ADVANCED DOWNLINK MACRODIVERSITY OPERATION

TECHNICAL FIELD

The present invention generally relates to communications technology, and more particularly to downlink macrodiversity in a wireless communication network.

BACKGROUND

Macrodiversity is a well-known technique for enhancing communication wireless communication networks such as cellular networks. In general, macrodiversity aims to combat fading and to increase the received signal strength in exposed positions in- between base stations or access points.

An important example of macrodiversity is soft handover or soft handoff. Soft handover refers to a feature used by e.g. UMTS (Universal Mobile Telecommunications System) and CDMA (Code Division Multiple Access) standards, where a mobile is simultaneously connected to two or more cells during a call. The mobile may thereby simultaneously receive signals from two or more base stations that are transmitting the same bit stream on the same channel. The mobile receiver can then combine the received signals in such a way that the bit stream can be decoded much more reliably than if only a single base station was transmitting to the mobile. If any one of these signals fades significantly, there will be a relatively high probability of having adequate signal strength from at least one of the other base stations.

Macrodiversity has also been proposed in combination with space-time codes in reference [I]. Here, space-time codes are used to achieve coded transmit macrodiversity on the cellular downlink.

However, a more straightforward type of macrodiversity is to send the same signal from different base stations in different orthogonal resources (e.g. in time, frequency or code domain) and use some suitable combining scheme in the receiver such as maximum ratio combining or interference rejection combining.

Traditional types of macrodiversity schemes normally reject interfering signals carrying what is regarded as irrelevant information. In rejecting, or seeing interfering signals as detrimental, the total energy that is transmitted is underutilized with the consequence that the raw potential for increasing throughput, delay, power (or energy) is not fully exploited.

Hence, it would be of interest to find an even more resource-efficient macrodiversity scheme.

SUMMARY

The present invention overcomes these and other drawbacks of the prior art arrangements.

It is a general object of the present invention to improve the macrodiversity operation of a wireless communication network such as a cellular network.

In particular it is desirable to provide a highly resource-efficient downlink macrodiversity scheme.

It is a specific object to provide an improved method and system for downlink macrodiversity operation of a wireless communication network.

It is a specific object to provide a wireless communication network configured for improved macrodiversity operation.

It is another specific object to provide a base station adapted for downlink macrodiversity operation in cooperation with at least one similar base station.

These and other objects are met by the invention as defined by the accompanying patent claims.

The invention considers a network coding topology for so-called multi-user downlink macrodiversity involving at least two base stations and at least two user terminals. In this context, a basic idea is to provide a macrodiversity operation scheme in which different data packets are designated for and sent to different user terminals, and at least part of the sent data packets, designated for at least two different user terminals, are jointly encoded and concurrently transmitted as common signal information from at least two base stations and received by each one of the considered user terminals so that at least one of the user terminals can detect its own designated data. In this way, a highly resource- efficient downlink macrodiversity operation is provided.

Preferably, the considered base stations receive signal information representative of data streams designated for respective user terminals from a common network unit, and each one of the base stations transmits signal information representative of data designated for a specific user terminal or specific user terminals. On the receiving side, each one of the considered user terminals store, if received, signal information, referred to as own designated signal information, representative of at least part of its own designated data. In addition, each one of the user terminals store, if received, signal information, referred to as overheard signal information, representative of at least part of data designated for at least one other user terminal. The base stations then concurrently transmit common signal information representative of jointly encoded redundancy data designated for at least two different of the considered user terminals. Each one of the user terminals receive and store signal information, referred to as redundancy signal information, representative of jointly encoded redundancy data, enabling at least on of the user terminals to detect at least part of its own designated

data based on the redundancy signal information together with overheard signal information and/or own designated signal information.

For example, the concurrent transmission of common signal information representative of jointly encoded redundancy data can be performed based on feedback from the receiving side or performed automatically as part of an autonomous transmission scheme.

In a related aspect, the invention also provides a base station adapted for downlink macrodiversity operation in cooperation with one or more similar base stations.

Other advantages offered by the invention will be appreciated when reading the below description of embodiments of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention, together with further objects and advantages thereof, will be best understood by reference to the following description taken together with the accompanying drawings, in which:

Fig. 1 is a schematic diagram illustrating a wireless communication network with a network part configured for downlink macrodiversity operation according to an exemplary embodiment of the invention.

Fig. 2 is a schematic flow diagram illustrating a method for improved downlink macrodiversity operation according to an exemplary embodiment of the invention.

Fig. 3 is a schematic diagram illustrating a wireless communication network with a general network part configured for downlink macrodiversity operation according to another exemplary embodiment of the invention.

Fig. 4 is a schematic diagram illustrating the basic operation of a considered part of a communication network according to a particular example of the invention.

Fig. 5 is a schematic diagram illustrating an example of joint encoding as a weighted linear combination of the considered regular data packets.

Fig. 6 is a schematic diagram illustrating a wireless communication network with a network part, divided into groups, configured for downlink macrodiversity operation according to yet another exemplary embodiment of the invention.

Fig. 7 is a schematic diagram illustrating the base station and user terminal configuration in a considered part of a wireless communication network according to a preferred exemplary embodiment of the invention.

Fig. 8 is a schematic diagram illustrating an example of how a large number of base stations arranged in a hexagonal cell pattern may transmit, over time, packets belonging to different data streams according to a particular example of the invention.

Fig. 9 is a schematic diagram of an exemplary model for simple evaluation of the invention.

Fig. 10 is a schematic diagram illustrating the relative performance gain with respect to throughput.

DETAILED DESCRIPTION OF EMBODIMENTS

Throughout the drawings, the same reference characters will be used for corresponding or similar elements.

Fig. 1 is a schematic diagram illustrating a wireless communication network with a network part configured for downlink macrodiversity operation according to an exemplary embodiment of the invention. In contrast to classical downlink macrodiversity that merely considers a single receiving node, e.g. a mobile station receiving information from two or three base stations, the present invention considers a macrodiversity topology involving multiple receiving nodes, so-called multi-user downlink macrodiversity. In a wireless communication network, a considered part of the network comprises at least two base stations 10-1, 10-2 and at least two user terminals 20-1, 20-2 such as mobile stations or similar user equipment. Due to the macrodiversity structure, the base stations 10-1, 10-2 receive information from some common point in the network, such as a common network unit 30. In this novel network topology for downlink macrodiversity, different data packets are designated for and sent to different user terminals, and at least part of the sent data packets, designated for at least two different user terminals 20-1, 20-2, are jointly encoded and concurrently transmitted as common signal information from at least two base stations 10-1, 10-2 and received by each one of the considered user terminals so that at least one of the user terminals can detect its own designated data.

The overall benefit is improved throughput performance and reduced power/energy usage in the network.

Fig. 2 is a schematic flow diagram illustrating a method for improved downlink macrodiversity operation according to an exemplary embodiment of the invention. In step Sl, the considered base stations receive signal information representative of data streams designated for respective user terminals from a common network unit. In step S2, each one of the base stations transmits signal information representative of data designated for a specific user terminal or specific user terminals. On the receiving side, each one of the considered user terminals store, if received, signal information, referred to as own designated signal information, representative of at least part of its own designated data in step S3. In addition, each one of the user terminals store, if

received, signal information, referred to as overheard signal information, representative of at least part of data designated for at least one other user terminal in step S4. In step S5, the base stations then concurrently transmit common signal information representative of jointly encoded redundancy data designated for at least two different of the considered user terminals. In step S6, once again on the receiving side, each one of the user terminals receive and store signal information, referred to as redundancy signal information, representative of jointly encoded redundancy data. In this way, at least one user terminal is able to detect at least part of its own designated data based on the redundancy signal information together with overheard signal information and/or own designated signal information in step S7.

Fig. 3 is a schematic diagram illustrating a wireless communication network with a general network part configured for downlink macrodiversity operation according to another exemplary embodiment of the invention. In the example of Fig. 3, the considered part of the overall network has been generalized to include a number M of base stations and a number N of user terminals such as mobiles or similar user equipment (UE), where M ≥ 2, and N > 2, and M may be different from N. This setting allows a whole set of M base stations to concurrently transmit common signal information representing jointly encoded redundancy data designated for two or more of the set of N user terminals.

Optionally, as indicated by dashed lines in Fig. 3, the user terminals on receiving side may send feedback information to the base stations and/or the common network unit. Feedback messages from the user terminals to the base stations may be used to enable the base stations to determine which packets have been received correctly by the intended receiving terminals and which have not. In an exemplary embodiment, the concurrent transmission of common signal information from the base stations is triggered by an indication that at least one of said user terminals could not receive its own designated data. In this case, the jointly encoded redundancy data is preferably generated at least partly based on feedback information from at least a subset of the user terminals. Reliable transfer is enabled by encoding packets with an error detecting

code, such that the receiving side can detect erroneous or lost packets. Data sequence integrity can for example be ensured by sequential numbering of packets and applying certain transmission rules.

It should however be understood that the concurrent transmission of common signal information representative of jointly encoded redundancy data can be performed automatically as part of an autonomous transmission scheme without the need for any feedback.

The data, normally FEC (Forward Error Correction) encoded, may be transmitted on different (orthogonal) resources, e.g. in time, frequency or code, or alternatively on the same resource but employing e.g. MUD (Multi-User Detection) in receivers to separate the information received. Each user may potentially receive at least part of its own designated data, and may also overhear and stores at least part of the data sent from some other base station (in some other cell) and designated for some other user. Then, redundancy common for two (or more) users is sent from at least two base stations (in different cells). As mentioned above, the need for sending redundancy can be classified by one or more cases: i) it is always sent, ii) it is triggered by the fact that one or more of the user terminals could not receive its/their intended data. The second alternative is of particular interest since it only provides redundancy in cases when needed. On the other hand, the first alternative does not require any feedback channels. The redundancy can then be combined with previously received designated data and/or overheard data. The following important scenarios can be given:

• At least one of the user terminals detects at least part of its own designated data based on the redundancy signal information and overheard signal information.

• At least one of the user terminals detects at least part of its own designated data based on the redundancy signal information, overheard signal information and part of its own designated signal information.

• Each one of the user terminals detects at least part of its own designated data based on the redundancy signal information together with overheard signal information and/or part of own designated signal information.

• Each one of the user terminals detects at least part of its own designated data based on the redundancy signal information and the overheard signal information.

• Each one of the user terminals detects at least part of its own designated data based on the redundancy signal information, the overheard signal information and part of own designated signal information.

Optionally, any common information such as said redundancy may, to improve transmission reliability, be sent with a transmit diversity scheme such as Alamouti diversity [2], linear delay diversity, cyclic delay diversity [3] or any other Space Time Code scheme, but here adapted for distributed operation over multiple base stations.

Fig. 4 is a schematic diagram illustrating the basic operation of a considered part of a communication network according to a particular example of the invention. In the simplest scenario, two base stations are communicating with two user terminals. In this example, after receiving information from some common point and/or unit in the network, a first base station sends data Dl in resource 1, and a second base station sends data D2 in resource 2. Both base stations then concurrently send redundancy information in resource 3, where Dl and D2 are jointly encoded. The receivers preferably use the common redundancy information in combination with overheard information and/or part of their designated information in order to determine their own designated data.

For example, the joint coding can be made through weighted linear combinations of (substrings of) packets, or by generating joint redundancy data bits by traditional

forward error correction techniques such as Low Density Parity Check (LDPC) codes, Turbo Codes, Convolution codes, Block codes, Product codes and so forth based on two or more data packets. Typically, any of the above coding schemes operates in a finite field meaning that the numbers that may be operated on are a finite number. For instance, if the field size is two, the finite field addition is a bitwise XOR operation.

In general, each individual redundancy data packet is normally based on at least part of a previously transmitted packet, and can be provided in coded or uncoded form.

In a preferred embodiment of the invention, the information pieces, here referred to as regular data packets, are jointly encoded by a coding scheme operating with finite fields. The composite data packet formed by joint encoding is normally formed as a weighted linear combination of the considered regular data packets, for example as illustrated in Fig. 5. Preferably, the weights w are selected based on feedback information from the at least a subset of the user terminals such that a new linearly independent combination of regular data packets is formed.

Preferably, the combining is performed on soft information, such as Log-Likelihood Ratio (LLR) representation of bits. Different combining strategies may be used, e.g. through so called belief propagation algorithm. Another approach is briefly mentioned: For the case where two regular packets are combined with a bitwise XOR packet, and 4QAM or 2 PAM is used, a combiner operating similarly to e.g. iterative LDPC (Low Density Parity Check) decoders may be used. The soft bit combining process is then described by ¹ :

L _u ¹ ' ¹ = L _n ^u ₊ In a Ihx ⁶ I tM . \ , (1)

Depending on definition of how 1 and 0 are modulated, the ratio in (1) may be the inverse.

where Z ₁₁ is the LLR for a bit for the D ₁ packet sent to a first user, Z ₂₁ is the LLR for a bit for the D ₂ packet sent to the first user, and Z ₃₁ is the LLR for a bit for the D ₁ θ D ₂ packet sent to the first user.

Further information on joint encoding and the formation of composite data packets, although in a different field of application, can be found in reference [4], which relates to scheduling and coding in communication systems utilizing Automatic Repeat reQuest (ARQ), especially for multiple unicast ARQ, and reference [5], which relates to optimization of the performance for reliable multiple unicast flows.

The term concurrent transmission generally implies that common redundancy information is sent on a common channel such that the common signal information from two or more base stations can be received at the user terminals for cooperative macrodiversity interaction. In other words, the common signals are substantially overlapping in time for cooperative interaction at the receiver side. Preferably, although not necessarily, the common signals are transmitted on the same frequency.

The data to be transferred may include one or more data units, and both single-user detection and multi-user detection are possible detection alternatives that can be selected according to application and design choice.

The detection can be done per bit or symbol or per sequence of bits or symbols, for a single user or for multiple users. The detection may take place on coded information or on information bits. This means that the detected information may in fact be demodulated coded information and/or both demodulated and decoded information. Although it may be useful to store at least part of demodulated coded information for subsequent processing in the detection/decoding process, the normal procedure is to finally decode the desired designated information.

Fig. 6 is a schematic diagram illustrating a wireless communication network with a network part, divided into groups, configured for downlink macrodiversity operation according to yet another exemplary embodiment of the invention. In this particular example, the considered network part is divided into two groups, a first group A and a second group B. Here, group A comprises a number M of base stations lO-A-1 to 10-A-M, and a number N of user terminals 20- _^ 4-7 to 20-A-N. Group B comprises a number K of base stations 10-5-7 to 10-B-K, and a number L of user terminals 20-5-7 to 20-B-L. Each group may be configured for multi-user downlink macrodiversity operation in the same or similar manner as above. Naturally, more than two groups may be considered and configured for multi-user macrodiversity operation on the downlink.

In general, a first group of base stations may concurrently transmit first common signal information representative of jointly encoded redundancy data respectively designated for a first set of user terminals, and a second group of base stations may concurrently transmit second common signal information representative of jointly encoded redundancy data respectively designated for a second set of user terminals. The first and second groups of base stations may transmit the first and second common signal information in the same time resource, or in different time resources. Also, the first and second sets of user terminals may be partly overlapping.

Fig. 7 is a schematic diagram illustrating the base station and user terminal configuration in a considered part of a wireless communication network according to a preferred exemplary embodiment of the invention. In this example, which preferably is based on feedback, each base station 10 basically comprises a data buffer module 11, an acknowledgment (ACK) register 12, a scheduler 13, and a packet encoder 14 as well as some form of transmitting functionality.

The data buffer module 11 receives data packets from a common network unit (not shown in Fig. 7) and temporarily holds data packets for subsequent scheduling/encoding and transmission to various user terminals.

Typically, each base station 10-1 to 10-M first sends one or more so-called regular data packets. As feedback information becomes available from the user terminals, each base station may form and send one or more composite data packets, each composite data packet being formed by jointly encoding redundancy data, in the form of already sent data packets, designated for at least two different user terminals.

The scheduler 13 considers which data packets that reside in the data buffer module 11, and also considers ACK information from the different user terminals stored in the ACK register 12. Basically, the scheduler 13 performs the scheduling of regular data packets available in the data buffer module 11 by determining a set of coding weights based on the ACK feedback in the ACK register 12. The scheduler 13 informs the data packet encoder 14, which forms or encodes a composite data packet as a weighted linear combination of regular data packets based on selected coding weights.

Supplementary knowledge may also be used for the scheduling. Supplementary information may include, but is not limited to: QoS requirements for the possibility to cater for QoS scheduling aspects, as well as status of individual packets such as their time to live value.

Each user terminal 20-1 to 20-N preferably comprises receiving means, which provides the necessary functionalities for performing the actual reception, a data packet decoder 21 with feedback functionalities, as well as a data buffer module 22. When a composite packet is received, the decoder 21 normally identifies which packets have been encoded together. Using the data packets already received and stored in the buffer module 22 the decoder 21 tries to decode the composite multicast packet to extract individual regular data packets, and particularly to retrieve own designated data packets. The decoder 21 is also configured for issuing and handling feedback messages.

Fig. 8 is a schematic diagram illustrating an example of how a large number of base stations arranged in a hexagonal cell pattern may transmit, over time, packets belonging to different data streams according to a particular example of the invention.

In a more general setting, applied to a cellular system, large number of base stations and user terminals such as mobile stations or similar user equipment may be involved over time where different groups of cells jointly encode the same messages. In Fig. 8 it is shown how packets belonging to different data flows A to I are sent from different base stations where the base stations, in this example, are placed in a hexagonal cell pattern. Each piece of information, such as a packet, is also enumerated, i.e. Al means a first packet of data flow A. It is also illustrated that packets sent in different cells may be jointly encoded into composite packets comprising two, three or possibly more packets. Further, each message may be weighted with some weights to increase the likelihood of, or to ensure, that new linear combinations are received over time. Whenever a composite packet is sent from multiple base stations, space time coding may be used to further improve the reception quality.

Fig. 9 is a schematic diagram of an exemplary model for simple evaluation of the invention. A basic performance evaluation for the orthogonal resource scenario has been performed, and will now be described briefly with reference to Fig. 9. The purpose is to illustrate the relative throughput gain over a more traditional macrodiversity scheme, in this case also employing communication over orthogonal resources. More specifically, the throughput for a symmetrical system is assessed where the probability for reception from the desired base station (BS) is p _a and the probability for overhearing a data transmission from the other BS is p _b . When receiving redundancy concurrently from both base stations BS ₁ and BS ₂ , the probability for reception is p ' _b , where p ' _b ≥ p _b due to diversity and potentially more total power being invested. We will assume that the user terminals, or mobile stations (MS) MS ₁ and MS ₂ are associated to the "best" base station such that/? _α > p _b holds.

First, the throughput for the scenario where packets are retransmitted without exploiting the overheard information is:

T ^{Res) = P _a . (2)

The throughput for the system in Fig. 9 where redundancy information and overheard information are exploited through network coding according to the invention and assuming a large quantity of packets sent, can be calculated as:

In deriving (3), it is assumed that the BS knows what packets have been received by which MSs. It is further assumed that when sending redundancy information, we may assume that p \ = P _b since p \ is lower bounded by p _b . Hence, using this value ofp ' _b provides a lower bound of the performance.

The relative performance gain of (2) over (1) vs. p _a and p _b is shown as contour plots in Fig. 10. As we assumed p _a >p _b , the only permissible area of operation is below thep _a = P _b line drawn in Fig. 10. It is evident that one can make a small, but useful gain of order 5-20%, at least for MS placed very close to cell borders where the reception probabilities would be of nearly the same magnitude. With more nodes, as well as more BSs, the gain will be even higher. One may, similar to MU-ARQ [4] and [5], expect that the throughput increases when the number of users increases.

There is likely to be additional performance benefits by, as optional embodiments suggest, combining network coded transmissions with space time coding or other TX diversity schemes.

The core notion of network coding is to allow and encourage mixing of data at intermediate network nodes. A receiver sees these data packets and deduces from them the messages that were originally intended for the data sink. Traditionally, network coding has been applied in fixed wired, non-overhearing, error free, and feedback free networks and in particular specific topologies like the multicast networks, i.e. where the same data is delivered to all nodes in a network.

The embodiments described above are merely given as examples, and it should be understood that the present invention is not limited there to. Further modifications, changes and improvements which retain the basic underlying principles disclosed and claimed herein are within the scope of the invention.

REFERENCES

[1] Coded Transmit Macrodiversity: Block Space-Time Codes over Distributed

Antennas, by Y. Tang and M.C. Valenti.

[2] US Patent 6,185,258

[3] US Patent 6,842,487

[4] International Patent Application PCT/SE2005/001144 published as WO 2007/008123 Al.

[5] Multi-User ARQ, by P. Larsson, N. Johansson, Conference proceedings of VTC2006spring, Melbourne, 7-10 May 2006.