Back to the List of the Granted Patents Click here to download KE000108 PDF
(11) Patent Number: KE 108
(45) Date of grant: 17/03/2009
(51) Int.C1.3: H 04N 7/10
(21) Application Number: 1999/000294
(22) Filing Date: 05/10/1999
(73) Owner: Hussein Mohammed Virani; of, P.O Box 22023 Nairobi, Kenya and Fuaad Omar AI-Kizim of, P.0 Box 46694 Nairobi., Kenya
(72) Inventor: Fuaad Omar Al-Kizim; Hussein Muhammad Virani and Athman Mohamed Athman Ali
(54) Title: MULTICAST BROADBAND INTERNET OVER HYBRID FIBRE AND/OR COAXIAL CABLE
The present invention relates to data transmission system having a logical bus structure (figure 1.) The transmission system of the present invention offers dedicated access (full time). The model of the present data transmission system can utilize any media capable of transmission electronic signals, and bandwidth free from the dependence of transmission resources. Information is carried on a multiplexing technique, allowing subscribers to share the bandwidth of the single transmission media. While Internet packets are being delivered to the subscribers, non-internet services are being carried on the same transmission media simultaneously and concurrently. The model is capable of sharing and delivering Internet without disruption to the host transmission media of Television Broadcast, Voice and data Telephony or Power Transmission among others. The host services and the Internet services are transmitted without one affecting the other and without depending on the transmission rate of the host services.
MULTICAST BROADBAND INTERNET OVER HYBRID AND / OR COAXIAL CABLE.
BACKGROUND OF INVENTION
Access to the Internet falls into two broad categories: dedicated access and dial-up access. With dedicated access, the computer is directly connected to the Internet via a router, or the computer is part of a network linked to the Internet. With dial-up access, a computer connects to the Internet with a temporary connection, generally over a telephone line using a modem a device that converts the electrical signals from a computer into signals that can be transmitted over traditional telephone lines. A modem is needed because computers are digital, meaning that their signals are made up of discrete units, while most telephone lines are analog, meaning that they carry signals that are continuous instead of discrete. Once a signal has traveled over the telephone line, a second modem is required at the other end of the line to reconvert the transmitted signals from analog to digital. A great many companies, called Internet Service Providers (ISPs), provide dial-up access to the Internet for a fee.
All data transmitted over the Internet is divided up into small units of information called
packets, each of which is labeled with a unique number indicating its place in the data
stream the flow of information between computing devices. When the various packets that make up a set of data arrive at their destination, they are re-assembled using the unique labels given them. If part of the network over which the packets are sent is malfunctioning, or down, special automatic features of the Internet's routing equipment re-route the packets so that they travel over functioning portions of the network. Other features make sure that all the data packets arrive intact, automatically requesting those missing or incomplete packets is re-sent from the source. This system, called packet switching, uses a series of protocols, or rules, known as TCP/IP (Transmission Control Protocol/Internet Protocol).
In-spite of tremendous advances that have been made in improving of cable communications, various challenges are yet to be overcome. Television, for example, has been termed by critics as a one way street. Programs are broadcast to viewers, but no mechanism is in built for viewers to talk back to the producers, or to send their own signals to the network. This is true with typical cable TV implementations, which transmit data from the head end to the subscriber's end. Thus referred to as the downstream direction or transmission.
This invention provides a data transmission system whereby cable TV network can transmit data signals in two directions; upstream and downstream. The HFC network is one of the implementations of such a two-way traffic cable TV network system enabling upstream, subscriber to provider transmission. To achieve this, the cable TV network has to be modified and this is the spirit of the present invention.
In accordance to the present invention, spectrum of bandwidth is allocated for the signals
travelling in upstream direction. Normally the frequency range from 5-42 MHz is used for the upstream data transmission. This frequency is usually broken down into 6MHz/8MHz channels (Giving 4 usable upstream channels)
The invention further relates to use of amplifiers on the existing cable television network that are capable of separating the upstream and the downstream signals and amplifying each 10 direction separately, in the right frequency range.
The downstream transmission from the head-end is broadcast. That is the same signal is
sent on all the cables whether subscribed or not. In contrast upstream transmission is inherently personal cast i.e. each subscriber is trying to place a signal on the upstream channels. Hence some form of access method is needed to arbitrate/multiplex which signal is actually being carried.
The system in accordance to the present invention involves having a model, as depicted in figure 1, laid on a logical Bus architecture or topology. (In contrast to Star in PSTN) The bus comprises of an assortment of cabling (coax, HFC, Twisted pair UTP etc.)in its various forms colors combinations provided they carry sufficient or more than sufficient bandwidth. The Bus comprises of the Aerial Plant, Bridger, amplifier, Bridger tap, Brouters, Bridging Amplifiers, drops, drop cables, connectors, FDDI’s feeder cables, feeder lines, local loops, LPF's, main trunks, Multiplexers, NUI, repeaters, Shared wire Networks, Splitters, switches, taps, thick wires, Thin wires, transmission Links, trunk amplifiers, trunk cables, trunk lines that runs throughout the community, placed at various distances, quantities etc. to achieve geographical coverage. In accordance to one embodiment, the model may be further extended from the IPOP through T1 lines to serve a different community serviced by a separate head-end but linked to the Mother IPOP as shown in figure 2.
Working of the Model
The Cable Company acts, as its own ISP it establishes a connection to the Internet via a 35 router and receives data over a 1.5Mbps or 45Mbps (T1/T3) line leased from local Phone Company. Data has to be converted from IP packets to coax signals. The overall structure or topology of the cable network follows a Tree and branch architecture. In each community the "Head-end" (the originating point of the Cable TV signals) is installed to receive satellite and traditional over the Air broadcast television signals.
These signals are then carried over to the subscribers' homes over coaxial cable that runs
from the head-end throughout the community. (Or runs partly on fiber optic cables and
partly on coaxial cables in case of an HFC network) Each 6MHz/8MHz TV channel is transmitted in analog form over (Not necessarily the same) 6 MHz/8Mhz of enclosed spectrum on the (single) cable. This is achieved by using a multiplexing scheme such as FDMA. Different channels are sent on different frequency offsets, for example, 6MHz 12MHz 24MHz 30Mhzetc) as depicted in figures 3 and 4.
The speeds of this access depend on the equipment used to modulate the digital computer information onto the CATV analog TV channels. Such equipment can provide bandwidths ranging from 500kbps to 10Mbps currently.
To deliver digital data, the head-end controller modulates the IP packets (received from the leased Tl/T3 line), encodes the data and broadcasts the signal down the cable on an unused channel. This forms the downstream. Generally, the head-end controller also acts as a traffic administrator using a special control channel. The head-end controller tells each subscriber interface device when it can transmit, on which frequency and for how long (to form the upstream). When the subscriber switches on his/her computer, the interfacing device scans all its assigned channels to locate the control channel, which can be identified because it uses a unique header signal. Then using information obtained from this control channel the computer goes online.
To achieve geographical coverage of the community, the cables from the head-end are split or branched. However, the content (signals) in these branches are not "split" i.e. all the subscribers receive all the downstream all the time. Thus, this forms logical bus architecture. Since these channels from the head-end carry both TV and internet data signals there has to be a device at the subscriber end which selects the internet data content and sends it to the computer. So when the CATV cable enters the subscriber premises it is split and one branch goes to the computer(s) and the other branch goes to the TV(s).
HFC cable TV's "tree and branches" architecture makes it hard to provide a signal of consistent quality along the length of the cable (from the head-end to the subscriber). Each time the cable is split the signal becomes weaker. Hence, amplifiers are needed to boost the signal strength. These amplifiers have to be able to support both upstream and downstream transmission of data.
The subscriber-interfacing device is a cross between a network adapter and a modem. It has to be able to scan the channels from the incoming cable to locate the control channel. It also has to be able to modulate upstream data and to de-modulate downstream data. Since the head-end controller transmits data downstream in a broadcast manner, the subscriber interface device just tunes in to the assigned downstream channels and demodulates the data. For upstream data, the subscriber interface has to monitor the control channel for instructions from the head-end controller on when and on what frequency it can transmit.This is because all the subscribers use the upstream channels to send data to the head-end.
The transmitted signal from the subscriber interface device can be so strong that any TV sets connected on the same cable might be disturbed. The isolation done by the splitter is normally not enough to prevent this. So an extra high-pass filter is connected to the TV branch of the cable and hence only high-frequency TV signals pass into the TV set and all low frequency upstream transmissions from the computer do not have any effect since they are filtered out and directed to the Head-end.
The subscriber interface device can be of two types: - External: External to the computer and is connected to the computer's system unit via a network interface card (NIC).Internal: maybe a PCI add-on card for a personal computer.
The external interface device can be connected to the personal computer via a network interface card, which is fast enough to absorb data at the rates that the subscriber interface device receives data from the coaxial (CATV) cable. Normally, an l0BaseT (Ethernet) N1C will suffice, though 100BaseT (fast Ethernet) or any network interface card that can handle the data transmission/receiving rates can be used.
The subscriber interface device contains the following components: (Also referred to as the cable modem)
Tuner: The tuner connects directly to the CATV cable. Normally, a tuner with a built-in
duplexer is used so that we can provide both upstream and downstream through the same tuner. The tuner must be of sufficiently good quality to be able to receive the digitally modulated QAM signals.
The Modulator: In the receive direction, the signal goes through a demodulator which normally consists of an analog to digital (A/D) converter, QAM-64/256 demodulator, MPEG frame synchronization and finally Reed-Solomon error correction.
The Burst modulator: In the transmit direction, a burst modulator feeds the tuner. The burst modulator does Reed-Solomon encoding of each burst, modulation of the QPSIC/QAM-16 on the selected frequency and the digital to analog (D/A) conversion, The output signal is fed through a driver with a variable output level so that the signal level can be adjusted to compensate for the unknown cable loss.
The MAC: A Media Access control mechanism is in between the receive and the transmit paths. This can be implemented as hardware or as a combination of hardware and software.
The interface: The data from the MAC passes into the subscriber's computer via a network interface, be it Ethernet, fast Ethernet, PCI or USE.
Normally, one head-end controller can drive about 1000-2000 simultaneous subscriber interface devices on a single 6MHz/8MHz channel. If more subscriber interface devices are required the number of channels used by the head-end controller are increased (in 6MHz or 8MHz increments)
Downstream: Is the term used for the signal received, by the subscriber interface device from the head-end controller. The electrical characteristics of this signal are out lined below:
Bandwidth: 6MHz 8MHz
Modulation: 64 QAM with 6 bits per symbol or 256 QAM with 8 bits per symbol.
The data rate depends on the modulation and the bandwidth as shown below:
64 QAM 256QAM
6MHz 31.2Mbps 41.6Mbps
8MHz 41.4Mbps 55.2Mbps
Note that a symbol rate of 6.9 Msym/s is used for the 6MHz bandwidth and 5.2Msym/s for the 6MHz bandwidth to take care of the transmissions overheads.
All subscribers receive the downstream data and each subscriber interface device filters out its data from the stream of data broadcast by the head-end controller.
Upstream: Is the term used for the signal transmitted by the subscriber interface device, to the head-end controller (The cable provider end) Upstream data is always sent in bursts so
that many subscriber interface devices can transmit data using the same frequency. The frequency range for the upstream is typically 5-65MHz or 5-42MHz. The bandwidth per channel may be e.g. 2MHz for a 3Mbit/s QPSK channel. The electrical characteristics of the upstream signal are summarized below:
Frequency: 5-65MHz 5-42MHz
Bandwidth: 2MHz 2MHz
Modulation: QPSK with 2 bits per symbol or 16-QAM with 4 bits per symbol.
One downstream is normally paired with a number of upstream channels to achieve the balance in data bandwidths required. Each modem transmits bursts of data in time slots that 25 can be marked as "ranging", "contention" or "reserved".
Reserved time slots: Are slots that can be used only by one particular subscriber interface
device to transmit data upstream; no other subscriber interface device can transmit in that
time slot. The bead-end controller allocates the time slot to the various subscriber interface devices.
Contention slots: Are slots, which are open; all subscriber interface devices can transmit in that slot. If two (or more) subscriber interface devices transmit in that slot the data is
garbled. The head-end controller will send a signal if it does not receive any signal in that slot. Hence, the subscriber interface devices will know that they have to re-transmit whatever data they had sent within that slot. These types of slots are used for short transmissions.
Ranging slots: Are used by the head-end controller to synchronize the subscriber interface devices with it. The other purpose of the ranging slots is to ensure that all subscriber interface devices transmit at the same signal level, which is essential for detection of "collisions" of data in the contention slots and also for the optimal performance of the upstream demodulator at the head-end controller.
Cable bandwidth is allocated dynamically. When a subscriber is not transferring traffic he does not consume transmission resources refer to figure 5.
The Media Access control mechanism is normally implemented in hardware or as a combination of hardware and software. The MAC allows the media to be shared in a reasonable way. Both the head-end controller and the subscriber interface device implements the protocols to do:
Ranging to compensate for different cable losses.
Ranging to compensate for different cable delays.
Assignment of the frequencies to the subscriber interface devices (done by the head-end controller). The subscriber interface devices first listen to the downstream to collect information about where and how to answer. Then it signs onto the system using the assigned frequency.
Allocate the time slots for the upstream data transmission (done by the head-end controller).
At the subscriber end, the number of computers connected to a single subscriber interface
device can range from just one or more than one; many computers can be connected to a
single interface by having one computer which acts as a gateway for the rest. The network connections between the gateway and the other computers has to be fast enough to match the data rates being produced by the subscriber interface device.
For electric-power transmission, three-wire cables sheathed with lead and filled with oil under pressure are employed for high-voltage circuits; secondary distribution lines usually employ insulated single-conductor cables. In residential electric wiring, B-X cable is often used. This type of cable contains two insulated conductors, which are wound with additional layers of insulation and covered with a helical wound strip of metal for protection. The ignition cable used to carry high-voltage current to the spark plugs of an internal-combustion engine is a single-conductor cable; it is covered with cloth impregnated with shellac for insulation.
The fiber cabling that can be used in the HFC network (refer to fig.) can be classified according to the following:
• The type of cabling connection.
• The type of fibre cable.
• The way the data is transmitted using the fibre cable.
Type of cabling connection:
The HFC network uses both coaxial and fibre optic cabling, hence the name. The fibre cabling may be used either on the ISP side of the network (refer to fig.) or on the subscriber end of the network (refer to fig.).
On the ISP end, the fibre cabling is used either internally within the ISP (to connect the servers and the routers to the Tl/T3 high-speed lines) or as a fast network interface between the ISP and the coaxial cabling to which the subscriber is connected. Thus, by the use of fibre cabling, the ISP ensures that the speeds at which the data is being received from the Tl/T3 links is being matched by the speeds at which the data is transmitted to the subscriber. The ISP can choose to have both its internal network and the interfacing network (between the ISP and the Coaxial part of the network) to be on fibre. Altematively, he may choose only to have the interface between the subscriber and him to be on fiber and the internal network to be using some other high-speed network. In the latter case, however, the ISP has to have the right interfacing devices between his internal high-speed network (non-fibre) and the interfacing fibre network.
On the subscriber end, the subscriber usually uses a very high-speed non-fibre network e.g. Base-T or 100Base-T, but he also has the option of using a fibre network as an interface between his computer(s) and the cable modem. This is particularly in the case of individuals/organizations that would like to utilize few cable modem connections for a large internal (their) network. In such configurations, the subscriber LAN/MAN has to be able to match the high data speeds at which the cable modem is receiving data.
Type of fibre cabling used:
Fibre cabling can be classified according to the way they have been constructed.
In the HFC network, the following types of cables might be used to provide the cabling.
• Distribution fibre optic cabling.
• Short-span aerial fibre optic cabling.
• Simplex fibre optic cabling.
• Heavy duty fibre optic cabling.
Note that the types of fibre optic cabling that can be used in the HFC network are not restricted to the ones mentioned above. In fact, any fibre optic cable that can support the speeds at which the HFC network is supposed to work (or even higher speeds!) can be used as the cabling.
Another thing to note is that the HFC network allows many subscribers to be using the network cabling to transmit data simultaneously. Hence, the fiber optic cable used in the network should ideally be a multi-mode fiber optic cable. However, single-mode fiber optic cabling can be used as long as the resulting network can support multiple users simultaneously.
The kind of protection that the cabling should use can be either tight buffered or loose tube. However, any other kind of protection (whether existing or not) may be used. When tight buffered the individual fibre is covered directly with a thin layer of thermoplastic material. In loose tube fibre cabling protection, one or more fibres can be contained within a loose tube which is filled with thixotropic gel.
How the HFC Cable Transmits the Data:
Light travels at a constant speed of approx. 300, 000Krn/s. in a denser medium such as glass or water (as compared to air) it moves slower. This is evident from the fact that light refracts towards the normal as it enters the denser medium (refer to fig.). The refractive index (n) of a transparent material is the ratio of the speed of light in a vacuum to the speed of light in that medium.
If light moves from a denser medium into a less dense one it will be refracted up to a certain angle of incidence (called the critical angle) beyond which total internal reflection (TIR) occurs (refer to fig.). Fibre optic cables uses TIR to trap light within a glass optical fibre and transmit it over long distances. A fibre with the right optical characteristics can hence guide light carrying much more information faster than the electrons in a copper wire.
The attenuation of the silica-based fibre, as used for telecommunication and data transfer, changes with the wavelength at which the information is relayed. One of three transmission windows are used, 850, 1300 and 1550 nanometer wavelengths, are used.
There are two categories of optical fibre, referred to as multi-mode and single-mode. Multi-mode allows several modes to be transmitted simultaneously while single-mode fibre can transmit only one mode at a time. Both the single-mode and the multi-mode type of fibre cabling can be used in the HFC network.
1. A data transmission system having a logical bus structure and offering dedicated access characterized in that it is a system providing multiple simultaneous services on same cable and which is capable of having a downstream as well as an upstream signals concurrently.
2. A data transmission system of claim 1 having transmission speeds bit rate of between about 10 to about 300 times faster than using typical PSTN networks.
3. A data transmission system of 1 further including amplifiers capable of maintaining strength of said downstream and said upstream signals.
4. A data transmission system of claim I having a means for the community to connect to the internet by sharing a single Low, Medium or High bandwidth medium either combinations, jointly or individually in any ratio.
5. A data transmission system of in claim 1 providing real full time connections, provided that the Head end & subscriber are powered without utilization of 25 transmission resources.
6. A data transmission system of claim 1 with means to transmit on a Logical bus, Tree branch architecture, which is capable of delivering full content to every subscriber whether online or offline at all time.
7. A data transmission system of in claim 1 further including a Frequency Division
Multiple Access (FDMA) and Carrier sense multiple access.
8. A data transmission system of claim 1 having dynamic bandwidth allocation.
9. A data transmission system of claim 1 further comprising a means for multicasting
data in packets on a channel over a single data transmission medium.
10. A data transmission system of claim 1 further comprising a filtering means for directing particular channels.
11. A data transmission system of claim l having ability to branch architecture of plant by physically splitting a portion of said upstream or downstream signal to travel on branches.
13. A data transmission system of claim 1 with means for the Head end to administer
14. A data transmission system of claim 1 with e functions of the time slots & their
15. A data transmission system as in claim (1) with a function for the channel assignment of the head-end by assigning channels and unique headers to each subscriber.
16. A data transmission system of claim 1 with Transmission mediums capabilities, whether coaxial, fiber optic, CAT family, drop wires and all or any medium capable of transmitting multiple broadcast of any shape, size color, material, combination of materials, physical layers in medium.
17. The logical bus structure of the data transmission system in accordance with all preceding claims from the head-end, to the aerial plant, underground plant, pole to pole plant, pole to subscriber, subsequent splitter to serve the different functions at subscriber end together with all interfacing connecting devices therein.
MULTICAST BROADBAND INTERNET OVER HYBRID/COAXIAL CABLE ABSTRACT
The present invention relates to a data transmission system having a logical bus structure
(figure 1). The transmission system of the present invention offers Dedicated Access Full Time. The model of the present data transmission system can utilize any data transmission media capable of transmitting electronic signals, and bandwidth free from dependence on Transmission Resources. Information is carried on Multiplexing technique, allowing subscribers to share the Bandwidth of the single transmission media. While Internet packets are being delivered to the subscribers, non-Internet services are being carried on the same transmission media simultaneously and concurrently.
The model is capable of sharing and delivering Internet without disruption to the host
transmission media of television broadcast, Voice and Data Telephony or power transmission cable. Host service and the Internet service are transmitted without one affecting the other and without depending on transmission rate of Host signal.