认知无线网络中的单用户与多用户频谱功率协同分配算法优化
35 浏览量
更新于2024-08-28
1
收藏 1.72MB PDF 举报
认知无线电网络(Cognitive Radio, CR)是现代无线通信领域的一个重要分支，其核心思想是允许非授权用户(认知用户)在未被主要用户占用的频谱资源上进行通信，从而提高频谱利用效率。本文主要关注的是在CR网络中，如何在满足主要用户低干扰的同时，合理分配认知用户的功率和频谱资源，以实现通信效用的最大化。
首先，研究者利用著名的注水定理(Water-Filling Algorithm)，提出了一种针对单个认知用户的新型功率分配算法。注水定理是一种经典的资源分配策略，在信号处理中常用于功率分配，以最大化系统的信道容量，同时考虑到干扰约束。在这个算法中，通过对认知用户施加适当的干扰限制，确保主要用户的服务质量，同时最大化认知用户的可用资源。
接着，文章进一步探讨了在有两个认知用户的场景下，通过考虑认知用户的信噪比增益，设计了一个基于最优频率划分的子载波和功率分配算法。这种方法旨在优化频谱利用，确保每个用户在最佳的频段上运行，从而提升整体系统的性能。
随着研究的深入，作者将这个单用户策略扩展到了多用户环境，提出了两种不同的频谱共享策略：贪婪算法和并行算法。贪婪算法通常在每一阶段选择局部最优解，而并行算法则可能涉及到全局搜索，以期找到更优的解决方案。这两种算法在处理多用户竞争时，能够平衡各个用户的需求，提高整体的频谱和功率分配效率。
理论分析与仿真实验的结果显示，这些算法不仅有效地保护了主要用户的权益，降低了对它们的干扰，而且能有效解决认知用户在有限功率和频谱资源下的分配问题。这表明，通过合理的联合功率和频谱分配策略，认知无线电网络能够在频谱稀缺的环境中实现高效、公平的资源共享。
总结来说，本文的主要贡献在于提出了一种结合了注水定理和频率优化的联合功率和频谱分配方法，适用于多用户认知无线电网络，这为改善网络性能，降低干扰，提升系统整体效用提供了实用的策略。随着未来无线通信对频谱效率的要求不断提高，这种创新的方法具有重要的实际应用价值。
Yutao Liu et al.: Joint power and spectrum allocation algorithm in cognitive radio networks 693
max
p
i,j
U =
N
i=1
U
i
=
N
i=1
M
i
j=1
B
M
log
2
(1 + K
i
· γ
i,j
)
s.t.
M
i
j=1
p
i,j
P
i
,p
i,j
0, ∀i ∈ N
σ
2
+ p
j
H
j,0
I
j
, ∀j ∈ M and j is on
(5)
where p
j
is the cognitive user’s total transmission power
in the jth subcarrier, and H
j,0
is the channel gain from the
cognitive user’s transmitter to the primary base station in
subcarrier j.
Because each subcarrier can support only one cogni-
tive user besides the primary users, the spectrum sharing
problems come down to: the cognitive users appropriately
allocate transmission power to the subcarriers which they
obtained in order to optimize the utilities. Next, we first
overview the water-filling theorem, and then propose a
novel one-user water-filling algorithm to solve the power
allocation problem for the cognitive users.
For the ith cognitive user, assume its subcarrier set is
{1, 2,...,M
i
}, from (4) we can obtain the transmission
power in each subcarrier.
p
i,j
=min
I
j
− σ
2
H
j,0
,
μ − h
−1
i,j
+
(6)
where the first item in the bracket denotes the interfer-
ence constraint for primary users, (x)
+
=max(x, 0).
μ = λ
−1
B/(M · ln 2) is the water-level, λ is the Lagrange
multiplier, and h
i,j
= K
i
H
i,j
/σ
2
denotes the ith user’s
gain-to-noise ratio in the jth subcarrier.
3.2 Water-filling algorithm
Thus, the optimal transmission rate for the cognitive users
can be obtained by power water-filling (or water-pouring),
as we can see in Fig. 1. When h
−1
i,j
is smaller, the cor-
responding transmission power assigned to this subcarrier
is higher; as h
−1
i,j
increases, the transmission power is re-
duced simultaneously. Especially when h
−1
i,j
μ, because
Fig. 1 Water-filling theor em
of the poor channel state, it will assign no transmission
power in the jth subcarrier, i.e., the cognitive user does
not transmit signals in this subcarrier. Distinguish from
the traditio nal water-filling theorem , the interference con-
straint for the primary users should be taken into account
(e.g., subcarrier 2 and 4 in Fig. 1), i.e., the transmission
power should not exceed (I
j
− σ
2
)/H
j,0
in subcarrier j
when the primary user is on.
Obviously, when the utility for the ith user is optimal,
M
i
j=1
p
i,j
= P
i
, from ( 6) we can see that p
i,j
is a function of
the water-level μ, so the constraint can be expressed as
f(μ)=
M
i
j=1
p
i,j
− P
i
=
M
i
j=1
min
I
j
− σ
2
H
j,0
,
μ − h
−1
i,j
+
− P
i
(7)
where f is a monotonic increasing function of variable
μ (subject to the interference constraint, the monotonicity
can not be strict in some subcarriers). Therefore, we can
obtain μ from (7), in this way, we propose a novel one-user
water-filling algorithm as shown in Table 1.
Table 1 The one-user water-filling algorithm
Input: M
i
, H
i,j
,H
j,0
∀j ∈ M
i
,BER
i
, σ
2
and f.
(i) Initialization: obtain the gain-to-noise ratio h
i,j
= K
i
H
i,j
/σ
2
.
(ii) Sort the subcarriers such that {h
−1
i,j
} is in increasing order,
i.e., h
−1
i,j
h
−1
i,j+1
for any j ∈{1,...,M
i
− 1}.
(iii) Set μ = h
−1
i,M
i
and obtain the function value about f according
to (7). If f(h
−1
i,M
i
) > 0,setM
i
= M
i
− 1 and return to step
(iii), otherwise, set x = M
i
andgotostep(iv).
(iv) Set num = x +1, and then search the water-level
μ ∈ (h
−1
i,x
,h
−1
i,x+1
) according to f(μ)=0.Ifμ<h
−1
i,x+1
,
set num = x and obtain the final μ.
(v) Obtain the transmission power transmission power strategies
according to (6).
Output: power allocation strategies p
i,j
and the number of allocated
subcarriers num.
Next, we’ll demonstrate the effectiveness of the one-
user water-filling algorithm. Assume the number of sub-
carriers the user actually used is
˜
M
i
,since{h
−1
i,j
} is in-
creasing in j,weget
μ − h
−1
i,j
>0, 1 j
˜
M
i
0,
˜
M
i
<j M
j
⇔ h
−1
i,
˜
M
i
<μ h
−1
i,
˜
M
i
+1
(8)
where
h
−1
i,
˜
M
i
+1
→ +∞
when
˜
M
i
= M
i
.
Since f(h
−1
i,M
i
+1
) > 0 and f is a monotonic increasing
function, the object for the algorithm is to find suitable
剩余10页未读，继续阅读
2015-12-28 上传
点击了解资源详情
2021-08-10 上传
2021-03-10 上传
2021-01-19 上传
2021-08-11 上传
2021-05-24 上传
点击了解资源详情
weixin_38637805
- 粉丝: 4
- 资源: 952
我的内容管理
展开
最新资源
- BottleJS快速入门：演示JavaScript依赖注入优势
- vConsole插件使用教程：输出与复制日志文件
- Node.js v12.7.0版本发布 - 适合高性能Web服务器与网络应用
- Android中实现图片的双指和双击缩放功能
- Anum Pinki英语至乌尔都语开源词典：23000词汇会话
- 三菱电机SLIMDIP智能功率模块在变频洗衣机的应用分析
- 用JavaScript实现的剪刀石头布游戏指南
- Node.js v12.22.1版发布 - 跨平台JavaScript环境新选择
- Infix修复发布：探索新的中缀处理方式
- 罕见疾病酶替代疗法药物非临床研究指导原则报告
- Node.js v10.20.0 版本发布，性能卓越的服务器端JavaScript
- hap-java-client：Java实现的HAP客户端库解析
- Shreyas Satish的GitHub博客自动化静态站点技术解析
- vtomole个人博客网站建设与维护经验分享
- MEAN.JS全栈解决方案：打造MongoDB、Express、AngularJS和Node.js应用
- 东南大学网络空间安全学院复试代码解析
