Search

Resources allocation (RA) is a challenging task in many fields and applications including communications and computer networks. The conventional solutions of such problems usually come with a time and memory cost, especially for massive networks such as Internet-of-Things (IoT) networks. In this paper, two RA deep network models are proposed for enabling a clustered underlay IoT deployment, where a group of IoT nodes are uploading information to a centralized gateway in their vicinity by reusing the communication channels of conventional cellular users. The RA problem is formulated as a two-dimensional matching problem, which can be expressed as a traditional linear sum assignment problem (LSAP). The two proposed models are based on the recurrent neural network (RNN). Specifically, we investigate the performance of two long short-term memory (LSTM) based architectures. The results show that the proposed techniques could be used as replacement of the well-known Hungarian algorithm for solving LSAPs due to its ability to find the solution for the problems with different sizes, high accuracy, and very fast execution time. Additionally, the results show that the obtained accuracy outperforms the state-of-the-art deep network techniques. [ABSTRACT FROM AUTHOR]

Internet-of-Things (IoT) deployment underlaying cellular communication have been drawing increasing attention in recent years. In this work, we consider Device-to-Device (D2D) communication technology for enabling uplink transmission in multiple IoT groups/clusters at which the resource block (RB) of each cellular user is reused by multiple IoT devices (IoTDs). Joint resources and power allocation optimization problem is formulated for maximizing both the IoTDs connectivity and cell throughput under an interference constraint. A heuristic algorithm, namely greedy iterative matching (GIM), is proposed as a sub-optimal solution for the resource allocation problem while a fixed-power margin technique is used for the cellular users' power allocation. Simulations results show that the proposed GIM algorithm provides enhanced cell throughput gain and accessibility to IoTDs. The proposed reuse model is applicable in a network at which the number of IoTDs deployed within multiple groups is larger than the available cellular reuse partners. [ABSTRACT FROM AUTHOR]