Dynamic allocation of frequency spectrum

The present invention relates to methods and arrangements for dynamically allocating communication system frequency spectrum after interference between transmission medias using the frequency spectrum.

Resource handling and resource utilization are fundaments of any modern communication. In the case of Very-high-speed Digital Subscriber Line VDSL modems, this fundamental issue is represented by frequency planning. The frequency planning in VDSL has to take many aspects into consideration: political, economical, and practical. Different frequency plans fit different services favored by one or the other operator and practical issues introduce many side constraints. Examples of these practical aspects are the location of HAM bands for radio amateurs and the frequency plans of other broadband systems, for instance Asymmetrical Digital Subscriber Line ADSL. Most common systems of today, including VDSL, use fixed allocation of the frequency spectrum. This leads to under-utilization of the medium.

Signaling in the copper plant is subject to cross-talk, which means that communication on a given line is disturbed by signals transmitted on other lines. The coupling between lines in general increases with frequency but can vary substantially from line to line. However, into any given line, the cross-talk is usually strong only from a small number of other lines.

Modern Digital Subscriber Line DSL systems are designed to be limited by either near-end cross-talk NEXT or far-end cross-talk FEXT. For FEXT limited systems, frequency planning is necessary in order to control cross-talk, that is, to avoid NEXT. While NEXT-limited systems do not need any spectrum management, they could still benefit from transmit-power management. In FEXT-limited systems, upstream and downstream traffic is typically assigned to different frequency bands. Without this frequency division duplexing FDD, the NEXT would severely reduce the available capacity in the network. As an analogy to this type of multi-user interference, we can compare with the situation in a cellular radio-communication system. The copper lines in a DSL system interfere significantly with only a few neighboring lines and thereby function as directional antennas do in a cellular system. However, cellular radio systems usually have much more sophisticated resource allocation mechanisms than only FDD. Correspondingly, the lack of efficient resource allocation mechanisms in the copper plant leads to a significant under-utilization of the available resources. In a U.S. Pat. No. 6,167,095 is disclosed a method for variably allocating upstream and downstream communication spectra. In the US patent, upstream and downstream portions of the frequency spectrum are allocated based on computed signal quality parameters representative of the line conditions. The signal must be characterized before the allocation of frequency bands in the spectra takes place. Chaotic resource allocation will potentially lead to system breakdown. To avoid system breakdown, coordination between the modems is necessary. Without the present invention, this in general requires a central site determining the upstream and downstream carrier frequencies depending on the line noise characteristics.

The present invention solves problems related to dynamic and efficient allocation of frequency spectrum without using a central coordinating device. Using cross-talk coupling between medias in order to make some of the medias observant of a possible disturbing situation solves the problem. The invention uses the fact that interference and cross-talk coupling is symmetric in the sense that if interference arises on a disturbed first transmission media due to cross-talk coupling of transmission on a disturbing second transmission media, a cross-talk coupling also exists in the converse direction. Due to the symmetry in the cross-talk coupling, transmission on the first media will arise via the cross-talk coupling as interference on the second media. According to the invention, a media control signal is sent on the disturbed media after cross-talk interference. The media control signal then reproduces and is detected on the disturbing media. Depending on whether the involved medias are used in a defined default direction or in a direction opposite the default direction, different measures will be taken.

More in detail in a first embodiment, cross-talk interference is detected on a disturbed transmission media used by a disturbed set of modem pair in a communication system. A disturbing transmission media, used by a disturbing set of modem pair in the system, causes the interference due to a cross-talk coupling. A media control signal is sent by the disturbed set of modem pair on the disturbed media. The media control signal is reproduced on the disturbing media due to the cross-talk coupling and is detected by the disturbing set of modem pair.

More in detail in a second embodiment, collision avoidance data is transmitted in resources of frequency spectrum on a disturbing transmission media by a receiving-end modem of a disturbing set of modem pair. The purpose of the collision avoidance data is to avoid collision in the form of strong NEXT interference. Cross-talk interference is detected on a disturbed transmission media used by a disturbed set of modem pair. The collision avoidance data on the disturbing media causes the interference due to a cross-talk coupling. A media control signal is sent by the disturbed set of modem pair on the disturbed media. The media control signal is reproduced on the disturbing media due to the cross-talk coupling and is detected by the disturbing set of modem pair.

More in detail in a third embodiment, modem-pairs in the system that are influenced by interference are synchronized in time. Collision avoidance data is then simultaneously transmitted by receiving-end modems of so called disturbing set of modem pairs at predefined first set of timeslots in the frequency spectrum on disturbing transmission medias. Cross-talk interference is detected on a disturbed transmission media used by a disturbed set of modem pair. The collision avoidance data from one of the disturbing medias caused the interference. A media control signal is sent at a predefined second set of timeslots by the disturbed set of modem pair on the disturbed media. The media control signal is reproduced on the disturbing media and is detected by the disturbing set of modem pair.

More in detail in a fourth embodiment, a further modem-pair, a so-called non-disturbing modem-pair detects a second reproduced media control signal. The second control signal is reproduced from the already mentioned media control signal sent by the disturbed set of modem pair on the disturbed media (see the earlier embodiments). The media control signal comprises in this embodiment information about transmission direction of the disturbed transmission media and is reproduced on the non-disturbing media due to a second cross-talk coupling. Due to the carried transmission direction information, the non-disturbing modem-pair can choose to either ignore the signal or to take measures to handle the situation.

The object of the present invention is to use capacity in the network in the sense that users in the system dynamically can seize the capacity they need if this disturbs no one else.

One advantage with the invention is that capacity can be allocated flexibly and according to need.

Another advantage is that no central coordinating device is needed.

Yet another advantage is that it is possible to increase traffic load without loosing control over the system. For moderate overall traffic loads, media utilization can be much better than static frequency planning. With increasing traffic load, the performance goes down towards that of a static frequency planning.

The invention will now be described more in detail with the aid of preferred embodiments in connection with the enclosed drawings.

FIG. 1A shows a block schematic illustration of a communication system comprising modem pairs having receiving end modems and sending end modems.

FIG. 1B shows the frequency spectrum divided into an example set of different frequency channels each having a defined default direction.

FIG. 2 shows a block schematic illustration of a communication system with a disturbing modem pair and a disturbed modem pair, each modem pair having a receiving end modem and a sending end modem. Signaling directions in the figure disclose the signaling in a first embodiment.

FIG. 3 shows a flow chart illustrating the first embodiment when an intention signal is used.

FIG. 4 shows a block schematic illustration of a communication system with two modem pairs. Signaling directions in the figure disclose the signaling in a second embodiment when collision avoidance data has been sent.

FIG. 5 shows a block schematic illustration of a communication system comprising modem pairs. Signaling directions in the figure disclose the signaling in a third embodiment of the invention whereby the modems have been synchronized in time.

FIG. 6 shows a block schematic illustration of a communication system comprising a disturbing modem pair, a disturbed modem pair and a non-disturbed modem pair, in a fourth embodiment of the invention.

FIG. 1A discloses an example broadband wireline communication system. The system in FIG. 1A is a copper wire installation that comprises a central office CO or a cabinet where modems are co-located. It is to be noted that this is an example and that the invention is not restricted to wireline systems. Additionally, instead of having a central office, more important is that modems are within cross-talk range from each other. The system comprises modem pairs A 1 -A 2 , B 1 -B 2 , C 1 -C 2 , D 1 -D 2 , E 1 -E 2 , F 1 -F 2 , G 1 -G 2 , H 1 -H 2 , each pair including a receiving end modem and a sending end modem. The modem pairs communicate on transmission media in either an upstream or downstream direction. In FIG. 1A , it can be seen that modem ends A 1 , B 1 , C 1 , D 1 , E 1 , F 1 , G 1 , H 1 of the modem pairs are co-located while the other modem ends A 2 , B 2 , C 2 , D 2 , E 2 , F 2 , G 2 , H 2 are at different locations. When a sending end modem transmits, it may cause near-end cross-talk NEXT to receiving end modems located close to it when using the same frequency band. Assume, for example, that the sending end modem H 1 transmits to its receiving end modem H 2 . If the modem pair F 1 -F 2 at the same time is using the same frequency band in the opposite direction, a receiving end modem F 1 most likely will be disturbed by NEXT.

FIG. 1B discloses with arrows an example of how default transmission is defined for different frequency bands. From a resource sharing point-of-view in a time-asynchronous system, the relevant measure of what resources a modem uses for transmission is the distribution of power over the frequency band. When one modem wants to change the power and/or frequency allocation used on the line, then some control information describing the new transmission setup could be communicated to the other end. This control information may be communicated on a pair of pre-allocated control channels, for instance the two lowermost bands CH 1 and CH 2 . Since in the present scenario there is no central resource controller, it is desirable to simplify the spectrum management scheme. One suitable way of doing this is to pre-define a set of frequency bands, each having a default transmission direction, i.e. either upstream or downstream. One example of a possible frequency layout is shown in FIG. 1B . The two lowermost bands CH 1 and CH 2 in the figure are set up for fixed use (i.e., in these bands it is never allowed to transmit in the opposite direction to the default). All other bands CH 3 , CH 4 , CH 5 and CH 6 are set up for a dynamic upstream or downstream use. A modem is allowed to freely use a frequency band in the default direction, possibly under some power restrictions. Under some harder restrictions, a modem can be allowed to use some frequency bands also in the reverse of the default direction. The idea is to allow a single user to use all available resources if no one else needs them at a given point in time. A modem that is using a frequency band in its opposite direction must be prepared to back off from that band if someone else wants to use it in the default direction.

As soon as resources are not needed any more, they should be released to allow other users to access them. Obviously this will cause variations in the cross-talk on the line. All users will however benefit from a significant reduction of the average interference level compared to the case when all modems are transmitting with a constant power whether they have data to transmit or not.

A prerequisite for the invention is that one or several resources in each frequency band regularly are set aside. These resources could be either sub frequency bands or time slots. Such a (small) resource could, for instance, be a few consecutive symbols out of a thousand. That is, after each thousand symbols, a small number of symbols (such as four) could not be used for regular data transmission. For instance, after one thousand symbols have been transmitted, one symbol period could be left silent, the next used in the opposite direction of the one thousand first, one more silent, and the another thousand symbols in the first direction, and so forth. The small resource can also be used to listen to cross-talk signals from other modems. One or several such small resource could also be a (small) portion of the spectrum in a frequency band. These small frequency portions of the spectrum should then always or at parts of the time be used in the opposite direction. Optionally, several such small resources could be avoided when placing data. A purpose of the unused resources and the reversing of transmission direction will be more obvious when the different embodiments are further analyzed later in the description.

FIG. 2 discloses in a first embodiment of the invention, the modem pairs A 1 -A 2 and B 1 -B 2 of the modem pairs already shown in FIG. 1A . Each modem pair comprises a sending end modem and a receiving end modem. Two of the modem ends A 1 and B 1 of the modem pairs A 1 -A 2 and B 1 -B 2 are located within cross-talk range, as an example of location within the central office CO, and the two other modem ends A 2 and B 2 are located outside the central office. All of the modem ends A 1 , A 2 , B 1 and B 2 comprise each one, a transmitter TXA 1 , TXA 2 , TXB 1 and TXB 2 and a receiver RXA 1 , RXA 2 , RXB 1 and RXB 2 . Each modem end A 1 , A 2 , B 1 and B 2 has the opportunity to become, within each frequency band, either a sending end modem or a receiving end modem. The current traffic situation decides which side of the modem pair that constitutes sending end or receiving end side. The modem ends A 1 and A 2 are in this example connected via a wire CH-X and thereby constitute the modem pair A 1 -A 2 . The modem ends B 1 and B 2 are connected via a wire CH-Y and constitute the modem pair B 1 -B 2 . The tube form symbol T symbolizes that the wires are close to each other within cross-talk range and that cross-talk coupling and interference can occur. In the example that now will be discussed, the modem end A 2 is transmitting on frequency channel CH 4 (shown in FIG. 1B ). The modem end A 2 thereby constitutes the sending end modem. The modem end A 1 receives the transmission from A 2 and thereby constitutes the receiving end side. As can be seen in FIG. 1B , frequency channel CH 4 is defined to have the upstream direction as default direction. Upstream direction is in this example the direction towards the central office CO. The modem pair A 1 -A 2 thereby communicates in the default direction.

A method according to the first embodiment of the invention will now be discussed. The method shows how cross-talk is used to communicate with a disturbing source and to send signals from a disturbed modem pair to a disturbing modem pair. Unfilled arrow symbols in FIG. 2 represent transmission signals after having passed the cross-talk coupling. From now on the modem pair A 1 -A 2 is called the disturbed set of modem pair A 1 -A 2 and the modem pair B 1 -B 2 is called the disturbing set of modem pair B 1 -B 2 . The method comprises the following steps:

The sending end modem A 2 of the disturbed set of modem pair A 1 -A 2 transmits data 10 on the frequency channel CH 4 to the receiving end side A 1 . Due to the long distance between the end modem B 1 of the disturbing set of modem pair B 1 -B 2 and the sending end modem A 2 of the disturbed set of modem A 1 -A 2 , B 1 will not hear the transmission from A 1 -A 2 . For the same reason the end modem B 2 will not hear the transmission from A 1 -A 2 . The modem B 1 starts to transmit data 11 on its wire CH-Y on the same channel CH 4 as the sending end modem A 2 is transmitting on. This since B 1 is unaware of the fact that the channel frequency CH 4 is already occupied. The receiving end modem A 1 detects cross-talk interference CTI on wire CH-X on the disturbed channel CH 4 . The cross-talk interference is caused by the transmission on wire CH-Y used by the disturbing set of modem pair B 1 -B 2 . This due to the cross-talk coupling T between CH-Y and CH-X. The receiving end modem A 1 transmits a media control signal CS on the frequency channel CH 4 . As already have been mentioned when FIG. 1 was discussed, all modem pairs have the opportunity to reverse direction during short time intervals. A possible variation would be that also A 2 transmits a media control signal. The media control signal CS is reproduced from wire CH-X to wire CH-Y and becomes a reproduced media control signal CS′. This due to fact that interference is symmetric in the sense that if there is cross-talk coupling from CH-Y to CH-X, there is also cross-talk coupling in the opposite direction, from CH-X to CH-Y. The reproduced media control signal CS′ is detected by the disturbing sending end modem B 1 . The disturbing set of modem pair B 1 -B 2 leaves the channel frequency CH 4 . The disturbing modem pair is using the frequency channel CH 4 in a direction opposite the defined default direction (see FIG. 1B ) and is forced to take the consequences of this and leave the channel. Another possible scenario would be that the frequency channel CH 4 was defined as having the opposite direction as default direction. The set of modem pair A 1 -A 2 then used the frequency channel CH 4 in a direction opposite the default direction while the modem pair B 1 -B 2 was using the channel in the defined default direction. In this case the modem pair A 1 -A 2 will leave the channel CH 4 . Optionally a media control signal can be sent by A 1 -A 2 indicating that it is most desirably for A 1 -A 2 to keep the channel. If the pair B 1 -B 2 then has the possibility to find another channel, this may happen.

As an alternative to the shown embodiment, B 1 instead, before starting to transmit its data 11 , sends an intention signal IS in one resource within CH 4 , thereby announcing its intention to start transmitting. The modem pair A 1 -A 2 can then follow the above pattern before actual data transmission by B 1 -B 2 commences. This may prevent potential loss of data caused by collision. This alternative will now be exemplified by an example below.

In FIG. 3 the most essential steps of the described first embodiment when using the intention signal is disclosed in a flow chart. The flow chart is to be read together with the earlier shown figures. The most essential steps are as follows:

The frequency channels CH 3 -CH 6 in the frequency spectrum are defined regarding default directions. This is shown in FIG. 3 by a block 101 . The sending end modem A 2 transmits data 10 on the frequency channel CH 4 to the receiving end side A 1 . The modem pair A 1 -A 2 is using the channel in the defined default direction. This is shown in the figure by a block 102 . The modem B 1 intends to start transmitting data 11 on its wire CH-Y on the same channel CH 4 as the sending end modem A 2 is transmitting on. The modem pair B 1 -B 2 will use the channel in a direction opposite the defined default direction. Before starting to transmit the data, B 1 sends an intention signal IS in the direction opposite the default direction. This is shown in the figure by a block 103 . The receiving end modem A 1 detects cross-talk interference CTI on wire CH-X on the disturbed channel CH 4 . The cross-talk interference is caused by the intention signal on wire CH-Y. This is shown in the figure by a block 104 . The receiving end modem A 1 transmits a media control signal CS on the wire CH-X. This is shown in the figure by a block 105 . The media control signal CS is reproduced from wire CH-X to wire CH-Y and becomes a reproduced media control signal CS′. This is shown in the figure by a block 106 . The reproduced media control signal CS′ is detected by the disturbing sending end modem B 1 . This is shown in the figure by a block 107 . The disturbing set of modem pair B 1 -B 2 leaves the channel CH 4 . This is shown in the figure by a block 108 .

FIG. 4 discloses in a second embodiment of the invention, the modem pairs A 1 -A 2 and B 1 -B 2 already shown in FIG. 2 . A method according to the invention will now be disclosed in connection with FIG. 4 . In the method, the end modem B 2 is acting as sending end modem and transmits data 11 to its receiving end modem B 1 on the frequency channel CH 5 . The method will show, like before, how detection of cross-talk is used to communicate with a disturbing source and to send signals from a disturbed set of modem pair to a disturbing set of modem pair. The modem pair A 1 -A 2 is like before called the disturbed set of modem pair and the modem pair B 1 -B 2 is called the disturbing set of modem pair. The method comprises the following steps:

As mentioned, the sending end modem B 2 transmits data 11 to its receiving end modem B 1 on the wire CH-Y. Neither A 1 nor A 2 is affected by cross-talk due to the long distance from B 2 to both A 1 and A 2 . If the end modem A 1 has the intention to start sending to its receiving end modem A 2 on the same frequency as B 2 is using, i.e. CH 5 , the cross-talk affecting B 1 most likely will be catastrophically strong. This situation is to be avoided. The receiving end modem B 1 now temporarily becomes sending end modem and transmits during a short time interval collision avoidance data CAD in a resource of the frequency spectrum on channel CH 5 in direction towards B 2 which now temporarily has become receiving end modem. As already have been mentioned earlier in the description, also receiving end modems occasionally during short time intervals regularly can become sending end modems. This temporarily change will occur regularly. The end modem A 1 detects cross-talk interference CTI on wire CH-X on the channel CH 5 . The cross-talk interference is caused by transmission of collision avoidance data on wire CH-Y, due to the cross-talk coupling. The end modem A 1 of the disturbed set of modem pair A 1 -A 2 transmits a media control signal CS on the frequency spectrum CH 5 . A possible variation would be that also A 2 transmits a media control signal. The media control signal CS is reproduced from wire CH-X to wire CH-Y and becomes a reproduced media control signal CS′. This due to fact that interference is symmetric in the sense that if there is cross-talk coupling from CH-Y to CH-X, there is also cross-talk coupling in the opposite direction from CH-X to CH-Y. The reproduced media control signal CS′ is detected by the disturbing receiving end which now again is modem B 1 . Like earlier mentioned, different scenarios now can take place depending on if modem pair A 1 -A 2 or modem pair B 1 -B 2 are communicating in the default direction or in a direction opposite the default direction.

FIG. 5 discloses a third embodiment of the invention. The modem pairs already shown in FIG. 4 are also shown in FIG. 5 . A third modem pair C 1 -C 2 can also be seen in the figure. The third modem pair C 1 -C 2 is of the same type as A 1 -A 2 and B 1 -B 2 and is thereby considered to already have been described. It is of course possible to have also other modem pairs in the system but in an attempt to make the figure clearer other modems have been left out. Transmission on media CH-X, CH-Y, CH-Z is in this embodiment divided into timeslots. A first set of timeslots is defined for collision avoidance data and a second set of timeslots for media control signals. Collision avoidance data is then transmitted in the predefined set of timeslots in the frequency spectrum of disturbing transmission medias, by receiving-end modems of so called disturbing set of modem pairs. In this example, end modems B 2 and optionally C 2 are the sending end modems sending regular data, while the end modems B 1 and C 1 are the receiving end modems sending collision avoidance data CAD. The method comprises the following steps:

Collision avoidance data CAD are transmitted at predefined first set of timeslots of the frequency spectrum of the medias CH-Y, CH-Z, in this example by the receiving-end modems B 1 , C 1 . Cross-talk interference CTI is detected at the predefined first set of timeslots on a disturbed media CH-X by the disturbed receiving end modem A 1 . The interference is received from a disturbing media CH-Y used by a disturbing set of modem pair B 1 -B 2 . This due to the sending of collision avoidance data CAD by B 1 and due to a cross-talk coupling T. A media control signals CS is transmitted at a predefined second timeslot of the frequency spectrum by the disturbed receiving end modem A 1 on the disturbed media CH-X. A reproduced media control signal CS′ is detected at the predefined second timeslot on the disturbing media CH-Y by B 1 of the disturbing set of modem pair B 1 -B 2 .

In order to avoid destructive NEXT interference, all involved modem pairs A 1 -A 2 , B 1 -B 2 and C 1 -C 2 in the system that are subject to interference should be synchronized in time. A description of symbol synchronization in a DMT system with cross-talk interference can be found in the European patent application EP 1093248. All embodiments where frequency bands have been divided into sub bands are applicable also when the frequency bands have been divided into timeslots.

FIG. 6 discloses in a fourth embodiment, an example of a situation where a non-disturbing media accidentally is affected by the media control signal. In FIG. 6 , the modem pairs A 1 -A 2 and B 1 -B 2 are shown together with a third modem pair C 1 -C 2 . A 1 and B 1 are within cross-talk range from each other as well as A 1 and C 1 , while B 1 and C 1 are not within cross-talk range from each other. A 1 -A 2 is called the disturbed modem pair. B 1 -B 2 is called the disturbing modem pair, and C 1 -C 2 is called the non-disturbing modem pair.

The situation that now will be discussed follows the earlier shown procedure discussed in the first embodiment and shown in FIG. 2 , i.e. the situation when A 2 transmits data 10 in an upward direction, and B 1 transmits data 11 in a downward direction. A 1 detects cross-talk interference CTI due to the cross-talk coupling T between CH-Y and CH-X and transmits a media control signal CS. However, due to a second cross-talk coupling T 2 between CH-X and a non-disturbing media CH-Z used by the non-disturbing modem pair C 1 -C 2 , the media control signal is also reproduced to CH-Z and becomes a second reproduced media control signal CS″. C 1 hereby assumes that it has disturbed a modem pair—which is not true. In order to avoid this situation, the media control signal carries information about the current transmission direction of the disturbed modem pair A 1 -A 2 that was sending the signal.

In the frequency sub band case, the media control signal carries information about the current transmission direction of the disturbed modem pair by configuration of sub bands in the frequency band in the spectrum. In the time synchronized case, the media control signal instead demonstrates the direction by the way the time slot pattern is configured when sent by the disturbed modem pair. If the indicated transmission direction is the same direction as the direction the modem pair C 1 -C 2 is using, the non-disturbing modem pair C 1 -C 2 will ignore the second reproduced media control signal CS″.

It is to be noted that a possible scenario would be that a reproduced media control signal is detected only on the non-disturbing media but not on the disturbing media. This, if for example the disturbing media for some reason is disturbed or turned off before the reproduced media control signal is detected.

Different variations are of course possible within the scope of the invention. The media control signal can for example be repeated on the disturbed transmission media by the signal sending modem. This is for redundancy purpose and prevents unintentional loss of the signal due to a disturbance before it has been received. In another example, in order to point out to other transmission medias in the communication system that a sudden change of transmission direction is about to take place in the system, a reproduced media control signal CS′, CS″ received by a transmission media is forwarded to at least one other closely related transmission media in the system. The reproduced media control signal is in this way propagated through the communication system.

The invention is in other words not limited to the above described and in the drawings shown embodiments but can be modified within the scope of the enclosed claims.