常用算法的计算复杂度

穷举法 (Method of Exhaustion)

穷举法的计算复杂度就是穷举次数 N N N

二分法 (Bisection)

二分法的计算复杂度是 O ( log ⁡ 2 ( L 0 / e ) ) \mathcal{O} (\log_2(L_0/e)) O(log2(L0/e)) , 其中 L 0 L_0 L0 是最初的长度间隔, e e e 是要求的精度。

线性规划(LP)

The complexity of a standard linear problem is of order O ( a 2 b ) \mathcal{O} (a^2b) O(a2b), where a a a is the number of variables and b b b is the number of constraints.
For instance, the problem (P2.1) in ref [2], it is easy to show that the complexity of (P2.1) is of order O ( ( M N + K N + K + K M N ) ( K M ) 2 ) \mathcal{O} \left ( (MN + KN + K+ KMN)(KM)^2 \right) O((MN+KN+K+KMN)(KM)2).
Note that (P2.1) is a linear function which only consists of the scheduling variables.

  • 变量个数:KM,为什么不是KMN?
  • 约束(10) 的个数 M N + K N MN+KN MN+KN
  • 约束(11) 的个数 K K K
  • 约束(14) 的个数 K M N KMN KMN

半定规划(SDP)

解决半定规划问题的计算复杂度为 O ( ( M + 1 ) 4.5 ) \mathcal{O} ((M+1)^{4.5}) O((M+1)4.5),其中 M M M代表变量个数 【8】。

逐次凸逼近(SCA)

SCA算法的计算复杂度和 “迭代次数( I I I)”,以及每次迭代中的 “更新变量的个数( N K NK NK)”有关;计算复杂度的表达式为 O ( I ( N K ) 3 ) \mathcal{O} (I(NK)^3) O(I(NK)3).

块坐标下降(BCD)

BCD algorithm 即 Alternating algorithm.

  • BCD算法的收敛要求每一个块的优化要达到最优解,而实际问题中每个块问题往往是非凸的,用凸优化技术SCA求解得到的往往是次优解,所以在证明算法的收敛性的时候,需要自行证明一下优化目标函数的非减、非增,证明是存在 upper bound 或 lower bound,如此可证明问题收敛。
  • 当BCD算法的某个块问题是采用SCA算法来解决的,则该块算法的计算复杂度和解决单独问题的SCA算法一样:计算复杂度为 O ( N 3.5 ) \mathcal{O} (N^{3.5}) O(N3.5) 【1】, N N N为变量个数。
  • 在保证问题收敛后,解决每个块问题的算法的复杂度进行相加,即为BCD算法的复杂度。
  • 如果算法的外部有循环 N o u t N_{out} Nout时,还需要在内部循环算法复杂度的基础上乘以 N o u t N_{out} Nout

内点法(Interior Point Method)

内点法的计算复杂度为 O ( n 3.5 log ⁡ ( 1 / ϵ ) ) \mathcal{O} (n^{3.5}\log(1/ \epsilon)) O(n3.5log(1/ϵ)),其中 n n n 为变量维度, ϵ \epsilon ϵ 为目标精度 【7】。

待整改

参考文献

【1】S. Boyd and L. Vandenberghe, Convex Optimization. Cambridge University Press, 2004.
【2】A. Bejaoui, K. Park and M. Alouini, “A QoS-Oriented Trajectory Optimization in Swarming Unmanned-Aerial-Vehicles Communications,” in IEEE Wireless Communications Letters, vol. 9, no. 6, pp. 791-794, June 2020, doi: 10.1109/LWC.2020.2970052.
【3】With Wireless Power Transfer," in IEEE Internet of Things Journal, vol. 8, no. 10, pp. 7833-7848, 15 May15, 2021, doi: 10.1109/JIOT.2020.3041303.
【4】C. Zhan and Y. Zeng, “Aerial–Ground Cost Tradeoff for Multi-UAV-Enabled Data Collection in Wireless Sensor Networks,” in IEEE Transactions on Communications, vol. 68, no. 3, pp. 1937-1950, March 2020, doi: 10.1109/TCOMM.2019.2962479.
【5】 R. Duan, J. Wang, C. Jiang, H. Yao, Y. Ren and Y. Qian, “Resource Allocation for Multi-UAV Aided IoT NOMA Uplink Transmission Systems,” in IEEE Internet of Things Journal, vol. 6, no. 4, pp. 7025-7037, Aug. 2019, doi: 10.1109/JIOT.2019.2913473.
【6】Q. Hu, Y. Cai, A. Liu, G. Yu and G. Y. Li, “Low-Complexity Joint Resource Allocation and Trajectory Design for UAV-Aided Relay Networks With the Segmented Ray-Tracing Channel Model,” in IEEE Transactions on Wireless Communications, vol. 19, no. 9, pp. 6179-6195, Sept. 2020, doi: 10.1109/TWC.2020.3000864.
【7】https://blog.csdn.net/weixin_39274659/article/details/121834231
【8】Luo Z Q , Ma W K , So M C , et al. Semidefinite Relaxation of Quadratic Optimization Problems[J]. IEEE Signal Processing Magazine, 2010, 27(3):20-34.

SDP算法(半定规划松弛)是一种用于求解二次优化问题的方。它使用半正定松弛来将原始的离散问题转化为半定规划问题。引用中提到了一种基于块坐标下降方的SDP求解算法,旨在降低计算复杂度。此外,引用和引用提供了一些经典著作和博客文章,可供进一步了解SDP算法的详细细节和示例代码。<span class="em">1</span><span class="em">2</span><span class="em">3</span> #### 引用[.reference_title] - *1* [SDP半正定规划的低复杂度求解:基于块坐标下降(Block Coordinate Descent)](https://blog.csdn.net/weixin_39274659/article/details/121834231)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v92^chatsearchT3_1"}}] [.reference_item style="max-width: 33.333333333333336%"] - *2* [经典的SDR算法: 用半正定松弛 ( Semidefinite Relaxation) 求解二次优化问题](https://blog.csdn.net/shixin_0125/article/details/125117466)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v92^chatsearchT3_1"}}] [.reference_item style="max-width: 33.333333333333336%"] - *3* [半正定松弛(SDR)2021-11-06](https://blog.csdn.net/weixin_44655342/article/details/121175761)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v92^chatsearchT3_1"}}] [.reference_item style="max-width: 33.333333333333336%"] [ .reference_list ]
评论 7
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值