Spread Spectrum Code-Division-Multiple-Access (SS-CDMA) Lightwave Communication System

The present invention relates to a spread spectrum code-division-multiple-access (SS-CDMA) lightwave communication system. More particularly, in the present system, spread spectrum techniques are used to implement a CDMA scheme on, for example, a single-mode optical waveguide arrangement wherein the transmission bandwidths of the various users do not necessarily coincide but are randomly scattered over the enormous channel bandwidth of the waveguide arrangement.

The use of optical fibers in communications systems is increasing from year to year. The multiple access, link length variation, etc. nature of such optical fiber systems places certain requirements on signaling and detection schemes to insure that a satisfactory predetermined error rate is achieved under varied conditions at the receivers. To achieve satisfactory separation in multiple concurrent transmissions on a fiber optic system, various modulation or multiplexing techniques have been employed which are at times similar to techniques found in microwave systems.

Spread spectrum modulation techniques have been used in terrestrial and satellite microwave communication systems to achieve security and satisfactory interference rejection at the receivers. In this regard see, for example, U.S. Pat. Nos. 4,112,372 issued to J. D. Holmes et al. on Sept. 5, 1978; 4,164,628 issued to C. R. Ward et al. on Aug. 14, 1979; and 4,470,138 issued to F. S. Gutleber on Sept. 4, 1984. More recently, spread spectrum modulation techniques have been suggested for use in optical communication systems. In this regard see, for example, the articles "Spread-Spectrum Signaling For Distributed Fiber Optical Systems" by C. A. Laber et al. in IEEE Transactions of Aerospace and Electronic Systems, Vol. AES-16, No. 4, July 1980, at pages 440-445; and "Spread-Spectrum Multiple Access Data Loop" by S. Hasegawa et al. in NEC Research & Development, No. 71, October 1983, at pages 48-57.

Optical single-mode fibers have a bandwidth that is measured in tens of THz (1 THz=1000 GHz). However, even the fastest optical fiber system of today leaves this large bandwidth mostly untapped, as the maximum bit rate in such a fiber system is measured in Gbit/sec. Among the various obstacles in the way of achieving efficient bandwidth utilization through Wavelength-Division-Multiplexing WDM), an important one is the lack of stability of semiconductor laser. A typical laser frequency will vary as a result of temperature variations, aging, and other causes in the order of several hundred Gigahertz. Therefore, in a Wavelength-Division-Multiplexing (WDM), also known as a Frequency-Division-Multiplexing (FDM), system, one must allow for a separation of several hundred GHz between adjacent channels to avoid possible interference due to one carrier drifting into the band used by an adjacent signal. This results in highly inefficient usage of the available fiber bandwidth, particularly in the case where each user only requires a small bandwidth. As a result, WDM is highly inefficient in the situations typical of local-area networks, where there may be hundreds or even thousands of users, each user requiring a relatively low bit rate.

The problem remaining in the prior art is to provide an optical communication system which uses the bandwidth of an optical waveguide arrangement, such as a single-mode optical fiber, in a substantially more efficient manner.

The foregoing problem in the prior art is solved in accordance with the present invention which relates to a technique for transmitting a spread spectrum code-division-multiple-access, (SS-CDMA) in a lightwave communication system and a transmitter and receiver designed to operate in such a system. More particularly, a lightwave communication technique and system, and an associated transmitter and receiver design, is disclosed wherein spread spectrum and CDMA techniques are used in addition to random Wavelength-Division-Multiplexing (WDM) in, for example, a single-mode optical waveguide arrangement, such that the transmission bandwidths of the various users, and their associated transceivers, do not necessarily coincide but are randomly scattered over the enormous channel bandwidth of the lightwave tranmission medium of the system to reduce interference and efficiently use the lightguide spectrum.

It is an aspect of the present invention to provide a WDM, SS-CDMA, lightwave communication system which permits receivers therein to track the drift of the carrier frequency of a signal it is receiving, and maintain good reception, even if the carrier frequency drifts into the bandwidth used by another transmitter/receiver pair. It is a further aspect of the present invention to provide a spread-spectum lightwave communication system wherein the spectrum of the underlying electromagnetic field is spread and de-spread, as opposed to the spreading and de-spreading of the spectrum of just the light-intensity signal.

Other and further aspects of the present invention will become apparent during the course of the following description and by reference to the accompanying drawings.

Referring now to the drawings:

FIG. 1 is a block diagram of a WDM, SS-CDMA, lightwave communication system arrangement including an exemplary transceiver arrangement in accordance with the present invention;

FIG. 2 illustrates a plurality of exemplary WDM, SS-CDMA, signal spectra disposed relative to the overall spectrum of a received signal before de-spreading of the desired signal at a receiver;

FIG. 3 illustrates the exemplary spectrum of a received signal after de-spreading of the desired signal at a receiver; and

FIG. 4 is an alternative arrangement of the receiver of FIG. 1.

In accordance with the present invention, a lightwave communication system uses spread spectrum code-division-multiple-access (SS-CDMA) techniques along with wavelength-division-multiplexing (WDM), otherwise known as Frequency-Division-Multiplexing (FDM), to achieve a lightwave communication system which efficiently uses the available optical spectrum of the associated lightguide while providing adequate rejection of interference caused by one signal drifting into the bandwidth being used by another transmitter/receiver pair. The present invention can be used to implement, for example, a local area network capable of supporting thousands of simultaneous two-way communication channels at a relatively low data rate, e.g., 10 Mbit/sec each, with no blocking, no hierarchy, and no centralized control. Also, since the communication channels are supported in a time-parallel manner, extraordinary throughput can be achieved without requiring unreasonable hardware speed. The present network architecture can, therefore, also be used to realize a switching system capable of establishing thousands, or even tens of thousands, of high-data-rate (e.g., 10 Mbit/sec) interconnections.

FIG. 1 is a block diagram of a preferred embodiment of the present optical communication system including an examplary transceiver arrangement in accordance with the present invention, showing all possible elements even though certain elements can be combined or are optional. In FIG. 1, a plurality of N optical transceivers 10.sub.1 to 10.sub.N are coupled to an optical waveguide arrangement 11, to permit communications between pairs of the transceivers 10.sub.i, or between a transceiver and an external network via an optional network interface unti 12 where such inter-network communication is desired. Optical wavegide arrangement 11 can be any optical device that combines the light signals from all the transmitters 13 of transceivers 10.sub.1 to 10.sub.N, and network interface 12 if present, and distributes the resulting superposition of signals among all receivers 14 of the N transceivers, and network interface 12, if present. In a preferred embodiment of the system, the optical waveguide arrangement 11 might consist of a large star coupler, with an input fiber from each transmitter 13, and network interface 12 if present, and an output fiber to each receiver 14, and network interface 12 if present. Alternative implementations might be as an optical bus or loop or any arrangement of optical devices that achieves the same function.

Each transceiver 10.sub.i comprises a transmitter 13, for transmitting a signal from an associated system user over optical waveguide arrangement 11, and a receiver 14 which receives all WDM, SS-CDMA, signals propagating on optical waveguide arrangement 11 and de-spreads the desired WDM, SS-CDMA signal for delivery to the associated system user.

An exemplary transmitter 13 in each of transceivers 10.sub.i is shown in FIG. 1 as comprising a laser source 20 which has its signal modulated by a light modulating means 21 internally or externally to source 20. Laser source 20 can be implemented by, for example, a semiconductor laser or any other suitable light source which is a coherent system that acts like an oscillator for generating a random carrier frequency. The lightwave signal of laser source 20 is modulated by modulating means 21 using any one of a variety of methods well-known in the art, including direct modulation of the laser current or external amplitude or phase modulation of the light. The modulating signal is either a digital or an analog signal from the associated system user. This results in an modulated output signal destined for another system user or to an external network, when network interface 12 is present in the system. The modulated output signal from the directly or externally modulated laser source 20 is, therefore, an up-converted signal in a frequency band determined by laser source 20 which is shown in FIG. 1 as being delivered to a spectrum spreading device 22 which for purposes hereinafter will be considered as a Pseudo-Noise (PN) scrambling means 22.

In PN scrambling means 22, a predetermined high-frequency pseudo-random noise code, generated by a PN sequence generator 23 under the control of a controller 24, further modulates the modulated information wave from the laser source 20/light modulator 21 combination to produce an output signal that is spread over a greater bandwidth than the information wave from modulated laser source 20, using any well known spreading technique. The output signal from PN scrambling means 22 is coupled into optical waveguide arrangement 11 via any well-known optical coupling device. It is to be understood that the sequence of laser source 20, light modulating means 21 and PN scrambling means 22, as shown in FIG. 1, could be modified such that the light signal from the laser source could have its spectrum spread before modulating the resulting spread spectrum carrier signal with the user signal. It is to be further understood that PN scrambling means 22 and PN sequence generator 23 are but one spectrum spreading technique that could be used in practicing the present invention. Alternatively, PN scrambling means 22 and PN sequence generator 23 could be replaced by a Frequency-Hopping (FH) code Generator 26 which is used to modulate the light beam of laser source 20 and produce an output light beam wherein the output carrier frequency is hopped in a predetermined sequence. The Frequency-Hopping technique, as a spectrum spreading function, is well known in the art. Regardless of the spectrum spreading techniques used, a (possibly attenuated) replica of the SS-CDMA output signal from a transmitter 13 appears at each output 30 of optical waveguide arrangement 11 along with the SS-CDMA signals from all the other transmitters. An exemplary frequency spectrum of multiple WDM, SS-CDMA, signals, present at each output 30 of optical waveguide arrangement 11 at any point in time, is shown in FIG. 2.

The signals propagating through optical waveguide arrangement 11 are received by each of receivers 14 of transceivers 10.sub.1 to 10.sub.N. In a receiver 14, the signal from optical waveguide arrangement 11 is received in a PN descrambling means 31 where the received WDM, SS-CDMA, signal, as shown in FIG. 2, is de-spread using a PN code sequence which is synchronized and matched to the PN code sequence used by a predetermined transmitter 13 of another one of transceivers 10.sub.1 to 10.sub.N forming the two-way communication. The PN code sequence used by PN descrambling means 31 for de-spreading the receiving signal is obtained from PN sequence generator 32 which is under the control of controller 33. The de-spread signal at the output of PN descrambling means 31 can comprise the exemplary configuration shown in FIG. 3, where the desired signal appears as a narrowband signal 39 and the other received transmissions form a smooth low-level background across the spectrum and appear as noise. It must be understood that the desired signal will only appear as a narrowband signal 39 if the PN code sequences used by the associated transmitter and receiver are the same and synchronized in their application to the spreading and de-spreading operation.

The output signal from PN descrambling means 31 is delivered to a narrowband receiver 34 which maintains itself locked to the narrowband signal of FIG. 3 by any suitable technique as, for example, Automatic Frequency Control (AFC), and demodulates the acquired signal to a corresponding analog or digital signal for transmission to the destined user via lead 35. It is to be understood that narrowband receiver 34 can be realized in a variety of ways, such as through the heterodyne or homodyne, detection or through external optical filters.

Where it is desired that the present system also have the capability of communicating with an external network, an optional network interface 12 is connected to optical waveguide arrangement 11 via appropriate coupling means at points 25 and 30 for the coupling of a WDM, SS-CDMA, signal onto optical waveguide arrangement 11 and the reception of WDM, SS-CDMA, signals from the waveguide arrangement, respectively. When present, network interface 12 functions to convert signals between the external network format to that for transmission on optical waveguide arrangement 11 in any suitable manner. For purposes of discussion hereinafter, network interface 12 will not be included in the system description, but it is to be understood that if such interface were present, it would act in a manner similar to that described for any one of transceivers 10.sub.1 to 10.sub.N for providing an appropriate 2-way communication link with the external network.

For the steady state operation, where various transmitters and receivers are already communicating with each other and there is no interest by any transmitter to acquire a 2-way communications link, an exemplary spectrum as shown in FIG. 2 would exist, less the dashed narrowband acquisition signals. Under such steady-state conditions, for each one-way communication link the PN sequence generator 23 in the transmitter at one end of the link and the PN sequence generator 32 in the receiver at the other end of the link must maintain mutual synchronization in order to obtain the narrowband signal 39 in the spectrum shown for the desired signal in FIG. 3. To achieve this, either the PN sequence generator 23 or the PN sequence generator 32 in each transceiver 10 must be "turnable" (i.e., programmable) to permit matching of its PN code sequence to that generated by the other PN sequence generator at the other end of the communciation link. Furthermore, either the laser source 20 of each transceiver 10 or the narrowband receiver 34 of each transceiver 10 must tunable over the full frequency range of interest, in order to allow full network interconnectivity, i.e., any receiver can receive the signal from any transmitter.

A preferred arrangement would be to have each narrowband receiver 34 at each of the transceivers selectively tunable to any frequency within a certain predetermined frequency range, while the laser source 20 of each transceiver 10.sub.i is not tunable and generates an arbitrary or undetermined frequency which is not constrained in any way. Indeed, the oscillation frequency of laser source 20 might even drift erratically and fall on top of the frequency generated by the laser source 20 of another transceiver without any adverse effects on the communication links. The range of tunability of each narrowband receiver 34 must be large enough to include all the frequencies of all laser sources 20 present in the system.

It is understood that in the foregoing discussion the term "frequency" refers to the underlying optical carrier and is of the order of 230,000 GHz for a wavelength of 1.3 .mu.m. The usage of frequency as a paramater instead of wavelength is motivated by the particular type of signal processing that makes frequency, and not wavelength, the important parameter. Additionally, in the above-mentioned preferred arrangement, it is also preferable that the PN Sequence Generator 23 at the associated transmitter 13 be programmable to match a PN code sequence generated by the PN Sequence Generator 32 of a desired receiver 14 in another transceiver 10, but that the PN Sequence Generator 32 at the receiver 14 would have to then be capable of synchronization to the PN Sequence Generator 23 at transmitter 13 of the link. It is to be understood that the present system is not limited to such preferred arrangement but that other permutations of the tunability of laser sources 20, narrowband receivers 34, and PN Sequence Generators 23 and 32 could also be used to achieve proper system operation. Controllers 24 and 33 can be used to achieve the proper tunability and synchronization in the associated PN Sequence Generators 23 and 32 and narrowband receiver 34 as will be explained hereinafter.

In a start-up condition, which can occur at any time, a first and a second transceiver are attempting to gain access to each other for establishing a 2-way link between the associated users of the system while others are already communicating via optical waveguide arrangement 11 as described above for the steady state condition. To begin acquisition, one technique would be to hae user "A", associated with, for example transceiver 10.sub.1 notify its controller 24, via a signaling message, that user "A" wishes to establish a link with user "B" associated with, for example, transceiver 10.sub.N. Controller 24 can comprise, for example, a simple microcomputer or other device which begins by disabling the PN Sequence Generator 23. As a result, the acquisition signal that user "A" transmits over optical waveguide arrangement 11 is not spectrally spread but, rather, it looks like the narrowband acquisition signals 40 and 41 shown in FIG. 2. This signal does not interfere with already established communication links which are using spread spectrum and, thereby, are immune to narrowband interference.

Laser source 20 at transceiver 10.sub.1 is modulated by an introductory message from user "A", which introductory message can comprise certain preamble information including the address of the calling transceiver (10.sub.1) and the address of the called transceiver (10.sub.N). The modulated narrowband introductory signal is sent over optical waveguide arrangement 11 as a narrowband acquisition information signal centered on the carrier frequency generated by laser source 20 of transceiver 10.sub.1. This narrowband acquisition information signal from transmitter 13 of transceiver 10.sub.1 appears as, for example, the narrowband signal 40 in FIG. 2 when looking at the overall frequency spectrum of optical waveguide arrangement 11. Such narrowband acquisition signal is only transmitted by the calling transceiver during acquisition, which, generally, takes a small amount of time when compared to the time the communication link is maintained after acquisition has been completed.

Each of the receivers 14 of transceivers 10.sub.1 to 10.sub.N receives the narrowband introductory message 40 from transceiver 10.sub.1 along with all other 2-way communications propagating on optical waveguide arrangement 11 from the already established communication links. If the called transceiver 10.sub.N is not already busy on another communication link, then under the control of its controller 33 of receiver 14, its narrowband receiver 34 is caused to continuously scan the frequency spectrum in order to detect a narrowband acquisition information signal similar to signal 40 of FIG. 2 somewhere in the spectrum of optical waveguide arrangement 11. When an exemplary signal 40 is received, controller 33 detects, via its feedback path 36, that such narrowband signal exists and causes narrowband receiver 34 to lock onto the carrier frequency of the narrowband signal to permit decoding the message contained therein. If the address contained in the decoded acquisition information message corresponds to another transceiver, then this transceiver will unlock its narrowband receiver 34 and resume scanning the spectrum, repeating the above process until it obtains an acquisition information message indicating that this is the transceiver being called.

After locking onto an acquisition information signal and finding that the decoded acquisition information message indicates that the transceiver is being called, controller 33 of the called transceiver 10.sub.N notifies the associated controller 24 in transmitter 13 that the transceiver is being called by transceiver 10.sub.1. Such notification causes controller 24 to disable PN Sequence Generator 23 while sending a proper acquisition answer message to the light modulating means 21, which includes the addresses of the called and calling transceiver. Such action causes transmitter 13 of transceiver 10.sub.N to transmit a narrowband acquisition answer signal, using the arbitrary or undetermined carrier frequency generated by its laser source 20, onto optical waveguide arrangement 11. Such narrowband acquisition answer signal can take the exemplary form of signal 41 in FIG. 2, which is, in general, at a frequency different from that of signal 40.

While transceiver 10.sub.1 is transmitting its narrowband acquisition information signal indicating it wishes to establish a 2-way call with transceiver 10.sub.N, controller 33 of transceiver 10.sub.1 is causing its narrowband receiver 34 to scan the spectrum looking for a narrowband signal such as signal 41 in FIG. 2. When it detects a narrowband signal, it causes narrowband receiver 34 to lock onto the carrier of such narrowband signal and decodes the message therein to see if it is an answer signal from transceiver 10.sub.N. If the detected narrowband signal is other than an answer signal from transceiver 10.sub.N, then controller 33 of transceiver 10.sub.1 unlocks its narrowband receiver 34 to continue the scanning and detection process.

Once transceiver 10.sub.1 detects that the received narrowband signal is an answer signal from transceiver 10.sub.N, a 2-way link is established since transceiver 10.sub.1 is locked onto the carrier frequency of the laser source 20 of transceiver 10.sub.N, and vice versa, due to the two acquisition information signals. The next and final step consists of switching to spread-spectrum communication on the newly achieved two-way link by activating PN scramblers 22 and descramblers 31 at both ends of the link. The final step begins with the controllers at the two ends of the link exchanging information about the PN sequence to be used for scrambling. For example, each transmitter could send to the receiver 14 at the other end the exact PN bit sequence to be used for descrambling. Alternatively, a short identifier or seed for generating the sequence could be sent. Such technique would allow the use of very long PN sequences without requiring the controllers to maintain a large directory of PN sequences in use in the system.

Next, the two transceivers must switch on their PN scramblers 22 and descramblers 31, and achieve pairwise synchronization of the associated PN code sequences. As soon as PN scramblers 22 are activated, the two narrowband acquisition signals 40 and 41, shown dashed in FIG. 2, disappear from optical waveguide arrangement 11 and are replaced by their spread-spectrum counterparts which look like the other signals already present on the waveguide arrangement 11, shown solid in FIG. 2. When proper synchronization of the PN sequence is achieved, the condition will be characterized by the appearance of narrowband signal 39 in the spectrum after PN descrambler 31, as shown in FIG. 3. The two transceivers can coordinate the transition to PN scrambling and descrambling through the already-established two-way link, so that the PN sequences will be approximately synchronized when the transition takes place. Final adjustment of the synchronization can then be achieved quickly through a simple search procedure. For example, controller 33 directs PN sequence generator 32 to slowly advance or retard the position of the generated PN code until narrow signal 39 appears in the spectrum, indicating that proper synchronization has been achieved. The two-way link is now operating in the steady-state WDM, SS-CDMA transmission format.

If, however, transceiver 10.sub.10 is presently busy on a call, it will not be sweeping or scanning the frequency spectrum looking for an acquisition information signal, but will be locked onto an existing WDM, SS-CDMA signal forming a current 2-way communication. As a result no answer will be sent by transceiver 10.sub.N back to transceiver 10.sub.1 within a reasonable period of time, and user 1 is thereby notified that his call cannot presently be completed.

It is to be understood that the above described technique for acquiring a 2-way link is merely provided for exemplary purposes and not for purposes of limitation since other techniques can also be used. For example, if transceiver 10.sub.1 wishes to establish a 2-way link with transceiver 10.sub.N transceiver 10.sub.1 can generate an acquisition information signal which is encoded with the (perhaps abbreviated) PN code sequence used by transceiver 10.sub.N, rather than being a narrowband uncoded signal. The PN code sequence of transceiver 10.sub.N can be easily determined from a look-up table in an associated memory of controller 24 of transceiver 10.sub.1. Meanwhile, each receiver 14 of all non-busy transceivers is sweeping or scanning the spectrum while changing the phase of its PN code sequence for each sweep of the spectrum by, for example, skipping one symbol. Within a short time, the PN Sequence Generator 32 at receiver 14 of transceiver 10.sub.N will detect a narrowband signal 39 as shown in FIG. 3, because the PN Sequence Generator 32 at transceiver 10.sub.N will become synchronized with the PN Sequence Generator 23 in transmitter 13 of transceiver 10.sub.1 when the phase of the received PN code sequence matches the generated PN code sequence at PN descrambling means 31 of transceiver 10.sub.N. At this point transceiver 10.sub.N is locked onto the frequency and is synchronized with the PN code sequence transmitted by transceiver 10.sub.1. At such occurrence, transceiver 10.sub.N transmits an acquisition answer signal, which is also WDM, SS-CDMA, encoded, back to transceiver 10.sub.1 where a similar process is repeated in transceiver 10.sub.1 until both transceivers are frequency and PN code sequence locked. Then 2-way steady state communications can proceed.

Availability of timing could be an expeditious feature to provide. Otherwise, acquisition is twodimensional (frequency and timing). Special signals marking the edges of the timing band could also be used to refresh bit and chip timing at the start of each frequency sweep. Thus the timing acquisition problem is mitigated so long as the hardware is kept stable enough to maintain timing during each sweep.

It is to be understood that the above-described embodiments are simply illustrative of the principles of the invention. Various other modifications and changes may be made by those skilled in the art which will embody the principles of the invention and fall within the spirit and scope thereof. For example, as shown in FIG. 4, the receiver 14 can comprise the structure wherein the signal from output 30 of waveguide arrangement 11 is received in the narrowband receiver 31 where the received lightwave signal is converted into an electrical signal which is then de-scrambled in PN de-scrambler 34 using the PN sequence generated by PN sequence generator 32. Additionally, in transmitter 13, light modulator 21 and PN scrambling means can be interchanged or any one or both parts may be incorporated as part of laser source 20.