Design of Cost-Efficient Optical Fronthaul for 5G/6G Networks: An Optimization Perspective

. 2022 Dec 1;22(23):9394. doi: 10.3390/s22239394

Algorithm 1 Cost-effective optical fronthaul design algorithm based on K-means clustering.

Input:B, P, N, $κ_{N}$ , $C_{f}$ , $C_{o}$ , $C_{p}$ , $C_{n}$ , $C_{b}$ , $C_{a}$ , $C_{O & M}$ , $C_{S r}$ , E, $E_{l}$ , $E_{o}$ , $E_{n}$ , $E_{cool}$ , $D 1_{m a x}$ , $D_{m a x}$ , maximum number of replications

Output: Optimal optical fronthaul deployment, optimal TCO

1:
Calculate the number of required splitters $κ_{P}$ by dividing $κ_{N}$ by $ζ$
2:
while the constraints of Equations (3)–(6), and Equations (13)–(15) not satisfied do
3:
Choose random power splitters locations
4:
for maximum number of replications do
5:
for all $n \in κ_{N}$ do
6:
for all $p \in P$ do
7:
$\sum_{n = 1}^{N} \sum_{p = 1}^{P} {(| N_{n} - P_{p} |)}^{2}$
8:
Based on the shortest distance, find the optimal fiber deployment between each RRH and the nearest splitter (distribution fiber)
9:
end for
10:
end for
11:
end for
12:
end while
13:
Find the optimal locations required splitters ( $κ_{P} \in P$ )
14:
Calculate the total cost ( $T C O_{1}$ )
15:
Choose random BBU pool location
16:
for all $κ_{P} \in P$ do
17:
for all $b \in B$ do
18:
19:
$\sum_{p = 1}^{P} \sum_{b = 1}^{B} {(| P_{p} - B_{b} |)}^{2}$
20:
if the constraints of Equations (7)–(10), and (13)–(16) are satisfied then
21:
Based on the shortest distance, find the optimal fiber deployment between each p and the nearest BBU pool (feeder fiber)
22:
end if
23:
end for
24:
end for
25:
Find the optimal locations of the required BBU pools ( $κ_{P} \in B$ ).
26:
According to the constraints given by Equations (11) and (12), calculate the cost of the required number of BBUs, AWGs and OLTs
27:
Calculate the optimized total cost ( $T C O 2$ )
28:
Calculate the final optimized total cost ( $T C O$ = $T C O 1$ + $T C O 2$ )