Energy Aware Dynamic Mode Decision for Cellular D2D Communications by Using Integrated Multi-Criteria Decision Making Model

1. Introduction

Device-to-device (D2D) communication is one of the promising approaches to communication between radio devices directly, for future generations. It plays an important role in upcoming cellular networks. In LTE-advanced the main challenges are to recover the local area services and enhance spectrum efficiency. In LTE-advanced, D2D communication helps to achieve those challenges (Sree Harsha & Tirupal, 2014). The 3^rd generation partnership project (3GPP) standardization is the most important standardization for 5G systems. Multiple numbers of solutions working on current cellular network model to increase the performance, such a way, this work has been defined as optimal D2D communication with enhanced mode selection. It is used in different fields like public safety, social services, network traffic offloading and proximity-based services.

Device-to-device communication provides direct communication between nearby devices without using a base station (BS) (Asadi, Wang, & Mancuso, 2014; Feng, Lu, Yuan-Wu, Li, Li & Feng, 2014). In normal cellular systems, all communications travel over/across the core network, even if the communication parties are close to each other. This type of architecture suitable for low data rate services like text messages and voice calls. In present days, cellular users use high data rate services (video streaming, online gaming, and proximity-based social services). These services could be in the range of direct communication, i.e. D2D communications. It mainly achieves three goals i.e. hop gain, reuse gain and proximity gain. Hop gain comes from the fact that D2D communications only use one hop to communicate with devices directly. Reuse gain is introduced when D2D users reuse the channels of existing cellular resources with proper interference. The proximity gain is attained due to communications occurred in close proximity. Mainly D2D communication has 2 phases. (1) peer-discovery where devices aware of their locations and also detects nearby devices (or) services. For discovering nearby devices D2D uses two approaches. They are direct discovery approach and centralized discovery approach. (2) Communication establishment where D2D devices exchange their information by using the D2D link. For communication, D2D exchange their data directly (or) by using an optimized path through the eNB. D2D communication networks consist of 2 main structures (1) stand-Alone D2D (without infrastructure) (2) network-Assisted D2D (with infrastructure).

The main difference between these structures is the existence of infrastructure. In stand-Alone D2D, systems organize the communications without infrastructure. In network-Assisted D2D, communications are organized with the help of infrastructure. D2D network architecture consists of D2D applications, network management and D2D area network. D2D area network consists of a swarm of devices, data aggregators and gateways. Swarm devices directly send the data from one device to another device with the help of D2D links. Data aggregators collect data from different nodes and send them to the accessed network by using D2D gateways. The accessed network may be a wired or wireless network. Network management consists of an access network and core network. Core network connected to service providers who manage the D2D services.