An Energy Efficient Design of Computation Offloading Enabled by UAV

. 2020 Jun 13;20(12):3363. doi: 10.3390/s20123363

Algorithm 1 SCA-based alternating algorithm for

P 1

Input:

A_{k}

, K, N,

Δ

δ

q_{k}

, P, B,

σ^{2}

P_{i}

P_{0}

U_{t i p}

v_{0}

d_{0}

ρ

, s, A,

V_{m a x}

ε

and tolerant threshold

ξ

ξ_{1}

;

1:
2:
Initialize: Iterative number $i = 1$ , local point $q_{u} {[n]}^{(1)}$ , $E E_{U}^{1} = 0$ ;
3:
4:
repeat
5:
6:
Solve P2.2 by applying CVX under the given trajectory $q_{u} {[n]}^{(i)}$ and get the optimal bits allocation in each slot $I_{k}^{u *} [n]$ , $I_{k}^{c *} [n]$ and $I_{k}^{d *} [n]$ ;
7:
8:
Update the iterative number $i = i + 1$ ;
9:
10:
Let $I_{k}^{u, i} [n] = I_{k}^{u *} [n]$ , $I_{k}^{c, i} [n] = I_{k}^{c *} [n]$ and $I_{k}^{d, i} [n] = I_{k}^{d *} [n]$ ;
11:
12:
repeat
13:
14:
Initialize: Iterative number $l = 1$ , local point $q_{u} {[n]}^{l}$ and $y {[n]}^{l}$ , $E E_{U}^{' i, 1} = 0$ ;
15:
16:
Solve P3.1 by applying CVX under the given bits allocation in each slot $I_{k}^{u, i} [n]$ $I_{k}^{c, i} [n]$ $I_{k}^{d, i} [n]$ and get the optimal trajectory of the UAV $q_{u}^{*} [n]$ ;
17:
18:
Update $l = l + 1$ ;
19:
20:
Let $q_{u}^{i, l} [n] = q_{u}^{*} [n]$ ;
21:
22:
Get $E E_{U}^{' i, l}$ ;
23:
24:
until $E E_{U}^{' i, l} - E E_{U}^{' i, l - 1} \leq ξ_{1}$
25:
26:
Obtain the energy efficiency of the UAV $E E_{U}^{i}$ by (14);
27:
28:
until $E E_{U}^{i} - E E_{U}^{i - 1} \leq ξ$
29:
30:
Let $I_{k}^{u} [n] = I_{k}^{u, i} [n]$ , $I_{k}^{c} [n] = I_{k}^{c, i} [n]$ , $I_{k}^{d} [n] = I_{k}^{d, i} [n]$ , $q_{u} [n] = q_{u}^{i, l} [n]$ ;
31:
Output: $I_{k}^{u} [n]$ , $I_{k}^{c} [n]$ , $I_{k}^{d} [n]$ and $q_{u} [n]$ .
32: