WO2022170719A1 - 基于有效区域的多目标优化设计方法 - Google Patents

基于有效区域的多目标优化设计方法 Download PDF

Info

Publication number
WO2022170719A1
WO2022170719A1 PCT/CN2021/102975 CN2021102975W WO2022170719A1 WO 2022170719 A1 WO2022170719 A1 WO 2022170719A1 CN 2021102975 W CN2021102975 W CN 2021102975W WO 2022170719 A1 WO2022170719 A1 WO 2022170719A1
Authority
WO
WIPO (PCT)
Prior art keywords
bacteria
bacterial
set1
optimal
fitness value
Prior art date
Application number
PCT/CN2021/102975
Other languages
English (en)
French (fr)
Inventor
张士兵
吴建绒
嵇雪
郭莉莉
包志华
Original Assignee
南通大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 南通大学 filed Critical 南通大学
Priority to AU2021221501A priority Critical patent/AU2021221501B2/en
Publication of WO2022170719A1 publication Critical patent/WO2022170719A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14Network analysis or design
    • H04L41/142Network analysis or design using statistical or mathematical methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/004Artificial life, i.e. computing arrangements simulating life
    • G06N3/006Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W24/00Supervisory, monitoring or testing arrangements
    • H04W24/02Arrangements for optimising operational condition
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W52/00Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/04TPC
    • H04W52/18TPC being performed according to specific parameters
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00Reducing energy consumption in communication networks
    • Y02D30/70Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • the invention relates to the field of communication system design, in particular to a multi-objective optimization design method of a cognitive communication system.
  • the present invention proposes a multi-objective optimal design method based on an effective area.
  • the design method can ensure that the Pareto optimal solution obtained by optimization is set in the effective area of the system, the convergence speed is fast, and the complex multi-objective optimization problem faced by the current communication system design is effectively solved.
  • Cognitive radio has the ability to perceive its own environment, the ability to learn about environmental changes, the ability to mine spectrum resources, and the ability to reconfigure system functions. Applying the cognitive radio technology to the emergency communication system to build a cognitive radio-based emergency communication system-cognitive emergency communication system can meet the requirements of large capacity, high density and low delay of emergency communication systems in emergency scenarios.
  • the cognitive emergency communication system Considering that most emergency communication users in the cognitive emergency communication system use rechargeable batteries as the power source, it is difficult to charge at the disaster site, so when designing the cognitive emergency communication system, the main goal of the system design is to achieve emergency communication. Maximize the user's transmission rate while maximizing the life cycle of emergency communication users. This is a typical multi-objective optimization problem for communication systems.
  • the third method has been widely used in multi-objective problems dealing with conflicting objectives, such as multi-objective bacterial foraging algorithm and multi-objective particle swarm optimization algorithm.
  • multi-objective bacterial foraging algorithm and multi-objective particle swarm optimization algorithm.
  • how to ensure that the solution of multi-objective optimization is within the effective area of the communication system to meet the actual scene requirements of the communication system, achieve the global optimum, and further accelerate the convergence speed of the optimization algorithm is a difficult problem that has not been completely solved.
  • the present invention proposes a multi-objective optimal design method based on an effective area.
  • the design method can ensure that the Pareto optimal solution obtained by optimization is set in the effective area of the system, the convergence speed is fast, and the complex multi-objective optimization problem faced by the current communication system design is effectively solved.
  • the present invention adopts the multi-objective optimization design method based on the effective area, and adopts the bacterial foraging algorithm to jointly optimize the multiple design objectives of the communication system by using the optimal design objective function of the communication system as the fitness value of the bacterial trend movement.
  • Bacteria swim to the global optimal position using adaptive step size and direction within the set effective area, and at the same time use dynamic retention ratio to continuously update the bacterial colony, and finally obtain the optimal design of the system by finding the best harmonic solution of the bacterial colony Program.
  • the method of the invention applies the effective area to the multi-objective optimization design of the communication system, and uses the set effective area to screen the bacteria, so that the bacteria always swim to the global optimal position with an adaptive step size and direction in the set effective area , quickly reach the global optimum, and solve the complex multi-objective optimization design problem faced by the communication system design.
  • This can produce the following beneficial effects:
  • step size and direction of bacterial trend movement are related to the global optimal bacterial position, which ensures that the multi-objective optimization design scheme of the system is globally optimal;
  • the direction of bacterial trend movement is related to the number of bacterial trend operations and the individual historical optimal position of bacteria, which speeds up the algorithm convergence speed;
  • the bacterial update adopts the dynamic retention ratio method, which improves the diversity of the multi-objective optimization design scheme of the system
  • FIG. 1 is a schematic diagram of the emergency communication system model of the present invention.
  • FIG. 2 is a comparison of the convergence of the solution of the present invention and other algorithms.
  • Fig. 3 is the design flow chart of the scheme of the present invention.
  • FIG. 1 is a schematic diagram of the model of the cognitive emergency communication system of the present invention.
  • the cognitive emergency communication system includes at least 3 cognitive emergency users and 2 main users.
  • the present invention implements a multi-objective optimization design method based on an effective area, including the following steps:
  • Step 1 Determine the system optimization goal
  • an appropriate cognitive emergency user transmit power P s is designed to maximize the cognitive emergency communication user's life cycle while maximizing the cognitive emergency user transmission rate.
  • f 1 is the optimal design objective function of the cognitive emergency user transmission rate
  • f 2 is the optimal design objective function of the cognitive emergency communication user life cycle
  • B is the system channel bandwidth
  • p 0 is the channel idle probability
  • p 1 is the channel busy probability
  • P s is the transmit power of the cognitive emergency user
  • h ss is the end-to-end power gain of the cognitive emergency user
  • h sp is the system cognitive
  • Q v is the maximum transmit power of the cognitive emergency user
  • P p is the primary user transmit power
  • h ps is the interference power gain of the primary user to the cognitive emergency user
  • I th is the primary user
  • the optimization target number of the system is determined according to the actual communication system requirements, and the more the target number is.
  • Step 2 Set the effective area of system optimization design
  • the search area for the optimal solution of the system optimization design is set as f 1min ⁇ f 1 ⁇ f 1max , f 2min ⁇ f 2 ⁇ f 2max , the Pareto optimal set of its distribution is set1, and the system is optimized
  • the effective area for designing the optimal solution is f 1min ⁇ f 1 ⁇ a, f 2min ⁇ f 2 ⁇ b 2
  • the Pareto optimal set of its distribution is set2
  • the screening area for the optimal solution of the system optimization design is f 1min ⁇ f 1 ⁇ a, f 2min ⁇ f 2 ⁇ b 1 , b 1 ⁇ b 2
  • the Pareto optimal set of its distribution is set3, where f 1min , f 1max , f 2min , and f 2max are the optimal design for system optimization
  • the boundary of the solution search area, f 1min , f 2min , a, b 2 are the boundaries of the effective area of the optimal solution of the system optimization design
  • the search area, effective area and screening area range are set according to the business characteristics and application scenarios of the actual communication system.
  • f 1min -28.5
  • f 1max 0
  • f 2min -1
  • f 2max 0
  • a -14
  • b 1 -0.65
  • b 2 -0.6.
  • Step 3 target optimization algorithm selection
  • the transmission power P s of the system cognitive emergency user is taken as the position value of the bacterial movement
  • the objective functions f 1 and f 2 of the system optimization design are taken as the fitness value F 1 and the fitness value F 2 of the bacterial group (bacteria collection) movement, respectively.
  • the bacterial foraging algorithm optimizes the transmit power P s of the cognitive emergency user in the system, and the optimal reconciliation solution of the bacterial motion position obtained is the optimal transmit power of the cognitive emergency user in the emergency communication system.
  • the fitness value F 1 and the fitness value F 2 are the fitness value sets of the flora corresponding to the objective functions f 1 and f 2 of the system optimization design, respectively.
  • the system target optimization algorithm may adopt the bacterial foraging algorithm, or may adopt other optimization algorithms, such as the ant colony algorithm.
  • Step 5 Calculate the bacterial fitness value
  • the fitness value of the bacteria is calculated according to the optimization objective of the system, and the fitness value of the bacteria movement is in a one-to-one correspondence with the objective of the system optimization design.
  • Step 6 Determine the global optimal bacterial position gbest
  • (i) Construct the Pareto optimal set set1 of the search area.
  • the selection of the global optimal bacterial position gbest is selected according to the area set in step 2, which directly affects the swimming step length and direction of the bacteria in the chemotaxis operation.
  • the swimming step size of bacteria is adaptive, and its size depends on the difference between the global optimal bacterial position gbest selected in step 6 and the current position, which ensures that the bacteria swim to the global optimal position.
  • Step 10 If i ⁇ 1, go back to (ii) in step 9.
  • Step 11 Update the global optimal bacterial position value gbest
  • Step 12 Update the next trend movement direction of all bacteria:
  • c 1 , c 2 are the acceleration constants (c 1 , c 2 ⁇ 0) of bacterial tropism movement
  • r 1 , r 2 are random numbers in [0,1]
  • w is the inertia weight of bacterial tendency to move
  • w max is the maximum value of bacterial tendency to move inertia weight
  • w min is the minimum value of bacterial tendency to move inertia weight.
  • c 1 0.6
  • c 2 0.4
  • w max 0.9
  • w min 0.4.
  • the direction of bacterial trend movement is not only related to the global optimal position and the individual historical optimal value, but also to the original trend movement direction of the bacteria, and its weight is related to the number of trends.
  • the bacterial replication operation adopts the method of dynamic retention ratio, which improves the diversity and uniformity of the algorithm solution.
  • Step 18 Determine the optimal transmit power.
  • the location value where the best reconciled bacteria is located is the optimal transmit power for the cognitive emergency user.
  • the multi-objective optimization design method of this embodiment is simulated and tested, and the test results are shown in Table 1 and FIG. 2 .
  • the number of trending operations adaptively updates the step size and direction of bacterial swimming, which speeds up the convergence of the algorithm and ensures the global optimality of the best harmonic solution.
  • the effectiveness of the embodiment of the present invention (the number of bacteria in the Pareto optimal set set2 and the number of bacteria in the Pareto optimal set set1) The ratio of numbers) is increased from 37.59% and 38.65% to 61.18% respectively, and the convergence of the embodiment of the present invention is reduced from more than 2000 iterations to about 500 iterations. It can be seen that, compared with the existing multi-objective optimization design method, the embodiment of the present invention significantly improves the effectiveness and convergence of the Pareto optimal set of the algorithm.
  • the invention is applicable to the system of multi-objective optimal design and also to the system of single-objective optimal design.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Algebra (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Probability & Statistics with Applications (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本发明涉及一种基于有效区域的多目标优化设计方法,采用细菌觅食算法将通信系统优化设计目标函数作为细菌趋向运动的适应值对通信系统的多个设计目标进行联合优化。细菌在设置的有效区域内使用自适应步长和方向向全局最优位置泳动,同时采用动态保留比例不断更新菌群,最后通过找出菌群的最佳调和解求得系统的最佳设计方案。为了确保系统多目标最佳优化设计方案在系统的有效区域内,本发明在细菌趋向运动的过程中引入有效区域。本发明通过自适应步长和方向使得细菌快速达到全局最优,有效解决了通信系统设计面临的复杂多目标优化设计难题。

Description

基于有效区域的多目标优化设计方法 技术领域
本发明涉及通信系统设计领域,具体地说涉及一种认知通信系统的多目标优化设计方法。为了克服上述现有技术的不足,本发明提出了一种基于有效区域的多目标优化设计方法。该设计方法能够保证优化得到的帕累托最优解集在系统的有效区域内,收敛速度快,有效解决了当前通信系统设计面临的复杂多目标优化难题。
背景技术
我国幅员辽阔,地理、气候条件复杂,是全球遭受自然灾害最严重的国家之一。及时构建稳定、可靠的应急通信系统是有效实施重大自然灾害现场救援的首要条件。基于基础设施的常规通信系统(如移动通信)往往由于其基础设施在重大灾害时遭到破坏,很容易陷入系统瘫痪或系统过载,无法满足应急场景下呈指数级增长的突发通信业务需求;而一些专用通信系统(如卫星通信)在紧急通信期间业务容易过载、非紧急通信期间频谱资源空闲等问题。
认知无线电具有对自身环境的感知能力、对环境变化的学习能力、对频谱资源的挖掘能力和系统功能的可重构能力。将认知无线电技术应用到应急通信系统中,构建基于认知无线电的应急通信系统-认知应急通信系统,可以满足紧急场景下应急通信系统的大容量、高密度、低时延的要求。
考虑到在认知应急通信系统中大多数应急通信用户是以可充电的电池作为电源,在灾害现场充电比较困难,因此在设计认知应急通信系统时,系统设计的主要目标是在实现应急通信用户传输速率最大化的同时最大化应急通信用户的生命周期。这是一个典型的通信系统多目标优化问题。
现有解决多目标优化问题方法主要有三种:(1)将最重要的因素视为优化目标,其他的视为约束;(2)通过为每个目标分配不同的权重,将多目标问题转换为单个目标问题;(3)找到一组由同时满足所有目标的满意解组成的帕累托最优集。如果能确定优化目标间有明确的优先级关系,一般采用第一种方法。第二种方法对优化线性组合目标非常有用,但是在实际应用场景中,由于环境和限制条件的模糊性,难以确定精确的加权比值,从而导致优化结果不理想。与前两种方法不同,第三种方法在处理目标冲突的多目标问题中得到了广泛应用,例如多目标细菌觅食算法和多目标粒子群算法等。但是,如何保证多目标优化的解在通信系统有效区域内以满足通信系统实际场景需求,达到全局最优,进一步加快优化算法的收敛速度,是一个尚未彻底解决的难题。
发明内容
为了克服上述现有技术的不足,本发明提出了一种基于有效区域的多目标优化设计方法。该设计方法能够保证优化得到的帕累托最优解集在系统的有效区域内,收敛速度快,有效解决了当前通信系统设计面临的复杂多目标优化难题。
为了达到上述目的,本发明基于有效区域的多目标优化设计方法,采用细菌觅食算法将通信系统优化设计目标函数作为细菌趋向运动的适应值对通信系统的多个设计目标进行联合优化。细菌在设置的有效区域内使用自适应步长和方向向全局最优位置泳动,同时采用动态保留比例不断更新菌群,最后通过找出菌群的最佳调和解求得系统的最佳设计方案。
本发明方法将有效区域应用到通信系统的多目标优化设计中,利用设置的有效区域对细菌进行筛选,使得细菌始终在设置的有效区域内以自适应步长和方向向全局最优位置泳动,快速达到全局最优,解决了通信系统设计面临的复杂多目标优化设计难题。由此可产生如下的有益效果:
(1)通过设置细菌趋向运动有效区域,确保了系统的多目标最佳优化设计方案在系统的有效区域内;
(2)细菌趋向运动的步长和方向与全局最优细菌位置有关,确保了系统的多目标优化设计方案是全局最优的;
(3)细菌趋向运动的步长和方向是自适应的,确保了细菌始终快速向全局最优位置泳动;
(4)细菌趋向运动的方向与细菌的趋向操作次数有关、与细菌的个体历史最优位置有关,加快了算法收敛速度;
(5)细菌更新采用动态保留比例方法,,提高了系统的多目标优化设计方案的多样性;
(6)有效区域与自适应步长、动态保留比例相结合,避免了算法搜索的盲目性,缩短了算法搜索的时间,提高算法的搜索精度。
附图说明
下面结合附图对本发明作进一步的说明。
图1是本发明应急通信系统模型示意图。
图2是本发明方案与其他算法的收敛性比较。
图3是本发明方案的设计流程图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述。
如图1所示为本发明认知应急通信系统模型示意图。认知应急通信系统中至少包括3个认知应急用户和2个主用户,本发明实施基于有效区域的多目标优化设计方法,包括如下步骤:
步骤1、确定系统优化目标
根据认知应急通信系统的业务特性,设计合适的认知应急用户发射功率P s,在最大化认知应急用户传输速率的同时最大化认知应急通信用户的生命周期。本系统的优化设计目标有两个,最大化认知应急用户传输速率和最大化认知应急通信用户的生命周期,即
Figure PCTCN2021102975-appb-000001
Figure PCTCN2021102975-appb-000002
或者:
Figure PCTCN2021102975-appb-000003
Figure PCTCN2021102975-appb-000004
式中,f 1为认知应急用户传输速率的优化设计目标函数,f 2为认知应急通信用户生命周期的优化设计目标函数,B为系统信道带宽,
Figure PCTCN2021102975-appb-000005
为系统信道高斯噪声功率,p 0为信道空闲概率,p 1为信道忙碌概率,P s为认知应急用户发射功率,h ss为认知应急用户端到端功率增益,h sp为系统认知应急用户对主用户的干扰功率增益,Q v为认知应急用户的最大发射功率,P p为主用户发射功率,h ps为主用户对认知应急用户的干扰功率增益,I th为主用户接收机的干扰门限。在 本实施例中,B=2MHz,
Figure PCTCN2021102975-appb-000006
p 0=0.4,p 1=0.6,h ss=0.6,h sp=0.5,Q v=1W,P p=1W,h ps=0.5,I th=0.4W。
本步骤中,系统的优化目标数是根据实际通信系统需求而确定的,目标数越多。
步骤2、设置系统优化设计有效区域
根据应急通信系统业务特性,设置系统优化设计最优解的搜索区域为f 1min≤f 1≤f 1max,f 2min≤f 2≤f 2max,其分布的帕累托最优集为set1,系统优化设计最优解的有效区域为f 1min≤f 1≤a,f 2min≤f 2≤b 2,其分布的帕累托最优集为set2,系统优化设计最优解的筛选区域为f 1min≤f 1≤a,f 2min≤f 2≤b 1,b 1≤b 2,其分布的帕累托最优集为set3,其中f 1min、f 1max、f 2min、f 2max为系统优化设计最优解搜索区域的边界,f 1min、f 2min、a、b 2为系统优化设计最优解的有效区域的边界,f 1min、f 2min、a、b 1为系统优化设计最优解的筛选区域的边界。搜索区域、有效区域和筛选区域范围根据实际通信系统的业务特性和应用场景而设置的。在本实施例中,中f 1min=-28.5,f 1max=0,f 2min=-1,f 2max=0,a=-14,b 1=-0.65,b 2=-0.6。
步骤3、目标优化算法选取
将系统认知应急用户的发射功率P s作为细菌运动的位置值,系统优化设计目标函数f 1和f 2分别作为菌群(细菌集合)运动的适应值F 1和适应值F 2,则选取细菌觅食算法对系统认知应急用户的发射功率P s进行优化,所得的细菌运动位置的最佳调和解就是应急通信系统中认知应急用户的最佳发射功率。其中,适应值F 1和适应值F 2分别是菌群对应于系统优化设计目标函数f 1和f 2的适应值集合。本步骤中,系统目标优化算法可以采用细菌觅食算法,也可以采用其他的优化算法,例如蚁群算法。
步骤4、初始化
假设有I个细菌参与觅食优化算法,构成细菌集合set。对集合set中的每个细菌进行编号。在可行区域[0,Q v]范围内随机生成每个细菌的初始位置
Figure PCTCN2021102975-appb-000007
在-1和1之间随机生成每个细菌的初始运动方向Δ(i),i=1,2,3,…,I;设细菌泳动的最大步数为N s,细菌的最大趋向次数为N c,细菌的最大复制次数为N re,细菌的最大迁移次数为N ed,迁移概率为P ed。在本实施例中,I=100,即有100个细菌参觅食优化算法,N s=5,N c=1000,N re=4,N ed=2,P ed=0.25。
步骤5、计算细菌适应值
根据细菌位置计算所有细菌的适应值。细菌i的适应值
Figure PCTCN2021102975-appb-000008
Figure PCTCN2021102975-appb-000009
的计算公式如下:
Figure PCTCN2021102975-appb-000010
本步骤中,细菌的适应值是根据系统的优化目标来计算的,细菌运动的适应值与系统优化设计的目标是一一对应的。
步骤6、确定全局最优细菌位置gbest
(i)构建搜索区帕累托最优集set1。在细菌集合set中两两比较细菌的适应值。如果细菌i的两个适应值
Figure PCTCN2021102975-appb-000011
Figure PCTCN2021102975-appb-000012
都不大于且不同时等于细菌m的两个适应值
Figure PCTCN2021102975-appb-000013
Figure PCTCN2021102975-appb-000014
i=1,2,...,I,m=1,2,...,I且m≠i,那么细菌m被细菌i支配,将细菌m丢弃。在剩余的细菌中重复这样的细菌适应值比较,直到被其它细菌支配的细菌都被丢弃。最后由这些剩余的非支配细菌构成搜索区域中分布的帕累托最优集set1。
(ii)对集合set1中的所有细菌进行编号,编号为n,n=1,2,...,N,N为集合set1中细菌的个数,N的大小根据集合set1中的实际细菌个数动态变化。
(iii)根据细菌适应值F 1的大小,对集合set1中细菌按升序进行排序,构成新的集合set1 1,并按升序对set1 1中的所有细菌进行重新编号,编号为u,u=1,2,...,N,同时保留细菌在集合set1中的编号n。
(iv)根据细菌适应值F 1计算集合set1 1中细菌u,u=2,3,...,N-1,与它前后两个细菌的间隔
Figure PCTCN2021102975-appb-000015
其中,
Figure PCTCN2021102975-appb-000016
表示集合set1 1中细菌u与它前后两个细菌的间隔,同时细菌u在集合set1中的编号为n,f 10max、f 10min分别为细菌适应值F 1的最大值和最小值。
(v)根据细菌适应值F 2的大小,对集合set1中细菌按升序进行排序,构成新的集合set1 2,并按升序对set1 2中的所有细菌进行重新编号,编号为v,v=1,2,...,N,同时保留细菌在集合set1中的编号n。
(vi)根据细菌适应值F 2计算集合set1 2中细菌v,v=2,3,...,N-1,与它前后两个细菌的间隔
Figure PCTCN2021102975-appb-000017
其中,
Figure PCTCN2021102975-appb-000018
表示集合set1 2中细菌v与它前后两个细菌的间隔,同时细菌v在集合set1中的编号为n,f 20max、f 20min分别为细菌适应值F 2的最大值和最小值。
(vii)令
Figure PCTCN2021102975-appb-000019
Figure PCTCN2021102975-appb-000020
为无穷大。根据
Figure PCTCN2021102975-appb-000021
Figure PCTCN2021102975-appb-000022
计算最优集set1中细菌n的拥挤距离
Figure PCTCN2021102975-appb-000023
(viii)根据细菌的拥挤距离大小,对集合set1中细菌按降序进行排序。
(ix)根据细菌i的适应值
Figure PCTCN2021102975-appb-000024
Figure PCTCN2021102975-appb-000025
对最优集set1中的细菌进行筛选,保留筛选区域(f 1min≤f 1≤a,f 2min≤f 2≤b 1)内的细菌,构成筛选区域中分布的帕累托最优集set3。
(x)选择最优集set3中拥挤距离前10%的个体组成集合gbestpool,从集合gbestpool中 随机选择一个细菌的位置作为全局最优细菌位置gbest。
本步骤中,全局最优细菌位置gbest的选取是根据步骤2中设置的区域进行筛选选取的,它直接影响到趋化操作中细菌的泳动步长和方向。
步骤7、细菌迁移
令l=1,在集合set中对所有细菌执行迁移操作。
步骤8、细菌复制
令k=1,在集合set中对所有细菌执行复制操作。
步骤9、细菌趋向
令j=1,集合set中所有细菌执行趋向操作。
(i)令i=0,细菌i泳动。
(ii)令i=i+1,sn=0。记录细菌i在执行趋向操作前的位置信息
Figure PCTCN2021102975-appb-000026
及其适应值:
Figure PCTCN2021102975-appb-000027
计算细菌i的泳动步长:
Figure PCTCN2021102975-appb-000028
(iii)细菌泳动。细菌i泳动一次,更新其位置:
Figure PCTCN2021102975-appb-000029
(iv)边界控制。检查细菌i是否在可行区域内
如果
Figure PCTCN2021102975-appb-000030
则细菌i从区域边界的下界作反方向泳动:
Figure PCTCN2021102975-appb-000031
如果
Figure PCTCN2021102975-appb-000032
则细菌i从区域边界的上界作反方向泳动:
Figure PCTCN2021102975-appb-000033
(v)更新细菌i的适应值f 1 i
Figure PCTCN2021102975-appb-000034
Figure PCTCN2021102975-appb-000035
(vi)如果
Figure PCTCN2021102975-appb-000036
(t=1,2)(即f 1 i
Figure PCTCN2021102975-appb-000037
都不大于且不同时等于
Figure PCTCN2021102975-appb-000038
Figure PCTCN2021102975-appb-000039
),则
Figure PCTCN2021102975-appb-000040
sn=sn+1;否则,sn=N s
(vii)如果sn<N s,返回步骤9中的(iii)。
(viii)选取细菌i的个体历史最优值pbest i
比较细菌i在趋向操作前后的适应值。如果细菌i在趋向操作后的两个适应值f 1 i
Figure PCTCN2021102975-appb-000041
都不大于且不同时等于细菌i在趋向操作前的适应值
Figure PCTCN2021102975-appb-000042
Figure PCTCN2021102975-appb-000043
那么此时细菌i的位置就是个体历史最优值pbest i,即
Figure PCTCN2021102975-appb-000044
如果细菌i在趋向操作前的两个适应值
Figure PCTCN2021102975-appb-000045
Figure PCTCN2021102975-appb-000046
都不大于且不同时等于细菌i在趋向操作后的适应值f 1 i
Figure PCTCN2021102975-appb-000047
那么细菌i在趋向操作前的位置就是个体历史最优值pbest i,即pbest i=P old;否则,细菌i在
Figure PCTCN2021102975-appb-000048
和P old之间等概随机选择其中的一个作为其个体历史最优值pbest i
本步骤中,细菌的泳动步长是自适应的,其大小取决于步骤6中选取的全局最优细菌位置gbest与当前位置差值,保证了细菌向全局最优位置泳动。
步骤10、如果i<I,返回步骤9中的(ii)。
步骤11、更新全局最优细菌位置值gbest
(i)更新搜索区中帕累托最优集set1
在细菌集合set中两两比较细菌的适应值。如果细菌i的两个适应值f 1 i
Figure PCTCN2021102975-appb-000049
都不大于且不同时等于细菌m的两个适应值f 1 m
Figure PCTCN2021102975-appb-000050
i=1,2,...,I,m=1,2,...,I且m≠i,那么细菌m被细菌i支配,将细菌m丢弃。在剩余的细菌中重复这样的细菌适应值比较,直到被其它细菌支配的细菌都被丢弃。最后由这些剩余的非支配细菌构成搜索区域中分布的帕累托最优集set1。
(ii)对集合set1中的所有细菌进行编号,编号为n,n=1,2,...,N,N为集合set1中细菌的个数。
(iii)根据细菌适应值F 1的大小,对集合set1中细菌按升序进行排序,构成新的集合set1 1,并按升序对set1 1中的所有细菌进行重新编号,编号为u,u=1,2,...,N,同时保留细菌在集合set1中的编号n。
(iv)根据细菌适应值F 1计算集合set1 1中细菌u,u=2,3,...,N-1,与它前后两个细菌的间隔
Figure PCTCN2021102975-appb-000051
(v)根据细菌适应值F 2的大小,对集合set1中细菌按升序进行排序,构成新的集合set1 2,并按升序对set1 2中的所有细菌进行重新编号,编号为v,v=1,2,...,N,同时保留细菌在集合set1中的编号n。
(vi)根据细菌适应值F 2计算集合set1 2中细菌v,v=2,3,...,N-1,与它前后两个细菌的间隔
Figure PCTCN2021102975-appb-000052
(vii)令
Figure PCTCN2021102975-appb-000053
Figure PCTCN2021102975-appb-000054
为无穷大。根据
Figure PCTCN2021102975-appb-000055
Figure PCTCN2021102975-appb-000056
计算最优集set1 0中细菌n的拥挤距离
Figure PCTCN2021102975-appb-000057
(viii)根据细菌的拥挤距离大小,对集合set1中细菌按降序进行排序;
(ix)根据细菌i的适应值f 1 i
Figure PCTCN2021102975-appb-000058
对最优集set1中的细菌进行筛选,保留筛选区域(f 1min≤f 1≤a,f 2min≤f 2≤b 1)内的细菌,更新筛选区域中分布的帕累托最优集set3;
(x)选择最优集set3中拥挤距离前10%的个体组成集合gbestpool,从集合gbestpool中随机选择一个细菌的位置作为全局最优细菌位置gbest;
步骤12、更新所有细菌的下一次趋向运动方向:
Figure PCTCN2021102975-appb-000059
Figure PCTCN2021102975-appb-000060
其中i=1,2,...,I,c 1、c 2为细菌趋向运动的加速常数(c 1,c 2≥0),r 1、r 2为[0,1]内的随机数,w为细菌趋向运动的惯性权重,w max为细菌趋向运动惯性权重的最大值,w min为细菌趋向运动惯性权重的最小值。本实施例中,c 1=0.6,c 2=0.4,w max=0.9,w min=0.4。
本步骤中,细菌趋向运动的方向不仅与全局最优位置、个体历史最优值有关,还与细菌原有的趋向运动方向有关,并且其权重与趋向次数有关。
步骤13、如果j<N c,令j=j+1,返回步骤9中的(i),对所有细菌继续执行下一轮趋向操作;否则结束本轮趋向操作,进入复制操作。
步骤14、复制操作
(i)根据细菌的适应值计算细菌集set中所有细菌的健康值J 1health、J 2health和J health,细菌i的健康值分别为:
Figure PCTCN2021102975-appb-000061
其中i=1,2,...,I。
(ii)更新细菌集set
将细菌集合set中的细菌按其健康值J 1health进行升序排列,选择其中的前50%-t个体组成pop 1;将细菌集合set中的细菌按其健康值J 2health升序排列,选择其中的前t个体组成;将细菌集合set中的细菌按其健康值J health升序排列,选择其中的前50%个体组成pop 3。更新后的细菌集set由集合pop 1、pop 2和pop 3合并构成,其中
Figure PCTCN2021102975-appb-000062
λ为共同调节系数以保证在第一次复制时,即k=1时,t的值为0.25。本实施例中,λ=0.321。
本步骤中,细菌的复制操作采用了动态保留比例的方法,改进了算法解的多样性和均匀性。
步骤15、如果k<N re,令k=k+1,返回到步骤9,对所有细菌执行下一轮趋向操作;否则结束本轮复制操作,进入迁移操作。
步骤16、迁移操作
(i)细菌集set中每个细菌均以概率P ed离开集合set而消失,同时在可行区域[0,Q v]范围内随机生成与消失细菌数量相同的细菌加入细菌集set,保持细菌总数稳定。如果l<N ed,令l=l+1,返回到步骤8,对所有细菌执行下一轮复制操作和趋向操作;否则结束迁移操作。
步骤17、选取最佳调和解
(i)根据细菌适应值对帕累托最优集set1中的细菌进行筛选,保留适应值在有效区域(f 1min≤f 1≤a且f 2min≤f 2≤b 2)内的细菌,得到有效区域中分布的帕累托最优集set2,并对集合set2中的所有细菌重新编号,编号为m,m=1,2,...,M,M为集合set2中的细菌个数,M的大小根据集合set2中的实际细菌个数动态变化。
(ii)计算帕累托最优集set2中第m个细菌的隶属度
Figure PCTCN2021102975-appb-000063
其中
Figure PCTCN2021102975-appb-000064
Figure PCTCN2021102975-appb-000065
(iii)在帕累托最优集set2中选取隶属度最大的细菌作为细菌觅食算法的最佳调和解。
步骤18、确定最佳发射功率。
最佳调和解细菌所在的位置值就是认知应急用户的最佳发射功率。
对本实施例的多目标优化设计方法进行仿真测试,测试结果见表1和图2。
表1 算法的有效性
平均值 本发明方案 多目标细菌觅食算法 多目标粒子群算法
set1 245.35 247.25 175.8
set2 148.7 92.9 67.9
比值 0.6118 0.3759 0.3865
结果表明采用本案的多目标优化设计方法,根据设置的有效区域筛选参与优化算法的细菌,提高了算法帕累托最优集的有效性,根据全局最优细菌位置、细菌个体历史最优值以及趋向操作次数自适应更新细菌泳动步长和方向,加快了算法的收敛性,保证了最佳调和解的全局最优性。与目前常用的多目标细菌觅食算法和多目标粒子群算法相比较,本发明实施例的有效性(帕累托最优集set2中细菌的个数与帕累托最优集set1中细菌的个数之比)分别从37.59%和38.65%提高到61.18%,本发明实施例的收敛性由2000次以上迭代下降为500次迭代左右。可见,本发明实施例与已有的多目标优化设计方法相比显著提高了算法帕累托最优集的有效性和收敛性。本发明适用于多目标优化设计的系统,也适用于单目标优化设计的系统。
除上述实施例外,本发明还可以有其他实施方式。凡采用等同替换或等效变换形成的技术方案,均落在本发明要求的保护范围。

Claims (7)

  1. 基于有效区域的多目标优化设计方法,所述通信系统中至少包括1个认知应急用户和1个主用户,认知应急用户有数据业务要传输,其特征在于所述多目标优化设计方法包括如下步骤:
    步骤1、确定系统优化目标
    根据下式设计认知应急用户发射功率P s
    Figure PCTCN2021102975-appb-100001
    Figure PCTCN2021102975-appb-100002
    式中,f 1为认知应急用户传输速率的优化设计目标函数,f 2为认知应急通信用户生命周期的优化设计目标函数,B为系统信道带宽,
    Figure PCTCN2021102975-appb-100003
    为系统信道高斯噪声功率,p 0为信道空闲概率,p 1为信道忙碌概率,P s为认知应急用户发射功率,h ss为认知应急用户端到端功率增益,h sp为系统认知应急用户对主用户的干扰功率增益,Q v为认知应急用户的最大发射功率,P p为主用户发射功率,h ps为主用户对认知应急用户的干扰功率增益,I th为主用户接收机的干扰门限;
    步骤2、设置系统优化设计有效区域
    设置系统优化设计最优解的搜索区域为f 1min≤f 1≤f 1max,f 2min≤f 2≤f 2max,其分布的帕累托最优集为set1,系统优化设计最优解的有效区域为f 1min≤f 1≤a,f 2min≤f 2≤b 2,其分布的帕累托最优集为set2,系统优化设计最优解的筛选区域为f 1min≤f 1≤a,f 2min≤f 2≤b 1,b 1≤b 2,其分布的帕累托最优集为set3,其中f 1min、f 1max、f 2min、f 2max为系统优化设计最优解搜索区域的边界,f 1min、f 2min、a、b 2为系统优化设计最优解的有效区域的边界,f 1min、f 2min、a、b 1为系统优化设计最优解的筛选区域的边界;
    步骤3、目标优化算法选取
    将系统认知应急用户的发射功率P s作为细菌运动的位置值,系统优化设计目标函数f 1和f 2分别作为菌群运动的适应值F 1和适应值F 2,选取细菌觅食算法对系统认知应急用户的发射功率P s进行优化,所得的细菌运动位置的最佳调和解就是应急通信系统中认知应急用户的最佳发射功率;其中,适应值F 1和适应值F 2分别是菌群对应于系统优化设计目标函数f 1和f 2的适应值集合;
    步骤4、初始化
    假设有I个细菌参与觅食优化算法,构成细菌集合set,对集合set中的每个细菌进行编号,在可行区域[0,Q v]范围内随机生成每个细菌的初始位置
    Figure PCTCN2021102975-appb-100004
    在-1和1之间随机生成每个细菌的初始运动方向Δ(i),i=1,2,3,…,I;设细菌泳动的最大步数为N s,细菌的最大趋向 次数为N c,细菌的最大复制次数为N re,细菌的最大迁移次数为N ed,迁移概率为P ed
    步骤5、计算细菌适应值
    根据细菌位置计算所有细菌的适应值,细菌i的适应值
    Figure PCTCN2021102975-appb-100005
    Figure PCTCN2021102975-appb-100006
    的计算公式如下:
    Figure PCTCN2021102975-appb-100007
    步骤6、确定全局最优细菌位置gbest
    (i)构建搜索区帕累托最优集set1,在细菌集合set中两两比较细菌的适应值,如果细菌i的两个适应值
    Figure PCTCN2021102975-appb-100008
    Figure PCTCN2021102975-appb-100009
    都不大于且不同时等于细菌m的两个适应值
    Figure PCTCN2021102975-appb-100010
    Figure PCTCN2021102975-appb-100011
    i=1,2,...,I,m=1,2,...,I且m≠i,那么细菌m被细菌i支配,将细菌m丢弃,在剩余的细菌中重复这样的细菌适应值比较,直到被其它细菌支配的细菌都被丢弃,最后由这些剩余的非支配细菌构成搜索区域中分布的帕累托最优集set1;
    (ii)对集合set1中的所有细菌进行编号,编号为n,n=1,2,...,N,N为集合set1中细菌的个数;
    (iii)根据细菌适应值F 1的大小,对集合set1中细菌按升序进行排序,构成新的集合set1 1,并按升序对set1 1中的所有细菌进行重新编号,编号为u,u=1,2,...,N,同时保留细菌在集合set1中的编号n;
    (iv)根据细菌适应值F 1计算集合set1 1中细菌u,u=2,3,...,N-1,与它前后两个细菌的间隔
    Figure PCTCN2021102975-appb-100012
    其中,
    Figure PCTCN2021102975-appb-100013
    表示集合set1 1中细菌u与它前后两个细菌的间隔,同时细菌u在集合set1中的编号为n,f 10max、f 10min分别为细菌适应值F 1的最大值和最小值;
    (v)根据细菌适应值F 2的大小,对集合set1中细菌按升序进行排序,构成新的集合set1 2,并按升序对set1 2中的所有细菌进行重新编号,编号为v,v=1,2,...,N,同时保留细菌在集合set1中的编号n;
    (vi)根据细菌适应值F 2计算集合set1 2中细菌v,v=2,3,...,N-1,与它前后两个细菌的间隔
    Figure PCTCN2021102975-appb-100014
    其中,
    Figure PCTCN2021102975-appb-100015
    表示集合set1 2中细菌v与它前后两个细菌的间隔,同时细菌v在集合set1中的编号为n,f 20max、f 20min分别为细菌适应值F 2的最大值和最小值;
    (vii)令
    Figure PCTCN2021102975-appb-100016
    Figure PCTCN2021102975-appb-100017
    为无穷大,根据
    Figure PCTCN2021102975-appb-100018
    Figure PCTCN2021102975-appb-100019
    计算最优集set1 中细菌n的拥挤距离
    Figure PCTCN2021102975-appb-100020
    (viii)根据细菌的拥挤距离大小,对集合set1中细菌按降序进行排序;
    (ix)根据细菌i的适应值
    Figure PCTCN2021102975-appb-100021
    Figure PCTCN2021102975-appb-100022
    对最优集set1中的细菌进行筛选,保留筛选区域(f 1min≤f 1≤a,f 2min≤f 2≤b 1)内的细菌,构成筛选区域中分布的帕累托最优集set3;
    (x)选择最优集set3中拥挤距离前10%的个体组成集合gbestpool,从集合gbestpool中随机选择一个细菌的位置作为全局最优细菌位置gbest;
    步骤7、细菌迁移
    令l=1,在集合set中对所有细菌执行迁移操作;
    步骤8、细菌复制
    令k=1,在集合set中对所有细菌执行复制操作;
    步骤9、细菌趋向
    令j=1,集合set中所有细菌执行趋向操作;
    (i)令i=0,细菌i泳动;
    (ii)令i=i+1,sn=0,记录细菌i在执行趋向操作前的位置信息
    Figure PCTCN2021102975-appb-100023
    及其适应值:
    Figure PCTCN2021102975-appb-100024
    计算细菌i的泳动步长:
    Figure PCTCN2021102975-appb-100025
    (iii)细菌泳动,细菌i泳动一次,更新其位置:
    Figure PCTCN2021102975-appb-100026
    (iv)边界控制,检查细菌i是否在可行区域内
    如果
    Figure PCTCN2021102975-appb-100027
    则细菌i从区域边界的下界作反方向泳动:
    Figure PCTCN2021102975-appb-100028
    如果
    Figure PCTCN2021102975-appb-100029
    则细菌i从区域边界的上界作反方向泳动:
    Figure PCTCN2021102975-appb-100030
    (v)更新细菌i的适应值f 1 i
    Figure PCTCN2021102975-appb-100031
    Figure PCTCN2021102975-appb-100032
    (vi)如果
    Figure PCTCN2021102975-appb-100033
    即f 1 i
    Figure PCTCN2021102975-appb-100034
    都不大于且不同时等于
    Figure PCTCN2021102975-appb-100035
    Figure PCTCN2021102975-appb-100036
    Figure PCTCN2021102975-appb-100037
    sn=sn+1;否则,sn=N s
    (vii)如果sn<N s,返回步骤9中的(iii);
    (viii)选取细菌i的个体历史最优值pbest i
    比较细菌i在趋向操作前后的适应值,如果细菌i在趋向操作后的两个适应值f 1 i
    Figure PCTCN2021102975-appb-100038
    都不大于且不同时等于细菌i在趋向操作前的适应值
    Figure PCTCN2021102975-appb-100039
    Figure PCTCN2021102975-appb-100040
    那么此时细菌i的位置就是个体历史最优值pbest i,即
    Figure PCTCN2021102975-appb-100041
    如果细菌i在趋向操作前的两个适应值
    Figure PCTCN2021102975-appb-100042
    Figure PCTCN2021102975-appb-100043
    都不大于且不同时等于细菌i在趋向操作后的适应值f 1 i
    Figure PCTCN2021102975-appb-100044
    那么细菌i在趋向操作前的位置就是个体历史最优值pbest i,即pbest i=P old;否则,细菌i在
    Figure PCTCN2021102975-appb-100045
    和P old之间等概随机选择其中的一个作为其个体历史最优值pbest i
    步骤10、如果i<I,返回步骤9中的(ii);
    步骤11、更新全局最优细菌位置值gbest
    (i)更新搜索区中帕累托最优集set1
    在细菌集合set中两两比较细菌的适应值,如果细菌i的两个适应值f 1 i
    Figure PCTCN2021102975-appb-100046
    都不大于且不同时等于细菌m的两个适应值f 1 m
    Figure PCTCN2021102975-appb-100047
    i=1,2,...,I,m=1,2,...,I且m≠i,那么细菌m被细菌i支配,将细菌m丢弃,在剩余的细菌中重复这样的细菌适应值比较,直到被其它细菌支配的细菌都被丢弃,最后由这些剩余的非支配细菌构成搜索区域中分布的帕累托最优集set1;
    (ii)对集合set1中的所有细菌进行编号,编号为n,n=1,2,...,N,N为集合set1中细菌的个数;
    (iii)根据细菌适应值F 1的大小,对集合set1中细菌按升序进行排序,构成新的集合set1 1,并按升序对set1 1中的所有细菌进行重新编号,编号为u,u=1,2,...,N,同时保留细菌在集合set1中的编号n;
    (iv)根据细菌适应值F 1计算集合set1 1中细菌u,u=2,3,...,N-1,与它前后两个细菌的间隔
    Figure PCTCN2021102975-appb-100048
    (v)根据细菌适应值F 2的大小,对集合set1中细菌按升序进行排序,构成新的集合set1 2, 并按升序对set1 2中的所有细菌进行重新编号,编号为v,v=1,2,...,N,同时保留细菌在集合set1中的编号n;
    (vi)根据细菌适应值F 2计算集合set1 2中细菌v,v=2,3,...,N-1,与它前后两个细菌的间隔
    Figure PCTCN2021102975-appb-100049
    (vii)令
    Figure PCTCN2021102975-appb-100050
    Figure PCTCN2021102975-appb-100051
    为无穷大,根据
    Figure PCTCN2021102975-appb-100052
    Figure PCTCN2021102975-appb-100053
    计算最优集set1 0中细菌n的拥挤距离
    Figure PCTCN2021102975-appb-100054
    (viii)根据细菌的拥挤距离大小,对集合set1中细菌按降序进行排序;
    (ix)根据细菌i的适应值f 1 i
    Figure PCTCN2021102975-appb-100055
    对最优集set1中的细菌进行筛选,保留筛选区域(f 1min≤f 1≤a,f 2min≤f 2≤b 1)内的细菌,更新筛选区域中分布的帕累托最优集set3;
    (x)选择最优集set3中拥挤距离前10%的个体组成集合gbestpool,从集合gbestpool中随机选择一个细菌的位置作为全局最优细菌位置gbest;
    步骤12、更新所有细菌的下一次趋向运动方向:
    Figure PCTCN2021102975-appb-100056
    Figure PCTCN2021102975-appb-100057
    其中i=1,2,...,I,c 1、c 2为细菌趋向运动的加速常数(c 1,c 2≥0),r 1、r 2为[0,1]内的随机数,w为细菌趋向运动的惯性权重,w max为细菌趋向运动惯性权重的最大值,w min为细菌趋向运动惯性权重的最小值;
    步骤13、如果j<N c,令j=j+1,返回步骤9中的(i),对所有细菌继续执行下一轮趋向操作;否则结束本轮趋向操作,进入复制操作;
    步骤14、复制操作
    (i)根据细菌的适应值计算细菌集set中所有细菌的健康值J 1health、J 2health和J health,细菌i的健康值分别为:
    Figure PCTCN2021102975-appb-100058
    其中i=1,2,...,I。
    (ii)更新细菌集set
    将细菌集合set中的细菌按其健康值J 1health进行升序排列,选择其中的前50%-t个体组成pop 1;将细菌集合set中的细菌按其健康值J 2health升序排列,选择其中的前t个体组成;将细菌集合set中的细菌按其健康值J health升序排列,选择其中的前50%个体组成pop 3。更新后的细菌集set由集合pop 1、pop 2和pop 3合并构成,其中
    Figure PCTCN2021102975-appb-100059
    λ为共同调节系数以保证在第一次复制时,即k=1时,t的值为0.25;
    步骤15、如果k<N re,令k=k+1,返回到步骤9,对所有细菌执行下一轮趋向操作;否则结束本轮复制操作,进入迁移操作;
    步骤16、迁移操作
    (i)细菌集set中每个细菌均以概率P ed离开集合set而消失,同时在可行区域[0,Q v]范围内随机生成与消失细菌数量相同的细菌加入细菌集set,保持细菌总数稳定,如果l<N ed,令l=l+1,返回到步骤8,对所有细菌执行下一轮复制操作和趋向操作;否则结束迁移操作。
    步骤17、选取最佳调和解
    (i)根据细菌适应值对帕累托最优集set1中的细菌进行筛选,保留适应值在有效区域(f 1min≤f 1≤a且f 2min≤f 2≤b 2)内的细菌,得到有效区域中分布的帕累托最优集set2,并对集合set2中的所有细菌重新编号,编号为m,m=1,2,...,M,M为集合set2中的细菌个数;
    (ii)计算帕累托最优集set2中第m个细菌的隶属度
    Figure PCTCN2021102975-appb-100060
    其中
    Figure PCTCN2021102975-appb-100061
    Figure PCTCN2021102975-appb-100062
    (iii)在帕累托最优集set2中选取隶属度最大的细菌作为细菌觅食算法的最佳调和解;
    步骤18、确定最佳发射功率
    最佳调和解细菌所在的位置值就是认知应急用户的最佳发射功率。
  2. 根据权利要求1所述基于有效区域的多目标优化设计方法,其特征在于:步骤2中,系统优化设计的搜索区域、有效区域和筛选区域是根据实际通信系统的业务特性和应用场景而设置的。
  3. 根据权利要求1所述基于有效区域的多目标优化设计方法,其特征在于:步骤3中,系统目标优化算法采用细菌觅食算法或蚁群算法。
  4. 根据权利要求1所述基于有效区域的多目标优化设计方法,其特征在于:步骤5中,细菌的适应值根据系统优化设计目标函数计算获得,细菌运动的适应值与系统优化设计的目标是一一对应的。
  5. 根据权利要求1所述基于有效区域的多目标优化设计方法,其特征在于:步骤6中,全局最优细菌位置gbest根据步骤2中设置的区域进行筛选选取。
  6. 根据权利要求1所述基于有效区域的多目标优化设计方法,其特征在于:步骤9中,细菌的泳动步长是自适应的,其大小取决于步骤6中选取的全局最优细菌位置gbest与当前位置差值。
  7. 根据权利要求1所述基于有效区域的多目标优化设计方法,其特征在于:步骤14中,细菌的复制操作采用了动态保留比例的方法。
PCT/CN2021/102975 2021-02-09 2021-06-29 基于有效区域的多目标优化设计方法 WO2022170719A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU2021221501A AU2021221501B2 (en) 2021-02-09 2021-06-29 Multi-objective optimization design method based on effective area

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110177324.2A CN112822058B (zh) 2021-02-09 2021-02-09 基于有效区域的多目标优化设计方法
CN202110177324.2 2021-02-09

Publications (1)

Publication Number Publication Date
WO2022170719A1 true WO2022170719A1 (zh) 2022-08-18

Family

ID=75864617

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/102975 WO2022170719A1 (zh) 2021-02-09 2021-06-29 基于有效区域的多目标优化设计方法

Country Status (2)

Country Link
CN (1) CN112822058B (zh)
WO (1) WO2022170719A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112822058B (zh) * 2021-02-09 2021-11-16 南通大学 基于有效区域的多目标优化设计方法

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070230597A1 (en) * 2006-04-03 2007-10-04 Kabushiki Kaisha Toshiba Dynamic optimisation of block transmissions for interference avoidance
US20100081449A1 (en) * 2008-09-30 2010-04-01 Motorola, Inc. Method and apparatus for optimizing spectrum utilization by a cognitive radio network
US20140087749A1 (en) * 2012-09-21 2014-03-27 Yuan Ze University Three layer cascade adaptive neural fuzzy inference system (anfis) based intelligent controller scheme and device
CN106954255A (zh) * 2017-03-28 2017-07-14 吉林化工学院 基于自适应细菌觅食的认知无线电功率控制方法及系统
CN110376897A (zh) * 2019-08-02 2019-10-25 西安建筑科技大学 一种基于ga-bfo的家庭能量多目标优化方法
CN111327376A (zh) * 2020-03-04 2020-06-23 南通大学 基于认知无线电的应急通信网络的频谱接入方法
CN111787545A (zh) * 2020-07-14 2020-10-16 南通大学 基于能量收集的全双工认知中继功率分配方法
CN112822058A (zh) * 2021-02-09 2021-05-18 南通大学 基于有效区域的多目标优化设计方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7873223B2 (en) * 2002-10-15 2011-01-18 Definiens Ag Cognition integrator and language
CN103987051B (zh) * 2014-04-28 2017-05-31 南京邮电大学 基于无线传感器网络可靠感知的频谱共享方法
CN104467999B (zh) * 2014-11-18 2017-02-22 北京邮电大学 一种基于量子蛙跳的频谱感知算法
CN104796901B (zh) * 2015-04-21 2018-04-10 上海交通大学 分布式频谱资源动态分配的实现方法
CN105420760B (zh) * 2015-11-06 2018-05-01 重庆科技学院 基于自适应步长细菌觅食算法的铝电解多参数优化方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070230597A1 (en) * 2006-04-03 2007-10-04 Kabushiki Kaisha Toshiba Dynamic optimisation of block transmissions for interference avoidance
US20100081449A1 (en) * 2008-09-30 2010-04-01 Motorola, Inc. Method and apparatus for optimizing spectrum utilization by a cognitive radio network
US20140087749A1 (en) * 2012-09-21 2014-03-27 Yuan Ze University Three layer cascade adaptive neural fuzzy inference system (anfis) based intelligent controller scheme and device
CN106954255A (zh) * 2017-03-28 2017-07-14 吉林化工学院 基于自适应细菌觅食的认知无线电功率控制方法及系统
CN110376897A (zh) * 2019-08-02 2019-10-25 西安建筑科技大学 一种基于ga-bfo的家庭能量多目标优化方法
CN111327376A (zh) * 2020-03-04 2020-06-23 南通大学 基于认知无线电的应急通信网络的频谱接入方法
CN111787545A (zh) * 2020-07-14 2020-10-16 南通大学 基于能量收集的全双工认知中继功率分配方法
CN112822058A (zh) * 2021-02-09 2021-05-18 南通大学 基于有效区域的多目标优化设计方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
JI XUE; CHEN JIAJUN; ZHANG SHIBING; ZHANG XIAOGE: "Mission Policy and Optimization Objectives in Cognitive Emergency Communication Networks", 2019 6TH INTERNATIONAL CONFERENCE ON SYSTEMS AND INFORMATICS (ICSAI), IEEE, 2 November 2019 (2019-11-02), pages 984 - 988, XP033724073, DOI: 10.1109/ICSAI48974.2019.9010074 *

Also Published As

Publication number Publication date
CN112822058B (zh) 2021-11-16
CN112822058A (zh) 2021-05-18

Similar Documents

Publication Publication Date Title
Karen The politics of class, race, and gender: Access to higher education in the United States, 1960-1986
WO2022170719A1 (zh) 基于有效区域的多目标优化设计方法
CN108123828B (zh) 一种基于接入用户移动性预测的超密集网络资源分配方法
Ji et al. Particle swarm optimization for mobile ad hoc networks clustering
Yangyang et al. Particle swarm optimization for base station placement in mobile communication
CN109872033A (zh) 一种机场机位分配方法
CN107949007A (zh) 无线缓存系统中基于拍卖理论的一种资源分配算法
CN115471016A (zh) 一种基于cisso与daed的台风预测方法
CN106535279A (zh) 基于cp神经网络的车载自组网交叉路口预测路由方法
Afuzagani et al. Optimizing BTS placement using hybrid evolutionary firefly algorithm
CN112887943B (zh) 一种基于中心度的缓存资源分配方法及系统
CN117709628A (zh) 一种基于大数据的公共服务设施用地规划方法
Liu et al. Robust power control for clustering-based vehicle-to-vehicle communication
CN108366410B (zh) 一种面向lte小站密集组网的同步方法
CN115545106A (zh) 一种多无人机中的AoI敏感的数据收集方法及系统
CN111698690B (zh) 基于改进小生境遗传的认知无线网络频谱分配方法
CN109120438B (zh) 一种机会网络下的数据协同传输方法及系统
Pinagapany et al. Solving channel allocation problem in cellular radio networks using genetic algorithm
CN108901074A (zh) 一种基于布谷鸟搜索算法的移动用户频谱分配方法
Tong et al. Cooperative spectrum sensing based on a modified shuffled frog leaping algorithm in 5G network
CN106358302B (zh) 一种无线传感器网络多收发器多信道分配算法
CN108834158A (zh) 一种用于超密集组网的干扰管理方法
CN111314157A (zh) 一种水质监测机器人集群通信网络节点高效部署方法
CN110381540B (zh) 基于dnn的实时响应时变文件流行度的动态缓存更新方法
CN110621042B (zh) 一种宽带数据流量分配和时限控制方法

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2021221501

Country of ref document: AU

Date of ref document: 20210629

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21925372

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21925372

Country of ref document: EP

Kind code of ref document: A1