MLS
theory basics
Maximum Length Sequence

Resources

MLS
in a nutshell...
MLS
is an abbreviation for Maximum Length Sequence. It is
basically a pseudorandom sequence of pulses.
Nowadays Maximum
Length Sequence
measurements are quite standard in many different application
fields. One of them is acoustics.
Using MLS techniques, it is possible to perform quasianechoic
measurements of a loudspeaker without having to place
it inside an anechoic
chamber (a room free from echoes and reverberations). The impulse
response can be easily windowed in the time domain, in order
to analyze the signal and reject the reflections from the walls
of the room.
Moreover
the room impulse response itself (and all the related
parameters such as reverberation time) can be measured.
The
MLS method can also be used to analyze
and obtain information about the impedance or the absorption
coefficient of a surface.
The Norder MLS sequence is periodic with period (2^N)1.
Different MLS sequences of the same order can exist.
They can be easily obtained by using a shiftregister with different
feedback taps.
The
MLS signal can be used to measure the response of any type of
LTI (Linear Time Invariant)
system. The
impulse response of the system can be easily obtained
by computing the crosscorrelation between input and output
signals.
For
LTI systems, the frequency response can
thus be computed by performing
a FFT of the windowed impulse response.
The
MLS technique
has many advantages when compared with other methods
of measuring the response
of a system.
Among them the following:
 The
MLS has a quasiflat power spectrum. The spectrum envelope
follows a square(sin(x)/x) law and falls by about
1.6 dB at 1/3 of the sampling
rate.
 MLS
technique
rejects
the DC component of the sampled
signal.
 MLS
measurements have a very
high Signal/Noise
ratio.
The crosscorrelation used to compute the impulse response
reduces
all background noise (uncorrelated with MLS), so that measurements
can be performed also in noisy environments. The use of averaging
techniques can furhter increase the S/N ratio.
 The
measured distortion of the system is spread throughout the
computed impulse response. Every MLS sequence has his own
characteristic distorsion pattern: more measurements on the
same system with different MLS sequences (of the same order)
allow an easy recognition of the distortions.
Some
precautions must be taken when using the MLS method:
 The
MLS signal length must be longer than the impulse response
of the system under test or
have the same length.
If these conditions are not satisfied, some parts of the computed
impulse response will overlap (timealiasing).
 The
system under test must be timeinvariant, at least during
the measure interval.
More
information and references about MLS measurement theory can
be found in the bibliography
section.
This
page is also available for download as PDF file
