elet2002(1)

Upload: krishnaveni-subramani-s

Post on 01-Jun-2018

226 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/9/2019 elet2002(1)

    1/3

    -1 o

    -0.5

    0

    0.5

    1 o

    time, ns

    I

    I

    -1.0 I I I I 1 I 1

    Fig 1 Near-peak values of an incident

    t)

    with parameters in/2 z =

    13.48 GHz and w,,/271=

    18.01

    GHz

    100

    I

    .

    predicted

    .

    .

    .

    .

    .

    .

    0 0.005 0.01 0 0.015

    depth,

    m

    Pig. 2 Normalised energies and peak powers of a pulse in

    tD

    and another

    in

    cL

    overlap (centre), are e x p ( - 2 r x ) (top), and are

    ?

    e x p ( - 2 r x )

    (bottom)

    O O L500

    Debye

    /

    ,+

    , Lorentz

    4

    1 1 ' 1 1 ' ' ' 1 1 '0 60 8 100

    frequency, GHz

    Fig 3

    Curves k,(o) or models

    cD

    and

    cL,

    with exponential decay ofpe aks

    verijied by eight examples described n text regarding rectangles

    Going beyond the analysis, a more sensitive measure of exponential

    decay is

    - I /€

    = [ln (x)]/dx. This is times the x-dependent

    slope (in dB/cm) of (x) when graphed in dB. Thus, - I / € is the local

    exponential decay rate of . For the pulses of Figs. 1 and 2, the

    computed - I/aries from 1185-1257/m (51-55 dB/cm) for the

    first

    60

    dB

    of

    energy attenuation. This is in the range [ 2 y y ] f

    101 1485/m (4 4 6 4 dB/cm) suggested but not yet predicted by

    theory.

    Fig. 3 shows ranges of local exponential decay rates

    - P / / P

    of peaks

    P for the first 60dB of peak-power attenuation. The large, marked

    rectangle, e.g. signifies that the incidentfwith spectrum 40-70 GHz has

    a peak that decays in the Lorentz model

    c

    at an exponential rate that

    varies from 709-868/m. The dashed curve shows this is within the

    range

    [T,r]

    f

    508-1247/m for

    E L .

    This example and the seven

    others in Fig. 3 verify the 5 e x p ( - r x ) prediction for peaks. The

    spectrum for the small, marked rectangle is 4-6 GHz. Linearity and

    Fig. 3 thereby yield

    two

    four-parameter families of examples of peaks

    that decay faster than ex p( -r .r ), with bandwidths over three and four

    octaves.

    Conclusions: In this Letter a practical model

    of

    pulses in far-field,

    lossy materials is used to show that exponential decay is typical. This

    is verified by many numerical examples, with bandwidths up to four

    octaves. Two other numerical observations have not yet been

    explained by analysis. First, analysis has not explained why

    e x p ( - 2 y x ) is a bound in Fig. 2. Secondly, the local exponential

    decay rates of energies and peaks are within bounds suggested, but not

    yet predicted, by analysis. The analytical results in this Letter,

    however, are all numerically verified.

    A consequence of recent, practical interest is that algebraic decay no

    longer seems to be a useful design principle for radar penetration of far-

    field, lossy clutter. Although algebraic decay might be recovered from

    exponential decay in a mathematical limit w in 0, this would not

    escape the real difficulties of low radiation efficiency and low resolution

    posed by near-DC signals.

    Acknowledgments: The

    US

    Air Force Office of Scientific Research

    supported this work. H. Steyskal contributed many useful conversa-

    tions.

    EE

    2002

    Electronics Letters Online No: 20020482

    DOI: 10.1049/el:20020482

    T.M. Roberts (Antenn a Technology Branch, Air Force Research

    LaboratorylSNHA,

    80

    Scott Dr ive, Hanscom AFB, M A

    01

    731 , U SA )

    E-mail:

    robertst@maxwell rl plh af mil

    28 February 2002

    References

    1 BRILLOUIN L.:

    'Wave propagation and group velocity' (Academic Press,

    1960)

    2

    OUGHSTUN K.E. and

    SHERMAN,

    G.c.: Electromagnetic pulse propagation

    in causal dielectrics' (Springer-Verlag, 1994)

    3

    KELBERT

    M. and SAZONOV 1.: 'Pukes and other wave processes in fluids'

    (Kluwer Academic, 1996)

    4 ROBERTS T.M.

    and PETROPOULOS P.G.: 'Asymptotics and energy

    estimates for electromagnetic

    pulses in

    dispersive media', Opt Soc

    Am. A 1996, 13, (6), pp. 12041217

    High voltage pulse generation

    M. Petkovsek, J. Nastran,

    I .

    Voncina, P. Zajec, D. Miklavcic

    and

    G. Sersa

    A new topology of a high voltage source with a variable output pulse

    pattern and featuring an independent adjustment

    of

    the magnitude,

    repetition frequency and pulse duration is presented. The power stage

    of the source consists of eight individual unipolar sources that can be

    arbitrarily connected in series to obtain the desired output voltage

    pulse

    of

    several amps and with extremely high

    duldt.

    Introduction: Owing to the tremendous increase

    of

    applications in

    oncology, genetics and cell biology, high voltage pulse sources

    capable of delivering AC

    or

    DC currents of several amps are the

    subject

    of

    intensive investigations. Several authors have reported that

    the application of short high voltage pulses transiently increases the

    permeability of the cell membrane

    [1-31,

    The so-called electropora-

    tion, or electropermeabilisation, has become an effective tool for the

    internalisation

    of

    various molecules, especially anti-cancer drugs and

    gene material , into the biological cells. The efficiency of such

    680 ELECTRONICS LETERS 4th July2 2 Vol 38 No 74

    mailto:[email protected]:[email protected]

  • 8/9/2019 elet2002(1)

    2/3

    -1 o

    -0.5

    0

    0.5

    1 o

    time, ns

    I

    I

    -1.0 I I I I 1 I 1

    Fig 1 Near-peak values of an incident

    t)

    with parameters in/2 z =

    13.48 GHz and w,,/271=

    18.01

    GHz

    100

    I

    .

    predicted

    .

    .

    .

    .

    .

    .

    0 0.005 0.01 0 0.015

    depth,

    m

    Pig. 2 Normalised energies and peak powers of a pulse in

    tD

    and another

    in

    cL

    overlap (centre), are e x p ( - 2 r x ) (top), and are

    ?

    e x p ( - 2 r x )

    (bottom)

    O O L500

    Debye

    /

    ,+

    , Lorentz

    4

    1 1 ' 1 1 ' ' ' 1 1 '0 60 8 100

    frequency, GHz

    Fig 3

    Curves k,(o) or models

    cD

    and

    cL,

    with exponential decay ofpe aks

    verijied by eight examples described n text regarding rectangles

    Going beyond the analysis, a more sensitive measure of exponential

    decay is

    - I /€

    = [ln (x)]/dx. This is times the x-dependent

    slope (in dB/cm) of (x) when graphed in dB. Thus, - I / € is the local

    exponential decay rate of . For the pulses of Figs. 1 and 2, the

    computed - I/aries from 1185-1257/m (51-55 dB/cm) for the

    first

    60

    dB

    of

    energy attenuation. This is in the range [ 2 y y ] f

    101 1485/m (4 4 6 4 dB/cm) suggested but not yet predicted by

    theory.

    Fig. 3 shows ranges of local exponential decay rates

    - P / / P

    of peaks

    P for the first 60dB of peak-power attenuation. The large, marked

    rectangle, e.g. signifies that the incidentfwith spectrum 40-70 GHz has

    a peak that decays in the Lorentz model

    c

    at an exponential rate that

    varies from 709-868/m. The dashed curve shows this is within the

    range

    [T,r]

    f

    508-1247/m for

    E L .

    This example and the seven

    others in Fig. 3 verify the 5 e x p ( - r x ) prediction for peaks. The

    spectrum for the small, marked rectangle is 4-6 GHz. Linearity and

    Fig. 3 thereby yield

    two

    four-parameter families of examples of peaks

    that decay faster than ex p( -r .r ), with bandwidths over three and four

    octaves.

    Conclusions: In this Letter a practical model

    of

    pulses in far-field,

    lossy materials is used to show that exponential decay is typical. This

    is verified by many numerical examples, with bandwidths up to four

    octaves. Two other numerical observations have not yet been

    explained by analysis. First, analysis has not explained why

    e x p ( - 2 y x ) is a bound in Fig. 2. Secondly, the local exponential

    decay rates of energies and peaks are within bounds suggested, but not

    yet predicted, by analysis. The analytical results in this Letter,

    however, are all numerically verified.

    A consequence of recent, practical interest is that algebraic decay no

    longer seems to be a useful design principle for radar penetration of far-

    field, lossy clutter. Although algebraic decay might be recovered from

    exponential decay in a mathematical limit w in 0, this would not

    escape the real difficulties of low radiation efficiency and low resolution

    posed by near-DC signals.

    Acknowledgments: The

    US

    Air Force Office of Scientific Research

    supported this work. H. Steyskal contributed many useful conversa-

    tions.

    EE

    2002

    Electronics Letters Online No: 20020482

    DOI: 10.1049/el:20020482

    T.M. Roberts (Antenn a Technology Branch, Air Force Research

    LaboratorylSNHA,

    80

    Scott Dr ive, Hanscom AFB, M A

    01

    731 , U SA )

    E-mail:

    robertst@maxwell rl plh af mil

    28 February 2002

    References

    1 BRILLOUIN L.:

    'Wave propagation and group velocity' (Academic Press,

    1960)

    2

    OUGHSTUN K.E. and

    SHERMAN,

    G.c.: Electromagnetic pulse propagation

    in causal dielectrics' (Springer-Verlag, 1994)

    3

    KELBERT

    M. and SAZONOV 1.: 'Pukes and other wave processes in fluids'

    (Kluwer Academic, 1996)

    4 ROBERTS T.M.

    and PETROPOULOS P.G.: 'Asymptotics and energy

    estimates for electromagnetic

    pulses in

    dispersive media', Opt Soc

    Am. A 1996, 13, (6), pp. 12041217

    High voltage pulse generation

    M. Petkovsek, J. Nastran,

    I .

    Voncina, P. Zajec, D. Miklavcic

    and

    G. Sersa

    A new topology of a high voltage source with a variable output pulse

    pattern and featuring an independent adjustment

    of

    the magnitude,

    repetition frequency and pulse duration is presented. The power stage

    of the source consists of eight individual unipolar sources that can be

    arbitrarily connected in series to obtain the desired output voltage

    pulse

    of

    several amps and with extremely high

    duldt.

    Introduction: Owing to the tremendous increase

    of

    applications in

    oncology, genetics and cell biology, high voltage pulse sources

    capable of delivering AC

    or

    DC currents of several amps are the

    subject

    of

    intensive investigations. Several authors have reported that

    the application of short high voltage pulses transiently increases the

    permeability of the cell membrane

    [1-31,

    The so-called electropora-

    tion, or electropermeabilisation, has become an effective tool for the

    internalisation

    of

    various molecules, especially anti-cancer drugs and

    gene material , into the biological cells. The efficiency of such

    680 ELECTRONICS LETERS 4th July2 2 Vol 38 No 74

    mailto:[email protected]:[email protected]

  • 8/9/2019 elet2002(1)

    3/3

    -1 o

    -0.5

    0

    0.5

    1 o

    time, ns

    I

    I

    -1.0 I I I I 1 I 1

    Fig 1 Near-peak values of an incident

    t)

    with parameters in/2 z =

    13.48 GHz and w,,/271=

    18.01

    GHz

    100

    I

    .

    predicted

    .

    .

    .

    .

    .

    .

    0 0.005 0.01 0 0.015

    depth,

    m

    Pig. 2 Normalised energies and peak powers of a pulse in

    tD

    and another

    in

    cL

    overlap (centre), are e x p ( - 2 r x ) (top), and are

    ?

    e x p ( - 2 r x )

    (bottom)

    O O L500

    Debye

    /

    ,+

    , Lorentz

    4

    1 1 ' 1 1 ' ' ' 1 1 '0 60 8 100

    frequency, GHz

    Fig 3

    Curves k,(o) or models

    cD

    and

    cL,

    with exponential decay ofpe aks

    verijied by eight examples described n text regarding rectangles

    Going beyond the analysis, a more sensitive measure of exponential

    decay is

    - I /€

    = [ln (x)]/dx. This is times the x-dependent

    slope (in dB/cm) of (x) when graphed in dB. Thus, - I / € is the local

    exponential decay rate of . For the pulses of Figs. 1 and 2, the

    computed - I/aries from 1185-1257/m (51-55 dB/cm) for the

    first

    60

    dB

    of

    energy attenuation. This is in the range [ 2 y y ] f

    101 1485/m (4 4 6 4 dB/cm) suggested but not yet predicted by

    theory.

    Fig. 3 shows ranges of local exponential decay rates

    - P / / P

    of peaks

    P for the first 60dB of peak-power attenuation. The large, marked

    rectangle, e.g. signifies that the incidentfwith spectrum 40-70 GHz has

    a peak that decays in the Lorentz model

    c

    at an exponential rate that

    varies from 709-868/m. The dashed curve shows this is within the

    range

    [T,r]

    f

    508-1247/m for

    E L .

    This example and the seven

    others in Fig. 3 verify the 5 e x p ( - r x ) prediction for peaks. The

    spectrum for the small, marked rectangle is 4-6 GHz. Linearity and

    Fig. 3 thereby yield

    two

    four-parameter families of examples of peaks

    that decay faster than ex p( -r .r ), with bandwidths over three and four

    octaves.

    Conclusions: In this Letter a practical model

    of

    pulses in far-field,

    lossy materials is used to show that exponential decay is typical. This

    is verified by many numerical examples, with bandwidths up to four

    octaves. Two other numerical observations have not yet been

    explained by analysis. First, analysis has not explained why

    e x p ( - 2 y x ) is a bound in Fig. 2. Secondly, the local exponential

    decay rates of energies and peaks are within bounds suggested, but not

    yet predicted, by analysis. The analytical results in this Letter,

    however, are all numerically verified.

    A consequence of recent, practical interest is that algebraic decay no

    longer seems to be a useful design principle for radar penetration of far-

    field, lossy clutter. Although algebraic decay might be recovered from

    exponential decay in a mathematical limit w in 0, this would not

    escape the real difficulties of low radiation efficiency and low resolution

    posed by near-DC signals.

    Acknowledgments: The

    US

    Air Force Office of Scientific Research

    supported this work. H. Steyskal contributed many useful conversa-

    tions.

    EE

    2002

    Electronics Letters Online No: 20020482

    DOI: 10.1049/el:20020482

    T.M. Roberts (Antenn a Technology Branch, Air Force Research

    LaboratorylSNHA,

    80

    Scott Dr ive, Hanscom AFB, M A

    01

    731 , U SA )

    E-mail:

    robertst@maxwell rl plh af mil

    28 February 2002

    References

    1 BRILLOUIN L.:

    'Wave propagation and group velocity' (Academic Press,

    1960)

    2

    OUGHSTUN K.E. and

    SHERMAN,

    G.c.: Electromagnetic pulse propagation

    in causal dielectrics' (Springer-Verlag, 1994)

    3

    KELBERT

    M. and SAZONOV 1.: 'Pukes and other wave processes in fluids'

    (Kluwer Academic, 1996)

    4 ROBERTS T.M.

    and PETROPOULOS P.G.: 'Asymptotics and energy

    estimates for electromagnetic

    pulses in

    dispersive media', Opt Soc

    Am. A 1996, 13, (6), pp. 12041217

    High voltage pulse generation

    M. Petkovsek, J. Nastran,

    I .

    Voncina, P. Zajec, D. Miklavcic

    and

    G. Sersa

    A new topology of a high voltage source with a variable output pulse

    pattern and featuring an independent adjustment

    of

    the magnitude,

    repetition frequency and pulse duration is presented. The power stage

    of the source consists of eight individual unipolar sources that can be

    arbitrarily connected in series to obtain the desired output voltage

    pulse

    of

    several amps and with extremely high

    duldt.

    Introduction: Owing to the tremendous increase

    of

    applications in

    oncology, genetics and cell biology, high voltage pulse sources

    capable of delivering AC

    or

    DC currents of several amps are the

    subject

    of

    intensive investigations. Several authors have reported that

    the application of short high voltage pulses transiently increases the

    permeability of the cell membrane

    [1-31,

    The so-called electropora-

    tion, or electropermeabilisation, has become an effective tool for the

    internalisation

    of

    various molecules, especially anti-cancer drugs and

    gene material , into the biological cells. The efficiency of such

    680 ELECTRONICS LETERS 4th July2 2 Vol 38 No 74

    mailto:[email protected]:[email protected]