WO2018210462A1

WO2018210462A1 - Transmission power distribution

Info

Publication number: WO2018210462A1
Application number: PCT/EP2018/055569
Authority: WO
Inventors: Pål FRENGER; Erik G LARSSON; Emil BJÖRNSON
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2017-05-19
Filing date: 2018-03-07
Publication date: 2018-11-22

Abstract

A method (200) of operating an access point (A j ) of a distributed MIMO system is disclosed. The distributed MIMO system serves a number (N) of wireless devices (u 1,...,u N ).The method comprises obtaining (220), for each wireless device (u k ), a channel quality metric (β k,j ) indicative of a channel quality between the access point (A j ) and the wireless device (u k ). The channel quality may e.g. be an SNR or a path loss. The method further comprises determining (230), based on priority parameters (p 1 ,...,PN) associated with the wireless devices ( Ui >...,U N )and the channel quality metrics (β k, j ), a distribution of transmission power from the access point to the different wireless devices(u 1 ,...,u N ).The priority parameter (pk) associated with a wireless device (Uk) is indicative of a priority of an upcoming transmission from the distributed MIMO system to that wireless device (Uk).

Description

TRANSMISSION POWER DISTRIBUTION Technical field

The present disclosure relates to distribution of transmission power in access points, such as in access points of distributed MIMO (Multiple-Input Multiple-Output) systems. Background

Conventional cellular networks consist of a set of base stations equipped with an array of co-located antenna elements, each forming one or multiple antenna ports. When a user has data packages to receive in the downlink, it is first associated with one of the base stations and then it is scheduled for transmission on a block of time- frequency resources. In these resource blocks, the serving base station array forms a beam towards the user terminal, with a spatial signature that is selected based on the spatial position of the user and co-users that are active in the same block. The beam is typically selected to balance between high received signal power at the user and little interference towards the co-users. Power control is also important to limit interference between users. With co-located antenna elements, the path gain to a given user is roughly the same for all antenna elements, thus the same power control coefficient can be used for all antenna elements.

As the cellular networks are densified, the inter-cell interference become a major issue and the topology of the network may have to be changed; a conventional cellular architecture with co-located antennas is not necessarily optimal. Cell- free massive MIMO (also known as "distributed antenna system" or "distributed massive MIMO") is a technology for this situation. In this concept, many physically separated access points are deployed within a conventional cell and there might be no explicit cell boundaries. Each user is served by a subset of such access points, typically the ones that have a sufficiently high SNR (Signal-to- Noise Ratio) to the user. That subset is partially overlapping between neighboring users and thus the access points cannot always be divided into disjoint sets, as is the case in

conventional cellular networks. Two benefits of cell-free massive MIMO is that interference can be coordinated over the entire network, to avoid inter-cell interference, and an increased level of macro-diversity helps to mitigate shadow fading.

Existing solutions for scheduling and power allocation are inefficient or non-applicable in this scenario:

1) The cell association conventionally associates the user with a predefined set of access points. In a cell- free system, in contrast, where the access points in the network cannot be divided into disjoint clusters, the cell association is fundamentally different and a mechanism to associate users with individual antennas is required.

2) The scheduling cannot be made on a per-cell basis, when there are no disjoint cells.

3) Since the access point antennas that serve a given user can have very different path gains to the user, it may be unsuitable to use the same power control coefficient for all antennas.

Summary

Embodiments herein aim at providing transmission-power distribution suitable for distributed MIMO systems.

According to a first aspect, there is provided a method of operating an access point of a distributed MIMO system. The distributed MIMO system serves a number of wireless devices. The method comprises obtaining, for each wireless device, a channel quality metric indicative of a channel quality between the access point and the wireless device. The channel quality may e.g. be an SNR or a path loss. The method further comprises determining, based on priority parameters associated with the wireless devices and the channel quality metrics, a distribution of transmission power from the access point to the different wireless devices. The priority parameter associated with a wireless device is indicative of a priority of an upcoming transmission from the distributed MIMO system to that wireless device.

The method may comprise obtaining, for each wireless device, said priority parameter from a central unit of the distributed MIMO system.

The step of determining may comprise taking a decision locally in the access point on the distribution of the transmission power. Taking said decision may comprise maximizing a utility function, which is a function of the distribution of transmission power, the priority parameters, and the channel quality metrics.

In some embodiments, the method comprises receiving, from each wireless device, a pilot signal transmission. The method may comprise determining the channel quality metric based on the received pilot signal transmission.

According to an exemplary embodiment, the determining step comprises sending the channel quality metrics to a central unit of the distributed MIMO system and receiving a decision on the distribution of transmission power from the central unit of the distributed MIMO system.

According to a second aspect, there is provided a method of operating a central unit of a distributed MIMO system. The distributed MIMO system comprises a plurality of access points and said central unit. The access points are configured to operate according to the above exemplary embodiment of the method of the first aspect. The method of the second aspect comprises receiving the channel quality metrics from the access points of the distributed MIMO system, taking decisions on the distributions of transmission power for each of the access points, and sending, to each of the access points, the respective decision on distribution of transmission power. The step of taking a decision may comprise maximizing a utility function, which is a function of the distributions of transmission power, the priority parameters, and the channel quality metrics.

According to a third aspect, there is provided a method of operating a distributed MIMO system comprising a plurality of access points and a central unit. The method comprises performing, in each of the access points, the method of the first aspect. The method may comprise sending, from the central unit to each of the access points, the priority parameters.

According to a fourth aspect, there is provided a method of operating a distributed MIMO system comprising a plurality of access points and a central unit. The method comprises performing, in each of the access points, the method of the above exemplary embodiment of the first aspect. The method further comprises performing, in the central unit, the method of the second aspect.

According to a fifth aspect, there is provided an access point for a distributed MIMO system. The access point comprises a transceiver for communicating wirelessly with wireless devices. Furthermore, the access point comprises a processing circuit configured to perform the method of the first aspect.

According to a sixth aspect, there is provided a cluster comprising a plurality of access points of the fifth aspect.

According to a seventh aspect, there is provided a distributed MIMO system comprising the cluster of the sixth aspect and a central unit. The central unit may be configured to send the priority parameters to each of the access points in the cluster.

According to an eighth aspect, there is provided a central unit for a distributed MIMO system. The central unit comprises a communication interface for communicating with access points of the distributed MIMO system. Furthermore, the central unit comprises a processing circuit configured to perform the method of the second aspect.

According to a ninth aspect, there is provided a computer program product comprising computer program code for executing the method of the first aspect when said computer program code is executed by a programmable processing circuit of the access point. According to a tenth aspect, there is provided a computer program product comprising computer program code for executing the method of the second aspect when said computer program code is executed by a programmable processing circuit of the central unit.

According to an eleventh aspect, there is provided a computer readable medium, such as a non-transitory computer-readable medium, having stored thereon the computer program product of the ninth or the tenth aspect.

Brief Description of the Drawings

Figs. 1-3 illustrate an example.

Figs. 4-6 show block diagrams.

Figs. 7-10 show flowcharts.

Figs. 11-12 schematically illustrate computer-readable media and processing circuits.

Detailed Description

In some scenarios of this disclosure, each access point obtains a list of the users that have data packets to receive. The list contains a priority indicator of the packet. Each access point also obtains a path gain indicator for the user's channel. Each access point independently selects a subset of the users that it will serve (transmit to) and divides its power between these users in dependence of said priority indicators and path gain indicators. The selection is thus made to maximize a local utility function that takes the priorities and path gains into account. Each access point acquires the data packets that it will transmit and sends them at the selected power level. The transmission is carried out phase-coherently among the access points.

The term "access point" is used in this disclosure. Sometimes, "antenna" or "antenna element" is used with the same meaning in the field of MIMO transmissions.

In some embodiments, time is slotted. In each slot a subset of the users having data packages are scheduled for transmission. There are at least two access points that can send the data packages. Each access point independently acquires information about the packages, which includes a priority indicator, and information regarding the channels of the

corresponding users, which includes the path gain. The priority of the packets is typically obtained by a central entity, or central unit, while the path gain for each user may be determined independently by each access point.

The priority indicator can be determined based on service slice information, user priority, event priority, service type, package size, latency requirement, reliability requirement, etc. The path gain can be an average or instantaneous measurement of the channel quality. Each access point has a local utility function that depends on the priority indicator, path gain of all users, and the transmit power that it allocates to the users. Each access point independently maximizes its utility function by selection of the transmit powers. Users that are allocated non-zero powers, or a power larger than a threshold, are considered served by the access point. Users which, according to said utility function optimization should be given a power level lower than a threshold may be given zero power in order to reduce the complexity in the distributing information to be transmitted to the network nodes.

The access point acquires the data packages for all users that it serves through the backhaul network. It then transmits the data packages according to the power level it has selected. Each access point might simultaneously send multiple data packages to multiple users using different pre-coder (phase and amplitude scaling) values for each data package.

In one embodiment, the priority of user k is measured by numerical value p_k between zero and one, where a larger value implies higher priority. Each access point has a power budget and the jth access point assigns a fraction _{k J}- > 0 to user k, under the power budget constraint∑_{fe k J}-≤ 1. One possible utilit function is the priority- weighted sum rate

(1) where /?_fe > 0 is the SNR between user k and the jth access point. This utility function is maximized by a weighted water filling power allocation. Note that the priority values for user k are denoted p_k in (1) indicating that for some access points j this value may be set to 0. Normally the access points that considers serving user k would use the same priority value P_k ⁼ P_k - Determining if an access point j shall consider serving user k can e.g. be based on whether the corresponding /?_fc -value is greater than a threshold.

For example, consider the setup of Fig. 1, with two users and eight access points. As a realistic elucidating example, the SNR between a user and an access point is proportional to the inverse of the squared distance. In Fig. 1, the distance units have been normalized such that a distance of 1 unit distance corresponds to an SNR of 1. The setup in Fig. 1 is merely an example used for illustration. In an actual implementation, the SNR values would normally be measured quantities.

If user 1 has priority p = 0.7 and user 2 has priority p₂ = 0.5, then by maximizing the utility function independently at each access point the resulting power allocation becomes as shown in Fig. 2. With this utility function, user 1 gets all the power from the 4 closest antennas while user 2 gets all the power from the 3 closest antennas. There is also one antenna that transmits to both users. Due to the higher priority, a larger amount of power is allocated to user 1. This distributed power allocation can be compared with the baseline scheme that allocates equal power allocation to both users, since they have both been scheduled.

The resulting rates for the two users using this example are illustrated in Fig. 3.

As can readily observed from Fig. 3, both users benefit from the optimized distributed power allocation, as compared to equal power allocation, but the user with high priority gets a higher rate since more power is allocated to this user.

It should be noted that Eq. (1) is merely an example of a utility function. Other alternatives could be to define a utility function as e.g. a sum of the _{k J}^_{k J}- values from all access points j in (1) before applying the log function; a priority- weighted average SINR; maximizing the number of users with a priority higher than a threshold that can achieve a certain rate;

maximizing a priority- weighted average user rate, etc. Any utility function that considers both the channel and the user-priorities ( ^ ) and calculates a system performance metric (e.g. percentile user rate, average user rate, average or median SNR or SNIR, sum capacity, etc) may be used.

In another embodiment, the access point adapts its decision based on decisions in previous time slots. The access point obtains indicators of the result of previous decisions, which may include HARQ (hybrid automatic repeat request) feedback and information about the decisions taken by other access points. Without loss of generality this can be embedded in a priority value that is dynamically updated in each transmission time interval.

In another embodiment, the user mobility level is obtained and the utility function depends also on this variable. The access point might be more likely to transmit to users that have high mobility, to facilitate handover between access points in the network.

Description of block diagrams

Fig. 4 is a block diagram of a distributed MIMO system according to an embodiment. The distributed MIMO system may e.g. be a distributed massive MIMO system. It comprises a plurality of access points A , ... , A_K. These access points A , ... , A_K correspond to the

"antennas" described above in the context of Fig. 1. The plurality of access points A_lt ... , A_K form a cluster 50 of access points. The distributed MIMO system further comprises a central unit 100. The central unit 100 provides backhaul, implement functionalities in higher layer protocols (TCP/IP, PDCP, RLC, MAC) and may also perform a large part of the base-band physical layer processing such as channel coding and decoding, modulation, etc. The central unit 100 may also coordinate calculations that are performed relatively seldom; such as determining which access points that should serve which users; ensure that the nodes are properly calibrated and synchronized; assign pilots to users to be used for channel estimation; make handover decisions to other central units in the vicinity; etc.

In the example shown in Fig. 4, the distributed MIMO system serves a number N of wireless devices u , ... , u_N, such as mobile phones or other types of UEs (User Equipment). These wireless devices correspond to the "users" described above in the context of Fig. 1. Upcoming transmissions to the different wireless devices u , ... , u_N may have different priorities. In this disclosure, a priority parameter p_k (indicated in Fig. 4) is associated with a wireless device u_k and is indicative of a priority of an upcoming transmission from the distributed MIMO system to that wireless device u_k. Note that the priorities, and thus the priority parameter values, may change over time. In many cases the priorities are determined on what event in the user equipment that triggered the communication. For example, a transmission related to a user may be given a high priority if it is related to a channel measurement or a control signal message and it may be given a lower priority if it is related to a best effort data service. Real time services may have time-dependent priorities in the sense that a packet priority depends on how much time it is left before the packet becomes outdated. Re-transmissions may be given higher priorities than transmissions of new data packet, etc. Priority values for the same user may be different for uplink and downlink transmissions. In case the user priorities change slowly the access points may be provided the user priority values periodically on a time scale slower than the transmission time interval. To enable fast updates of user priority values the priority values may alternatively be provided with the data to be transmitted in each transmission time interval.

Furthermore, the communication channel, or communication path, between a given access point A_j and a given wireless device u_k has a certain channel quality, such as path loss, SNR, or the like. In this disclosure, a channel quality metric is denoted /?_fe and is indicative of a channel quality between the access point A_j and the wireless device u_k. The channel quality metric /?_fe may e.g. be a numerical value of the path loss or SNR.

Fig. 5 is a simplified block diagram of an access point A_j according to an embodiment. It comprises a transceiver a transceiver 52 configured to communicate wirelessly with wireless devices u_k. Furthermore, it comprises a processing circuit 54 configured to perform a method 200 described below with various embodiments. The processing circuit 54 may be a programmable processing circuit 54, and may e.g. comprise a processor 56 and memory 58. The memory 58 may e.g. store computer program code executable by the processor 56. The access point A_j may also comprise a communication interface 55 configured to communicate with the central unit 100, as indicated in Fig. 5.

Fig. 6 is a simplified block diagram of the central unit 100 according to an embodiment. It comprises a communication interface 102 configured to communicate with access points A_lt ... , A_K of the distributed MIMO system, e.g. through the communication interface 55 (Fig. 5). Furthermore, it comprises a processing circuit 104. The processing circuit may be configured to control the access points A , ... , A_K in various ways. For example, it may be configured to provide data to be transmitted to the wireless devices u , ... , u_N. It may be configured to send the priority parameters p , ... , p_N to each of the access points A , ... , A_K . It may be configured to perform a method 300 described below. The processing circuit 104 may be a programmable processing circuit 104, and may e.g. comprise a processor 106 and memory 108. The memory 108 may e.g. store computer program code executable by the processor 106.

Description of flow charts

According to some embodiments, there is provided a method 200 of operating an access point A_j. Flowcharts illustrating some such embodiments are shown in Figs. 7-9.

According to the embodiment illustrated in Fig. 7, operation of the method 200 is started in step 210. Step 220 comprises obtaining, for each wireless device u_k, the channel quality metric /?_fe (the index j is used here to identify the access point among all access points of the distributed MIMO system). Step 230 comprises determining, based on (or "in response to") the priority parameters p , ... , p_N associated with the wireless devices u , ... , u_N, and the channel quality metrics /?_fe , a distribution of transmission power from the access point to the different wireless devices u_lt ... , u_N. For example, the access point A_j may have a total transmission power P_tot to be distributed among the upcoming (concurrent) transmissions to the different wireless devices u , ... , u_N. Step 230 may comprise determining fractions

, ... , _N of the total power P_tot to be assigned to, or used for, the transmissions to the different wireless devices u , ... , u_N. A boundary condition of∑_fe a_k = 1 applies in some embodiments. A boundary condition of∑_{fe k} < 1 applies in some embodiments. The operation is then ended in step 240. When the transmission power distribution has been determined, the access point A_j then uses this distribution of the transmission power when transmitting to the different wireless devices u_k. As illustrated in Fig. 8, the method 200 may additionally comprise a step 215 of obtaining, for each wireless device u_k, said priority parameter p_k from the central unit 100.

Step 220 may e.g. comprise receiving a pilot signal transmission (or similar reference signal transmission) from each wireless device u_k. Furthermore, step 220 may comprise, for each wireless device u_k, determining the channel quality metric /?_fc based on the received pilot signal transmission.

In some embodiments, the decision on the distribution of transmission power is taken locally in each access point A_j. In such embodiments, step 230 may comprise taking a decision locally in the access point A_j on the distribution of the transmission power. This may e.g. be done by maximizing a utility function, which is a function of the distribution of transmission power, the priority parameters p_lt ... , p_N, and the channel quality metrics β-L , ... , /?_w , as discussed above in the context of the example given with reference to Figs. 1- 3. An example of such a utility function is given by Eq. 1 above.

In other embodiments, the decision on the distribution of transmission power is taken centrally in the central unit 100. In such embodiments, step 230 may comprise steps 232 and 234 illustrated in Fig. 9. Step 232 comprises sending the channel quality metrics /?_fc to the central unit 100. Step 234 comprises receiving the decision on the distribution of transmission power from the central unit 100.

A related method 300 of operating the central unit 100 according to an embodiment is illustrated in Fig. 10. In Fig. 10, the operation of the method 300 is commenced in step 310. Step 320 comprises receiving the channel quality metrics /?_fe from the access points

A , ... , A_K. Step 330 comprises taking decisions on the distributions of transmission power for each of the access points A , ... , A_K. Step 340 comprises sending, to each of the access points A , ... , A_K, the respective decision on distribution of transmission power. That is, the decision on the transmission power distribution relating to a particular access point A_j, taken by the central unit 100, is sent to that particular access point A_j. The operation is then ended in step 350.

Similar to embodiments where the decisions are taken locally by the access points A_j, step 330 of taking the decisions may comprise maximizing a utility function, which is a function of the distributions of transmission power, the priority parameters p_lt ... , p_N , and the channel quality metrics /?_fe . When the decision is taken centrally by the central unit 100, the decision can be taken based on a larger set of channel quality metrics, namely Uy{/?i ,■■■ , N }- The central unit 100 thus has the opportunity to decide a transmission power distribution that is better, or "more optimal", in a global sense (i.e. for the whole distributed MIMO system) compared with if the decisions are made locally in the access points Aj . However, an advantage of taking the decisions locally in the access points Aj is that that results in less signaling overhead between the central unit 100 and the access points Aj .

The methods 200 and 300 described above may be repeated as necessary. For example, they may be repeated when channel quality metrics /? _fe have changed, when priorities have changed, when wireless devices enter or leave the coverage are of the distributed MIMO system, for each data packet to be transmitted, etc.

Description of computer-readable media figures

As described above with reference to Figs. 5 and 6, the processing circuits 54 and 104, or parts thereof, may be implemented with programmable and/or configurable hardware units, such as but not limited to one or more field-programmable gate arrays (FPGAs), processors, or microcontrollers. Thus, the processing circuits 54 and 104 may be a programmable processing circuits. Hence, embodiments of the present disclosure may be embedded in computer program products, which enables implementation of the methods and functions described herein, e.g. the embodiments of the methods 200 and 300 described with reference to Figs. 7-10.

According to some embodiments, there is provided a computer program product comprising computer program code for executing the method 200 when said computer program code is executed by the programmable processing circuit 54 of the access point Aj . The computer program code may be stored on a computer readable medium 1000, as illustrated in Fig. 1 1. The computer-readable medium 1000 may e.g. be a non-transitory computer-readable medium. The computer program code may be loadable into the memory 58 in order to be executed by the processor 56.

According to some embodiments, there is provided a computer program product comprising computer program code for executing the method 300 when said computer program code is executed by the programmable processing circuit 104 of the central unit 100. The computer program code may be stored on a computer readable medium 2000, as illustrated in Fig. 12. The computer-readable medium 2000 may e.g. be a non-transitory computer-readable medium. The computer program code may be loadable into the memory 108 in order to be executed by the processor 106.

The present disclosure has been presented above with reference to specific embodiments. However, other embodiments than the above described are possible within the scope of the disclosure. Different method steps than those described above, performing the method by hardware or software, may be provided within the scope of the disclosure. The different features and steps of the embodiments may be combined in other combinations than those described.

The term "comprises/comprising" when used in this disclosure is taken to specify the presence of stated features, integers, steps, or components, but does not preclude the presence or addition of one or more other features, integers, steps, components, or groups thereof.

Claims

1. A method (200) of operating an access point (A_j) of a distributed MIMO system, wherein the distributed MIMO system serves a number (N) of wireless devices (u , ... , u_N), the method comprising:

obtaining (220), for each wireless device (u_k), a channel quality metric (/?_fc ) indicative of a channel quality between the access point (A_j) and the wireless device (u_k); and

determining (230), based on priority parameters (p , ... , p_N) associated with the wireless devices (u , ... , u_N), wherein the priority parameter (p_k) associated with a wireless device (u_k) is indicative of a priority of an upcoming transmission from the distributed MIMO system to that wireless device (u_k), and the channel quality metrics (/?_¾, ), a distribution of transmission power from the access point to the different wireless devices (u , ... , u_N).

2. The method of claim 1 , comprising:

obtaining (215), for each wireless device (u_k), said priority parameter (p_k) from a central unit (100) of the distributed MIMO system.

3. The method of claim 1 , wherein the determining step (230) comprises:

sending (232) the channel quality metrics ( _k ) to a central unit (100) of the distributed MIMO system; and

receiving (234) a decision on the distribution of transmission power from the central unit (100) of the distributed MIMO system.

4. The method of claim 1 , wherein the step (230) of determining comprises:

taking a decision locally in the access point (Aj) on the distribution of the transmission power.

5. The method of claim 4, wherein the step of taking said decision comprises:

maximizing a utility function, which is a function of the distribution of transmission power, the priority parameters (p , ... , p_N), and the channel quality metrics ■■■ , /?w_,y)-

6. The method of any preceding claim, comprising receiving, from each wireless device (u_k), a pilot signal transmission.

7. The method of claim 6, comprising, for each wireless device (u_k), determining the channel quality metric based on the received pilot signal transmission.

8. A method of operating a central unit (100) of a distributed MIMO system, the distributed MIMO system comprising a plurality of access points (A , A_K) and said central unit (100), wherein the access points (A , ... , A_K) are configured to operate according to the method of claim 3, the method comprising:

- receiving (320) the channel quality metrics (/?_fe ) from the access points (A , ... , A_K) of the distributed MIMO system;

- taking decisions (330) on the distributions of transmission power for each of the access points (A , ... , A_K); and

- sending (340), to each of the access points (A , ... , A_K), the respective decision on distribution of transmission power.

9. The method of claim 8, wherein the step of taking a decision comprises:

maximizing a utility function, which is a function of the distributions of transmission power, the priority parameters (p , ... , p_N), and the channel quality metrics (Uy{/?i ,■■■ , /?_N })-

10. A method of operating a distributed MIMO system comprising a plurality of access points (A , ... , A_K) and a central unit (100), comprising

performing, in each of the access points (A_lt ... , A_K), the method of any one of the claims 1-7.

1 1. The method of claim 10, comprising

sending, from the central unit (100) to each of the access points (A_lt ... , A_K), the priority parameters (p , ... , p_N).

12. A method of operating a distributed MIMO system comprising a plurality of access points (A , ... , A_K) and a central unit (100), comprising

performing, in each of the access points (A , A_K), the method of claim 3; and performing, in the central unit (100), the method of claim 8 or 9.

13. The method of any preceding claim, wherein the channel quality is an SNR or a path loss.

14. An access point {Aj) for a distributed MIMO system, comprising

a transceiver (52) for communicating wirelessly with wireless devices (u_k); and a processing circuit (54) configured to perform the method of any one of the claims 1-7.

15. A cluster (50) comprising a plurality of access points (A , ... , A_K) of claim 14.

16. A distributed MIMO system comprising the cluster (50) of claim 15 and a central unit (100).

17. The distributed MIMO system of claim 16, wherein the central unit (100) is configured to send the priority parameters (p_1; ... , p_N) to each of the access points (A , ... , A_K) in the cluster (50).

18. A central unit (100) for a distributed MIMO system, comprising

a communication interface (102) for communicating with access points (A_lt ... , A_K) of the distributed MIMO system; and

a processing circuit (104) configured to perform the method of claim 8 or 9.

19. A computer program product comprising computer program code for executing the method of any one of the claims 1 - 7 when said computer program code is executed by a programmable processing circuit (54) of the access point {Aj).

20. A computer program product comprising computer program code for executing the method of any one of the claims 8 - 9 when said computer program code is executed by a programmable processing circuit (104) of the central unit (100).

21. A computer readable medium (1000, 2000) having stored thereon the computer program product of claim 19 or 20.