A Random Access Protocol for Crowded Massive MIMO Systems Based on a Bayesian Classifier

Bueno, Felipe Augusto Dutra; Yamamura, Cézar; Goedtel, Alessandro; Marinello, José Carlos

doi:10.1109/lwc.2022.3207069

Cited by 2 publications

(12 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we report the results of our proposed NN-based RA method applied in an overcrowded M-MIMO scenario, similar to that of [ 16 , 17 , 20 , 21 ]. Our outcomes are presented in terms of confusion matrices, ANAAs, FFAAs, latencies, and throughputs.…”

Section: Results and Discussionmentioning

confidence: 99%

“…Retransmission of the pilot signal occurs when the hypothesis test

is true [ 16 ], and the pilot signal is not retransmitted when

is true:

where

is a bias parameter. It is worth noting that ( 9 ) and ( 10 ) mean that the UE only retransmits when its own signal strength is larger than half of the contending UEs’ signal strengths, which is a sufficient but not necessary condition for being the strongest contender [ 17 ]. In addition, the bias parameter can be calibrated to adjust the system behavior; for instance, to maximize the average number of resolved collisions, to minimize the occurrence of false positives (or negatives), or to ensure that at most one UE will transmit the pilot in the third step of the SUCRe protocol when

for all k [ 16 ].…”

Section: Materials and Methodsmentioning

confidence: 99%

“…In addition, the bias parameter can be calibrated to adjust the system behavior; for instance, to maximize the average number of resolved collisions, to minimize the occurrence of false positives (or negatives), or to ensure that at most one UE will transmit the pilot in the third step of the SUCRe protocol when

for all k [ 16 ]. A suitable value of

is proposed as

in [ 16 ], where the factor

multiplies the standard deviations of

centered around its mean value

and

is the average UL interference, assumed to be known at the UE [ 16 , 17 ]. It is given by

, where the expectation is computed with respect to user locations and shadow-fading realizations.…”

Section: Materials and Methodsmentioning

confidence: 99%

“…The motivation for the strongest-user retransmission criterion is that it will always have only one strongest user in a pilot collision, and it is possible to evaluate a decentralized test to let the user know if it is the strongest contender without additional signaling overhead. Despite being able to address up to 90% of all collisions, the SUCRe protocol still suffers from a significant number of false-negative cases, as highlighted in [ 17 ]. This is due to its inability to resolve pilot collisions where the strongest UE’s signal strength is lower than 50% of the sum of the signal strengths of contending UEs.…”

Section: Introductionmentioning

confidence: 99%

“…Finally, the authors of [ 17 ] propose a GB RA protocol that employs statistical techniques to address collisions in a decentralized manner at the UE level. This protocol uses a Bayesian classifier (BC) to identify the strongest user and replaces the retransmission rule of the SUCRe protocol.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Neural Network-Based Random Access Protocol for Crowded Massive MIMO Systems

Bueno,

Yamamura,

Scalassara

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

Fifth-generation (5G) and beyond networks are expected to serve large numbers of user equipments (UEs). Grant-based random access (RA) protocols are efficient when serving human users, typically with large data volumes to transmit. The strongest user collision resolution (SUCRe) is the first protocol that effectively uses the many antennas at the 5G base station (BS) to improve connectivity performance. In this paper, our proposal involves substituting the retransmission rule of the SUCRe protocol with a neural network (NN) to enhance the identification of the strongest user and resolve collisions in a decentralized manner on the UEs’ side. The proposed NN-based procedure is trained offline, admitting different congestion levels of the system, aiming to obtain a single setup able to operate with different numbers of UEs. The numerical results indicate that our method attains substantial connectivity performance improvements compared to other protocols without requiring additional complexity or overhead. In addition, the proposed approach is robust regarding variations in the number of BS antennas and transmission power while improving energy efficiency by requiring fewer attempts on the RA stage.

show abstract