COORDINATED LINEAR BEAMFORMING IN DOWNLINK MULTI-CELL WIRELESS NETWORKS

US 20090296650A1
Filed: 06/03/2009
Published: 12/03/2009
Est. Priority Date: 06/03/2008
Status: Active Grant

First Claim

Patent Images

1. ) A method to optimize wireless communication for a plurality of mobile wireless devices, comprising:

a. determining a structure for downlink beamforming vectors optimizing the weighted sum-rate across multiple orthogonal resource slots in the time, frequency or code domain; and

b. distributing a non-convex optimization exploiting the determined structure over a plurality of base stations.

View all claims

2 Assignments

Timeline View

Assignment View

0 Petitions

Accused Products

Abstract

System and methods are disclosed for optimizing wireless communication for a plurality of mobile wireless devices. The system uses beamforming vectors or precoders having a structure optimal with respect to the weighted sum rate in a multi-cell orthogonal frequency division multiple access (OFDMA) downlink. A plurality of base stations communicate with the mobile devices and all base stations perform a distributed non-convex optimization exploiting the determined structure.

88 Citations

View as Search Results

27 Claims

1. ) A method to optimize wireless communication for a plurality of mobile wireless devices, comprising:
- a. determining a structure for downlink beamforming vectors optimizing the weighted sum-rate across multiple orthogonal resource slots in the time, frequency or code domain; and
  
  b. distributing a non-convex optimization exploiting the determined structure over a plurality of base stations.
- View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26)
- - 2. ) The method of claim 1, wherein the non-convex optimization is responsive to one or more of:
    - a set of channel estimates;
      
      user priorities; and
      
      maximum transmit power available at each base-station.
  - 3. ) The method of claim 1, wherein the non-convex optimization utilizes Karush-Kuhn-Tucker (KKT) equations.
  - 4. ) The method of claim 1, wherein the distributing over the base stations includes providing a small number of channel estimates at each base station and limited message passing among adjacent base stations.
  - 5. ) The method of claim 1, comprising obtaining channel estimates.
  - 6. ) The method of claim 1, comprising obtaining predicted channel estimates responsive to an average scheduling delay parameter.
  - 7. ) The method of claim 1, wherein the mobile wireless device comprises a multiple-input single-output (MISO) system wherein each user is equipped with a single receive antenna and receives a single data stream from its serving base station via space-division multiple-access (SDMA).
  - 8. ) The method of claim 1, comprising determining the optimal structure of the beamforming vectors optimizing the weighted sum-rate in a MISO system.
  - 9. ) The method of claim 8, comprising determining a leakage matrix.
  - 10. ) The method of claim 9, wherein determining a leakage matrix L_m,k(n) comprises:
    - $L_{m, k} (n) = \sum_{j = 1}^{M} \sum_{\underset{(j, u) \neq (m, k)}{u \in B_{j} (n)}} P_{j, u} (n) G_{m, u} (n) \in C^{P}$ $where$ $P_{j, u} (n) = \frac{α_{u} (n) S I N R_{j, u} (n)}{1 + \sum_{l = 1}^{M} \sum_{s \in B_{l} (n)} w_{l, s}^{H} (n) G_{l, u} (n) w_{l, s} (n)}$ B_m(n) is the set of users served by base station m on slot n;
      
      N is the total number of orthogonal resource slots;
      
      M is the number of coordinated base stations;
      
      P is the number of transmit antennas at each base station;
      
      w_m,k(n)ε
      
      C^Pdenotes the P-dimensional beam vector used by the base station m to serve user kε
      
      B_m(n) on slot n;
      
      W={w_m,k(n), kε
      
      B_m(n), m=1, . . . , M, n=1, . . . N} indicates the collection of all beam vectors;
      
      α
      
      _k(n)>
      
      0 is the priority assigned by the scheduler to user k on slot n;
      
      P_m,maxis the maximum transmit power of base station m;
      
      SINR_m,k(n) is the signal-to-interference-plus-noise ratio for user k served by base station m on slot n and is given by $S I N R_{m, k} (n) = \frac{w_{m, k}^{H} (n) G_{m, k} (n) w_{m, k} (n)}{1 + \sum_{j = 1}^{M} \sum_{\underset{(j, u) \neq (m, k)}{u \in B_{j} (n)}} w_{j, u}^{H} (n) G_{j, k} (n) w_{j, u} (n)}$ $with G_{m, k} (n) = h_{m, k} (n) h_{m, k}^{H} (n) \in C^{P \times P} .$ h_m,k(n)ε
      
      C^Pis the P-dimensional channel vector between base station j and user k and slot n (which includes small-scale fading, large scale fading and path attenuation) normalized by the standard deviation of the received noise.
  - 11. ) The method of claim 10, comprising determining $i_{m, k}$
    - ( n ) = ∑
      
      l = 1 M 
      
      ∑
      
      s ∈
      
      B l 
      
      ( n ) ( l , s ) ≠
      
      ( m , k ) 
      
      w l , s H 
      
      ( n ) 
      
      G l , k 
      
      ( n ) 
      
      w l , s 
      
      ( n ) , 
      
      k ∈
      
      B m 
      
      ( n ) , m = 1 , …
      
      
      
      , M , n = 1 , …
      
      
      
      
      
      N .
  - 12. ) The method of claim 8, comprising determining λ
    - ₁, . . . , λ
      
      _Mand ${{\hat{B}}_{m, k} (n; λ_{m}), k \in B_{m} (n), n = 1, \dots N}$ $using {\langle {\hat{β}}_{m, k} (n; λ_{m}) \rangle}^{2} = \frac{{(\begin{matrix} α_{k} (n) h_{m, k}^{H} (n) {\hat{T}}_{m, k}^{†} (n; λ_{m}) \\ h_{m, k} (n) - {\hat{i}}_{m, k} (n) - 1 \end{matrix})}^{+}}{{\langle h_{m, k}^{H} (n) {\hat{T}}_{m, k}^{†} (n; λ_{m}) h_{m, k} (n) \rangle}^{2}}$ $and \sum_{n = 1}^{N} \sum_{k \in B_{m} (n)} \underset{\underset{f_{m} (n; λ_{m})}{}}{{ {\hat{β}}_{m, k} (n; λ_{m}) {\hat{T}}_{m, k}^{†} (n; λ_{m}) h_{m, k} (n) }^{2}} = P_{m, \max}$ where {circumflex over (T)}_m,k(n;
      
      λ
      
      _m)={circumflex over (L)}_m,k(n;
      
      λ
      
      _m)+(λ
      
      _mln
      
      2)I_p, (•
      
      )^†indicates the pseudo-inverse and x⁺=max{x,0} and where {î
      
      _m,k(n),{circumflex over (L)}_m,k(n)} denotes most recent values of {i_m,k(n),L_m,k(n)}.
  - 13. ) The method of claim 8, comprising updating beam vectors as
    w_m,k(n)={circumflex over (β
    - )}_m,k(n;
      
      λ
      
      _m){circumflex over (T)}_m,k^†(n;
      
      λ
      
      _m)h_m,k(n).
  - 14. ) The method of claim 8, comprising choosing an initial feasible set of beam vectors.
  - 15. ) The method of claim 14, wherein the initial feasible set is selected after splitting power across the available slots.
  - 16. ) The method of claim 14, comprising performing channel-matched beamforming to select the initial feasible set of beam vectors.
  - 17. ) The method of claim 14, comprising performing in-cell zero-forcing beamforming to select the initial feasible set of beam vectors.
  - 18. ) The method of claim 14, comprising performing maximum signal-to-leakage-plus-noise ratio (MSLNR) beamforming to select the initial feasible set of beam vectors.
  - 19. ) The method of claim 1, comprising providing an equivalent MISO system wherein each mobile user has multiple-receive antennas and performs rank-one receive beamforming before detection.
  - 20. ) The method of claim 1, wherein the mobile wireless device comprises a multiple-input multiple-output (MIMO) system and wherein each user is equipped with multiple receive antennas and receives multiple data streams from a serving base station via linear precoding.
  - 21. ) The method of claim 1, comprising determining an optimal structure of the linear precoders in a MIMO system.
  - 22. ) The method of claim 20, comprising determining a leakage matrix.
  - 23. ) The method of claim 22, wherein determining a leakage matrix L_m,k(n) comprises determining:
    - $L_{m, k} (n) = \sum_{j = 1}^{M} \sum_{\underset{(j, u) \neq (m, k)}{u \in B_{j} (n)}} \frac{α_{u} (n)}{\ln (2)} H_{m, u}^{H} (n) ({R_{j, u} (n)}^{- 1} - {(\begin{matrix} R_{j, u} (n) + H_{j, u} (n) \\ Q_{j, u} (n) H_{j, u}^{H} (n) \end{matrix})}^{- 1}) H_{m, u} (n), where$ $R_{m, k} (n) = \sum_{j = 1}^{M} \sum_{\underset{(j, u) \neq (m, k)}{u \in B_{j} (n)}} H_{j, k} (n) Q_{j, u} (n) H_{j, k}^{H} (n) + I_{N_{k}} .$ B_m(n) is the set of users served by base station m on slot n;
      
      N is the total number of orthogonal resource slots;
      
      M is the number of coordinated base stations;
      
      P is the number of transmit antennas at each base station;
      
      N_kis the number of receive antennas of user k, Q_j,u(n)=W_j,u(n)W_j,u^H(n)ε
      
      C^P×
      
      Pis a covariance matrix to be optimized;
      
      W_m,k(n)ε
      
      C^P×
      
      D^m,k⁽ⁿ⁾, and 1≦
      
      D_m,k(n)≦
      
      P indicates the number of data stream delivered to user kε
      
      B_m(n) on slot n by base station m.;
      
      Q={Q_m,k(n)ε
      
      Q, kε
      
      B_m(n), m=1, . . . , M, n=1, . . . N} indicates the collection of the positive-semidefinite covariance matrices;
      
      α
      
      _k(n)>
      
      0 is the priority assigned by the scheduler to user k on slot n;
      
      P_m,maxis the maximum transmit power of base station m; and
      
      H_j,k(n)ε
      
      C^N^k^×
      
      Pis the channel matrix between base station j and user k and slot n (which includes small-scale fading, large scale fading and path attenuation) normalized by the standard deviation of the received noise.
  - 24. ) The method of claim 20, comprising solving a following convex optimization problem at each base-station m, for a given {L_m,k(n)} and {R_m,k(n)}:
    - $\underset{{Q_{m, k} (n)}}{argmax} \sum_{n = 1}^{N} \sum_{k \in B_{m} (n)} [\begin{matrix} α_{k} (n) \log_{2} \langle \begin{matrix} I_{N_{k}} + H_{m, k} (n) Q_{m, k} (n) \\ H_{m, k}^{H} (n) {R_{m, k} (n)}^{- 1} \end{matrix} \rangle - \\ tr (L_{m, k} (n) Q_{m, k} (n)) \end{matrix}], s . t . \sum_{n = 1}^{N} \sum_{k \in B_{m} (n)} tr (Q_{m, k} (n)) \leq P_{m, \max}, Q_{m, k} (n) \in \overline{Q}, \forall k \in B_{m} (n) \forall n$
  - 25. ) The method of claim 20, comprising selecting an initial feasible set of positive semi-definite covariance matrices after splitting power across available slots.
  - 26. ) The method of claim 25, comprising performing Maximum Signal-to-Leakage-plus-Noise Ratio (MSLNR) precoding to select the initial feasible set of positive semi-definite covariance matrices.

27. ) A communication system, comprising:
- a. a plurality of mobile devices using linear precoding matrices having a predetermined structure optimal with respect to a weighted sum rate in a multi-cell orthogonal frequency division multiple access (OFDMA) downlink; and
  
  b. a plurality of base stations communicating with the mobile devices, all base stations performing a distributed a non-convex optimization exploiting the predetermined structure.

Specification

Resources

Litigation Campaign Assessment

Current Assignee
NEC Corporation
Original Assignee
NEC Laboratories America Inc (NEC Corporation)
Inventors
Wang, Xiaodong, Prasad, Narayan, Venturino, Luca

Granted Patent

US 8,218,422 B2
Time in Patent Office

Days
Field of Search
US Class Current

370/330
CPC Class Codes

H04W 16/28 using beam steering

COORDINATED LINEAR BEAMFORMING IN DOWNLINK MULTI-CELL WIRELESS NETWORKS

First Claim

2 Assignments

0 Petitions

Accused Products

Abstract

88 Citations

27 Claims

Specification

Solutions

Use Cases

Quick Links

COORDINATED LINEAR BEAMFORMING IN DOWNLINK MULTI-CELL WIRELESS NETWORKS

First Claim

2 Assignments

Subscription Required

Subscription Required

0 Petitions

Subscription Required

Accused Products

Subscription Required

Abstract

88 Citations

27 Claims

Specification

Subscription Required

Solutions

Use Cases

Quick Links