What Makes a GNSS Signal Great? @ the IGS Analysis Center Workshop 2-6 June 2008, Miami Beach Larry Young (not an expert, just a user) Jet Propulsion Lab.

  • Published on
    18-Jan-2016

  • View
    217

  • Download
    0

Transcript

  • What Makes a GNSS Signal Great?

    @ the IGS Analysis Center Workshop2-6 June 2008, Miami BeachLarry Young(not an expert, just a user)Jet Propulsion Lab

  • AgendaCurrent and planned GNSS signalsCarrier frequency impactCode rate impact (Code vs semi-codeless)BPSK vs BOCData vs pilot

  • Current and planned signalsGPS C/A, P(Y), M, [L1C planned] at L1 (1575.42 MHz)L2C, P(Y), M at L2 (1227.6 MHz)Planned L5 (1176.45 MHz)GLONASS Signals corresponding to C/A, P(Y), near L1 (FDMA)Signal corresponding to P(Y), near L2 (FDMA)Helpful change to CDMMA is plannedGALILEO Rich selection of at least 6 signals at or near L1, L5 (called E5a and E5b), and E6 (1278.75 MHz)We hope all signals will be made available to science usersBeidou, etc not treated here

  • Carrier frequency impactAt least two frequencies are required for accurate work to remove delay/phase advance from the ionosphere. For example, an ion-free combination for pseudorange isPc = [1+(F22)]/ (F12-F22)*P1 - F22/ (F12-F22)*P2In order to reduce the magnification of measurement errors in P1 and P2 by this linear combination, the factor F22/ (F12-F22) should be as small as possible.Carrier phase trilaning: Form dual widelanes with the ambiguities resolved, then form an ionosphere-free linear combination of those. The table (next vg) compares use of L2 vs E6 for the third frequencyAt least one of us favors investigation of a much higher frequency, near 5115 MHz for example, to form very precise carrier phase observables with very small ionospheric effects, enabling compact actively steered arrays that produce multiple beams toward the satellites and null forming toward multipath sources.

  • Trilane frequency comparison(Good choice, ESA!)

    L1 (in units of 10.23 MHz)

    L5

    L2, E6,

    or CS

    Error in ion-free trilane (Assumed carrier error at each freq is 1 mm)

    154

    115

    120 (L2)

    11 cm

    154

    115

    125 (E6)

    6.7 cm

    154

    115

    500 (C-band)

    0.3 cm [OFF-TOPIC for today!]

  • Code rate impact (Code vs semi-codeless)

    Advantages to higher chip rates For a given ratio of signal bandwidth to chip rate, the errors due to system noise are inversely proportional to the chip rate.Multipath errors are less, in some cases, for higher chip rates. (See next VG that shows identical MP for small delays.)We recommend that studies be performed to see to what extent multipath from greater distances affects the estimated parameters at IGS sites.

  • Same MP error for delays normally encountered (needs study)

  • Code rate impact (Code vs semi-codeless) continued

  • Pilot signals, and BPSK vs BOCMany new codes are pilots with no data bits, allowing long coherent integrations for acquisition and tracking of weak signals. BOC and MBOC signals will be exciting to exploitBetter precisionLower MP (for long delays)+

  • Pseudorange MP for 10 MC/s BPSK vs BOC(10,5)-10 dB MP, 32 ns E-L lag spacing

  • Carrier MP for 10 MC/s BPSK vs BOC(10,5) -10 dB MP, 32 ns E-L lag spacing

  • ConclusionThe variety of carrier frequencies will allow new approaches, such as trilaning, to be developedThe new codes, especially high-rate BOC codes such as the E5a and E5b pair will offer unprecedented pseudorange precisionThere is still plenty of room for improvement!

Recommended

View more >