FIBER OPTIC COMMUNICATIONS USING OPTICAL SINGLE SIDEBAND TRANSMISSION AND DIRECT DETECTION
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to optical fiber communications, and more particularly, to optical fiber communications systems which use optical single sideband transmission and direct detection.
2. Description of the Related Art
As the result of continuous advances in technology, particularly in the area of networking, there is an increasing demand for communications bandwidth. For example, the growth of the Internet, home office usage, e-commerce and other broadband services is creating an ever-increasing demand for communications bandwidth. Widespread deployment of new bandwidth-intensive services, such as xDSL, will only further intensify this demand. Moreover, as data-intensive applications proliferate and data rates for local area networks increase, businesses will also demand higher speed connectivity to the wide area network (WAN) in order to support virtual private networks and high-speed Internet access. Enterprises that currently access the WAN through Tl circuits will require DS-3, OC-3, or equivalent or higher data rate connections in the near future. As a result, the networking infrastructure will be required to accommodate greatly increased traffic.
Optical fiber is a transmission medium that is well-suited to meet this increasing demand. Optical fiber has an inherent bandwidth which is much greater than metal-based conductors, such as twisted pair or coaxial cable. There is a significant installed base of optical fibers and protocols such as SONET have been developed for the transmission of data over optical fibers. Typical communications system based on optical fibers include a transmitter, an optical fiber, and a receiver. The transmitter converts the data to be
communicated into an optical form and transmits the resulting optical signal across the optical fiber to the receiver. The receiver recovers the original data from the received optical signal. Recent advances in transmitter and receiver technology have also resulted in improvements, such as increased bandwidth utilization, lower cost systems, and more reliable service.
However, current optical fiber systems also suffer from drawbacks which limit their performance and/or utility. For example, optical fibers typically exhibit dispersion, meaning that signals at different frequencies travel at different speeds along the fiber. More importantly, if a signal is made up of components at different frequencies, the components travel at different speeds along the fiber and will arrive at the receiver at different times and/or with different phase shifts. As a result, the components may not recombine correctly at the receiver, thus distorting or degrading the original signal. In fact, at certain frequencies, the dispersive effect can result in destructive interference at the receiver, thus effectively preventing the transmission of signals at these frequencies. For example, if the optical signal is a double sideband signal, the upper and lower sidebands are located at different frequencies and, as the result of dispersion, may destructively interfere during detection at a square-law receiver. Dispersion effects may be compensated by installing special devices along the fiber specifically for this purpose. However, the additional equipment results in additional power loss (e.g., insertion loss) as well as in additional cost, and different compensators will be required for different types and lengths of fiber. Other fiber effects, such as fiber nonlinearities, can similarly degrade performance.
In addition, although optical fibers have an inherently large bandwidth available for the transmission of data, constructing transmitters and receivers which can take advantage of this large bandwidth can be problematic. Current approaches, such as the on-off keying and time-division multiplexing of signals used in the SONET protocols, cannot be extended to higher speeds in a straightforward manner. This is because current electronics technology limits the speeds at which these approaches can be implemented and electronics fundamentally will not have sufficient bandwidth to fill the capacity of a fiber.
Even if this were not a limitation, current modulation schemes are also spectrally inefficient. For example, SONET is largely based on on-off keying. On-off keying
typically uses a train of square wave pulses, which results in large sidelobes in the frequency domain. As a result, the power spectrum does not drop off quickly and signals must be further separated in frequency in order to avoid interference. Separation means that the available bandwidth is not being used efficiently. More data can be transmitted in less bandwidth by using more efficient modulation schemes. As another example, double sideband modulation is typically used to impress the data-carrying electrical signal onto an optical carrier. This is partly because Mach-Zehnder modulators (MZMs) and other commonly available modulators inherently generate double sideband modulation. It is also partly because double sideband signals can be straightforwardly recovered using direct detection receivers, which are easier to implement given the currently available technology. Double sideband modulation, however, requires twice as much spectral bandwidth than single sideband modulation, thus increasing the spectral inefficiency.
Current optics technology also prevents the full utilization of a fiber's capacity. For example, in wavelength division multiplexing, signals are placed onto optical references of different wavelengths and all of these signals are transmitted across a common fiber. However, the components which combine and separate the different wavelength signals currently place a lower limit on the spacing between wavelengths, thus placing an upper limit on the number of wavelengths which may be used.
The ever-increasing demand for communications bandwidth further aggravates many of the problems mentioned above. In order to meet the increasing demand, it is desirable to increase the data rate of transmission across each fiber. However, this typically can only be achieved by either increasing the bandwidth being utilized and/or by increasing the spectral efficiency of the transmission scheme. Increasing the bandwidth, however, aggravates frequency-dependent effects, such as dispersion. Increasing the spectral efficiency requires moving to different transmission schemes.
Thus, there is a need for optical communications systems which more fully utilize the available bandwidth of optical fibers. There is further a need to reduce or eliminate the deleterious effects caused by fiber dispersion and to combat the many drawbacks mentioned above.
SUMMARY OF THE INVENTION
In accordance with the present invention, an optical communications system is for transmitting information across an optical fiber. The system includes a single sideband transmitter coupled to a direct detection receiver via the optical fiber. The single sideband transmitter generates an optical data signal, which includes an optical carrier and a primary optical sideband containing the information to be transmitted. The non-primary optical sideband is either attenuated or non-existent. The optical data signal is transmitted from the single sideband transmitter to the direct detection receiver, which recovers the information from the received optical data signal.
In one aspect of the invention, the direct detection receiver includes a square law detector (e.g., a PIN diode) coupled to a demodulator. The square law detector mixes the optical carrier with the primary optical sideband to produce an electrical signal which includes a frequency down-shifted version of the primary optical sideband. The demodulator recovers the information from the frequency down-shifted version of the primary optical sideband. Here, the optical carrier plays a role analogous to that of a local oscillator in a heterodyne detection system. In a further refinement of this case, the frequency down-shifted version of the primary optical sideband includes a tone and an electrical sideband containing the information. The demodulator includes an electrical mixer, which mixes the tone with the electrical sideband to produce a frequency component containing the information (e.g., a component located at the difference frequency of the tone and the electrical sideband).
In another aspect of the invention, the optical carrier is coherent with the primary optical sideband. The direct detection receiver includes a square law detector for coherently mixing the optical carrier with the primary optical sideband to produce an electrical signal containing the information.
In yet another aspect, the single sideband transmitter receives an optical reference located at a frequency fo and an electrical reference located at a frequency fc'. In this aspect, the optical carrier is located substantially at a frequency fo + j fc' / k, where j and k are integers. The optical carrier and the primary optical sideband are separated in frequency by an amount Δf which is substantially equal to a non-zero integer multiple of fc' / k.
Another aspect of the invention concerns specific designs for the single sideband transmitter or for the direct detection receiver. For example, in one approach, the single sideband transmitter is based on an electrical modulator coupled to an optical single sideband modulator. In another approach, an optical frequency shifter and an optical modulator are coupled in series. In a third approach, a source produces an optical reference which is split into two optical references, one used to produce the optical carrier and the other to produce the primary optical sideband. In yet another approach, the two optical references are generated by two wavelength-locked optical sources rather than by splitting a single optical reference. These and other approaches, including approaches for the direct detection receiver, are more fully discussed below.
The present invention has many advantages. For example, the use of a single optical sideband significantly reduces the effects of dispersion. With double optical sidebands, dispersion can lead to destructive interference between the two sidebands during detection at a square-law receiver. If the two optical sidebands are of equal strength, then the destructive interference can virtually eliminate the entire signal if the dispersion causes a relative phase shift of approximately π between the two sidebands. In the single sideband case, the non-primary optical sideband is either attenuated or eliminated, so the effect of destructive interference is significantly reduced. Transmission of a single optical sideband also requires half the spectral bandwidth compared to double optical sidebands but conveys the same amount of information. Thus, the overall spectral efficiency is roughly doubled.
In addition, the overall transmission scheme uses a direct detection receiver but is able to obtain many of the performance advantages of heterodyne detection. One advantage of this approach is that a direct detection receiver is simpler than a heterodyne receiver. For example, no local oscillator is required. Another advantage is that in the heterodyne case, the local oscillator should be polarization-matched to the received optical sideband in order to achieve high mixing efficiency. This is problematic since there is no inherent relationship between the polarizations of the primary optical sideband and that of the local oscillator since one originates at the transmitter and the other at the receiver. In contrast, when the optical carrier plays a role analogous to the local oscillator, it is easier to achieve this polarization matching since both signals are generated by the transmitter (and may even
be derived from the same optical reference) and propagate through the optical fiber together.
In further accordance with the invention, a method for transmitting information across an optical fiber includes the following steps. An optical data signal is generated. The optical data signal includes an optical carrier and a primary optical sideband which contains the information to be transmitted. The optical data signal is transmitted across the optical fiber. The information is recovered from the transmitted optical data signal using direct detection.
BRIEF DESCRIPTION OF THE DRAWING
The invention has other advantages and features which will be more readily apparent from the following detailed description of the invention and the appended claims, when taken in conjunction with the accompanying drawing, in which:
FIG. 1 is a block diagram of a system 100 according to the present invention;
FIG. 2 is a flow diagram illustrating a method 200 for transmitting an information signal across a fiber, according to the present invention;
FIG. 3 is a diagram of one embodiment of optical transmitter 114 using an electrical modulator;
FIGS. 4A-4B are diagrams of embodiments of optical transmitter 114 using an optical frequency shifter and an optical modulator coupled in series;
FIGS. 5A-5C are diagrams of embodiments of optical transmitter 114 using two optical references;
FIG. 6A is a block diagram of one embodiment 690 of demodulator 190 based on squaring a signal containing a tone and a sideband;
FIG. 6B is a block diagram of one embodiment 635 of direct detection receiver 130 based on squaring the received optical data signal;
FIG. 7A is a block diagram of another embodiment 790 of demodulator 190 based on multiplying a tone with a sideband;
FIGS. 7B-7D are block diagrams of more embodiments 735 of direct detection receiver 130;
FIG. 8 is a block diagram of yet another embodiment 890 of demodulator 190 using separate extraction paths to process different sidebands;
FIG. 9 is a block diagram of one embodiment 990 of demodulator 890 based on multiplying a tone with a sideband;
FIG. 10 is a block diagram of another system 1100 according to the invention;
FIG. 11 is a graph illustrating the spectra of the composite optical signal for system
1100;
FIG. 12 is a block diagram of yet another system 1200 according to the invention; and
FIGS. 13A-13B are graphs illustrating the spectra of various signals in system 1200.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
FIG. 1 is a diagram of a system 100 according to the present invention. System 100 includes a single sideband (SSB) transmitter system 110 coupled to a direct detection receiver system 130 by optical fiber 120. The SSB transmitter system 110 preferably includes an optical source 112, an electrical oscillator 113, and a SSB optical transmitter 114. The SSB optical transmitter 114 includes an optical reference port 122 coupled to the optical source 112 and also includes a data port 124. In the preferred embodiment shown in FIG. 1, the SSB optical transmitter 114 also includes an electrical reference port 123 coupled to the electrical oscillator 113. In a preferred embodiment, the direct detection receiver system 130 includes a square law detector 137 coupled to a demodulator 190, although other approaches will be apparent as will be discussed further below. System 100
is used to transmit information from transmitter system 110 to receiver system 130 via fiber 120.
With reference to the flow diagram of FIG. 2 as well as to FIG. 1, system 100 operates as follows. The frequency spectrum of an example electrical data signal is shown by spectrum 140, which is characterized by a frequency fs. The frequency fs could be zero, for example, if the electrical data signal is based on on-off keying. The electrical data signal 140 contains the information to be transmitted and maybe any of a variety of signals. For example, it may be a single high speed data stream. Alternately, it may contain a number of data streams which are time-division multiplexed together, for example, if 64 OC-3 data streams are combined together to form a single OC-192 signal, which serves as the basis for the electrical data signal 140. As another example, the electrical data signal may include a number of constituent signals, each of which occupies a different frequency band within spectrum 140. In other words, the constituent signals may be frequency division multiplexed together. Other types of electrical data signals 140 and methods for combining constituent signals to form the electrical data signal 140 will be apparent. For clarity, the electrical data signal 140 will always be represented in this discussion by a single frequency lobe although it is to be understood that this is not necessarily the case.
Optical source 112 generates an optical reference 132 at a frequency fo, which is received 202 by optical transmitter 114 via its optical reference port 122. Examples of optical sources 112 include solid state lasers and semiconductor lasers. The electrical data signal 140 is also received 206 by optical transmitter 114 via the data port 124.
The optical transmitter 114 generates 208 an optical data signal 142 from the received signals. The optical data signal 142 includes an optical carrier 146 and exactly one primary optical sideband 144. The optical carrier 146 is derived from the optical reference 132 and the optical sideband 144 contains the information to be transmitted. Throughout this disclosure, the primary optical sideband 144 typically will be shown as the upper sideband, but it should be understood that the lower sideband may be used as well. Furthermore, for clarity as in FIG. 1, the discussion will often depict the non-primary optical sideband as non-existent. It should be understood that this is not a requirement of the invention. In many embodiments, it is sufficient to simply attenuate the non-primary
optical sideband relative to the primary sideband. For example, any reduction of the non- primary optical sideband will lead to a corresponding reduction in destructive interference due to dispersion.
In the preferred embodiment shown in FIG. 1, the electrical oscillator 113 generates an electrical reference 133 at a frequency fc', which is received by optical transmitter 114 via its electrical reference port 123. The electrical reference 133 is used to separate the optical carrier 146 from the primary optical sideband 144 in frequency. Typically, the frequency offsets and frequency separations in system 100 will be based on a reference frequency fc, which in turn will be based on the electrical reference fc'. For example, in this particular embodiment, fc = fc', the optical carrier 146 is located substantially at a frequency fo + m fc where m is an integer, and the optical carrier 146 and the optical sideband 144 are separated in frequency by an amount Δf which is substantially equal to a non-zero integer multiple of fc.
In the example of FIG. 1, the optical carrier 146 is located at the same frequency fo as the optical reference 132 and the optical sideband 144 is located at the frequency fo + fc. As will be apparent from the following examples, this is not necessarily the case. For example, the optical sideband 144 could just as well be located at fo - fc or the offset Δf could be any other non-zero integer multiple of fc, although offsets Δf of + fc and - fc are generally preferred. Similarly, the optical carrier 146 could be located at any of the frequencies fo + m fc, although the frequencies fo - fc, fo, and fo + fc are generally preferred. In addition, although in this and all of the following examples, fc = fc', this is not a requirement of the invention. It is used purely for convenience. For example, alternate embodiments may use various harmonics and/or subharmonics to generate the frequency offsets from electrical reference fc'. In one approach, harmonic or subharmonic generators may be used to mix the electrical reference fc' up to its harmomc frequencies or down to its subharmonics, with the harmonic or subharmonic effectively used as the reference frequency fc. Alternately, the mixing processes in optical SSB transmitter 114 may be based on harmonics or subharmonics of fc'. In all of these cases, a more accurate statement is that the optical carrier 146 is located substantially at a frequency fo + j fc' / k where j and k are integers, and the optical carrier 146 and the optical sideband 144 are separated in frequency by an amount Δf which is substantially equal to m fc' / n where m and n are
integers. However, even these mathematical relationships are not required since methods other than those based on harmonics and subharmonics may be used. Furthermore, although a significant portion of this discussion will follow the mathematical relationships set forth above, this is not meant to limit the invention. For example, alternate embodiments may generate the optical data signal 142 in other ways or even without the use of an electrical reference 133.
Note that the term "sideband" in the phrase "primary optical sideband 144" refers to the relationship between the optical sideband 144 and the optical carrier 146. For example, referring to the example of FIG. 1, the optical data signal 142 includes primary optical sideband 144 located at fo + fc, which means that the non-primary sideband located at fo - fc would be attenuated with respect to the primary optical sideband 144. Although the non- primary optical sideband will often be described as being eliminated or as non-existent, it is to be understood that this means that the non-primary optical sideband is not especially relevant to the recovery of the information by direct detection receiver system 130. For example, the non-primary optical sideband may be severely attenuated or may simply be filtered out or otherwise ignored by receiver system 130. As a result, the optical data signal 142 shall often be referred to as a single sideband (SSB) signal.
This label does not imply, however, that the optical data signal 142 itself only includes SSB components. For example, the optical data signal 142 in FIG. 1 includes a double sideband (DSB) modulated version of the original electrical data signal 140, with upper sideband 145U, lower sideband 145L, and its own carrier 147. The two sidebands 145 are located at fo + fc +fs and fo + fc -fs, respectively. Other embodiments will use other spectra. For example, the primary optical sideband 144 could include the two sidebands 145 but no carrier 147, or it could include just a single sideband (either 145L or 145U), with or without carrier 147. As another example, it could further include frequency tone(s) located at frequencies other than the carrier 147 frequency. These tones may be used by the direct detection receiver 130. As a final example, the primary optical sideband 144 could include multiple tone(s) and/or sideband(s) generated from multiple incoming electrical data signals 140. For further examples, see FIGS. 10 and 15 of co-pending U.S. Patent Application Serial No. 09/728,373, "Optical Communications Using Heterodyne
Detection," by Ting K. Yee and Peter H. Chang, filed Nov. 28, 2000, which subject matter is incorporated by reference herein in its entirety.
Nor does the SSB label mean that only SSB optical modulation may be used within optical transmitter 114 to produce the optical data signal 142. For example, optical transmitter 114 may utilize DSB optical modulators, as will be the case in several examples below, so long as the final optical data signal 142 has only a single primary optical sideband 144.
Optical transmitter 114 may be conceptualized as implementing three functions. First, it produces the optical carrier 146. In preferred embodiments, the optical carrier 146 is based on the optical reference 132 received from source 112, possibly frequency shifted by mixing with the electrical reference 133 from the oscillator 113. Second, the optical transmitter 114 produces the primary optical sideband 144, which contains the inforaiation to be transmitted, and attenuates or eliminates the non-primary optical sideband. In preferred embodiments, this is achieved by modulating an optical signal (typically a signal containing either the optical reference 132 or a frequency-shifted version of the optical reference 132) with the electrical data signal 140 received via data port 124. Third, the optical transmitter 114 separates the optical carrier 146 and optical sideband 144 in frequency. This typically is achieved by proper selection of the frequencies for the previous two functions. Note that while this three function model is useful for understanding the function of optical transmitter 114, it is not meant to imply that each of these functions is implemented separately by separate components within optical transmitter 114. As will be seen below, in many embodiments, the functions are implemented simultaneously by components working in concert with each other.
The use of SSB transmission has many advantages. First, it significantly reduces the effects of dispersion. With double optical sidebands, dispersion can lead to destructive interference between the two sidebands during detection at a square-law receiver. If the two optical sidebands are of equal strength, then the destructive interference can virtually eliminate the entire signal if the dispersion causes a relative phase shift of approximately π between the two sidebands. In the single sideband case, the non-primary optical sideband is either attenuated or eliminated, so the effect of destructive interference is significantly
reduced. Second, the transmission of one sideband requires half the spectral bandwidth as the transmission of two sidebands but conveys the same amount of information. Thus, the overall spectral efficiency is roughly doubled.
Returning again to FIG. 2, the single sideband optical data signal 142 is transmitted 230 over fiber 120 to direct detection receiver system 130. Current optical fibers have two spectral regions which are commonly used for communications: the 1.3 and 1.55 micron regions. At a wavelength of 1.3 micron, transmission of the optical signal is primarily limited by attenuation in the fiber 120; dispersion is less of a factor. Conversely, at a wavelength of 1.55 micron, the optical signal will experience more dispersion but less attenuation. Hence, the optical data signal preferably has a wavelength either in the 1.3 micron region or the 1.55 micron region and, for long distance communications systems, the 1.55 micron region is generally preferred.
The direct detection receiver system 130 receives 235 the transmitted optical data signal 142 and then recovers 240 an electrical signal 170 which contains the original information. In the specific embodiment shown in FIG. 1, the square law detector 137, preferably a PIN diode, receives 235 the incoming optical data signal 142. Demodulator 190 then recovers 240 an electrical signal 170 which contains the original information. In FIG. 1, the recovered electrical signal 170 is the same as the original electrical data signal 140, although this is not necessarily the case.
In one embodiment, the optical carrier 146 plays a role similar to that of a local oscillator in a heterodyne system. Square law detector 137 produces a photocurrent which is proportional to the intensity of signal 142, which effectively mixes together the various frequency components in spectrum 142. The resulting electrical signal has a number of frequency components located at different frequencies, with the components of interest shown by spectrum 150. Spectrum 150 is similar to spectrum 142, but frequency downshifted to a frequency Δf, which is the difference between the frequency of the optical carrier frequency 146 and that of the optical sideband 144. Note that in the example of FIG. 1, both sidebands 155L and 155U and carrier 157 have been frequency downshifted compared to optical data signal 142. Demodulator 190 then further processes spectrum 150 to recover the original information.
More specifically, in one approach, the demodulator 190 uses the carrier 157 as a reference tone. It mixes at least one of the sidebands 155 with the tone 157 to produce a number of frequency components, including one frequency component 170 located at the difference frequency between the relevant sideband 155 and tone 157. This difference component 170 contains the electrical data signal 140, although it may be offset in frequency from the original frequency fs, depending on the frequencies of the sideband 155 and tone 157. Frequency components other than the difference component 170 may be used to recover the electrical data signal. For example, the mixing typically also produces a sum component located at the sum of the frequencies of the relevant sideband 155 and tone 157, and the electrical data signal 140 may be recovered from this sum component rather than the difference component. If more than one sideband 155 is processed by demodulator 190, each sideband 155 preferably is processed separately from the others in a manner which prevents destructive interference between the sidebands, although in many applications the sidebands 155 may be close enough in frequency that destructive interference due to dispersion is not a significant factor.
Each of FIGS. 3-5 includes both a particular embodiment of optical transmitter 114 and a specific example of that embodiment. The details of the specific example are not intended to limit the general description of the embodiment. FIGS. 6-9 are examples of various embodiments of demodulator 190 and/or direct detection receiver 130. Many of these embodiments are illustrated using the example of FIG. 1 in which optical data signal 142 includes two sidebands 145 located on either side of tone 147. The invention, however, is not limited to this specific example. Modulation schemes besides double sideband may be used (e.g., single sideband). Similarly, the tone 147 may be located at other frequencies and/or multiple or no tones 147 may be used. The examples in FIGS. 3-9 are used purely to illustrate the principles of various aspects of the invention. Other embodiments and corresponding modifications to the examples will be apparent.
FIG. 3 is a diagram of one embodiment 314 of optical transmitter 114 using an electrical modulator. Transmitter 314 includes an electrical modulator 310 coupled to an optical single sideband (SSB) modulator 320. The electrical modulator 310 is also coupled to the data port to receive the electrical data signal 140 and to the electrical reference port to
receive the electrical reference 133 at frequency fc'. The optical SSB modulator 320 is also coupled to the optical reference port to receive the optical reference 132 at frequency fo.
Optical transmitter 314 operates as follows. The electrical modulator 310 modulates the electrical reference 133 with the electrical data signal 140 to produce an intermediate electrical signal 330. The intermediate electrical signal 330 includes a component which contains the information to be transmitted and is located substantially at the frequency fc in this example. In the preferred embodiment of transmitter 314 shown in FIG. 3, the electrical modulator 310 is a mixer which produces a double sideband modulated version 330 of signal 140, including carrier. Spectrum 330 shows the two sidebands 331U and 331L centered about the carrier at frequency fc. Other types of electrical modulators 310 will be apparent. For example, the electrical modulator 310 could be a SSB modulator rather than a DSB modulator. Alternately, a filter could be used to attenuated one of the sidebands 331, effectively converting the DSB modulator to an SSB modulator. The modulated signal 330 may or may not include its own carrier at frequency fc and/or other tones at other frequencies.
Optical SSB modulator 320 receives the intermediate electrical signal 330 and uses it to SSB modulate the optical reference 132, thus producing the optical data signal 142. The SSB modulator 320 may be implemented in a number of ways. For example, acousto- optic modulators can be configured for SSB modulation. Alternately, other optical modulators which normally produce DSB modulation, such as Mach-Zehnder modulators (MZM) and electro-absorption modulators, can be combined with a filter which attenuates one of the two sidebands produced. This effectively results in a SSB modulator. For other examples of SSB modulators, see also U.S. Patent No. 5,301,058, "Single Sideband Optical Modulator for Lightwave Systems" by Olshansky; U.S. Patent No. 5,101,450, "Quadrature Optical Phase Modulators for Lightwave Systems" by Olshansky; M. Izutsu, et al., IEEE J. Quantum Electron. Vol. QE-17, No. 11, pp. 2225-2227, Nov. 1981; M. Izutsu, et al., Optics Letters, Vol. 7, No. 11, pp. 549-551, 1982; and S. Shimotsu, et al., "LiNbO3 Optical Single- Sideband Modulator," Postdeadline paper PD16-1, Optical Fiber Communications Conference (OFC 2000), March 7-10, 2000. The teachings of all of the foregoing are incorporated herein by reference. The filter need not be placed immediately after the DSB modulator. Instead, the unwanted sideband may propagate through a portion of the system
before being filtered out. The SSB modulation results in the single optical sideband 144 of signal 142. The optical carrier 146 is also produced by the SSB modulation. Many optical modulators are "leaky" or can be operated in a leaky mode, meaning that they will pass some of the incoming optical reference 132, resulting in optical carrier 146.
FIG. 3 shows a specific embodiment in which the optical SSB modulator 320 includes an MZM 322 coupled to a filter 324. An MZM, if biased at its quadrature point, will produce a large optical carrier in addition to a DSB signal. In theory, MZMs can also be biased to suppress the optical carrier although, in practice, it is difficult to entirely eliminate the optical carrier. Moving the bias away from this point results in a stronger optical carrier, although still significantly weaker than if the MZM were biased at its quadrature point. In FIG. 3, the MZM 322 is so biased. The result is a DSB signal 343U,343L with a reduced optical carrier 344. Filter 324 attenuates the lower sideband 343L, producing the optical data signal 142. The upper sideband 343U serves as the primary optical sideband 144 and the "leaked" optical reference 132 serves as the optical carrier 146. Note that the primary optical sideband 144 contains a double sideband modulated signal produced by the electrical modulator 310. The lower sideband 343L can also be used as the optical sideband 144.
FIGS. 4A-4B are diagrams of embodiments of optical transmitter 114 using an optical frequency shifter and an optical modulator (including both SSB and DSB optical modulators) coupled in series. The transmitter 314 of FIG. 3 was based on one electrical modulator and one optical modulator. The embodiments of FIGS. 4 are based on two optical modulators coupled in series (the frequency shifter is basically a type of optical modulator). Generally speaking, both embodiments 414 and 454 operate similarly. The series combination of optical modulator and frequency shifter is coupled to the optical reference port to receive the optical reference 132. The optical modulator is coupled to the data port and impresses the electrical data signal onto the optical stream. The optical frequency shifter is coupled to the electrical reference port and introduces the frequency separation between the optical sideband 144 and the optical carrier 146 of the optical data signal 142. The major difference between the two embodiments 414,454 is the order of these operations.
Turning first to FIG. 4A, in embodiment 414, the optical modulator 420 comes before the optical frequency shifter 430. In the specific example of transmitter 414 shown in FIG. 4A, the optical modulator 420 is an optical SSB modulator. In alternate embodiments, other optical modulators, including optical DSB modulators, may be used. The optical frequency shifter 430 is used primarily to accomplish frequency shifting of signals. Optical modulators can be used for this purpose and so are one example of optical frequency shifters. Since the frequency shifting is typically from one frequency location to a single other frequency location, optical SSB modulators are preferred. In the specific example of FIG. 4A, the optical frequency shifter 430 is a leaky optical SSB modulator. Other optical modulators, including those based on MZM and acousto-optic modulators, may also be used to implement optical frequency shifter 430.
The specific transmitter 414 shown in FIG. 4A operates as follows. Optical SSB modulator 420 modulates the optical reference 132 with the electrical data signal, producing an optical intermediate signal 422 which in this case is a SSB signal plus carrier. The leaky optical SSB modulator 430 frequency shifts the optical intermediate signal 422 by an amount +fc to produce the primary optical sideband 144 located at fo + fc. The sideband 144 could alternately be located at fo - fc or at fo + m fc for other values of m. The modulator 430 also leaks a portion of the SSB signal 422, and the carrier part of this leaked portion serves as the optical carrier 146.
In embodiment 454 of FIG. 4B, the optical frequency shifter 470 comes before the optical modulator 460. In the embodiment of transmitter 454 shown in FIG. 4B, the optical modulator 460 is a MZM. The optical frequency shifter 470 includes a leaky MZM 472 and a filter 474, but the filter 474 is placed after the MZM 460 rather than directly after MZM 472. Other embodiments of transmitter 454 may use other implementations of optical frequency shifter 470 and optical modulator 460. Amplifiers, EDFAs in this embodiment, may also be placed along the various signal paths to provide amplification.
The specific embodiment of transmitter 454 shown in FIG. 4B operates as follows. Leaky MZM 472 "modulates" the optical reference 132 with the electrical reference 133. The result is the intermediate optical signal 477, which includes an upper and lower "sideband" 473U,473L as well as the leaked carrier 474. Note that although the effect of
MZM 472 has been explained in terms of "modulation" and "sidebands," the net effect is that MZM 472 has frequency shifted the optical reference 132 by an amount +fc to location 473U. It has also frequency shifted the optical reference 132 by an amount -fc to location 473L (but this component will be filtered out by filter 474, which may also be positioned at different locations along the signal stream). The optical modulator 460 modulates the intermediate signal 477 with the electrical data signal. In this case, the MZM 460 produces three sets of DSB-modulated signals, one for each frequency in signal 477. The filter 474 attenuates the unwanted component 463L located at fo - fc, leaving the two components 464 and 463U located at fo and fo + fc, respectively. The carrier part of DSB signal 464 is used as the optical carrier 146 and the DSB signal 463U serves as the primary optical sideband 144. Note that the optical data signal 142 includes only a single optical sideband 144, even though that single sideband contains a DSB signal within it and optical DSB modulation was used by both MZMs 472 and 460 to produce the single sideband 144.
FIGS. 5A-5C are diagrams of embodiments of optical transmitter 114 using two optical references. In the examples of FIGS. 3-4, the optical carrier 146 was produced by allowing a single optical reference 132 (or a frequency-shifted version of the optical reference 132 in alternate embodiments) to leak through the modulators. In FIGS. 5A-5B, the incoming optical reference 132 is split into two optical references 521,522. One optical reference is used to produce the optical carrier 146 and the other is used to produce the primary optical sideband 144 (or precursors of these). These two components are then recombined, leading to the final optical data signal 142. In FIG. 5C, rather than splitting an incoming optical reference into two portions, two sources generate two separate optical references, one to produce optical carrier 146 and the other to produce the primary optical sideband 144.
In FIG. 5A, embodiment 514 includes a splitter 502, a carrier arm 504, a sideband arm 506, and an optical combiner 508. The optical splitter 502 is coupled to the optical reference port. Each of the carrier arm 504 and the sideband arm 506 is coupled to the optical splitter 502 on one end and to the combiner 508 on the other end. The sideband arm 506 is also coupled to the electrical reference port and to the data port. The splitting ratio of the optical splitter 502 may vary from one embodiment to the next, depending on other system considerations, particularly since amplifiers and/or attenuators may be used to adjust
the relative strengths of the signals in the two arms 504,506. In a preferred embodiment, in the final optical data signal 142, the power in the optical carrier 146 is roughly equal to that in the optical sideband 144. The optical combiner 508 preferably is a fiber coupler, due to its low cost and applicability to fiber systems. Other types of splitters and combiners may be used, including for example waveguide-based devices and those based on bulk optics.
Transmitter 514 operates as follows. The optical splitter 502 splits the optical reference 132 into two optical references 521,522. The optical reference 521 propagates through the carrier arm 504, which in this case is primarily a length of optical fiber, to the combiner 508. This is not meant to imply that carrier arm 504 cannot include other devices, as it often will. Similarly, sideband arm 506 may also include devices which are not explicitly shown in the figure. For example, both carrier arm 504 and sideband arm 506 likely may also include amplifiers or fixed or variable attenuators in order to adjust the power of optical reference 521 or 522, respectively. The optical reference 521 is used as the optical carrier 146 in the final optical data signal 142.
Optical reference 522 enters the sideband arm 506 and is used to produce the primary optical sideband 144. In this embodiment, the sideband arm 506 produces an optical sideband 144 which is frequency offset from the original carrier frequency by +fc. For example, since all of the embodiments shown in FIGS. 3-4 generate an optical sideband 144 located at fo + fc, they are all suitable for use as the sideband arm 506. The embodiments in FIGS. 3-4 also generate signals located at fo, which are not necessary for transmitter 514, so they may be improved for use as sideband arm 506 by eliminating this frequency component. This may be done, for example, by eliminating the leakiness which gives rise to this frequency component or by filtering out this component.
In the specific embodiment of transmitter 514 shown in FIG. 5 A, the transmitter of FIG. 4B is used as the sideband arm 506. MZM 472 preferably is no longer leaky and frequency shifts the optical reference 132 by an amount +fc to location 473U and by an amount -fc to location 473L (but this component is filtered out by filter 474). MZM 460 modulates each tone 473 with the electrical data signal, producing two DSB -modulated signals 463U and 463L. The filter 474 attenuates the unwanted component 463L, leaving the desired component 463U which is used as the optical sideband 144. Note that the filter
474 could be located in any number of places downstream of MZM 472, including between MZM 472 and MZM 460, between MZM 460 and combiner 508 and also after combiner 508 as is shown in FIG. 5A.
The optical combiner 508 combines the optical carrier 146 and the optical sideband 144 into the optical data signal 142.
In FIG. 5B, embodiment 554 includes a splitter 502, a carrier arm 504, a sideband arm 506, and an optical combiner 508 coupled in the same manner as in FIG. 5A. However, in this embodiment, the carrier arm 504 is coupled to the electrical reference port and includes an optical frequency shifter 560 for frequency shifting the optical reference 521 away from the reference frequency fo. The sideband arm 506 includes an optical modulator 570 for modulating the subcarrier 522 with the electrical data signal. The result is an optical sideband 144 which is located substantially at the original reference frequency fo and a primary optical sideband 144 which is separated from fo.
More specifically, the specific embodiment of transmitter 454 shown in FIG. 5B operates as follows. In the carrier arm 504, the optical frequency shifter 560 includes a
MZM 562 coupled to a filter 564. The MZM 562 frequency shifts the optical reference 521 by an amount -fc to location 563L. It also frequency shifts the optical reference 521 by an amount +fc to location 563U, but this component is filtered out by filter 564. The result is a frequency component 563L at fo - fc which is used as the optical carrier 146. In the sideband arm 506, the optical modulator 570 is an optical SSB modulator, which applies SSB modulation to subcarrier 522. The combiner 508 combines the signals from the two arms 504,506, producing the optical data signal 142. Again, note that other types of frequency shifter and 560 and optical modulators 570 would be appropriate, including for example optical DSB modulators.
In addition, filter 564 may be placed in various locations, for example after the combiner 508 rather than before it as is shown in FIG. 5B. In fact, additional filters may be placed at various locations throughout the different embodiments of transmitter 114 in order to attenuate unwanted frequency components. For example, if the optical carrier 146 is located at frequency fo + m fc and the optical sideband 144 is located at frequency fo + n fc, it may be desirable to place filters at the frequencies fo + j fc, where j is not equal to m or n,
in order to attenuate unwanted components at these frequencies. Alternately, bandpass filters may be placed at the frequencies fo + m fc and/or fo + n fc.
Embodiment 584 in FIG. 5C is analogous to embodiment 514 of FIG. 5 A, except that the two optical references 521 and 522 are produced by two wavelength-locked optical sources 112A and 112B rather than splitting the output of a single optical source 112, as in FIG. 5 A. More specifically, in addition to the two optical sources 112A,112B and the corresponding wavelength locker 586, embodiment 584 also includes a carrier arm 504, a sideband arm 506, and an optical combiner 508. Each of the carrier arm 504 and the sideband arm 506 is coupled to one of the optical sources 112 on one end and to the combiner 508 on the other end. The sideband arm 506 is also coupled to the data port. The wavelength locker 586 is coupled between the two optical sources 112 and also coupled to the electrical reference port.
Transmitter 584 operates as follows. The wavelength locker 586 ensures that the frequency separation of the two optical sources 112 is generally constant. In the example of FIG. 5C, laser source 112A generates an optical reference 521 at frequency fo. The wavelength locker 586 receives taps from outputs of both optical sources 112A and 112B and provides feedback to laser source 112B to ensure that its output frequency is locked to approximately fo + fc. Phase-locked loops may be used as the basis of wavelength locker 586. In a preferred embodiment, the wavelength locker 586 is based on frequency-locked loops, including for example those disclosed in co-pending U.S. Patent Application Serial No. yyy, "Wavelength-Locking of Optical Sources," by Shin-Sheng Tarng, et al., filed on the same date as this application, which subject matter is incorporated by reference herein in its entirety. In alternate embodiments, the wavelength locker 586 could lock the output of optical source 112A to that of optical source 112B, or lock both outputs to a third master oscillator.
The remainder of transmitter 584 operates similarly to transmitter 514. Optical reference 521 propagates through the carrier arm 504, which in this case is primarily a length of optical fiber, to the combiner 508. Carrier arm 504 and sideband arm 506 often will also include devices, for example amplifiers and/or fixed or variable attenuators, for adjusting the power of optical reference 521 or 522. The optical reference 521 is used as
the optical carrier 146 in the final optical data signal 142. Optical reference 522 enters the sideband arm 506 and is used to produce the primary optical sideband 144. In this specific embodiment, the sideband arm 506 includes a MZM 460 (although other types of modulators may be used), which modulates the optical reference 522 located at frequency fo + fc to produce the DSB-modulated signal 463. This is used as the primary optical sideband 144. The sideband arm 506 may also include amplifiers and/or attenuators. The optical combiner 508 combines the optical carrier 146 and the optical sideband 144 into the optical data signal 142.
The example of FIG. 5C illustrates how the approach shown in FIG. 5 A, which uses a single optical source and an optical splitter, may be implemented using two optical sources instead. This principle is equally applicable to the other approaches shown in FIG. 5B and/or described in the text. In addition, all of the examples of FIGS. 3-5 are based on externally modulating an optical signal produced by a separate optical source, but the principles illustrated are also applicable to implementations based on internally modulating an optical source.
Moving now to the receiver side, the exact design of the direct detection receiver 130 and demodulator 190 will depend in part on the structure of optical data signal 142. FIGS. 6 A, 7 A, 8 and 9 are embodiments which continue the specific example of FIG. 1, in which the primary optical sideband 144 itself includes two sidebands 145 and a tone 147 (which in this example is the original carrier for the two sidebands 145, although the tone(s) could be located at other frequencies in alternate embodiments). In these examples, the optical carrier 146 plays a role similar to a local oscillator in a heterodyne detection receiver and is used to frequency down-shift the primary optical sideband 144. The demodulator 190 then mixes the down-shifted sidebands 155 with tones 157 to recover the original information. In the examples of FIGS. 6 A, 7 A, 8 and 9, the two sidebands 155 are processed separately in order to avoid dispersion effects. However, in alternate embodiments, they could be processed together using standard double sideband demodulation techniques.
The approach of using the optical carrier 146 to play a role similar to that of a local oscillator has several advantages over traditional heterodyne detection systems (i.e., those
which use a local oscillator). One advantage is that in either approach, the primary optical sideband 144 should be polarization matched to the optical carrier 146 or local oscillator, respectively, in order to achieve high mixing efficiency. In traditional heterodyne systems, this is problematic since there is no inherent relationship between the polarization of the primary optical sideband 144 and that of the local oscillator, particularly since the primary optical sideband 144 is generated by the SSB transmitter 110 and the local oscillator would be generated by the receiver. In the current approach, the optical carrier 146 and the primary optical sideband 144 are both generated by the SSB transmitter 110. Hence, it often will be easier to polarization match these two components, for example if they are both based on the same optical reference 132, and propagation through fiber 120 typically will not significantly degrade this matching. Another advantage is that the receiver system 130 is simplified since a local oscillator is not required.
The double sideband plus tone example is used in FIGS. 6 A, 7A, 8 and 9 because it illustrates principles which can be applied to other types of optical data signals 142. For example, the demodulators of FIGS. 6A and 7A would be directly applicable if the primary optical sideband 144 contained only a single sideband (say 145U) and tone 147, since these two examples specifically concern the processing of a single sideband plus tone. Furthermore, the principles illustrated in these examples may easily be extended to primary optical sidebands 144 which contain sidebands from multiple signals and/or multiple tones.
In an alternate embodiment, when the optical carrier 146 and tone are substantially phase-locked, for example if they are both from the same laser source, the optical carrier 146 plays the role of a tone with respect to the sideband. In this case, the general principles illustrated in FIGS. 6-9 are also applicable by recognizing that in these embodiments the optical carrier 146 functions as the tone 147. In particular, FIGS. 6B and 7B illustrate demodulators for this case.
FIG. 6A is a block diagram of one embodiment 690 of demodulator 190 based on squaring a signal containing a tone and a sideband. Here, the term "sideband" is meant to refer, for example to sidebands 155 rather than optical sideband 144. Demodulator 690 includes a bandpass filter 610, a square law device 620, and a low pass filter 630 coupled in series. The filters 610, 630 may be implemented in many different ways, for example, by a
DSP chip or other logic device implementing a digital filter, a lump LC filter, a surface acoustic wave filter, a crystal-based filter, a cavity filter, or a dielectric filter. Other implementations will be apparent. The square law device 620 also may be implemented in many different ways. A diode is one common implementation.
Demodulator 690 recovers the electrical data signal 140 from electrical signal 150 as follows. Bandpass filter 610 frequency filters one of the sidebands and one of the tones from electrical signal 150. In this example, signal 150 includes two sidebands 155 and tone 157. Bandpass filter 610 passes the upper sideband 155U and the tone 157, and blocks the lower sideband 155L, thus producing spectrum 660. The square law device 620 squares the filtered components 660, resulting in spectrum 670. Spectrum 670 includes frequency components 672 located at the difference of frequencies between sideband 155U and tone 157, and also frequency components 674 located at the sum of these frequencies. Low pass filter 630 selects the difference components 672, thus recovering the electrical data signal 140.
Selection of the difference components 672 rather than the sum components 674 is advantageous because the difference components 672 typically are located at a lower frequency, allowing for simpler electronics. In addition, processing a single sideband 155U, rather than both sidebands 155U and 155L together, prevents any potential destructive interference between the sidebands, as may be caused by frequency dispersion effects.
Note that demodulator 690 is also directly applicable to cases in which the primary optical sideband 144 contains a single sideband plus tone. For example, referring to FIG. 1, consider the case where the primary optical sideband 144 does not include lower sideband 145L, then the signal 150 entering the demodulator 690 would be the same as signal 150 but without lower sideband 155L. It would still contain the tone 157 and upper sideband 155U. The lack of lower sideband 155L, however, does not affect the operation of demodulator 690 and, therefore, demodulator 690 will correctly process the single sideband signal.
Referring to FIG. 6B, now consider a situation in which the optical data signal 142 only includes the optical carrier 146 and one sideband, say the upper sideband 145U. In other words, it does not include either the lower sideband 145L or the tone 147. In this case, the optical data signal 142 of FIG. 6B is similar to signal 660 in FIG. 6 A except that
the optical carrier 146 plays the role of the tone 157, and the optical data signal 142 is located at optical frequencies whereas the signal 660 is typically at RF frequencies. Because of this similarity, the basic principles illustrated by demodulator 690 may be used to implement a direct detection receiver 130 for this type of optical data signal 142, as is shown by direct detection receiver 635 in FIG. 6B.
Specifically, direct detection receiver 635 includes a square law detector 136 coupled to an optional low pass filter 630. It operates analogously to demodulator 690. The square law detector 136 plays a role analogous to square law device 620. It mixes the incoming signals, resulting in difference component 672. Note that the optical carrier 146 and primary optical sideband 144 preferably are coherent to produce a proper difference component. The low pass filter 630 selects the difference components 672, thus recovering the electrical data signal 140 and eliminating higher order square-law products that may result from the MZM. In this case, the difference components 672 may be located at fairly high frequencies, so intermediate mixing may be desirable in order shift them to lower frequencies.
FIG. 7A is a block diagram of another embodiment 790 of demodulator 190 based on multiplying a tone with a sideband. This extractor 790 includes two bandpass filters 710 and 712, a multiplier 720 and a low pass filter 730. The two bandpass filters 710, 712 are each coupled to receive the incoming electrical signal 150 and are coupled on their outputs to multiplier 720. The multiplier is coupled to low pass filter 730.
Bandpass filter 710 selects a tone 157 and bandpass filter 712 selects one of the sidebands 155. In this specific example, the tone 157 and upper sideband 155U are the selected components. Multiplier 720 multiplies the tone 157 against the selected sideband 155U, resulting in a signal with a sum component 774 and a difference component 772, as in FIG. 6. Low pass filter 730 selects the difference component 772, thus recovering the electrical data signal 140. For the same reasons as demodulator 690, demodulator 790 is also directly applicable to cases in which the primary optical sideband 144 contains a single sideband plus tone (e.g., no sideband 155L).
FIG. 7B is a block diagram of a direct detection receiver 735 which is analogous to demodulator 790. The receiver 735 includes a wavelength-selective optical splitter 702.
One arm of the optical splitter 702 is coupled to an optical combiner 708 via an optical amplifier 704. The other arm is coupled directly to the optical combiner 708. The output of the optical combiner 708 is coupled to the receiver 635 of FIG. 6B.
In this case, as with FIG. 6B, the optical data signal 142 includes a coherent optical carrier 146 and single sideband, say the upper sideband 145U. However, unlike FIG. 6B, the optical carrier 146 is weaker than would be preferred for direct input to receiver 635. This may be done, for example, in order to reduce the overall power transmitted over fiber 120 or to maintain certain power ratios between the optical carrier 146 and primary optical sideband 144. The direct detection receiver 735 effectively amplifies the optical carrier 146 with respect to the primary optical sideband 144 and then feeds this modified signal to receiver 635.
More specifically, the optical splitter 702 sends the optical carrier 146 to the amplifier 704. A narrowband amplifier may be used to also introduce some wavelength selectivity. The amplified carrier is received by the optical combiner 708, which also receives the primary optical sideband 144 via the other arm of optical splitter 702. The amplified carrier and primary optical sideband are recombined by optical combiner 708 and then fed to receiver 635 to be processed as previously described.
FIG. 7B illustrates one approach for amplifying the optical carrier 146 relative to the primary optical sideband 145U. Other approaches will be apparent. For example, in FIG. 7C, the optical splitter 702 is implemented as a conventional optical splitter 762, one output of which is coupled to a narrowband filter 764. The splitter 762 divides the incoming optical data signal 142 into two signals. The narrowband filter 764 attenuates the optical sideband 145U, thus leaving primarily the optical carrier 146. In a variation of FIG. 7C, a narrowband amplifier, such as a Brillouin amplifier, is used in place of the narrowband filter 764 and amplifier 704. In this case, the narrowband amplifier primarily amplifies the optical carrier 146, boosting its strength relative to the optical sideband 145U. Generally speaking, in these embodiments, if signals from the two arms are recombined by optical combiner 708 (e.g., if a sideband 145U propagates through each arm), then they preferably are phase matched in order to avoid destructive interference.
In FIG. 7D, the entire splitter-combiner portion is replaced by a narrowband amplifier 766. Again, the amplifier 766 primarily amplifies the optical carrier 146, achieving the desired result.
FIG. 8 is a block diagram of yet another embodiment 890 of demodulator 190 using separate extraction paths for different sidebands. Example 890 includes two extraction paths 850A and 850B, and a combiner 860. Each extraction path 850 receives the incoming electrical signal 150 and is coupled on the output side to combiner 860.
Each extraction path 850 processes a different sideband within the electrical signal 150 to recover electrical data signals 140A and 140B, respectively. As an example, extraction path 850A might process the upper sideband 155U; whereas extraction path 850B processes the lower sideband 155L. Both extraction paths 850 may use the same tone in their processing, or they may use different tones. Combiner 860 receives the recovered electrical data signals 140A and 140B and constructively combines them to produce a resultant difference component 140C, which contains the original electrical data signal. The difference components 140A and 140B typically may be phase shifted with respect to each other in order to align their phases before they are combined; the amount of the phase shift may be frequency-dependent. If difference components 140 are located at difference frequencies, combiner 860 may also frequency shift them to a common frequency before combining.
In a preferred embodiment, each path 850 is based on the approach of demodulator
690 of FIG. 6A, except that each extraction path 850 is designed to process a different sideband. Thus, for example, the bandpass filter 610 for extraction path 850A may be tuned to select the optical tone 157 and upper sideband 155U; whereas the bandpass filter 610 for extraction path 850B might select the tone 157 and lower sideband 155L. Alternately, each extraction path 850 may be based on the approach of demodulator 790 of FIG. 7A.
FIG. 9 is a block diagram of one embodiment 990 of demodulator 890 in which the extraction paths 850 share components, although the sidebands are still processed separately. In this embodiment, each of the extraction paths 850 is based on demodulator 790. Extraction path 850A processes the upper sideband 155U; whereas extraction path 850B processes the lower sideband 155L. Both extraction paths use the same tone 157.
Hence, they may share a common bandpass filter 710, which selects the tone 157. In other words, the extraction paths are overlapping. The tone 157 is then fed to both multipliers 720 in each respective extraction path 850.
Combiner 860 includes a phase shifting element 912 and an adder 914. Phase shifting element 912 phase shifts the difference component 140A produced by extraction path 850A so that it is in phase with the difference component 140B produced by extraction path 850B. Adder 914 then adds the two in-phase components to produce the resulting difference component 140C.
In FIGS. 3-9, the tone 157 was located between the two sidebands 155. This is a typical result if the carrier used to produce the sidebands 155 is also used as the tone 157. However, the tone 157 may also be located at other frequencies. For example, see FIGS. 10 and 15 of co-pending U.S. Patent Application Serial No. 09/728,373, "Optical Communications using Heterodyne Detection," by Ting K. Yee and Peter H. Chang, filed Nov. 28, 2000, which subject matter is incorporated by reference herein in its entirety.
FIGS. 10 and 12 are block diagrams of further example systems 1100 and 1200 according to the invention. Example system 100 used a single direct detection receiver system 130 with an optical data signal 142 having a single optical carrier 146, in order to illustrate the basic principles of the invention. Systems 1100 and 1200 use multiple direct detection receivers, each directed to a different optical carrier within the optical data signal. For clarity, in FIGS. 10 and 12, the term "receiver" is used to describe receivers based on direct detection of SSB transmitted signals, such as receiver system 130 in FIG. 1 and its variants described in FIGS. 2-9.
In FIG. 10, system 1100 includes a transmitter subsystem 1102 coupled to a receiver subsystem 1104 via an optical fiber 120. Briefly stated, the transmitter subsystem 1102 encodes information to be transmitted onto an optical signal. For reasons which will become apparent below, this optical signal is referred to as a "composite optical signal." The composite optical signal is transmitted across the fiber 120 and received by the receiver subsystem 1104. The receiver subsystem 1104 recovers the original information from the composite optical signal.
In more detail, the transmitter subsystem includes SSB transmitters 11 lOA-11 ION which are optically coupled to an optical combiner 1112. SSB transmitter 110 of FIG. 1 and its variants are suitable for use as a SSB transmitter 1110. Each SSB transmitter 1110 produces an optical data signal which includes an optical carrier and a primary optical sideband, as discussed previously in the context of FIG. 1, et seq. However, each transmitter 1110 uses a different wavelength λι.-λN for its optical carrier frequency so as to spectrally separate the individual optical data signals. Combiner 1112 optically combines these optical data signals to produce the composite optical signal. Examples of combiners 1112 include 1 :N power combiners (i.e., not wavelength selective) and WDM multiplexers.
FIG. 11 shows the spectra of an example composite optical signal. Transmitter
1110A produces signal 1420A with optical carrier 1410A at wavelength λi. and primary optical sideband 1412A. Similarly, transmitters 1 HOB- 11 ION produce signals 1420B- 1420N. For clarity, each of the sidebands 1412A-1412N will be referred to as subbands of the composite optical signal.
On the receive side, the receiver subsystem 1104 includes a wavelength-selective optical splitter 1132 coupled to direct detection receivers 1130A-1130N. Receiver 130 of FIG. 1 and its variants are suitable for use as a receiver 1130. The optical splitter 1132 splits the composite optical signal into N optical signals, from which the encoded information is recovered. In a preferred embodiment, there is a one-to-one correspondence between transmitters 1110 and receivers 1130. In other words, transmitter 1110A produces signal 1420A and optical splitter 1132 sends signal 1420A to receiver 1130A, which recovers the subband produced by transmitter 1110A. A similar relationship exists for the other transmitters 1110 and receivers 1130. In a preferred embodiment, the optical splitter 1132 is a WDM demultiplexer. Alternately, it may be a conventional power splitter coupled to filters.
In FIG. 12, system 1200 includes a transmitter subsystem 1202 coupled to a receiver subsystem 1204 via an optical fiber 120. In system 1100 of FIG. 10, each transmitter 1110 received an electrical data signal and generated a corresponding optical data signal with a primary optical sideband. The optical data signals were optically combined to produce the composite optical signal. In system 1200, electrical data signals and tones are electrically
combined to produce a composite electrical signal, which is then converted to optical fonn to produce the composite optical signal. Approaches which use a mix of electrical and optical combining will be apparent. FIGS. 13A-13B illustrate the frequency spectra of various signals in system 1200. For clarity, only the relevant portions of these spectra are depicted in the figures.
In more detail, transmitter subsystem 1202 includes two electrical transmitters 1208A-1208B which are electrically coupled to an FDM multiplexer 1209, which in turn is coupled to transmitter 1210. Each electrical transmitter 1208 includes the same construction as element 245 in FIG. 6B of co-pending U.S. Patent Application Serial No. 09/405,367, "Optical Communications Networks Utilizing Frequency Division
Multiplexing," by Michael W. Rowan, et al., filed Sept. 24, 1999 (hereafter, the "FDM Application"), which subject matter is incorporated by reference herein in its entirety. In brief, electrical transmitter 1208 includes a QAM modulator (included in element 640 of FIG. 6B) coupled to an FDM multiplexer (elements 642 and 644 in FIG. 6B). Each electrical transmitter 1208 receives a number of incoming electrical low-speed channels 1222. The QAM modulator applies a QAM modulation to each incoming low-speed channel. The FDM multiplexer combines the QAM-modulated low-speed channels using FDM techniques to form an electrical channel 1224A-1224B. The frequency spectra of signals 1224A and 1224B are shown in FIG. 13A. See also FIG. 10D, et seq. in the FDM Application for a specific example.
The FDM multiplexer 1209 combines the two channels 1224 into a single electrical signal, which for convenience will be referred to as the composite electrical signal 1226. It does this using conventional FDM techniques, frequency shifting some or all of the channels 1224 to higher carrier frequencies. For example, referring again to FIG. 13 A, channel 1224A is not frequency shifted, as shown by spectra 1234A, but channel 1224B is frequency shifted to a higher frequency range, as shown by spectra 1234B. In addition, tones 1237 are added after this frequency shifting. In other embodiments, the tones 1237 may be added at different times during the signal processing and/or different channels may share a common tone. In addition, the electrical transmitters 1208 may include frequency shifters to move the spectral location of channels 1224, for example if they would otherwise overlap with the tones 1237. h the embodiment shown in FIG. 12, the FDM multiplexer
1209 also includes optional spectral filters 1235, which filter out unwanted frequency components.
Transmitter 1210 is an E/O converter, which in this embodiment includes a laser 1212 and a SSB optical modulator 1214. The laser 1212 generates an optical reference at a frequency fo and the modulator 1214 modulates the optical reference with the electrical high-speed channel 1226. The result is a composite optical signal 1242, as shown in FIG. 13B. Note that the composite optical signal 1242 includes two tones 1237 and two subbands 1224. Tone 1237A and subband 1224A form one optical carrier-primary optical sideband pair. Tone 1237B and subband 1224B form a second optical carrier-primary optical sideband pair. By using two (or more) electrical transmitters 1208, system 1200 operates at double (or more) the data rate compared to a system with only a single electrical transmitter 1208.
On the receive side, the receiver subsystem 1204 includes an optical splitter 1232 which is optically coupled to two direct detection receivers 1230A-1230B, each of which is coupled to an electrical receiver 1238. The wavelength-selective splitter 1232 and direct detection receivers 1230 operate as in FIG. 10. Receiver 1230A recovers subband 1224A as signal 1254A, and receiver 1230B recovers subband 1224B as signal 1254B. The electrical receivers 1238 reverse the functionality of electrical transmitters 1208, separating the incoming electrical signal 1254 into its constituent low-speed channels 1256. Accordingly, each receiver 1238 includes the same construction as element 240 in FIG. 6 A of the FDM Application. An FDM demultiplexer (elements 624 and 622 in FIG. 6A) frequency demultiplexes the electrical signal 1254 into separate electrical channels, each of which is then QAM demodulated by a QAM demodulator (included in element 620 in FIG. 6A).
System 1200, like the other systems described, is an example. The invention is not limited to the specific numbers of transmitters and/or receivers, frequency ranges, data rates, etc. Other variations will be apparent. For example, a transmitter subsystem 1202 operating at a first wavelength X\ could be used as the transmitter system 1100A in system 1100, a second transmitter subsystem 1202 operating at wavelength λ2 as transmitter system 1100B, and so on, with corresponding changes on the receive side. In this way, systems 1100 and 1200 can be combined to yield an even higher data rate system.
Although the invention has been described in considerable detail with reference to certain preferred embodiments thereof, other embodiments are possible. Therefore, the scope of the appended claims should not be limited to the description of the preferred embodiments contained herein.