CA2186817A1 - Optical communication system utilizing photonic patterns - Google Patents

Abstract

A system for transmitting data from a plurality of channels from a first location to at least another location utilizes photonic patterns for transmission encoding and reception decoding of the data. Optical signals corresponding to data channels are each encoded with a different photonic pattern. The signals are then modulated and are then multiplexed into a single signal for transmission on a single waveguide to a distant location. At the distant location after the signal is received, a photonic pattern recognizing circuit, recognizes the particular photonic patterns and detects/decodes the data.

Description

21~61317 An Optical Communication System Utilizing Photonic Patterns Field of the Invention 5 This invention relates generally to optical communications systems and more particularly to a system of transmitting and receiving data utilizing photonic patterns.

Background of the Invention 10 Using optical signals as a means of carrying channelized information at high-speeds through free space or within optical waveguides is preferable over schemes such as microwave links, coaxial cables, and twisted copper wires since propagation loss is lower, and optical systems are immune to the Electro-Magnetic Interference (EMI), and have higher channel capacities. These high speed optical systems have signaling 15 rates of several mega-pulses per second to several tens of giga-pulses per second.
Optical communication systems and fiber optic networks are fast-growing and are nearly ubiquitous in communication networks. The expression herein "Optical communication system" and "Fiber optics systems" relates to any system that usesoptical signals at any wavelength to convey inforrnation between two points by any 20 means. Optical systems and networks are described in Gower, Ed. Optical communication Systems, (Prentice Hall, NY) 1993, and by P.E. Green, Jr in "Fiberoptic networks" (Printice Hall New Jersey) 1993, which are incorporated here as reference.

25 High speed (data) signals are plural signals that are formed by the aggregation of several data sources that share a transmission medium which send their data to adestination at a distant location. In optical communication technology there are two fundamental multiplexing schemes: Time Division Multiplexing (TDM) and Wavelength Division Multiplexing (WDM).

In TDM, data from each data channel is transmitted periodically in a pre-specified 5 time-slot and in a predetermined time-frame. The tr~n~mis~ion speed of a TDM system can be increased by using narrow pulses with high peak powers. However, TDM
systems impose a technological problem. In a TDM system, at a first stage of demultiplexing, electro-optic switches must operate at a speed of the total data rates;
However, currently, it is not practicable to operate electro-optic circuits efficiently at 10 rates higher than 20 Gbit/sec, thus, limiting the number of high-speed channels. Also in TDM systems a clock signal at the highest data rate is required to synchronously multiplex and demultiplex signals at these high data rates.

In WDM each high speed data channel transmits its information at a particular 15 wavelength, for example, ~i. A plurality of data channels each being assigned a predetermined wavelengths, are multiplexed and are launched into an optical waveguide. At a receiver end, channels are separated by narrow band optical filters and then detected or used for further processing. In practice, the number of channels in a WDM system are limited as a result of crosstalk, limited bandwidth of optical 20 amplifiers and/or optical fiber non-linearities. Moreover such systems require an accurate band selection, stable tunable lasers or filters, and spectral purity that increase the cost of WDM systems and add to their complexity.

25 Space division multiplexing, is a scheme in which individual optical channels are physically separated in space by tr~nsmis~ion on separate waveguides, for example, optical ribbons. This approach suffices for short distances but tends to be costly and in some instances undesirable. For instance, in supercomputer interconnections a word of data to be transmitted at a very high speed must be sent parallel and simultaneously with other data words through an optical fiber to prevent bit skew and differential latency. Optical ribbons are not appropliate for this application for various reasons such as the unavoidable difference between the physical parameters such as dispersion 5 coefficients within different fibers in a ribbon. Thus it would be preferable to transmit many parallel bits through a single optical waveguide with an alternative technology.

Notwithstanding, other factors must also be considered.. For example it is preferable for a plurality of different users to be able communicate with each other as directly 10 and quickly as possible. Transmission, in principal, has been based on the use of TDM and/or WDM technology. Networks typically use coding and protocols for communication and are designed to allow users to communicate with each other.
Hence large number of independent parallel channels are often dedicated for userconnections and simplicity in networking. There are many versions of optical network 15 architectures which generally apply one of the above mentioned fundamental multiplexing methods, TDM or WDM. Some of the approaches for addressing a network user involve wavelength routing using wavelength conversion or Code Division Multiple Access (CDMA). Wavelength routing network architectures are complex and costly, on the other hand, Code Division Multiple Access can provide a 20 large number of (virtual) channels, nevertheless, this is not a bandwidth efficient way of communication within a network and eventually the speed of a network will degrade. In time domain CDMA (TD-CDMA) each user receives its respective data ina pre-specified format of pulses. As the number of users increase the length of the code increases to transmit each bit of information. This, in turn, significantly decreases the 25 data rate of each (user) channel. The same is true for the CDMA codes employing WDM technology.

In summary, for transmission of high speed plural signals and efficient optical networking it is desirable to have a large number of simultaneous parallel high speed independent channels that can be easily separated after being combined. Given the present commercially available technology, TDM and WDM cannot fulfill this objective inexpensively. As a result, high speed optical networking progress has been limited and prototype systems remain costly.
s It is therefore an object of the invention to provide a system and method for carrying many high-speed channels in a relatively simple and cost-effective manner.

Summary of the Invention 10 In accordance with the invention there is provided, a method of transmitting data from a plurality of channels from a first location to at least another location comprising the steps of:
a) providing a plurality of optical signals each optical signal corresponding to at least one of the plurality of data channels, each having at least a different photonics pattern;
15 b) mo~lul~ting the optical signals;
c) multiplexing the optical signals into at least a first multiplexed optical single signal;
d) transmitting the multiplexed optical signal to at least another location;
e) recognizing the photonic patterns corresponding to each channel from the multiplexed optical signal and, detecting the data;
In accordance with the invention there is further provided, a method of transmitting data from a first location to another location comprising the steps of:
a) providing a plurality of optical signals b) passing each of said optical signals through a photonic pattern encoder to encode 25 each of the signals with a different photonic pattern.
b) mo(lnl~ting the optical signals;
c) multiplexing the optical signals into a multiplexed optical signal;
d) transmitting the optical signal to another location;

21 ~681 7 e) recognizing the photonic patterns at the other location and decoding the data within each of the optical signals In accordance with another aspect of the invention there is provided, an opticalS communications system comprising:
a photonic pattern encoder for encoding data at a transmitting end; and, a photonic pattern or decoder for decoding data at a receiving end.

In accordance with the invention there is provided a system for transmitting data from 10 a plurality of channels from a first location to at least another location comprising:
a) means for providing a plurality of optical signals each optical signal corresponding to at least one of the plurality of data channels, each having at least a different photonics pattern;
b) means for mocl~ ting the optical signals;
15 c) means for multiplexing the optical signals into at least a first multiplexed optical single signal;
d) transmitting means for transmitting the multiplexed optical signal to at least another location;
e) means for recognizing the photonic patterns corresponding to each channel from 20 the multiplexed optical signal and, for detecting the data.

This invention provides a method for communication that can carry a plurality optical signals for example, M channels. for over an optical path, where M is a whole number greater than or equal to 2. The communication is provided by utili7.ing photonic25 patterns.

A photonic pattern (PP) is a combination of number of photons of different colors (i.e.
combination of optical powers at different wavelengths). A PP can be represented by a continuous or a discrete diagram which indicates a (normalized) number of photons in for example, a vertical axis (see Fig. 1) versus the wavelength or frequency otherwise referred to as color, along horizontal access. Each optical source or filter has a characteristic wavelength power spectrum, which can be referred to as wavelengthspectrum and herein we refer to as a Photonic Pattern.

The communication system in accordance with this invention comprises a Photonic Pattern encoder and a Photonic Pattern recognizer or decoder. At the transmitter end, each of M channels is encoded with a Photonic Pattern as its unique signature.

10 The Photonic Pattern encodes each channel by N optical sources (1 < N ~M ), and an integrated optic circuit comprising optical waveguide components such as filters, couplers, splitters and multipliers. M Photonic Patterns can be generated from Mdifferent light sources. Conveniently, optical sources may include LED's, lasers, all types of lamps, an or a white source; other light sources regardless of radiation 15 wavelength and the spectral width may be utilized. Integrated optic circuits herein refer to optical waveguide components that can be made by essentially any available technology on glass, semiconductor, electro-optic materials, or any other new material that integrated optical components can be made of. The Photonic Pattern encoder may be static or programmable. When it is programmable the multipliers, filters and 20 couplers are tunable and made from electro-optics materials such that their optical characteristics can be varied by applying an external electrical field.

Each signature is modulated in an On and Off keying (OOK) format by the electrical data channel. The modulation can be direct or external. In direct modulation the output 25 of the optical source is directly controlled by the electrical data within the same device.
An External modulator controls the passage of optical signal through itself. After modulation, the optical channels in the form of Photonic Patterns are combined and are send launched into a single optical waveguide. Separation of the channels is then 21 868l7 performed at a receiver by computing the cross-correlation of an incoming pluralsignal with the known signature of each channel. The decoder recognizes the presence and absence of a Photonic Pattern in the received plural signal and hence the information therein. It is proved that the decoding or recognition of Photonic Patterns 5 is possible by arithmetic linear scalar operations that can be performed by passive optical components. The decoder is conveniently implemented by use of integratedoptic technology or other means for performing Photonic Pattern recognition such as those described hereafter.

10 In this way communication is realized lltilizing a Photonic Pattern code for each channel. Since the number of recognizable Photonic Patterns are endless even for a limited range of wavelengths, a large number of parallel channels can be realized and therefore the total throughput of the optical network and link can be higher than with the aforementioned prior art technologies.
It is also noted that each channel may address a users at a plurality of destinations thus providing networking capability; each user would be able to recognize its own ~igns~tllre ignoring the signature of others.

20 Brief Description of Dl,.willgs:
Exemplary embodiments of the invention will now be described by way of the accompanying drawings, in which:

Fig. 1 a to ld graphically illustrate four different examples of Photonic Patterns 25 diagrammatically;

Fig. 2 is a Schematic representation of an optical communication system having Mdata channels according to the present invention, Fig. 3. diagrammatically shows direct modulation of M sources and shaping their corresponding Photonic Patterns;

Fig. 4 is an external modulation circuit for multiplexing M data channels with MPhotonic Patterns into a single waveguide;

Fig. S is circuit for producing MPhotonic Patterns from a single light source;

Fig. 6 is a block diagram of an MPhotonic Pattern generator from N input patterns;

Fig. 7 is a graphical layout for generating M Photonic Patterns from N Photonic input patterns using optical splitters or couplers;

Fig. 8 is a diagram of receiver node of an optical channel demultiplexer with balanced detectors and low noise electrical amplifiers for recognizing presence or absence of a Photonic patterns in an incoming optical signal comprising a plurality of channels encoded by plural patterns;

Fig. 9 is a schematic diagram of an integrated optic portion of the Photonic Pattern recognizer shown in Fig. 8; and, Figs. lOa to lOd are graphs of samples of Photonic Patterns for a four channel communication system that can be obtained with the Photonic Patterns of waveguide directional couplers with different length in the wavelength range of 1530 nm to 1560.

Detailed Description Channel Multiplexing and propagation through an optical waveguide link s Referring now to Fig. la, lb, lc and ld examples of Photonic Patterns are shown diagrammatically. A Photonic Pattern indicates a number of photons of each colorpresent in a group of photons of an optical signal at any instant of time. Referring to Fig. 2, M (M>2) high speed electronic data channels are shown. Each channel has a 10 Photonic Pattern as its signature. Thus, each channel transmits its data by switching "on" and "off" its corresponding Photonic Pattern.

At the receiver end, a Photonic Pattern recognizer recognizes the presence of a Photonic Pattern in the incoming plural signal and decodes the data of the 15 corresponding channel.

Once the shape of each of the Photonic Patterns is known, each corresponding to one of a plurality data channels, and a plurality of photons is received, recognition is possible and a determination is made as to whether any Photonic Pattern corresponds 20 to a known pattern related to that particular channel. As a means of explanation, and for simplicity, consider a system having only one color and for example, four separate channels wherein data corresponding to the four channels is to be transmitted along a single waveguide simultaneously. Further, let us assume that there is advanced knowledge that channel one always transmits its data by sending only one photon 25 corresponding to a one binary bit (correspond to logic 1 ) and no photons for a zero binary-bit. Let us further assume that channel two sends two photons for a one-bit and no photons for the zero-bit. Channel three kansmits four photons for a one-bit and no photons for a zero-bit and channel four transmits eight photons for a one-bit and nothing for a zero-bit. In this system if the receiver receives 3 photons it can be assumed that only channel one and two have sent a one-bit. If 4 photons are received, it can be assumed that only channel three could have sent these photons and so on.
Furthermore, it is evident that channel 3 cannot send 3 photons for its one-bit high or that channel four cannot send 5 or 6 or 7 photons as its unique ~ign~tllre, otherwise 5 one could not uniquely determine among the plurality of received photons, channels they correspond to..

In this example, the number of photons of each subsequent channel are a power of 2.
However, in general one can choose any number of photons as the signature of its10 channels as long as none of them can be expressed as the summation of other channels' assigned number of photons. Further, in order to increase the number of channels, the number of photons corresponding to certain channels become large and consequently these channels must transmit at high energy to represent their data.
Alternatively, to increase the number of the channels while maintaining the bit energy 15 of the channels at a relatively low level, photons of different colors can be used. For example if another color is used with another 4 level (1,2,4 and 8 photons) the system can distinguish 16 different bi-color channels.

Of course, this scheme can be extended to a continuous color spectrum with an endless 20 combination of numbers of photons of different colors that in the form of Photonic Patterns.

To have M photonics patterns M separate sources with different Photonic Patternscan be utilized, or M optical sources with an identical Photonic Pattern with each of 25 them followed by an optical filter with a different Photonic Pattern, or by using a smaller number of sources and combining the Photonics Pattern to produce other ones.
Optical filters are referred to herein as any optical device that has a characteristic Photonic Pattern. Such optical filters can be static or tunable.

Two methods are provided for constructing a Photonic Pattern communication systems. Referring now to Fig. 3., a method is provided in which direct modulation of the optical signal is employed prior to making a unique Photonic Pattern for each 5 channel. Fig. 3 further shows an instance where the sources are directly modulated with the electronic data and their Photonic Pattern is shaped by a set of optical filters with Photonic Patterns offj(i=l...M). The light source may be a wide band or narrow band. Typically narrowband sources in the form of laser diodes can be used such as those commercially available from Hewlett Packard, Fujitsu, New Focus etc.
10 Alternatively, wide band sources may be used, such as light emitting diodes (LEDs) Yet alternatively, a white light source can be used. . High power wide-band sources may be provided in the form of optical amplifiers that produce large amplified spontaneous emission ASE. Turning now to Fig. 4, an alternate scheme is shown inwhich the pre-processed unique Photonic Patterns are modulated externally. The 15 External modulators modulate the optical signal as opposed to modulating the optical source as is done in the direct modulation system. External modulators are manufactured in the form of bulk and integrated optics with different configurations from electro-optic and electro-absorption materials such as lithium niobate, LiNbO3, and Multiple Quantum Wells (MQW) as described in New Focus, Inc., "Practical Uses 20 and applications of Electro-Optic Modulators," Application Note 2.

In general, a channel's Photonic Pattern can be static or dynamic. When it is static the shape of Photonic Patterns are fixed. Alternatively, in a dynamic system the shape of the patterns can be changed with time.
In the case of pre-processed Photonic Patterns, statically or dynamically, for the external modulation, two methods are provided to obtain a particular set of photonic Patterns:

1. Refering now to Fig. 5, a single source with a Photonic Pattern of gO(~) is provided.
The pattern is shaped independently by M optical filters to make M new Photonic Patterns gl to gM. . f j indicates the Photonic Patterns of the optical device that shape the input Photonic Patterns.
5 In this case Wti(~)= psgo(~)fti(~)= psgi(~) i=1,2,...M and ~o-~S~S~o+A~ (1) where W,j is the Photonic Pattern of channel i, ~0 is the central wavelength, A~ is linewidth spectral width of the Photonic pattern in wavelength domain, and Ps is a constant that is proportional to the signal actual power.
10 Mathematically each Photonic Pattern, i.e. W,j, can be regarded as a vector (or a point) in a Hilbert space.

2. Referring to Fig. 6, a small number of base Photonic Patterns N are used, where N
is greater than one and equal to or smaller than M. M new Photonic Patterns are produced by a linear combination of the initial N Photonic Patterns. This method is 15 suitable when a large number of filters cannot be used, or when it is desirable to use a programmable Photonic pattern generator. In this instance the channel Photonicpatterns can be expressed by N

wti(~)= ps~Aijgoi(~) Psgi(~) i=1,2,...M (2) j=l 20 where A jj are elements of a norm~li7~d MxN matrix that will produce M new Photonic pattern from N input. This matrix is expressed by -Al I, ..... AI N
A21, A22, A2N (3) AMI ~ - - - - - - - - - - AMN

Where A is a norm:~li7e~1 matrix its elements being smaller than unity. Optical filters and linear combiners and A jj are static or dynamic coefficients, which may all be implemented by passive waveguide technology. Fig. 7 shows a layout for 5 implementation and realization of relations (2) and (3). Each input Photonic Pattern is split to at most Mbranches, from each input photonic pattern one branch is connected to the block of multipliers. Each multiplier block has at most N inputs, each of them is multiplied by the corresponding factor (A j,), and they are combined to make one of the W,j. The elements of Matrix A can be an optical coupler or splitter with one uncoupled 10 end. Thus, the output of each coupler or splitter is the input multiplied by a factor which is less than one. The multipliers can be programmable using tunable couplers that are made of electro-optics materials.

The optical device for adding the signals and may be waveguide directional couplers 15 or grating couplers or any device capable of adding several optical signals into a one single optical beam.

The plural signal may then be amplified and transmitted to a destination along an optical path which may be guided (e.g. fiber optic links) or unguided (e.g. free space 20 propagation including communication between satellites). The amplifier may be any type of optical amplifier capable of operating in the operating wavelength region.
These include semiconductor amplifiers and rare earth ion doped waveguide amplifiers such as erbium, neodymium, praseodymium, ytterbium, or mixtures thereof.

25 The plural signal at the kansmitter end, WA~T~ is represented by M

WMT(~ qti(~)Wti(~) i=1,2,.. M (4) i=l where q,j(~ ) is the time domain data signal corresponding to channel I and is expressed by:

qti(T) = di(T)s(T,~Tti) (5) where dj(~ ) is the binary data (i.e. 0 or 1) and sj(~, A~,j ) is the shape of the pulse in the time domain which is sent for the logic 1, and ~,j is the initial pulse-width at half maximum at the transmitter end. The multiplexed signal that contains the information of M channels propagates within the optical fiber (i. e. the link) between the two 10 neighboring nodes. At the receiver end a circuit is required to separate and demulitiplex the channels. If the peak power level of the multiplexed signal is low enough then the propagation of multiplexed signal will be in the linear regime and therefore the propagation of each Photonic pattern through the fiber link with length L
can be considered separately. The Photonic pattern of channel i, wrjJ after propagation 15 and at the destination can be expressed by Wri(~T) = Wti(~)hi(~)qri(T) i=1,2,...M (6) where hj(~) is the spectral response of the fiber of length L in the wavelength domain 20 and qrj(~ ) is the pulse shape at the receiver that is expressed by qri(T) = di(T)s(l, Alri) (7) which is similar to the (5) except that the pulse-width at the receiver may havechanged due to the wavelength dependent nature of the optical path due to wavelength dispersion. When optical fibers are used as the optical path, the fiber link is managed 25 to have a zero dispersion for the operating wavelength range. Disperson can be lessened or minimi7ed by splitting the fiber link to two sections each section having a same dispersion coefficient but with an opposite sign.

- Demultiplexing Fig. 8 shows, an optical circuit for recognizing Photonic Patterns and detecting the 5 presence of particular Photonic Patterns corresponding to a data channel, at a receiver end. When the received plural optical signal contains Photonic Patterns corresponding to channel no. i and a predetermined noise threshold is exceeded, the signal being deemed other than noise, an electrical signal is present at the output of the electrical amplifier indicating the presence of Photonics Pattern no. i in the incoming plural 10 optical signal.

Referring to the Fig. 9, for recognition a Photonic Pattern in the incoming plural signal, the incoming signal is first split into K branches each followed by an optical filter, where K is greater ~N where an optical filter is disposed at each branch . Then 15 the filtered signal at each branch is split again to at most M branches. The output of each branch is then multiplied by a factor B~k. Referring to Fig. 8, the output of all the positive and negative factors are added together separately from which the summation of negatively or positively multiplied signals is directed to a corresponding photodetector in the balanced receiver.

Referring to Fig. 9, at the receiver another M set of Photonic patterns are synthesized in terms of the Photonic Patterns of the filters in the receiver, so that we have, K

Wrl (~) = Pr ~ Blk frk (~) 1=1,2,. . .M (8) k=l 25 where B,k are normalized constant or variable coefficients andfrk(~ ) are the Photonic Patterns of the filters in the receiver side (Refer to Fig. 9) and wrl are the M new of Photonic patterns synthesized from the incoming multiplexed signal by the optical filters and passive multipliers at the receiver. It is desirable for this set of Photonic Patterns to closely resemble the original incoming set of Photonic patterns.
Mathematically to measure their resemblance we can calculate their cross-correlation:

(X' N K
R~ ) = Wri Wrl = qri(~ Aijgoj(~)h(~)~Blkfrk(~
_~o i= I k= I
i,l= 1,2,.., M (9) N K o~
= qri (1)~ Aij ~ Blk Jgo j (~)h(~) frk (~)d~
i=l k=l _~o The cross-correlation matrix S then can be written R (I ) = [ A . G . B ~ ]. dia g ( qi(l )) (10) 10 where A is an M x N matrix introduced in (3), G is a N x K which is given by _ J go1(~) h (~) frl(~) d~,........................ .- - -- Jgol (~)h(~) frK (~)d~
oo oo oo o~ o~
Jgo2()~)h(~)frl(~)d)~ Jgo2(~)h(~)fr2(~)d~ Jgo2(~)h(~)frK(~)d~
o~ _o~ --oo o~ oo JgoN (~)h(~) frl (~)d~,.......................... ...--- - - JgON (~)h(~) frK (~)d~
oo oo (1 1) l S and B is a MxK matrix and is given by Bl l, .. - - - - - -BIK
B = B21 ,B22 ,--------B2K (12) BMI - - - - - - - - BMK

Therefore the cross-correlation matrix R(~) is an MXM matrix. In order to recover the unique data corresponding to each channel R(T) should be equal to the diag(q,(l)). I, S where I is an MxMunity matrix. Hence, we must have ¦AGBT=1 ¦

(13) 10 This provides a design tool for calculating the matrices A and B and consequently the individual values of the multiplication coefficient in the Figs. 7 and 9. As long as this relation is satisfied the channels can be successfully demultiplexed without any cross-talk. Thus, this relation provides a guide line for selecting each channel's Photonic Pattern and that of the receiver filters, i.e. ~(~) and frk(~). Thus, if (13) is satisfied the 15 Photonic Pattern reconstruction of channels is successfully without any cross-talk from other channels. In other words photons of the Photonic Patterns of the other channels have no effect on any other channel.

Without loss of generality we can put A=l then the relation (13) takes a simpler form:
IG.BT _ 11 (14) In this instance we are only using the N basic Photonic patterns ,i.e., gj(~)=gOj(~).The number of optical filters in the receiver end,frk(~), can still can be arbitrary as long as 25 relation (14) is satisfied. If N filters are used at the receiver both G and B become NxN matrices. In this case as long as matrix G has an inverse we can demultiplex the channels (i.e. separate the energies of different Photonic patterns). Relation (14) shows that the ith row of G is orthogonal to all the columns of BT except the column i.

5 Turning to Fig. 9, for implementation of relation (14) it can be noticed that some of the elements of matrix B are negative. Although energies and photons are non-negative quantities, electronics can be used to provide the required minus sign.Therefore, a differential detector is utilized in which all the positive elements are directed to one photodiode while all the negative elements directed to an other 10 photodiode Now with reference to Fig. 8 This circuit in fact utilizes an opto-electronic means for recognition of the Photonic Patterns the incoming signal iscomposed of.

In summary, and in terms of mathematics the invention provides a communication 15 system wherein for each channel a point exists in a multi-dimensional space with a particular set of basis functions. It is noted that the same point in space can be mapped by other sets of basis functions. The point randomly becomes on and off according to the data of the channel. The plural signal subsequently carries the messages of several of these points. At the receiver end demultiplexing of the channels is performed by 20 recognizing if the point is on or off by recognizing the location of the all the points that exist in the incoming plural signal.

If the elements of matrix B differ by a small amount, relation (14) cannot be completely satisfied and crosstalk will exist between the different channels. If Bo is the 25 theoretical value of matrix B that can satisfy relation (14) and B are the values yielded after implementation then we can show B as following B=Bo+~ (15) where ~ represents the deviation of B from its theoretical value. The error that will introduce the cross-talk is E = G . ~

5 the cross-talk from other channels to channel i is defined by M

Qi = ~ Eij - Eii (16) j=l the cross-talk in dB then is defined by l+E
Q~(dB)= lOlog1o M (17) ~ Eij Eii to minimi7e the cross talk from other channels in relation 16 we have to minimi7~ the Gij = min imum possible (for example << 1) i ~ j . (18) ii 1 5 where x Gij = ¦gi(~)h(~)frj(~)d~ (19) --x Equation (18) is another guide-line for designing the Photonic patterns g(~)s andf(~)s 20 to make the communication system robust against the changes in the matrix B. When the optical fiber link is modeled with an attenuator, relation (19) then implies that the optimum choice forfrj (~) is to be exactly similar in shape to the base PhotonicPatterns, g~ ) in the general case, i.e. formula (13), or to the Channel's Photonic Patterns gj(~) in the case of formula (14).

5 A simple method for designing the Photonic Patterns To design the Photonic Patterns we first determine several points for each Photonic pattern and then make a Photonic Pattern curve as close as possible to the points.

Photonic Pattern Photonic pattern Photonic pattern Photonic pattern Photonic pattern (PP) No. value at the value at the value at the value at the wavelength ~, wavelength ~2 wavelength ~3 wavelength ~4 PPI forChl, g~(~) 1 2 4 8 PP2 forCh2,g2(~) 2 4 8 PP3 forCh3,g3(~) 4 8 1 2 PP4 for Ch4, g,(~) 8 1 2 4 Table 1.

Table one shows how points are created for a four channel system. In general if there 15 are M channels, M points should be determined with different values for the Mdifferent wavelengths (colors). The first choice of M values yield the first Photonic Pattern. Subsequent Photonic Patterns are created by shifting and circulation of the M
values to the next wavelength, as shown in table one.

20 The first M values are albilldly however it is advantageous to utilize values capable of shaping corresponding Photonic patterns that can be implemented easily with the Photonic Patterns of optical sources and filters. For example if it is desired to produce Photonic Patterns by using optical couplers, then it is better to select these M values on a sinusoidal curve, because the Photonic Patterns (wavelength spectrum) of couplers varies sinusoidally with wavelength (color).

Referring to Fig. 10, a set of Photonic Patterns is shown that are designed based on the above method and the Photonic Patterns that can be produced from optical waveguide couplers having different lengths.

10 Applications:

1) Any high-speed point to point data communication with several channels. That includes the data links between the major routing and switching nodes, conveying a word of high-speed computer in parallel form.
2) Optical multiple access applications [12] either point to multi-point or broadcasting and all the application that have been recognized for optical Code Division Multiple Access networks [13,14].

20 3) Using the optical communication system in different network topologies such as ring, star and tree configuration or any other that uses optical signal transmission using this techniques for making the plural signal or separate them by the provided theory [7,8,12].

25 4) This theory and the optical circuits in Fig. 8 can also be used for processing and recognition of Photonic Patterns of any optical source that carry information like the application fiber-optics array sensors [15].

This invention provides an optical communication system based on Photonic Pattern encoding and recognition.

An embodiment of the invention further provides a single broadband Continuous wave 5 optical source in conjunction with a number of filters with an arbitrary spectrum and a number of static/programmable arithmetic linear operations, performed by optical and electro-optical waveguide components, to make the desired Photon Pattern, i.e. data channel signature.

10 With a small number of sources a large number of signatures, i.e. large number of channel to address a large number of destinations in an optical network.

This invention further provides is a circuit for recognizing Photonic Patterns. The circuit can be manufactured from optical and electro-optical waveguide components.
Of course, numerous other embodiments may be envisaged without departing from the spirit and scope of the invention.
References 1. Gower, Ed. Optical Communication Systems, (Prentice Hall, NY) c. 1983, pp.1-25.
2. "Fiber Optic Networks", Paul E Green, Jr. Prentice Hall Inc. New Jersey, 1993, pp.3-19 3. Hatami-Hanza, A. Mostof1, and P.L. Chu, "A multilevel soliton communication system," in 8th IEEE LEOS 1995 Annual Meeting, San Francisco, USA, 1995, vol. 2,paper OCS3.4, pp. 35-36.

4. L. A. Bergman, A. Mendez, and L. S. Lome, "Bit-Parallel Wavelength Links for High Performance Computer Networks," SPIE Critical Review, vol. CR62, p210, Jan.96.

5. R. M. Gagliardi and A. J. Mendez, "Performance Improvement with hybrid WDM
and CDMA optical communication," Proceeding SPIE vol. 2690, 1996, pp. 88-96.

6. Zou et al," Limitation in 10 Gb/s WDM optical-fiber transmission when using variety of fiber types to manage dispersion and nonlinearities," IEEE Journal ofLightwave Technology, Special issue on Multiwavelength Optical Technology and Network June 1996, pp. 1127-1135.

7. See all the papers in the Journal of Selected area of communication Special issue on optical networking, June 1996.

8. See all the paper in the IEEE Journal of Lightwave Technology, Special issue on Multiwavelength Optical Technology and Network June 1996.

9. New Focus, Inc., "Practical Uses and applications of Electro-Optic Modulators,"
Application Note 2.

10. "Fiber Optic Networks", Paul E Green, Jr. Printice Hall Inc. New Jersey, 1993, pp.

11. "Fiber Optic Networks", Paul E Green, Jr. Printice Hall Inc. New Jersey, 1993, pp.

12. "Fiber Optic Networks", Paul E Green, Jr. Printice Hall Inc. New Jersey, 1993, pp.
417-489.

13. J. Salehi, "Code division multiple-access techniques in optical fiber networks-Part I: Fundamental Principles," IEEE Trans. Commun. vol. 37, no. 8, pp. 824-833, Aug.
1989.

14. J. Salehi and C. Brackett, "Code division multiple-access techniques in optical fiber networks-Part II: System perforrnance analysis," IEEE Trans. Commun. vol. 37, no. 8, pp. 834-842, Aug. 1989.
M. H. Maher, K. Tabrizi, J. D. Prohaska, E. Snitzer, Fiber Bragg gratings for Civil Engineering Applications," SPIE vol. 2682, pp. 298-302, 1995.

Claims (4)

1. A method of transmitting data from a plurality of channels from a first location to at least another location comprising the steps of:
a) providing a plurality of optical signals each optical signal corresponding to at least one of the plurality of data channels, each having at least a different photonics pattern;
b) modulating the optical signals;
c) multiplexing the optical signals into at least a first multiplexed optical single signal;
d) transmitting the multiplexed optical signal to at least another location;
e) recognizing the photonic patterns corresponding to each channel from the multiplexed optical signal and, detecting the data.

2. A method of transmitting data from a first location to another location comprising the steps of:
a) providing a plurality of optical signals b) passing each of said optical signals through a photonic pattern encoder to encode each of the signals with a different photonic pattern.
b) modulating the optical signals;
c) multiplexing the optical signals into a multiplexed optical signal;
d) transmitting the optical signal to another location;
e) recognizing the photonic patterns at the other location and decoding the data within each of the optical signals.

3. An optical communications system comprising:
a photonic pattern encoder for encoding data at a transmitting end; and, a photonic pattern or decoder for decoding data at a receiving end.

4. A system for transmitting data from a plurality of channels from a first location to at least another location comprising:
a) means for providing a plurality of optical signals each optical signal corresponding to at least one of the plurality of data channels, each having at least a different photonics pattern;
b) means for modulating the optical signals;
c) means for multiplexing the optical signals into at least a first multiplexed optical single signal;
d) transmitting means for transmitting the multiplexed optical signal to at least another location;
e) means for recognizing the photonic patterns corresponding to each channel from the multiplexed optical signal and, for detecting the data;