Blind Ocean Acoustic Tomography presentation at the ENEA Workshop on Ocean Acoustic Tomography: results and perspectives 28 November 22 Santa Teresa, Lerici, Itália by S.M. Jesus and C.Soares SiPLAB, FCT - Universidade do Algarve Campus de Gambelas, PT-8 Faro, Portugal
Blind Ocean Acoustic Tomography S.M. Jesus and C. Soares {sjesus,csoares}@ualg.pt SiPLAB, FCT - Universidade do Algarve, Campus de Gambelas, PT - Faro, Portugal Supported under ATOMS, FCT contract PDCTM/P/MAR/15296/1999 and TOMPACO, CNR, Italy.
Outline the game of the name and motivation the INTIFANTE sea trial environmental inversion active tomography passive tomography summary
source receiver X Ocean Acoustic Tomography - synoptic acoustic pressure y r( ζ ) Correlator Optimization p( ζ ) Acoustic Model ζ new environment source signal s(t) (r,z) ^ ζ
Motivation decrease the dependency on source knowledge (emitted signal and source position) increase the (time-space) adaptivity ability to use non-co-operative (noise) sources
INTIFANTE Experimental Site -9 3' -9 15' -9 ' -8 45' -8 3' 39 ' 39 ' 38 45' 38 45' LISBON Setubal 38 3' 38 3' Sesimbra Espichel Cape INTIFANTE 99 Area 38 15' 38 15' 38 ' 38 ' -9 3' -9 15' -9 ' -8 45' -8 3'
Geometry and Bathymetry
Vertical Line Array
Radio Buoy: deployment and setup
Recovering the Telemetry Unit
Sound source and emitted signals
Events 2 and 5: bathymetry and runs
Temperature evolution (XBT)
119 156 151 1515 Soundspeed (m/s) 1 N ˆN = min n=1 λ 2 n { >.8} N M m=1 λ 2 m 8 Depth (m) 6 Depth (m) 4 n=1 α nun ceof = ĉ + N 2 Empirical Orthogonal Functions 2 4 6 8 1 119 1st EOF 2nd EOF.2.1.1 Soundspeed (m/s)
Depth (m) 6 119 Physical model - NW/NE track 156 1518 m/s VA Source (63 m) Sediment 2. m 175 m/s α=.8 db/λ ρ=1.9 g/cm 3 18 m/s α=.8 db/λ ρ=1.9 g/cm 3 Subbottom 2 5.7 Range (km)
Inversion results for Event 2 Bartlett Power 1.9.8.7.6.5.4.3.2.1 289.58 289.6 289.62 289.64 289.66 Range (km) 5.5 5 4.5 4 3.5 3 2.5 2 1.5 1.5 289.58 289.6 289.62 289.64 289.66 Depth (m) 1 2 3 4 5 6 7 8 9 1 289.58 289.6 289.62 289.64 289.66 Sensordepth (m) 95 94 93 92 91 9 89 88 87 86 85 289.58 289.6 289.62 289.64 289.66 Speed in sed. (m/s) 17 168 166 164 162 16 158 156 154 152 289.58 289.6 289.62 289.64 289.66 Sediment thickness (m) 14 12 1 8 6 4 2 289.58 289.6 289.62 289.64 289.66 Speed in sub bottom (m/s) 19 185 18 175 17 165 16 155 289.58 289.6 289.62 289.64 289.66 Array tilt (rad).4.3.2.1.1.2.3 289.58 289.6 289.62 289.64 289.66 α 1 2 15 1 5 5 1 15 2 289.58 289.6 289.62 289.64 289.66 2 15 1 5 α 2 5 1 15 2 289.58 289.6 289.62 289.64 289.66
Depth (m) Physical model - NW/NE track 156 1518 m/s VA 6 Source (63 m) 119 Sediment 2. m 165 m/s α=.8 db/λ ρ=1.9 g/cm 3 α=.8 db/λ ρ=1.9 g/cm 3 18 m/s Subbottom 2 5.7 Range (km)
Inversion results for Event 5 1 5.5 1.9 5 2.8 4.5 3 Bartlett Power.7.6.5.4.3 Range (km) 4 3.5 3 2.5 2 Depth (m) 4 5 6 7.2 1.5 8.1 1 9 29.94 29.96 29.98 291 291.2.5 29.94 29.96 29.98 291 291.2 1 29.94 29.96 29.98 291 291.2 95 94 17 168 14 Sensordepth (m) 93 92 91 9 89 88 87 Speed in sed. (m/s) 166 164 162 16 158 156 Sediment thickness (m) 12 1 8 6 4 86 85 29.94 29.96 29.98 291 291.2 154 152 29.94 29.96 29.98 291 291.2 2 29.94 29.96 29.98 291 291.2 19.3 2 Speed in sub bottom (m/s) 185 18 175 17 165 16 Array tilt (rad).2.1.1.2 α 1 15 1 5 5 1 15 155 29.94 29.96 29.98 291 291.2.3 29.94 29.96 29.98 291 291.2 2 29.94 29.96 29.98 291 291.2 2 15 1 5 α 2 5 1 15 2 29.94 29.96 29.98 291 291.2
Tide evolution and VA moving
Depth (m) 12 13 14 15 16 17 18 Temperature ( o C) 12 13 14 15 16 17 18 Temperature ( o C) 1 1 8 XBT Thermistors 8 XBT Thermistors 6 Depth (m) 6 4 4 2 2 Estimated temperature profiles - Event 2 & 5
NRP D. Carlos I Overall length (m) 68 Beam (m) 13 Gross displa. (ton) 28 Two diesel-electric (HP) 8 Max speed (kn) 11 Crew 34 Scientific 15
Militar Long. (km) x 124 126 128 13 132 134 149 x ULVA 2:14 Militar Lat. (km) 15 3:14 2:59 151 Event 6 x 152 x x x x 2:44 153 Event 2 NW leg Event 5 NE leg Event 6: bathymetry and source run Depth (m) 13 12 11 1 9 8 7
NRP D. Carlos I speed and heading 14 (a) Ship speed (kn) 12 1 8 6 4 2 291.9 291.1 291.11 291.12 291.13 291.14 291.15 291.16 4 (b) Bearing (deg) 3 2 1 291.9 291.1 291.11 291.12 291.13 291.14 291.15 291.16 Julian date
Depth (m) Physical model and search parameters 6 119 Source (3 m) Sediment 156 1518 m/s 2. m 165 m/s 18 m/s VA α=.8 db/λ ρ=1.9 g/cm 3 α=.8 db/λ ρ=1.9 g/cm 3 Symbol Unit Search int./steps α1 m/s -2 2 64 α2 m/s -2 2 64 sr km.5 3.5 64 sd m 1 1 32 rd m 85 95 32 θ rad -.3.3 32 Subbottom 2 5.7 Range (km)
185 1 2 3 4 5 6 7 Frequency (Hz) 19 Power (db) 195 2 291.1 291.17 291.114 291.121 291.128 291.135 2 3 291.15 291.1 1 2 3 4 5 6 7 Frequency (Hz) 25 Frequency (Hz) 359 4 291.11 291.115 291.125 291.12 25 2 195 19 185 18 5 7 6 db NRP D. Carlos I radiated noise
Inversion results for Event 6 Bartlett Power 1.8.6.4.2 291.12 291.11 291.118 291.126 291.134 Range (km) 3.5 3 2.5 2 1.5 1.5 291.12 291.11 291.118 291.126 291.134 Depth (m) 1 2 3 4 5 6 7 8 9 1 291.12 291.11 291.118 291.126 291.134 Sensordepth (m) 95 9 85 291.12 291.11 291.118 291.126 291.134 Array tilt (rad).3.2.1.1.2.3 291.12 291.11 291.118 291.126 291.134 α 1 2 15 1 5 5 1 15 2 291.12 291.11 291.118 291.126 291.134 α 2 2 15 1 5 5 1 15 2 291.12 291.11 291.118 291.126 291.134 Depth (m) 2 4 6 8 1 291.12 291.11 291.118 291.126 291.134 1516 1514 1512 151 158 156 154
Linear broadband data model: Theoretical background Y (θ) = [Y T (ω1) Y T (ωl)] T = H(θ)S + U Minimum variance signal estimator: Ŝ(ˆθ) = [(H H (ˆθ)H(ˆθ)] 1 H H (ˆθ) N 1 Y n(θ) n= Broadband conventional processor: Pinc(θ) = K k=1 Ŝ(ω k) 2 H H (ωk, θ)ĉy Y (ωk, ωk)h(ωk, θ) H(θ)Ŝ 2
Frequency (Hz) 2 3 4 5 6 7.1.2.3 Likelihood.4.5.6.7.8.9 1 k=1 V (ω, k) K v(ω) = K Likelihood measure: V (ω, k) = 1 T T [Y (ω, t, k) µ Y (ω, k)] 2 dt, Variance at the k th hydrophone: Frequency selection
α 2 Sensordepth (m) Bartlett Power 1.8.6.4.2 95 9 85 2 15 1 291.112 291.117 291.123 291.128 291.134 291.112 291.117 291.123 291.128 291.134 5 5 1 15 2 291.112 291.117 291.123 291.128 291.134 13 14 15 16 17 18 Temperature o C 1 in focus off focus XBT 8 Depth (m) 6 4 2.3 2 291.112 291.117 291.123 291.128 291.134 291.112 291.117 291.123 291.128 291.134.2 15 Array tilt (rad).1 5 1.1 α 1 5 1.2 15.3 2.5 1 291.112 291.117 291.123 291.128 291.134 291.112 291.117 291.123 291.128 291.134 1 1.5 Range (km) 2 Depth (m) 2.5 3 2 3 4 5 6 7 8 9 1 3.5 Event 6: Inversion results
27 359 719 2 3 4 5 6 7 Frequency (Hz) 291.11 291.115 291.12 291.125 291.13 Event 6: Ship power spectrum estimation
Summary This is a preliminary test for ship noise tomography in a (assumed) unknown environment. strong correlation between ship speed/emitted power and accurate source tracking when in focus geometrical and environmental parameter assume credible values more extensive data for oceanographic features observation enhanced search of ship noise spectral components