Analysis of Data Dissemination and Control in Social Internet of Vehicles
ABSTRACT
To achieve end-to-end delivery in intermittently connected mobile Internet of Vehicles (IoV) networks, epidemic routing is proposed for data dissemination at the price of excessive buffer occupancy due to its store-and-forward nature. Typically, epidemic routing should be controlled to reduce system resource usage (e.g., buffer occupancy) while simultaneously providing data delivery with differentiated level of statistical guarantee. With the aid of social connectivity a vehicles, the control of data dissemination could be benefited from the property of instant end-to end communication in Social IoV (SIoV). In particular, social links are leveraged to deliver control message for balancing the tradeoffs between buffer occupancy and data delivery reliability for supporting data dissemination in SIoV. In this paper, we proposed two representative schemes: the global timeout scheme and the anti packet dissemination scheme respectively for lossy and lossless data delivery where control messages are delivered in social-based end-to-end and local based ad-hoc fashions. For lossy data delivery, our investigation shows that with the suggested global timeout value, the per node buffer occupancy only depends on the maximum tolerable packet loss rate and pair wise meeting rate, providing principles toward mission-critical protocols. For lossless data delivery, our analytical results show that the buffer occupancy can be significantly reduced via fully antipacket dissemination, providing efficient end-to-end communication. The developed tools therefore offer new insights for epidemic routing protocol designs and performance evaluations for SIoV.
EXISTING SYSTEM :
Acting as a key enabler for the intelligent transportation, in recent years Internet of vehicles (IoV) receives lots of attention due to the functionalities of ubiquitous information exchange and content sharing a vehicles . Typically, smart vehicles are equipped with advanced technologies so that vehicle-to-vehicle (V2V) communications with nearby vehicles are established and vehicular ad-hoc networks (VANETs) are formed. By further leveraging the Vehicle-to-infrastructure (V2I) communications with Road Side Unit (RSU), smart vehicles could establish social relationships with other objects (both vehicles and RSUs) and form an overlay social network. , with heterogeneous links (i.e,, both ad hoc connectivity and social relationship), data search and dissemination could be facilitated to achieve the requirements of various vehicular applications . Epidemic routing is known to be a promising candidate toward end-to-end data delivery in an intermittently connected VANET. Since an end-to-end path between the source and the destination vehicles might not exist at any time instance in such networks, the data are delivered in a store-and forward fashion, that is, all vehicles encountering the source vehicle participate in relaying the data to other vehicles until the data are received by the destination vehicle. Although such a data delivery scheme reduces the end-to-end latency and spares the need for routing table updates, it inevitably induces tremendous buffer occupancy for each relaying vehicle. Therefore, striking the balance between buffer occupancy and delivery reliability is of utmost importance in epidemic
routing protocol design in VANET. In the SIoV setting, we consider the practical scenario where the social links between vehicles serve a function in control channels (i.e., exchanging low-overhead control signals) and the data transportation is fulfilled by local store-and-forward contacts. With the aid of social-based and local-based links in SIoV, in this paper we investigate two SIoV-enabled data dissemination control paradigms in epidemic routing and provide a tractable performance analysis of buffer occupancy and delivery reliability tradeoffs. As the data delivery dynamics of store-and-forward routing schemes much resemble the spreads of epidemics , throughout this paper we use the terminology from epidemiology to model epidemic routing. Analogously, avehicle/node is in the infected state if it receives the data and has the ability to deliver the data to surrounding nodes. A vehicle/node is in the recovered state if it is immune to the data (i.e., it refuses to receive the data). A vehicle/node is in the susceptible state if it is neither in the infected state nor in the recovered state (i.e., it will participate in data delivery after receiving the packet). This epidemic model is known as the susceptible-infected-recovered (SIR) model . Due to the spreading nature, before the data reaches the destinationnode the average number of infected nodes (i.e., the nodes who have received the data) increases monotonically with time. After the destination successfully receives the data, the relaying packets buffered at intermediate nodes become redundant and are expected to be removed. The deletion of packet for a node can be viewed as undergoing the transition from infected state to recovered state, and thus the immunity mechanisms in epidemiology can be applied to resolve excessive buffer occupancy problem in IoV networks.
PROPOSED SYSTEM :
In this paper, we proposed two representative schemes: the global timeout scheme and the anti packet dissemination scheme respectively for lossy and lossless data delivery where control messages are delivered in social-based end-to-end and local based ad-hoc fashions. For lossy data delivery, our investigation shows that with the suggested global timeout value, the per node buffer occupancy only depends on the maximum tolerable packet loss rate and pair wise meeting rate, providing principles toward mission-critical protocols. For lossless data delivery, our analytical results show that the buffer occupancy can be significantly reduced via fully antipacket dissemination, providing efficient end-to-end communication.
CONCLUSION :
To understand the performance tradeoffs between buffer occupancy and delivery reliability for epidemic routing supporting various applications in SIoV networks, we use an SIR model to characterize the state evolution equations of global timeout scheme and antipacket dissemination scheme. For lossy data delivery, we prove the scalability and ubiquity of the global timeout scheme by providing a closed-form expression for optimal global timeout value. With proper selection of the global timeout value as suggested in this paper, the per-node buffer occupancy is shown to only depend on the maximum packet loss rate and pairwise meeting rate, irrespective of the node population, which is crucial for SIoV protocol design. For lossless data delivery, we show that the buffer occupancy can be significantly reduced if every node participates relaying the antipackets to other nodes. In particular, end-to-end data transportation is guaranteed while minimizing the buffer occupancy via antipacket dissemination, which in turn optimizes simultaneous mission transmissions. Consequently, this paper provides performance evaluations and protocol design guidelines for epidemic routing, and offers new insights in buffer occupancy and data delivery reliability analysis toward SIoV networks.