Certains contenus de cette application ne sont pas disponibles pour le moment.
Si cette situation persiste, veuillez nous contacter àObservations et contact
Note: Texte fondé sur des processus automatiques de reconnaissance optique de caractères. Seule la version PDF a une valeur juridique



This application claims the benefit of U.S. Provisional Application No. 62/185,675, filed on Jun. 28, 2015.


[0001] The disclosed inventions relate generally to wireless communication, and in particular, to the mechanism for a Base Station (BS) to acquire the Channel State Information (CSI) of each User Equipment (UE) in a wireless communication system with reciprocal channels.


[0002] In a large-scale Multiple-Input Multiple- Output (MIMO) or massive MIMO system, the BS is equipped with tens to several hundreds of transmitting antennas. It has received enormous attention due to its capability to provide linear capacity growth without the need of increased power or bandwidth. This is realized by employing Multi-User MIMO (MU-MIMO) to serve multiple UEs simultaneously in the same Resource Blocks (RBs). In this system, the BS groups UEs at each scheduling slot and transmits data to them on the same time and frequency resource. In order to remove the mutual interference among these UEs and maximize the multi-user sum rate, the BS needs to know the CSI in the downlink and the Channel Quality Information (CQI) of each UE in particular. However, it is infeasible to obtain the CSI directly by sending reference pilots in the downlink because of two reasons: 1. the large number of antennas would cause a very large system overhead for reference signals in the downlink; 2. A larger number of bits is needed to quantize the CSI accurately, which would cause overload of the feedback channel in the uplink. As a result, the reciprocal property of a wireless channel, such as in a TDD system or in an FDD system using switching to create channel reciprocity between the uplink and downlink as described in our patent application PCT/US 14/71752, can be employed to reduce the reference signal overhead. In such a system, a UE sends a Sounding Reference Signal (SRS) to the BS in the uplink. The BS then estimates the uplink CSI through the received SRSs and uses it to estimate the downlink CSI based on channel reciprocity.


[0003] This application provides a method to allocate radio resources used for the SRS, with multiplexing patterns of SRSs of multiple UEs. In this method, some special symbols in an uplink subframe are reserved for SRS. In addition, to avoid the inter-cell interference, the radio resources of these symbols are multiplexed among the neighboring BSs through Time-

Division Multiplexing (TDM) or Frequency-Division Multiplexing (FDM). Moreover, FDM or Code-Division Multiplexing (CDM) is used to multiplex the resources among UEs belonging to the same BS.


[0004] The aforementioned implementation of the invention as well as additional implementations will be more clearly understood as a result of the following detailed description of the various aspects of the invention when taken in conjunction with the drawings. Like reference numerals refer to corresponding parts throughout the several views of the drawings.

[0005] Fig. 1 shows the block diagram illustrating the frame structure and the locations of SRSs in a TDD wireless communication system.

[0006] Fig. 2 shows the block diagram illustrating the time-frequency structure of one SRS symbol.

[0007] Fig. 3 shows the block diagram illustrating one deployment example where TDM is used to multiplex the SRS symbols among three neighboring BSs.

[0008] Fig. 4 shows the block diagram illustrating one deployment example where FDM is used to multiplex the SRS symbols among three neighboring BSs.

[0009] Fig. 5 shows the block diagram illustrating the use of FDM to multiplex the SRS Resource Units (RUs) among UEs belonging to the same BS.

[0010] Fig. 6 shows the flowchart illustrating the process of a BS grouping UEs sending SRSs on the same resources.


[0011] Fig. 1 illustrates the frame resource structure of a wireless communication system. In such a system, one radio frame 1 includes K subframes 2, where each subframe includes Nsym consecutive Orthogonal Frequency-Division Multiplexing (OFDM) symbols 3. Among the K subframes, K2 subframes are used for uplink signal transmission, i.e., uplink subframes, while the rest K3 = K — K2 subframes are used for downlink signal transmission, i.e., downlink subframes, in TDD systems. A total of NsRS consecutive OFDM symbols are reserved for SRS symbols 4 in one of the uplink subframes. Fig. 2 illustrates the time-frequency structure of a SRS symbol. In the frequency domain, the iVused usable subcarriers 5 of an OFDM symbol are divided in to M segments with L = Nuse^/M subcarriers in each segment, where L is assumed to be an integer. A segment is called an SRS RU 6.

[0012] The SRSs are transmitted in an uplink subframe with a period of T subframes. These NsRS consecutive OFDM symbols may be chosen as the first or the last NsRS OFDM symbols of an uplink subframe. These iVSRS consecutive symbols are called the SRS region. Two methods are presented to share the SRS region by the neighboring NBS BSs in order to avoid inter-cell interference.

[0013] Method- 1.

[0014] In Method-1, every NSRS/NBS SRS symbols are reserved for a different BS, e.g., in Fig. 3, the SRS symbols 40-2 are used for the BSs 0-2 respectively where iVSRS = ^BS = 3 in this example. Note that NsRS and NBS are properly chosen to ensure that NSRS/NBS is an integer. One way to determine the SRS symbol indices is that the SRS symbols with indices from— (BS_ID mod JVBS) to— [(BS_ID mod JVBS) + 1] - 1 are reserved for the BS with the identification number BS_ID, where mod represents the modular arithmetic operation.

[0015] Method-2.

[0016] In Method-2, all the iVSRS SRS symbols are divided into NBS segments in the frequency domain, where each segment includes M/NBS SRS RUs. Note that M and NBS are properly chosen to ensure that M/NBS is an integer. Then, each segment is reserved for a different BS. One way to determine the segment index used by the BS with the identification number BS_ID in the nsvm th SRS symbol is ( BSID + — — ) mod iVBS , nsvm =


0, -" , NSRS— 1 . Fig. 4 illustrates an example where NBS = 3 and NsRS = 3 hence each segment includes 1 SRS RU 6.

[0017] A SRS channel is defined as the radio resource unit on which the SRS is carried so that the channel between each transmitting antenna of each UE and all receiving antennas of the associated BS could be estimated.

[0018] Because of the flat fading feature, channel coefficients of all subcarriers in one SRS RU are almost the same, hence only one SRS channel is needed for one transmitting antenna of a UE. This means that one SRS RU can provide L SRS channels. In order to avoid the inter-channel interference, two methods are proposed to multiplex these L SRS channels.

[0019] Method-1 FDM.

[0020] In Method-1, each SRS channel is assigned for a different subcarrier in one SRS RU. Fig. 5 illustrates an example of this option where each SRS RU 6 provides L SRS channels for L UEs with each SRS channel being assigned for a different subcarrier 5 in one SRS RU.

[0021] Method -2 CDM.

[0022] In Method-2, each SRS channel uses all the L subcarriers in one SRS RU, while the signal sequences transmitted by the L SRS channels are mutually orthogonal, e.g., the signal sequences for the L channels are s1( ··· , sL, and they satisfy the constraints

where the function (a, b) denotes the inner product of vectors a and b in the complex Hilbert space. The L signal sequences constitute a sequence set.

[0023] To homogenize the interference from neighboring BSs on each SRS channel, d different sequence sets where d < L + 1 are predefined so that the correlation coefficient between the kth sequence of the ith set skl and the Zth sequence of the y'th set satisfies

where Ski is the Kronecker delta function. The mapping relation between the sequence set index and the BS_ID should follow some predefined functions, e.g., setting the index of the sequence set allocated to the BS with identification number BS_ID as QBSJD/LJ mod d) . The tables I and II provide examples of sequence sets when L = 4 and L = 8 respectively.

[0024] The BS determines the number of SRS channels allocated to each UE according to two factors: 1. The number of receiving antennas of a UE nrx ; 2. The bandwidth, i.e., nRU consecutive SRS RUs in the frequency domain, allocated to a UE in the downlink transmission. Then, the number of SRS channels allocated to the UE is calculated as SRSchan = rx RU > and the location information of the SRS channels is informed to the UE through the downlink control channel.

[0025] The BS determines the UEs allocated to the same SRS RU according to the estimated CQI as shown in Fig. 6. Specifically, after the process begins 7, the BS first determines the CQI of each UE through the uplink signals, e.g., ranging signals and control channel signals 8. Then, the BS groups the UEs according to their CQI 9. After that, the UEs in the same group are allocated to the same SRS RU 10, before the process ends 11. For example, the UEs among which the maximum CQI difference is less than a predefined threshold value are allocated into the same SRS RU. The BS could estimate the CQI of UEs through the uplink control channels, e.g., the physical ranging channel and the physical uplink control channel.

Table I. The sequence set with L = 4.

Table II. The sequence set with L = 8 .