Contents I. Introduction
Wireless communication is growing day by day. Till 2019, it has reached 168Exabytes per month. This creates a high load on the base station which requires higher power. Thus we need some technique to offload the base station. D2D communication allows direct communication between two nearby devices without going through the base station, so it can be used for cellular offloading [1]. Device-to-device communication consists of two types: Inband device-to-device communication and Outband device-to-device communication. When the D2D link operates in the cellular spectrum and cellular user devices (CUDs) share their spectrum with device users (DUs), it is called Inband D2D communication. When D2D links use spectrum, which is not used for cellular operation it is called Outband D2D communication. In Inband mode, there are two types-underlay and overlay. In underlay, the cellular spectrum is shared by device-to-device users. In the overlay, there is a dedicated cellular spectrum assigned to device-to-device users. In overlay device-to-device communication since there is a dedicated spectrum given to device-to-device users, it does not provide spectral efficiency[2]. So a lot of research is taking place in underlay device-to-device communication. In underlay device-to-device communication, it provides spectral efficiency and cellular offloading but it causes interference to cellular users. The issue of the pairing of device-to-device links with a cellular user in underlay mode is critical [3]. In [4] a three-step method has been proposed to produce maximum throughput of the network. After a D2D pair is allowed to share the spectrum of a cellular user, power allocation of the cellular user and corresponding device-to-device user is done. Thereafter a maximum weight matching Bipartite algorithm is applied to produce maximum throughput. In [6] a resource sharing optimization problem is discussed. It has been assumed that the base station is capable of assigning optimal resource between the cellular user and the device-to-device user. Two optimization methods have been discussed. Greedy sum-rate maximization method and sum-rate maximization with rate constraints. Two scenarios have been discussed: a single-cell scenario and a WINNER II A1 office scenario consisting of multi-cell environments. In [7] a scenario in which a D2D pair shares an uplink spectrum has been considered. Each D2D pair can share one uplink resource. A mixed-integer nonlinear programming problem (MINLP) has been formed for uplink channel reuse sharing. The Hungarian algorithm is used to find the optimal solution for this problem. To solve this Hungarian problem, a suboptimal heuristic algorithm is discussed. In [8] a Bipartite matching graph theory has been applied to solve an optimal resource allocation problem. Both downlink and uplink cases have been discussed. M number of cellular users and N number of D2D users has been taken. A D2D group shares the same uplink or downlink spectrum. Intra-cell interference has been considered. In [9] a power control that is distributed has been applied to reduce interference in underlay device-to-device communication. Stochastic geometry has been used to solve this model. Coverage probability and spectral efficiency increase in this paper when more device users operate, the quality of service of the cellular user is the same. In [10] also a distributed power control scheme is used for the cellular network-assisted D2D communication. A cellular user and a D2D pair have been considered in a multi-cell scenario. Intercell and intracell interference occur when the OFDM resource block of the cellular user is shared with the D2D pair. These interferences are reduced by reducing the power of D2D pair such that the SINR of the cellular user is maintained at a target level. In [11] power control of the D2D pair is being done based on the receiver sensitivity of the D2D pair receiver. ITU-R and METIS-2020 channel models have been used for communication. Intercell and intracell interference have been considered. It is shown that the cell sum rate increases when D2D operates in underlay mode compared to plane cellular communication. In [12] also, power control is done based on the receiver sensitivity of the D2D pair receiver. Here the METIS channel model has been used. In [13] an approach based on joint power and channel allocation problem has been discussed to maximize the total throughput of the cell. With constraints of rate and outage probability on the cellular user and constraints of latency on the D2D user. In [14] maximum sum rate of the system has been found based on joint power and channel allocation algorithm. Maintaining a minimum SINR for cellular and D2D users both. Here a new pairing scheme for device-to-device with cellular users in the uplink spectrum has been presented. The objective is to pair the device-to-device user with the cellular user such that it does not degrade the performance of the cellular users. To reduce computation complexity the work has been divided into two parts. First, a set of D2D link has been found for each of cellular user. After that, it has been investigated that how to optimize pairing selection between the device-to-device user and the cellular user from the previously established candidate set.
