WO2001043440A1

WO2001043440A1 - Method and system for reducing data errors in digital communications

Info

Publication number: WO2001043440A1
Application number: PCT/US2000/033711
Authority: WO
Inventors: Phillip K. Freyman
Original assignee: Motorola Inc.
Priority date: 1999-12-13
Filing date: 2000-12-13
Publication date: 2001-06-14
Also published as: EP1159828A1; AU2091801A; EP1159828A4

Abstract

A system (10) for reducing data errors in a digital data stream includes a receiver (122), an error detection circuit (123), and a processor (100). One or more communication channels transmit redundant data slots to the receiver (122). In response to the redundant data slots, the error detection circuit (123) generates data valid signals corresponding to the communication channels. The data valid signals are set based on the contents of the redundant data slots. In response to the data valid signals from the error detection circuit (123), the processor (100) generates a non-redundant data stream including data slots selected from either of the incoming communication channels. In this manner, corrupted data slots can be removed, resulting in an output non-redundant data stream having fewer data errors.

Description

METHOD AND SYSTEM FOR REDUCING DATA ERRORS IN DIGITAL COMMUNICATIONS

TECHNICAL FIELD OF THE INVENTION

The invention generally relates to digital communication systems, and in particular, to a system and method for improving the reliability of data transmission over a communication system utilizing a digital protocol.

BACKGROUND OF THE INVENTION

Digital communication systems are ubiquitous, acting as the main data thoroughfares in virtually all major communication networks. One goal of digital communication systems is to provide reliable transport of information. However, in any communication system, situations arise where communication links become impaired, resulting in the loss of data. Such impairments can be caused by any of the number of the events, such as a break in a transmission line, a nearby lightning strike causing electromagnetic interference channel fade, an equipment failure, or the like.

In some types of networks, such as cable television networks, an impairment such as a line break results in a temporary loss of service to subscribers. Depending on how the robust the system is, the loss of service may be momentary, or may last for hours. Other impairments, such as lightning strikes, channel fade or marginally operational transceivers do not always result in a loss of service. Instead, they often degrade the quality of service by introducing data errors into data streams. In cable networks, these data error can result in noise that degrades the audio /video quality received on a subscriber' s premises.

In microwave systems it is common practice to use antenna " space diversity" to overcome impairment problems in microwave transmissions in wireless environments. Space diversity physically places two antennas more than 1/4 wavelength apart. This then utilizes the physical fact that a channel fade in one physical path will be different than the channel fade on second physical path. In general, the two antennas both drive a single receiver and the selection of the antennas is based on radio frequency (RF) signal strength or RF signal-to-noise ratio. In either case, the recovered information is based solely on the RF or analog information from one receiver and is not constructed from redundant digital information.

In some wireless broadcast services, multiple receivers are located in physically different locations in a geographical area to ensure that at least one RF path can access at least one receiver. This arrangement protects the system from RF coverage holes, fading, and the like. The received outputs from each site are communicated, either by another radio /microwave link or landline, to a central point and submitted to a device typically refered to as a " comparator" . The comparator processes the multiple input signals for signal quality. Signal quality is usually based on RF level or signal-to-noise ratio. Newer digital systems often use bit error rates as a quality metric. Based on the definition of " best" channel, the comparator selects the best signal source and sends it to the receiving user. Accordingly, data is received over a single channel that is susceptible to data loss caused by temporary impairments.

Accordingly, a need exist for an improvement to digital communication systems that increases the reliability and quality of data transmissions by reducing the effects of temporary impairments to the system.

BRIEF DESCRIPTION OF THE DRAWING:

The accompanying drawings provide an understanding of the invention as described in an embodiment to illustrate the invention and serve to explain the principles of the invention. FIG. 1 is a block diagram illustrating a communication system in accordance with the embodiment of the present invention;

FIG. 2 is a diagram illustrating a data frame format useable with the communication system shown in FIG. 1; FIG. 3 is a block diagram of the cable access unit (CAU) included in the system of FIG. 1;

FIG. 4 is a detailed block diagram of an exemplary line card includable in the CAU of FIG. 3;

FIG. 5 is a conceptual diagram illustrating reconstruction of downstream data traffic performed by a CAU on redundant input data streams that are aligned in time;

FIG. 6 is a conceptual diagram illustrating reconstruction of downstream data traffic performed by a CAU on redundant input data streams that are offset in time; FIG. 7 is a conceptual diagram illustrating the reconstruction of downstream data traffic performed by a CAU on a single input data stream having redundant data frames;

FIG. 8 is a conceptual diagram illustrating upstream data transmission using redundant transmitters; and FIG. 9 is a flow chart diagram illustrating a method of transmitting upstream data using redundant data paths.

DETAILED DESCRIPTION OF AN EMBODIMENT OF THE INVENTION

The following description of the disclosed communication system is exemplary and intended to provide further explanation of the invention as claimed. The present invention relates to an improved communication system that utilizes redundant data reception to reduce data errors caused by impairments to the communication system. According to one aspect of the invention, the system can include one or more communication channels for transmitting singular or redundant data slots, a receiver for receiving the data slots, an error detection circuit (error detector) coupled to the receiver, and a processor coupled to the error detection circuit. The error detection circuit can be configured to generate data valid signals corresponding to each communication channel. In response to each data slot, the error detector sets one of the data valid signals to indicate whether the respective data slot contains valid data. The data valid signals are received by the processor, which in response generates a non-redundant data stream by selecting ones of the incoming data slots as a function of the data valid signals. In this manner, a corrupted incoming data stream can be reconstructed into an output data stream having fewer or no data errors. This technique of protecting real-time data is particularly advantageous in communication systems carrying streaming data, such as real-time audio or video, because it does not rely on complex coding schemes that increase latency and processing delays.

Turning now to the drawings, and the particular FIG. 1, a communication system 10 in accordance with an embodiment of the invention is illustrated. The communication system 10 includes a headend 12, a fiber node 16, and a cable access unit (CAU) 20. A fiber optic network 17 permits communication between the headend 12 and the fiber node 16, while a coaxial cable network 23 permits communications between the fiber node 16 and the CAU 20. Networks, such as the one illustrated in FIG. 1, comprising a combination of optic fiber and coaxial cable are commonly referred to as hybrid fiber/coax (HFC) networks.

Although an HFC network is used to illustrate the operation of claimed invention, it will be apparent that the invention is not limited to HFC networks and can be readily applied in systems using other communication mediums, such as purely optic fiber networks, wireless networks, or any combination thereof.

The communication system 10 is capable of carrying almost any type of information including telephony and audio /visual data traffic. In the example shown, the system 10 carries cable telephony and cable television (CATV) transmissions between service providers and a subscriber premises 22. For broadband cable telephony service applications, the system 10 generates multiple RF carriers. The carriers designate communication channels for transporting voice traffic between the headend 12 and the CAU over the HFC network 25. The CAU 20 can be located at or near the subscriber premises 22. The CAU 20 can be configured to provide conventional POTS (plain old telephone system) services to a subscriber telephone 26. As a result of the bidirectional HFC bandwidth capacity, the system 10 provides economical and efficient voice transmission when compared to a conventional twisted wire pair.

To transmit information over the network 25, the system 10 can rely on industry standard channelization and bandwidth allocation for CATV.

In this scheme, CATV signals are all located between 50-500 MHZ in North

America, while upstream traffic (information transferred from the subscriber premises 22 to the headend 12) is allocated to 5-40 MHZ in North America and 5-65 MHZ internationally. Downstream traffic (information flowing from the headend 12 to the subscriber premises 22) is allocated to 50-750 MHZ in North America and 85-750MHz internationally. In North America, non-CATV downstream traffic can be allocated between 500-750 MHZ. Multiple RF carriers representing different communications channels can be located within the bandwidths allocated to upstream and downstream data traffic. Accordingly, the network 25 can simultaneously carry a plurality of data streams in either the upstream or downstream direction. The fiber optic network 17 can include two or more fibers for carrying data between the headend 12 and the fiber node 16. Each fiber typically carries data in one direction, either upstream or downstream. The fiber network

17 can be a conventional optical network, such as one used in existing CATV cable networks.

The fiber node 16 can be a commercially available electronic device for interfacing the fiber optic network 17 to the coaxial cable network 23. Fiber node 16 includes circuitry for converting downstream lightwave signals into RF signals which are then output to the coaxial cable 21. The fiber 16 node also includes circuitry for converting upstream RF signals from the coaxial network 23 to lightwave signals output to the fiber optic network 17.

The tap 18 permits individual CAUs to be connected to the coaxial network 23. Generally, the tap 18 includes a conventional co-axial transformer (not shown) that permits the RF signals pass between the coaxial cable 21 and the CAU 23. Although not shown, more than one tap and CAU can be connected to the coaxial cable 21.

The headend 12 can be an operations center for a CATV system. It can act as a receiving station for television transmissions, by way of satellite or other means, and can provide an interconnection point between a cable operator' s local network and a national or international telecommunications networks, as illustrated by the connection to a local digital switch (LDS) 14. Generally, the headend 12 serves a predefined geographic area and can distribute multiple signals throughout the area over the HFC network 25 to the subscribers. The HFC network 25 has the capability of carrying standard and interactive video, voice telephony and high-speed data services from the headend 12. Television, telephone, and data signals can be received at the headend 12, then collected, processed, amplified, combined and distributed to subscribers through the two-way HFC network 25. The headend 12 includes a cable control frame (CCF) 28, a combiner /splitter 32, and a fiber-optic transceiver 23. A plurality of cable control units (CCU) 30 are included in the CCF 28 and coupled to the LDS 14 of a telephone service provider. The CCF 28 and CCUs 30 are commercially available from Motorola, Inc., while the combiner /splitter 32 and fiber optic transceivers can be implemented using conventional, commercially-available components commonly used in existing CATV cable networks.

The CCF 28 act as a gateway between the LDS 14 and the HFC network 25. The CCF 28 can include a Bellcore GR-303 or ETSI V5.2 rrunking interface, permitting it to operate with almost any central office digital telephone switch used as the LDS 14. The CCF 28 can support multiple downstream RF channels with the channels connecting to individual combiners (not shown) included in the combiner /splitter 32. The combiners can group the downstream telephony signals with CATV signals for transmission over the HFC network 25. The combiner outputs are provided to optical transmitters (not shown) in the fiber optic transceiver 33 via coaxial cables. The optical transmitters then convert downstream RF signals to lightwave signals for transmission over the HFC network 25.

Upstream RF signals arrive at the headend 12 via the fiber optic network 17 as lightwave signals. At the headend 12, fiber optic receivers (not shown) in the fiber optic transceiver 33 convert the lightwave signals into electrical RF signals. Coaxial cables can route the RF signals directly to the CCF

28.

Each CCU 30 contains span line processing, call processing, RF modulation circuitry, a serving area matrix, and alarms for monitoring telephone calls. Calls originating from the LDS 14 on analog or digital subscriber lines can be connected to the intended recipient subscriber by routing DS0/E0 subscriber lines to the CCF 28. The calls are dynamically assigned to the DS1/E1 span line servicing the intended recipient through the HFC network 25 by one of the CCUs 30. The CCF 28 routes the voice signal of the caller by assigning the call to a data slot defined by a time division multiplexed (TDM) protocol. The TDM data slots are modulated on an RF carrier, then combined with the CATV RF signals by the combiner/splitter 32 before entering the HFC network 25. Each RF carrier for telephony can be 600 kHz wide, thus carrying multiple simultaneous calls. The combined RF signals are converted into lightwave signals by the transceiver 33, and then reconverted from lightwave signals to RF signals at the fiber node 16. Each 600 kHz channel can carry eight data slots to support eight separate 64 kbps channels for voice. Accordingly, standard 64 kbps telephony signals can be carried through the HFC network 25 to the subscriber premises 22. Each 64 Kpbs signal, embedded in each RF carrier can be decoded into voice audio using a standard telephony CODEC at the CAU 20.

Each 600 kHz channel represents either upstream or downstream traffic. The TDM protocol can be used to carry downstream traffic, and a time division multiple access (TDMA) protocol can be used to carry upstream traffic over the HFC network 25. Data can be modulated onto the RF channels using

\4 shift differential quadrature phase shift keying (DQPSK). Using this modulation scheme, a channel symbol of a rate of 384 Ksps, corresponding to a channel bit rate of 768 Kbps, can be achieved. With the TDM/TDMA protocol, each channel transfers 400 data frames per second, each data frame consisting of eight time (data) slots or the equivalent of eight voice channels. The format of the TDM/TDMA data frames is shown in FIG. 2.

FIG. 2 conceptually illustrates a portion 60 of the data stream carried by one of the channels. The data stream is represented as a continuous sequence of data frames, each having a predetermined number of data slots. The CCF 28 and CCUs 30 include hardware and software for implementing the TDM/TDMA protocol. They also control, the allocation of data slots and data transfers over the HFC network 25 by way of the TDM/TDMA protocol. The illustrated portion 60 of the data stream includes two data frame 62-64, each having eight data slots 0-7. Each data slot can be dynamically allocated to a call. Also, each data slot comprises a header 66, a data payload 68, and an error checking code 70, such as a cyclic redundancy code (CRC). The error checking code 70 is preferably computed as a function of the data payload 68. Alternatively, the error checking code 70 can be computed based on the contents of both the header 66 and the data payload 68. The error code can be computed for each data slot by the CCU 30 assigned to control the particular data transmission session in which the slot is included. The error code can be generated using any appropriate error coding scheme, such as a checksum function, a CRC function, or the like. Although the present invention is not limited to any particular technique of error detection, the CCUs 30 preferably use a CRC polynomial of 201177 octal to compute the error code. The CRC polynomial is represented by the following equation: g (x) = l + X⁷+ X¹⁰ + X¹²+ X¹³ + X¹ + χl⁵ + χl6 + χl⁷ _{+ χ}18 _{+ χ}19 _{+ χ}20 (1)

The TDM/TDMA protocol permits each non-CATV channel to carry telephony voice or data between the headend 12 and the CAU 20. To accomplish this, the header 66 can include a destination identification (ID) which can be determined by one of the CCUs 30 at the time of a call setup. The destination ID indicates a particular CAU line card for receiving the data slot, as will be discussed below in greater detail. The data payload 68 can include any number of data bits. However, using the TDM/TDMA protocol and RF modulation scheme described herein, the data payload 68 includes 160 bits of data.

The present invention is not limited to any particular transmission protocol or modulation scheme. Other transmission schemes, channel spacing, capacities, and the like can be used without departing from the scope of the claimed invention. Further, a wireless link can be equivalently substituted for the coaxial network 23. This can be accomplished by placing antennas at the CAU 20 and fiber node 16 and adapting the RF transceivers at both locations to support wireless communications. The RF modulation scheme disclosed herein can support the TDM/TDMA protocol over such wireless link. FIG. 3 is a detailed block diagram illustrating the CAU 20 shown in

FIG. 1. The CAU 20 can include one or more line cards 100 coupled to a first RF interface 106 and a second RF interface 108. Redundant bus A 102 connects the line cards 100 to the first interface 106 and a first power supply 130 while redundant bus B 104 connects the line cards 100 to the second RF interface 108 and a second power supply 132. Each line card 100 can provide a predetermined service to a customer premises. For example, a line card can be configured to provide POTS telephony service at the subscriber premises 22. Alternatively, other line cards can provide modem services for data transfer, ISDN, RS232, nx 64kbps data, H328 streaming video, coin phone, VoIP, or the like.

The first and second interfaces 106-108 permit redundant data streams to be received by the line cards 100. Each line card can process the incoming redundant data streams to generate a non-redundant output data stream having fewer data errors. This can be accomplished as follows. The redundant incoming data streams can be received on separate

RF paths 110-112. These RF paths can be separate physical communication paths, such as redundant coaxial cables connecting the CAU 20 to the HFC network 25 as shown. In this arrangement, a single RF channel can be transmitted down two hardware paths. The single RF channel carries singular (non-redundant) data slots. The redundant data streams are created by transmitting the RF channel over the two paths.

Alternatively, a single physical link may be connected to both interfaces 106, 108 to provide a singular stream of data slots to the redundant interfaces. This architecture supports redundancy of the receiver hardware and permits the CAU 20 to continue functioning in the event of receiver failure or receiver performance degradation. Receiver preformance degradation occurs when a receiver degrades over time and can not capture the incoming data stream without errors. In a further alternative embodiment, the " paths" can be two different channels modulated by different RF carriers onto a single coaxial cable. In this scheme, only one RF interface is needed in the CAU 20 to receive the incoming data streams. Using only one RF interface, the RF receiver included therewith is configured to respond to both carriers in order to generate the redundant data streams. This can be accomplished by including in the RF receiver an RF splitter and a pair of receivers, each receiving a respective channel from the splitter, and in response, generating a respective digitized data stream. In either the two interface or single RF interface scheme, the channels carry the same data stream. These data streams are transported using the TDM protocol described above.

In the two RF interface CAU 20 shown in FIG. 3, downstream RF signals to the CAU 20 are received at the first and second receivers 122, 126. The RF receivers 122, 126 can be conventional receivers configured to demodulate RF signals that are modulated according to the modulation parameters described above in connection with FIG. 1. The receivers 122, 126 output the digitized redundant data streams formatted into the data frames of the TDM protocol. The redundant TDM data streams are then distributed to the line cards over the respective buses 102, 104.

In addition to generating the TDM protocol data stream, the receivers 122, 126 can also generate data valid signals corresponding to each channel. The data valid signal from each receiver 122, 126 is also distributed to each of the line cards 100 by way of the respective buses 102, 104. The data valid signal from each of the receivers indicates the validity of the contents of each TDM data slot output by the receiver. Accordingly, if the contents of the data slot received by the receiver is corrupted or otherwise contains data errors, the data valid signal is set to alert the line cards 100 of this condition.

To determine whether data slots are corrupted, the receivers 122, 126 each include an error detection circuit 123, 125. Each circuit 123, 125 performs an error checking function on each incoming slot to generate an error code. This error code can then be compared to the error code 70 included in the incoming data slot. If the error codes do not match, the data valid signal is set to indicate that the data slot contains corrupted data. The error detectors 123, 125 can perform a CRC check on the payload data 68 included in each data slot in accordance with equation 1. The output of the error detector can then be compared to the CRC error code included in the data slot to generate the data valid signal. Each of the line cards 100 can be assigned a predetermined destination ID (TID_x). By comparing their assigned destination IDs to the destination ID included in the header 66 of each incoming data slot, the line cards respond only to those data slots directed to them. The destination IDs are dynamically assigned to the line cards 100 by the controllers 114-116. During call set up, each controller 114-116 can negotiate with a CCU 30 at the headend 12 to assign data slots to respective ones with the line cards 100. The controllers 114-116 use the TDM/TDMA protocol to communicate with the CCU 30. To accomplish communication with the CCU 30, each controller 114-116 has a designated destination ID (CAU ID) that is entered into the CCU at initial installation of the CAU 20.

Each line card 100 can generate a non-redundant data stream including data slots selected from either of the redundant incoming data streams. The data slots are selected based on the data valid signals from the receivers 122, 126. If a data slot is corrupted on one RF channel, it can be replaced by a redundant data slot received from the other RF channel. In this manner, corrupted data slots are filtered out of the data stream. FIG. 4 is a detailed block diagram illustrating an example architecture of one of the line cards shown in FIG. 3. The line card 150 includes a microprocessor 152, a memory 154, a CODEC 158, first transceiver 160 and a second transceiver 162. A standard microprocessor bus 156 is used to connect the components included in the line card 150. Incoming data slots and data valid signals from the redundant buses 102-104 are received by the transceivers 160-162. The transceivers 160-162 can be implemented using commercially available digital bus transceivers. The memory 154 can store a software program executable by the microprocessor 152 for directing the microprocessor 152 to process each of the incoming data slots and their corresponding data valid signals. The processor 152 can first detect the header 66 of an incoming data slot and then compare its assigned destination ID to that contained in the header 66. If the microprocessor 152 determines that the data slot is directed to the line card 150, it continues to process the incoming data slot. Otherwise it ignores the incoming data slot.

Provided that the incoming redundant data slots are directed to the line card 150, the microprocessor 152 then checks the data valid signals to determine whether either of the redundant data slots received from buses A and B 102, 104 are corrupted. If the data valid signal corresponding to a particular data slot indicates that the slot is corrupted, the microprocessor 152 discards that data slot, and instead responds to the redundant data slot provided by the other bus. If both redundant data slots are corrupted, then the microprocessor 152 defaults to one channel and selects that data slot as its output. However, if an uncorrupted data slot is available, the microprocessor 152 transfers this data slot to the CODEC 58 by way of the bus 156. In this manner, the microprocessor 152 reconstructs a non-redundant output stream from the incoming redundant slots. The non-redundant stream can be generated by the microprocessor 152 using any one of the software processes described herein in connection with FIGS. 5-7. The CODEC 158 can be a commercially available standard CODEC for performing digital /analog conversion and pulse code modulation (PCM) decoding and encoding for standard telephony. The CODEC 158 outputs an analog audio signal in response to non-redundant digital data it receives under the control of the microprocessor 152. The power supplies 130, 132 independently supply power to a respective interface 106, 108 as well as the line cards 100. Inclusion of redundant power supplies permits the CAU 20 to continue operation should one of the interfaces 106, 108 or power supplies 130, 132 fail. The power supplies 130, 132 can be implemented using commercially available standard power supplies suitable for communication electronics.

Figure 5 is a conceptual diagram illustrating reconstruction of the data stream, such reconstruction being performed the processor 152 on one of the line cards 100. A pair of redundant data streams 200-202 are received by a line card over the redundant RF channels. The data slots received at the first RF interface 106 are designated A_0-A_7 in the data stream 200. The data slots received by the second RF interface 108 are likewise indicated by B_0-B_7 shown in the data stream 202. For each incoming data slot, error detection circuit 123, 125 that receives the slot generates a respective data valid signal. As illustrated, the error detection circuit generates a data valid signal shortly after receiving a corresponding data slot. Accordingly, the data valid signal for the RX_A Downstream data 200 indicates that data slots A_2 and A_3 are corrupted. The data valid signal for RX_B Downstream data 202 indicates that data slot B_4 is corrupted. In response to the incoming redundant data streams and the data valid signals, the line cards reconstruct non-redundant output datastream 204. In the example shown, the default incoming data stream is received by the first interface 106. Thus, the line card substitutes data slots B_2 and B_3 for the corrupted slots A_2 and A_3, respectively. In FIG. 5, the incoming redundant data streams are time aligned, meaning the corresponding redundant data slots arrive at the interfaces 106, 108 at approximately the same time. Figure 6 illustrates a conceptual diagram illustrating the reconstruction of redundant data streams that are not time aligned. The default redundant data stream 220 is received at the first interface 106. The second redundant data stream 222 is delayed in time relative to the first data stream. The second data stream is received at the second interface 112. Although the redundant data stream can be offset any number of data slots, the example shown in FIG. 6 illustrates a delay offset of one-half a data frame, or four data slots. Using offset redundant data streams is advantageous in that data corrupted by impairments lasting less than one half data frame can be recovered. Temporary impairments of this nature are generally caused by power surges and lightning strikes. Delaying the redundant data stream RX_B 222 by one-half data frame under the TDM protocol described above generally does not cause human perceivable delay in real-time audio applications, such as telephony voice. The data stream RX_A 220 at the first interface 106A acts as the default input data stream. Thus, to reconstruct the non-redundant output data stream 224, the line card replaces corrupted data slots in the RX_A data stream 220 with valid data slot from the RX_B data stream 222 received at the second interface 108. FIG. 7 illustrates a conceptual diagram of transferring redundant data streams using a single RF channel. In this arrangement, redundant half- data frames (four data slots) are repeated in sequence on a single RF carrier to form a redundant data stream 240. As shown, a first half-data frame 244 (slots AO.O - A0.3) is transmitted followed by a second half-data frame from 246 (slots BO.O - B0.3) which is in turn followed by a third half-data frame 248 and so on. The B data frames carry information that is redundant to the A data frames. The sequence of A and B data frames is repeated throughout the data stream. In the example shown, the A data stream is the default data stream. Accordingly, the line card replaces corrupted data slots in the A data streams with non-corrupted data slots from B data stream to reconstruct a non-redundant output data stream 242. In the example shown, the data valid signal indicates that data slot A0.1 is corrupted. Accordingly, in the output data stream, data slot A0.1 is replaced by its corresponding redundant data slot B0.1. The redundancy scheme shown in FIG. 7 has advantage in that it requires only one RF receiver at the CAU. It is also advantageous in that it permits redundant data to be offset in time. Accordingly, for transmission disruptions lasting less than one-half data frame, corrupted data can be recovered. FIG. 8 illustrates a conceptual diagram illustrating upstream data transmission using the redundant transmitters 124, 128 included in the CAU 20. The transmitters 124, 128 can be implemented using off-the-shelf components configured to operate in accordance with the RF transmission parameters discussed above in connection with FIG. 1. To transmit data upstream to the headend 12, the CAU 20 relies on the TDMA protocol discussed in connection with FIG. 1. Access to the upstream data slots is controlled by one of the CCUs 30. To be allocated an upstream data slot, one of the controllers 114, 116 requests access by transmitting an access request to the CCU 30. Which one of the controllers 114, 116 that actually performs the request is determined by which one is activated at the time the access request is made. Generally, a predetermined one of the controllers 114, 116 acts as a default activate controller. If the default controller fails, the other controller takes over. An upstream data slot for an active line card is negotiated by the active controller included in the CAU 20 and one of the CCUs 30. After an upstream data slot is assigned to the line card, the line card can transmit upstream data using the RF interface associated with the active controller. When the line card is ready to transmit an upstream data slot, it generates a transmit enable signal over the active bus A 102 or bus B 104. The transmit enable signal alerts the active transmitter 118 and 120 that there is data from the line card ready to be transmitted over the HFC network 25. Upon receiving the data from the line card, the respective transmitter 118 or 120 outputs the data, modulated onto an RF carrier, on a respective one of the interfaces 118, 120. The RF signal representing the transmitting data is then received by the CCF 28 and processed by a CCU 30. In response to each received data slot, the CCU 30 generates an acknowledgment (ACK) signal, which is transmitted downstream to the line card to indicate that the upstream slot was successfully received at the CCU 30. In response to the ACK signal, the processor 152 in the line card can generate a subsequent transmit enable signal for transferring the next upstream data slot.

Each transmitted data slot is paired in time with ACK signal generated by the CCU 30. If the transmitting line card does not receive an ACK within a predetermined time period following the transmission of the data slot, the line card processor 152 does not generate a subsequent transmit enable signal. The predetermined time period can be the duration of one data slot. Upon failing to receive a timely ACK signal, the line card switches to a redundant transmitter and begins transmitting using it. FIG. 8 illustrates a transmitter switch over. A line card generates two redundant uplink data streams 260-262 which are respectively transferred over bus A 102 and bus B 104 to the transmitters 124, 128. In the example shown, the first transmitter TX_A 118 is enabled to transmit the first two data slots L_0-L_1. The Transmit_A enable signal indicates transmission over the first interface 106 for slots L_0-L_1. However, after transmission of data slot L_l, an ACK signal is not received from the CCU 30, and therefore, the line card begins transmission over the redundant interface 108. The line card accomplishes this by generating the Transmit_B_Enable signals to active the second transmitter 120. Data slots L_2 and L_3 are transmitted over the second interface 108. Following transmission of the data slot L_2, the line card does not receive an ACK signal from the CCU 30. In response, the line card switches back to the first transmitter 118 to transmit the remaining data slots L_4-L_7.

FIG. 9 is a flow chart diagram of a method 300 for transmitting upstream data using redundant data channels. In step 302, a data slot is transmitted upstream by a line card using a default transmitter selected from either the first interface 106 or the second interface 108. After transmitting the data slot, the line card waits for an acknowledgment (ACK) signal generated by one of the CCUs 30.

In step 306, a check is made to determined whether the ACK signal was received by the line card. If so, the line card continues using the default transmitter (308); if not, the line cards selects the redundant transmitter for subsequent transmission of upstream data slots (step 310).

It should be appreciated that a wide range of changes and modifications may be made to the embodiment of the invention as described herein. Thus, it is intended that the foregoing detailed description be regarded as an illustrative rather than limiting and that the following claims, including all equivalents are intended to define the scope of the invention.

What is claimed is:

Claims

1. A system, comprising: one or more communication channels for transmitting a plurality of redundant data slots; at least one receiver for receiving the redundant data slots by way of the communication channels; at least one error detection circuit, operatively coupled to the at least one receiver, for generating a plurality of data valid signals corresponding to the communication channels, the at least one error detection circuit setting one of the data valid signals to a predetermined value for each of the redundant data slots based on the contents thereof; a processor, operatively coupled to the at least one error detection circuit, for generating a non-redundant data stream including ones of the redundant data slots selected from the communication channels based on the data valid signals.

2. The system of claim 1, further comprising: a CODEC, operatively coupled to the processor, for converting the non-redundant data stream to an analog format.

3. The system of claim 1, wherein the data slots include data selected from the group consisting of streaming audio, streaming video and streaming data.

4. The system of claim 1, wherein each of the redundant data slots includes a data payload and a cyclic redundancy code (CRC).

5. The system of claim 4, wherein the controller includes means for setting the data valid signals based on the CRCs included in the redundant data slots.

6. The system of claim 1, wherein the redundant data slots are arranged into at least one data frame.

7. The system of claim 1, wherein the communication channels exist on separate physical communication paths.

8. The system of claim 1, wherein the communication channels utilize a wireless link.

9. The system of claim 1, wherein the communications channels are logical channels existing on the same physical communication path.

10. The system of claim 1, wherein at least one of the redundant data slots transferred on one of the communication channels is time delayed relative to a corresponding redundant data slot transferred on another of the communication channels.

11. A cable access unit, comprising: a first receiver for demodulating a first radio frequency (RF) input to generate a first sequence of data slots received over a first physical communication path; a first error detection circuit for generating a first data valid signal based on the first sequence of data slots, the first data valid signal indicating the validity of the contents of each data slot included in the first sequence of data slots; a second receiver for demodulating a second RF input to generate a second sequence of data slots received over a second physical communication path, the second sequence of data slots including data redundant to those included in the first sequence of data slots; a second error detection circuit for generating a second data valid signal based on the second sequence of data slots, the second data valid signal indicating the validity of the contents of each data slot included in the second sequence of data slots; and a line card capable of generating a non-redundant data stream including at least one data slot selected from the first sequence of data slots or second sequence of data slots based on the first data valid signal and the second data valid signal.

12. The cable access unit of claim 11, further comprising: a first bus coupling the first error detection circuit, the first receiver, and the line card.

13. The cable access unit of claim 11, further comprising: a second bus coupling the second error detection circuit, the second receiver, and the line card.

14. The cable access unit of claim 11, further comprising: a first power supply operatively coupled to the first receiver, the first error detection circuit, and the line card.

15. The cable access unit of claim 11, further comprising: a second power supply operatively coupled to the second receiver, the second detection circuit, and the line card.

16. The cable access unit of claim 11, further comprising: a first transmitter, operatively coupled to the line card, for modulating data slots for transmission as RF output over the first physical medium; and a second transmitter, operatively coupled to the line card, for modulating data slots for transmission as RF output over the second physical medium.

17. The cable access unit of claim 16, wherein the line card includes means for selecting the first transmitter or the second transmitter based on transmission reliability.

18. A system, comprising: means for transmitting a plurality of redundant data slots over one or more communication channels; means for receiving the redundant data slots from the transmitting means; means for generating a plurality of data valid signals and setting each of the data valid signals to a predetermined value for each of the redundant data slots based on the contents thereof; and means for generating a non-redundant data stream including ones of the redundant data slots selected from the communication channels based on the data valid signals.

19. The system of claim 18, wherein the means for transmitting includes means for generating a sequence of singular data slots and means for transmitting the singular data slots over a plurality of separate physical communication paths to produce the redundant data slots.

20. A method for correcting errors in streaming data transferred by a digital communication system, comprising: receiving a plurality of redundant data slots containing the streaming data over a plurality of communication channels; generating a plurality of data valid signals corresponding to the communication channels; setting each of the data valid signals for each of the redundant data slots based on the contents thereof; and correcting errors in the streaming data by generating a non- redundant data stream including valid ones of the redundant data slots selected from the communication channels based on the data valid signals.

21. The method of claim 20, wherein the digital communication system includes a hybrid fiber coax (HFC) network.

22. The method of claim 20, further comprising: generating a cyclic redundancy code (CRC) based on a data payload included in a one of the redundant data slots; comparing the CRC to a predetermined CRC included in the one of the redundant data slots; and setting one of the data valid signals based on the comparison of the CRC and the predetermined CRC.

23. A system, comprising: a cable control unit (CCU) for generating downstream data traffic including a plurality of redundant data slots; a fiber optic transmitter, operatively coupled to the CCU for generating lightwave signals in response to the downstream data traffic; a fiber node for generating an RF signal in response to the lightwave signal; a fiber optic network, operatively coupled to the fiber optic transmitter and the fiber node, for carrying the light wave signal; a cable access unit (CAU) for generating a non-redundant data stream in response to the RF signal, the cable access unit being capable of demodulating the RF signal to recover the redundant data slots and being capable of deriving the non-redundant data stream from the redundant data slots; and a cable network, operatively coupled to the fiber node and the

CAU, for carrying the RF signal.

24. The system of claim 23, further comprising a local digital switch for (LDS) communicating with the CCU.

25. The system of claim 23, further comprising: a combiner /splitter, operatively coupled to the fiber optic transmitter, for combining the downstream data traffic with CATU signal.