Self-optimization of coverage and capacity based on a fuzzy neural network with cooperative reinforcement learning

EURASIP Journal on Wireless Communications and Networking

Table 1 Simulation parameters

Parameter	Value
Carrier frequency	2.0 GHz
Channel bandwidth	10 MHz
Propagation loss	−128.1−37.6 log10(d), d in km
Shadowing standard deviation	8 dB
Shadowing correlation distance	50 m
Shadowing correlation	Intersite, 0.5; intersector, 1.0
Penetration loss	20 dB
Thermal noise density	−174 dBm/Hz
BS maximum transmit power	46 dBm
Maximum antenna front-to-back attenuation A_m	25 dB
Maximum vertical slant angle attenuation SLA_v	20 dB
Horizontal half-power beam width φ_{3d B}	70°
Vertical half-power beam width θ_{3d B}	10°
BS and UE antenna height	32 m; 1.5 m
Power allocation on channels	Equal allocation
Self-optimization time	100 ms
interval(learning time interval)
KPI weight factor ω	0.8
JKPI weight factor α	0.5
Initial greed action factor ε	0.1
Reducing rate of greed action	0.001/s
factor
Discount factor γ	0.7
Learning rate for Q-learning ξ	0.5
Parameter learning factor η	0.2