Multichannel MAC and Cognitive MAC

It is known that inter-vehicle communication systems are proposed for safety applica- tions and trac enhancements. However, non-safety applications, or commercial applica- tions, have been proposed in the literature to provide eective use of the DSRC spectrum [54, 55, 56]. The best solution to provide non-safety realtime applications is through the multichannel communications. It has been explained in Section 2.4.2 that DSRC allows multiple channel communications by operating only one channel at a time. This is because there is only one radio used in DSRC. The dicult part of designing a multichannel MAC protocol is to provide the none safety communication while meeting the QoS of the safety communications. As discussed in Section 2.5.1, Su and Zhang propose a multichannel pro- tocol that operates in two dierent radios simultaneously. This type of communication is considered a multichannel operating protocol, but it is also considered costly since each

vehicle should be equipped with two transceivers.

Wang and Hassan propose a framework that performs periodic channel switching over DSRC to provide a concurrent safety and non-safety applications in [54]. It is found that, during rush hours of trac, the non-safety applications can be extremely restricted to assure the QoS of the safety applications. An interesting conclusion is that using simple techniques can increase the commercial non-safety applications opportunities. Moreover, the authors suggested the use of an adaptive scheme to perform dynamic adjustments to the control channel interval to support the switching between safety and commercial applications.

In [56], multichannel single radio MAC is proposed to provide V2V and V2I communi- cations. The protocol aims at providing concurrent safety and commercial services. The work is rst presented with the basic idea in [55]. However, in [56], the authors extend their work with proof of theorems of the design, and performance evaluation with the IEEE 802.11 DCF used in the ad-hoc mode and PCF used in the centralised mode. The protocol is tested in three congurations: the DCF-only, the PCF-only in the hotspot area, and dedicated coordinating AP (DCAP) protocol conguration. The DCAP conguration is based on the DCF, PCF, and spatial division functions. These congurations are simulated using ns-2 in a four-lane highway scenario with high density of a vehicle ow. The DCAP conguration is shown to oer more consistent QoS than the other two congurations.

A cognitive MAC protocol for VANETs (CMV) based on cognitive radio management is proposed in [57]. In CMV, the protocol provides long-term and short term spectrum access which is applied in vehicular communications channels. The protocol is applied on DSRC channels. The cognitive radio management is used to improve the capacity in long-term spectrum access. For short-term access, the wideband spectrum pooling is used. The protocol showed a signicant improvement in throughput compared with other multichannel protocols.

The IEEE 1609.4 draft standard for VANET [58] denes the sync interval which consti- tutes of control channel interval (CCH interval) and service channel interval (SCH interval) as shown in Fig 3.1. The IEEE 1609.4 standard denes the time division scheme for WAVE radios to alternatively switch between CCH and SCH during sync interval to support dif- ferent applications concurrently [59]. The start of a sync intervals are synchronised with the Coordinated Universal Time (UTC) second and multiples of 100 ms thereafter. The sync intervals constitute of 100ms which has 50 ms CCH interval and 50 ms SCH interval [60]. According to WAVE, Wave Basic Service Set (WBSS) is dened by set of nodes which cooperates with each other, and consists of one provider (the node that has services to oer) that is WBBS initiator and one or more WBSS users. An example scenario is

Figure 2.7: Example scenario for VANET.

shown in Fig 2.7

Several MAC protocols have been proposed based on IEEE 1609.4 standards to improve the protocol performance and reliability. The IEEE 1609.4 standard has been studied by Qi et al. [59] where the author shows that in IEEE 1609.4 standard the channel is heavily under utilised. The extended SCH interval has been proposed in [61], which increase the saturation throughput of the SCH but does not improve the reliability of the system for safety applications in terms of end to end (E2E) delay, throughput and jitter. Cognitive MAC protocol dene by Seung et al. [62] which improves system throughput but degrades the performance of safety data delivery. The novel MAC protocol for vehicular mesh network by Yunpeng, Zeng and Stibor et alia. [63] enhances the performance of non safety application. Furthermore, the Hardware Constrained Cognitive MAC has been proposed in [64], which shows signicant improvement in channel utilization and good performance for non safety users, but again the HC-MAC did not improve system performance for safety users. The above mentioned MAC protocols for VANET focus to improve the overall system performance only for IP services. However, none of them showed any improvement for safety related application, which is the most crucial application for VANET. In this paper we propose the novel mechanism that is based on IEEE 1609.4 and guarantee the reliability of the system performance for safety trac with a better channel utilization.

interval, so there are 10 sync intervals per second. This is motivated by a desire to map a sync interval to the generally assumed 10Hz vehicle safety messaging rate. It is possible that more than two providers try to access any of the six SCH. To avoid collision between WSA frames on CCH during the CCH interval random back-o scheme is being used. Moreover, when any provider listens to the successful WSA/WSAR handshake within the same CCH interval and also intends to broadcast WSA frames adjust it's SCH in order to avoid collision and if the successful WSA/WSAR handshake is conducted, the provider chooses dierent SCH during the SCH interval. Furthermore, if the provider fails to conduct WSA/WSAR handshake within the CCH interval then it has to wait until the next sync interval and then that node will make have to use larger CW size to make second WSA announcement. After the above mentioned management procedures, the vehicles will change into the SCH agreed by them, during SCH interval and start the data transmission. In addition, large amount of data required several sync intervals to complete the transmission.