GB2457011A - Duplexing communications to terminal classes in both time and frequency domains - Google Patents

Abstract

Access to a network is controlled by allocating terminals in communication with the network controller to first and second terminal classes, e.g. by dividing a cell area into virtual inner and outer zones (fig.6), and then performing duplexing on both terminal classes. Terminals in the first terminal class, e.g. inner zone, duplex communication in time domain (TDD) in a first frequency band. Terminals in the second terminal class, e.g. outer zone, duplex communication between a second frequency band (FDD DL) and a time slot of said first frequency band (TDD UL). The duplexing of the first terminal class includes dynamically allocating time slots in the first frequency band based on the requirements for communication to and from the terminals in both of the first and second terminal classes. This arrangement is particularly advantageous for asymmetric applications, i.e. where uplink (UL) and downlink (DL) transmissions are at different data rates, e.g. internet connections or broadcast data.

Description

Wireless Communications Apparatus The present invention is directed to wireless communications and particularly, but not exclusively, to communications between wireless communications apparatus in a network of mobile terminals and base stations.

Successive developments in wireless networking communications have aimed to improve spectral efficiency and network coverage. This is particularly true of cellular networks, where network design involves cost and efficiency based trade offs in planning the positioning and capabilities of wireless communications stations establishing cells in the network.

For those networking technologies known in the art as Next Generation Networking (NON) (which may be in development, already developed and launched, or indeed merely envisaged), and generally referred to as Beyond 3G (B3G) or 4G, the challenging objective has been to achieve high spectral efficiency and macroscopic coverage with guaranteed quality of service (Q0S) for multimedia applications. To strive towards that objective, research and development has focused upon the physical layer (PHY) in the commonly adopted OSI model. This has included, but has not been restricted to, multi-carrier modulation (MCM) schemes, multiple-input multiple-output (MIMO), interference management, advanced channel coding algorithms and cross-layer optimisation. However, little investigation so far has been made into the choice of duplex schemes.

Bandwidth requirements for new networking protocols have increased exponentially from one generation' to the next. For instance, a 2G system might typically have a bandwidth requirement of 30-200 Id-li whereas for a 3G system a bandwidth requirement of 15-20 MHz would be possible. Therefore, a high demand of use of the radio spectrum is created by increasingly sophisticated network service protocols.

There are primarily two types of duplex methods to separate uplink (UL) and downlink (DL) communication signals, namely Time Division Duplex (TDD) and Frequency Division Duplex (FDD). Each has its respective merits and demerits, depending on the particular application for which they are employed. However, as a general rule, TDD has been favoured for low-mobility and smaller coverage areas, whilst FDD has generally been the choice for higher mobility and wider area deployments.

This is because FDD systems traditionally had an interference avoidance advantage over TDD systems in cellular deployment scenarios. FDD systems had additional protection against interference at the base station from neighbouring base stations. The reason for this is that base station transceiver to base station transceiver interference is line of sight, whereas interference from subscriber stations to the base station transceiver is often non-line of sight. Because FDD systems separate transmit from receive paths on separate frequencies, this particular interference problem can be minimised. In contrast, TDD uses spectrum very efficiently, translating to more capacity and users (and hence greater revenue potential). This is because it utilises only one channel, enabling 100% use of the available spectrum, regardless of the actual downstream-to-upstream channel utilisation pattern.

Although FDD is (as a general rule) the favoured choice for cellular systems, interference restrains the performance of the method. Hence, reducing interference in a cellular system is the most effective approach to increasing radio capacity and/or transmission data rate in such a wireless environment.

Code-division duplexing (CDD), as described in "The most spectrum-efficient duplexing system: CDD" (Lee, W.C.Y.; IEEE Comm. Mag., Volume 40, Issue 3, March 2002 Page(s) 163 -166.), emerged as an innovative solution that could eliminate system interference, as long as the codes are chosen from a set of smart codes. Large Area Synchronous (LAS) Codes are smart codes that can reduce interference very effectively. The effectiveness of smart codes applied to TOD makes it the right choice in cellular systems. The application of LAS Codes in a TDD system (called ID-LAS system) creates a CDD system.

Using this approach, any single (unpaired) spectrum band with a bandwidth equal to 1.6 MHz can be used for the CDD application. Implementation of the system is straightforward so a handset implementing the design can be of relatively simple construction. Since no duplexer is needed, the cost and size of the handset can be reduced. Power consumption (a critical design factor for wireless communications handsets) can be lowered due to there being only one single spectrum band operable.

CDD is similar to TDD as shown in figure 1, the ideal system for asymmetrical traffic.

Whereas CDD has proved useful, the present application puts forward proposals making use of other technologies.

Wireless duplexing has been traditionally implemented by dedicating two separate frequency bands -one band for the uplink and one band for the downlink. This arrangement of frequency bands is known as "paired spectrum", and the two bands are separated by a "guard band" which provides isolation of the two signals, as shown in figure 2.

Duplex communications can also be achieved in time rather than by frequency. In this approach, the uplink and the downlink are defined on the same frequency, but they are switched very rapidly. In successive (and very short) time slots the channel sends the uplink signal, and then the downlink signal. Because this switching is performed very rapidly, the impression is given that one channel is acting as both an uplink and a downlink simultaneously. This approach is therefore a favourable choice for unpaired spectrum.

As mentioned, TDD systems require a guard time (instead of a guard band) between transmit and receive streams. The TX/RX Transition Gap (TTG) is a gap between downstream transmission and the upstream transmission. This gap allows time for the base station to switch from transmit mode to receive mode and subscribers to switch from receive mode to transmit mode. During this gap, the base station and subscriber are not transmitting modulated data but are simply allowing the base station transmitter carrier to ramp down, the TX IRX antenna switch to actuate, and the base station receiver section to activate. The TTG has a variable duration that is an integer number of physical time slots (PS). The TTG starts on a PS boundary.

The TTG is equal to the following value: TTG (in seconds) = 2 x (maximum link distance in Kin) / (speed of light) + modem TX/RX transition ITO (in PS) = TTG (seconds) / (4 x symbol rate) Figure 3, which is a schematic diagram of wireless communications between a handset and a base station of a wireless communications network, shows data transmission is symmetric if the data in the downlink and the data in the uplink is transmitted at the same data rate. This will probably be the case for voice transmission -substantially the same amount of data is sent both ways. However, for internet connections or broadcast data (e.g., streaming video), it is likely that more data will be transmitted in the downlink direction, and hence these are considered to be asymmetric.

FDD transmission is not well suited to asymmetric applications as it uses equal frequency bands for the uplink and the downlink, resulting in a waste of valuable spectrum. On the other hand, TDD does not have this fixed structure, and its flexible bandwidth allocation is more appropriately suited to asymmetric applications, such as internet connection. For example, TDD can be configured to provide 3 84kbps for the downlink (the direction of the majority of data transfer), and 64kbps for the uplink (where the traffic largely comprises requests for information and acknowledgements).

US Patent Application US2006/O 126546 illustrates a number of time/frequency configurations intended to provide duplexing of uplink arid downlink activity. As illustrated in figure 2A of that document, FDD UL and FDD DL bands are provided across a time slot and, within the same timeslot, a further band is allocated at a further frequency band, this further band being itself duplexed between TDD DL and TDD UL bands. This is therefore a hybrid approach, using both FDD and TDD. In this approach, a macro cell is defined and within the macro cell a plurality of micro cells are

S

defined. The macro cell uses the FDD resources for uplink and downlink, while the micro cells use the FDD UL resource together with the additional TDD DL and UL resources provided by the further band.

Aspects of the present invention aim to provide advantages of both TDD and FDD schemes and to increase flexibility and efficiency of mobile communication systems.

An aspect of the invention provides a wireless communication controller operable to control access to a network by one or more networking terminals, said controller comprising means for defining in an available frequency spectrum first and bandwidth portions for communication, time domain duplexing means for allocating timeslots in said first bandwidth portion for time domain duplexing between establishment of uplink communication from a terminal to said controller and establishment of downlink communication to a terminal from said controller, and means for establishing downlink communication in said second bandwidth portion, wherein said time domain duplexing means is operable to determine the duration of said timeslots on the basis of predicted, anticipated or requested traffic requirements.

An aspect of the invention provides a wireless communication controller operable to control access to a network by one or more networking terminals, said controller comprising means for defining in an available frequency spectrum first and second bandwidth portions for communication, frequency domain duplexing means operable to allocate said first bandwidth portion into first and second sub-portions, said allocation being on the basis of traffic requirements, said second sub-portion being allocated for downlink communication to a terminal from said controller, time domain duplexing means for allocating timeslots in said first sub-portion for time domain duplexing between establishment of uplink communication from a terminal to said controller and establishment of downlink communication to a terminal from said controller, and means for establishing downlink communication in said second bandwidth portion, wherein said time domain duplexing means is operable to determine the duration of said timeslots on the basis of predicted, anticipated or requested traffic requirements.

An aspect of the invention provides a method of controlling access to a network by one or more terminals networking with a network controller, said method comprising: allocating terminals in communication with said network controller to first and second terminal classes; firstly duplexing communication with terminals in said first terminal class in time domain in a first frequency band; secondly duplexing communication with terminals in said second terminal class between a second frequency band and a time slot of said first frequency band; wherein said first duplexing includes dynamically allocating time slots in said first frequency band based on the basis of requirements for communication to and from said terminals in connection with both of said first and second duplexing.

An aspect of the invention provides an opportunistic duplexing system aiming to combine advantages of both TDD and FDD schemes and to increase flexibility and efficiency of mobile communication systems.

An aspect of the invention provides, for multilevel Quality of Service (QoS) requirements, dynamic adjustment of bandwidth allocation to users according to the underlying network condition so as to increase bandwidth utilization.

An aspect of the invention provides a controller operable to alternating band allocation to achieves reciprocity; by such an approach the channel can be estimated in each band when it is used for uplink and then exploited when it is to be used for downlink.

An aspect of the invention provides a method of scheduling communication in a wireless communication network, the method being subject to a flexibility metric, as an additional parameter to QoS parameter.

An aspect of the invention provides access control to a network by one or more networking terminals comprises allocating terminals in communication with said network controller to first and second terminal classes and then performing duplexing on both terminal classes. Terminals in the first terminal class duplex communication in time domain in a first frequency band. Terminals in the second terminal class duplex communication between a second frequency band and a time slot of said first frequency band. The duplexing of the first terminal class includes dynamically allocating time slots in the first frequency band based on the basis of requirements for communication to and from said terminals in connection with both of the first and second duplexing.

An aspect of the invention provides, in accordance with any of the foregoing, computer program means operable to configure suitable general purpose computer equipment to operate in accordance with the invention. The computer program means can be provided in the form of a product, such as storage medium (e.g. an optical disk), an electronic means such as a dedicated integrated circuit (ASIC, FPGA etc.) or a downloadable product.

Specific embodiments of the invention will now be described with reference to the accompanying drawings, in which: Figure 1 illustrates, for background purposes, forms of duplexing in use in examples of

the prior art;

Figure 2 illustrates, schematically, a form of frequency domain duplexing in accordance

with a prior art example;

Figure 3 illustrates, schematically, examples of symmetric and asymmetric traffic in a Figure 4 illustrates a hybrid duplexing arrangement in accordance with a first embodiment of the invention; Figure 5 illustrates time domain allocation of a frame in accordance with the first embodiment of the invention; Figure 6 illustrates allocation of a cell of an embodiment of the invention to inner and outer zones; Figure 7 illustrates a hybrid duplexing arrangement in accordance with a second embodiment of the invention; Figure 8 illustrates a hybrid duplexing arrangement in accordance with a third embodiment of the invention; Figure 9 illustrates a terminal access request prioritisation method in accordance with a fourth embodiment of the invention; Figure 10 illustrates a user priority database for use in the method of figure 9; and Figure 11 illustrates a time domain allocation of a frame in accordance with use of the method of the fourth embodiment of the invention.

Figure 4 illustrates the nature of the specific embodiment of the invention in terms of a an arrangement of uplink and downlink allocations in a time/frequency domain. The duplexing system in accordance with the specific embodiment defines two frequency bands; one frequency band is allocated for both uplink and downlink communications, which is identical to TDD mode, and the other band is devoted to downlink communication only. In the illustrated example, the latter band is higher than the former, but the reader will understand that this is not essential.

The two frequency bands are denoted as the TDD band and the FDD DL band respectively, as shown in figure 4. Uplink communication is available only in the TDD UL, while downlink traffic can be sent in either of the illustrated TDD DL or FDD DL bands. In the system, based on the location of terminals (for instance, mobile devices), the cell area and terminals are divided into two viriual groups or zones. Conceptually, a terminal in the virtual inner and virtual outer zones are configured to send uplink signals in the TDD UL band while those in the virtual outer zone are configured to communicate with the base station in the TDD UL and FDD DL bands. As a result, a terminal at the virtual inner zone communicates in TDD mode while a terminal in the virtual outer zone communicates in FDD mode. As the terminal moves inwards or outwards within a cell, it needs to change from FDD mode to TDD mode or vice versa.

This is further illustrated in figure 5, in which successive time slots of a frame are shown, across five of which time slots a TDD DL band is defined, followed by a TDD UL band over two successive time slots, and a TDD UL (outer zone) band over a further two successive time slots. The bands are separated from each other in time by relatively short guard intervals. Although the durations of the bands are shown as such, these durations are adaptive to the needs of the system, and particularly the asymmetry of communication. The maimer in which the band durations can be adapted will be described in due course.

As illustrated in figure 6, a cell of a cellular network is shown with a base station (BS) at its centre. With regard to the transmission/reception profile of the BS, and the presence of other (not illustrated) neighbouring base stations defining their own cells, an outer boundary of the cell is nominally hexagonal. An inner virtual zone of the cell is defined, outside which an outer zone of the cell is defined.

A mobile terminal can be identified as being a member of a virtual group. In the illustrated example, two virtual groups are defined, corresponding to the inner and outer virtual zones as indicated. Membership of one or other virtual group can be based on one or more of the following information available at the BS: * Location information of users; * Received power; and * Received signal to noise ratio (SNR) at the terminals.

In one configuration of the embodiment, all of the above are factors in designating membership of a virtual group by a terminal.

The decision of the virtual grouping of the users is also secondarily dependent on: * Quality of Service (QoS) requirements; and * Data rate requirements.

Thus, while the virtual groups are nominally in correspondence with the inner and outer virtual zones, this is not a strict correspondence and other factors may influence the allocation of a terminal to a particular virtual group.

It has been forecasted, for example within the UMTS Forum, that future wireless systems will operate with a 2.3:1 ratio of downlink activity to uplink activity, to enable video streaming, interactive web browsing and so forth. However, it will be understood that this will depend largely on emerging applications and standards. Hence, to meet these requirements, the present embodiment employs TDD DL and FDD DL for downlink communication, while uplink traffic can be sent in either of the two TDD UL bands.

By allocating the entire FDD DL band for terminals in the outer cell, or otherwise designated as members of the outer virtual group, the system will be able to meet the Q0S requirements for edge of cell terminals with requests for continuous transmission.

In a variation on the above embodiment, to increase spectral efficiency, the bandwidth can be redistributed between UL and DL to optimise QoS provision further. Although the size of the frequency spectrum physically limits the capacity of the radio networks, effective solutions can optimise usage of available capacity and resources. As mentioned earlier, in FDD, the UL and DL channels are assigned a fixed bandwidth and hence the bandwidth cannot be transferred between the two links, even though the data rate requirements on the DL will generally be more than the UL requirements.

In a network supporting multilevel QoS, bandwidth allocation to users should be adjusted dynamically according to the underlying network condition so as to increase bandwidth utilization and service provider's revenue. Generally, the minimum band separation between the UL and DL can be smaller at the BS than at the MS, because a 13S will most likely comprise a more sophisticated and therefore sharper RF filter.

When the demand for DL transmission increases substantially, and rapidly while that for UL traffic remains low, less time and/or bandwidth can be allocated to UL and more to DL to accommodate this surge in traffic. This can also maintain the quality of delay sensitive service by instantaneously lending resources from one direction to another, as shown in figures 5 and 7. For example, when a large DL packet urgently needs to be delivered to the receiver, an additional time slot or fractional bandwidth can be added to the DL subframe or bandwidth to accommodate it.

To address the need for various applications and usage model requirements in a flexible manner, there needs to be a support for a wide range of frame sizes. Table 1 below sets out Mobile WiMax parameters as an example:

Table 1

Parameters Values System Channel Bandwidth (MHz) 1.25 5 10 20 Sampling Frequency (F in MHz) 1.4 5.6 11.2 22.4 FFT Size (NFjrr) 128 512 1024 2048 Number of Sub-Channels 2 8 16 32 Sub-Carrier Frequency Spacing 10.94 kHz Useful Symbol Time (Tb=lIf) 91.4 s Guard Time (T8Tb/8) 11.4 ItS OFDMA Symbol Duration (T5Tb+Tg) 102.9 p.S Taking, for example, a 2048 FFT size, the number of OFDM symbols in the short frame size, (for example 2 ms), will be very small for narrow bandwidths (less than 2 OFDM symbols for 1.25 MHz band) which makes the short frame sizes practically unusable due to high overhead. One of the advantages of flexible bandwidth allocation is to guarantee a lower bound on the number of OFDM symbols per frame, which is particularly a problem for small bandwidth and frame sizes. Without dynamically allocating bandwidth with regard to traffic demand, performance will be reduced and/or implementation cost will be increased for low-size channel bandwidths.

Furthermore, if the modified approach described above is further modified as follows, the two duplexing schemes so described can be combined in a more effective manner.

Viewing the abovementioned scheme from a macroscopic perspective, effectively the scheme has a pair of spectrum blocks that have been re-assigned into a combination of TDD and FDD bands. However, instead of reserving a block for TDD UL & DL, and the other one for FDD DL, their use is alternated every T sec, as shown in figure 8.

This scheme will thus be able to achieve reciprocity and the channel can be estimated in each band when it is used for uplink and then exploited when it is to be used for downlink.

For this to operate effectively, the wireless channel must itself be reciprocal. However, if the mobile terminal moves faster than the rate at which the channel is updated, time variation will destroy the channel reciprocity. Hence, to maintain reciprocity, time variation must be negligible while the channel coefficients are estimated in the up-link and the precoded data are then transmitted in the down-link.

Low mobility can be realised by a simple change in the burst structure, where the pilot signals for the channel estimation are initially transmitted from the MS to the BS. In contrast, at higher mobility, changes in the link direction must be realised very quickly.

The challenge is then to continue to identify the signals, to estimate the channel with good quality and to transmit a useful amount of data in a fraction of the channel coherence time.

However, these requirements may not be substantialLy stricter than in conventional wireless systems which use the channel information only at the receiver. The pilot signals need only be transmitted from the terminal towards the base station, either in the up-link (as in the conventional system) or prior to the down-link, and the data being received within the coherence time. Such a roundtrip time (RTT) needs two times the propagation delay.

lfRTT<T Enable switching else Disable switching The following further additional operational features of the previously described embodiments provide a manner of allocating available radio resources to user terminals, such that their requirements are fulfilled and, at the same time, the time and frequency resources are used with the maximum possible efficiency. A scenario is also presented for the MS and BS negotiation on QoS before the service can be offered to the user terminal.

Each user terminal first submits its requested QoS profile to the network. The requested QoS profile for the nth user is defined as rQoS=[rR b, rBER] , where rRb is the requested bit rate and rBER shows the maximum acceptable bit error rate. In other words, a QoS profile describes the data throughput and the reliability requirements for the specified user terminal. The network then reviews all submitted profiles, considering the extent of existing resources, and returns an offered QoS profile, oQoS, together with a cost of service, CoS, to each user. If the user terminal accepts one of these oQoS profiles, a "contract" will be signed between the user terminal and the network, and the communication begins.

Scheduling user transmissions in time can be used to differentiate between service classes or to increase fairness among users. If some services do not require an instantaneous data rate but rather an average rate, i.e. a quantity of data is delivered over a certain time interval, then more flexible use of the spectrum can be considered. intra-cell interference can also be efficiently mitigated by utilizing the possibility of scheduling the transmissions within the cell, in addition to established power control techniques. Scheduling is highly related to rate control, and could be considered a special case of the latter, since setting the data rate and power to zero can be regarded equivalent to scheduling. For a highly asymmetric traffic (downlink dominated) scenario, scheduling dowrilinlc traffic locally in each cell could be a simple and attractive candidate scheme for supporting non-real time data services.

A suitable algorithm for scheduling in accordance with the specific embodiments of the invention set out above will now be described, with reference to figures 9 to 11 of the drawings.

In step SI, the various user requests are prioritised on the basis of Q0S requirements.

This prioritisation is stored in a user priority database 100 as illustrated in figure 10.

Then, in step S2, the Coherence time T is calculated for each of the active users. Table 2 sets out an example of the coherence times at varying velocities for a 3.5 GHz system: Velocity (km/hr) fm (Hz) Coherence Time (sec) 3 9.722 1.84E-02 162.037 1.1OE-03 324.074 5.52E-04 648.148 2.76E-04 In conventional scheduling algorithms, the highest priority user, for example MS1, is scheduled first for DL transmission (MS, -> Q0S_POl), followed by subsequent lower priority users. However, in accordance with the presently described algorithm, the following additional metrics are used for setting bounds on scheduling, which as well as adding flexibility, will also increase the efficiency and capacity of the system.

Therefore, the difference, Tm, between the start of the TDD UL slot and the end of the TDD DL slot, is calculated. Moreover, the difference, between the end of the TDD UL slot and the start of the TDD DL slot, is determined.

In step S3, if Tm < Ta,, then the user is assigned a higher flexibility level, implying that a user with the highest QoS requirement might be assigned a lower priority whilst scheduling. This is stored in the user priority database 100.

If there are multiple users that meet the aforementioned condition, users with a higher QoS requirement will be prioritised.

In step S4, if Tm> T1, then Tm,,, is calculated. If Tm1,, < Ta,, then the priority of MS, is reassigned (in step S5) and scheduled for transmission (ensuring it is within the T) so as to accommodate higher mobility users, that can be scheduled for transmission at this location with lower value of coherence time. This is indicated in figure 10 by means of arrows showing reassignment of scheduling to user terminals based on this flexibility criterion.

If Tnrn> T, then the TDD scheme fails and the scheme reverts back to FDD. The process is repeated for all user terminals.

So effectively, the user terminal with the highest flexibility can be scheduled last, whilst the user terminal with lowest flexibility is scheduled first, without compromising their respective QoS requirements. However, it is worth noting here that despite this scheduling being based on a flexibility metric, it does also take into account the QoS provisions at the start of the process.

The foregoing description should be read as an example of implementation of the invention, and should not be read as being prescriptive of an particular features. The reader will understand that the spirit and scope of the invention is to be read in accordance with the accompanying claims appended hereto.

Claims (20)

1. CLAIMS: 1. A wireless communication controller operable to control access to a network by one or more networking terminals, said controller comprising: bandwidth allocation means operable to allocate terminals in communication with said controller to first and second terminal classes; first duplexing means operable to duplex communication with terminals in said first terminal class in time domain in a first frequency band; second duplexing means operable to duplex communication with terminals in said second terminal class between a second frequency band and a time slot of said first frequency band; wherein said first duplexing means is operable to allocate dynamically time slots in said first frequency band based on the basis of requirements for communication to and from said terminals by both of said first and second duplexing means.
2. 2. A controller in accordance with claim 1 wherein said first duplexing means is operable to allocate a first time slot for down.link communication with a terminal in said first terminal class, a second time slot for uplink communication with a terminal in said first terminal class, and to reserve a third time slot for use by said second duplexing means.
3. 3. A controller in accordance with claim 1 or claim 2 wherein said second duplexing means is further operable to allocate, adaptively, a band portion of said first frequency band, for duplexing communication with terminals in said second class, said band portion allocation overriding said first duplexing means.
4. 4. A controller in accordance with claim 3 wherein said allocated band portion of said first frequency band is allocated for downlink communication with a terminal of said second terminal class.
5. 5. A controller in accordance with any preceding claim wherein said time slot of said first frequency band allocated for duplexed communication with a terminal of said second terminal class is allocated for uplink communication from said terminal.
6. 6. A controller in accordance with any preceding claim wherein said second frequency band is allocated for downlinjc communication with a terminal of said second terminal class.
7. 7. A controller in accordance with any preceding claim wherein said bandwidth allocation means is operable to allocate terminals to one of said terminal classes on the basis of one or more of: Location information of a terminal; Received power at a terminal or at said controller; and Received signal to noise ratio (SNR) at a terminals.
8. 8. A controller in accordance with claim 7 wherein said first terminal class comprises terminals which, having regard to said one or more criteria, have more robust communication characteristics with said controller than terminals of said second class.
9. 9. A controller in accordance with any preceding claim operable to alternate allocation of said first and second frequency bands.
10. 10. A controller in accordance with any one of the preceding claims and further comprising terminal channel access allocation means operable to determine, on the basis of received request information from terminals in said network, the relative quality of service requirements of said terminals, to prioritise said terminals for access to said channel on the basis of said relative quality of service requirements, and to modi& said prioritisatjon on the basis of the extent to which a terminal channel access requirement is flexible.
11. 11. A controller in accordance with claim 10 wherein said received request information comprises information defining the nature of data to be transmitted to said terminal and wherein said controller is operable to determine a flexibility of said terminal channel access requirement on the basis of said data nature information.
12. 12. A method of controlling access to a network by one or more terminals networking with a network controller, said method comprising: allocating terminals in communication with said network controller to first and second terminal classes; firstly duplexing communication with terminals in said first terminal class in time domain in a first frequency band; secondly duplexing communication with terminals in said second terminal class between a second frequency band and a time slot of said first frequency band; wherein said first duplexing includes dynamically allocating time slots in said first frequency band based on the basis of requirements for communication to and from said terminals in connection with both of said first and second duplexing.
13. 13. A method in accordance with claim 12 wherein said first duplexing comprises allocating a first time slot for downlink communication with a terminal in said first terminal class, a second time slot for uplink communication with a terminal in said first terminal class, and reserving a third time slot for use in said second duplexing.
14. 14. A method in accordance with claim 12 or claim 13 wherein said second duplexing further comprises allocating, adaptively, a band portion of said first frequency band, for duplexing communication with terminals in said second class, said band portion allocation overriding said first duplexing.
15. 15. A method in accordance with claim 14 wherein said allocated band portion of said first frequency band is allocated for downlink communication with a terminal of said second terminal class.
16. 16. A method in accordance with any one of claims 12 to 15 wherein said time slot of said first frequency band allocated for duplexed communication with a terminal of said second terminal class is allocated for uplink communication from said terminal.
17. 17. A method in accordance with any one of claims 12 to 16 wherein said second frequency band is allocated for downlink communication with a terminal of said second terminal class.
18. 18. A method in accordance with any one of claims 12 to 17 wherein said allocating of bandwidth comprises allocating terminals to one of said terminal classes on the basis of one or more of: Location information of a terminal; Received power at a terminal or at said controller; and Received signal to noise ratio (SNR) at a terminals.
19. 19. A computer program product comprising processor executable in.tructions to configure a network controller as a controller in accordance with any one of claims I to 11.
20. 20. A computer program product comprising processor executable instructions which, when executed, cause a computer to perform the method of any one of claims 12 to 18.