United States Patent: 5,808,767

[US Patent & Trademark Office, Patent Full Text and Image Database]

( 1 of 1 )

United States Patent	*5,808,767*
Williams , et al.	September 15, 1998

Fiber optic network with wavelength-division-multiplexed transmission to customer premises

Abstract

A fiber optic network comprises an optical fiber connection (one fiber or two) from a central office to an intelligent interface device in the subscriber's premises. The central office includes a serving node transceiver providing communication links to/from at least a narrowband switch and a broadband switch for providing narrowband and broadband service routing. The network includes at least one passive power splitter/combiner for passing all wavelengths on the optical fiber connection between the serving node transceiver and the intelligent interface devices. All wavelengths are provided to each customer and bandwidth on the optical fiber loop is dynamically allocated for individual services on demand through two-way wavelength division multiplexing and demultiplexing as well as any necessary signal format conversions. The network has media access control functionality and utilizes a dynamic media access control procedure for allocation of the bandwidth.

Inventors:	Williams; Larry (Olney, MD); Little; Dave (Columbia, MD); Lucas; Amos (Baltimore, MD); Burton; William (Ashton, MD)
Assignee:	Bell Atlantic Network Services, Inc (Arlington, VA)
Appl. No.:	656880
Filed:	May 30, 1996

Current U.S. Class: 398/70; 398/71; 398/72

Intern'l Class: H04B 010/00

Field of Search: 359/115,118,124,125,164,165,167,152

References Cited [Referenced By]

U.S. Patent Documents

4135302	Jan., 1979	Kronman et al.	32/57.
4506387	Mar., 1985	Walter	455/612.
4709418	Nov., 1987	Fox et al.	455/612.
4763317	Aug., 1988	Lehman et al.	370/58.
4768188	Aug., 1988	Barnhart et al.	370/80.
4891694	Jan., 1990	Way	358/86.
4994909	Feb., 1991	Graves et al.	358/86.
5121244	Jun., 1992	Takasaki	359/161.
5133079	Jul., 1992	Ballantyne et al.	455/4.
5136411	Aug., 1992	Paik et al.	359/125.
5175639	Dec., 1992	Takasaki	359/118.
5181106	Jan., 1993	Sutherland	358/86.
5189673	Feb., 1993	Burton et al.	370/110.
5221983	Jun., 1993	Wagner	359/125.
5262883	Nov., 1993	Pidgeon	359/125.
5272556	Dec., 1993	Faulkner et al.	359/125.
5293376	Mar., 1994	White	370/54.
5303229	Apr., 1994	Withers et al.	370/58.
5325223	Jun., 1994	Bears	359/137.
5436748	Jul., 1995	Uinel et al.	359/125.
5469283	Nov., 1995	Uinel et al.	359/118.
5526350	Jun., 1996	Gittins et al.	370/58.

Primary Examiner: Bacares; Rafael
Attorney, Agent or Firm: Lowe, Price, LeBlanc, & Becker

Claims

We claim:

1. A fiber optic communications network comprising:

a serving node telecommunications transceiver transmitting a first plurality of optical communications and information signals;

a plurality of customer premises equipment systems, each one of said customer premises equipment systems comprising:

a customer optical transceiver for receiving said first plurality of optical communications and information signals transmitted by said serving node telecommunications transceiver, for transmitting a second plurality of optical communications and information signals to said serving node telecommunications transceiver, and for performing wavelength division multiplexing and demultiplexing of said first and second plurality of optical communications and information signals, and

a terminal device providing communication services requiring different bandwidth allocations or a plurality of terminal devices requiring different respective bandwidth allocations;

a fiber optic transmission path connecting said serving node telecommunications transceiver to said customer optical transceiver in each one of said plurality of customer premises equipment systems, said serving node telecommunications transceiver controllably transmitting said first plurality of optical communications and information signals to each optical transceiver at a plurality of wavelengths over said fiber optic transmission path and controllably receiving said second plurality of optical communication and information signals from each customer optical transceiver over said fiber optic transmission path;

a plurality of optical elements, said optical elements splitting the first plurality of optical communication and information signals transmitted by said serving node telecommunications transceiver into a plurality of split optical signals, said split optical signals containing all of the wavelengths of said first plurality of optical communication and information signals transmitted by said serving node telecommunications transceiver, and forwarding each one of said split optical signals to a respective one of said customer premises, and said plurality of optical elements combining said second plurality of optical communication and information signals received from said customer premises into a combined optical signal and forwarding said combined optical signal to said serving node telecommunications transceiver; and

a media access controller connected to and controlling said customer optical transceiver, so as to dynamically provide said different bandwidth allocations over said fiber optic transmission path connecting said serving node telecommunications transceiver to said plurality of customer premises by assigning one or more of said plurality of wavelengths to said customer optical transceiver.

2. A fiber optic communications network according to claim 1, wherein said plurality of optical elements comprises a plurality of passive power optical elements.

3. A fiber optics communications network as recited in claim 1, wherein said fiber optic transmission path comprises at least one full duplex optical fiber.

4. A fiber optic communications network according to claim 3, wherein said plurality of optical elements comprises a plurality of passive power integrated splitter/combiners.

5. A fiber optics communications network as recited in claim 1, wherein said fiber optic transmission path comprises simplex optical fibers.

6. A fiber optic communications network according to claim 5, wherein said plurality of optical elements comprises a plurality of passive power splitters and a plurality of passive power combiners.

7. A fiber optics communications network as recited in claim 1, wherein said serving node telecommunications transceiver comprises a narrowband switch providing voice grade telephone service routing and a broadband switch providing routing for broadband services.

8. A fiber optics communications network as recited in claim 7, wherein said serving node telecommunications transceiver further comprises an asynchronous transfer mode switch.

9. A fiber optics communications network as recited in claim 8, wherein said serving node telecommunications transceiver transmits optical service signals, containing network identifiers different from said first or second plurality of optical communications and information signals, over said fiber optic transmission path; and

wherein each of said customer optical transceivers receive said optical service signals, converts said optical service signals into electrical signals and selectively delivers the electrical signals to electrical devices in the customer premises in accordance with said network identifiers.

10. A fiber optics communications network as recited in claim 7, wherein:

narrowband communications for said customer premises is permanently allocated to one of said plurality of wavelengths; and

said media access controller dynamically provides said different bandwidth allocations for said broadband services.

11. A fiber optics communications network as recited in claim 7, wherein said serving node telecommunications transceiver further comprises an optical switch providing optical signals from an optical network.

12. A fiber optics communications network as recited in claim 1, wherein said first plurality of optical communications and information signals from said telecommunications central office are transmitted simultaneously on said plurality of wavelengths and each one of said customer optical transceivers contains an optical receiver receiving all of said plurality of wavelengths and controllably discriminating said plurality of wavelengths.

13. A fiber optics communications network as recited in claim 12, wherein said customer optical transceiver comprises at least one service definition module.

14. A fiber optics communication network as recited in claim 13, wherein said at least one service definition module is responsive to signalling communications received over said optical fiber for controlling said wavelength dependent multiplexer and demultiplexer in accordance with said signalling communications.

15. A fiber optics communications network as recited in claim 1, wherein said first plurality of optical communications and information signals from said serving node telecommunications transceiver are transmitted simultaneously on at least three different wavelengths and each one of said customer optical transceivers performs format conversions of said first plurality of optical communications and information signals from said serving node telecommunications transceiver.

16. A fiber optics communications network as recited in claim 1, wherein the media access controller performs a dynamic media access control procedure.

17. A fiber optics communications network as recited in claim 16, wherein said dynamic media access control procedure supplies individual services and dynamically provides said different bandwidth allocations upon demand by a customer.

18. A fiber optics communications network as recited in claim 8, wherein said different bandwidth allocations include the dynamic allocation of said plurality of wavelengths.

19. A fiber optics communications network as recited in claim 18, wherein said dynamic allocation of said plurality of wavelengths is performed in response to a demand by said customer.

20. A fiber optics communications network as recited in claim 18, wherein said dynamic media access control procedure includes the dynamic allocation of bit rates of said first or said second plurality of optical signals in response to a demand by said customer.

21. A fiber optics communications network as recited in claim 1, wherein at least one of said customer optical transceivers comprises a SONET multiplexer.

22. A fiber optics communications network as recited in claim 16, wherein said media access controller is addressable by said telecommunications central office.

23. A fiber optics communications network as recited in claim 16, wherein said media access controller performs said dynamic media access control procedure by causing said customer optical transceiver to transmit a bandwidth request signal at a common channel signalling wavelength over said fiber optic transmission path, said common channel signalling wavelength being separate from said plurality of wavelengths carrying said first or second plurality of optical communication and information signals.

24. A fiber optics communications network as recited in claim 23, wherein said media access controller communicates with said serving node telecommunications transceiver to determine which one of said plurality of wavelengths carrying said first or second optical communications and information signals are idle and allocating said idle wavelengths.

25. A fiber optics communications network as recited in claim 23, wherein said bandwidth request signal is transmitted in response to user inputs.

26. A fiber optics communications network as recited in claim 25, wherein said user inputs are received from a remote control.

27. In a fiber optic communications network including a serving node telecommunications transceiver, a plurality of customer premises equipment systems, each system including a customer optical transceiver, a media access controller, and at least one terminal device requiring different respective bandwidth allocations, and a fiber optic transmission path including a plurality of optical elements, for connecting the serving node telecommunications transceiver to each customer optical transceiver, a method comprising the steps of:

transmitting a first plurality of optical communication and information signals at a plurality of wavelengths from the serving node telecommunications transceiver;

splitting the first plurality of optical communication and information signals into a split plurality of optical signals, the split optical signals containing all of the wavelengths of the first plurality of optical communication and information signals;

controllably receiving said first plurality of optical communications and information signals at each customer optical transceiver;

performing wavelength division demultiplexing of said first plurality of optical communications and information signals;

performing wavelength division multiplexing of a second plurality of optical communications and information signals;

transmitting the second plurality of optical communications and information signals from at least one customer optical transceiver to the serving node telecommunications transceiver;

combining the transmitted second plurality of optical communication and information signals into a combined optical signal and forwarding the combined signal to the serving node telecommunications transceiver; and

dynamically controlling the customer optical transceiver to provide different bandwidth allocations over the fiber optic transmission path to the at least one customer optical transceiver.

28. The method of claim 27, wherein the step of dynamically controlling includes the step of dynamically providing the different bandwidth allocations upon demand by customer.

29. The method of claim 27, wherein the step of dynamically controlling includes the step of dynamically allocating the plurality of wavelengths.

30. The method of claim 27, wherein the step of dynamically controlling includes the step of dynamically allocating bit rates of the first or the second plurality of optical signals in response to a demand by a customer.

31. The method of claim 27, wherein the step of dynamically controlling includes the steps of:

causing said customer optical transceiver to transmit a bandwidth request signal at a common signalling wavelength over the fiber optic transmission path;

determining which one of the plurality of wavelengths carrying the first or second optical communications and information signals is idle; and

allocating said idle wavelength.

Description

FIELD OF THE INVENTION

This invention relates generally to the distribution of information and communications services via an optical fiber network. In particular, the invention relates to the control of wavelength division multiplexing in optical fiber connections of a network in order to more efficiently and cost-effectively distribute information.

CROSS REFERENCE TO RELATED APPLICATIONS

This application is related to U.S. application Ser. No. 08/656,879, filed May 30, 1996.

BACKGROUND OF THE INVENTION

The distribution of high bandwidth information, such as video, is frequently carried out over so-called hybrid fiber/coaxial (HFC) systems. These systems generally distribute the high bandwidth information in one direction only. The fiber optics are connected to the head end of the system at the information source and transport a large number of individual signals or channels over the majority of the distance between the head end and the user locations. The fiber optics usually terminate at a point relatively close to a user location or group of user locations and are transported over coaxial cables from the termination point to the user location or group of user locations.

A typical video signal distribution system for distributing TV programs from a central location is depicted in U.S. Pat. No. 4,994,909 to Graves et al. In that system, the television signals are transmitted by fiber optic connections to an interface network 30 and are transmitted by coaxial cables from there to a number of television receivers.

Many of the HFC systems are referred to as fiber to the curb systems. See, for example, U.S. Pat. No. 5,262,883 to Pidgeon and U.S. Pat. No. 5,133,079 to Ballantyne et al. Such systems typically use an optical to electrical converting terminal at the curb, sometimes referred to as an optical network unit (ONU), and deliver voice and video downstream to the home from the terminal using coaxial cable or twisted copper pair technology. U.S. Pat. Nos. 5,181,106, 5,189,673 and 5,303,229, commonly assigned to Alcatel Systems, discuss a network having a telecommunications central office connected to remote terminals by fiber optics with the connections to the customers over twisted pair (narrowband) and coaxial cable (broadband). POTS decoding and D/A conversion take place at the ONU and the digital signals limited to video information are decoded and D/A converted at the set top in the CPE.

The installation of ONUs and other active electronic units outside the premises (on telephone poles, electric curbside units, etc.) in fiber to the curb systems is labor and cost intensive and exposes them to inclement weather. The installation of electronic units outside the premises to provide conversion to wire transport also erases the advantage of additional security provided by the extreme difficulty in tapping fiber optics. Even those fiber optic systems having fiber optic drops from the curb to the home, such as in U.S. Pat. No. 5,325,223 to Bears, convert the optical signals to electrical signals at the curb units for the purpose of multiplexing and demultiplexing.

Typical fiber to the curb systems also are not efficient from the network side or the customer side in their use of bandwidth over the fiber, thus ultimately increasing the cost of the system. From the network perspective, the systems do not use bandwidth in an effective way. They also tend to be very rigid and to follow standard North American digital hierarchies or vendor proprietary hierarchies. Such interfaces also tend to be very channelized and to switch large chunks of bandwidth in a discrete type manner regardless of the customer's actual bandwidth needs. The upstream transmission is usually restricted to signalling messages, and the upstream transmission rate back into the network from the customer is extremely bandwidth limited and is not truly dynamic.

Any switching of signals or bandwidth typically occurs at the headend with a point to point transmission system from the headend to the ONU located close to the end user. The optical fiber or fiber optic pair to the ONU carries a very high bandwidth rate which is then subdivided into smaller bandwidths for each one of a plurality of customer premises. For example, the full bandwidth on the fiber may be OC-12, and that bandwidth may be subdivided into 8 different segments, each segment transporting information for one premises over coaxial cable or twisted copper pairs to that premises. See for example, copending, commonly assigned, patent application Ser. No. 08/413,215 filed on Mar. 28, 1995 entitled "Full Service Network with Distributed Architecture".

For video and POTS, all of the switching usually occurs at the headend with channelized transmission and access back through the active electronics at the ONU or elsewhere in the network. For example, in U.S. Pat. No. 5,136,411 to Paik et al, electrical service request signals from subscribers are converted into optical service request signals and sent to a headend terminal. The headend terminal then selects among a number of channels for transmission to the subscribers. In U.S. Pat. No. 4,506,387 to Walter and U.S. Pat. No. 4,709,418 to Fox et al, users can send an upstream request for a selected video program to be delivered downstream at a high rate over fiber optics to a receiving unit at the user's location where the video signals are then displayed. While channels or video programs can be selected, there is nevertheless no dynamic relationship in the allocation of fiber optic bandwidth between the headend and the customer.

From the customer side, the restricted upstream transmission prevents customers connected to fiber to the curb systems from being able to generate signals on the network for two-way transmission in a manner analogous to home page providers on the Internet. It would be preferable to have a system which enables a customer to send as well as receive, for example, full motion video to and from another location and which facilitates transmission of the content and routing signals.

Distribution systems which utilize optical fibers for the entire transmission path to the home are known but are generally regarded as being substantially more expensive and less flexible than HFC or fiber to the curb systems.

U.S. Pat. No. 4,891,694 to Way pertains to a fiber optic cable television distribution system in which each customer location is connected to a remote terminal via a dedicated optical fiber. Television signals are transmitted by optical fiber to a remote terminal and into CATV tuners. An optical fiber connects the remote terminal to the customer location, however the remote terminal is not the interface to the home.

U.S. Pat. No. 5,272,556 to Faulkner et al pertains to distributing HDTV signals from a transmitter at a head station to a number of receivers located at customer stations along an optical network.

U.S. Pat. No. 4,135,302 to Cutler pertains to a broadcasting system in which a signal path between a central station and a plurality of subscribers includes fiber optic transmission lines. The optical fibers extend over the whole length of the transmission paths between an electro-optical transducer at the central station and a distribution point or photo-sensitive detector of each of the plurality of subscribers.

Although these optical fiber systems are capable of delivering a large amount of information, they nevertheless offer little flexibility in switching between multiple channels or services.

Some fiber optic systems utilizing wavelength division multiplexing and providing telephone, data and video services are known in the prior art. For example, U.S. Pat. Nos. 5,121,244 and 5,175,639 to Takasaki suggest a fiber optic system having auxiliary fiber optic transmission lines for use when upgrading to higher bandwidth services. Although the fiber optic lines may be wavelength division multiplexed, the multiplexers and demultiplexers at both ends merely switch the entire bandwidth output of each fiber optic line. There is no provision made for altering or allocating different customer bandwidths over a single fiber optic line. Takasaki, for example, suggests providing auxiliary transmission lines for flexibility in providing high bandwidth services.

U.S. Pat. No. 5,221,983 to Wagner also discusses a fiber to the home system utilizing wavelength division multiplexing. A central office is connected to a plurality of remote nodes by a fiber. Each node connects to a subscriber premises via another fiber. The central office contains wave-division multiplexing modules, and their output is connected by fibers to the remote node.

U.S. Pat. No. 4,763,317 to Lehman et al discusses a communication network structured to carry both wideband and narrowband communications via optical fibers to the home. Examples of the various services that the network provides include telephony, audio, telemetry, packet-switched interactive data, facsimile, one-way video (TV), restricted one-way video (video-on-demand) and video conferencing. Network interface equipment at the subscribers' premises connects to remote network nodes via distribution optical fibers. Each distribution optical fiber is wavelength-division multiplexed and carries modulated (pulse-analog, pulse-code, or differential pulse-code) wideband digital channels as well as a multiplexed channel comprising 32 time-division-multiplexed narrowband digital channels. One of the narrowband channels carries all signaling messages.

Each remote node in the Lehman et al system comprises a digital space-division switch for wideband channels, and a digital time-division switch for narrowband channels. All switches are controlled by a central node control complex over a control bus and its extensions. Signaling messages are transferred between the signaling-message-carrying narrowband channels and the central node complex by a subscriber signaling subsystem, via the narrowband switch and the control bus. The central node optionally includes interfaces to other communication systems, as well as trunk communication fiber and Common Channel Inter office Signaling (CCIS) fiber connections to other central nodes of the network. The central node in FIG. 2 has fibers that transmit to the remote nodes, and the remote nodes have fiber connections to various subscribers.

Lehman et al is limited in flexibility because it utilizes a predefined static channelization on the loop fiber. If the service or bandwidth requirements of a subscriber change quickly, there is no means for establishing various connections through allocated bandwidth on the fiber(s) between the home and the central office. In particular, there is no means provided which make it possible to dynamically allocate bandwidth on the fiber distribution loops on demand.

Completely fiber optic systems such as the one in Lehman et al have limited or no built-in functionality for dynamic access control by the subscriber. Thus, there is no functionality that would allow a subscriber served by a fiber optic system to, for example, request, receive and pay for only the bandwidth which he desires.

Furthermore, completely fiber optic systems are usually limited to a specific service context such as either a fiber video system or a fiber POTS system. There is a need for an interface which can multiplex/demultiplex as required whatever fiber optic signal is sent, thus enabling more transparent access to different information sources with different bandwidth requirements.

BRIEF SUMMARY OF THE INVENTION

It is an object of the present invention to provide methods and apparatus which will overcome the problems and disadvantages of known fiber optic networks and will meet the needs discussed above.

It is an object of the present invention to deliver many different services to the customer using a single fiber optic network.

It is also an object of the invention to use fiber transport from a central office to an interface type device located within the customer premises.

It is a further object of the present invention to deliver service signals over a fiber transport, using various network identifiers that take the service signals through encoding technology such as ATM, to intelligent network interfaces at the customer premises which optically receive those service signals and deliver them to electrical devices in the customer premises.

It is also an object of the present invention to provide a fiber optic system having a dynamic bandwidth interface with the fiber at the point of end use having true dynamic bandwidth for whatever service is subscribed to or for whatever customer access is required.

It is a further object of the present invention to provide an intelligent interface inside the customer premises and to provide multi-service distribution inside the customer premises, including dynamic distribution between or within services.

A further object is to provide a long-term costeffective solution by delivering all wavelengths of a wavelength-division-multiplexed fiber optic network to all customers and using electronics at the customers' premises to discriminate the wavelengths in order to manage the bandwidth of the network.

A preferred embodiment of the invention provides an optical fiber connection (one fiber or two) of a common plurality of wavelengths from a serving node transceiver in a central office to each one of a plurality of intelligent interface devices in the customers' premises. The central office includes at least a narrowband switch and a broadband switch. The narrowband switch provides voice grade telephone service routing. The broadband switch provides routing for video services and may comprise an Asynchronous Transfer Mode (ATM) switch, an optical switch or the like. Preferably, the central office also includes a packet data switch.

At the subscriber premises, the intelligent interface device provides the connection to the optical fiber and performs two-way multiplexing and demultiplexing as well as any necessary signal format conversions. The intelligent interface device may also provide similar two-way optical to electrical conversion and interfacing for ISDN, telemetry, packet data, etc. Traditional media (twisted pair, coaxial cable, etc.) may be used inside the customer premises for the delivery of the various services.

The intelligent interface device in the subscriber premises also provides a broadband connection. The broadband link within the customer premises may take the form of an optical network, or the on-premises broadband distribution may rely on wireless transmissions. Alternatively, the interface may include equal division splitters to provide separate links serving individual terminal devices.

The intelligent interface device is thus analogous to an electrical distribution panel, in that the interface device distributes demultiplexed signals throughout the house. If multiple RF signals are delivered, they are converted and modulated at the interface. In this regard, the interface device basically performs the functionality analogous to an optical network unit (ONU) and/or a network interface module (NIM) in a terminal, except now at a central point in the home.

The network has media access control functionality and utilizes a dynamic media access control procedure. Bandwidth on the optical fiber loop is dynamically allocated to individual services on demand. The optical fiber(s) can carry three different wavelengths, and allocation of bandwidth can include wavelength selection as well as bit rate allocation.

A media access control unit in the interface controls communication over the fiber loop and allows total control of the optical fiber bandwidth allocation to each service. The customer may carry out any desired or necessary communications processing at the interface rather than at a settop box for a TV or VCR. For example, at the interface, conversion functions could be made to, for example, get picture in picture, watch and tape television programs at the same time, watch television and receive data, etc.

The central office includes a similar interface device connected to the subscriber's optical loop. The interface device in the central office provides two-way conversion between optical and electrical signals and performs the multiplexing and demultiplexing of the signals carried on the loop. The central office interface couples the various signals on the loop to and from the voice grade narrowband switch, the broadband switch and the packet data switch, as needed.

The intelligent interface device preferably receives power from the AC power grid in the subscriber's premises. Alteratively, the interface device may receive power from the central office via a twisted wire pair, e.g. the old telephone loop replaced by the installation of the optical fiber loop. The interface may also control any alarming, power monitoring or other utility monitoring.

Additional objects, advantages and novel features of the invention will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following or may be learned by practice of the invention. The objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a simplified diagram of an illustrative embodiment of an optical fiber-to-the-home network in accordance with the invention in which wavelength division multiplexed information is transmitted to a customer premises.

FIG. 2 is a four layer stack diagram, useful in explaining the functions performed by the elements comprising the fiber optic network shown in FIG. 1.

FIG. 3 is a simplified block diagram of matching fiber optic interfaces used in the network shown in FIG. 1.

FIG. 4 is an illustrative block diagram of a media access controller used in the network shown in FIG. 1.

FIG. 5 is an illustrative diagram of wavelength routing employed in another embodiment of the invention.

FIG. 6 is an illustrative diagram of a SONET optical network that may be optimally used with the dynamic bandwidth access procedure providing a plurality of services to the customer's premises.

FIG. 7 shows the IID 101 of the SONET network embodiment of FIG. 6 in more detail.

FIGS. 8A, 8B and 8C show exemplary CPE interfaces for residential, small business and large business customers, respectively.

FIG. 9 shows an alternative embodiment of the invention in which multiple wavelengths are delivered to each customer in the network.

FIG. 10 shows a preferred embodiment of CPE transceiver 907 in FIG. 9.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

An exemplary fiber optics network having dynamic bandwidth allocation according to the invention is illustrated in FIG. 1. The Intelligent Interface Device (IID) 101 is one element of the customer premises equipment (CPE). Although illustrated simply as an optical fiber from a telecommunications central office (CO) to the CPE in FIG. 1, it should be understood that the fiber optic transmission path 104 may be composed of any number of duplex fibers or any plurality of simplex fibers, including a fiber optic bundle. It should also be understood that FIG. 1 is a simplified illustration of a network and that the fiber optic network may be of any configuration and may be extremely complicated.

The IID 101 contains a media access controller (MAC) 102, connected directly to fiber optic transmission path 104, which converts input optical signals from fiber optic transmission path 104 to electrical signals and converts electrical signals to optical signals.

A similar media access controller 105 on the network side preferably connects to a plurality of fiber optic transmission paths, 104 to 104.sub.N, each one of which is connected to any number of different customer premises, 1 to M. In an embodiment having multiple customer premises locations, passive optical splitters and couplers (not shown) are preferably used to optically distribute all of the signals on a single fiber optic transmission path. Each fiber optic transmission path 104 to 104.sub.N, is preferably full duplex with two logical paths.

The serving node, preferably a telephone central office, comprises media access controller 105 and any number and variety of switched inputs providing communications and information signals. The network shown in FIG. 1 includes a Switched Multi-megabit Data Services (SMDS) network 106, a narrowband telephone switch 107, a broadband switch 108, an optical switch 109, and a data packet switch 110. The media access controller 105 receives four input rails, each having a different service through one of the switches or network connections, and selectively couples each one to a different wavelength transmitted on the fiber optic transmission paths 104 to 104.sub.N.

Regardless of the network architecture or other characteristics, the transmission path over the fiber optic medium 104 is dynamically managed, and the bandwidth is dynamically allocated on a regular basis, by media access controller 105 at the network side and each one of media access controllers 102 to 102.sub.M on the subscriber side. Preferably, the management and allocation of bandwidth is carried via a logical control link from one of the media access controllers 102 at the customer premises back to the CO, and control information is provided to the media access controllers to dynamically manage the channelization and assign bandwidth.

INTELLIGENT INTERFACE DEVICE

An important feature of the fiber optics network of the invention is the intelligent interface device (IID) 101, preferably comprising a full duplex media access controller (MAC) 102 and one or more customer interfaces, located at the customer premises. The IID 101 is connected to at least one fiber optic network through MAC 102 and to any number of different media in the customer premises through appropriate customer interfaces.

The MAC 102 contains an optical receiver which receives light from fiber optic transmission path 104 and then converts it directly into corresponding electrical signals. An optical transmitter converts electrical input signals to light transmitted directly into fiber optic transmission path 104. MAC 102 and its optical receiver and transmitter are permanent components of the intelligent interface device.

The customer interfaces preferably comprise a number of respective service definition modules, 103 to 103-M, such as those available from Broadband Technologies (BBT) for use with their FLX fiber-to-the curb systems. The modules are not permanent. Each of these modules can be individually loaded into the IID between MAC 102 and the customer interface, replaced, upgraded by the customer as desired, etc. Description thereof is provided in the Executive Overview of the FLX System, Section BBT-200-901, Planning and Engineering, Issue 2.3X, October 1992, pp. 3-28, which is hereby incorporated in its entirety by reference. The service definition may provide conventional services. For example, a telephone service definition module would comprise circuitry for all of the necessary BORSCHT (battery, overvoltage, ringing, supervision, coding, hybrid and test) functionality to provide normal telephone line service to standard customer premises telephones connected to the interface via a twisted wire pair. An ISDN module would comprise the structure and functions of an ISDN PC card, such as the CyberSpace Freedom.TM. series of ISDN cards available from ISDN*tek.

However, the modules 103 to 103-M in the preferred embodiment of the invention are not limited to the functions of conventional network communications interfaces. They can support a number of different higher level functions and perform other functions such as decoding of the image compression scheme for full motion video proposed by the Motion Picture Experts Group (MPEG), image processing, etc. For example, a digital video service definition module would perform the functions of a set-top box and, in particular, a digital entertainment terminal (DET). The architecture and functional details of an exemplary DET can be found in commonly assigned copending U.S. patent application Ser. No. 08/380,755 filed on Jan. 31, 1995 and entitled "Digital Entertainment Terminal with Channel Mapping" or commonly assigned copending U.S. patent application Ser. No. 08/498,265 filed on Jul. 3, 1995 and entitled "Downloading Operating System Software through a Broadcast Channel", both of which are hereby incorporated by reference in their entirety.

Although the service definition modules 103 to 103-M are preferred, the IID 101 can also include network interface module (NIM) and/or processing circuits, such as a MPEG decoder, for receiving broadcast or video-on-demand services. The modules also support a variety of other communications services, such as telemetry meter reading, energy saving, remote control, etc. The modules support a plurality of different physical CPE media such as ISDN, rf coaxial cable, digital coaxial cable, local fiber optics, twisted copper wire pairs. In order to facilitate routing, the customer interfaces may include, or be connected to, optical combining networks so that the optical signals can be bused around the CPE analogous to electrical signals in copper wires.

The IID 101 includes means to receive selection signals from a user via remote control, direct entry or through the distribution network internal to the CPE. The IID responds by transmitting appropriate data signals over a narrowband signaling wavelength channel on the fiber optic transmission path 104 to the corresponding MAC 105. The MAC 105 is also addressable and/or assigned a time slot for signalling communications. If the data represents bandwidth request, wavelength request or bit rate request selection signals, the MAC 105 dynamically responds to that data by allocating bandwidth, wavelengths or bit rate connections as outlined above, and stores data identifying each subscriber's selections or preferences for subsequent communications.

The IID 101 can also be provided with a number of interfaces generally designed according to the needs of the user. FIGS. 8A, 8B and 8C show three different optical transceivers. FIG. 8A depicts a transceiver for residential uses. That transceiver contains a low speed SONET terminal multiplexer which takes the OC-12 two-way and provides a two-way split down to a DS1 internal interface. Standardized service definition module cards provide an RJ-11 jack for conventional telephone service, an ISDN external interface and an ATM data is made available on twisted copper wire pair for interactive video, etc. FIG. 8B illustrates a transceiver designed for small business. That transceiver generally is the same as the residential transceiver in FIG. 8A, except for the addition of a high capacity DS-3 output that segregates it in one or more high cap services. The transceiver of FIG. 8C is designed for large business and would have a high capacity OC-3C optical interface and an SMDS interface. These interfaces can go into a Private Branch Exchange (PBX).

The logical functions of the IID can best be understood by reference to the four layer diagram shown in FIG. 2. The layers 1-4 in FIG. 2 are not actual elements or functions. Bottom layer 1 in FIG. 2 represents the physical layer and is comprised of the medium and the terminating connectors for the medium, for example, a twisted copper telephone wire pair and associated RJ-11 jacks. The MAC described above is a level 1 device, it converts the optical signals into electrical signals and vice versa.

The next layer, layer 2, is referred to as the data link layer and is the logic that directly controls the physical layer. The link layer is comprised of, for example, the input electrical control signals for the optical transmitter, either digital or analog. This layer performs the wavelength multiplexing and demultiplexing functions of IID 101 by selectively controlling the optical transmitter and receiver of the MAC 102.

The next higher layer, network protocol layer 3, accomplishes the signalling, switching and connection of services to create the network. Although shown as a single layer in FIG. 2, it is the most complicated of the layers and may be composed of several sub-layers. It includes, for example, communications protocols and discrete switch layers for disparate services such as Ethernet Local Area Network (LAN), FM, Switched Mega-bit Data Services (SMDS), frame relay, etc. In the network, the corresponding signalling layer also includes Level 1 gateways for interactive video and SS7 networks for setting up connections across virtual switches for narrowband services. In the IID, the layer 3 functionally communicates with such signalling layer elements of the network.

Layer 4 rides on top of all of these other layers and constitutes the actual data being communicated from one party to another on the network. The payload, such as a video signal, can exist and remains the same in different network set-ups corresponding to layer 3 (ATM or SONET), different physical links comprising layer 2 (wavelength division multiplexing or time division multiplexing), and different physical medium in layer 1 (either fiber optic, electrical wire, wireless transmission, etc.).

Layers 1-4 represent different levels of complexity in which there can be alternative implementations. For example, video can be transported over either coaxial cable or fiber optics (different physical layer). The signals can be wavelength division multiplexed or time division multiplexed on the physical medium. The multiplexing can occur in an ATM network or a SONET network.

Most of the expense incurred in the installation of large bandwidth fiber optic networks is associated with the physical equipment (fiber optics, optical splitters, O/E converters, etc.) in physical layer 1. Cost effective advantages can be achieved by effectively using the bandwidth through optimum development of the equipment, protocols and programmed logic corresponding to layers 2 and 3, in particular, the switching of different bandwidth services and services operating differently at the upper level, network level 3. Current fiber optic systems typically handle level 3 operations by a number of discrete switches, each corresponding to a respective service such as Ethernet LAN, FM, SMDS, frame relay, etc., which are located exclusively in the central office. Such level 3 operations are handled or controlled, at least in part, by switches and other equipment within the intelligent interface device of the preferred embodiments.

Media access controller 102 of IID 101 shown in the preferred embodiment of FIG. 1 contains an interface between physical layer 1 and the higher layers and carries out level 1 functions. It channels optical and electrical signals of varying bit rates and formats to and from the physical medium.

The downstream optical signals can be broken down into individual electrical bits, frames or packets, which are output to be decoded in level 3 by the service definition modules 103 to 103-M of the IID 101, and then delivered to the electrical devices in the CPE and converted to physical layer 4 output. For example, the optical signals can represent MPEG encoded video, then can be converted to electrical signals and decoded by an MPEG decompression circuit and D/A converted into a baseband video signal in a service definition module and then delivered to a television set and converted to a displayed image.

In the preferred embodiment shown in FIG. 1, fiber optic transmission path 104 is preferably a single multi-fiber optical pipe, and media access controllers 102 and 105 carry out wavelength division multiplexing and time division multiplexing of the signals transported on the pipe. Spatial division is also carried out by passively splitting the different fibers of the pipe and terminating them at different customer premises.

The wavelength division multiplexing is carried out to the maximum extent safely permitted by the physical properties of the fiber optics. Presently available equipment is capable of effectively operating in three separate wavelength regions: 850 nm, 1300 nm, and 1550 nm. Each wavelength region can support many separate discrete wavelengths. For example, in the 1300 nm region, there is a 1290 nm wavelength signal, a 1300 nm wavelength signal and a 1310 nm wavelength signal. The increments of permissible wavelength divisions and the amount of loss or dispersion vary in the different wavelength regions. For example, the 1550 nm region has the lowest loss, but the 1300 nm region has the best dispersion characteristics.

Of course, there may be more than the three specific wavelength regions mentioned above, and each wavelength region may have wavelength divisions controlled down to the 10ths or hundreds or thousands of Angstroms. In such a case, a larger number of wavelength multiplexed signals are available for simultaneous use.

FIG. 4 shows a preferred embodiment of media access controller 105. A separate laser is used for each wavelength. The lasers can each be on separate semiconductor chips, but the optical transmitter in the preferred embodiment is comprised of a single integrated (IC) chip 401. The IC 401 has an array of 4, 8 or 16 lasers, each one of which produces pulses at a different wavelength within a common wavelength region. The lasers are not variable or tunable, but the transmitting wavelength(s) are selectable.

The physical light pulses produced by the lasers on IC 401 are actually coupled by integrated optics and supplied directly into a single fiber 402 so that the light is already wave division multiplexed when it is input into the fiber. The bandwidth of the combined output of the four lasers is in the 10-50 Gb/s range using presently available commercial equipment. If even more bandwidth or multiplexing is necessary, two laser array ICs may be used with an optical combiner/coupler to feed the output of both into one optical fiber.

All of the lasers operating in a single wavelength region are preferably made from a common substrate (for example, indium phosphate for lasers in the 1300 nm wavelength region) so that the relative difference in wavelength between the lasers is very stable. For example, if there is an increase in temperature of an IC which causes the wavelength of a 1300 nm laser to increase to 1304 nm, the output of a 1310 nm wavelength laser on the same chip will similarly increase to approximately 1314 nm. Therefore, the difference between the output wavelengths will stay approximately constant at 10 nm and will not decrease to 6 nm. The wavelengths consequently always have guardbands to prevent intermodulation.

The semiconductor lasers operating at discrete wavelengths within different wavelength regions are fabricated on different semiconductor chips, each chip corresponding to one of the wavelength regions. The outputs of each chip are then combined in an optical combiner/coupler onto the fiber. However, if fabrication technologies develop to the point where lasers operating at wavelengths of different wavelength regions can be placed on the same substrate, then a single chip can be used.

The optical receiver 403 has to be able to selectively receive and detect each one of the wavelengths on fiber optic 104. Receiver 403 could be controlled so that the light from optical fiber 104 goes to a detector that is only sensitive to the desired wavelength. However, it is preferable for optical receiver 403 to receive all of the different wavelengths on the fiber 104. A wavelength dependent splitter 404 selectably and optically splits the wavelengths and sends each wavelength to a separate detector 406.sub.1 to 406.sub.4. Each one of the separate detectors is generally comprised of the same broadband detector, such as an avalanche diode, which has a relatively flat sensitivity response for all of the wavelengths on fiber 104.

The illustration of the laser array constituting the optical transmitter and the wavelength dependent splitter of the optical receiver in FIG. 4 is essentially functional. Similarly, I/O and media access logic section 407 is illustrated as merely performing the function of obtaining and forwarding the appropriate inputs to the optical transmitter and accepting the outputs of the optical receiver. However, I/O and media access logic section provides switchability and performs the level 2 and 3 functions described elsewhere in this application.

An important feature of the invention is that the central office media access controller 105 can best be thought of as having a matching interface with each one of a plurality of media access controllers 102 in corresponding customer premises. FIG. 3 illustrates a pair of such matching optical interfaces 301, preferably SONET interfaces, connected via optical fiber 104. The optical interfaces are preferably identical to each other, the only difference being between the I/O and media access logic section of the central office media access controller 105 and the CPE media access controller 102.

The IID 101 interacts with the central office in a dynamic media access control procedure to decide the allocation of wavelengths and bandwidths to customers for each call or service at required bit rates. In particular, the dynamic bandwidth assignment can be carried out intra-session at the customer's request. For example, when a customer on his or her computer connected to an information service through the network desires to transmit or receive a large amount of large bandwidth data such as video, he can request that the bandwidth to his premises be substantially increased. If the customer has an optical customer interface on his or her IID and an optical card in the computer, the bandwidth can be increased up to the bandwidth of the fiber coming into the premises.

The following execution of a narrowband telephone call is given as an example of the dynamic bandwidth allocation procedure in the embodiment of FIG. 1. When a telephone set 111 goes off-hook a relay closes and current flows through the telephone and twisted wire pair 112. The intelligent interface device 101 recognizes this as an off-hook condition. In response, the interface 101 transmits a request for dial tone upstream over fiber 104, via some established signaling channel and predefined signaling protocol. If the bandwidth necessary for the telephone call is available on fiber 104, the interface at the central office or corresponding MAC 105 allocates a wavelength and a time slot on fiber 104 to the desired telephone call and transmits back a signaling message identifying the allocated wavelength and time slot to the interface device 101 at the customer premises. The interface at the central office concurrently establishes a connection to narrowband telephone switch 107. At this point, the fiber optic system has established a telephone grade link through from the telephone set 111 to the narrowband telephone switch 107, and the switch 107 can accept dialed digits and complete the call through the PSTN in the normal manner. Similar procedures are used to obtain broadband channels and associated signaling channels on fiber 104 and establish connections to appropriate broadband switching systems, such as broadband switch 108, to set up sessions with video information providers, e.g. for video on demand type services.

When bandwidth allocation is employed using wavelength division multiplexing (WDM), a large bandwidth, completely fiber optic system becomes much more economically feasible. The physical interfaces necessary for fiber optic transmission of 2 gigabits per second (2 Gb/s) over a single wavelength (optical transmitter, fiber optic and optical receiver) are very expensive. It is cheaper to use, for example, four sets of 500 Mb/s interface equipment for four separate wavelengths on the fiber.

In addition to transmitting and receiving light in different wavelength bands as described above, the media access controllers 102 and 105 also controllably perform the modulation and demodulation of the light on a selectable basis. They carry out coherent transmission and subcarrier modulation of light on the fiber, with ample consideration given in order to prevent beats from occuring between wavelengths. The use of N wavelengths in an optical fiber generally multiplies the bandwidth of the fiber by a factor of N. The number N which is practically obtainable is restricted by the physical properties of either the transmitter, the fiber optic or the receiver. The number may increase with improvements in one or all of these elements, but the distinguishing characteristics of this invention will remain for all values of N.

Because of the cost associated with the physical transceivers necessary to convert the optical signals of very high bandwidths to electrical signals and vice versa, wavelength division multiplexing creates a cheaper transmission path for distribution than time division multiplexing in fiber optic networks. The large savings achieved at the physical layer, layer 1 in FIG. 2, more than compensates for the complexity introduced into layers 2 and 3. The different wavelengths may be used for different services or for different customer premises. Separation of wavelengths for each customer provides security in the distribution of information because one customer premises does not receive the information selectively distributed to another customer premises.

A preferred embodiment for distribution (separation and routing) of the wavelength multiplexed signals is illustrated in FIG. 5. The optical transmitter 401 of MAC 105 is connected to the optical receiver in each one of a plurality of customer premises and transmits and receives, full duplex, a multiple number (n) of different wavelengths. A passive wavelength dependent splitter 501 receives the fiber optic signals from the optical transmitter 401 and separates the transmitted optical signals by wavelength into two legs, thereby demultiplexing the wavelength division multiplexed optical signals according to wavelength. Each leg is input to another wavelength dependent splitter 502 which splits the leg into two wavelengths.

Each one of the separated wavelengths is forwarded to a corresponding CPE wherein it is received by one of the MAC's 102.sub.1 to 102.sub.4. The structure of the physical interface of MAC 102 does not need to be different than the physical interface of the central office MAC 105. The optical receiver is controlled to be responsive to the separated wavelength and the laser in the transmitter array which corresponds to the separated wavelength is selected and controlled for transmitting optical signals to MAC 105. Multiplexing and demultiplexing no longer must be carried out by IID 101. Dynamic media access control occurs through control of the allocation of wavelengths to each one of the plurality of customer premises and/or to each one of a plurality of different requested services further up in the network.

SONET NETWORK

A synchronous optical network (SONET), connected to the CPE according to the present invention and utilizing the intelligent interface device of FIG. 4, is shown in FIG. 6.

The basic module or first level of the Synchronous Optical Network (SONET) signal is called the Synchronous Transport Signal-Level 1 (STS-1). The STS-1 has a bit rate of 51.84 Mb/sec and is synchronous. The STS-1 signal is formed from a sequence of repeating frames. The STS-1 frame is illustrated in FIG. 2 of U.S. Pat. No. 5,293,376. The STS-1 frame structure can be drawn as 90 columns by 9 rows of 8-bit bytes. The order of transmission of the bytes is row by row, from left to right across the columns, with one entire frame being transmitted every 125 micro-seconds. The 125 micro-second frame period supports digital voice signal transport encoded using 1 byte/125 micro-seconds=64 kb/s. The first three columns of the STS-1 frame contain section and line overhead bytes. The remaining 87 columns form the STS-1 Synchronous Payload Envelope (SPE). The SPE carries SONET payloads including 9 bytes of path overhead. The STS-1 can carry a clear channel DS3 signal (44.736 Mb/s) or, alternatively, a plurality of lower-rate signals such as DSO, DS1, DSlC, and DS2 by dividing the Synchronous Payload Envelope into a plurality of fixed time slots. For example, 648 DSO signals fit into the SPE of an STS-1 signal.

Higher rate SONET signals are obtained by byte interleaving N frame aligned STS-1 signals to form an STS-N signal in accordance with conventional SONET technology. An STS-N signal may be viewed as having a repetitive frame structure, wherein each frame comprises the overhead bits of N STS-1 frames and N synchronous payload envelopes. For example, three STS-1 signals may be multiplexed by a multiplexer into an STS-3 signal. The bit rate of the STS-3 signal is three times the bit rate of an STS-1 signal and the structure of each frame of the STS-3 signal comprises three synchronous payload envelopes and three fields of overhead bits from the three original STS-1 signals. When transmitted using optical fibers, the STS-N signal is converted to optical form and is designated as the OC-N signal. A more detailed description of SONET appears in U.S. Pat. No. 5,293,376 and in Davidson et al, The Guide to SONET, Telecom Library Inc., 1991, both of which are hereby incorporated by reference in their entirety.

The SONET network of FIG. 6 preferably comprises a plurality of fiber optic hubs 601.sub.1, 601.sub.2, to 601.sub.N on the SONET network side of the fiber optic medium of ring 602. Each fiber optic hub can drop and add optical signals to and from the SONET ring 602. The SONET ring 602 may receive signals from a variety of input sources and distributes those signals through the media access controllers 105.sub.1 to 105.sub.N. Each one of the respective media access controllers 105.sub.1 to 105.sub.N contains the same optical transmitter 401 and optical receiver 403 shown in FIG. 4. However, the I/O and media access control logic sections are different and provide a switchable connection to SONET gateway mux 603 through SONET ring 602 instead of to the switches shown in FIG. 1. The SONET gateway multiplexer 603 has a plurality of network interface cards 604 for receiving a variety of inputs. For example, a narrowband switch 605 provides narrowband telephone communications from telephone network 606. An ATM switch 607, connected to SONET gateway multiplexer 603, couples ATM signals to and from ATM switch 607 on the SONET ring 602. The links to SONET gateway mux 603 and the ATM switch 607 may be those in copending, commonly assigned, patent application Ser. No. 08/413,215 filed on Mar. 28, 1995 entitled "Full Service Network with Distributed Architecture", which is hereby incorporated by reference in its entirety.

These inputs to the gateway 603 preferably include narrowband transport for voice and narrowband data services. A digital switch or an analog implementation of a Service Switching Point (SSP) switch 605 provides standard type plain old telephone service (POTS) for telephone customers 606. The digital POTS switch provides a DS1 type digital input/output port through interfaces conforming to either TR008 or TR303. The DS1 goes directly to a SONET gateway multiplexer 603. The multiplexer may also receive telephone signals in DS1 format from an analog switch through a central office terminal. The central office terminal converts analog signals to digital and digital signals to analog as necessary to allow communication between the analog switch and the rest of the network.

The signals are made available at one of the SONET drop/add hubs 601 and wavelength multiplexed by the corresponding MAC 105 (possibly time division multiplexed with a number of DS1 signals) for transmission over one wavelength on the fiber optic transmission path 104 to IID's 101.

A preferred embodiment of the IID 101 for use in the SONET network of FIG. 6 is shown in FIG. 7. Each IID 101 preferably contains an optical interface 701 for two-way conversion between optical and electrical signals, a SONET terminal multiplexer 702, and a plurality of service definition module cards 703.sub.1 to .sup.703.sub.M. Preferably, one of the service definition modules 703 provides telephony service through SONET terminal mux 702 in compliance with TA-TSY-000303, IDLC System Generic Requirements, Objectives and Interface: Feature Set C--SONET interface (Supplement 2); Bellcore, Issue 3, December 1987, hereby incorporated by reference. The standard protocol for MUX 702 to signal with SONET drop/add fiber hubs 601.sub.1 to 601.sub.N, and the functionality of MUX 702 are defined in TA-TSY-000253, SONET Transport Systems: Common Generic Criteria; Bellcore, Issue 5, February 1990. Other services are also provided by other service definition modules 703 through SONET terminal mux 702. One of the service definition module cards preferably provides ATM cells as output with STS-1 interface on twisted pair to an ATM Packet Assembler/Disassembler (PAD) 704 and a terminal device 705.

The ATM switch 607 receives and outputs broadcast video signals 609 and/or video signals from a Level 2 Gateway 610 under the control of Level 1 Gateway 608. The ATM switch 607 uses known encoding technology to provide various network identifiers (e.g., VPI/VCI values) that take the service signals through the fiber optic network to corresponding broadband electrical devices at the customer premises. The network identifiers are preferably used only in ATM service signals at different network nodes in a manner consistent with that in copending, commonly assigned, patent application Ser. No. 08/413,215 filed on Mar. 28, 1995 entitled "Full Service Network with Distributed Architecture", hereby incorporated by reference in its entirety. The Intelligent Interface Devices 101 which optically receive those service signals, convert them to electrical signals and deliver them to electrical devices in the customer premises based in part on the network identifiers.

Of course, the broadband SONET network shown in FIG. 6 may have any number of connected sources and end users and may deliver any type of, or any number of types of, broadband information. For example, the network could also transmit digital multimedia information and the customer premises equipment could include a personal computer and computer monitor instead of, or in addition to, a DET and television. The broadband information is made available at one of the SONET drop/add hubs 601 and wavelength multiplexed by the corresponding MAC 105 (possibly time division multiplexed with a number of DS1 signals) for transmission over one wavelength fiber optic transmission path 104.

The network executes the dynamic media access control in three phases. As an illustrative example, consider a telephone call. A twisted pair connection is established through the customer premises to the interface device when a telephone goes off-hook. A channel is developed on a wavelength for the signaling communications of the dynamic media access control procedure carried out between each IID 101 and a corresponding MAC 105 using either an asynchronous optical protocol, or the synchronous (SONET) OC rate transport. The signaling information may be broadcast over all of the fiber optic couplings, 612.sub.1 to 612.sub.N to optical coupler 611 in a time division multiplexed channel. In this case, each one of the IID's 101 is addressable and/or is assigned a predetermined time slot and includes means to selectively receive and decode the signalling data indicating the bandwidth allocation for that IID 101. Using the signalling channel, the MAC 102 in the interface 101 then negotiates with MAC 105 to obtain a bandwidth path for the voice on an assigned wavelength on fiber path 104. The MAC 102 performs multiplexing and demultiplexing in accordance with the decoded data obtained and stored by the IID 101, to provide two-way communication via the allocated bandwidth.

The fiber optic network of FIG. 9 comprises a serving node transceiver 901, a bi-directional wavelength division multiplexed optical fiber 902 and a 1 to 2 passive power splitter/combiner 903. Transceiver 901 receives all of the various available services (video, telephony data, POTS, etc.) as inputs 909.sub.1 to 909.sub.N controllably interconnected to a number of lasers (not shown), each operating at a wavelength available in fiber optic 902.

In the downstream direction from serving node transceiver 901 to each one of the CPE's 907, passive power splitter/combiner 903 splits light from optical fiber 902 and couples the light into two fibers 904. Each fiber 904 has all of the wavelengths on it and is, in turn, coupled to one of the passive power splitter/combiners 905, which again splits the light into four equal parts for transmission over four identical fibers 906 with all wavelengths on each fiber. Each one of the fibers 906 provides a downstream light input to a respective one of the four CPE's 907.

In the upstream direction from CPE's 907 to serving node transceiver 901, passive power splitter/combiners 905 combine light from four fibers 906 into light for transmission over a single fiber 904. The passive power splitter/combiner 903 combines the light from two fibers 904 into light for transmission over one fiber 902 to the transceiver 901.

All of the fibers 902, 904 and 906 in the embodiment shown in FIG. 9 are bidirectional. The passive power splitter/combiners are preferably purely passive, carry all wavelengths to all customers and to all transceivers and combine all return wavelengths from all of the customers and the transceivers. The elements form a shared bus architecture with a fiber bus wherein the fiber optic links carry all frequencies over fixed distances.

An alternative embodiment of the network can employ two separate uni-directional transmission paths: one in the direction from serving node transceiver 901 to CPE's 907 and the other in the direction from CPE's 907 to serving node transceiver 901. Each transmission path uses a respective set of fibers carrying traffic in only one direction and uses either splitters or combiners instead of integrated splitter/combiners. Although such an embodiment will generally be more expensive than an embodiment with bi-directional fibers, it is not critical that the splitters and combiners be passive.

Another alternative embodiment of the network can serve a larger number of CPE's 907, such as nine or sixteen, simply by using a different number of 1 to 2 passive power splitter/combiners or by using 1 to 3 passive power splitter/combiners or 1 to 4 passive power splitter/combiners instead of the 1 to 2 passive power splitter/combiners.

There are similarities and differences between the embodiment shown in FIG. 9 and an embodiment in which wavelengths are discriminated in the network and only one single wavelength or wavelength region is delivered to each CPE (e.g., FIG. 5 in which one wavelength or wavelength region is provided to each premise and carries all of the information for that premise and all other wavelengths in the network are blocked) . The transceiver 901 at the serving node in the embodiment of FIG. 9 is very similar to the transceiver of media access controller 105 in the embodiment of FIG. 1, the only slight difference being in the logic of the control program or software so that the network is managed a little differently.

The discrimination of wavelengths by the CPE 907 is crucial to the embodiment of FIG. 9 and an important difference lies in the controlling management of multiple wavelengths by the I/O and media access control logic section in CPE transceiver 908. The relationship between the optical transceiver and the media access control logic section is discussed above with respect to FIG. 4. Each transceiver 908 in the CPE 907 of FIG. 9 is able to access all wavelengths. When a particular subscriber or particular subscriber device needs bandwidth and/or a communication link, media access controller 911 performs a management function that allocates and controls the bandwidth and/or communication link to the transceiver 908. The media access controller 911 also controls the input and output of electrical signals to the optical transceiver 908 in order for them to be transmitted and received over the bandwidth and/or communication link that it has allocated and controlled. The management function preferably also specifies the wavelength which is going to be used and, if it is time-division-multiplexed, the timeslots of that wavelength that are going to be used as well. The optical transceiver 908 performs only the functions of O/E conversion, E/0 conversion, transmission of optical signals and reception of optical signals.

Preferably, the transceiver 908 has a tunable laser for transmission back to the network as well as a tunable detector of some sort for the detection of optical signals on the specified wavelength or multiple wavelengths as shown previously in FIG. 4. The tunable laser is preferably a selectable laser array in which each one of a plurality of lasers operates at a predetermined wavelength.

In one embodiment, all of the downstream traffic to a CPE 907 is allocated on only a single wavelength and transceiver 908 of that CPE "tunes into" that wavelength in order to derive all of the POTS, ISDN, ATM, video, etc., traffic off of that one wavelength. The single wavelength is not shared with other CPE's while in use. The downstream wavelength appears at different homes, but it is not assigned to different homes at the same time. The one wavelength is used up to its capacity. If the CPE's traffic demands went beyond the capacity of the single wavelength, then the serving node of the network allocates the customer a second wavelength which is used in the same way as the single wavelength. The upstream traffic is preferably allocated to the same wavelength as the downstream traffic, but it can also be allocated to a different wavelength. The upstream traffic is inherently received only at the serving node transceiver and cannot be received by different homes. (In an alternative embodiment, the video traffic can be one wavelength, POTS on another wavelength, etc.)

A subscriber gets basic low bandwidth services, and the CPE transceiver 908 makes calls through network connections and carries out two-way communications all the way to network transceiver 901 according to the following description made with reference to FIG. 10. The functionality of transceiver 908 is expanded to have POTS delivered into the house on twisted wire pair. The preferred embodiment shown in FIG. 10 includes an ISDN terminal adapter 1001 that provides conventional ISDN on the central office side and looks like a POTS line on the customer side. ISDN terminal adapter 1001 converts from POTS signaling to ISDN Q.921/Q.931 signalling, performs analog-to-digital and digital-to-analog conversion and has a plurality of different call control message sets installed. The central office switches (not shown) also have the ISDN protocols installed.

When a phone at CPE 907 goes off-hook, the terminal adapter exchanges messages with the switch using the D-channel of the ISDN to more or less get the switch's attention. A Link Access Procedure-D (LAP-D) protocol allows a link to serve multiple users. The switch then establishes call set-up, sends a message instructing the terminal adapter 1001 to provide dial tone on twisted wire pair and sets up a DS-1 path.

Preferably, the terminal adapter provides a primary rate interface (PRI) ISDN (23B+D) on a DS-1 channel on the network side. The D channel is used for signalling the call control. A SONET multiplexer 1002 time-division-multiplexes (TDM) the DS-1 up to a higher rate and also provides DS-3 and OC-3 interfaces to provide high bit rate data services such as video services. All of the different types of traffic, including the PRI ISDN signal, are time-division-multiplexed by TDM SONET MUX 1002; and the multiplexed signal is used to modulate a laser which outputs signals on a single respective wavelength .lambda..sub.1 to the network.

Each TDM SONET MUX 1002 is preferably assigned a different wavelength for signalling to avoid collisions at optical combiners. However, a common default wavelength may be used by each of the receivers as a common signaling channel with a random access technique such as slotted ALOHA in which each transceiver 908 is active during a certain time interval.

A small amount of certain traffic, such as OC-1 carrying the POTS or ISDN service, may be permanently assigned to the primary wavelength .lambda..sub.1 in a static mode so as to provide continuous services. All services to and from the CPE can be handled up to the OC-1 rate before any bandwidth adjustments are made. The idle channel traffic is going to the D channel. The only traffic going to the network in an idle condition is the idle signalling channel information in wavelength .lambda..sub.1.

Additional narrowband services and all broadband services are dynamically allocated bandwidth through control of TDM SONET MUX 1002. Bandwidth is dynamically allocated through the manipulation of wavelengths within the fibers and time-division-multiplexing within wavelengths. The dynamic allocation includes a negotiation between a CPE media access controller (CPE MAC) 911 providing program control of transceiver 908 in CPE 907 and a serving node media access controller (MAC) 910 providing program control of serving node tranceiver 901. Although shown in FIG. 9 as being connected by a separate token bus 912, CPE MAC 911 and MAC 910 communicate over wavelength .lambda..sub.1+1 in the fiber optic network. Serving node transceiver 901 has ultimate control over delegating the bandwidth and CPE transceiver 908 is more or less the master allotting bandwidth for corresponding CPE 907.

As mentioned above, all wavelengths .lambda..sub.1 to .lambda..sub.N carry data and are delivered to each one of CPE transceivers 908. The additional wavelength .lambda..sub.N+1 is preferably used as a common signalling wavelength which carries all the media access bids for control. Each CPE transceiver 908 has a tunable optical transmitter and a tunable discrete receiver which are capable of discriminating all of the various wavelengths including the media access wavelength.

When a terminal device in the CPE 907 requests or needs wavelength(s) in addition to the statically allocated wavelength to carry out a service, the transceiver 908 determines which one of the wavelengths is idle through negotiations transmitted over the media access wavelength and assigns the idle wavelength(s) to transceiver 908.

The negotiations over the media access wavelength can be made in accordance with any one of a number of protocols used to avoid collisions, such as LAP-D or slotted aloha. A preferred embodiment of the invention uses a token bus operating in the 1 megahertz or higher range so that it seems almost instantaneous to each customer. The network signalling system in control of assigning the idle wavelengths is also the token controller and it passes out the token to each one of the terminal devices. The token is the limiter for coverage of the media access controller module set. It gives each terminal device, at different times, the option to send a signalling message using a token.

Although shown in the figures and frequently referred to as a house or other location, it is to be understood throughout this application that the customer premises may constitute any type of premises such as a school, an office, an apartment building, an office building, etc. The CPE may also be comprised of a plurality of premises, separated by a small distance such as a housing subdivision, a university campus, etc. The CO may be a telephone company central office containing other equipment and providing other services. Although shown and referred to as a central office, the CO may comprise any serving node which is geographically distant from the CPE.

While the foregoing has described what are considered to be preferred embodiments of the invention, it is understood that various modifications may be made therein and that the invention may be implemented in various forms and embodiments, and that it may be applied in numerous applications, only some of which have been described herein. It is intended by the following claims to claim all such modifications and variations which fall within the true scope of the invention.

* * * * *

Current U.S. Class:	398/70; 398/71; 398/72
Intern'l Class:	H04B 010/00
Field of Search:	359/115,118,124,125,164,165,167,152