README.md

Reproducing "UAV Trajectory Planning for Data Collection from Time-Constrained IoT Devices"

The first unofficial implementation of a paper with the titled "UAV Trajectory Planning for Data Collection from Time-Constrained IoT Devices". In this repository, I implemented a sub-optimal algorithm based on successive convex approximation (SCA) as introduced by Samir et al. (2019). Specifically, I reproduced the Figure 3 in the paper, where the goal is to optimize the UAV trajectory and allocation of resources to maximize the total number of served IoT devices within a flight mission duration based on a given set of target time constraints.

Paper: M. Samir, S. Sharafeddine, C. M. Assi, T. M. Nguyen and A. Ghrayeb, "UAV Trajectory Planning for Data Collection from Time-Constrained IoT Devices," in IEEE Transactions on Wireless Communications, vol. 19, no. 1, pp. 34-46, Jan. 2020, doi: 10.1109/TWC.2019.2940447. > https://ieeexplore.ieee.org/document/8842600

Summary

I summarize the paper and the implementation in this slides

How to execute the code

1. Install CVX on MATLAB : http://cvxr.com/cvx/download/
2. Change rician variable in the main.m to true or false as needed
3. Execute the matlab code, i.e., main.m

If the plot is not displayed after the iteration is completed, please execute this in the command window:

hold on; scatter(X, Y, 10, 'b');scatter(device_X, device_Y,50, 'r'); xlim([0 800]);ylim([0 800]); text(device_X, device_Y, split(num2str(deadline))); hold off

Reproduced Results

Rician Channel

Path-loss Channel