Hunter interrogates the network, listens to how the network responds, and then allows you to find the exact source of the problem. It provides information on the type of impairment, tells you which nodes are in the most dire need, and provides statistical information on the duration and the time of day during which the impairment is most prevalent. The CNR of each return channel is used as a guideline to create alarms. Impulse noise, ingress and CPD can be discovered and recognized precisely. For ingress and impulse noise, this is the basis behind what we term Intelligent Monitoring, in which alarms are only set off in situations where an impairment is actually network-affecting and fixing the impairment would offer immediate benefit. This complements the predictive maintenance benefits of Hunter, mitigating CPD impairments before they even have a chance to become network affecting.
For detecting and ranging CPD (or any non-linear distortion), we use the existing QAM channels as radar signals to find the time distance to certain types of impairments within an HFC network. We then convert this time distance to linear distance, which tells the user how far away from the impairment they currently are. If electronic maps are available, we can also query a database generated from these maps. This database contains time distances to every device within the network. The database query outputs the number of devices that match the measured time distance within a certain tolerance. We then use field hardware and software to find the exact root cause of the network impairment.
There is a lot of equipment on the market that can tell a cable operator if they have a problem with a particular node. Many manufacturers make gear to look at various signal parameters and set alarms when certain numbers go above arbitrary thresholds. But often the operator already knows they have a problem – modems are not responding, customers are calling, etc. Hunter is the only system with the ability to tell the operator WHERE the problem is located. Additionally, we make use of a concept called Intelligent Monitoring, which sets alarms on ingress and impulse noise only when the impairment is network-affecting.
The core function of Hunter is the detection of intermodulation distortions. The system precisely measures the time delay of the signal to determine the exact source location of non-linear distortions like CPD. However, the CPD that we can see is a little different than the CPD that operators have become accustomed to seeing over the years. The CPD that is visible using traditional test equipment is, relatively speaking, very high level CPD. This is the common recurring 6MHz pattern (NTSC) in the return band. Hunter provides the technology to measure disturbances down to -70dB to -80dB – well below the cable system noise floor. This allows us to see any and all CPD-related problems in the plant, whether they are currently system affecting or not. It also allows us to locate CPD after it has seemingly disappeared to other, conventional equipment. Part of CPD impairment is what we term Digital CPD – this is CPD that is caused by intermodulation of QAM channels, and it often appears as broadband noise. It is very common for this type of CPD to be misdiagnosed as noise. We also provide data on the presence of impulse noise.
I-Scout probes are small devices that are permanently installed at various network points to assist in the quick identification of problem sources within specific areas of your network. Our probes work by applying a very small and imperceptible AM modulation of only .3dB at a very low frequency between 2 and 4 Hz. The modulation is invisible to all other equipment, but the Hunter can use it to mark signals coming from up to 32 selected zones within the node. The benefit is that instead of starting the search for the impairment source at the fiber node, the user can start the search much deeper into the network. I-Scout probes can be used to locate CPD and steady sources of ingress. Using i-Scout can significantly reduce the time required to fix a problem.
The presence of CPD is an indication of an impairment in the network. Some argue that there is no benefit in fixing low-level CPD, but when you locate a device in the field that is causing the low-level CPD and see how the device or connector is deficient, nobody will dispute the fact that the device or connector needs to be fixed or replaced. CPD is the result of a deficiency within a connector or device that is likely being attacked by the environment. The only certainty is that the problem will not fix itself, and that over time it will further deteriorate and become more of a problem.
Hunter is perfectly suited for the all-digital world. Nonlinear distortions like CPD often appear as broadband noise and will be indistinguishable from ingress. Hunter is the only tool that is able to differentiate between these two very different issues. It is an invaluable tool in troubleshooting an all-digital network. Additionally, for Hunter to operate, all analog channels are removed prior to the correlation process through filtering, so all digital operation is no different from the status quo system operation.
We have found problems in every type of device installed in a network. We have located nodes without seizure screws, amplifiers with hand-tightened connectors, rusted seizure screws, splitters where spiders had made a home, underwater amplifiers and taps, faulty terminators, bad F-connectors, oxidized taps and splitters, bad splices, overdriven drop amplifiers and line extenders, and bent stingers on pin connectors. In every system with Hunter installed, we have located devices that were having network-affecting problems and needed to be replaced.
Time and time again, when we go into the network to find the root cause of network impairments, the impairment is also a source of ingress. This makes sense when you think about it – devices that are sources of CPD have a problem: either a flaw in the connection interface of the device or a breach in the moisture seal that has allowed oxidation to take place. If there is a path where moisture can get into the unit, there is also an ingress path. Frequently, when you find and fix a CPD source or sources with Hunter, there is a positive impact on the system’s noise floor performance.
Additionally, much of the ingress seen within the return path is not truly ingress, but rather digital CPD. This is CPD caused from intermodulation of the digital channels that appears as wideband noise. Often this is mistakenly attributed as ingress.
The i-Scout can locate ingress zones when the ingress signal has a stable and non-pulsing nature. This is due to the demodulation process which takes 150 seconds. Impulse noise, by definition, is not stable. Therefore, it cannot be located using i-Scout Probes. I-Scout Probes can detect CPD and stable ingress sources.
Every operator has their own unique problems and goals, so of course benefits are realized in a variety of different ways. One customer’s system improved their node certification process from 2-3 nodes per month to 8-10 per week. They went from being way behind on their telephony launch schedule to way ahead. Because of this, they were able to launch early, thereby exceeding their revenue projections. Another system decreased the number of line techs from 8 to 5 within a three-month period, and was able to move the freed-up workers to revenue-generating tasks. The customer also fixed intermittent problems that had been ignored for years, drastically reducing the number of service calls and repeat calls. Overall, systems have seen improvements such as decreased service calls, fewer repeat calls, significantly improved noise floors, enhanced data and telephony performance, increased job satisfaction, improved network uptime, decreased operating costs, and significantly more efficient technical staff. A major European MSO realized the following improvements after one year of diligent use:
- (a) SDMH (severely degraded modem hour) errors per active broadband customer reduced by 69%
- (b) Average CER (code error rate) reduced by 51%
- (c) CNR < 25dB improved by 35%. This is percent of time C to N ratio is below 25dB.
- (d) Number of HOT SPOT NODES, number of nodes in ‘immediate action category’ – reduced by 38%. This is a measure of the quality of the signal severely affecting customer service.
Another way of looking at it is that the SDMH per active customer went from 0.94 hours/week in 2010 to 0.30 hours/week in 2011; to 0.09 hours/week in 2012 – an improvement (reduction) of a remarkable 90
Yes. Quite often, the root cause of the return path noise problem is also the root cause of the CPD problem. This is why we highly recommend that if a node has a noise problem, you immediately fix all CPD sources on that node. There isn’t necessarily a correlation between CPD severity and ingress, however. In many instances, Hunter has located relatively low-level CPD sources that turned out to be major ingress sources. This relationship between CPD and ingress is why we added the return channel CNR measurement feature to the Hunter System. Here, charts are available that show the relationship between CPD and the carrier to noise ratio (CNR) of return channel. In the vast majority of cases these two metrics will track each other. In fact, our first recommendation is always to fix nodes where CPD and CNR alarms are consistent and indicated together.
For a full Hunter System, seven major pieces of hardware and software are required: the headend radar, the headend server, the return path switches, the Quiver field tool, Quiver Navigator software (optional), and Xcor Client software. An additional software set called MHV Server provides the ability to view the status of multiple installed Xcor Hunter hubs on a single screen. This is intended for users that want to maintain system visibility over a large installed footprint. It is also possible to utilize the Quiver field tool as a stand-alone device.
The headend radar, the server, and the return path switches are all installed at the hub. The server contains the database and program that drive the entire system. The architecture of the system is based on a client/server scheme. The Xcor Client software can be installed on any Windows PC and connects to the server via IP. It provides the user with statistical information about the status of each node as pertains to the existence and magnitude of CPD, CNR, the presence of impulse noise, laser clipping, and the presence of node outages. This statistical information is gathered and processed in the server and used as an input to determine which nodes contain CPD, and which are affected such that the CNR is low or in critical condition. The Quiver and Quiver Navigator are field tools that are used to locate the exact location of the impairment. The Xcor Admin software is used to access the server in order to make configuration and setting changes related to system operation.
Just one. We’ve gone to great effort to make our system easy to use. Traditional CPD hunting techniques require multiple people. Typically, one technician pulls returns, physically disconnecting parts of the network – think of what happens to the phone call or data experience when this happens – and another watches a spectrum analyzer to see if the problem goes away. It is an archaic process that also only finds stable and very high level CPD.
Typically, a technician can find a problem related to CPD in about an hour.
Generally speaking, when long-time cable technicians and engineers think about CPD, they think about high-level CPD. This is the CPD that can be viewed on a spectrum analyzer. Whenever you can see high-level CPD, it is most likely already network-affecting. When you hear someone say they don’t have a CPD problem, what they really mean to say is that they don’t see high-level CPD very often. They can’t say they don’t have a low-level CPD problem for the simple reason that without Hunter or Quiver, they have no way to measure it. Low-level CPD is what we’re calling CPD that is below the noise floor of the system. All high-level CPD starts out as low-level CPD – that’s why it makes sense to address low-level CPD before it becomes network-affecting. This concept is what we call predictive maintenance. Through predictive maintenance, problems can be found and fixed on a much less frantic basis, and they can also be found before they impact the network and customers. The high-level CPD is usually already affecting CNR, while the low-level CPD will affect CNR at some point in the future.
Yes. We have seen this problem many times and in many different forms. In one system, we saw a large number of house amps that had problems – they were easily overloaded and became network-affecting. We located them and proved that they were a problem. In another system we found that all of a particular brand of nodes were set up incorrectly and were generating beats on the system. In yet another system we found that a particular brand of multi-tap had saturated ferrites, and that they generated significant beats from time to time. In many systems, incorrectly aligned and overloaded amplifiers have been found. Without our equipment, these problems would have been very difficult to find. An overloaded amplifier is a nonlinear distortion because new carriers are generated. It is similar to CPD in that CPD is also a nonlinear distortion. That is why our system finds these types of impairments.
Yes. We have e-mail and SNMP alarm features that are built into the Hunter System. The alarm is easily configurable and customized to provide the information you need.
Nodes are calibrated to speed up the scanning process by calculating the exact distance to the fiber node. Since CPD cannot occur within fiber, it does not make sense to scan this portion of the network. Whenever you calibrate a node, the calibration information is automatically transferred to the device database. The information is used by the system and the next time the network is scanned, the scan only takes place in the RF portion of the plant. By speeding up the scanning process, the wait time is minimized for field technicians who are calibrating or using the headend view feature. This is especially important for hubs with large node counts. For hubs with less than 100 nodes, calibration is unnecessary.
The nodes are the only devices that need to be calibrated. It is typically a one-time event, unless some fiber distances have changed.
During the calibration process, Quiver simulates a tiny non-network affecting source of CPD. To distinguish our calibration CPD from other CPD sources, we alter the waveform in a particular fashion. In the continual scanning process, Hunter monitors each node for CPD. Whenever Hunter detects CPD with this certain altered characteristic, it marks and records this CPD as a calibration signal.
Hunter can work without electronic maps. The downside of this is that technicians will likely need to spend a little more time in the troubleshooting process and need to make a few more stops to determine the direction of the impairment origin.
Map accuracy isn’t a big deal. What it comes down to is that if the maps are off it will take a little longer with Quiver and Quiver Navigator to track down the problem. Perhaps a technician will have to go to one or two more locations before finding the problem.
Arcom Digital has programs, processes and staff in place to quickly and efficiently convert updated maps. The maps and updated Hunter database are loaded onto the customer’s Hunter server. Everything is done electronically, including the remote update of the server with the new database.
No, i-Scout can only be used on a full Hunter installation.
DOCSIS 3.1 has robust time and frequency synchronization of OFDM signals at the CMTS and CM using Precision Time Protocol (PTP), which offers significant benefit. Using a fully DOCSIS 3.1 compliant CMTS with full access to the signal samples as provided for in the specification, Hunter will be able to migrate to a software only product, with no headend hardware.