System for optical time domain multiplexing of digital signals

Details System for optical time domain multiplexing of digital signals System for optical time domain multiplexing of digital signals

1998-05-11

2001-07-17

Pascal, Leslie (Department: 2633)

Optical: systems and elements

Deflection using a moving element

Using a periodically moving element

C359S199200, C359S199200, C359S199200, C359S199200

Reexamination Certificate

active

06262823

ABSTRACT:

FIELD OF THE INVENTION
This invention relates generally to optical signaling, and more particularly to time multiplexing optical signals.
BACKGROUND OF THE INVENTION
Single mode, fiber-optic cables have an inherent theoretical bandwidth that is more than a thousand times higher than that of electrical cables. One optical fiber could potentially carry a data stream at about ten Tera-bits per second. This is more than the total bandwidth of electronic switches used to interconnect multiple computers (hosts or “nodes”) in high end servers, such as Digital's Memory Channel, IBM's SP-2 family of super-computers, and SGI's Cray T3E. Therefore, systems of “nodes” interconnected by fiber-optic cables are ideal for a very high performance system area networks (SAN), network of workstations (NOW), multi-processor communication subsystems, Internet switches, routers, and the like.
However, data are generally produced by electronic systems and need to be delivered at their destination in electronic form. Consequently, the theoretical bandwidth of a fiber-optic cable is not directly attainable because of the limitation imposed by the need to convert between signaling in the electrical domain and signaling in the optical domain. This is analogous to trying to pump water through a fire hose that is connected to straws at each end.
The general structure of an optical communication system
100
is depicted in FIG.
1
. At a transmitting side, multiple, independent data streams, each of which operates at a speed attainable by electronic means, for example a source node
10
, are provided on multiple input channels
101
. An optical multiplexer
110
combines the multiple data streams onto a single optical transmission medium
115
, generally a fiber-optic cable. At the receiving side, a demultiplexer
120
separates the combined data steam into multiple, independent data streams that are subsequently delivered to their intended destination
20
via output channels
102
.
It should be noted that both the multiplexer
110
and the demultiplexer
120
may actually include a number of independent components, each responsible for only one data stream. It is further possible to combine or split the signal on the transmission medium
115
into several parts. In the case where the signals are split, each part contains all of the combined information, hence passively distributing an optical signal to multiple destinations is functionally equivalent to broadcasting.
FIG. 2
shows the schematic for the equivalent broadcasting system
200
with input channels
201
, multiplexers
210
, optical broadcast medium
215
, demultiplexers
220
, and output channels
202
. The electronic nodes are not depicted.
To overcome the intrinsic limitations of the electro-optical interfaces, it is possible to multiplex multiple, independent data streams onto one fiber by using multiple different wavelengths, or by interleaving data in time. The first technique is called wavelength division multiplexing (WDM), and the second technique is called optical time domain multiplexing (OTDM). There are also other forms of optical multiplexing techniques under investigation, for example, ones based on spread-spectrum techniques, however, WDM and OTDM are the only techniques that have currently been reduced to practical applications.
Wavelength division multiplexing is similar to a conventional broadcast system. There, the electromagnetic spectrum is divided into multiple channels with a different frequency assigned to each channel. The transmitters and receivers can selectively operate on specific channels while sharing the same transmission medium without interference. Fiber-optic transmission systems using WDM have been demonstrated with up to 128 channels.
However, WDM systems are rather costly because they require highly frequency-stable lasers on the transmitting side, and very selective filters on the receiving side. Moreover, at this point in time, WDM systems either cannot change channels at all, or require a relatively long time to tune the transmitters and receivers to different frequencies, as long as many milli-seconds.
As an advantage, WDM system can operate over very long distances, and are ideal for wide area networks (WANs), telephony, cable TV systems, etc. WDM systems are not compelling at the system area network level due to their high cost, a limited number of channels, and their limited ability to switch channels. Because WDM systems cannot easily reallocate bandwidth to meet rapidly changing demands, they are best suited in situations with fairly static communication patterns.
FIG. 3
is a timing diagram of an optical domain multiplexed broadcast signal
300
for an example eight node network. A system that can produce the signal
300
is described below with reference to FIG.
4
. In
FIG. 3
, time is indicated along the x-axis
301
, and the relative intensity (amplitude) of the signal
300
is indicated along the y-axis
302
. In the broadcast signal
300
, pulses
310
,
320
,
330
,
340
, and so forth, are called framing pulses, and pulses
311
-
318
are data pulses, i.e., there are eight data pulses for each framing pulse.
Logical ones and zeros are respectively indicated by solid and open data pulses, i.e., the presence or absence of light pulses. The framing pulses are always ones. Generally, the framing pulses have a greater intensity than the data pulses so that they can readily be discerned. The rate of the framing pulses determines the bit rate of the network. The relative offset of the data pulses with respect to the framing pulses determines the OTDM channels to which transmitters and receivers can tune.
As shown in
FIG. 4
, the data from many sources are sequentially interleaved on a single medium, for example, an optical fiber. Because the system operates with a fixed bit rate (frequency), the time between two consecutive bits from the same source is a constant: the bit-time. The bit-time is the inverse of the transmission rate of each channel.
The data from all channels are transmitted sequentially, starting with the first bit of channel “0” (
311
), the first bit of channel “1” (
312
), etc. After the first bit from all channels have been sent, the cycle repeats with the next bit, and so on. The total number of channels in the system is a constant. The bit-time multiplied by the number of channels determines the total bandwidth of the system (total bit rate (TBR). The framing pulses delineate the channel multiplexing sequence. For example, each framing pulse may precede the data from channel “0.” Each receiving node in the system can use the framing pulses to synchronize its selection mechanism based on a relative time offset from the framing pulses.
Because of the speeds involved, multiplexing and demultiplexing must be performed optically. Each transmitting node must insert a data bit in its assigned channel time slot, similarly, each receiver extracts its data from one of the channel time slots. As an advantage, an OTDM system can utilize many channels, and it is possible to change channels quickly, for example, in a few nano-seconds as opposed to many milli-seconds for an WDM system. In other words, changing channels in an OTDM system can be about a million times faster than changing channels in a WDM system). As is the case for a WDM system, the multiplexing and demultiplexing functions can be distributed such that each device that is interconnected has one multiplexing and one demultiplexing device as part of its network interface. Such a configuration is shown in FIG.
4
.
FIG. 4
shows the basic structure of an optically coupled network
400
that can produce the multiplexed signals as shown in FIG.
3
. This system is disclosed in U.S. Pat. No. 5,493,433 issued to Prucnal et al. on Feb. 20, 1996. Framing pulses as shown in
FIG. 3
are generated by a single modelocked, pulse compressed laser source
413
of a “hub”
410
. The pulse rate is equal to the bit rate that is used by each of the attached nodes, for example, 1.3 Gbits/sec. The pulse width of each light burs

Affiliated with

Also associated with

Rating

System for optical time domain multiplexing of digital signals does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with System for optical time domain multiplexing of digital signals, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System for optical time domain multiplexing of digital signals will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2484728