Browse Prior Art Database

Method for Increasing the Tuning Rate of a Filter with Inertia

IP.com Disclosure Number: IPCOM000106440D
Original Publication Date: 1993-Nov-01
Included in the Prior Art Database: 2005-Mar-21
Document File: 4 page(s) / 153K

Publishing Venue

IBM

Related People

Sivarajan, KN: AUTHOR [+2]

Abstract

In some optical WDM networks, the receiver at a station scans all possible wavelengths periodically in order to determine whether there is a connection request for it on any of them and, if a connection request is detected, stops the scan and attempts to lock onto that wavelength. If the scanning is done sufficiently, rapidly the filter overshoots due to its inertia and locks onto a nearby wavelength. The proposed method enables faster scanning with continued ability of locking to the correct transmission despite the filter's inertia.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 31% of the total text.

Method for Increasing the Tuning Rate of a Filter with Inertia

      In some optical WDM networks, the receiver at a station scans
all possible wavelengths periodically in order to determine whether
there is a connection request for it on any of them and, if a
connection request is detected, stops the scan and attempts to lock
onto that wavelength.  If the scanning is done sufficiently, rapidly
the filter overshoots due to its inertia and locks onto a nearby
wavelength.  The proposed method enables faster scanning with
continued ability of locking to the correct transmission despite the
filter's inertia.

      In some optical WDM networks, e.g., RAINBOW [*], the optical
filter at a receiver scans all possible transmit wavelengths in order
to determine whether the transmission on any of them is intended for
it and then locks onto that wavelength.  In order to be able to tune
rapidly from one wavelength to another it is desirable to scan as
fast as possible.  The principal limitation on fast scanning is the
inertia of the optical filter; e.g., the mechanical inertia of the
piezo-electrically tuned Fabry Perot filter in RAINBOW.  Let N be the
number of wavelengths in the network and assume one is scanning from
wavelength 1 through N and has to lock to wavelength i.  If one stops
the scan and attempts to lock the filter on detecting the connection
request on wavelength i, because of the inertia of the filter, if one
scans sufficiently rapidly, one might lock to wavelength i+1 or i+2
rather than wavelength i.  It is desirable to have a mechanism by
which one can lock to the correct wavelength in the presence of the
inertia of the filter while scanning as rapidly as possible.  In the
above description, the control of the filter has a feedback loop that
was assumed in order to lock the filter (and keep it locked).  The
same problem tuning of the filter is implemented inan open-loop
fashion.  In addition, this problem is not specific to optical
filters but is present in the case of any filter that has to scan and
then stop but due to the inertia of the filter its response "lags"
the tuning input (voltage in the case of the RAINBOW Fabry Perot
filter).  Nevertheless, for the sake of concreteness, we will
describe our invention in the context of a voltage-controlled optical
filter.

      The method proposes that instead of stopping the scan on
detecting a connection request, the scan is reversed for a short
period of time which depends on the measured/calculated/estimated
inertia of the filter) and then stop and close the loop.  This
enables locking to the correct wavelength while scanning at a much
faster rate as follows.

      Let V(t) be the voltage applied (measured in volts) to the
filter at time t and let x(t) be the wavelength (measured in nm) the
filter is tuned to.  Assume that the filter tunes K nm/volt.  Let the
voltage input to the filter be ramped at R volts/ms and let m = K R.
Typical values of these para...