WO2007103410A2

WO2007103410A2 - Multi-channel differentially encoded phase shift keyed signal receivers

Info

Publication number: WO2007103410A2
Application number: PCT/US2007/005778
Authority: WO
Inventors: David O. Caplan
Original assignee: Massachusetts Institute Of Technology
Priority date: 2006-03-08
Filing date: 2007-03-08
Publication date: 2007-09-13
Also published as: WO2007103410A3

Abstract

An optical, multi-channel, Differential Phase Shift Keying (DPSK) receiver demodulates multiple Wavelength Division Multiplexed (WDM) channels using at least one interferometer. This distributes expense of the interferometer(s) over all channels of an optical signal, allowing for deployment of cost-effective, scalable, wideband, WDM DPSK systems. For example, for an 80 channel WDM link, the receiver uses a single interferometer instead of eighty interferometers and associated stabilization hardware, dramatically reducing size, weight, power, and cost. The receiver is architecturally compatible with existing interferometer technologies so previous development and qualification efforts can be leveraged. This allows for expedited technology insertion into existing optical communications networks, including terrestrial and space-based optical networks.

Description

MULTI-CHANNEL DIFFERENTIALLY ENCODED PHASE SHIFT KEYED

RECEIVERS

RELATED APPLICATIONS

This application claims the benefit of U.S. Provisional Application, 60/780,447, filed on March 8, 2006, and is a continuation-in-part of U.S. Patent Application No. 11/022,344, now allowed, filed on December 23, 2004, which claims the benefit of U.S. Provisional Application No. 60/639,183, filed on December 23, 2004. The entire teachings of the above applications are incorporated herein by reference.

GOVERNMENT SUPPORT

The invention was supported, in whole or in part, by a grant under Contract No. F19628-00-C-0002 from the United States Air Force. The Government has certain rights in the invention.

BACKGROUND OF THE INVENTION Recently, optical Differential Phase Shift Keying (DPSK) modulation has received considerable attention by the telecommunications industry primarily due to its increased sensitivity over commonly used On-Off-Keying (OOK) and its reduced peak power, which mitigates nonlinear effects in fiber-optic applications. This has led to demonstrated utility in long haul applications, with experiments confirming more than 3 Tb/s capacity using 80 Wavelength Division Multiplexed (WDM) DPSK channels. It is expected that the first wideband telecommunications fiber optic links using WDM DPSK will be deployed by 2006. DPSK is also an attractive modulation format for high-rate spectrally efficient Free Space Optical (FSO) links because the increase in sensitivity over OOK allows for a corresponding reduction in costly transmitter power.

DPSK conveys information by encoding a relative phase difference between two optical bits. Differential encoding simplifies the receiver by eliminating the need for a stable absolute phase reference that is required for coherent homodyne Phase Shift Keyed (PSK) modulation. Instead of mixing the received signal with a local oscillator phase reference to determine its phase, DPSK mixes the incoming signal with a delayed version of itself. Consequently the DPSK receiver is often referred to as a "self-homodyne" receiver.

SUMMARY OF THE INVENTION DPSK is often used to refer to differentially encoded binary PSK (DBPSK) in which the binary information preferably corresponds to either a 0 or 180 degree (π) relative phase difference, which can be demodulated using a single delay-line interferometer. However, in systems where spectral efficiency is important, more than two relative phases can be differentially encoded. Differential Quadrature Phase Shift Keying (DQPSK), for instance, conveys 2-bits of information per symbol by differentially encoding one of four relative phases, preferably 0, 90, 180 or 270 degrees [0, π/2, π, 3π/2], which can essentially be demodulated using two DBPSK receivers consisting of interferometers shifted by +π/4 and -π/4 so that they are orthogonal (π/2 out of phase). While less sensitive and more complex than DBPSK, DQPSK requires half the bandwidth to convey the same information, (see, e.g., R. A. Griffin, R. I. Johnstone, R. G. Walker, J. Hall, S. D. Wadsworth, K. Berry, A. C. Carter, and M. J. Wale, "10 Gb/s optical differential quadrature phase shift key (DQPSK) transmission using GaAs/AlGaAs integration," in OFC, 2002). The spectral efficiency can be improved further by extending the alphabet to 8- DPSK, 16-DPSK, or more generally 2^k-DPSK, providing 3, 4, and k-bits/symbol respectively, corresponding to a 1/k reduction in bandwidth required to communicate at a given data rate.

The benefits of optical DPSK come at the cost of increased complexity, requiring a phase modulator in the transmitter and optical interferometer(s) and balanced detection in the receiver in order to derive maximum benefit. Of these elements, the interferometer is the most technically challenging and the least mature. Control hardware is necessary to ensure stable operation, which requires that the arms of the interferometer be stable to small fractions of a wavelength. As a result, carefully designed thermo-mechanical packaging is necessary in addition to stabilization electronics, adding to size, weight, power, and cost. An optical, multi-channel, Differential Phase Shift Keying (DPSK) receiver, and corresponding method, employing the principles of the present invention demodulates multiple Wavelength Division Multiplexed (WDM) channels using a single interferometer for DBDSPK and two interferometers for DQPSK. A DPSK receiver using a single interferometer for WDM demultiplexing is achieved by constraining the received wavelength spacing (Δv_Ch) and leveraging the interferometer's periodic transfer function to perform demodulation on all channels (λ's). This distributes the expense of the interferometer over all channels, allowing for deployment of cost-effective, scalable, wideband, WDM DPSK systems. For example, for an 80 channel receiver, size, weight, and power (SWAP), and costs are significantly reduced through use of a single interferometer instead of eighty interferometers and associated stabilization electronics. The receiver is architecturally compatible with existing interferometer technologies, so previous development and qualification efforts can be leveraged. Such leveraging allows for expedited technology insertion into existing optical communications networks, including terrestrial and space-based optical networks.

Accordingly, one embodiment of an optical receiver or corresponding method for demodulating optical signal(s) having DPSK channels according to the principles of the present invention includes a delay line interferometer that (i) demodulates optical signal(s) having DPSK channels to optical signal(s) having channels modulated in intensity, and (ii) outputs the demodulated optical signal(s) onto at least one main output optical path. The receiver also includes channel selectors, such as optical filters or Wavelength Division de-Multiplexers (WDMs), in the main output optical path(s) directing channels onto tributary optical paths. The channels may be predefined wavelengths or wavelength ranges (e.g, 1557.0nm +/- 2.0nm). The receiver may include optoelectronic converters in the tributary optical paths that convert the demodulated optical signal into respective, corresponding, electrical signals.

The delay line interferometer may be a one-bit delay line interferometer, a multiple bit delay line interferometer, or selectably adjustable delay line interferometer to interfere optical signal pulses offset by selectable numbers. The - A -

interferometer may also include an electronically tunable phase shifter for stabilization and/or for receiving channels of different wavelengths.

The interferometer may be adapted to demodulate optical signals having wavelength spacing between carrier wavelengths defining the channels with an integer multiple of a channel rate. For example, for wavelengths on a 100 GHz International Telecommunications Union (ITU) grid (i.e., 100 GHz channel separation), standard Synchronous Optical Network (SONET) rates of 2.5 Gbps and 10 Gbps, for instance, factor evenly into the 100 GHz spacing. Therefore, these rates are compatible with this multi-wavelength DPSK receiver design. In another example, 40 Gbps channel rates requires 200 GHz channel spacing, which is also compatible with the ITU grid and the multi-wavelength DPSK receiver.

The interferometer may also be adapted to demodulate optical signals having wavelength spacing between carrier wavelengths defining the channels evenly divisible by an odd number of half channel rates of the optical signal. For example, 40 Gbps goes into 100 GHz channel spacing exactly 2 ¹A times. The received

Signal-to-Noise Ratio (SNR) remains intact, but the received data is inverted — a condition that can be anticipated or detected and corrected via post processing. In one embodiment, detection electronics perform the post processing by correcting polarity of electrical signals of received optical channels as necessary, which may correspond to predefined optical channels. In another embodiment, optical elements perform the post processing by correcting polarity of the phase demodulated optical signal. With this capability, a multi-wavelength DPSK receiver can receive all channels without any penalty whenever the channel spacing is evenly divisible by the half channel rate. For prior art, single channel, DPSK receivers, the interferometer may track on the single carrier wavelength to optimize demodulator performance and compensate for transmitter or interferometer drift. However, for multiple-channel reception with a single interferometer, the interferometer can only adjust the period (i.e., wavelength or channel separation) and/or shift the entire comb of channels. Thus, a single interferometer cannot track and lock onto arbitrarily spaced wavelengths simultaneously in order to optimize performance. Therefore, the optical receiver according to the principles of the present invention may also include a feedback processor that generates signals transmitted to transmitter(s) of the optical signal(s) to cause the transmitter(s) to tune carrier wavelengths defining the channels. Tuning the carrier wavelengths allows for independent channel adjustment and alignment to the interferometer for improved demodulation performance. With this feedback, independent transmitter wavelengths can be precisely aligned to the interferometer, despite potential drift in the transmitter wavelengths relative to the interferometer. Such feedback, for example, could be used to compensate for laser aging or for Doppler shifts.

In one embodiment, the optical receiver includes a low noise optical amplifier that receives the optical signal(s) and outputs the received, amplified, optical signal(s) to the interferometer, where "optical signal(s)" in this case refers to the collection of one or more WDM channels being received. The low noise optical amplifier may be an Erbium Doped Fiber Amplifier (EDFA) or other optical amplifier known in the art. To support interferometer control, the interferometer or orthogonally phased interferometers may receive a pilot signal or plurality of pilot signals that serve(s) as reference wavelengths that the interferometer(s) can lock-on to for stabilization and/or alignment to incoming optical signal(s). The pilot signal(s) may be locally resident or distributed, may be tunable, and may be selected to be outside the standard communications band so as not to impact the available communications spectrum. As an example, the pilot tone to which the interferometer locks-on can be a dominant incoming channel that may be selectable by the network. With feedback, the interferometer and all other channels including locally resident references can be spaced at wavelengths relative to the dominant signal. In the event the dominant channel drops out, locally resident pilot tone(s) or other incoming channels, which have been aligned relative to the dominant channel can take over as the reference, allowing for continued wavelength alignment and control. The use of multiple pilot tones can be used to improve interferometer control and provide a means of measuring interferometer parameters, such as Free Spectral Range (FSR). The receiver may be used in various applications, such as an optical regenerator. The receiver may also be used in various optical networking environments, such as Free Space Optic (FSO) and fiber optic environments. Within those environments, the networks may be point-to-point networks, mesh networks, ring networks, broadcast networks, multi-access networks, and so forth.

BRIEF DESCRIPTION OF THE DRAWINGS The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of preferred embodiments of the invention, as illustrated in the accompanying drawings in which like reference characters refer to the same parts throughout the different views. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention.

Figs. IA, IB, and 1C are a schematic diagrams of an optical, multi-channel, Differential Phase Shift Keying (DPSK) receiver according to the principles of the present invention;

Fig. 2A is a schematic diagram of an example optical delay line interferometer that may be used in the receiver of Figs. 1 A or 1 B;

Fig. 2B is a plot of sensitivity as a function of normalized carrier frequency offset (e.g., center wavelength) for the interferometer of Fig. 2A and receiver of Figs. IA or IB;

Fig. 2C is a schematic diagram of example optical delay line interferometers that may be used in an orthogonal DPSK receiver such as the quadrature or DQPSK receiver of Fig. 1C;

Fig. 3A is a spectral diagram indicating channel wavelength spacing as an integer multiple of a channel rate that optimizes use of one embodiment of the receiver of Figs. IA or IB; Fig. 3B is a spectral diagram indicating channel wavelength spacing as an integer multiple of a half channel rate that optimizes use of an alternative embodiment of the receiver of Figs. IA or IB;

Fig. 3 C is a spectral diagram indicating targeted wavelength spacing on the ITU grid that optimizes use of an alternative embodiment of the receiver of Figs. IA or IB, illustrating the use of a pilot tone to lock the interferometer to a known reference and wavelength measurement to generate corrective feedback to transmitters of a DPSK optical signal; Fig. 4 is a schematic diagram of network nodes employing the receiver of Figs. IA or IB;

Fig. 5 A is an illustration of a Free Space Optic (FSO) application in which the receiver of Figs. IA or IB is deployed in mobile platforms such as satellites or aircraft; and

Fig. 5B is a network diagram of an optical communications network application in which the receiver of Figs. IA or IB is deployed in network nodes.

DETAILED DESCRIPTION OF THE INVENTION

A description of preferred embodiments of the invention follows. Fig. IA and Fig. I B are schematic diagrams of example embodiments of an optical, multi-channel, binary DPSK receiver 100 according to the principles of the present invention. The receiver 100 receives optical signal(s) 105 from optical transmitter(s) 90. The optical signal(s) 105 may have many DPSK channels defined by distinct wavelengths, λ i,λ2, ... , λ_n. In one embodiment, the optical signal(s) 105 are received by an optical amplifier 110, such as a low noise Erbium Doped Fiber Amplifier (EDFA) 110, that may be separate from the receiver 100 or integrated into the receiver 100. The EDFA 1 10 outputs amplified optical signal(s) 105 to a first optical splitter 115, which directs a portion of the amplified optical signal(s) 105 to a binary DPSK demodulator 199, including an interferometer 120, such as a 1-bit delay line interferometer 120, which receives the amplified signals via a primary input line 1 18. The delay line interferometer 120 is sometimes referred to as a "delay and multiply demodulator." The remaining portion of the amplified optical signal(s) 105 is directed to a wavemeter 170. The receiver 100 may leverage the periodic transfer functions (cos or sin ) from either arm of the interferometer 120 to perform a delay line (e.g., one bit) demodulation on all channels (λ's) with a single interferometer. Such leveraging is preferably performed by defining the wavelength spacing (Δ v_Ch) to be an integer multiple of a channel rate (R).

Δ v_Ch = /wR, where m is a positive integer. (Equation 1)

The interferometer 120 demodulates the optical signal(s) 105 by interfering the received optical signal pulses from each channel with an offset version of itself, where the offset is equal to the interferometer delay. The delay is usually a one bit delay (i.e., interfering adjacent signal pulses), but more generally can be an integer multiple of the one bit delay duration (i.e., interfering non-adjacent pulses). The interference converts the optical signal in each channel from being differentially phase modulated to being intensity modulated. An example of an interferometer that may be used in the receiver 100 is described below in reference to Fig. 2A.

Continuing to refer to Fig. 1 , the interferometer 120 outputs a processed form of the optical signal(s) 105 onto first and second main optical paths 132-a and 132-b, respectively (collectively, optical paths 132). The demodulated optical signal(s) on the two main optical paths 132 are complementary; if the optical signal pulses in each channel constructively interfere (i.e., they are in-phase), the interferometer 120 directs the phase demodulated optical signals onto one main optical path (132-a), and if the optical signal pulses destructively interfere (i.e., they are out-of-phase), the interferometer 120 directs the phase demodulated optical signal onto the other main optical path (132-b).

The optical paths 132 traverse a channel selector network 130 shown as 1 x (n+1) Wavelength Division de-Multiplexers (WDMs) 131-a, 131-b (collectively 131) (e.g., prisms, diffraction gratings, or Arrayed Waveguide Gratings (AWGs) in Fig. IA, and alternatively defined by a series of cascaded fiber Bragg grating (FBG) WDM filter pairs 133-la/133-lb, 133-2a/133-2b, . . . , 133-na/133-nb (collectively, WDM filters 133) in Fig. IB. The WDMs 131 (Fig. IA) or filter pairs 133 (Fig. IB) is/are coupled to respective tributary optical paths 140-la/140-lb, 140-2a/140-2b, . . . , 140-na/140-nb (collectively, tributary optical paths 140).

The tributary paths for each channel are sent to post processing elements 160, which may be all optical, or optoelectronic, which includes elements such as communications electronics (not shown), balanced detection hardware 150, and detection electronics 155, as illustrated in Fig. IB. The communications electronics pass network traffic to network communications systems (Figs. 4, 5A, and 5B). The detection electronics 155 may include peak RF power detection (for signal-to- interferometer alignment), clock recovery, Forward Error Correction (FEC) decoding hardware, and so forth. As shown in Fig. IA, each of the post processing elements 160-1, 160-2, ... 160-n, (collectively 160) may send channel performance information or metrics, such as Bit Error Rate (BER) and peak RF power, to a control processor 125 via bus lines 165-1, 165-2, ..., 165-n (collectively 165). Other inputs to the control processor 125 may include measurements by a wavemeter 170 of a pilot tone 108 and incoming optical signal(s) wavelength information, which can be used along with the BER and peak RF power, among other information or metrics, to determine feedback signal(s) 185 and 188. The feedback signals 185 and 188 are communicated from the control processor 125 to the pilot tone generator 107 and optical transmitter(s) 90 via feedback paths 190 and 195, respectively. The feedback signals 185 and 188 are generated to control the pilot tone 108 and incoming received signal wavelengths, respectively.

For control purposes, an optical pilot tone generator 107 generates an optical pilot or reference tone 108 at wavelength λ_p that may be injected into the interferometer 120 via a secondary input line 119 at an output of a second optical splitter 1 16. A portion of the pilot tone 108 may also be directed via the optical splitter 1 16 to the wavemeter 170 for wavelength measurement. Pilot tone outputs from the interferometer 120 are directed through the channel selector network 130. The a-side and b-side WDM pilot tone outputs on tributary optical paths 140p-a and 140p-b, respectively (collectively 14Op), are directed to a control processor 125, which may measure a contrast between the pilot tone outputs on tributary optical paths 14Op. The measured contrast (D), which is the normalized difference between the pilot tone outputs on the tributary optical paths 140ρ:

D = (140p-a - 140p-b)/(140p-a + 140p-b),

is a function of the pilot tone wavelength λ_p (or equivalently center frequency) and the interferometer bias (e.g., interferometer phase), which is discussed in further detail below in reference to Figs. 2A and 2B. For a given pilot tone wavelength, which may either be known or measured, the pilot tone contrast ratio is a function of the interferometer bias. Therefore, sending pilot tone outputs on the tributary optical paths 14Op to the control processor 125 enables the pilot contrast to be determined, which can provide feedback parameters that may be used to measure and the control interferometer 120, as discussed above. Referring to Fig. IB, each of the tributary optical paths 140 undergo balanced detection by balanced detectors 150, the output of which is electrically coupled to detection electronics 155-1, 155-2, .... 155-n (collectively 155). Balanced detection can be accomplished in many ways, for example, using balanced detector pairs that output a difference photocurrent directly or using discrete photodetectors that output their respective photocurrents to electronic elements, such as a differential amplifier, which subsequently performs the differencing. For wavelengths on the International Telecommunications Union (ITU) grid (e.g., 100 GHz channel separation), Time Division Multiplexed (TDM) architecture standard Synchronous Optical Network (SONET) rates of 2.5 Gbps and 10 Gbps, for instance, factor evenly into the 100 GHz ITU spacing. Therefore, these standard rates and wavelength spacing are compatible with this multi-wavelength DPSK receiver design. In another example, 40 Gbps channel rates requires 200 GHz channel spacing, which is also compatible with the interferometer 120 in the optical receiver 100 and the ITU grid.

If, however, the channel spacing is evenly divisible by an odd number of half channel rates, for example, 40 Gbps goes into 100 GHz channel spacing exactly 2 ¹A times, the received Signal-to-Noise Ratio (SNR) remains intact, but the received data is inverted on every other channel (see also Fig. 3) — a condition that can be anticipated or detected and corrected via a polarity corrector 162-1 (e.g., a conditional inverter) in the post processing electronics 160 or optically inverted prior to detection by suitable optical element(s) 162-2, 162-3. An example of such an optical element is a controllable delay line interferometer in which the relative optical phase between the two arms can be switched by approximately a half- wavelength of the carrier frequency, or odd multiples thereof. The interferometer

120, for instance, could act as an optical DPSK inverter in this manner, but since it is processing all channels simultaneously, it cannot perform polarity correction selectively. With polarity correction capability, a multi-wavelength DPSK receiver can receive all channels without any penalty whenever the channel spacing is evenly divisible by the half channel rate (R/2), shown in the equation below.

Δv_ch = /wR/2, with polarity correction (m is a positive integer). (Equation 2) Referring to Fig. IB, the multichannel binary DPSK demodulators of Fig IA and IB can be extended to accommodate multichannel orthogonal DPSK signals such as quadrature DPSK or DQPSK signals using two binary DPSK demodulators 199-a and 199-b, with interferometers 120-a with a phase offset of π/4 and 120-b with a phase offset of -π/4. Further detail regarding the interferometer phase offset is given in Fig. 2C.

Fig. 2A is a schematic diagram of a representative delay line interferometer 120 composed of a first 50% optical splitter 210 that accepts primary and secondary inputs 118, 1 19 and splits them equally between two internal paths or "arms" 205, 206 with relative time delay τ. The optical signals in the two paths 205, 206 are recombined with a second 50% optical splitter 220. The interferometer 120 (i) may be constructed as a Mach-Zehnder (as shown) or Michelson interferometer, (ii) may be waveguide, free-space, or fiber based, and/or (iii) may include Faraday rotator elements to achieve polarization independence. Fig. 2B graphically illustrates signal loss for wavelength misalignment (i.e., the carrier wavelength is offset from the optimum wavelength) given polarity correction. The transfer functions of the two interferometer output arms 132a, 132b are periodic and complementary, following a cos²(πΔf/FSR+Δφ) and sin²(πΔf/FSR+Δφ) dependence, respectively. The biasing term Δφ is a measure of the relative optical phase between the internal interferometer arms and is typically controlled to maximize the signal interference in one or both of the output arms 132a, 132b. When the bias Δφ = 0, the term Δf is the frequency deviation from optimum alignment of the incoming signal to the interferometer, which is periodic. FSR is the interferometer Free Spectral Range, which is the interferometer spectral period, the inverse of the interferometer time delay (τ).

The theoretical contrast between the two output arms 132a, 132b is calculated according to the following equation: D = cos²(πΔf/FSR+Δφ) - sin²(πΔf7FSR+Δφ) = cos(2(πΔf/FSR+Δφ)). The contrast determines the impact on the signal-to-noise ratio (SNR) of the interferometer 120. When the bias Δφ = 0, the contrast and SNR go to zero when Δf = FSR/4 (or odd multiples thereof), corresponding to the 3 dB point of the transfer function of both arms. As a result, interchannel crosstalk between channels with nearly this spacing is reduced. Beyond the Δf = FSR/4 point, the data starts to invert.

The signal loss is represented as the solid line with diamonds; the solid line represents the interferometer cos²() transfer function; the dashed line represents the interferometer sin²() transfer function. A performance penalty is incurred with the multi-channel DPSK receiver 100 whenever the condition of Equation 2 presented above is not met since it is not possible for all channels to align to the interferometer simultaneously. This can occur, for example, when commonly used 7%-overhead G.709 compliant Forward Error Correction (FEC) coding is used with standard SONET rates, which brings a 10 Gbps SONET data rate to a 10.7 Gbps coded channel rate. These rates do not factor evenly into the 100 GHz ITU grid spacing, so it is impossible for all of the ITU grid-based WDM channels to align with the periodicity of the interferometer 120.

Fig. 2C is a schematic diagram of a representative orthogonal DPSK demodulator such as a quadrature DPSK or DQPSK demodulator 222 composed a splitter 129 followed by two delay line interferometers. A first interferometer 120-a has a π/4 phase shift and a second 120-b has a -π/4 phase shift relative to the first so that the two are orthogonal. Interferometer 120-a demodulates a first series of DBPSK signals and 120-b demodulates a second series of DBPSK signals that are orthogonal to the first.

Fig. 3 A is a graphical representation of a channel spacing evenly divisible by an integer multiple of a channel rate, superimposed on the interferometer transfer function for one output arm, (e.g., cos²). In this case, Xi, X₂, and X_n are spaced by integer multiples of the channel rate. Fig. 3B is a graphical representation of a channel spacing evenly divisible by the half channel rate, superimposed on the interferometer transfer function for both output arms, (i.e., cos² and sin²). In this case, λi and X_n are at an odd half channel rate spacing multiple (requiring polarity correction), and X₂ is spaced by both integer and half rate multiples of the channel rate. Fig. 3C illustrates the use of the pilot tone to lock the interferometer to a known reference. Fig. 3C also illustrates the generation of feedback signals that can be used to tune transmitted channels λι,λ₂, and X_n to the desired wavelengths, in this case aligned to the interferometer on ITU grid channels 57, 58, and 59. A 40 GHz FSR interferometer is used to receive, for example, 40 Gbps DPSK data on the 100 GHz ITU grid. As in Fig. 3B, λi and X_n are at an odd half channel rate spacing multiple being located on an inverted fringe (requiring polarity correction), and λ₂ is spaced by both integer and half rate multiples of the channel rate. The pilot tone is set to 195,970.0 GHz, and the interferometer phase bias is controlled using a phase shifter (not shown) so that the measured normalized contrast D(λ_p, Δφ) for the given pilot tone wavelength is adjusted to a target value, i.e., D(λ_p, Δφ) = T, where {-1 < T < 1 }. Continuing to refer to Fig. 3C, the target contrast is zero, which places an inverted fringe precisely on ITU channel #59 (195,900 GHz) where λι is expected. Similarly, a positive fringe is aligned to ITU channel #58 (195,800 GHz) where λ₂ is expected, and another inverted fringe is aligned to ITU channel #57 (195,700 GHz) where λ₃ is expected. Note that the polarity of the ITU channels received by the interferometer could be inverted by simply shifting the pilot tone an odd multiple of FSR/2, (e.g., λ_p= 195,950.0, 195,910.0, or 195,870.0).

Any of the methods known in the prior art, such as proportional feedback control, can be used to control the pilot tone 108. For example, an updated error term can be defined as the difference between the measured value of D and the target T, i.e., Ej₊i = Dj - T. The phase Δφ of the interferometer, controlled by a phase shifter, can be augmented by an increment that is proportional to the error, i.e., Δφj₊i = Δφi + g*Ej, where g is an appropriately chosen proportionality constant. The phase of the interferometer 120 may be iteratively updated until the phase error reaches an acceptably small level and converges in a stable manner to a unique phase, at which point, the interferometer 120 is "locked" to the target contrast D and a corresponding phase, Δφ.

As indicated'in Fig. 3C, the incoming channel wavelengths are measured by wavemeter 170 to be 195,902.2, 195,800.0, and 195,698.2 GHz for λi, λ₂, andλ,,, respectively. The control processor 125 calculates the wavelength error for each channel, which corresponds to wavelength error Δ's of 4.2 GHz, 0.0 GHz, and —1.8 GHz for λi, λ₂, and X_n.. The control processor 125 sends accurate wavelength error correction information 188 back to the transmitters 90 to achieve rapid convergence to optimum link performance. Note that error information may be useful for calibrating local wavelengths and measurement hardware, such as the local pilot tone generator 107 or wavemeter 170. For instance, if all the incoming channels yield the same error (e.g., Δ| through Δ_n = 4.5 GHz), this can be an indicator that either the pilot tone generator 107 or the wavemeter 170 may need calibration. Alternatively, rather than correct all the transmitters with a common wavelength error, the control processor 125 may shift the pilot tone 108 by 4.5 GHz to compensate for the error common to all the transmitters.

To avoid performance penalties when using the multi-channel DPSK receiver 100, the half channel rate and channel spacing can be forced to factor evenly (upholding Equation 2) by either: (i) adjusting the channel spacing to be a multiple of the channel rate, abandoning, for example, a standard such as the ITU grid if necessary, or (ii) adjusting the half channel rate to be an even factor of the channel spacing, abandoning, for example, SONET or G.709 standards, if necessary. Note that applying strong Rate Vi codes to SONET data rates yields channel rates consistent with the multi-wavelength DPSK receiver and standard ITU grid channel spacing.

When misalignment cannot be avoided, e.g., if conforming with existing standards is a priority, performance penalties can be constrained to an acceptable level by operating in a regime where the incoming channel wavelength and the interferometer alignment are close enough so that the misalignment penalty or conformance concerns become negligible.

For instance, a 10.7 Gbps channel rate can be received by a 10.7 GHz FSR interferometer, which can accept optical center frequencies every 5.35 GHz (assuming the ability to anticipate or detect and correct inverted data). While most of the interferometer fringes do not align exactly to the 100 GHz ITU grid, none of the fringes are more than 5.35/2 or ~2.7 GHz from the ITU grid, with the average deviation of ~1.4 GHz. Note that if a smaller deviation is required, the deviation can be reduced by a factor of n by using Non-Adjacent (NA)-DPSK with an n NA-pulse separation, which corresponds to an interferometer FSR narrowed by a factor of n. Thus, for 10.7 Gbps and n=2, transmitted wavelengths can be tuned to align precisely with the interferometer 120, providing penalty-free performance while maintaining a sub-GHz average deviation from the ITU grid. Note that for transmission of harmonically related channel rates (e.g., 2.5 Gbps, 10 Gbps, and 40 Gbps), a single interferometer 120 with delay equal to an integer multiple of the lowest channel rate bit period may also be used to demodulate multi-rate optical signals (e.g., n-NA-DPSK channels with varying n). Similarly, a pair of orthogonally phased delay line interferometers can be used to demodulate multi-rate orthogonal DPSK signals such as-DQPSK optical signals. In this manner, one or two interferometers may used to demodulate multiple orthogonal DBDPSK or DQPSK WDM channels having multiple rates simultaneously. For example, a single interferometer with a 400 psec delay can be used to demodulate multiple optical DPSK signals with ITU grid compliant 100 GHz spaced carrier wavelengths carrying DBPSK data at 2.5 Gbps, 10 Gbps, and/or 40 Gbps rates. With a second orthogonally phased interferometer, the pair could demodulate 100 GHz spaced carrier wavelengths carrying DQPSK data at 5 Gbps, 20 Gbps, and/or 80 Gbps rates. Fig. 4 is a schematic diagram of a two node network 400 that includes two network nodes 405-1 and 405-2 (collectively, network nodes 405). Each of these network nodes 405 includes transmitter(s) 90 and a receiver 100, as described in reference to Figs. IA and IB. The network nodes 405 communicate via optical paths 410, which may be free space or fiber optic optical paths. In operation, the transmitter(s) 90 transmits WDM DPSK signal(s) 105 via the optical path 410 to a receiver 100. Proper wavelength spacing is achieved in this embodiment optionally through use of a feedback signal 188, which is fed back by the receiver 100 to the transmitter 90. The feedback signal 188 may be determined through use of the wavemeter 170, which measures the wavelength of each channel (e.g., 1555.000 nm vs. 1554.800 nm). In another embodiment, a reference or pilot tone 108 may be temporarily or continuously introduced into the interferometer 120. The wavelength of the reference or pilot tone 108 may be known in advance or measured with the wavemeter 170. In such an embodiment, the reference or pilot tone 108 can be used for stabilization or to lock the interferometer 120 to a preselected comb of wavelength(s) at which incoming channel(s) are expected to be.

Given knowledge of (i) the pilot tone wavelength, (ii) the interferometer FSR, and (iii) the targeted locking position within the FSR, which are parameters that can be known or measured locally at the receiver 100, the position of the entire comb of interferometer channels (i.e., location of periodic interferometer peaks and troughs) can be determined. This information may be relayed back to the transmitter(s) as target wavelength information or may be combined with measurement of incoming wavelengths to generate an error signal indicating the error between the received and targeted wavelength for each channel.

The error or a representation of the error is fed back to the transmitter(s) 90 of the optical signal(s) for correction of the given channel's wavelength. Optionally, the error signal(s) may be used to align the interferometer 120 to the incoming channels, thereby minimizing the aggregate error. As an example, this capability may be particularly beneficial if all incoming wavelengths deviate from the optimum by the same amount, a condition that can occur whenever the incoming wavelengths are (properly) spaced at an integer multiple of the interferometer FSR. In this case, it may be preferable to tune the interferometer 120 to the incoming signals rather than tuning all the incoming signals to the interferometer 120.

In yet another embodiment, other information available to the receiver may be used to determine the contents of the feedback signal. Examples of such information includes (i) a bit error rate (BER) or other metrics associated with detection of the optical signal(s) 105, such as FEC based error rate estimates or Doppler shifit(s) that can be measured, for example, via the wavemeter 170

(optically) or via clock recovery offset(s), or (ii) overhead or data contained in the optical signal(s) 105. In such an embodiment, the transmitter(s) 90 of the optical signal(s) 105 may "step around" the wavelength of the channel(s) until a minimum bit error rate, for example, is found. Other min/max search techniques known in the art may also be employed.

The feedback signals 185, 188 (Fig. IA) may be implemented in many different forms. For example, the feedback signals 185, 188 may be analog, digital, or packetized. The feedback signals may be communicated via electrical, Radio Frequency (RF), or optical paths using applicable communications techniques and protocols. For example, the feedback signals 185, 188 may be handled via a network-level maintenance and control channel, traffic channel, or other suitable communications channel. The feedback signals 185, 188 may be a command signal or informational signal, depending on the processing capability associated with the transmitter(s) 90. The feedback signals 185, 188 may be included in overhead or payload portions of communications packets. The feedback signals 185, 188 may be transmitted over the optical communications path 410, optical maintenance path (not shown), electrical communications path (not shown), electrical maintenance path (not shown), or other path(s) configured for transporting the feedback signals 185, 188. Since optical transmitter wavelength(s) are generally stable (i.e., have a slow drift rate), correcting the transmitted wavelength(s) can be done relatively infrequently (i.e., every few minutes, hours, days, weeks, months, or years) depending on the transmitter 90 and possibly environmental conditions at the transmitter(s) 90, receiver 100, or optical paths therebetween and be sent from the control processor 125 on a periodic, aperiodic, event driven, or request driven basis.

Fig. 5A is an example communications network 500 in which the receiver 100 may be employed. The communications network 500 is a Free Space Optic (FSO) network. The optical signal 105 is transmitted by the transmitters 90 between mobile platforms, such as communications satellites 501-1 and 501-2, and received by the DPSK receiver(s) 100 in a manner as described above. The optical receiver 100 can also be employed by other platforms, such as aircraft, and relatively stable links, such as building-to-building and ship-to-shore FSO links. Fig. 5B is another example of a network in which the DPSK receiver 100 may be employed. The optical network 515 includes optical nodes 405-1, 405-2, 405-3, and 405-4 (collectively, nodes 405). Between each of these nodes 405 are fiber optic or FSO communications paths 410. The optical communications network 515 is configured as a Unidirectional Path Switched Ring (UPSR) or Bi-directional Line Switched Ring (BLSR) that may be optically coupled to other rings or optical networks having other configurations.

It should be understood that there are many other optical communications applications in which the WDM receiver 100 may be employed and provide advantages as described above. While this invention has been particularly shown and described with references to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention encompassed by the appended claims.

For example, the interferometers 120 of Figs. 1 and 2 can be more than a 1- bit delay line interferometer. It may be a multiple bit (multi-bit) delay line interferometer in one embodiment or a dynamically selectable bit delay line interferometer in another embodiment. The pilot tone reference wavelength of Fig. IA, rather than being local, can be selected as one of the incoming channels. The interferometer 120 and optical channels 105 can align to wavelengths relative to the selected channel through feedback from the control processor 125. The main optical paths 132 (Figs. IA and IB) may also be a single main optical path instead of the two main optical paths. In such an embodiment, the single main optical path may carry a signal when the interference is constructive and no signal when the interference is destructive, or vice-versa. Because the two main optical path embodiment provides 3 dB more sensitivity than the single main optical path embodiment, the two main optical path embodiment is generally preferable.

However, the single main path embodiment may be easier to implement since it does not require both a-side and b-side elements (such as WDMs 131), balanced detection, or amplitude and time alignment of received photocurrents for each channel. The main optical paths 132 may also include more than two main optical paths. Such an embodiment may be used in cases where the interferometer interferes more than two optical signal pulses in the DPSK channels. For example, future noise reduction or security schemes may facilitate or warrant higher order DPSK demodulation, in which case, three, four, or more main optical paths 132 may be employed.

In Figs. IA and IB, instead of having a tributary optical path for each wavelength, multiple wavelengths may be directed onto the same tributary optical paths and separated onto yet other tributary optical paths to respective optical receivers for processing. In addition to use in the point-to-point network of Fig. 5 A and the ring network of Fig. 5B, the receiver 100 may be used in other network configurations, such as mesh networks, Bi-directional Line Switched Ring (BLSR) networks, broadcast networks, or multi-access networks.

Claims

CLAIMSWhat is claimed is:

1. An apparatus for demodulating an optical signal having Differentially Encoded Phase Shift Keying (DPSK) channels, the apparatus comprising: at least one delay line interferometer to (i) delay at least one channel in an optical signal by multiple bits, (ii) demodulate the optical signal from having DPSK channels to an optical signal having channels modulated in intensity and (iii) output the demodulated optical signal onto at least one main output optical path; and channel selectors in the at least one main output optical path to direct channels onto tributary optical paths.

2. The apparatus according to claim 1 further including optoelectronic converters in the tributary optical paths to convert the demodulated optical signal into respective, corresponding, electrical signals.

3. The apparatus according to claim 1 wherein the at least one delay line interferometer delays at least one channel by at least one bit.

4. The apparatus according to claim 1 wherein the optical signal includes at least one of the following: BDPSK, DQPSK, and/or 8-DPSK WDM optical signals.

5. The apparatus according to claim 1 wherein the delay line interferometer is selectably adjustable to interfere optical signal pulses offset by a selectable number.

6. The apparatus according to claim 1 wherein the interferometer includes an electronically tunable phase shifter to receive incoming wavelengths.

7. The apparatus according to claim 1 wherein wavelength spacing between carrier wavelengths defining the channels is an integer multiple of a channel rate of the optical signal.

8. The apparatus according to claim 1 wherein wavelength spacing between carrier wavelengths defining the channels is evenly divisible by an odd number of half channel rates of the optical signal.

9. The apparatus according to claim 8 further including a polarity corrector to correct polarity of the channels.

10. The apparatus according to claim 9 wherein the polarity corrector is configured to correct polarity of electrical signals corresponding to the channels.

11. The apparatus according to claim 9 wherein the polarity corrector is configured to optically correct polarity of the channels.

12. The apparatus according to claim 1 wherein channel rates are integer multiples of interferometer delay.

13. The apparatus according to claim 1 further including a feedback processor to generate signal(s) transmitted to a transmitter(s) of the optical signal(s) to cause the transmitter to tune carrier wavelengths defining the channels for adjusting wavelength position or channel separation.

14. The apparatus according to claim 1 further including a low noise optical amplifier to receive the optical signal and outputting the received amplified optical signal to the interferometer.

15. The apparatus according to claim 1 wherein the channel selectors include optical filters.

16. The apparatus according to claim 1 wherein the interferometer is configured to receive at least one pilot signal to allow for stabilization and control of the interferometer and wavelength alignment of the incoming optical signal(s).

17. The apparatus according to claim 1 used in an optical regenerator.

18. The apparatus according to claim 1 used in a free space optic communications network.

19. The apparatus according to claim 1 used in a fiber optic communications network.

20. The apparatus according to claim 1 wherein the optical signal includes multiple-rate WDM signals.

21. The apparatus according to claim 1 wherein two delay line interferometers are orthogonally phased

22. The apparatus according to claim 21 wherein the optical signals are WDM optical signals that are spaced by odd multiples of the interferometer FSR/4.

23. The apparatus according to claim 1 wherein two interferometers have different delays.

24. A method of demodulating an optical signal having Differentially Encoded Phase Shift Keying (DPSK) channels, the method comprising: delaying at least one first series of differentially encoded optical signal pulses in respective DPSK channels of the at least one first series of optical signals by a multiple bit duration to align the at least one first series optical signal pulses with at least one respective second series of differentially encoded optical signal pulses also in the respective DPSK channels; respectively interfering the at least one first and second series of optical signal pulses in an aligned state to demodulate the optical signals from having DPSK channels to at least one optical signal having channels modulated in intensity; outputting the demodulated optical signal onto respective main output optical paths; and directing channels in the respective main output optical paths onto tributary optical paths.

25. The method according to claim 24 further including converting the demodulated optical signal on the tributary optical paths into respective, corresponding, electrical signals.

26. The method according to claim 24 wherein demodulating the optical signal includes interfering adjacent optical signal pulses.

27. The method according to claim 24 wherein demodulating the optical signal includes interfering non-adjacent optical signal pulses.

28. The method according to claim 24 wherein demodulating the optical signal includes interfering optical signal pulses offset by a selectable number.

29. The method according to claim 24 wherein demodulating the optical signal includes tuning at least one interferometer for receiving incoming wavelengths.

30. The method according to claim 24 wherein wavelength spacing between carrier wavelengths defining the channels is an integer multiple of a channel rate of the optical signal.

31. The method according to claim 24 wherein delaying the at least one first series of optical signal pulses further comprises delaying each optical signal of the at least one first set a different amount.

32. The method according to claim 29 wherein the at least one delay line interferometer is controlled using a pilot signal.

33. The method according to claim 24 wherein wavelength spacing between carrier wavelengths defining the channels is evenly divisible by an odd number of half channel rates of the optical signal.

34. The method according to claim 24 wherein the delay is achieved with two orthogonally phased delay line interferometers.

35. The method according to claim 34 wherein wavelength spacing between carrier wavelengths defining the channels is an odd multiple of the delay line interferometer FSR/4.

36. The method according to claim 33 further including correcting polarity of the channels.

37. The method according to claim 33 wherein correcting the polarity of the channels includes correcting polarity of electrical signals corresponding to the channels.

38. The method according to claim 36 wherein correcting the polarity of the channels includes optically correcting polarity of the channels.

39. The method according to claim 24 wherein channel rate bit durations are integer multiples of demodulating delay.

40. The method according to claim 24 further including feeding back signal(s) to transmitter(s) of the optical signal(s) to cause the transmitter to tune carrier wavelengths defining the channels for adjusting wavelength position or channel separation.

41. The method according to claim 24 further including optically amplifying the optical signals prior to demodulating the optical signal.

42. The method according to claim 24 wherein directing the channels onto the tributary optical paths includes filtering the phase demodulated optical signals.

43. The method according to claim 24 further including controlling at least one demodulating interferometer based on at least one pilot signal.

44. The method according to claim 24 used in an optical regenerator.

45. The method according to claim 24 used in a free space optic communications network.

46. The method according to claim 24 used in a fiber optic communications network.

47. An apparatus for demodulating an optical signal having Differentially Encoded Phase Shift Keying (DPSK) channels, the apparatus comprising: means for delaying at least one first series of differentially encoded optical signal pulses in at least one respective DPSK channel of the optical signal by a multiple bit duration to align the at least one series of first optical signal pulses with at least one second series of differentially encoded optical signal pulses also in the at least one respective DPSK channel; means for interfering the at least one first and second series differentially encoded optical signal pulses in an aligned state to demodulate the optical signal from having DPSK channels to a respective optical signal having channels modulated in intensity; means for outputting the respective demodulated optical signal onto at least one respective main output optical path; and means for directing channels in the at least one respective main output optical path onto tributary optical paths.

48. A method for demodulating an optical signal having Differentially Encoded Phase Shift Keying (DPSK) channels, the method comprising: feeding back a signal to constrain wavelength spacing of DPSK channels in an optical signal; and interfering offset optical signal pulses in the optical signal to demodulate the optical signal from being DPSK modulated to being intensity modulated.

49. The method according to claim 48 further including directing channels in the demodulated optical signal from at least one main optical path onto tributary optical paths.

50. A system for demodulating an optical signal having Differentially Encoded Phase Shift Keying (DPSK) channels, the system comprising: means for feeding back a signal to constrain wavelength spacing of DPSK channels in an optical signal; and means for interfering offset optical signal pulses in the optical signal to demodulate the optical signal from being DPSK modulated to being intensity modulated.

51. The method according to claim 50 further including means for directing channels in the demodulated optical signal from at least one main optical path onto tributary optical paths.

52. An apparatus for demodulating an optical signal having Differentially Encoded Phase Shift Keying (DPSK) channels, the apparatus comprising: a processor to feed back a signal to constrain wavelength spacing of DPSK channels in an optical signal; and at least one interferometer that interferes offset optical signal pulses in the optical signal to demodulate the optical signal from being DPSK modulated to being intensity modulated.

53. The apparatus according to claim 52 wherein the at least one interferometer directs channels in the demodulated optical signal from at least one main optical path onto tributary optical paths.

54. An apparatus for demodulating an optical signal having Differentially Encoded Phase Shift Keying (DPSK) channels, the apparatus comprising: a least one delay line interferometer to (i) demodulate an optical signal from having DPSK channels to an optical signal having channels modulated in intensity (ii) output the demodulated optical signal onto at least one main output optical path, and (iii) receive at least one pilot signal to allow for stabilization and control of the at least one interferometer and wavelength alignment of the incoming optical signal(s); and channel selectors in the at least one main output optical path to direct channels onto tributary optical paths.