文章快速检索 高级检索

Application of millimeter wave beam coding technology in UAV intelligent swarm
XU Lei, ZHOU Lisha, LI Renjun, GU Cunfeng
Shanghai Institute of Mechanical and Electrical Engineering, Shanghai 201100, China
Abstract: Millimeter wave beam coding technology is considered as an important solution to the UAV intelligent trunking communication network because of its high speed and strong anti-interference ability. However, in the communication scenario of UAV intelligent swarm, the unstable jitter of the body caused by various reasons will cause small angle deflection of the communication beam, resulting in the decline of the communication quality, thus affecting the control and decision-making of the UAV swarm. To solve this problem, this paper proposes an adaptive beam design method for UAV swarm communication. First, the equivalent channel model is established according to the jitter of the beam measured by the sensor. Then, the objective function is established by using the quantized channel model parameters, and the ideal beam coding vector is obtained. On this basis, the geometric greedy algorithm is used to decompose it. The simulation results show that the proposed scheme can effectively improve the average communication rate and reduce the computational complexity compared with other coefficient decomposition algorithms.
Keywords: UAV swarm    millimeter wave    beam coding    beam deflection    adaptive beam

1 通信场景与系统建模 1.1 通信场景

 图 1 无人机集群通信场景 Fig. 1 UAV swarm communication scenario

1.2 等效信道模型

 $\mathit{\boldsymbol{H}} = \sqrt {\frac{{{M_{{\rm{BS}}}}{M_{{\rm{MS}}}}K}}{{1 + K}}} {a_0}{\mathit{\boldsymbol{d}}_{{\rm{BS}}}}(\theta )\mathit{\boldsymbol{d}}_{{\rm{MS}}}^*(\varphi )$ （1）

 ${\mathit{\boldsymbol{d}}_{{\rm{MS}}}}(\varphi ) = {\mathit{\boldsymbol{d}}_{{\rm{MS}}}}({\varphi _{\rm{h}}},{\varphi _{\rm{v}}}) = {\mathit{\boldsymbol{d}}_{\rm{h}}}({\varphi _{\rm{h}}}) \otimes {\mathit{\boldsymbol{d}}_{\rm{v}}}({\varphi _{\rm{v}}})$ （2）

 $\left\{ \begin{array}{l} {\mathit{\boldsymbol{d}}_{\rm{h}}}({\varphi _{\rm{h}}}) = \frac{1}{{\sqrt N }}[1,{{\rm{e}}^{{\rm{j}}k{d_{\rm{h}}}}}^{{\rm{sin}}{\varphi _{\rm{h}}}{\rm{cos}}{\varphi _{\rm{v}}}}, \cdots ,\\ {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {{\rm{e}}^{{\rm{j}}k(N - 1){d_{\rm{h}}}{\rm{sin}}{\varphi _{\rm{h}}}{\rm{cos}}{\varphi _{\rm{v}}}}}{]^{\rm{T}}} \in {{\bf{C}}^N}\\ {\mathit{\boldsymbol{d}}_{\rm{v}}}({\varphi _{\rm{v}}}) = \frac{1}{{\sqrt M }}{[1,{{\rm{e}}^{{\rm{j}}k{d_{\rm{v}}}}}^{{\rm{sin}}{\varphi _{\rm{v}}}}, \cdots ,{{\rm{e}}^{{\rm{j}}k(M - 1){d_{\rm{v}}}{\rm{sin}}{\varphi _{\rm{v}}}}}]^{\rm{T}}} \in {{\bf{C}}^M} \end{array} \right.$ （3）

 $\mathit{\boldsymbol{y}} = {\mathit{\boldsymbol{w}}^*}\mathit{\boldsymbol{Hc}}s + {\mathit{\boldsymbol{w}}^*}\mathit{\boldsymbol{n}}$ （4）

 图 2 混合波束成形结构 Fig. 2 Hybrid beamforming structure

 图 3 集群波束偏转情况 Fig. 3 Swarm beam deflection

 $\begin{array}{*{20}{l}} {{\mathit{\boldsymbol{d}}_{{\rm{MS}}}}(\varphi _{\rm{h}}^l(t),\varphi _{\rm{v}}^l(t)) = {\mathit{\boldsymbol{d}}_{{\rm{MS}}}}(\varphi _{{\rm{v,0}}}^l + {A^l}(t),}\\ {{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} \varphi _{{\rm{h,0}}}^l + {B^l}(t)) = {\mathit{\boldsymbol{d}}_{{\rm{MS,h}}}}(\varphi _{v,0}^l + {A^l}(t)) \otimes }\\ {{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\mathit{\boldsymbol{d}}_{{\rm{MS,v}}}}(\varphi _{{\rm{h,0}}}^l + {B^l}(t))} \end{array}$ （5）

 $\begin{array}{l} {\mathit{\boldsymbol{H}}_l}(t) = \sqrt {\frac{{{M_{{\rm{BS}}}}{M_{{\rm{MS}}}}{K_l}}}{{1 + {K_l}}}} {\mathit{\boldsymbol{d}}_{{\rm{MS}}}}(\varphi _{\rm{h}}^l(t),\varphi _{\rm{v}}^l(t)) \cdot \\ {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} \mathit{\boldsymbol{d}}_{{\rm{BS}}}^*(\theta _{\rm{h}}^l(t),\theta _{\rm{v}}^l(t)) = \sqrt {\frac{{{M_{{\rm{BS}}}}{M_{{\rm{MS}}}}{K_l}}}{{1 + {K_l}}}} \\ {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\mathit{\boldsymbol{d}}_{{\rm{MS,h}}}}(\varphi _{{\rm{h,0}}}^l(t) + {A^l}(t)) \otimes {\mathit{\boldsymbol{d}}_{{\rm{MS,v}}}}(\varphi _{{\rm{v,0}}}^l(t) + \\ {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {B^l}(t))\mathit{\boldsymbol{d}}_{{\rm{BS}}}^*(\theta _{\rm{h}}^l(t),\theta _{\rm{v}}^l(t)) \end{array}$ （6）
1.3 均值通信速率

 $\begin{array}{l} {R_l} = E({R_l}(t)) = E[{\rm{lo}}{{\rm{g}}_2}(|1 + {\rho _l}\mathit{\boldsymbol{R}}_{n,l}^{ - 1}\mathit{\boldsymbol{w}}_l^*{\mathit{\boldsymbol{H}}_l}(t){\mathit{\boldsymbol{c}}_l}\mathit{\boldsymbol{c}}_l^* \cdot \\ {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} \mathit{\boldsymbol{H}}_l^*(t){\mathit{\boldsymbol{w}}_l}|)] = E[{\rm{lo}}{{\rm{g}}_2}(|1 + {\rho _l}\mathit{\boldsymbol{R}}_{n,l}^{ - 1}\mathit{\boldsymbol{u}}_{{\rm{BB}},l}^*\mathit{\boldsymbol{W}}_{{\rm{RF}},l}^* \cdot \\ {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\mathit{\boldsymbol{H}}_l}(t){\mathit{\boldsymbol{F}}_{{\rm{RF}},l}}{\mathit{\boldsymbol{v}}_{{\rm{BB}},l}}\mathit{\boldsymbol{v}}_{{\rm{BB}},l}^*\mathit{\boldsymbol{F}}_{{\rm{RF}},l}^*\mathit{\boldsymbol{H}}_l^*(t){\mathit{\boldsymbol{W}}_{{\rm{RF}},l}}{\mathit{\boldsymbol{u}}_{{\rm{BB}},l}}|)] \end{array}$ （7）

 ${R_{{\rm{group}}}} = \sum\limits_{l = 1}^L {{R_l}}$ （8）

 $\begin{array}{l} (\mathit{\boldsymbol{F}}_{{\rm{RF}},l}^{{\rm{opt}}},\mathit{\boldsymbol{v}}_{{\rm{BB}},l}^{{\rm{opt}}}) = \mathop {{\rm{ argmax }}}\limits_{({\mathit{\boldsymbol{F}}_{{\rm{RF}},l}},{\mathit{\boldsymbol{v}}_{{\rm{BB}},l}})} ({R_{{\rm{ group }}}}) = \sum\limits_{l = 1}^L {\mathop {{\rm{ argmax }}}\limits_{({\mathit{\boldsymbol{F}}_{{\rm{RF}},l}},{\mathit{\boldsymbol{v}}_{{\rm{BB}},l}})} } ({R_l})\\ \begin{array}{*{20}{l}} {{\rm{s}}{\rm{.t}}{\rm{.}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {{[{\mathit{\boldsymbol{F}}_{{\rm{RF}},l}}]}_{:,i}} \in {\mathit{\boldsymbol{F}}_{{\rm{RF}}}}}&{i = 1,2, \cdots ,{N_{{\rm{RF}}}}}\\ {{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} \left\| {{\mathit{\boldsymbol{F}}_{{\rm{RF}},l}},{\mathit{\boldsymbol{v}}_{{\rm{BB}},l}}} \right\|_{\rm{F}}^2 = 1}&{l = 1,2, \cdots ,L} \end{array} \end{array}$ （9）

 $\begin{array}{l} (\mathit{\boldsymbol{F}}_{{\rm{RF}},l}^{{\rm{opt}}},\mathit{\boldsymbol{v}}_{{\rm{BB}},l}^{{\rm{opt}}}) = \mathop {{\rm{ argmax }}}\limits_{({\mathit{\boldsymbol{F}}_{{\rm{RF}},l}},{\mathit{\boldsymbol{v}}_{{\rm{BB}},l}})} {\left\| {{\mathit{\boldsymbol{c}}_{{\rm{opt}},l}} - {\mathit{\boldsymbol{F}}_{{\rm{RF}},l}},{\mathit{\boldsymbol{v}}_{{\rm{BB}},l}}} \right\|_{\rm{F}}}\\ \begin{array}{*{20}{l}} {{\rm{s}}{\rm{.t}}{\rm{.}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {{[{\mathit{\boldsymbol{F}}_{{\rm{RF}},l}}]}_{:,i}} \in {\mathit{\boldsymbol{F}}_{{\rm{RF}}}}}&{i = 1,2, \cdots ,{N_{{\rm{RF}}}}}\\ {{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} \left\| {{\mathit{\boldsymbol{F}}_{{\rm{RF}},l}},{\mathit{\boldsymbol{v}}_{{\rm{BB}},l}}} \right\|_{\rm{F}}^2 = 1}&{l = 1,2, \cdots ,L} \end{array} \end{array}$ （10）
2 混合波束编码算法

 步骤1 FRF=[] 步骤2 Fres=copt, l 步骤3 for i≤NRF do 步骤4 Fq=quantfy(Fres) 步骤5 FRF=FRF    Fq 步骤6 M=max|Fres(n), m=min|Fres(n)| 步骤7 J=find(Fres(J))≥(M+2)/2 步骤8 δ′=mean(Fres(J)/Fq(J)) 步骤9 if  |δ′|>(M+m)/2 then 步骤10 δ=(δ′/|δ′|)[(M+m)/2] 步骤11 else 步骤12 δ=δ′ 步骤13 else if 步骤14 Fres=Fres-δFq 步骤15 end for 步骤16 vBB=[(FRF*FRF)-1FRF*Flopt]/||Flopt||F

 图 4 GG算法残差更新几何方法示意图 Fig. 4 Schematic diagram of geometric method for updating residual of GG algorithm
3 仿真验证

3.1 不稳定角度指向的影响

 图 5 角度偏转对均值频谱效率的影响 Fig. 5 Influence of angle deflection on average spectrum efficiency
3.2 集群通信质量

 图 6 集群的均值总传输速率曲线 Fig. 6 Curves of average total transfer rate of swarm
3.3 计算复杂度

 图 7 运算时间变化 Fig. 7 Variation of operation time

4 结论

 [1] 贾顾伟, 汤俊, 白亮, 等. 面向时间协同的多无人机队形变换最优效率模型[J]. 航空学报, 2019, 40(6): 322599. JIA G W, TANG J, BAI L, et al. Time synergistic optimal efficienvy model for formation transformation of multiple UAVs[J]. Acta Aeronautica et Astronautica Sinica, 2019, 40(6): 322599. (in Chinese) Cited By in Cnki | Click to display the text [2] 陈川, 顾村锋. 基于分布式的隐蔽式探测技术研究[J]. 空天防御, 2018, 1(4): 48-51. CHEN C, GU C F. Concealed detection research based on distributed techniques[J]. Air & Space Defense, 2018, 1(4): 48-51. (in Chinese) Cited By in Cnki | Click to display the text [3] ZENG Y, ZHANG R, TENG J L. Wireless communications with unmanned aerial vehicles:Opportunities and challenges[J]. IEEE Communications Magazine, 2016, 54(5): 36-42. Click to display the text [4] SUN K P. Cooperative UAV search and intercept[D]. Toronto: University of Toronto, 2009. [5] XIAO Z, XIA P, XIA X G. Enabling UAV cellular with millimeter-wave communication:Potentials and approaches[J]. IEEE Communications Magazine, 2016, 54(5): 66-73. Click to display the text [6] SONG J, CHOI J, LOVE D J. Common codebook millimeter wave beam design:Designing beams for both sounding and communication with uniform planar arrays[J]. IEEE Transactions on Communications, 2017, 65(4): 1859-1872. Click to display the text [7] NOH S, ZOLTOWSKI M D, LOVE D J. Multi-resolution codebook and adaptive beamforming sequence design for millimeter wave beam alignment[J]. IEEE Transactions on Wireless Communications, 2015, 16(9): 5689-5701. Click to display the text [8] ALKHATEEB A, AYACH O E, LEUS G, et al. Channel estimation and hybrid precoding for millimeter wave cellular systems[J]. IEEE Journal of Selected Topics in Signal Processing, 2017, 8(5): 831-846. Click to display the text [9] ZHAO J, GAO F, WU Q, et al. Beam tracking for UAV mounted SatCom on-the-move with massive antenna array[J]. IEEE Journal on Selected Areas in Communications, 2018, 36(2): 363-375. Click to display the text [10] HUR S, KIM T, LOVE D J, et al. Millimeter wave beamforming for wireless backhaul and access in small cell networks[J]. IEEE Transactions on Communications, 2013, 61(10): 4391-4403. Click to display the text [11] ZHONG W, XU L, LU X, et al. Research on millimeter wave communication interference suppression of UAV based on beam optimization[C]//International Conference on Machine Learning and Intelligent Communications.Berlin: Springer, 2017. [12] 郑婷婷.面向高速移动平台的毫米波波束成形与波束追踪技术研究[D].北京: 北京邮电大学, 2018. ZHENG T T. Millimeter wave beamforming and beam tracking for high speed mobile platforms[D]. Beijing: Beijing University of Posts and Telecommunications, 2018(in Chinese). Cited By in Cnki | Click to display the text [13] KAUSHIK A, THOMPSON J, YAGHOOBI M. Sparse hybrid precoding and combining in millimeter wave MIMO systems[C]//Radio Propagation & Technologies for 5G. London: IET, 2017. [14] 仲伟志, 徐磊, 朱秋明, 等. 面向天线互耦效应抑制的波束预编码算法[J]. 西安电子科技大学学报, 2018, 45(6): 86-91. ZHONG W Z, XU L, ZHU Q M, et al. Beam precoding algorithm for antenna mutual coupling effect suppression[J]. Journal of Xidian University, 2018, 45(6): 86-91. (in Chinese) Cited By in Cnki | Click to display the text [15] ZHONG W Z, XU L. Research on millimeter wave communication interference suppression of UAV based on beam optimization[C]//International Conference on Machine Learning and Intelligent Communications, 2018: 472-481. [16] 徐俊.多无人机的组群飞行特性与控制分析[D].南京: 南京理工大学, 2017. XU J. Group flight characteristics and control analysis of multiple UAVs[D]. Nanjing: Nanjing University of Science & Technology, 2017. Cited By in Cnki | Click to display the text [17] 蔡青松.毫米波混合波束成形技术研究[D].西安: 西安电子科技大学, 2015. CAI Q S. Research on millimeter wave hybrid beamforming technology[D]. Xi'an: Xidian University, 2015(in Chinese). Cited By in Cnki (1) | Click to display the text [18] AYACH O E, RAJAGOPAL S, ABU-SURRA S, et al. Spatially sparse precoding in millimeter wave MIMO systems[J]. IEEE Transactions on Wireless Communications, 2014, 13(3): 1499-1513. Click to display the text
http://dx.doi.org/10.7527/S1000-6893.2019.23754

0

#### 文章信息

XU Lei, ZHOU Lisha, LI Renjun, GU Cunfeng

Application of millimeter wave beam coding technology in UAV intelligent swarm

Acta Aeronautica et Astronautica Sinica, 2020, 41(S1): 723754.
http://dx.doi.org/10.7527/S1000-6893.2019.23754