Random-Access NOMA in URLL Energy-Harvesting IoT Networks With Short Packet and Diversity Transmissions

Performance Analysis of URLL Energy-Harvesting Cognitive-Radio IoT Networks With Short Packet and Diversity Transmissions

“…Adopting a machine learning approach and fountain codes for adaptive channel assignment to reduce transmission latency, and ensure reliability [22] Spectrum sharing in underlay CR networks with ARQ Optimization of SU average transmit power under outage constraint for achievable rate maximization the blocklength as well as the number of packet replicas and required RBs are optimized for energy-efficiency maximization. It is worth-mentioning that this study is different from our URLLC-based previous works in [19], [20]. Specifically, this work considers a different network model that brings dynamic spectrum access (i.e.…”

“…Other works have considered resource allocation, network slicing, link adaptation, and scheduling for enhanced mobile broadband (eMBB) with URLLC requirements [16]- [18]. Our previous works [19], [20] analyzed the performance of random-access NOMA (RA-NOMA) with clustered IoT devices in URLL-EH-IoT networks, where analytical expressions for network metricssuch as average packet latency, reliability, and GoodPut-are derived. More importantly, the RA-NOMA scenario has been compared to its RA-OMA counterpart to illustrate the merits of NOMA over OMA.…”

“…• The IoT UEs energy-efficiency maximization problem subject to energy-causality and URLL constraints on the average packet latency and reliability is solved by utilizing the derived analytical expressions. Specifically, Cross-layer optimization framework for resource allocation and packet dropping policies to minimize transmit power [11] Coordinated multi-point transmission in 5G networks Optimization of blocklength and SINR to maximize resource availability [12] OFDMA-based 5G networks with HARQ Study the impact of system bandwidth, SINR, and HARQ [13] IoT in a dense smart factory Exploiting a graph-theoretical approach to improve spectral-efficiency [14] 5G centralized multi-cell radio access networks with HARQ Adopting centralized multi-cell scheduling algorithms for latency improvement [15] Point-to-point communication with IR and HARQ Adopting dynamic programming to optimize blocklength, power per round, and number of retransmissions for average energy minimization [19] Clustered uplink RA-NOMA IoT network with EH nodes Performance analysis and optimization of transmit power, number of replicas, and number of transmitted data bits per blocklength for GoodPut maximization [20] Clustered uplink RA-NOMA IoT networks Performance analysis and exploration of the effect of number of replicas, and number of data bits per blocklength on network metrics [21] Spectrum sharing in licensed and unlicensed mmWave bands…”

“…Specifically, this work considers a different network model that brings dynamic spectrum access (i.e. cognitive radio) with energyharvesting nodes into the IoT paradigm, while [19], [20] analyze network metrics in uplink RA-NOMA IoT networks with URLL requirements. The IoT UEs in this study are assumed to access the spectrum dynamically through spectrum sharing with primary users.…”

“…This makes the network analysis fundamentally different, since the random behavior of the primary users and their data traffic pattern greatly influence the transmission process of the IoT UEs. Furthermore, to lower the packet queueing delay, this work assumes that each IoT UE is able to select a number of RBs to simultaneously transmit multiple packets in one transmission time interval (TTI), as opposed to the randomly selected RB by clustered IoT UEs in [19], [20]. Additionally, the multi-user nature of the problem in hand, and the IoT UEs' interaction are prominent differences that affect the different IoT network metrics.…”