notesum.ai

Published at December 6

cs.IT

math.IT

Released Date: December 6, 2024

Authors: Pramesh Gautam¹, Ravi Sharan B A G², Paolo Baracca³, Carsten Bockelmann¹, Thorsten Wild², Armin Dekorsy¹

Aff.: ¹University of Bremen; ²Nokia Bell Labs Stuttgart; ³Nokia Standards

Parameter	Value
Deployment Parameters
Number of SNs, $N$	$16$
Number of UEs, $M$	$1$
Interfering SNs, $\|\hat{\mathscr{N}}(t)\|$	$4$
Deployment Area	$20\times 20~{}\text{m}^{2}$
Minimum separation distance,	$4$ m
Mobility Model	Random Directional Model
Cell Radius, $r$	$2$ m
Velocity, $v$	$2$ m/s
Channel and PHY Parameters
Carrier frequency	$6$ GHz
Number of sub-bands, $K$	$4$
Frequency reuse	$1/4$
Pathloss	3GPP InF-DL [22]
Shadow fading std. deviation	$4$ dB(LOS) $~{}\|~{}$ $7.2$ dB(NLOS)
Decorrelation distance,	10 m
Doppler frequency, $f_{d_{n}}$	$80~{}\text{Hz}$
Transmit Power	$0$ dBW
TTI duration	$0.1$ ms
Packet size	$160$ bits
DSSM and EKF Parameters
CQI type	wideband
MCS reference table	Table $5.1.3.1-3$ [23]
Number of MCS levels, $L$	$29$
SINR min-max difference, $\Delta$	$4.8$
ESM error variance, $\sigma_{G_{1}}^{2}$	$\Delta^{2}/(12\times L)$ [24]
CQI mapping error variance, $\sigma_{G_{2}}^{2}$	$2\times{10}^{-9}$
Noise Factor,	$10$ dB
Noise Bandwidth,	$50$ MHz
Process Noise Variance, $\sigma^{2}_{F}$	$0.0042$
Baseline Parameters
Moving-avg. smoothing factor	$0.01$
LSTM sliding window	$30$
LSTM prediction-step	$1$
LSTM training epochs	$200$