Multi-Cell Downlink Beamforming: Direct FP, Closed-Form FP, Weighted MMSE

Multi-User in each Cell, MISO
沈闓明代码

Direct FP

K. Shen and W. Yu, “Fractional Programming for Communication Systems—Part I: Power Control and Beamforming,” in IEEE Transactions on Signal Processing, vol. 66, no. 10, pp. 2616-2630, 15 May15, 2018, doi: 10.1109/TSP.2018.2812733.

${\sum }_{n=1}^N{\sum }_{k=1}^K{\log }_2\left(1+\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\right)$
the direct FP approach applies the multidimensional quadratic transform (Theorem 2) to each SINR term.
$f_q\left(\mathbf{W},\mathbf{Y}\right)={\sum }_{(n,k)}\log \left(1+2\mathrm{Re}\left\{y_{n,k}^H{\mathbf{w}}_{n,k}^H{\mathbf{h}}_{n,n,k}\right\}-{\left|y_{n,k}\right|}^2\left({\sum }_{(j,i)\ne (n,k)}{\left|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}\right|}^2+{\sigma }_{n,k}^2\right)\right)$

初始化${\mathbf{w}}_{n,k},\forall n,k$
重复

1. 更新$y_{n,k}^{\star }=\frac{{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}$ with ${\mathbf{w}}_{n,k},\forall n,k$
2. 给定$y_{n,k}$，求解问题，更新${\mathbf{w}}_{n,k}$
   max ⁡ { w n , k , y n , k }      f q ( W , Y ) s . t .    ∑ k = 1 K w n , k H w n , k ≤ p ˉ n , ∀ n = 1 , … , N , \begin{array}{l} \mathop {\max }_{\left\{ {{{\bf{w}}_{n,k}},{y_{n,k}}} \right\}} \;\;{f_q}\left( {{\bf{W}},{\bf{Y}}} \right)\\ {\rm{s}}.{\rm{t}}.\; \sum_{k = 1}^K {{\bf{w}}_{n,k}^H{{\bf{w}}_{n,k}}} \le {{\bar p}_n},\forall n = 1, \ldots ,N, \end{array} max{wn,k​,yn,k​}​fq​(W,Y)s.t.∑k=1K​wn,kH​wn,k​≤pˉ​n​,∀n=1,…,N,​
   the optimization problem is a convex problem of ${\mathbf{w}}_{n,k}$ when the auxiliary variable $y_{n,k}$ is held fixed.

直到$f_q\left(\mathbf{W},\mathbf{Y}\right)$收敛

Closed-Form FP

K. Shen and W. Yu, “Fractional Programming for Communication Systems—Part I: Power Control and Beamforming,” in IEEE Transactions on Signal Processing, vol. 66, no. 10, pp. 2616-2630, 15 May15, 2018, doi: 10.1109/TSP.2018.2812733.

${\sum }_{n=1}^N{\sum }_{k=1}^K{\log }_2\left(1+\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\right)$
Lagrangian Dual Transform (Multidimensional and Complex)
$f_r\left(\mathbf{W},\mathbf{U}\right)={\sum }_{(n,k)}\left(\log \left(1+u_{n,k}\right)-u_{n,k}+\left(1+u_{n,k}\right)\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\right)$
$\frac{\partial }{\partial u_{n,k}}{f}_r\left(\mathbf{W},\mathbf{U}\right)=0$
$u_{n,k}^{\star }={\gamma }_{n,k}=\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\in {\mathbb{R}}^1$
Quadratic Transform (Multidimensional)
$f_q\left(\mathbf{W},\mathbf{U},\mathbf{V}\right)={\sum }_{(n,k)}\left(2\sqrt{(1+u_{n,k})}\mathrm{Re}\left\{{\mathbf{w}}_{n,k}^H{\mathbf{h}}_{n,n,k}{v}_{n,k}\right\}-{\left|v_{n,k}\right|}^2\left({\sum }_{(j,i)}{\left|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}\right|}^2+{\sigma }_{n,k}^2\right)\right)+\mathrm{const}(\mathbf{U})$
$\frac{\partial }{\partial v_{n,k}}{f}_q\left(\mathbf{W},\mathbf{U},\mathbf{V}\right)=0$
$v_{n,k}^{\star }=\frac{\sqrt{(1+u_{n,k})}{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}$

Transformed Problem
max ⁡ { w n , k }      f q ( W , U , V ) s . t .    p ˉ n − ∑ k = 1 K w n , k H w n , k ≥ 0 , ∀ n = 1 , … , N , \begin{array}{l} \mathop {\max }_{\left\{ {{{\bf{w}}_{n,k}}} \right\}} \;\;{f_q}\left( {{\bf{W}},{\bf{U}},{\bf{V}}} \right)\\ {\rm{s}}.{\rm{t}}.\;{{\bar p}_n} - \sum_{k = 1}^K {{\bf{w}}_{n,k}^H{{\bf{w}}_{n,k}}} \ge 0,\forall n = 1, \ldots ,N, \end{array} max{wn,k​}​fq​(W,U,V)s.t.pˉ​n​−∑k=1K​wn,kH​wn,k​≥0,∀n=1,…,N,​

the Lagrangian function

$L(\mathbf{W},\mathbf{U},\mathbf{V},\mathbf{\eta })=f_q(\mathbf{W},\mathbf{U},\mathbf{V})+{\sum }_{n=1}^N{\eta }_n\left({\bar{p}}_n-{\sum }_{k=1}^K{\mathbf{w}}_{n,k}^H{\mathbf{w}}_{n,k}\right)$

the Lagrange dual function: maximizing the Lagrangian

g ( η ) = m a x { w n , k }    L ( W , U , V , η ) g\left( {\bm{\eta}} \right) = \mathop {{\rm{max}}}_{\left\{ {{{\bf{w}}_{n,k}}} \right\}} \;L({\bf{W}},{\bf{U}},{\bf{V}},{\bm{\eta}}) g(η)=max{wn,k​}​L(W,U,V,η)
$\frac{\partial }{\partial {\mathbf{w}}_{n,k}}L(\mathbf{W},\mathbf{U},\mathbf{V},\mathbf{\eta })=0\Rightarrow$
${\mathbf{w}}_{n,k}^{\ast }={\left(\sum _{(m,l)}\left({\mathbf{h}}_{n,m,l}{v}_{m,l}{v}_{m,l}^H{\mathbf{h}}_{n,m,l}^H\right)+{\eta }_n\mathbf{I}\right)}^{-1}\sqrt{(1+u_{n,k})}{\mathbf{h}}_{n,n,k}{v}_{n,k}$

the Lagrange dual problem: minimizing the Lagrange dual function

Lagrange multipliers are component-wise non-negative
${\min }_{\mathbf{\eta }\ge 0}g\left(\mathbf{\eta }\right)$

Closed-Form FP

初始化${\mathbf{w}}_{n,k},\forall n,k$
重复

1. 更新$u_{n,k}^{\star }={\gamma }_{n,k}=\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\in {\mathbb{R}}^1$ with ${\mathbf{w}}_{n,k},\forall n,k$
2. 更新$v_{n,k}^{\star }=\frac{\sqrt{(1+u_{n,k})}{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}$ with ${\mathbf{w}}_{n,k},\forall n,k$ and $u_{n,k},\forall n,k$
3. 更新${\eta }_n,\forall n$。利用二分法可以找到最小的${\eta }_n$ ，使得${\sum }_{k=1}^K{\mathbf{w}}_{n,k}^H\left({\eta }_n\right){\mathbf{w}}_{n,k}\left({\eta }_n\right)={\bar{p}}_n$ 。
   其中，${\mathbf{w}}_{n,k}\left({\eta }_n\right)={\left(\sum _{(m,l)}\left({\mathbf{h}}_{n,m,l}{v}_{m,l}{v}_{m,l}^H{\mathbf{h}}_{n,m,l}^H\right)+{\eta }_n\mathbf{I}\right)}^{-1}\sqrt{(1+u_{n,k})}{\mathbf{h}}_{n,n,k}{v}_{n,k}$
4. 更新${\mathbf{w}}_{n,k}^{\ast }={\left(\sum _{(m,l)}\left({\mathbf{h}}_{n,m,l}{v}_{m,l}{v}_{m,l}^H{\mathbf{h}}_{n,m,l}^H\right)+{\eta }_n\mathbf{I}\right)}^{-1}\sqrt{(1+u_{n,k})}{\mathbf{h}}_{n,n,k}{v}_{n,k}$ with $u_{n,k},\forall n,k$, $v_{n,k},\forall n,k$, and ${\eta }_n,\forall n$

直到$f_q\left(\mathbf{W},\mathbf{U},\mathbf{V}\right)$收敛

对偶变量或Lagrange multipliers的更新：KKT条件 ${\eta }_n\left({\bar{p}}_n-{\sum }_{k=1}^K{\mathbf{w}}_{n,k}^H{\mathbf{w}}_{n,k}\right)=0,\forall n=1,\dots ,N,$
${\sum }_{k=1}^K{\mathbf{w}}_{n,k}^H\left({\eta }_n\right){\mathbf{w}}_{n,k}\left({\eta }_n\right)$是关于${\eta }_n$的单调递减函数，利用二分法可以找到最小的${\eta }_n$ ，使得${\sum }_{k=1}^K{\mathbf{w}}_{n,k}^H\left({\eta }_n\right){\mathbf{w}}_{n,k}\left({\eta }_n\right)={\bar{p}}_n$ ：
当 ${\eta }_n<{\eta }_n^{\ast }$时，${\sum }_{k=1}^K{\mathbf{w}}_{n,k}^H\left({\eta }_n\right){\mathbf{w}}_{n,k}\left({\eta }_n\right)$ >基站最大发射功率${\bar{p}}_n$ 。
当${\eta }_n={\eta }_n^{\ast }$ 时，${\sum }_{k=1}^K{\mathbf{w}}_{n,k}^H\left({\eta }_n\right){\mathbf{w}}_{n,k}\left({\eta }_n\right)$ =基站最大发射功率${\bar{p}}_n$ 。
当 ${\eta }_n>{\eta }_n^{\ast }$时，${\sum }_{k=1}^K{\mathbf{w}}_{n,k}^H\left({\eta }_n\right){\mathbf{w}}_{n,k}\left({\eta }_n\right)$ <基站最大发射功率${\bar{p}}_n$ 。

${\overline{\mathbf{w}}}_{n,k}^{CFFP}=\frac{{\left({\sum }_{(j,i)}{|v_{j,i}|}^2{\mathbf{h}}_{n,j,i}{\mathbf{h}}_{n,j,i}^H+{\eta }_n\mathbf{I}\right)}^{-1}{\mathbf{h}}_{n,n,k}}{\Vert {\left({\sum }_{(j,i)}{|v_{j,i}|}^2{\mathbf{h}}_{n,j,i}{\mathbf{h}}_{n,j,i}^H+{\eta }_n\mathbf{I}\right)}^{-1}{\mathbf{h}}_{n,n,k}\Vert }$
$1+u_{n,k}=\frac{{\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}$

$$\begin{gathered} {\alpha }_{n,k}={\left|v_{n,k}\right|}^2=\frac{(1+u_{n,k}){|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\left({\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2\right)}^2}=\frac{{\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\left({\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2\right)}^2} \\ =\frac{1}{{\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}=\frac{{\gamma }_{n,k}}{{\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2} \end{gathered}$$

${\beta }_{n,k}=\sqrt{(1+u_{n,k})}{v}_{n,k}=\frac{{\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\frac{{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(j,i)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}=\frac{{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}$

WMMSE和Closed-Form FP是等价的

repeat
${\alpha }_{n,k}=\frac{1}{{\sum }_{j,i}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}=\frac{{\gamma }_{n,k}}{{\sum }_{j,i}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}>0$
${\beta }_{n,k}=\frac{{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}\in {\mathbb{C}}^1$
${\mathbf{w}}_{n,k}={\beta }_{n,k}{\left({\sum }_{(j,i)}{\alpha }_{j,i}{\mathbf{h}}_{n,j,i}{\mathbf{h}}_{n,j,i}^H+{\eta }_n\mathbf{I}\right)}^{-1}{\mathbf{h}}_{n,n,k}$
end

其中
${\overline{\mathbf{w}}}_{n,k}={\mathbf{w}}_{n,k}/\Vert {\mathbf{w}}_{n,k}\Vert =\frac{{\left({\sum }_{(j,i)}{\alpha }_{j,i}{\mathbf{h}}_{n,j,i}{\mathbf{h}}_{n,j,i}^H+{\eta }_n\mathbf{I}\right)}^{-1}{\mathbf{h}}_{n,n,k}}{\Vert {\left({\sum }_{(j,i)}{\alpha }_{j,i}{\mathbf{h}}_{n,j,i}{\mathbf{h}}_{n,j,i}^H+{\eta }_n\mathbf{I}\right)}^{-1}{\mathbf{h}}_{n,n,k}\Vert }$
$p_{n,k}={\mathbf{w}}_{n,k}^H{\mathbf{w}}_{n,k}={\Vert {\mathbf{w}}_{n,k}\Vert }^2={\left|{\beta }_{n,k}\right|}^2{\Vert {\left({\sum }_{(j,i)}{\alpha }_{j,i}{\mathbf{h}}_{n,j,i}{\mathbf{h}}_{n,j,i}^H+{\eta }_n\mathbf{I}\right)}^{-1}{\mathbf{h}}_{n,n,k}\Vert }^2>0$

${\gamma }_{n,k}\ll {\bar{p}}_n/{\sigma }_{n,k}^2$
$\sum _{j,i}{\left|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}\right|}^2+{\sigma }_{n,k}^2\ge {\sigma }_{n,k}^2\ll 1$
${\alpha }_{n,k}\ll {\bar{p}}_n/{\sigma }_{n,k}^2/{\sigma }_{n,k}^2\gg {\bar{p}}_n/{\sigma }_{n,k}^2$

Weighted MMSE

hm,n,k denote the downlink channel between BS m and UE k in cell n
wn,k denote the beamformer for UE k in cell n
the optimum Lagrange multiplier ${\mu }_n^{\star }$ can be determined efficiently by a bisection search method.
$u_{n,k}=\frac{{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2}\in {\mathbb{C}}^1$（图中h的下标打错了）
$v_{n,k}=1/\left(1-u_{n,k}^H{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}\right)\in {\mathbb{R}}^1$
${\mathbf{w}}_{n,k}={\left(\sum _{(m,l)}\left(u_{m,l}{u}_{m,l}^H{v}_{m,l}{\mathbf{h}}_{n,m,l}{\mathbf{h}}_{n,m,l}^H\right)+{\eta }_n\mathbf{I}\right)}^{-1}{\mathbf{h}}_{n,n,k}{u}_{n,k}{v}_{n,k}$
${\overline{\mathbf{w}}}_{n,k}^{WMMSE}=\frac{{\left({\sum }_{(j,i)}{|u_{j,i}|}^2{v}_{j,i}{\mathbf{h}}_{n,j,i}{\mathbf{h}}_{n,j,i}^H+{\eta }_n\mathbf{I}\right)}^{-1}{\mathbf{h}}_{n,n,k}}{\Vert {\left({\sum }_{(j,i)}{|u_{j,i}|}^2{v}_{j,i}{\mathbf{h}}_{n,j,i}{\mathbf{h}}_{n,j,i}^H+{\eta }_n\mathbf{I}\right)}^{-1}{\mathbf{h}}_{n,n,k}\Vert }$

其中，${\left|u_{j,i}\right|}^2{v}_{j,i}>0$，因为
$0<\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2}<1$
$v_{n,k}=1/\left(1-\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2}\right)>1>0$

$$\begin{gathered} {\alpha }_{n,k}={\left|u_{n,k}\right|}^2{v}_{n,k}=\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\left({\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2\right)}^2}\frac{1}{1-\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2}} \\ =\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\left({\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2\right)}^2}\frac{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2-{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}=\frac{{\gamma }_{n,k}}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2} \end{gathered}$$

$$\begin{gathered} {\beta }_{n,k}=u_{n,k}{v}_{n,k}=u_{n,k}/\left(1-u_{n,k}^H{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}\right) \\ =\frac{{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2}/\left(1-\frac{{|{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}|}^2}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2}\right) \\ =\frac{{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2}/\frac{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2}{{\sum }_{(m,j)}{|{\mathbf{h}}_{m,n,k}^H{\mathbf{w}}_{m,j}|}^2+{\sigma }_{n,k}^2} \\ =\frac{{\mathbf{h}}_{n,n,k}^H{\mathbf{w}}_{n,k}}{{\sum }_{(j,i)\ne (n,k)}{|{\mathbf{h}}_{j,n,k}^H{\mathbf{w}}_{j,i}|}^2+{\sigma }_{n,k}^2} \end{gathered}$$

原论文

Q. Shi, M. Razaviyayn, Z. -Q. Luo and C. He, “An Iteratively Weighted MMSE Approach to Distributed Sum-Utility Maximization for a MIMO Interfering Broadcast Channel,” in IEEE Transactions on Signal Processing, vol. 59, no. 9, pp. 4331-4340, Sept. 2011, doi: 10.1109/TSP.2011.2147784.

${\mathbf{V}}_{i_k}$ 表示基站k对用户$i_k$ 的波束成形
${\mathbf{H}}_{i_k,j}$ 表示从基站j到用户 $i_k$的信道
$u_{k,i}=\frac{{\mathbf{h}}_{k,k,i}^H{\mathbf{w}}_{k,i}}{{\sum }_{(j,l)}{|{\mathbf{h}}_{j,k,i}^H{\mathbf{w}}_{j,l}|}^2+{\sigma }_{k,i}^2}$