在无线传感器网络中，将无人飞行器（UAV）用作传感器节点（SN）的移动数据收集器是延长网络寿命的节能技术。考虑到SN-UAV链路的一般衰落信道模型，共同优化了SN的唤醒时间表和UAV的轨迹，以最大程度地降低所有SN的最大能耗，同时确保收集到所需的数据量每个SN可靠地发送。我们将设计公式化为混合整数非凸优化问题。通过应用连续凸优化技术，提出了一种有效的迭代算法来寻找次优解。数值结果表明，与基准方案相比，该方案实现了显着的网络节能。
clear all
close all
addpath('./src')
global u q0 qF
istrain = 0
if istrain == 1
%% Optimization
Q = {};
X = {};
Theta = {};
Lo = {};
for T = [40 50 100]
disp(['T =' num2str(T) ', start'])
[q, x, th, lo]= solveP1('T', T,'Sk',10*(10^6),'epsilon',10^-2);
disp(['T =' num2str(T) ', end'])
Q = cat(1, Q, {q});
X = cat(1, X, {x});
Theta = cat(1,Theta,{th});
Lo = cat(1,Lo, {lo});
end
else
load('./data/Fig1.mat')
parameter_setting('Sk',10*(10^6),'epsilon',10^-2);
end
%% figure 1.(a)
figure(11)
hold on
plot(Q{1}(1,:),Q{1}(2,:), '--b',Q{2}(1,:),Q{2}(2,:), '-.r',Q{3}(1,:),Q{3}(2,:), '-k');
scatter(u(1,:),u(2,:), 'xk');
legend('T=40s','T=50s','T=100s');
axis([ -800 800 -800 800])
grid on
title('Fig. 1.(a), UAV trajectory')
xlabel('x')
ylabel('y')
text(q0(1),q0(2),'\leftarrow Initial point');
text(qF(1)-300,qF(2),'Final point \rightarrow');
for k = 1: 4
text(u(1,k),u(2,k),['u_{', num2str(k), '}']);
end
drawnow
%% figure 1.(b)
figure(13)
Xr = X{2};
t = zeros(4,1);
for k = 1: 4
t(k) = graythresh(Xr(:,k));
end
t = min(t) * 1.2;
Xr(Xr >= t) = 1;
Xr(Xr < t) = 0;
plot([0:99]*50/99, Xr(:,1), '-r',[0:99]*50/99, Xr(:,2), '--b', [0:99]*50/99, Xr(:,3), ':k',[0:99]*50/99, Xr(:,4), '-.m');
legend('SN u_{1}','SN u_{2}','SN u_{3}','SN u_{4}','Location','north','NumColumns',4);
axis([ 0 50 0 1.25])
title('Fig.1(b), Wake-up schedule (T=50s)');
yticks([0 1])
yticklabels({'Sleep','Wake-up'})
drawnow
%% save datas
mkdir ./data
save('./data/Fig1.mat', 'X','Q','Theta');
完整代码或者代写添加QQ1575304183