Device-to-Device Wireless Communications
Device-to-Device (D2D) is one of the important proposed solutions to increase the capacity, offload the traffic, and improve the energy effciency in next generation cellular networks. D2D communication is known as a direct communication between two users without using cellular infrastructure networks. Despite of large expected bene fits in terms of capacity in D2D, the coexistence of D2D and cellular networks in the same spectrum creates new challenges in interference management and network design. To limit the interference power control schemes on cellular networks and D2D networks are typically adopted. Even though power control is introduced to limit the interference level, it does not prevent cellular and D2D users from experiencing coverage limitation when sharing the same radio resources. Therefore, the design of such networks requires the availability of suitable methods able to properly model the eff ect of interference in the presence of random terminals deployment. To this purpose in this PhD dissertation studied a new analytical model based on stochastic geometry to characterize the coverage probability on both cellular and D2D networks taking into account the impact of power control, shadowing and user’s random locations. The above mentioned work focus on static wireless networks while in dynamic mobility model user mobility poses several challenges especially at millimeter-wave (mmW). When transmitting at high frequencies, and using beam-forming techniques, the presence of obstacles and user mobility might severely impact the wireless link blockage properties thus translating into frequent handovers and channel estimate updates causing signifi cant signaling overhead. Unfortunately an analysis based on the coverage probability does not provide information about link blockages duration/rate as its evaluation usually does not include spatial/time correlation of the link state. In this context, the link lifetime represents an important performance index able to properly capture the dynamic behavior of the wireless link. For this reason, in this thesis I have also addressed the e ffect of blockages and user mobility in wireless networks on link lifetime. When users in D2D communications are machines, then I refer to machine-to-machine (M2M) communications. Examples of M2M networks are those devoted to improve the safety of read users like cars and cyclists. In the last part of the Thesis, the experimental activity carried out within the European H2020 project XCYCLE is reported. Such an activity regards the investigation of a communication and localization architecture to determine potentially dangerous situations and provide a real-time feedback to road users. Summarizing, the main contributions of this dissertation are: Chapter 1: Analytical Characterization of Device-to-Device and Cellular Networks Coexistence: In chapter 1 a new analytical framework based on stochastic geometry for the characterization of the reciprocal impact of D2D communications and an underlay cellular network in terms of coverage probability is presented. I consider a random number of D2D groups where in each group devices are distributed according to di fferent spatial distributions to model users’ behavior. The e ffect of power control, users’ spatial distribution, shadowing and random base station (BS) deployment are accounted for in the analysis and closed form expressions for coverage probability for both cellular network and D2D networks are derived. The validity of the framework developed is assessed via simulation in the numerical results where the eff ect of key system parameters as well as devices spatial distribution on cellular and D2D coverage is investigated and the amount of the traffic that could be offloaded through D2D communications is studied. Part of the materials (text, tables and illustrations) of this chapter have been published in [J1], [C1], [C2], @ IEEE. Chapter 2: Characterization of Link Lifetime in the Presence of Random Blocking Object-Part I: In chapter 2 the statistical characterization of the link lifetime is addressed by introducing a new mathematical framework to model randomly deployed obstacles distributed according to Poisson point process (PPP) and user’s mobility. I show that the link lifetime can be computed through a Markov chain model. In the numerical results the interplay of between obstacles’ density, transmission range and user’s speed is investigated for two di fferent mobility models. Part of the materials (text, tables and illustrations) of this chapter have been submitted for publication in [J2], @ IEEE. Chapter 3: Characterization of Link Lifetime in the Presence of Random Blocking Object-Part II: In chapter 3 I extend the model that I proposed in chapter 2 where the size of obstacles are taken into account. I derive an analytical framework to characterize the statistics of the link lifetime of a moving user in the presence of random obstacles with diff erent sizes for two diff erent mobility models. Using statistical geometry arguments, closed-forms of the cumulative distribution function and the average link lifetime are obtained as a function of the distance, user’s speed and direction, obstacles’ density and size. The analytical framework is validated through simulations and allows to get insights on the impact of system parameters on the link lifetime. Part of the materials (text, tables and illustrations) of this chapter have been submitted for publication in [J3], and [C3] @ IEEE. Chapter4:Device-to-Device (D2D) communication and localization for road users: In chapter 4, an ultrawide-band localization system and high-level architectures to improve the cyclists’ safety are presented. They consist of tags placed on bikes, whose positions have to be estimated, and anchors, acting as reference nodes, located at intersections and/or on vehicles. The peculiarities of the localization system in terms of accuracy and cost enable its adoption in enhanced risk assessment systems situated on the infrastructure/vehicle, depending on the architecture chosen, as well as real-time warning to road users. Part of the materials (text, tables and illustrations) of this chapter have been submitted for publication in [J4] @ IEEE.