.733-1* DETERMINATION OF THE G/T RATIO FOR EARTH STATIONS OPERATING IN THE FIXED-SATELLITE SERVICE (Question ITU-R 42/4 (1990))** Rec. ITU-R S.733-1 (1992-1993)

More information

The Measurement and Characterisation of Ultra Wide-Band (UWB) Intentionally Radiated Signals

The Measurement and Characterisation of Ultra Wide-Band (UWB) Intentionally Radiated Signals The Measurement and Characterisation of Ultra Wide-Band (UWB) Intentionally Radiated Signals Rafael Cepeda Toshiba Research Europe Ltd University of Bristol November 2007 Rafael.cepeda@toshiba-trel.com

More information

THE BENEFITS OF DSP LOCK-IN AMPLIFIERS

THE BENEFITS OF DSP LOCK-IN AMPLIFIERS THE BENEFITS OF DSP LOCK-IN AMPLIFIERS If you never heard of or don t understand the term lock-in amplifier, you re in good company. With the exception of the optics industry where virtually every major

More information

2015 The MathWorks, Inc. 1

2015 The MathWorks, Inc. 1 2015 The MathWorks, Inc. 1 What s Behind 5G Wireless Communications? 서기환과장 2015 The MathWorks, Inc. 2 Agenda 5G goals and requirements Modeling and simulating key 5G technologies Release 15: Enhanced Mobile

More information

Designing a Sky-Noise-Limited Receiver for LWA

Designing a Sky-Noise-Limited Receiver for LWA The Next Generation of Receivers for Low Frequency Radio Astronomy: Designing a Sky-Noise-Limited Receiver for LWA Steve Ellingson Contributions from D. Wilson, T. Kramer Virginia Tech ellingson@vt.edu

More information

Advances in Antenna Measurement Instrumentation and Systems

Advances in Antenna Measurement Instrumentation and Systems Advances in Antenna Measurement Instrumentation and Systems Steven R. Nichols, Roger Dygert, David Wayne MI Technologies Suwanee, Georgia, USA Abstract Since the early days of antenna pattern recorders,

More information

Traditional analog QDC chain and Digital Pulse Processing [1]

Traditional analog QDC chain and Digital Pulse Processing [1] Giuliano Mini Viareggio April 22, 2010 Introduction The aim of this paper is to compare the energy resolution of two gamma ray spectroscopy setups based on two different acquisition chains; the first chain

More information

ELEC RADAR FRONT-END SUMMARY

ELEC RADAR FRONT-END SUMMARY ELEC Radar Front-End is designed for FMCW (including CW) radar application. The output frequency of each RX provides range, speed, and amplitude information to DSP. It will detect target azimuth angle

More information

Frequency-Modulated Continuous-Wave Radar (FM-CW Radar)

Frequency-Modulated Continuous-Wave Radar (FM-CW Radar) Frequency-Modulated Continuous-Wave Radar (FM-CW Radar) FM-CW radar (Frequency-Modulated Continuous Wave radar = FMCW radar) is a special type of radar sensor which radiates continuous transmission power

More information

6. DSP Blocks in Stratix II and Stratix II GX Devices

6. DSP Blocks in Stratix II and Stratix II GX Devices 6. SP Blocks in Stratix II and Stratix II GX evices SII52006-2.2 Introduction Stratix II and Stratix II GX devices have dedicated digital signal processing (SP) blocks optimized for SP applications requiring

More information

EVLA Memo 151 EVLA Antenna Polarization at L, S, C, and X Bands

EVLA Memo 151 EVLA Antenna Polarization at L, S, C, and X Bands EVLA Memo 11 EVLA Antenna Polarization at L, S, C, and X Bands Rick Perley and Bob Hayward April 28, 211 Abstract The method described in EVLA Memo #131 for determining absolute antenna cross-polarization

More information

ECE 476/ECE 501C/CS Wireless Communication Systems Winter Lecture 6: Fading

ECE 476/ECE 501C/CS Wireless Communication Systems Winter Lecture 6: Fading ECE 476/ECE 501C/CS 513 - Wireless Communication Systems Winter 2005 Lecture 6: Fading Last lecture: Large scale propagation properties of wireless systems - slowly varying properties that depend primarily

More information

The Australian SKA Pathfinder Project. ASKAP Digital Signal Processing Systems System Description & Overview of Industry Opportunities

The Australian SKA Pathfinder Project. ASKAP Digital Signal Processing Systems System Description & Overview of Industry Opportunities The Australian SKA Pathfinder Project ASKAP Digital Signal Processing Systems System Description & Overview of Industry Opportunities This paper describes the delivery of the digital signal processing

More information

Update on Optics Design for SKA Reflector Antennas with Wide Band Single Pixel Feeds

Update on Optics Design for SKA Reflector Antennas with Wide Band Single Pixel Feeds Update on Optics Design for SKA Reflector Antennas with Wide Band Single Pixel Feeds William Imbriale (2), Germán Cortés (1), and Lynn Baker (1) (1) NAIC, Cornell University, Ithaca, NY, 14853, USA (2)

More information

EVLA Memo #119 Wide-Band Sensitivity and Frequency Coverage of the EVLA and VLA L-Band Receivers

EVLA Memo #119 Wide-Band Sensitivity and Frequency Coverage of the EVLA and VLA L-Band Receivers EVLA Memo #119 Wide-Band Sensitivity and Frequency Coverage of the EVLA and VLA L-Band Receivers Rick Perley and Bob Hayward January 17, 8 Abstract We determine the sensitivities of the EVLA and VLA antennas

More information

ECE 476/ECE 501C/CS Wireless Communication Systems Winter Lecture 6: Fading

ECE 476/ECE 501C/CS Wireless Communication Systems Winter Lecture 6: Fading ECE 476/ECE 501C/CS 513 - Wireless Communication Systems Winter 2004 Lecture 6: Fading Last lecture: Large scale propagation properties of wireless systems - slowly varying properties that depend primarily

More information

WHAT ARE FIELD PROGRAMMABLE. Audible plays called at the line of scrimmage? Signaling for a squeeze bunt in the ninth inning?

WHAT ARE FIELD PROGRAMMABLE. Audible plays called at the line of scrimmage? Signaling for a squeeze bunt in the ninth inning? WHAT ARE FIELD PROGRAMMABLE Audible plays called at the line of scrimmage? Signaling for a squeeze bunt in the ninth inning? They re none of the above! We re going to take a look at: Field Programmable

More information

UNIT- 7. Frequencies above 30Mhz tend to travel in straight lines they are limited in their propagation by the curvature of the earth.

UNIT- 7. Frequencies above 30Mhz tend to travel in straight lines they are limited in their propagation by the curvature of the earth. UNIT- 7 Radio wave propagation and propagation models EM waves below 2Mhz tend to travel as ground waves, These wave tend to follow the curvature of the earth and lose strength rapidly as they travel away

More information

Digital Logic, Algorithms, and Functions for the CEBAF Upgrade LLRF System Hai Dong, Curt Hovater, John Musson, and Tomasz Plawski

Digital Logic, Algorithms, and Functions for the CEBAF Upgrade LLRF System Hai Dong, Curt Hovater, John Musson, and Tomasz Plawski Digital Logic, Algorithms, and Functions for the CEBAF Upgrade LLRF System Hai Dong, Curt Hovater, John Musson, and Tomasz Plawski Introduction: The CEBAF upgrade Low Level Radio Frequency (LLRF) control

More information