Dynamic load balancing under partial service conditions

Information

  • Patent Grant
  • 10027588
  • Patent Number
    10,027,588
  • Date Filed
    Wednesday, October 28, 2015
    8 years ago
  • Date Issued
    Tuesday, July 17, 2018
    5 years ago
Abstract
Methods and apparatuses for controlling load balancing of modems may include receiving an indication of impairment of a channel, placing modems using the impaired channel into a partial service mode and removing the impaired channel from a list of channels available for load balancing, monitoring the quality of the impaired channel, and, responsive to detection of acceptable quality on the impaired channel, removing the modems from partial service mode and adding the channel back to the list of channels available for load balancing. Another method includes, but is not limited to, receiving an indication of a partial service mode for a modem, incrementing a partial service mode counter for a channel associated with the partial service mode indication, computing a partial service mode metric based in part upon the partial service mode counter, and responsive to a determination that the partial service mode metric meets a threshold, determining an allocation of modems to channels.
Description
FIELD

The field of this disclosure relates to load balancing of a set of modems in a network and, more particularly, relates to deferring load balancing under certain partial service conditions.


BACKGROUND

A typical cable network comprises a head end, which communicates with a cable modem termination system (CMTS) containing multiple transmitters and receivers. Each transmitter and receiver generally communicates with multiple cable modems (CMs) via a coaxial cable or hybrid fiber-coaxial network. The bandwidth between a CM and the CMTS is divided into a set of channels. For each node in the network, a receiver at the CMTS is generally allocated to each upstream channel that is in use, and a transmitter is allocated to each downstream channel.


Channel bonding in cable modems entered mainstream use with the introduction of the DOCSIS 3.0 specification. One or more bonded channels may be allocated to one or more of the CMs in communication with a CMTS. At times, the bonded channel(s) may become unsuitable for communication. To operate under these conditions, CMs may enter a partial service mode and discontinue use of the problematic channel, while continuing to operate using the remaining channels. The CM may continue ranging on the problematic channel, however, waiting for the partial service condition to be resolved such that it may again pass data on the channel.


A load balancing manager, detecting that there is little or no traffic on the channel impacted by a partial service condition, may consider the channel to be underutilized and actually assign additional modems to that channel. CMs newly moved by the load balancer to a bonding group that includes the channel impacted by the partial service condition may then also enter partial service mode because of the problem. Thus, the throughput of the CMs would be reduced, not increased, by the balancing operation. As the newly added modems also cease to pass traffic, the load balancer may be led to add more modems still to the impacted channel. Furthermore, modems entering partial service mode may shift more traffic to their other channels, potentially overloading those channels and starving modems for which only a single channel is available. When the CMs come out of partial service mode, the formerly impacted channel may quickly become overloaded, requiring further channel reallocations.


Thus, channel allocations using current load balancing techniques may not only be inefficient, but may also increase the chances of disruption of data and voice traffic.





BRIEF DESCRIPTION OF THE DRAWINGS


FIG. 1 is an exemplary schematic representation of a bi-directional cable system.



FIG. 2 illustrates an exemplary upstream module of a cable modem termination system.



FIG. 3 is a flowchart of a method for controlling the load balancing of a set of modems, wherein the method accounts for an impaired channel.



FIG. 4 is a flowchart of a method for controlling the load balancing of a set of modems by balancing modem counts, wherein the method accounts for partial service mode modems.



FIG. 5 is a flowchart of a method for controlling the load balancing of a set of modems by balancing utilization, wherein the method accounts for modems in partial service mode.





DETAILED DESCRIPTION

Disclosed herein are improved load balancing techniques, particularly in the presence of devices operating in a partial service mode. For example, the disclosed techniques may account for modems in partial service mode in the determination of whether to perform load balancing. Improving load balancing to account for partial service mode modems may reduce the number of problems that result from unnecessary or counterproductive modem moves, such as reduction of throughput, de-registration, and loss of data or voice traffic.


In embodiments, the method comprises, but is not limited to: receiving an indication of impairment of a channel, placing modems using the impaired channel into a partial service mode and removing the channel from a list of channels available for load balancing, monitoring the quality of the impaired channel, and, responsive to detection of acceptable quality on the impaired channel, removing the modems from partial service mode and adding the channel to a list of channels available for load balancing. By accounting for the impaired channel, the method prevents load balancing assignment of additional modems to the impaired channel, thereby avoiding unnecessary moves of modems and reducing the associated risk of data loss or disruption of transmissions that would occur under prior art load balancing methods.


In embodiments, the method comprises, but is not limited to: receiving an indication of a partial service mode for a first modem of a plurality of modems, incrementing a partial service mode counter for a first channel of the plurality of channels associated with the partial service mode indication responsive to the indication of the partial service mode, computing a partial service mode metric based in part upon the partial service mode counter, and, responsive to a determination that the partial service mode metric meets a threshold, determining an allocation of the plurality of modems to the plurality of channels, including the first channel. By accounting for the modems in partial service mode, the method avoids generating new load balancing channel allocations that would be counter-productive once those modems leave partial service mode, thereby avoiding unnecessary moves of modems and reducing the associated risk of data loss or disruption of transmissions that would occur under prior art load balancing methods.


In embodiments, the improved load balancing is performed by an apparatus comprising, but not limited to: a plurality of receivers for receiving data transmitted from a plurality of modems or a plurality of transmitters for transmitting data to a plurality of modems, a memory for storing a list of channels available for load balancing, and a processor for performing: receiving an indication of impairment of a channel, placing a plurality of modems using the impaired channel into a partial service mode and removing the channel from the list of channels available for load balancing, monitoring the quality of the impaired channel, and responsive to detection of acceptable quality on the impaired channel, removing the plurality of modems from a partial service mode and adding the channel to the list of channels available for load balancing.



FIG. 1 is an exemplary schematic representation of a bi-directional cable system. The cable modem termination system (CMTS) 120 communicates Internet Protocol (IP) traffic to and from a wide area network (WAN) 110, and to and from subscriber cable modems (CMs) 140 over the cable network 130. Cable network 130 may commonly be a coaxial network or hybrid fiber-coaxial (HFC) network. Cable modems 140 are connected directly or via wired or wireless local area networks (LANs) to customer premises equipment (CPE) 150.


Those of skill in the art will appreciate that “CMTS” generally refers to any suitable modem termination system, that the architecture illustrated is exemplary, and that any type of cabling may be used, such as coaxial wires, optical fibers, twisted pairs, and wireless connections. Upstream and downstream communications may be handled by the same or different processors, transceivers, and cabling. In embodiments, the CMTS comprises a Motorola BSR 64000 I-CMTS Chassis housing multiple Motorola RX 48 Decoupled Upstream Modules and Motorola TX 32 Decoupled Downstream Modules. In embodiments, the receiver or transmitter may be part of a combined video and data services platform. In embodiments, the CMs comprise Motorola SURFboard SB6120, SB6121, or SBG6580 modems.


The bandwidth between a CM 140 and the CMTS 120, such as those shown in FIG. 1, is divided into a set of channels. For each node in the network, a receiver 270 at the CMTS 120 is generally allocated to each upstream channel that is in use, and a transmitter is allocated to each downstream channel. The bandwidth available over each channel may be limited. Furthermore, the number of channels is generally much smaller than the number of connected CMs 140. Thus, multiple CMs 140 may share each channel. Sharing of a channel may be accomplished, for instance, using variants of time division multiple access (TDMA) or code division multiple access (CDMA) channel access methods.


In some cases, the full the bandwidth of a single channel may be insufficient to provide a desired level of service. Additionally, jitter and latency on a channel can greatly impact quality of service, particularly for voice and video services. Channel bonding may provide throughput and reduce transmission delays through the concurrent use of multiple channels. Channel bonding in cable modems entered mainstream use with the introduction of the DOCSIS 3.0 specification. A bonded CM 140 may send and receive data traffic on multiple channels in parallel, to and from the CMTS 120, allowing higher-speed operation. Furthermore, packet delays may be reduced since individual data packets may be segmented for transmission over the multiple channels and reassembled at the destination, reducing the transmission time for the packet.


At times, one or more of the bonded channels allocated to one or more of the CMs 140 in communication with a CMTS 120 may become unsuitable for communication. The communication problem may be caused by a localized hardware or software issue, affecting a subset of CMs 140 using the channel, or by general channel impairment, affecting all CMs 140 on the channel.


To operate under these conditions, CMs 140 may enter a “partial service mode.” In partial service mode, a CM 140 discontinues use of the problematic channel and continues operating using the remaining channels. The CM 140 may continue ranging on the problematic channel, however, waiting for the partial service condition to be resolved such that it may again pass data on the channel.


Utilization of certain channels can become higher than desired due to either the number of CMs 140 on the channel or the volume of data being passed by those CMs 140. Thus, cable operators may employ load balancing techniques to equalize traffic, generally balancing modem count or utilization across the available channels.


Load balancing is a feature of the CMTS 120 that controls dynamic changes to the set of downstream and upstream channels used by a CM 140. A Load Balancing Group (LBG) is a set of upstream and downstream channels over which a CMTS performs load balancing for a set of CMs. The CMTS generally attempts to balance load among all of the channels of each LBG. In cases where a single channel, upstream or downstream, is associated with more than one LBG, the CMTS might have to consider all LBGs for which such overlaps exist in its load balancing algorithm. In DOCSIS 3.0 environments, the CMTS may perform load balancing by using Dynamic Channel Change (DCC) or Dynamic Bonding Change (DBC) messages, by changing the Receive Channel Set (RCS) of the CM, by moving one or more service flows to different downstream channels within the current RCS of the CM, by changing the Transmit Channel Set (TCS) of the CM, or by moving one or more service flows to different upstream channels within the current TCS of the CM. The DOCSIS 3.0 specification currently does not place requirements on the specific algorithms used by the CMTS for load balancing, nor does it make a statement regarding the definition of “balanced” load. CMTS vendors are free to develop appropriate algorithms in order to meet market and deployment needs.



FIG. 2 illustrates an exemplary upstream module 200 of a CMTS 120 of FIG. 1. The upstream module 200 may perform various methods to control load balancing, which are adapted to account for partial service mode conditions. Processing unit 210 may include a microprocessor 220 for receiving information, such as instructions and data, from a ROM 240 or RAM 260. Upstream module 200 may also comprise a network module 230 for communications with other CMTS elements, such as corresponding downstream modules. Receivers 270 allow reception of data from CMs, including user data and control data, which may include indications of partial service mode conditions. The receiver portion of upstream module 200 may comprise multiple connectors, RF switches, tuners, and RF receivers. For example, the Motorola RX48 BSR 64000 Decoupled Upstream Module, which may be adapted for use with the disclosed techniques, provides eight physical RF ports, each with six upstream QAM receivers. In a corresponding exemplary downstream module, such as the Motorola TX32 BSR 64000 Decoupled Downstream Module, element 270 would be replaced with a transmission section. In other implementations, receivers and transmitters may be combined on the same card.


Processing unit 210 may be connected, directly or over a network, to a display 280, which may display status information. An input keypad 290 may also be connected, directly or indirectly, to processing unit 210 and may allow an operator to provide instructions, processing requests and/or data to upstream module 200. ROM 240 and/or RAM 260 may also carry instructions for microprocessor 220 to perform the processes illustrated in any or all of the following figures. Furthermore, any of the processes described below may be contained on a computer readable medium, which may be read by microprocessor 220. A computer readable medium may be any tangible storage medium capable of carrying instructions to be performed by a microprocessor, including a CD disc, DVD disc, magnetic or optical disc, tape, silicon based removable or non-removable memory. While microprocessor 220 is shown as being associated with an exemplary upstream module, it is understood that a common microprocessor may be shared amongst multiple upstream modules or combinations of upstream and downstream modules in CMTS 120.


As described above, performing load balancing when a significant number of modems are in a partial service mode can be counterproductive. Disclosed are embodiments for controlling the load balancing of modems, which account for modems in partial service mode.


Conditions causing entry into a partial service mode may be detected either by the CMTS 120 or by individual modems 140. In the first case, the load balancing manager may work in conjunction with a spectrum manager process, or other CMTS process, to monitor channel quality. If the spectrum manager detects that a particular channel is impaired, all of the CMs 140 using that channel may be put into partial service mode by the CMTS 120. However, this may cause the data utilization on that channel to be reduced, leading a normal load balancing algorithm to add CMs to that channel in an attempt to balance loads.


To address this problem, an improved load balancing method is desirable. FIG. 3 is a flowchart of such an improved method 300 for controlling the load balancing of modems, which accounts for a channel being impaired. The method may be performed within a CMTS 120, or particularly within an upstream module 200, a downstream module, or a combined upstream/downstream unit. Program code for the method may be stored on ROM 240 and/or RAM 260, and may be executed by microprocessor 220.


At 310, the CMTS 120 receives an indication of an impairment of one of the communication channels used for communications with CMs 140. The indication may be received from a process running on microprocessor 220, such as a spectrum management process, when a channel is determined to be impaired. The spectrum management process may generate the indication of impairment based upon information received from receivers or transmitters 270. The detection of impairment may be based upon metrics such as signal strength, signal-to-noise ratio, detected errors, or other indications generated by hardware or software.


At 320, the CMTS 120 places all modems associated with the impaired channel into a partial service mode. The CMTS 120 may accomplish this via the microprocessor 220 sending control messages to each of the affected CMs 140 via transmitters or network interface 230. The control messages may be sent to the CMs 140 using the modulation used for regular data transmission, such as 8-QAM, 16-QAM, 32-QAM, 64-QAM, 128-QAM, or 256-QAM, or using a different, more robust modulation scheme such as quadrature phase-shift keying (QPSK). The message may be sent via the regular data transmission path or via a separate control channel.


The entry of CMs into partial service mode may lead to an increase in traffic on the remaining non-impaired channels allocated to those modems. For instance, in a DOCSIS 3.0-capable system where the DOCSIS 3.0 modems each use two bonded channels, an impairment of one channel, and subsequent entry of those modems into partial service mode, will cause those modems to shift more of their traffic to their other channel.


At 330, the impaired channel is removed from the list of channels available to the load balancing process. With the impaired channel removed, CMs will not be moved by the load balancer to the impaired channel or to a bonding group containing the impaired channel. Channel availability may be stored as an array, or in another suitable form, in RAM or other memory accessible to the load balancing manager process. The actions of 320 and 330 may take place in either order or in parallel.


At 340, the quality of the impaired channel is monitored. The monitoring may be performed by a spectrum manager or other process on the CMTS 120. Metrics such as signal strength, signal-to-noise ratio, or detected errors, may be used in the monitoring. Monitoring signals may be generated by hardware, software, or a combination of the two.


At 350, the results of the monitoring of 340 are assessed. If the channel quality is deemed to be acceptable, the algorithm proceeds to 360 and those modems placed in partial service mode due to the channel impairment are taken out of partial service mode. At 370, the channel removed from the load balancing list at 330 is marked as being available again for use in load balancing. The actions of 360 and 370 may take place in either order, or in parallel.


If, at 350, the channel quality is still unacceptable, the algorithm returns to 340 for further monitoring. Thus, load balancing involving the impaired channel is delayed until the impairment is resolved and counterproductive moves of modems to the impaired channel are prevented.


The method of FIG. 3 may be performed by an apparatus such as the upstream module 200 of FIG. 2. The receivers 270 may receive data transmitted from modems on a plurality of channels. A random access memory 260 may store a list of channels available for load balancing. Microprocessor 220 may receive an indication of impairment of at least one channel. Responsive to the indication of impairment of the channel, the CMTS 120 may place those modems using the impaired channel into a partial service mode and remove the impaired channel from the list of channels available for load balancing. The microprocessor 220 may then monitor the quality of the impaired channel, or monitor indicators of the quality received from another component, and, responsive to detection of acceptable quality on the impaired channel, remove the modems from partial service mode and add the channel back to the list of channels available for load balancing. Program code related to execution of the method may be stored in RAM 260, ROM 240, or other memory.


In the example of FIG. 3, channel impairment is detected by the CMTS 120, generally affecting all CMs 140 that are using the impaired channel. In other cases, however, modems themselves detect problems, which may or may not be affecting other CMs, which necessitate entry into a partial service mode. These problems may be caused by localized noise issues, ranging issues, faulty splitters, or other software or hardware failures or malfunctions within the customer premises or in the network. Different methods are needed to address these cases, which may depend upon the load balancing technique being used.


One particular technique for load balancing is the balancing of modem counts across channels. If an assumption is made that sets of modems will, on average, produce, or consume, similar amounts of traffic, equalizing the numbers of modems on each channel would result in balanced utilization. Load balancing in this scenario is performed by moving modems between channels to balance the modem counts. When the difference in the number of modems between channels exceeds a threshold, modems are moved to rebalance the system.


As described above, modems in partial service mode may not appear to be using a channel, while in fact, they are in the process of ranging and could resume transmission on that channel in a short time. Performing load balancing using prior art algorithms that fail to account for the modems in partial service mode may therefore be counterproductive and lead to a worse load imbalance in the near future.


To address this problem, the load balancing manager may store a counter for each channel to keep track of the number of modems currently in partial service. When load balancing is triggered, the number of modems considered to be using each channel during the modem count comparison will be adjusted to include the modems in partial service. By including the partial service modems in the calculation, the load balancing manager will be less likely to initiate unnecessary modem moves that would likely need to be reversed when modems come out of partial service.



FIG. 4 is a flowchart of such a method 400 for controlling the load balancing of CMs, which accounts for partial service mode CMs. The method may be performed within a CMTS 120, or particularly within an upstream module 200, a downstream module, or a combined upstream/downstream unit. Program code for the method may be stored on ROM 240 and/or RAM 260, and may be executed by microprocessor 220.


At 410, the CMTS 120 receives an indication of a partial service mode for one of the modems. The indication may be received directly from the modem 140, as described above, due to detection by the modem of any inability to communicate effectively on one or more of its channels. The control messages may be sent to the CMTS 120 by the CM 140 using the modulation used for regular data transmission, such as 8-QAM, 16-QAM, 32-QAM, 64-QAM, 128-QAM, or 256-QAM, or using a different, more robust modulation scheme such as quadrature phase-shift keying (QPSK). The message may be received via the regular data transmission path or via a separate control channel. In a DOCSIS 3.0 system, the indication may take the form of a DOCSIS REG-ACK message if a channel is not acquired during registration, a DBC-RSP message if the channel is not acquired during Dynamic Bonding Change, or a CM-STATUS message if a channel becomes unusable during normal operation.


At 420, the CMTS increments a partial service mode counter for each affected channel in response to the indication of the partial service mode. The partial service mode counter may be stored in RAM 260 and may be adjusted by microprocessor 220. The partial service mode counter may be implemented as a separate count of partial service mode CMs that can be added to the active modem count for a total, or a total count of the combined number of active and partial service mode CMs.


At 430, a load balancing trigger is detected. In the case of a system with scheduled periodic load balancing, the trigger may be generated by a timer. The timer may be a process executed by processor 220. In other systems, the trigger may be related to utilization, modem count imbalances, or other indicators.


At 440, one or more differences are computed between the number of CMs associated with a channel and counts related to one or more other channels. Various difference calculations may be used. For instance, the difference may be computed as a difference between the modem count of the relevant channel and one other channel or an average modem count for all channels related to an LBG, or perhaps a difference between the modem count of the relevant channel and the minimum of the modem counts for the other channels in the LBG. In one implementation, the maximum and minimum modem counts for the relevant channels are determined and the difference between them used in the assessment.


At 450, if the difference in the counts of CMs, where the counts include the partial service mode CMs, exceeds a threshold, the algorithm proceeds to 460 and load balancing is performed. It should be noted that 450 may be equivalently implemented as a check of whether the difference meets or exceeds the threshold with the corresponding selection of a higher threshold. Furthermore, the decision may be equivalently considered in the opposite sense, wherein load balancing is deferred if the difference is below a threshold.


It should be noted that the load balancing itself may involve any set of two or more channels in the LBG. For instance, in one implementation, load balancing, once triggered, may potentially impact allocation of modems across all channels in the LBG. In other cases, the load balancing may simply move modems from the most heavily loaded channel to the least heavily loaded channel.


If the difference in modem count does not exceed a threshold when the partial service modems are considered, it could be counterproductive to rebalance. The partial service mode modems are expected to eventually recover and begin to pass data evenly on all of their upstream channels, causing the channels to go back to the balanced condition. Thus, if the difference threshold is not exceeded, rebalancing is delayed and the algorithm then returns to 430 to await the next trigger.


The method of FIG. 4, described above, is applicable particularly to systems where load balancing is based on modem counts. The method serves to prevent load balancing when the difference in counts of active modems is based upon the inactivity of modems in partial service mode, which are expected to return to active service in the near term, rather than upon a difference in the expected longer-term modem counts.


Balancing modem counts, however, will not always produce balanced utilization of channels. Therefore, in some systems, load balancing is performed not based upon differences in modem counts, but upon differences in data utilization. As above, modems in partial service mode may also impact whether load balancing should be performed or deferred in a system based upon utilization.


In such systems, modems in partial service mode may lead to an appearance that a channel is underutilized, when in fact the modems are likely to resume communications on the channel in the near term. Moving modems to such a channel could lead to counterproductive load balancing, unnecessary modem moves, and disruption of communications. Thus, a modified method, described below, is desirable for such systems.



FIG. 5 is a flowchart of such a method 500 wherein load balancing is deferred for a channel until the percentage of modems in partial service mode is below a threshold, or, equivalently, the percentage of active modems is above a threshold. The method may be performed within a CMTS 120, or particularly within an upstream module 200, a downstream module, or a combined upstream/downstream unit. Program code for the method may be stored on ROM 240 and/or RAM 260, and may be executed by microprocessor 220.


At 510, the CMTS receives an indication of a partial service mode for one of the modems. The indication may be received directly from the modem, as described above, due to detection by the modem of any inability to communicate effectively on one or more of its channels. In a DOCSIS 3.0 system, the indication may take the form of a DOCSIS REG-ACK message if a channel is not acquired during registration, a DBC-RSP message if the channel is not acquired during Dynamic Bonding Change, or a CM-STATUS message if a channel becomes unusable during normal operation.


At 520, the CMTS 120 increments a partial service mode counter for the relevant channels in response to the indication of the partial service mode from the CM. The partial service mode counter may be stored in RAM 260 and may be adjusted by microprocessor 220. The partial service mode counter may be implemented as a separate count of partial service mode CMs that can be added to the active modem count for a total, or a total count of the combined number of active and partial service mode CMs.


At 530, a load balancing trigger is detected. In the case of a system with scheduled periodic load balancing, the trigger may be received from a timer process. The timer may be a process executed by processor 220. In other systems, the trigger may be related to utilization or other indicators.


At 540, the percentage of CMs in partial service mode for the channel of interest is calculated. If a large percentage of CMs are in partial service mode, the utilization metrics for the channel are likely to be misleading. CMs that will likely be communicating on the channel in the near future are not presently communicating. Thus, moving CMs to the channel based on the temporary underutilization could be counterproductive.


At 550, if the percentage of CMs in partial service mode is below a chosen threshold, the algorithm proceeds to 560 and load balancing is conducted with the relevant channel included. For instance, if a 10% threshold is used, load balancing would be allowed for a channel where fewer than ten of a total of one hundred CMs were in partial service mode. The threshold percentage is an implementation decision and could be based on average utilization statistics of modems.


In other cases, the partial service mode metric may exceed the threshold. In these cases, it could be counterproductive to rebalance since the partial service CMs should eventually recover and resume communications on all of their channels, causing the channels to go back to a more balanced condition. Thus, in those cases, the algorithm returns to 530 without rebalancing the channel and awaits the next trigger. For instance, if a 10% threshold is used, load balancing would be deferred for a channel where one or more of a total of eight modems were in partial service mode.


In certain embodiments, the load balancing algorithm may equivalently compute the percentage of CMs that are not in partial service mode and require that the percentage exceed the threshold for rebalancing to occur. In some embodiments, load balancing will also be contingent upon a minimum utilization level regardless of the differences in utilization, or other conditions.


The methods of FIG. 4 and FIG. 5 may also be performed by the apparatus of FIG. 2. The receivers 270 may receive an indication of a partial service mode for a modem. Responsive to the indication of a partial service mode for the modem, processor 220 may increment a first partial service mode counter for channel associated with the partial service mode indication, compute a first partial service mode metric based in part upon the partial service mode counter, and responsive to a determination that the partial service mode metric meets a threshold, determine an allocation of modems to channels, including the first channel associated with the partial service mode indication. The indication of a partial service mode for a modem may alternatively be received via network interface 230. Program code related to execution of the method may be stored in RAM 260, ROM 240, or other memory.


While the principles of the disclosed techniques have been described above in connection with specific apparatus, it is to be clearly understood that this description is made only by way of example and not as a limitation on the scope. For instance, the methods described may be applied to load balancing of transmission or reception devices, and may apply to networks other than cable networks. Aspects of each of the described methods may be combined to provide improved load balancing in systems with partial service mode.


The foregoing detailed description has set forth various embodiments of the devices and or processes via the use of block diagrams, flowcharts, and examples. Insofar as such block diagrams, flowcharts, and examples contain one or more functions and or operations, it will be understood by those within the art that each function or operation within such block diagrams, flowcharts, or examples can be implemented individually or collectively by a wide range of hardware, software, firmware, or virtually any combination thereof. In one embodiment, several portions of the subject matter described herein may be implemented via Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs), digital signal processors (DSPs), or other integrated formats. However, those skilled in the art will recognize that some aspects of the embodiments disclosed herein, in whole or in part, can be equivalently implemented in integrated circuits or as one or more computer programs running on one or more computers. In addition, those skilled in the art will appreciate that the mechanisms of the subject matter described herein are capable of being distributed as a program product in a variety of forms, and that an illustrative embodiment of the subject matter described herein applies regardless of the particular type of computer readable medium used for distribution.


It is to be understood that the depicted architectures are merely exemplary and that in fact many other architectures can be implemented which achieve the same functionality. In a conceptual sense any arrangement of components to achieve the same functionality is effectively associated such that the desired functionality is achieved.


While particular aspects of the present subject matter described herein have been shown and described, it will be apparent to those skilled in the art that based upon the teachings herein, changes and modifications may be made without departing from the subject matter described herein and its broader aspects, and therefore, the appended claims are to encompass within their scope all such changes and modifications as are within the true spirit and scope of the subject matter described herein.

Claims
  • 1. A method of managing communication with a plurality of modems on a plurality of channels, said method comprising: receiving an indication of a partial service mode for a first modem of said plurality of modems;responsive to said indication of said partial service mode for said first modem, incrementing a first partial service mode counter for a first channel of said plurality of channels associated with said partial service mode indication;computing a first partial service mode metric based in part upon said first partial service mode counter; andresponsive to a determination that said first partial service mode metric meets a threshold, determining an allocation of said plurality of modems to said plurality of channels, including said first channel,wherein said first partial service mode metric meeting said threshold comprises a difference between a first modem count, comprising a sum of a count of active modems associated with said first channel and said first partial service mode counter for said first channel, and a second modem count, exceeding a predetermined threshold, andwherein said second modem count is one of: a count of active and partial service mode modems associated with a second channel, an average of a plurality of counts active and partial service mode modems associated with two or more of said plurality of channels, or a minimum count of active and partial service mode modems associated with said plurality of channels.
  • 2. The method of claim 1 wherein computing said first partial service mode metric is performed responsive to detecting a trigger condition for load balancing of said plurality of modems.
  • 3. The method of claim 1 wherein said first modem communicates on said first channel of said plurality of channels using one of a time division multiple access channel access method or a code division multiple access method.
  • 4. The method of claim 1 wherein said first partial service mode metric meeting said threshold comprises a difference between the first partial service mode metric associated with said first channel and a second partial service mode metric associated with a second channel being above a predetermined threshold.
  • 5. The method of claim 1 wherein said second modem count comprises modems in the partial service mode.
  • 6. The method of claim 1 wherein at least one modem of said plurality of modems communicates on at least two channels.
  • 7. The method of claim 1 wherein said determining an allocation of said plurality of modems to said plurality of channels comprises controlling a balance of the plurality of modems among said plurality of channels based on one of: the number of modems in communication on each channel or the data utilization of each channel.
  • 8. The method of claim 1 wherein said partial service mode metric is a percentage of modems associated with said first channel that are in a partial service mode.
  • 9. The method of claim 8 wherein said partial service mode metric meeting said threshold comprises said percentage of modems associated with said first channel that are in a partial service mode being below a predetermined percentage or said percentage of modems associated with said first channel that are not in a partial service mode being above a predetermined percentage.
  • 10. The method of claim 9 wherein said predetermined percentage is between 5% and 15%.
  • 11. The method of claim 1 wherein receiving an indication of the partial service for the first modem comprises receiving a message from a cable modem termination system (CMTS).
  • 12. The method of claim 1 wherein the first modem is at least one of said plurality of modems that communicates on at least two channels.
  • 13. The method of claim 1 wherein the first modem is at least one of said plurality of modems that communicates on said first channel of said plurality of channels using one of a time division multiple access channel access method or a code division multiple access method.
  • 14. An apparatus for communicating with a plurality of modems on a plurality of channels comprising: a plurality of receivers for receiving data transmitted from said plurality of modems on said plurality of channels or a plurality of transmitters for transmitting data to said plurality of modems on said plurality of channels;a memory for storing a list of channels available for load balancing; anda processor for:receiving an indication of a partial service mode for a first modem of said plurality of modems;responsive to said indication of said partial service mode for said first modem, incrementing a first partial service mode counter for a first channel of said plurality of channels associated with said partial service mode indication;computing a first partial service mode metric based in part upon said first partial service mode counter; andresponsive to a determination that said first partial service mode metric meets a threshold, determining an allocation of said plurality of modems to said plurality of channels, including said first channel,wherein said first partial service mode metric meeting said threshold comprises a difference between a first modem count, comprising a sum of a count of active modems associated with said first channel and said first partial service mode counter for said first channel, and a second modem count, exceeding a predetermined threshold, andwherein said second modem count is one of: a count of active and partial service mode modems associated with a second channel, an average of a plurality of counts active and partial service mode modems associated with two or more of said plurality of channels, or a minimum count of active and partial service mode modems associated with said plurality of channels.
CROSS-REFERENCE TO RELATED APPLICATION

This application is a divisional of U.S. patent application Ser. No. 13/728,220 filed on Dec. 27, 2012, entitled “Improved Dynamic Load Balancing Under Partial Service Conditions,” now issued as U.S. Pat. No. 9,203,639, the contents of which is incorporated herein by reference in its entirety.

US Referenced Citations (273)
Number Name Date Kind
3838221 Schmidt et al. Sep 1974 A
4245342 Entenman Jan 1981 A
4385392 Angell et al. May 1983 A
4811360 Potter Mar 1989 A
4999787 McNally et al. Mar 1991 A
5228060 Uchiyama Jul 1993 A
5251324 McMullan Oct 1993 A
5271060 Moran et al. Dec 1993 A
5278977 Spencer et al. Jan 1994 A
5347539 Sridhar et al. Sep 1994 A
5390339 Bruckert et al. Feb 1995 A
5463661 Moran et al. Oct 1995 A
5532865 Utsumi et al. Jul 1996 A
5557603 Barlett et al. Sep 1996 A
5606725 Hart Feb 1997 A
5631846 Szurkowski May 1997 A
5694437 Yang et al. Dec 1997 A
5732104 Brown et al. Mar 1998 A
5790523 Ritchie et al. Aug 1998 A
5862451 Grau et al. Jan 1999 A
5867539 Koslov Feb 1999 A
5870429 Moran et al. Feb 1999 A
5886749 Williams et al. Mar 1999 A
5939887 Schmidt et al. Aug 1999 A
5943604 Chen et al. Aug 1999 A
6032019 Chen et al. Feb 2000 A
6061393 Tsui et al. May 2000 A
6108351 Hardy et al. Aug 2000 A
6154503 Strolle Nov 2000 A
6229792 Anderson et al. May 2001 B1
6230326 Unger et al. May 2001 B1
6233274 Tsui et al. May 2001 B1
6240553 Son et al. May 2001 B1
6272150 Hrastar et al. Aug 2001 B1
6278730 Tsui et al. Aug 2001 B1
6308286 Richmond et al. Oct 2001 B1
6310909 Jones Oct 2001 B1
6321384 Eldering Nov 2001 B1
6330221 Gomez Dec 2001 B1
6334219 Hill et al. Dec 2001 B1
6377552 Moran et al. Apr 2002 B1
6385773 Schwartzman et al. May 2002 B1
6389068 Smith et al. May 2002 B1
6434583 Dapper et al. Aug 2002 B1
6445734 Chen et al. Sep 2002 B1
6456597 Bare Sep 2002 B1
6459703 Grimwood et al. Oct 2002 B1
6477197 Unger Nov 2002 B1
6480469 Moore et al. Nov 2002 B1
6483033 Simoes et al. Nov 2002 B1
6498663 Farhan et al. Dec 2002 B1
6512616 Nishihara Jan 2003 B1
6526260 Hick et al. Feb 2003 B1
6546557 Ovadia Apr 2003 B1
6556239 Al Araji et al. Apr 2003 B1
6556562 Bhagavath et al. Apr 2003 B1
6556660 Li et al. Apr 2003 B1
6559756 Al Araji et al. May 2003 B2
6563868 Zhang et al. May 2003 B1
6570394 Williams May 2003 B1
6570913 Chen May 2003 B1
6574797 Naegeli et al. Jun 2003 B1
6588016 Chen et al. Jul 2003 B1
6606351 Dapper et al. Aug 2003 B1
6611795 Cooper Aug 2003 B2
6646677 Noro et al. Nov 2003 B2
6662135 Burns et al. Dec 2003 B1
6662368 Cloonan et al. Dec 2003 B1
6671334 Kuntz et al. Dec 2003 B1
6687632 Rittman Feb 2004 B1
6690655 Miner et al. Feb 2004 B1
6700875 Schroeder et al. Mar 2004 B1
6700927 Esliger et al. Mar 2004 B1
6711134 Wichelman et al. Mar 2004 B1
6741947 Wichelman et al. May 2004 B1
6748551 Furudate et al. Jun 2004 B2
6757253 Cooper et al. Jun 2004 B1
6772388 Cooper et al. Aug 2004 B2
6772437 Cooper et al. Aug 2004 B1
6816463 Cooper et al. Nov 2004 B2
6839829 Daruwalla et al. Jan 2005 B1
6853932 Wichelman et al. Feb 2005 B1
6877166 Roeck et al. Apr 2005 B1
6895043 Naegeli et al. May 2005 B1
6895594 Simoes et al. May 2005 B1
6906526 Hart et al. Jun 2005 B2
6928475 Schenkel et al. Aug 2005 B2
6944881 Vogel Sep 2005 B1
6961314 Quigley et al. Nov 2005 B1
6961370 Chappell Nov 2005 B2
6967994 Boer et al. Nov 2005 B2
6973141 Isaksen et al. Dec 2005 B1
6985437 Vogel Jan 2006 B1
6999408 Gomez Feb 2006 B1
7002899 Azenkot et al. Feb 2006 B2
7010002 Chow et al. Mar 2006 B2
7017176 Lee et al. Mar 2006 B1
7032159 Lusky et al. Apr 2006 B2
7039939 Millet et al. May 2006 B1
7050419 Azenkot et al. May 2006 B2
7054554 McNamara et al. May 2006 B1
7058007 Daruwalla et al. Jun 2006 B1
7072365 Ansley Jul 2006 B1
7079457 Wakabayashi et al. Jul 2006 B2
7099412 Coffey Aug 2006 B2
7099580 Bulbul Aug 2006 B1
7139283 Quigley et al. Nov 2006 B2
7142609 Terreault et al. Nov 2006 B2
7152025 Lusky et al. Dec 2006 B2
7158542 Zeng et al. Jan 2007 B1
7164694 Nodoushani et al. Jan 2007 B1
7177324 Choudhunj et al. Feb 2007 B1
7197067 Lusky et al. Mar 2007 B2
7222255 Claessens et al. May 2007 B1
7227863 Leung et al. Jun 2007 B1
7242862 Saunders et al. Jul 2007 B2
7246368 Millet et al. Jul 2007 B1
7263123 Yousef Aug 2007 B2
7274735 Lusky et al. Sep 2007 B2
7315573 Lusky et al. Jan 2008 B2
7315967 Azenko et al. Jan 2008 B2
7400677 Jones Jul 2008 B2
7421276 Steer et al. Sep 2008 B2
7451472 Williams Nov 2008 B2
7492703 Lusky et al. Feb 2009 B2
7554902 Kim et al. Jun 2009 B2
7573884 Klimker et al. Aug 2009 B2
7573935 Min et al. Aug 2009 B2
7616654 Moran et al. Nov 2009 B2
7650112 Utsumi et al. Jan 2010 B2
7672310 Cooper et al. Mar 2010 B2
7684315 Beser Mar 2010 B1
7684341 Howald Mar 2010 B2
7716712 Booth et al. May 2010 B2
7739359 Millet et al. Jun 2010 B1
7742697 Cooper et al. Jun 2010 B2
7742771 Thibeault Jun 2010 B2
7778314 Wajcer et al. Aug 2010 B2
7787557 Kim et al. Aug 2010 B2
7792183 Massey et al. Sep 2010 B2
7856049 Currivan et al. Dec 2010 B2
7876697 Thompson et al. Jan 2011 B2
7953144 Allen et al. May 2011 B2
7970010 Denney et al. Jun 2011 B2
8000254 Thompson et al. Aug 2011 B2
8037541 Montague et al. Oct 2011 B2
8040915 Cummings Oct 2011 B2
8059546 Pai et al. Nov 2011 B2
8081674 Thompson et al. Dec 2011 B2
8116360 Thibeault Feb 2012 B2
8265559 Cooper et al. Sep 2012 B2
8284828 Cooper et al. Oct 2012 B2
8310926 Bernstein Nov 2012 B1
8345557 Thibeault et al. Jan 2013 B2
8788647 Nolle et al. Jul 2014 B1
20010055319 Quigley et al. Dec 2001 A1
20020038461 White et al. Mar 2002 A1
20020044531 Cooper et al. Apr 2002 A1
20020091970 Furudate et al. Jul 2002 A1
20020116493 Schenkel et al. Aug 2002 A1
20020154620 Azenkot et al. Oct 2002 A1
20020168131 Walter et al. Nov 2002 A1
20020181395 Foster et al. Dec 2002 A1
20030028898 Howald Feb 2003 A1
20030043732 Walton et al. Mar 2003 A1
20030067883 Azenkot et al. Apr 2003 A1
20030101463 Greene et al. May 2003 A1
20030108052 Inoue et al. Jun 2003 A1
20030120819 Abramson et al. Jun 2003 A1
20030138250 Glynn Jul 2003 A1
20030149991 Reidhead et al. Aug 2003 A1
20030158940 Leigh Aug 2003 A1
20030179768 Lusky et al. Sep 2003 A1
20030179770 Reznic et al. Sep 2003 A1
20030179821 Lusky et al. Sep 2003 A1
20030181185 Lusky et al. Sep 2003 A1
20030182664 Lusky et al. Sep 2003 A1
20030185176 Lusky et al. Oct 2003 A1
20030188254 Lusky et al. Oct 2003 A1
20030200317 Zeitak et al. Oct 2003 A1
20030212999 Cai Nov 2003 A1
20040015765 Cooper et al. Jan 2004 A1
20040042385 Kim et al. Mar 2004 A1
20040047284 Eidson Mar 2004 A1
20040052356 McKinzie et al. Mar 2004 A1
20040062548 Obeda et al. Apr 2004 A1
20040073937 Williams Apr 2004 A1
20040096216 Ito May 2004 A1
20040109661 Bierman et al. Jun 2004 A1
20040139473 Greene Jul 2004 A1
20040163129 Chapman et al. Aug 2004 A1
20040181811 Rakib Sep 2004 A1
20040202202 Kolze et al. Oct 2004 A1
20040208513 Peddanarappagari et al. Oct 2004 A1
20040233234 Chaudhry et al. Nov 2004 A1
20040233926 Cummings Nov 2004 A1
20040248520 Miyoshi Dec 2004 A1
20040261119 Williams et al. Dec 2004 A1
20050010958 Rakib et al. Jan 2005 A1
20050025145 Rakib et al. Feb 2005 A1
20050034159 Ophir et al. Feb 2005 A1
20050039103 Azenko et al. Feb 2005 A1
20050058082 Moran et al. Mar 2005 A1
20050064890 Johan et al. Mar 2005 A1
20050097617 Currivan et al. May 2005 A1
20050108763 Baran et al. May 2005 A1
20050122996 Azenkot et al. Jun 2005 A1
20050163088 Yamano et al. Jul 2005 A1
20050175080 Bouillett Aug 2005 A1
20050183130 Sadja et al. Aug 2005 A1
20050198688 Fong Sep 2005 A1
20050226161 Jaworski Oct 2005 A1
20050281200 Terreault Dec 2005 A1
20060013147 Terpstra et al. Jan 2006 A1
20060121946 Walton et al. Jun 2006 A1
20060250967 Miller et al. Nov 2006 A1
20060262722 Chapman et al. Nov 2006 A1
20070002752 Thibeault et al. Jan 2007 A1
20070058542 Thibeault Mar 2007 A1
20070076592 Thibeault et al. Apr 2007 A1
20070076789 Thibeault Apr 2007 A1
20070076790 Thibeault et al. Apr 2007 A1
20070086328 Kao et al. Apr 2007 A1
20070094691 Gazdzinski Apr 2007 A1
20070097907 Cummings May 2007 A1
20070133672 Lee et al. Jun 2007 A1
20070143654 Joyce et al. Jun 2007 A1
20070147489 Sun et al. Jun 2007 A1
20070177526 Siripunkaw et al. Aug 2007 A1
20070184835 Bitran et al. Aug 2007 A1
20070189770 Sucharczuk et al. Aug 2007 A1
20070206600 Klimker et al. Sep 2007 A1
20070206625 Maeda Sep 2007 A1
20070211618 Cooper et al. Sep 2007 A1
20070223512 Cooper et al. Sep 2007 A1
20070223920 Moore et al. Sep 2007 A1
20070245177 Cooper et al. Oct 2007 A1
20080056713 Cooper et al. Mar 2008 A1
20080062888 Lusky et al. Mar 2008 A1
20080075157 Allen et al. Mar 2008 A1
20080101210 Thompson et al. May 2008 A1
20080140823 Thompson et al. Jun 2008 A1
20080193137 Thompson et al. Aug 2008 A1
20080200129 Cooper et al. Aug 2008 A1
20080242339 Anderson Oct 2008 A1
20080250508 Montague et al. Oct 2008 A1
20080274700 Li Nov 2008 A1
20080291840 Cooper et al. Nov 2008 A1
20090031384 Brooks et al. Jan 2009 A1
20090103557 Hong et al. Apr 2009 A1
20090103669 Kolze et al. Apr 2009 A1
20090249421 Liu et al. Oct 2009 A1
20100014425 Cheng et al. Jan 2010 A1
20100083356 Steckley et al. Apr 2010 A1
20100095360 Pavlovski et al. Apr 2010 A1
20100154017 An Jun 2010 A1
20100157824 Thompson et al. Jun 2010 A1
20100158093 Thompson et al. Jun 2010 A1
20100223650 Millet et al. Sep 2010 A1
20110026577 Primo et al. Feb 2011 A1
20110030019 Ulm et al. Feb 2011 A1
20110069745 Thompson et al. Mar 2011 A1
20110080868 Krishnaswamy et al. Apr 2011 A1
20110110415 Cooper et al. May 2011 A1
20110194418 Wolcott et al. Aug 2011 A1
20110194597 Wolcott et al. Aug 2011 A1
20110197071 Wolcott et al. Aug 2011 A1
20110206103 Beser Aug 2011 A1
20110243214 Wolcott et al. Oct 2011 A1
20120054312 Salinger Mar 2012 A1
20120084416 Thibeault et al. Apr 2012 A1
20120147751 Ulm Jun 2012 A1
20130088961 Ramachandran Apr 2013 A1
Foreign Referenced Citations (19)
Number Date Country
69631420 Dec 2004 DE
1235402 Aug 2002 EP
1341335 Sep 2003 EP
55132161 Oct 1980 JP
04208707 Jul 1992 JP
6120896 Apr 1994 JP
6177840 Jun 1994 JP
09008738 Jan 1997 JP
9162816 Jun 1997 JP
10247893 Sep 1998 JP
11230857 Aug 1999 JP
2001044956 Feb 2001 JP
2003530761 Oct 2003 JP
2004172783 Jun 2004 JP
2004343678 Dec 2004 JP
0192901 Jun 2001 WO
0233974 Apr 2002 WO
2004062124 Jul 2004 WO
2009146426 Dec 2009 WO
Non-Patent Literature Citations (37)
Entry
Volpe, B., and Miller, W., “Cable-Tec Expo 2011: Advanced Troubleshooting in a DOCSIS© 3.0 Plant,” Nov. 14-17, 2011, pp. 17.
Wolcott, L., “Modem Signal Usage and Fault Isolation,” U.S. Appl. No. 61/301,835, filed Feb. 5, 2010.
Zhao, F., et al., “Techniques for minimizing error propagation in decision feedback detectors for recording channels,” IEEE Transactions on Magnetics, vol. 37, No. 1, Jan. 2001, pp. 12.
B. Volpe, “DOCSIS 3.0 Partial Service”, The Volpe Firm, Dec. 7, 2011, retrieved from the Internet at <http://volpefirm.com/docsis-3-0-partial-service/? on Mar. 24, 2015.
“A Simple Algorithm for Fault Localization Using Naming Convention and Micro-reflection Signature,” Invention Disclosure 60193, Cable Television Laboratories, Inc., Jun. 2008, p. 2.
“Data-Over-Cable Service Interface Specifications DOCSIS 3.0: MAC and Upper Layer Protocols Interface,” CM-SP-MULPIv3.0-I16-110623, Cable Television Laboratories, Inc., Jun. 2011, section 8, pp. 242-266.
“Data-Over-Cable Service Interface Specifications DOCSIS® 3.0—MAC and Upper Layer Protocols Interface Specification,” CM-SP-MULPIv3.0-I17-111117, Cable Television Laboratories, Inc., Nov. 17, 2011, pp. 770.
“DOCSIS Best Practices and Guidelines; Proactive Network Maintenance Using Pre-Equalization,” CM-GL-PNMP-V01-100415, Cable Television Laboratories, Inc., pp. 123.
“Pre-Equalization Based Pro-active Network Maintenance Process Model for CMs Transmitting on Multiple Upstream Channels,” Invention Disclosure 60203, Cable Television Laboratories, Inc., May 2009, pp. 2.
“Pre-Equalization based pro-active network maintenance process model,” Invention Disclosure 60177, Cable Television Laboratories, Inc., Jun. 2008, pp. 2.
“Proactive Network Maintenance Using Pre-Equalization,” DOCSIS Best Practices and Guidelines, Cable Television Laboratories Inc., CM-GL-PNMP-V02-110623, Jun. 23, 2011, pp. 133.
“Radio Frequency Interface Specification,” Cable Television Laboratories, Inc., Data-Over-Cable Service Interface Specifications DOCSIS 2.0, CM-SP-RFIv2.0-I06-040804, pp. 524.
Campos, L. A., et al., “Pre-equalization based Pro-active Network Maintenance Methodology,” Cable Television Laboratories, Inc., (presentation), 2012, pp. 32.
Howald, R. L., et al., “Customized Broadband—Analysis Techniques For Blended Multiplexes,” pp. 12.
Howald, R., “Access Networks Solutions: Introduction to S-CDMA,” Presentation to Society of Cable Telecommunications Engineers (SCTE) South Florida Chapter, Motorola, Inc., 2009, pp. 15.
Howald, R., “Upstream Snapshots & Indicators (2009),” Regional Samples, Presentation to Society of Cable Telecommunications Engineers (SCTE) South Florida Chapter, Jan. 2010, pp. 22.
Howald, R., et al., “Characterizing and Aligning the HFC Return Path for Successful DOCSIS 3.0 Rollouts,” Society of Cable Telecommunications Engineers (SCTE) Cable Tee Expo, Oct. 2009, pp. 66.
Howald, R., et al., “Docsis 3.0 Upstream: Readiness & Qualification,” pp. 17.
Howald, R., et al., “The Grown-Up Potential of a Teenage Phy,” pp. 65.
Howald, R.,“DOCSIS 3.0 Upstream: Technology, RF Variables & Case Studies,” Access Networks Solutions, 2009, presentation to Society of Cable Telecommunications Engineers (SCTE) South Florida Chapter, Jan. 2010, pp. 23.
Hranac, R., “Linear Distortions, Part 1,” Communication Technology, Jul. 1, 2005, accessed at www.cable360.net/print/ct/operations/testing/15131.html, pp. 6.
Liu, X., and Bernstein, A., “Variable Bit Rate Video Services in DOCSIS 3.0 Networks,” NCTA Technical Papers, 2008, pp. 12.
Motorola., “White Paper: Expanding Bandwidth Using Advanced Spectrum Management,” Sep. 25, 2003, pp. 12.
Newton's Telecom Dictionary, Sep. 1995, Flatiron Publishing, 9th Edition, pp. 216 and 1023, definitions of “carrier to noise ratio” and “signal to noise ratio”.
Patrick, M., and Joyce, G., “Delivering Economical IP Video over DOCSIS by Bypassing the M-CMTS with DIBA,” SCTE 2007 Emerging Technologies, Topic Subject: Service Velocity & Next Generation Architectures: How Do We Get There?, 2007, pp. 17.
Popper, A., et al, “An Advanced Receiver with Interference Cancellation for Broadband Cable Networks,” Juniper Networks, International Zurich Seminar on Broadband Communications Access 2002, pp. 23-1-23-6.
Popper, A., et al, “Ingress Noise Cancellation for the Upstream Channel in Broadband Cable Access Systems,” Juniper Networks, IEEE International Conference on Communications 2002, vol. 3, pp. 1808-1812.
Qureshi, S. U. H., “Adaptive Equalization,” IEEE, vol. 73, No. 9, Sep. 1985, pp. 1349-1387.
Ramakrishnan, S., “Scaling the DOCSIS Network for IPTV,” Cisco Systems, Inc., SCTE Conference on Emerging Technologies and the NCTA Cable Show, 2009, pp. 19.
Shelke, Y. R., “Knowledge Based Topology Discovery and Geo-localization,” Thesis, 2010, pp. 173.
Thompson, R., et al., “256-QAM for Upstream HFC,” Spring Technical Forum Proceedings, 2010, pp. 142-152.
Thompson, R., et al., “256-QAM for Upstream HFD Part Two,” SCTE Cable Tec Expo 2011, Technical Paper, pp. 22.
Thompson, R., et al., “Multiple Access Made Easy,” SCTE Cable Tec Expo 2011, Technical Paper, pp. 23.
Thompson, R., et al., “Optimizing Upstream Throughput Using Equalization Coefficient Analysis,” National Cable & Telecommunications Association (NCTA) Technical Papers, Apr. 2009, pp. 35.
Thompson, R., et al., “Practical Considerations for Migrating the Network Toward All-Digital,” Society of Cable Telecommunications Engineers (SCTE) Cable-Tec Expo, Oct. 2009, pp. 22.
Thompson, R., et al., “64-QAM, 6.4MHz Upstream Deployment Challenges,” SCTE Canadian Summit, Toronto, Canada, Technical Paper, Mar. 2011, pp. 25.
Ulm, J., et al., “Flexible Channel Bonding for IP Video Delivery Disclosure,” pp. 8.
Related Publications (1)
Number Date Country
20160050154 A1 Feb 2016 US
Divisions (1)
Number Date Country
Parent 13728220 Dec 2012 US
Child 14924920 US