WO2009155729A1 - Low-complexity multi-base-station mimo method and apparatus for up-link - Google Patents

Abstract

A low-complexity, multi-base-station MIMO method for use in a communication network includes: determining the cluster of the user equipment, said cluster includes performing MIMO base-station convergence for the user equipment; processing procedure for checking user equipment signal and cancelling the greatest interference between users; and the service base-station in the user cluster performs combination of interference-cancelled user signals coming from every base-station. A low-complexity multi-base-station MIMO apparatus for use in a communication network.

Description

 Low complexity multi-base station MIM0 method and device for uplink

 The present invention relates to the field of wireless communications, and in particular, to a low complexity multi-base station MIM0 method and apparatus for uplink. Background technique

 Compared to LTE (Long Term Evolution) R8, the average spectral efficiency and cell edge spectrum efficiency must be further increased by two to three times in LTE-advanced systems to meet the needs of IMT-advanced systems. A common view from major companies in 3GPP (Alcatel-Lucent is one of the main leaders) is that multi-base station MIM0 (Multiple Input Multiple Output) technology is a major candidate for improving system performance, especially edge performance. One.

 With multi-base stations, typical inter-cell interference is converted into signals, so signal combining can be performed to enhance system performance. However, due to centralized joint signal processing, significant backhaul costs are incurred between multiple sites, ie, information such as Instant Channel State Information (CSI) and symbols are transmitted, and the computational complexity of centralized processing is also unacceptable. Achieving a balance between system performance enhancements and a large amount of signaling overhead on the backhaul remains a challenging challenge.

 In an actual system, each user is different from the wireless channel of each base station in the entire wireless network. Therefore, users of different cells or even users of the same cell may have different coordinated cell clusters; and the coordinated cell cluster between different users may have only one or A limited number of base stations (BS) overlap. This feature is not utilized by the upstream network MIM0 scheme because it defines a cell-based coordinated cell cluster in which users in the cluster perform centralized signal processing; and because there is no cell-based inter-cluster communication defined, two adjacent cell users Although there is strong co-channel interference, interference cancellation or suppression cannot be performed because they belong to different cell clusters. This feature is partially utilized by the downlink cooperative MIM0, but because its scheme limits the transmission of channel information by users between base stations, it limits the interference suppression or cancellation of users between cells. For the uplink, we propose a distributed interference suppression or cancellation technique for users in a cluster across a community or across clusters on a publicly overlapping BS, which can effectively eliminate the same-frequency interference of the strongest users and improve The signal strength of each base station in the cluster of users to each coordinated cell.

 Alcatel-Lucent has proposed two multi-base station MIM0 solutions, 艮 P :

■ Network MIM0 for both uplink [1] and downlink In network MIMO, inter-cell interference is converted to a signal combination by coordination between multiple base stations (called clusters) and multiple users. The central processing unit is required to perform joint detection on users in the coordinated plurality of base stations. At the same time, since there is no inter-cluster communication, if users share only some of the base stations in the cluster, they cannot be jointly detected in the central unit, so the co-channel interference between them cannot be effectively suppressed. It has been confirmed that the network MIM0 can bring significant performance enhancement, but its scheme complexity is high, and it requires a large amount of backhaul bandwidth to transmit CSI and quantized received signals.

 ■ Collaboration for the downlink MIM0 [2]

 In cooperative MIMO, users are jointly served by multiple adjacent base stations, and each is coordinated.

The base station is tuned to use SDMA technology to serve multiple users within the cell. In the cooperative MIM0, since the instantaneous CSI signaling is not allowed to be transmitted between the base stations, the cross-cell multi-user M I M O cannot be effectively formed. Collaborative MIMO has low complexity, but its system performance improvement is not obvious.

 As shown in Figure 1, there are three cells that make up cluster C1 and cluster C2 (for comparison, the cluster size is simplified [1]). There are User 1 (service user of cell A) and users 2, 3, 4 (service users of cell B), and user 5 (service user of cell C). C1 consists of cells A and B and includes two users 1 and 2. Cluster 2 consists of cell B and cell C and includes only users 3, 5. User 4 belongs only to cell B. In network MIMO, only users 1 and 2 can be jointly detected in the central unit of cluster C1. In the uplink cooperative MIM0 (the concept of downlink cooperative MIM0 is applied to the uplink), users 2 and 3 or users 3 and 4 can be jointly detected, however, joint detection of users 1 and 3 or users is not allowed. 1 and 4, since User 1 and User 3 and 4 are in different cells, and instant CSI signaling between base stations is prohibited in cooperative MIM0. At the same time, the S D M A technology in the cooperative M I M O does not allow the base station to schedule the number of antenna users, so the interference cancellation between users is limited to the number of users that can be supported.

Multi-base station MIMO is to convert the interference of neighboring cells in the traditional cellular network into useful signals, and the strength of the wireless signal decays rapidly with distance, so the number of base stations or cells of the cell-based cluster defined in the network MIM0 Too large (for example, >= 7), at the same time, all users in the cell must choose the cluster. From the perspective of return cost and signaling overhead, there is no achievability in the actual system. Therefore, for an actual system with reasonable complexity, a user-based cluster (a coordination area composed of several base stations to implement a multi-base station MIM0) should be redefined. In addition, in [1], the definition of inter-cluster communication does not take into account complexity. This does not pose a major problem for network MIM0 with a large coordination area and including central users into the cluster. However, such inter-cluster communication will play an important role in newly defined clusters with appropriate complexity and should be addressed. For uplink cooperative MIMO, it prohibits instantaneous CSI signaling between base stations and SDMA (Space Division Multiple Access) based technology, which limits its system performance. For example, in Figure 1, the strong interference from cell 4 to user 1 to cell B is significant and cannot be effectively offset by cooperative MIM0.

 Therefore, a new multi-base station MIM0 technology is needed to meet the low bandwidth requirement and low processing complexity; and the inter-cell co-channel interference can be effectively overcome, and the edge spectrum efficiency is improved. The average spectral efficiency of the system can be improved by single-user and multi-user M I M O of a single base station. references:

 [1] Coordinating base stat ions for greater upl ink spectral efficiency : proportional ly fair user rates, Sivarama Venkatesan, Bell-Labs, ALU, PIRMC07

 [2] Collaborative MIMO based on multiple base station coordination, Yang Song, Liyu Cai, Keying Wu, Hongwei Yang, IEEE C802. 16m_07/162

Summary of the invention

 The present invention proposes a low complexity MIMO scheme for use in a multi-base station environment, particularly for the uplink, where multiple cells (base stations) can jointly combine received signals from one or a group of users in the cluster Without passing the user's instantaneous channel information (CSI). At the same time, in each cell (base station) in the cluster, adaptive receive beamforming can be used, for example, a SE (minimum mean square error) receiver, at the local base station to the users and clusters from the cluster but Users with overlapping cells jointly detect or perform interference suppression.

 The present invention is directed to a cell edge user, defining a dynamic cluster with a limited number of cell subsets for each user to implement multi-base station MIM0 transmission. Depending on the long-term CSI between the user and the cell (including path loss and shadow attenuation), the cluster including that size and the users belonging to the cluster should be updated in a slow manner (eg 5- lOOras). For users in the cluster, multi-cell pilots need to be defined, and the base stations within the cluster can also detect immediate CSI for all users in the cluster. This instant CSI information is stored in each base station and can be further transferred to the central unit of the cluster or depending on whether further centralized processing of the cluster users is required. In order to reduce the cost of backhaul and reduce the processing complexity, it is assumed hereinafter that CSI signaling is not required to be transmitted by backhaul.

The present invention uses adaptive receive beamforming at each base station to implement a cluster in the base station containing the base station Interference suppression or interference cancellation by users (services of the base station and/or non-serving users) and service users of the base station. In adaptive receive beamforming, only the users that cause the strongest inter-user interference are used for joint detection with the target user, which is most effective for reducing inter-cell and/or inter-cluster interference in the local base station.

 The present invention proposes a simple bit-level soft-combining to combine the received signals of users in the cluster to reduce the cost of backhaul. In practice, this is cost effective because in each base station, the local strongest inter-user interference has been mitigated and the user's signal to each base station in the cluster is enhanced, so a simple bit-level soft combination is sufficient to effectively improve system performance.

 Accordingly, in accordance with an aspect of the present invention, a low complexity multi-base station MIM0 method for use in a wireless communication network is provided, comprising the steps of: performing, at each base station, determining a cluster of all user equipments of the cell, The cluster includes a base station that performs multi-base station MIMO on the user equipment; processing steps, each base station detects a signal of the user equipment and cancels the strongest co-channel interference from different cells or different clusters; and the user service base station pairs each base station from the cluster The signals that are cancelled by the interference are combined.

 According to another aspect of the present invention, a low complexity multi-base station MIM0 device for use in a communication network is provided, wherein the device includes: determining means for determining a cluster of user equipment, the cluster including the The user equipment performs a base station set of multi-base station MIMO; the processing device is configured to detect a signal of the user equipment and perform interference cancellation; and the combining device combines the interference canceled signals from the respective base stations.

 According to still another aspect of the present invention, a base station including the above apparatus is provided.

 According to still another aspect of the present invention, a communication system including the above base station is provided. DRAWINGS

 The above and other features of the present invention will become more apparent from the detailed description of the appended claims.

 1 is a schematic diagram showing a multi-base station MIM0 scheme for uplink in the prior art; FIG. 2 is a diagram showing low complexity of the present invention with distributed inter-cell or inter-cluster user interference cancellation or suppression and then simple combining processing Schematic diagram of a multi-base station MIM0 scheme;

 3 is a flow chart showing the main steps of a low complexity multi-base station MIM0 method according to the present invention; and

4 is a block diagram showing a low complexity multi-base station MIM0 device in accordance with the present invention. detailed description

 2 is a low complexity multi-base with distributed inter-cell joint processing of the present invention

Schematic diagram of the station MIM0 scheme. As shown in FIG. 2, edge user 1 of cell A and edge user 2 of cell B form cluster C1, while user 4 of cell C and user 5 of cell B form cluster C2 because of clustering in [1] Definition, this is simpler.

 For users in the cluster, multi-cell pilots for uplink transmission should be defined so that the base station in the cluster can detect the channel information (CSI) of the user, and the cell B can also use the CSI of the user 1 to the cell B. Detected as h3 and h4. Cell A can detect h7 and h8 of user 4. As an example, User 1 is scheduled in Cell A, User 3 is scheduled in Cell B (unlike Cooperative M I M O , Cell B can also schedule User 3 and User 5 simultaneously), and User 4 is scheduled in Cell C. In this way, in the cell A, the local adaptive receive beamforming is used to jointly process the radio link between the user 1 and A and the radio link between the user 4 and the cell A. For the user 1, the neighbor can be effectively eliminated. The interference of the cell 4 and the neighboring cluster user 4, as well as the user 4, can effectively eliminate the interference of the user 1 in the cell A. The soft bit signals for the detected signals of User 1 and User 4 are then transmitted to the central processing unit of the user corresponding clusters C1 and C2 for further bit level soft combining, as with H A R Q combining. The central processing unit of each user corresponding cluster is located in the modified user service cell. This is also true for the local adaptive receive beamforming (e.g., wake up SE) for cell 3 for user 1 and cell B. This adaptive receive beamforming locally in each cell can effectively eliminate interference between two different families of users (such as user 1 in C 1 and user 4 in C 2 ) or interference suppression across cells (eg User 1 in the A cell and User 4 in the B cell, in the case where User 1 and User 4 are or do not form MU-MIMO, which is not possible with the existing network MIM0 or the cooperative MIM0, which is fully applicable to There may be only one or several cells overlapping between different user clusters, and joint detection or interference suppression between different users may be performed on overlapping cells. The proposed multi-base station MIM 0 method compensates the same-frequency interference of the strongest interfering users from different cells or different clusters on the radio link of each cell in the cluster, and then combines and receives the intra-cluster cell signals. Greatly helps improve system performance.

With scheduling methods (distributed iterative multi-cell scheduling or parallel distributed scheduling, etc.), the proposed transmission and detection can be used as a priori knowledge of the scheduler, or only as a priori knowledge of link adaptation. Among these two options, the scheme of Figure 2 can effectively improve the system spectral efficiency, especially the spectral efficiency of the edge users. The main steps of the low complexity multi-base station MIMO method of the present invention are described below with reference to FIG. Here, some details are given with the proposed method, and an example is given to analyze the potential performance improvement of the method.

 The method of Figure 3 begins at S300 and then dynamically defines a cluster based on each user, including a set of cells that need to cooperate for the user in the cellular network for multi-base station MIM0 transmission. The definition of a cluster is based on long-term channel state information (CSI) including path loss and shadow attenuation. Specifically, at S302, a threshold PLD defining a cluster for all users is constructed - threshold at S304, for each user i, cluster i is initialized, which includes a set of serving cells for user equipment i. At S306, the cluster is adjusted to find the neighboring cell, that is, the difference between the path loss (including the shadow weakening) between the serving cell and the non-serving cell of the user i is less than the threshold PLD-threshold, and then by joining the non-serving cell. Update cluster i; otherwise, if no suitable neighbor cell is found, no update is performed, or when the user equipment moves to change the serving cell and the non-serving cell, the non-serving cell is removed, and the user's new service is used. The cell is to update cluster 1. Need to mention extra, it may be necessary to reorganize the cluster, that is, if multiple users belong to the server cluster (meaning that their subset cells are the same, not just partially overlapping), then each of the multiple users will be The cluster is merged into only one cluster for the multiple users, which can be better scheduled to form a multi-base station multi-user MIM 0. In addition, each user's cluster is updated slowly (eg, at any time in 5 - 100ms).

 At S308, for a user in the cluster, a multi-cell pilot for uplink transmission is defined. A detectable interfering user (which may be across cells or clusters) is then identified at S310, and joint detection or interference suppression is performed in S3 1 2 . As described above with reference to FIG. 2, user 1 is scheduled in cell A, user 3 is scheduled in cell B, and user 4 is scheduled in cell C. The local adaptive receive beamforming is used in cell A to jointly handle the radio link between users 1 and A and the radio link between user 4 and cell A. Then, when S312 base station A detects user 1, it performs interference suppression on the strongest interfering user such as user 4, and in detecting user 4, interference suppression can be performed on user 1 or another user. Then, at step 314, the soft bit information for the detected signal is transmitted to the central processing unit of the cluster for further bit level soft combining.

 An example is given below to illustrate the comparison between the low complexity multi-base station MIM0 of the present invention and the existing network MIM0.

 It is assumed that User 1 and User 3 are scheduled in cells A and B. Each cell has only two receive antennas, and all users have only one Tx antenna. Therefore, the performance of the Sa SE receiver is equivalent to an MRC (Maximum Ratio Combining) receiver with an equivalent SINR of the sum of the SINR (signal to interference to noise ratio) for each channel.

1) If the network MIM0 is applied in a centralized joint process, then User 1 is centrally processed using a linear wake-up SE at cluster 1 with cells A and B, with an equivalent SINR_userl=|hl| ² + |h2| ² + |h3| ² + |h4| ² , And use user 3 as white interferer (white interferer);

User 3 in cell B is detected by MMSE, and user 1 is treated as white interference, and its equivalent SINR_user3=|h5| ² + |h6| ² .

 2) Using the solution proposed by the present invention:

First, the location of user 1 is detected by using 匪SE in cell A, and its equivalent SINR is: SINR_userl=|hl | ² + |h2) ² _;

 Then use the painting SE to jointly detect the user 1 and the user 3 in the cell B;

Thus, the equivalent SINR of user 1 in cell B is: SINR_userl>|h3| ² + |h4| ² because it cancels the interference of user 3;

The equivalent SINR of user 3 in cell B is: SINR_user3>|h5l ² + |h6| ² because it cancels the interference of user 1;

 In addition, the two signals of User 1 can be soft combined at the bit level like HARQ.

It can be seen that the equivalent SINR of the final user 1 is: SINR_userl>|hl| ² + |h2| ² + |h3| ²

+ |h4| ² , and SINR_ user3>|h5l ² + |h6| ² o

 4 is a block diagram showing a low complexity multi-base station MIMO device in accordance with the present invention. As shown, the low complexity multi-base station MIM0 device 400 includes determining means 402, processing means 404 and combining means 406. The determining means 402 is configured to determine a cluster of user equipment, the cluster comprising a base station that performs a multi-base station M I MO for the user equipment. Preferably, the determining means 402 comprises: an initialization module, configured to initialize a cluster of the user equipment to a serving base station including the user equipment; a change module, configured to change the cluster according to a predetermined rule; and an update module, periodically updating the user equipment Cluster. Preferably, the change module comprises: a threshold generation sub-module, generating a threshold for each user equipment; a calculation sub-module, calculating a difference in path loss between the user equipment to the serving base station and the user equipment to the non-serving base station; and an alternation And, in the case that the calculated difference is less than the threshold, the non-serving base station is added to the cluster of the user equipment.

 Processing device 404 is operative to detect signals from the user equipment and to counteract the strongest inter-user interference. Preferably, a local adaptive receive beamforming device, such as a linear minimum mean square error receiver or a nonlinear interference cancellation receiver, may be employed. The combining means 406 combines the signals canceled by the interference from the respective base stations. For example, the combining device can include a bit level soft combining device.

In summary, the present invention proposes a low complexity multi-site MM0 with multi-base station cooperation, which is only used For edge users (defined by the difference in path loss above), it significantly reduces the number of pilots, backhaul costs, and scheduling credits. In addition, the present invention proposes a distributed local adaptive receive beamforming to achieve interference cancellation between inter-cell or inter-cluster users, which can provide a more robust wireless link. The present invention also proposes a soft combination of user signals across cells (cluster) to achieve multi-base station MIMO transmission with low backhaul cost and low computational complexity, for example, 2*4M IMO is decomposed into two 1*2 MIMs. 0 matrix calculation, but each 1 x 2M IM 0 effectively eliminates the strongest local interference. Such local joint inter-cell or inter-cluster inter-user cancellation in each base station can compensate for performance loss due to centralized multi-base station processing. Thus, the present invention achieves comparable performance with much lower complexity than the uplink network MIM0.

 While the invention has been described in connection with the preferred embodiments of the embodiments of the embodiments of the invention . Therefore, the invention should be construed as limited by the appended claims and the appended claims.

Claims

A low complexity multi-base station MIM0 method for use in a communication network, comprising the steps of performing at each base station:  Determining a cluster of user equipment, the cluster including a base station set of M I M O for the user equipment  Processing steps to detect signals of the user equipment and offset the strongest inter-user interference;  Right  The serving base station in the user cluster combines the user signals from the respective base stations that cancel the interference. 2. The method of claim 1, wherein the step of determining a cluster comprises the following substeps:  Initializing a cluster of user equipment to a serving base station including the user equipment;  begging  Change the cluster according to predetermined rules; and  Regularly update the cluster of user devices.  3. The method according to claim 1, wherein the determining the cluster further comprises: merging clusters of each of the plurality of user devices into a plurality of user devices having the same cluster Only one cluster of the plurality of user devices.  4. The method according to claim 2, wherein the changing step comprises:  Define a threshold for each user device;  Calculating a difference in path loss between the user equipment to the serving base station and the user equipment to the non-serving base station;  If the calculated difference is less than the threshold, the non-serving base station is added to the cluster of user equipment.  The method according to claim 2, wherein the period of performing the updating step is any value from 5 ms to 100 ms.  The method according to claim 4, wherein the changing step further comprises:  When the serving base station becomes a non-serving base station due to the movement of the user equipment, the non-serving base station is removed from the cluster of the user equipment while its new serving base station is added to its cluster. 7. The method of claim 1, wherein the processing step is performed using local adaptive receive beamforming. 8. The method of claim 7, wherein the local adaptive receive beamforming comprises linear minimum mean square error or nonlinear interference cancellation reception.  9. The method according to claim 1, wherein the combining step comprises:  The bit-level soft combining is performed on the signals canceled by the interference from the respective base stations.  10. A low complexity multi-base station MIM0 device for use in a communication network, wherein the device comprises  Determining means for determining a cluster of user equipment, the cluster comprising a base station that performs a multi-base station M I M O for the user equipment;  Processing means for detecting signals of the user equipment and canceling the strongest inter-user interference; and combining means for combining the signals canceled by the interference from the respective base stations.  11. Apparatus according to claim 10, wherein said processing means comprises local adaptive receive beamforming means.  12. Apparatus according to claim 11 wherein said local adaptive receive beamforming means comprises a linear minimum mean square error receiver or a non-linear interference cancellation receiver.  13. The device according to claim 10, wherein the determining means comprises:  An initialization module, configured to initialize a cluster of user equipment to a serving base station including the user equipment; a change module, configured to change the cluster according to a predetermined rule;  Update the module to periodically update the cluster of user devices.  14. The device according to claim 13, wherein the change module comprises a threshold generation sub-module that generates a threshold for each user equipment;  Calculating a sub-module, calculating a difference in path loss between the user equipment to the serving base station and the user equipment to the non-serving base station;  The replacement sub-module adds the non-serving base station to the cluster of the user equipment if the calculated difference is less than the threshold.  15. Apparatus according to claim 10, wherein said combining means comprises bit level soft combining means.  A base station comprising the apparatus according to any one of claims 10 to 15.  A communication system comprising the base station according to claim 16.