METHOD AND SYSTEM FOR DECODING A MODULATED SIGNAL

Abstract

In a method and system for decoding a differential M-ary phase or quadrature amplitude modulated signal, the incoming signal is decoded according to a plurality of different decoding rules, wherein said plurality of decoding rules correspond to different values of a resulting frequency difference or mismatch between a signal frequency and a local oscillator reference frequency. The invention allows to increase a tolerance window for the maximal allowable frequency offset, and thus helps to speed up an initial locking process or to allow for equipment which has a lower tuning granularity.

Claims

1. A method for decoding a modulated signal, said method comprising: receiving an encoded signal, said encoded signal comprising a pair of an in-phase component and a quadrature component being extracted from a mixing signal, said mixing signal being obtained from mixing a received modulated signal at a first frequency with a local oscillator signal at a second frequency; and decoding said encoded signal according to a plurality of different decoding rules, said plurality of decoding rules corresponding to different values of a resulting frequency difference or mismatch between said first frequency and said second frequency.

2. The method according to claim 1, wherein said decoding step comprises a step of identifying a pre-determined reference outcome in said decoded signal.

3. The method according to claim 2, further comprising a step of determining said frequency difference or mismatch based on said identified reference outcome.

4. The method according to claim 2, further comprising a step of adjusting said frequency difference or mismatch in accordance with said identified reference 1 outcome, wherein said adjusting step preferably comprises a step of tuning a local oscillator that supplies said local oscillator signal.

5. The method according to claim 1, further comprising a step of selecting a decoding rule among said plurality of different decoding rules, and employing said selected decoding rule to decode subsequent encoded signals.

6. The method according to claim 1, wherein said decoding step comprises a step of assigning different information symbols to said encoded signal in accordance with different assignment rules.

7. The method according to claim 6, further comprising a step of comparing a plurality of sequences of said information symbols to a predetermined reference outcome, said plurality of sequences corresponding to said different assignment rules.

8. The method according to claim 1, wherein said decoding step comprises a step of assigning information symbols to said encoded signal in accordance with an assignment rule, and comparing said information symbols against a plurality of reference outcomes.

9. The method according to claim 1, wherein said modulated signal is an optical M-ary modulated signal, M being a positive integer, in particular an optical M-ary phase modulated signal or an optical M-ary quadrature amplitude modulated signal.

10. The method according to claim 1, wherein said in-phase component (32) and said quadrature component are extracted from said mixing signal by means of heterodyne detection.

11. A system for decoding a modulated signal, said system comprising: a receiving unit adapted to receive an encoded signal, said encoded signal comprising a pair of an in-phase component and a quadrature component being extracted from a mixing signal, said mixing signal being obtained from mixing a received modulated signal at a first frequency with a local oscillator signal at a second frequency; and a decoding unit adapted to decode said encoded signal according to a plurality of different decoding rules, said plurality of decoding rules corresponding to different values of a resulting frequency difference or mismatch between said first frequency and said second frequency.

12. The system according to claim 11, wherein said system comprises a local oscillator unit providing said local oscillator signal at said second frequency, and further comprises a mixing unit adapted to receive said modulated signal at said first frequency, to mix said modulated signal with said local oscillator signal and to extract said pair of said in-phase component and said quadrature component from said mixing signal, wherein said mixing unit is further adapted to provide said extracted pair to said decoding unit.

13. The system according to claim 11, wherein said decoding unit comprises a frame hunter unit adapted to compare said decoded signal to a pre-determined reference outcome.

14. The system according to claim 11, wherein said system (14) is adapted to implement a method according to claim 1.

15. A computer program product comprising computer-readable instructions that cause a computer coupled to a system according to claim 11 to implement on said system a method according to claim 1.

Description

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

[0043] The features and numerous advantages of the method and system for decoding a modulated signal according to the present invention will be best apparent from a detailed description of preferred embodiments with reference to the accompanying drawings, in which:

[0044] FIG. 1 is a schematic view of a system for transmitting a modulated signal which comprises a system for decoding the signal according to an embodiment of the present invention;

[0045] FIG. 2 is a schematic constellation diagram for an example in which two bits per symbol are transmitted, and with perfect frequency compensation;

[0046] FIG. 3 is a rotated constellation diagram that corresponds to an imperfect frequency compensation; and

[0047] FIG. 4 is a schematic diagram of a decoding system according to an embodiment of the present invention.

[0048] FIG. 1 is a schematic diagram of a system 10 for transmitting a quadrature amplitude modulated signal between a sender unit 12 and a receiver unit 14 over an optical channel 16, such as a fiber optical link. In the sender unit 12, two digital message bit streams 18, 18′ are encoded by changing (modulating) the amplitudes of two carrier signals 20, 20′. The two carrier signals 20, 20′, usually sinusoidal waves, are out of phase with each other by 90° and lead to quadrature amplitude modulation (QAM). In the sender unit 12, the modulated waves are summed up, and the resulting wave form is a combination of both phase-shift keying (PSK) and amplitude-shift keying (ASK), or phase modulation and amplitude modulation.

[0049] The resulting signal is sent via the optical channel 16 to the receiver unit 14. As illustrated in FIG. 1, the receiver unit comprises a mixing unit 22, a local oscillator unit 24 and a decoding unit 26. The mixing unit 22 receives the modulated optical signal at a first input 28, and has a second input 30 for a local oscillator signal supplied by the local oscillator unit 24. The local oscillator unit 24 may generate a local oscillator signal by means of a laser, and may provide the local oscillator signal to the second input 30, as is generally known in the art.

[0050] The mixing unit 22 serves as a coherent demodulator for demodulating the incoming optical signal. To this end, the incoming optical signal received at the first input 28 is superimposed with the local oscillator signal received at the second input 30. At the receiving photodiode, the superposition results in a current that is proportional to the product of the electrical field of the local oscillator laser and the electrical field of the incoming signal. In addition, the current oscillates with the difference of the angular frequency of the local oscillator signal and the incoming modulated signal,

I.sub.PD˜2*E.sub.LO*E.sub.signal*sin((ω.sub.LO−ω.sub.signal)*t+φ(t))

[0051] In this equation, I.sub.ND denotes the generated photocurrent, E.sub.LO the electric field of the local oscillator, E.sub.signal the electric field of the modulated incoming signal, ω.sub.LO the angular frequency of the local oscillator signal, and ω.sub.signal the angular frequency of the incoming modulated signal, t denotes the time and φ(t) describes the time-dependent phase which contains the transmitted data information.

[0052] As can be seen from the equation, the photodiode current depends on the differential angular frequency of the two light sources, ω.sub.LO−ω.sub.signal. This oscillating term is either completely deleted by equalizing both laser frequencies (so-called “homodyne coherent detection”), or it is electronically compensated by multiplying (“down-converting”) the received signal with a sine function and a cosine function with the difference frequency (“heterodyne detection”). In any case, the mixing unit 22 supplies a first signal 32 comprising the “in-phase” (I) component of the demodulated signal and a second signal 34 comprising the “quadrature” (Q) component of the demodulated signal. The in-phase component 32 and the quadrature component 34 are supplied to the decoding unit 26 for data recovery by means of digital signal processing (DSP).

[0053] In an M-ary differential modulation scheme, the transmitted information is encoded into the relative phase difference between two symbols, i.e., Δφ=φ(t.sub.2)−φ(t.sub.1). Depending on the number M of possible differential phases, log.sub.2(M) bits can be transmitted per symbol. As an example, for M=4 log.sub.2(4)=2 bits per symbol can be transmitted. In other words, for M=4, Δφ can have four different values, assuming an ideal signal and an ideal receiver. The bit stream is then reconstructed by assigning those four differential phase angle pairs to two data bits. An example is shown in the following Table 1:

TABLE-US-00001 TABLE 1 assignment of the detected bits for different values of φ φ (degrees) Bit “I” Bit “Q” 0 0 0 90 0 1 180 1 0 270 1 1

[0054] Table 1 merely shows one possible assignment, as an example. As long as a one-to-one correspondence between bit pairs {I,Q} and ranges of the phase difference φ is established, the assignment is completely arbitrary and can be chosen differently from Table 1. For instance, in an alternative example a “Gray Code” may be employed to assign the values such that only one bit value differs from line to line in Table 1.

[0055] In order to retrieve the phase difference y from the incoming signal, the decoding unit 26 samples the baseband signals 32, 34 preferably in the middle of the time slot assigned to each symbol. The phase of the optical signal field at the time t can then be calculated as φ.sub.t=arctan (I/Q). The phase difference can then be calculated as Δφ=φ(t.sub.2)−φ(t.sub.1), where Δt=t.sub.2−t.sub.1 denotes the symbol duration.

[0056] FIG. 2. shows an exemplary constellation diagram in which the I component and the Q component are plotted for a plurality of received signals, and bins or quadrants (delimited by broken lines) are assigned to bit pairs according to Table 1. The distribution of the signals shown in FIG. 2 is due to noise in the signal and the receiving equipment, which cannot be totally avoided in practice. The decoding unit 26 reconstructs the bit sequence from the constellation components I and Q according to Table 1 and outputs a pair of decoded signal bit streams 36, 36′ that reconstruct the message bit streams 18, 18′.

[0057] FIG. 2 represents a case of perfect frequency compensation, i.e., the difference angular frequency Δω=ω.sub.LO−ω.sub.signal=0. However, in practice, perfect frequency compensation can hardly be achieved, mainly due to fluctuations in the local oscillator laser or other device imperfections. An imperfect frequency compensation will result in a rotation of the received symbols by an angle α. This is schematically illustrated in FIG. 3.

[0058] The rotation angle α can be calculated as α=Δω*Δt (in radian). In other words, the frequency control can lock to multiples of Δω=n*π/(2*Δt). The value of a can be derived from the detected angles by for example determining the center-of-gravity of the symbol clouds of FIG. 3. Note, however, that it can only be determined modulo the number of possible phase angles. Hence, when four phase angles are possible, a can only be determined within +45°. Higher values of α fold back into the lower bins.

[0059] As described above with reference to FIG. 2 and Table 1, a certain bit pair is detected if the calculated differential phase angle lies within the boundaries of the respective bin. For instance, if a differential angle is between −45° and +45°, it lies within the 0° bin and hence is decoded according to Table 1 as the bit pair {0,0}, and similarly for the other bins {0,1}, {1,0}, and {1,1}. Thus, in a noiseless system, a can be allowed to vary between −45° and +45°. Depending on the signal-to-noise ratio of the system, in practice the allowable tolerance for a is smaller, usually at around +30°. For a digital quadrature phase shift keying system with a symbol rate of e.g. 622 Mbaud, this translates to a maximum allowable frequency offset of the local oscillator laser in the range of ±52 MHz. Hence, either the local oscillator laser or the down-converting frequency must be kept within this interval in order for the decoding to work reliably.

[0060] Controlling the local oscillator laser frequency or the down-converting frequency within approximately ±50 MHz is practically feasible, and has been shown to work in practice. However, when the system is started up, the local oscillator laser has to scan the available band for a signal. This scanning process needs to be relatively slow in order to get close enough to hit the right frequency window. A typical tunable laser covers an optical frequency band of about 4 THz or more. In order to scan that band with a resolution of about 100 MHz, about 40,000 steps are necessary. After each laser frequency step, the decoding unit 26 decodes the received bits and typically tries to recognize a specific pre-determined frame delimiter bit pattern in the received data stream. The bit stream consists of two bit pairs, and it is generally unknown where the bits start within this bit stream. In practice, four parallel frame hunter units may be employed to scan the bit stream for the frame delimiter pattern. Each frame hunter unit may use a different offset within the bit stream. Hence, even if the laser scans 100 steps per second, the full local oscillator laser scan could take up to 400 seconds in practice, which is too long for many practical applications.

[0061] The inventor found that the tolerance value for the maximum allowable frequency offset, and hence the speed of the initial locking process can be enhanced by adding one or more detection schemes that allow for higher values of the rotation angle α. Assuming that the rotation angle α exceeds 45°, all detected differential angles would fall into the wrong detection bin in FIG. 1. However, assuming that the angle α does not vary over time, this behavior is deterministic. For instance, assuming 45°<α≦135°, all detected signals are shifted clockwise by one bin, i.e., they fall into the neighboring bin along the clockwise direction. This can be accommodated by complementing Table 1 with a plurality of additional bit assignment patterns. For instance, three additional bit assignment patterns may be employed according to Table 2 to accommodate arbitrary values of a, wherein each decoding scheme corresponds to one detection bin in FIG. 1.

TABLE-US-00002 TABLE 2 three additional bit assignment patterns abs (α) ≦ 45° 45° < α ≦ 135° 135° < α ≦ 225° 225° < α ≦ 315° φ Bit Bit φ Bit Bit φ Bit Bit φ Bit Bit (degrees) “I” “Q” (°) “I” “Q” (°) “I” “Q” (°) “I” “Q” 0 0 0 0 1 1 0 1 0 0 0 1 90 0 1 90 0 0 90 1 1 90 1 0 180 1 0 180 0 1 180 0 0 180 1 1 270 1 1 270 1 0 270 0 1 270 0 0

[0062] As can be taken from Table 2, the first pattern for the range abs (α)≦45° is identical to Table 1. The three additional bit assignment patterns for the quadrants (i) 45°<α≦135°, (ii) 135°<α≦225° and (iii) 225°<α<315° are shifted versions of the first bit assignment pattern. Hence, these additional bit assignment patterns compensate for the rotation of the constellation by angles α with abs (α)>45°, and in effect rotate larger shift angles back into the first bin, so that bit values are assigned as if the absolute value of α were smaller than 45°.

[0063] The detected phases as such do not usually allow to infer the shift angle α. However, this may be achieved by checking the received signal for a pre-determined reference outcome, such as for a certain frame delimiter bit pattern which the sender unit 12 is known to transmit, as will now be explained with reference to FIG. 4.

[0064] FIG. 4 shows a detailed view of a decoding unit 26 according to an embodiment of the present invention. The decoding unit 26 comprises a copying unit 38, a plurality of frame hunter units 40a-40d, and a control unit 42.

[0065] The copying unit 38 copies the in-phase component 32 and the quadrature component 34 received from the mixing unit into four identical copies. These signal pairs are then supplied to four frame hunter units 40a to 40d that each analyze the incoming signals according to one of the four different bit assignment patterns in Table 2 and scan for the frame delimiter bit pattern. The frame delimiter pattern may be a pre-determined sequence of bits that the sender unit 12 is known to transmit as a reference pattern. Only one of the frame hunter units 40a to 40d will find the reference bit pattern, depending on the value of α, and will notify the control unit 42. For instance, assuming 45°<α≦135°, the frame hunter unit 40b that implements the second bit assignment pattern in Table 2 will detect the frame delimiter pattern, and will forward it to the control unit 42. The control unit 42 makes sure that all subsequent signals are decoded according to the corresponding bit assignment pattern, i.e., according to the second bit pattern in Table 2. The decoded bit values are output as the decoded signal bit stream 36, 36′.

[0066] Alternatively, in response to the range of a being detected by the control unit 42, the control unit 42 may send a control signal (not shown) to the local oscillator unit 24 to adapt the local oscillator frequency so that the shift angle α is reduced to the first quadrant abs (α)≦45°. All subsequent signals can then be decoded according to the first bit assignment pattern in Table 2, which corresponds to the original bit assignment pattern in Table 1.

[0067] In the embodiment described above with reference to FIG. 4 and Table 2, the frame hunter units 40a to 40d decode the incoming signal according to four different bit assignment schemes, and compare the outcome to a known frame delimiter pattern. Alternatively, the incoming signals 32, 34 may be decoded according to one fixed bit assignment pattern, such as the bit assignment pattern in Table 1 and may be compared to different (rotated) versions of the frame delimiter pattern. This provides another, equivalent means of determining the range of the shift angle α.

[0068] For simplicity, the examples described above assume M =4, and hence Δφ can have four different values. However, the same scheme can be employed for higher-order constellations that contain more data points.

[0069] In the examples described above, the invention has been described in the context of heterodyne detection. However, the same scheme can be employed for homodyne or intradyne detection.

[0070] The invention increases a tolerance window for the maximum allowable frequency offset, and hence can be used to achieve a faster initial scanning and locking to the transmitter data. At the same time, the invention allows to accommodate lasers with a lower tuning granularity, and hence enables faster and reliable coherent detection with less sophisticated and thereby cheaper equipment.

[0071] The description of the preferred embodiments and the accompanying drawings merely serve to illustrate the invention and the beneficial effects associated therewith, but should not be understood to imply any limitation. The scope of the invention is to be determined solely based on the appended set of claims.

Xieon Networks S.à.r.l.

X30344WO

[0072]

TABLE-US-00003 Reference Signs 10 system for transmitting a modulated signal 12 sender unit 14 receiver unit 16 optical channel 18, 18′ message bit streams 20, 20′ carrier signals 22 mixing unit 24 local oscillator unit 26 decoding unit 28 first input of mixing unit 22 for modulated optical signal 30 second input of mixing unit 22 for local oscillator unit 32 in phase (I) component of demodulated signal 34 quadrature (Q) component of demodulated signal 36, 36′ decoded signal bit stream 38 copying unit 40a, b, c, d frame hunter units 42 control unit