This paper presents a detailed study of uplink scheduling in narrowband internet of things (NB-IoT) networks. As NB-IoT devices need a long battery lifetime, we aim to maximize energy efficiency while satisfying the main requirements for NB-IoT devices. Also, as the NB-IoT scheduling problem is divided into link adaptation problem and resource allocation problem, this paper investigates the correlation between these two problems. Accordingly, we propose two scheduling schemes: the joint scheduling scheme, where the two problems are combined as one optimization problem, and the successive scheduling scheme that manages each problem separately but successively. Each scheme aims to maximize energy efficiency while achieving reliable transmission, satisfying delay requirements, and guaranteeing resource allocation specifications. Also, we investigate the impact of the selected devices to be served on the total energy efficiency. Accordingly, we propose two device selection techniques to maximize the total energy efficiency. The first technique exhaustively searches for the optimal devices, while the second sorts the devices based on a proposed priority score. The simulation results compare the successive and the joint scheduling schemes. The results show that the joint scheme outperforms the successive scheme in terms of energy efficiency and the number of served devices but with higher complexity. Also, the results highlight the impact of each proposed selection technique on the scheduling schemes’ performance.
This paper provides a detailed investigation on the performance of Narrowband IoT link adaptation under different approaches: Max-Rate approach, Min-Energy approach and the Max-Energy-Efficiency approach while considering realistic channel estimation. Simulation results study the impact of the received power on the performance of the link adaptation in terms of the achieved consumed energy, data rate, energy efficiency, and spectral efficiency. Particularly, we show that maximizing data rate in NB-IoT leads to maximizing the spectral efficiency except at extremely low received powers. Also, results focus on the impact of the data size on the performance of the link adaptation in NB-IoT. We show that larger data sizes lead to higher energy consumption, lower data rate and lower energy efficiency.
This paper provides a detailed study on NB-IoT link adaptation, for realistic channel estimation, under different objectives. NB-IoT supports link adaptation parameters of multiple types of resource units (RU) and number of repetitions to extend coverage while achieving reliable transmissions. The paper focuses on the impact of these parameters on NB-IoT performance. We provide a mathematical formulation of link adaptation following three different approaches: maximizing the rate, minimizing the energy, and maximizing the energy efficiency. Simulation results shed the light on the effect of each link adaptation parameter (modulation and coding scheme (MCS), number of repetitions and the type of RU) on the NB-IoT transmission in the different approaches and under different coverage extensions.
This paper studies the uplink resource allocation problem in NB-IoT networks. We aim to maximize the network’s energy efficiency to serve devices in need for long battery lifetime. Also, the network energy efficiency is affected by the associated UEs to candidates, carrying the scheduling information for UEs. For this reason, we investigate the joint resource allocation and Candidate-UE association problem in the uplink (UL) in NB-IoT. The resource allocation problem allocates the UEs with the suitable scheduling delay value and subcarrier indication field. The Candidate-UE association problem selects the appropriate UE for each candidate. The optimization variables involved in the joint problem are divided into binary association variables and binary resource allocation variables. This makes it hard to solve the joint problem since its optimization variables are of combinatorial nature. To this end, we formulate the joint problem as a four-sided matching game that ranks the UEs, candidates, the scheduling delay values and subcarrier indication fields while considering maximizing the energy efficiency as a preference metric. We then describe a four-sided matching game algorithm to solve the problem. The simulation results highlight the proposed algorithm’s effectiveness compared to other algorithms in the obtained network energy efficiency.
In this paper, we consider a relay-aided device-to-device (RA-D2D) communication link operating within a cellular network. The underlay mode where the D2D users share the uplink frequency resources of the cellular link is considered. An optimal algorithm for allocating the appropriate D2D transmit powers that maximize the ergodic capacity of the D2D link while ensuring the cellular communication requirements is proposed. The performance of the proposed scheme is then analyzed in terms of the ergodic capacity of the RA-D2D link. The improvement in the spectral efficiency is compared with the half-duplex (HD) and the full-duplex (FD) modes of D2D communications. Results show that the relay-aided system outperforms the HD system and even the FD system above a certain threshold of self-interference.