DECREASING THE RA COLLISION IMPACT FOR MASSIVE NB-IOT IN 5G WIRELESS NETWORKS

Abstract

To satisfy the gigantic need for Internet of things (IoT) applications, the third-generation partnership project (3GPP) has revealed the narrowband IoT (NB-IoT) standard. In any case, collisions in the radio access channel of NB-IoT can be extreme due to the numerous random-access (RA) activates by a massive number of NB-IoTs and the limited available radio resources. The RA procedure is one of the MAC-layer’s functions that initiates a contention-based setup to grant an uplink transmission. In this paper, the performance of a new RA procedure is investigated by introducing a modified backoff scheme to reduce the collision probability. The key mechanism of the proposed scheme is to perform an autonomous approach for determining the time for an NB-IoT to transmit in a collide environment. The proposed scheme can improve the overall throughput of the network and the NB-IoTs battery lifetime while prioritizing some QoS parameters such as favoring the NB-IoTs with heavier traffic loads. The probability of collision analysis is subjected to many operating parameters, including the backoff countdown probability, number of NB-IoTs, queue size and the contention window size. The system and link-level simulations are conducted to assess the proposed scheme with up to five thousand NB-IoTs per cell. The simulation results showed that the proposed scheme outperforms the conventional approach.

Keywords

• backoff
• collision
• mac
• random access
• window size