US20100067607A1

US20100067607A1 - All-optical balanced detection system

Info

Publication number: US20100067607A1
Application number: US12/284,195
Authority: US
Inventors: Mathias Westlund; Peter Andrekson; Henrik Sunnerud; Mats Skold
Original assignee: PicoSolve Inc
Current assignee: Exfo Electro Optical Engineering Inc
Priority date: 2008-09-18
Filing date: 2008-09-18
Publication date: 2010-03-18
Also published as: CN102150384B; US20110182573A1; WO2010033654A2; US9325428B2; WO2010033654A3; EP2351263A4; EP2351263B1; EP2351263A2; CN102150384A

Abstract

A two gate sampling system designed to perform sampled balanced detection of one or more input signal pairs. The present invention performs simultaneous sampling of both signals in each signal pair followed by digitization and combination of the sample pairs using software. By first sampling the signals and then combine the sampled into the corresponding balanced detected signal it is possible to avoid the bandwidth limitations and impedance problems introduced by traditional balanced detectors and electrical oscilloscopes. In particular for optical sampling gates very high bandwidth sampling gates can be designed without any impedance issues and hence almost perfect balanced detection reconstruction can be performed for very high speed signals. Balanced detection is becoming more and more important as the new phase modulated optical data signals are introduced to the market, such as e.g. PSK, DPSK, QPSK and DQPSK. The present invention is well suited for analysis of these new type of signals.

Description

TECHNICAL FIELD

The present invention relates to a sampling arrangement particularly well-suited for analysis of high speed data signals and, more particularly, to a sampling arrangement with two or more coupled sampling gates.

BACKGROUND OF THE INVENTION

Digital sampling is a technique used to visualize a time-varying waveform by capturing quasi-instantaneous snapshots of the waveform via, for example, a sampling gate. The gate is “opened” and “closed” by narrow pulses (strobes) in a pulse train that exhibit a well-defined repetitive behavior such that ultimately all parts of the waveform are sampled. The sampling implementation can either be real-time or equivalent-time, where real-time sampling refers to the case where the sampling rate is higher than twice the highest frequency content of the waveform under test (Nyquist sampling), while equivalent-time sampling uses an arbitrarily low sampling rate. However, equivalent-time sampling requires the measured waveform to be repetitive (in order to provide accurate signal reconstruction)—a fundamental limitation when compared to real-time sampling. The present invention is independent of the sampling rate, and hence, can be either real-time or equivalent-time sampling.
The recent advances in the field of optical communication with new, more complex, data modulation formats as a key technology has created a need for optical waveform characterization tools which are capable of extracting more information from the waveform than simply its power as a function of time.
In particular, many different modulation formats have been developed which use modulation of the phase of the optical carrier to encode the data to be transmitted. A few types of phase modulated signals have already been employed in commercial systems, such as differential phase shift keying (DPSK) and differential quaternary phase shift keying (DQPSK). For these differential modulation formats the data is encoded as the relative phase shift between consecutive symbols. In DPSK modulation schemes, for example, a π phase shift between bits represents a logical “1” and a zero phase shift represents a logical “0”. For DQPSK modulation, each symbol contains two bits of information by allowing four different relative phase changes between consecutive bits (e.g., 0, π/2, π and 3π/2).
FIG. 1 is used to further clarify the concept of phase-encoded modulation formats such as phase-shift keying (PSK), differential phase-shift keying (DPSK), and QPSK and DQPSK as defined above. For each type of modulation, the optical phase and amplitude of the data signal are visualized in constellation diagrams showing the optical field amplitude as the radial distance from origin R and the optical field phase as the angle φ. In FIG. 1, the logical marks (ones) and spaces (zeros) are represented as either absolute phase and amplitude levels (for PSK and QPSK formats, FIGS. 1( a) and (b), respectively), or as phase and amplitude transitions for the differentially-coded phase and amplitude levels (for DPSK and DQPSK formats, FIGS. 1( c) and (d), respectively). For D/QPSK each symbol contains, as shown, two bits of information. Therefore, four different logical phase and amplitude combinations are used to represent the “symbols” in either of these modulation format types.
It is to be noted that the amplitude of the data signal is constant for each of these phase-encoded modulation techniques. Hence, if only the power of the incoming signal is “detected” using a conventional photodetector-based o/e conversion device, the phase information will be lost. To extract the phase information, the signal needs to be mixed with an optical reference signal which converts the phase information into amplitude information. For differentially-modulated signals, delay interferometers (DIs), such as Mach-Zehnder interferometers (MZIs), Michelson interferometers, or the like, are commonly used in which the signal itself serves as reference after being delayed one (or more) bit periods. For absolute phase encoded signals (e.g. PSK or QPSK), an independent reference signal is necessary to extract the phase information from each bit.
The DI is an interferometric structure where the incoming optical waveform is equally split up in two paths and one path is delayed relative to the second path before recombining the two paths. The relative delay is coarsely set equal to an integer number of bit slots (most commonly one bit slot) and finely tuned to match a particular relative phase delay of the optical carrier. For example, in the DPSK case, the relative delay is a multiple of π in order to effectively translate the relative phase shifts between the symbols into a binary amplitude modulated signal. The DI has two output ports—a constructive interference port and a deconstructive interference port (the ‘destructive’ port outputting the complementary data of the ‘constructive’ port). In order to optimize a DPSK receiver in terms of signal sensitivity, both outputs from the DI are detected by a balanced detector structure.
In order to recover the data embedded in an incoming DQPSK signal, the signal is first evenly split so as to applied as “equal power” inputs into two separate DIs with different relative optical phase delays (+π/4+n*π and −π/4+m*π, where n and m are integers) and each DI pair of outputs is thereafter detected by a balanced detector structure. By properly choosing the relative phase delays, two bits of information per symbol can be separated and represented as one bit per balanced detector output. The amplitude modulated output from each balanced detector is thereafter sampled (for example, digital sampling) in order to visualize each bit's corresponding eye-diagram.
A major concern when using balanced detection for optical to electrical (o/e) conversion followed by electrical digital sampling is the influence of the measurement system on the measured waveform, which is known to introduce measurement error. In particular, balanced detection and electrical sampling suffers from two major limitations: (1) limited measurement bandwidth (currently <50 GHz); and (2) significant impedance mismatch, resulting in distortion in the measured waveform. For high speed signal characterization (10 GSymbols/s, 40 GSymbols/s or higher), these effects can influence the measurement results to such an extent that the measured waveform is dominated by the measurement system impulse response, which is unacceptable when needing to recover such high speed data signals.
Thus, a need remains in the art for an arrangement capable of characterizing (visualizing) high symbol rate optical signals without being hampered by the measurement system bandwidth or the distortion due to o/e conversion and related impedance matching issues.

SUMMARY OF THE INVENTION

The needs remaining in the prior art are addressed by the present invention, which relates to a sampling arrangement particularly well-suited for analysis of high speed data signals and, more particularly, to a sampling arrangement comprising two or more coupled sampling gates for recovering information from modulated input signals.
In accordance with the present invention, a sampling arrangement utilizes two (or more) separate sampling gates controlled by the same strobe frequency, f_s, to acquire samples from two (or more) input signals. The lengths of the paths to the sampling gates are adjusted by tunable (or fixed) delay lines so as to enable precise, time-overlapped sampling of all input signals. Examples of suitable delay line arrangements include a “fixed” delay line, a “set-and-forget” delay line and a “tunable” delay line.
In particular, for the application of measuring the output signal pairs of one or more delay interferometers (DIs), as in the case of DPSK and DQPSK signals, the delay lines are used to ensure that the time delay from the output of each DI to the two corresponding sampling gates are equal to within a fraction of the temporal resolution of the sampling gates. Hence, every pair of samples originating from the two outputs of each DI originates from the same time “slice” of the waveform under test. The acquired pairs of samples are then, after detection and analog-to-digital (A/D) conversion, combined in software to samples representing balanced detection of the sample pairs. With this scheme, the need to perform balanced detection in hardware is avoided. In particular, when using optical sampling gates, the sampling gate bandwidth (temporal resolution) can be extremely high and impedance mismatch is no longer an issue, since the sampling takes place in the optical domain.
In one embodiment of the present invention, the two sampling gates are used for more than one input signal pair, such as in the case for a DQPSK signal where after demodulation by two DIs the two output signal pairs are measured in order to present the eye-diagram of each bit in the 2 bits/Symbol DQPSK signal. By including, for example, optical switches before the two sampling gates, the DI output pairs can be measured by the sampling gates by switching in a predetermined fashion.
Another embodiment of the present invention includes a sampling gate to sample an external reference clock, which can be used to establish the time base for the acquired samples from the signal under test.
Other and further aspects and embodiments of the present invention will become apparent during the course of the following discussion and by reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

Referring now to the drawings,

FIG. 1 illustrates the modulation principle of four different phase encoded modulation formats visualized in constellation diagrams containing information about both the amplitude and phase of the optical field;

FIG. 2 illustrates a prior art arrangement for demodulating DPSK signals;

FIG. 3 shows an embodiment of the present invention for performing sampling in a demodulated DPSK signal, and also illustrates measured eye-diagrams of a demodulated 40 GSymbol/s DPSK signal using this embodiment of the present invention;

FIG. 4 shows the embodiment of the present invention from FIG. 3 with a typical signal DI demodulation setup in front of the four input ports of the present invention;

FIG. 5 illustrates a timing condition associated with the arrangement of FIG. 4; and

FIG. 6 illustrates an embodiment of the present invention using an external reference clock to synchronize the acquired samples.

DETAILED DESCRIPTION

Prior to describing the details of the exemplary sampling arrangement of the present invention, a prior art arrangement for demodulating DPSK-encoded signals will be reviewed with reference to FIG. 2. The DPSK signal is demodulated using a delay interferometer (DI) 10 having a relative delay difference between the two interferometer arms. DI 10 is shown as comprising a first signal path 12 and a second path 14. An incoming modulated DPSK signal passes through a splitter 16 such that an even power level of signal is directed into paths 12 and 14. Second path 14 includes a delay element 18, represented as a fixed amount of delay (25 ps in this example) and a variable amount of delay (shown as Δφ). The phase shift is selected such that a delay of an integer number of bits (generally a single bit) is obtained. The original and the phase-shifted versions of the DPSK-encoded signal are thereafter recombined in a signal combiner 20 and split along two separate output paths 22 and 24. As with splitter 16, the output signals along paths 22 and 24 comprise half of the power of the combined original/phase-shifted signals.
At the two outputs from DI 10, the phase information in the DPSK signal is converted into two amplitude modulated signal, a first “constructive interference” signal with power P_calong a first output path 22, and a second “destructive interference” signal (exhibiting the complementary information) at power P_dalong a second output path 24
With traditional techniques, these two output signals would be applied as inputs to a balanced opto-electronic detector, which would subtract the one signal from the other and convert the difference into the electronic domain, ideally providing an electrical signal representing P_c−P_d. In the prior art arrangement of FIG. 2, a pair of photodiodes 21 and 23 are used to provide this opto-electronic conversion. However, such o/e conversion techniques are limited in bandwidth and quality of impulse response. As a result, the electrical signal created after detection does not represent the ideal case, in particular for high speed signals.
In contrast, the present invention utilizes a sampling technique to individually measure the waveform on each DI output, in a manner to be described in detail hereinbelow. A software-embedded algorithm is then used to combine the samples in a manner which emulates the operation of an ideal balanced detector, performing the operation P_c−P_d, to create a sampled output waveform as shown in eye diagram 34 of FIG. 3. For input signals other than DPSK (such as DQPSK), similar reasoning applies but instead of having only two output signals after a single DI, there can be more DI's each having two output signals which can be taken care of by embodiments of the present invention to be described below.
In a preferred embodiment of the present invention, the sampling of the two DI output signals is performed in the optical domain, so as to completely remove the influence of the bandwidth limitations inherent in optical-electronic conversion and provide a final result which can be very close to the targeted ideal result P_c−P_d. However, the sampling technique of the present invention is not limited to the optical domain; electrical sampling techniques may be used in suitable applications (for example, lower speed applications).
FIG. 3 shows an embodiment of the present invention which utilizes the same incoming DPSK-encoded signal and demodulating arrangement including DI 10 as described above in association with FIG. 2. The embodiment of FIG. 3 may also be utilized if only one bit in a two-bits-per-symbol DQPSK signal is sampled. As will be described in detail below, a sampling arrangement 40 formed in accordance with the present invention is used in place of prior art o/e conversion arrangements to more accurately recover the data from the phase-encoded incoming signal. The “constructive” signal propagating along first output path 22 is shown as applied as an input to first signal port Al of arrangement 40. Similarly, the “destructive” signal propagating along second output path 24 is applied as an input to second signal port A2.
It is to be understood that the technique of the present invention can be scaled to support a larger number of input ports, as will be discussed in detail below. Moreover, the input signals can be either optical or electrical. In its most general form, the present invention is a combination of performing sampling of pairs of input signals in hardware and using software algorithms to combine the created samples into a single output corresponding to balanced detection of the input signal pairs.
Referring back to the particular embodiment of FIG. 3, such a result is shown as the simultaneous sampling of A1 and A2 and subsequent reconstruction of a sampled version of the power of A1-A2. At the core of the present invention is a pair of sampling gates 42 and 44 which are opened and closed by a common strobe source 46 representing a sampling frequency f_s. The sampling can be performed either in the electrical domain or in the optical domain, depending on the domain of the signals arriving at ports A1 and A2. However, to fully take advantage of the capability of the present invention, optical sampling is the preferred embodiment in order to eliminate all high speed electronics and o/e conversion. By digitally sampling the output waveform from the balanced detector structure, the corresponding electrical eye-diagram, showing logical binary amplitude levels corresponding to the phase transitions in the DPSK signal, can be visualized.
Optical sampling gates 42 and 44 may comprise any one of a wide variety of implementations using different nonlinear optical processes to create the gating functionality. Exemplary suitable components include, but are not limited to, four-wave mixing in fiber, sum-frequency generation in optical crystals and cross-phase modulation in fiber or semiconductor optical amplifiers. While strobe source 46 is illustrated as a single element, it is to be understood that separate strobe sources, having the same sampling frequency f_smay also be used, with each separate strobe source used to control a separate gate.
A key design parameter for the present invention is to facilitate alignment of the sampling times of gates 42 and 44 via strobe source 46 such that the two parts of the signal are synchronously sampled in order for combination in the software to be accurate. A delay line 48 is disposed at first input port Al and is used to adjust the distance (or time delay) from the input A1 to sampling gate 42, thereby adjusting the sampling time of gate 42 relative to the sampling time of gate 44. FIG. 5 will describe an example which highlights the condition for adjusting delay line 48. In general, the operation of delay line 48 can be either adjustable or fixed, depending on the measurement application.
The output samples from the sampling gates 42 and 44 are digitized by analog-to-digital converters (A/D) 50 and 52, respectively, and subsequently fed into a software processing and signal visualization system 54. The main functionality of system 54 related to the present invention is to combine the acquired sample pairs for each measurement in order to provide balanced detection functionality. Furthermore, the software can be used to visualize each measured input signal pair as the corresponding balanced detected signal. Eye diagram 28 of FIG. 3 is the sampled output associated with the constructive port, eye diagram 30 is the sampled output associated with the destructive port and, most importantly, eye diagram 34 is the resultant DPSK sampled information eye diagram, where each of these diagrams was created by system 54.
An alternative embodiment of the present invention allows for detection of the output samples from the sampling gates 42 and 44 using low bandwidth balanced receivers in order to perform the balanced detection in the hardware before digitizing the samples in an A/D converter.
It is to be understood that the present invention is independent of the particular method utilized to time stamp each sample. In particular, the technique of the present invention has been found to work for both real-time sampling and equivalent-time sampling, irrespective of the time-base design used for equivalent-time sampling.
FIG. 4 illustrates an embodiment suitable for measurement of, for example, QPSK or DQPSK signals, which requires the generation of two sample pairs for proper demodulation. As before, the incoming phase-encoded signal is split along signal paths 12 and 14 by a power splitter 16. In this case, however, the portion of the signal propagating along signal path 14 is thereafter applied as an input to a first DI 10-1, and the remaining portion propagating along signal path 12 is applied as an input to a second DI 10-2, as shown in FIG. 4. Each DI 10 includes a separate delay element 18, illustrated as delay element 18-1 (associated with DI 10-1) and delay element 18-2 (associated with DI 10-2). Delay elements 18-1 and 18-2 are shown as exhibiting the appropriate bit delays TS and phase relations Δφ1 and Δφ2 for demodulation of the input signal. In particular, for the case of DQPSK the phase relations can be, for example, Δφ1=+π/4 and Δφ2=−π/4, in order to separate each bit in the 2 bits/Symbol DQPSK data signal. By flipping the switches 56 and 58 in a predetermined manner, the outputs from each DI 10-1 and 10-2 can be measured separately, and with the present invention the corresponding balanced detected signals to each demodulated bit of the DQPSK signal can be visualized.
In this case, a set of four output signals have been created, a first signal pair A1 and A2 from DI 10-1 (similar to the embodiment of FIG. 3, as discussed above) and a second signal pair B1 and B2, from DI 10-2. In order to most efficiently utilize the elements of sampling arrangement 40, switches 56 and 58 are positioned at the entrance ports of arrangement 40, in front of sampling gates 42 and 44, in order to facilitate alternating sampling of the input signal pairs from DI's 10-1 and 10-2, that is, first A1, A2 and then B1, B2. As shown, a second delay line 48-2, associated with input B1 is included in the arrangement to provide the same synchronization activity as delay line 48 defined above and discussed in detail below.
FIG. 4 also points out that o/e conversion of the signals can be performed before the sampling takes place. In this case, a photodiode or other o/e conversion element is disposed along each signal path, and is collectively illustrated as conversion component 70 in FIG. 4. In this case, sampling gates 42 and 44 will comprise electronic sampling gates. The positioning of o/e conversion component 70 is flexible and can be either directly after DIs 10-1 and 10-2, or at any other point in front of sampling gates 42 and 44. Additionally, delay lines 48-1 and 48-2, as well as switches 56 and 58 can be either electrical or optical.
FIG. 5 illustrates the critical timing required between sampling gates 42 and 44 in order to sample each part of the generated signal pairs at the correct matching times. FIG. 5 is an extracted part of the embodiment of FIG. 4. In FIG. 5, the signal is not split up until point A, corresponding to the output of the DI 10-1. From this point on, it is critical that the time difference from point A until each part of the signal is sampled very close to equal in order to generate samples originating from the same time in the original signal. This is a condition in order to be able to combine the two samples in the software and emulate the balanced detection.
The timing condition can be expressed using the notations in FIG. 5 as |(T_c−t_c)−(T_d−t_d)|<Δτ, where T_cis the propagation time for the “constructive” signal from point A to sampling gate 42, t_cis the propagation time for the sampling strobe pulse from strobe source 46 to sampling gate 42, T_dis the propagation time for the “destructive” signal from point A to sampling gate 44, t_dis the propagation time for the sampling strobe pulse from strobe source 46 to sampling gate 44, and Δτ represents the temporal resolution of sampling gates 42 and 44. As alluded to above, delay line 48-1 plays a critical role to facilitate the fulfillment of this timing condition, in particular since the lengths of the conventional output fiber connectors of a delay interferometer is normally beyond the control of the constructor of the demodulating system present invention. However, with precise control of every other component within sampling arrangement 40, delay line 48 may be omitted, in particular for low bandwidth sampling gate solutions with high Δτ. The timing condition applies to all input signal pairs within the system of the present invention (e.g., in the arrangement of FIG. 4, a similar condition applies from point B to sampling gates 42 and 44).
It has been pointed out that the present invention is independent on the time-base design used to synchronize the acquired samples into a replica of the original signal. However, it should be noted that the present invention is compatible with U.S. Pat. No. 7,327,302, issued to M. Westlund et al. on Feb. 5, 2008, assigned to the assignee of this application and hereby incorporated by reference. FIG. 6 illustrates an embodiment of the present invention where an external reference signal source 60 is used to supply a gating control signal for the system, where the frequency f_cof reference clock signal C is directly related to the frequency of the demodulated signals appearing at ports A1, A2, B1 and B2. As shown, the reference clock output signal C from source 60 is sampled by a separate sampling gate 62, using the same strobe source 46. The generated clock samples are then digitized by an A/ID converter 64 and applied as an input to software processing system 54. With this input information, the timebase of the external clock can be determined by the embedded software algorithms. Since the frequency f_cof the external clock is directly related to the frequency of the input signal bit rate, the time-base of the external clock can be directly transferred to the recovered output signal signal.
It is to be understood that other advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details and representative embodiments shown and described herein. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the claims appended hereto.

Claims

1. A sampling arrangement for performing balanced detection of a pair of input data signals, the sampling arrangement comprising

at least one pair of sampling gates, a first sampling gate of said pair responsive to a first input signal of the pair of input signals and a second sampling gate of said pair responsive to a second input signal of said pair of input signals, each sampling gate creating a stream of samples representative of the respective input signal applied thereto;

a strobe source for controlling the at least one pair of sampling gates to open and close at a predetermined sampling frequency f_s;

an adjustable delay line disposed at the input of one sample gate of said at least one pair of sampling gates for minimizing differences in path lengths between each input signal and each sampling gate; and

a processor configured to perform balanced detection of the sampled pair of input signals to generate as an output a sampled balanced detected version of said input signal pair.

2. A sampling arrangement as defined in claim 1 wherein the at least one pair of sampling gates comprises a pair optical sampling gates.

3. A sampling arrangement as defined in claim 1 wherein the at least one pair of sampling gates comprises a pair of electrical sampling gates.

4. A sampling arrangement as defined in claim 1 wherein the strobe source comprises an optical component.

5. A sampling arrangement as defined in claim 1 wherein the strobe source comprises an electrical component.

6. A sampling arrangement as defined in claim 1 wherein the strobe source comprises a pair of strobe elements operating at the same frequency, each separate strobe element coupled to control a separate gate of the at least one pair of sampling gates.

7. A sampling arrangement as defined in claim 1 wherein the sampled pair of signals is represented by optical signals and the balanced detection of each sampled pair is performed by

using a optical-to-electrical balanced receiver having two optical inputs, S1 and S2, and an electrical output corresponding to S1-S2;

an A/D converter coupled to the output of the balanced receiver to convert an analog set samples into a digital representation thereof.

8. A sampling arrangement as defined in claim 1 wherein the sampled pair of signals is represented by electrical signals and the balanced detection of each sampled pair is performed by

using an electrical circuit having two electrical inputs, S1 and S2, and an electrical output corresponding to S1-S2;

9. A sampling arrangement as defined in claim 1 wherein the sampled pair of signals is represented by optical signals and the balanced detection of each sampled pair is performed by

detecting the first and second component of each sampled pair with separate optical-to-electrical receivers;

each component of the sampled pair are coupled to an A/D converter to convert the analog sets of samples into digital representations thereof;

the digitally represented pairs of sampled outputs are then subtracted on a sample by sample basis to form the reconstructed balanced detected signal.

10. A sampling arrangement as defined in claim 1 wherein the adjustable delay line is selected from the group consisting of: fixed delay, set-and-forget delay or tunable delay.

11. A sampling arrangement as defined in claim 10 wherein the adjustable delay line is selected from the group consisting of: optical delay lines and electrical delay lines.

12. A sampling arrangement as defined in claim 1 wherein

a single pair of optical sampling gates is responsive to a single pair of optical input signals to form a sampled pair of signals;

a single optical strobe source controls the single pair of optical sampling gates at a predetermined frequency;

an adjustable optical delay line is used to control the input signal pair paths differences to each sampling gate; and

the output sampled signals from the single pair of optical sampling gates are converted to electrical signals using separate optical-to-electrical receivers followed by A/D conversion of each signal, wherein the digitally represented signals are then subtracted to form the balanced detected version of the input signal pair.

13. A sampling arrangement as defined in claim 1 wherein

two pairs of optical sampling gates are responsive to two pairs of optical input signals to form two sampled pairs of signals;

a single optical strobe source controls the two pairs of optical sampling gates at a predetermined frequency;

adjustable optical delay lines are used to control each input signal pairs paths differences to each sampling gate; and

the output sampled signals after each pair of sampling gates are converted to electrical signals using separate optical-to-electrical receivers followed by A/D conversion of each signal wherein the digitally represented signals are then subtracted to form balanced detected versions of both input signal pairs.

14. A sampling arrangement as defined in claim 1 wherein

a single pair of optical sampling gates is responsive to at least two pairs of optical input signals to form sampled pairs of signals;

a single optical strobe source controls the at least two pairs of optical sampling gates at a predetermined frequency, the sampling arrangement further comprising:

optical switches disposed in the input signal path before each sampling gate associated with two or more inputs and a single output and controlled such that the optical switches select, in a predetermined order, which pair of input signals is sent to the sampling gates; and

adjustable optical delays disposed along each input signal pair to control the input signal pair paths differences to each sampling gate, wherein

the output sampled signals after the two sampling gates are converted to electrical signals using separate optical-to-electrical receivers followed by A/D conversion of each signal such that the digitally represented signals are then subtracted to form the balanced detected versions of the input signal pairs.

15. A sampling arrangement as defined in claim 1 wherein the arrangement further comprises

at least one additional sampling gate responsive to at least one input reference clock signals to form sampled reference clock signals wherein the at least one additional sampling gate is controlled by the same strobe source as the sampling gates responsive to the input signal pairs; wherein

the output sampled reference clocks after the sampling gates are converted to electrical signals using separate optical-to-electrical receivers followed by A/D conversion of each clock such that the digitally represented clock signals are used to create time-bases for the sampled signals to correctly position each sample in the reconstructed balanced detected versions of the input signal pairs.

16. A sampling arrangement as defined in claim 1 wherein the arrangement further comprises

a demodulation method for optical signals containing data encoded in the optical phase coupled to the input of the sampling arrangement, wherein after demodulation the input optical signal is split up in a number of input signal pairs containing the optical phase information converted into the form of amplitude modulation.

17. A sampling arrangement as defined in claim 16 wherein the demodulation method is designed to demodulate modulation formats selected from the group consisting differential phase shift keying (DPSK), differential quadrature phase shift keying (DQPSK), phase shift keying (PSK), quadrature phase shift keying (QPSK) or differential eight level phase shift keying (D8PSK)