Harmonic Analysis: Six-Segment Downhole Acoustic Waveguide - Measurement While Drilling Signal Analysis, Optimization and Design - page 48

2.1 MWD Fundamentals
In conventional Measurement-While-Drilling operations, the problems
associated with surface reflections, e.g., those at the mudpump, the desurger, the
rotary hose, and so on, are well known. Several echo cancellation schemes
using delay line models based on single and multiple transducers are available
for surface signal processing, these importantly acknowledging the perils and
subtleties associated with acoustic wave motions. These problems will be
compounded at higher data rates because they are necessarily associated with
higher frequencies and much shorter wavelengths. A single pressure transducer
installed arbitrarily on the standpipe is not likely to be optimal; careful planning
and the use of transducer arrays may be required. Many surface signal
processing ideas are appreciated by MWD designers, but perhaps are not as well
understood as they should be. As we will find in this topic, important and useful
models for forward and inverse applications can be developed using rigorous
solutions to the one-dimensional acoustic wave equation.
The technical and patent literature has generally focused on the
consequences of reflections as they affect surface signal processing. However,
as described in the chapter introduction, they are equally prominent downhole in
the way that they affect signal creation at higher data rates. For example,
consider a mud siren (or other pulser type) transmitting information and
employing a phase-shift-keying (PSK) scheme, but at carrier frequencies much
higher than those presently used. The “usual” wave will propagate uphole to the
surface with encoded data and it is this wave that is addressed in surface signal
processing. However, a downgoing wave is also created at the source that
reflects “at the drillbit” with amplitude and phase changes that depend on mud
sound speed, pulser location and the details of the bottomhole assembly and
annulus. This reflected wave (still “long” in the sense that its wavelength
greatly exceeds a typical diameter) passes through the mud siren (or other
pulser) and combines with later created waves that are moving upwards.
The net signal that travels up the drillpipe contains both the “usual” wave
and the more complicated reflected one - it is impossible to determine, at the
surface using present signal processing methods, which phase-shifts are real and
which are not. These “ghost reflections” cannot be eliminated without
additional information - these are not problematic if they are not created in the
first place, of course, and Chapter 10 suggests a practical solution using an
alternative “FSK” telemetry scheme. Because ghost echoes do exist,
improvements to existing continuous wave telemetry over the years have been
limited. This observation suggests the use of telemetry schemes which are more
robust from a downhole source oriented perspective - and perhaps, one that can
additionally increase transmission distance as well as data rate - by harnessing
the use of constructive wave interference. These applications suggested the
development of a math model that can be used in both telemetry job planning
and in high-data-rate tool design.