La Gazette de l' N° 22 Mensuel - Avril 2009 Bonjour à tous et à toutes !!! Voici le traditionnel mot du CEI qui accompagne chaque numéro de votre gazette préférée. Tout d'abord bravo à tous nos mi-thésards pour leur exposé de mi-thèse. En espérant que la dernière moitié de la thèse soit fructueuse. Ça y est, c'est le printemps ! Le soleil n'est pas encore vraiment au rendez-vous, mais ça ne saurait tarder. En avril
Sooner or later, every man in Australia runs into problems with impotency viagra australia like other bodily functions, must be in order.
Neighbor stability-based vanet clustering for urban vehicular environments
J Supercomput (2016) 72:161–176DOI 10.1007/s11227-015-1517-6 Neighbor stability-based VANET clustering for urban
Jung-Hyok Kwon1 · Hyun Soo Chang2 ·
Taeshik Shon2 · Jai-Jin Jung3 · Eui-Jik Kim1
Published online: 11 September 2015 Springer Science+Business Media New York 2015 Abstract In this paper, we propose a neighbor stability-based VANET clustering
(NSVC) that can efficiently deliver data in urban vehicular environments. The salient
features of urban vehicles are their high mobility and unpredictable direction of
movement, so vehicle-to-vehicle and vehicle-to-infrastructure (V2X) communication
should take into consideration the frequent changes in the topology of vehicular ad
hoc networks (VANETs). These technical challenges are addressed with NSVC by
including a neighbor stability-based VANET clustering scheme and the corresponding
supplementary transmission scheduling method. Thereby, NSVC supports fast cluster
formation, minimizes the number of cluster head elections, and moreover guarantees
the reliable delivery of data for emergency messages. The results of the simulation
indicate that NSVC achieves better network performance when compared to existing
Clustering · IEEE 802.11p · Neighbor stability · Urban environment · Information and communications technology (ICT) has advanced rapidly, and indus-try and academia have identified the Internet of things (IoT) as an emerging paradigm Department of Convergence Software, Hallym University, 1 Hallymdaehak-gil, Chuncheon-si,Gangwon-do 24252, South Korea Department of Computer Engineering, Ajou University, Suwon, South Korea Department of Multimedia Engineering, Dankook University, Yongin, South Korea J.-H. Kwon et al.
for future ICT convergence ,A number of communication standards that havebeen developed to date can be used to interconnect a plurality of IoT devices, and these include wireless body area networks (WBANs), wireless personal area networks(WPANs), wireless local area networks (WLANs), and wide area networks (WANs).
Nevertheless, these technologies need to be customized or merged according to theservice requirements of the given vertical application Particularly for automotiveservices, intelligent transportation systems (ITS), connected cars (V2X), or telematicsare being developed to support wide range of vertical applications including safety,traffic management, driver assistance, travel route optimization, etc. Thus, a numberof variations or configurations are required for existing communication standards.
Two communication standards, IEEE 1609.x and IEEE 802.11p, have been gener-ally considered for use in wireless access in vehicular environments (WAVE). IEEE1609.x covers resource management, security, networking service, and muti-channeloperation while IEEE 802.11p provides WLAN physical (PHY) and medium accesscontrol (MAC) layers that can support the baseline communications in a vehicularenvironment , This paper thus focuses on the network formation for data delivery between moving vehicles in urban traffic environments. In urban traffic environments, vehicles on theroad have different moving velocities and directions relative to each other, and so theirnetwork topology changes frequently. Thus, a vehicular communication system shouldbe designed to maintain reliable communication links. In ], vehicles frequentlyexchange small messages to their neighbors or with adjacent infrastructure to retainthe quality of wireless links. Vinel et al. ] proposed a long-term evolution (LTE)-based vehicular communication scheme that provides a reliable communication linkand in addition, enlarges the service coverage. However, this scheme is not suitable foruse in a real environment due to the high cost of LTE access. Jarupan et al. ] pro-posed infrastructure-based vehicular communication where vehicles can communicateusing roadside infrastructure without additional costs. This scheme highly dependenton fixed infrastructure, and thus involves high overhead due to the necessity to re-connect to the infrastructure. Cluster-based network formation approaches have beenmainly discussed in the literature to address the cost and infrastructure-dependency invehicular environments. Lin et al. proposed a lowest ID-based clustering scheme,but it suffers from the clustering overhead due to the frequent re-election of the clus-ter head (CH) because it does not consider the mobility of vehicles. Basu et al. ]proposed a more efficient clustering scheme that uses the relative mobility betweenthe neighboring vehicles. This scheme uses a clustering period with a fixed size thatmight degrade the network performance.
In this paper, we propose a neighbor stability-based VANET clustering (NSVC) for an urban vehicular environment. This seeks to address the aforementioned problemsby implementing two main mechanisms: (1) neighbor stability-based clustering tominimize the number of changes in topology and the delay in cluster formation, and(2) classified transmission scheduling to support the fast transmission of emergencymessages. The former uses the rate of change for the number of neighbors, whichis named as the neighbor stability (NS), to make a decision of whether or not toexecute the new CH election procedure. The latter uses polling-based channel access to guarantee the successful transmission of high-priority data.
Neighbor stability-based VANET clustering for urban vehicular.
We then evaluate the performance of NSVC by comparing it against those of existing schemes where the lowest ID and relative mobility are considered as criteria for CH election under a fixed clustering period. Simulations are then conducted in varioustraffic scenarios (e.g., varying the maximum velocity, numbers of vehicles, packet size). The result indicates that NSVC can achieve an improvement in performancein terms of CH lifetime, rate of the CH changes, overall network throughput, andend-to-end latency.
In the following section, we illustrate the design and the performance of our work in further details.
2 System model
The salient features of VANET, including the high mobility and unpredictable directionof movement, cause frequent changes in the topology and network disconnections. Tosolve these problems, a clustered network structure has been generally considered foruse in a VANET because it provides more reliable connectivity for a group of vehiclesFigure shows the general system architecture for vehicular services andcommunications using VANET where cluster members (CMs) communicate withtheir designated CH, and the CH can be connected to the service provider, trafficsurveillance center, or even general users via the Internet. The CH plays a role as anetwork coordinator. It specifically carries out the initiation, termination, routing, andscheduling for vehicle-to-vehicle (V2V) communications. Moreover, it communicateswith the infrastructure on behalf of the CMs in vehicle-to-infrastructure (V2I) usagescenarios.
In this paper, we assume an urban vehicular environment where a number of vehi- cles move with a different velocity and direction and where the structure of the road Fig. 1 System architecture for general vehicular service scenarios
J.-H. Kwon et al.
Fig. 2 Example of cluster-based vehicular system model in an urban area
is also complicated. As a result, the network topology changes and disconnectionsoccur frequently between the vehicles. This phenomenon is very critical for V2Xcommunication services.
Figure shows the cluster-based vehicular system model for an urban environment that is considered in this work. It consists of a set of infrastructure devices, which arenamed roadside units (RSUs), and a set of vehicles, named onboard units (OBUs), thatmove in different directions on a two-lane road in an urban traffic scenario. We assumethat the vehicles move in exactly the opposite directions on the two-lane road withdifferent velocities. When the vehicles reach an intersection, they wait for a certaintime to avoid collisions with other vehicles. For communications, we consider thatvehicles (i.e., OBUs) communicate with one-hop neighbor vehicles or infrastructure(i.e., RSUs) through a single channel, and they are equipped with multiple onboardequipment to support various forms of communication ].
3 Design of the NSVC
The NSVC is designed to provide reliable V2X (i.e., V2V and V2I communications)connectivity to vehicles in an urban traffic environment. To this end, we propose twomain mechanisms that are implemented in the NSVC: (1) neighbor stability-basedclustering that minimizes the number of changes in topology and the delay in clusterformation, and (2) classified transmission scheduling that supports fast transmissionsfor emergency messages, such as an eCall ].
3.1 Superframe structure
Figure shows the time-partitioned superframe structure of the NSVC that iscomposed with a beacon period (BP), clustering period (CP), high-priority data trans-mission period (HPDP), and general data transmission period (GDP). The duration ofthe superframe changes dynamically according to the traffic conditions. At the begin- ning of the superframe, each of the vehicles broadcasts a beacon frame that informs a Neighbor stability-based VANET clustering for urban vehicular.
Fig. 3 NSVC superframe structure
one-hop neighbor vehicle of its existence in the carrier sense multiple access with col-lision avoidance (CSMA/CA) manner through the use of BP. With this beacon frameexchange, every vehicle is assumed to be synchronized to each other. CP is the controlperiod where vehicles exchange control messages to set up a cluster, and as shown inFig. the CP is divided into two sub-periods: clustering decision (CD) and clusteringcontrol (CC). In the CD sub-period, the CH determines whether or not to execute anew CH election procedure, and CC is the duration to elect the new CH or mainte-nance of the cluster. Both HPDP and the GDP periods are the data periods for actualdata delivery. The data transmission in the HPDP is limited to emergency messages,such as an eCall, and it guarantees their successful transmission using polling-basedcommunication. Unlike HPDP, the GDP period accommodates all types of data. Vehi-cles can use GDP to communicate with each other in a contention-based CSMA/CAmanner 3.2 Neighbor stability-based clustering
As previously mentioned, the CP consists of two sub-periods: CD and CC. DuringCD, which is sub-period with fixed length, the existing CH determines whether or notto execute the new CH election procedure, and this consequently affects the length ofthe following CC sub-period. The flowchart in Fig. shows the clustering decisionprocedure. The notion of ‘neighbor stability (NS)' is used to make a decision of whetheror not to execute the new CH election procedure. During BP, the old CH counts thenumber of incoming/outgoing vehicles into/from its own cluster, respectively, andcalculates the rate of change for the number of neighbors, the NS value, which can beobtained as follows: (t) + Outgoing (t) i (t ) = IncomingiNi(t − 1) where NSi (t) is the NS value for vehicle ID i at the current beacon inter-val, Incoming ( i t ) and Outgoingi t ) are the number of incoming/outgoing vehicles into/from its own cluster coverage for vehicle ID i , and Ni (t − 1) denotes the numberof neighbors in the previous beacon interval. In the CD sub-period, the CH vehicle of the previous beacon interval selectively broadcasts two types of messages, a new J.-H. Kwon et al.
Fig. 4 Clustering decision procedure
CH elect (New_ch_elec) or an old CH retain (Old_ch_ret), depending on the decisionthat has been made. If the NSi (t) exceeds a pre-determined threshold, the old CHbroadcasts a New_ch_elec message to its CM vehicles, and otherwise broadcasts anOld_ch_ret message. According to the decision of the CH, all vehicles within a clusterselect two length types for the subsequent CC sub-period: Long-type CC and Short-type CC. When the CM vehicles receive a New_ch_elec message from the existing CHvehicle, they set their CC period to Long-type to elect the new CH vehicle. If none ofthe messages have been received during a certain time limit, the CC period is also setto a Long-type for the same reason. Meanwhile, the reception of Old_ch_ret messagemeans that the old CH vehicle of the previous beacon interval continues as a CH inthe current beacon interval. Thus, in this case, the CC period is set to Short-type onlyfor the cluster association of the newly entering vehicles.
Figure shows the CH election procedure in further detail. Upon receiving the New_ch_elec message, the vehicles start the CH election procedure during the CC sub-period. First, each vehicle calculates its own NSi (t) value and composes a neighborvehicle table (NVT). The NVT entry includes the vehicle ID, CH ID, direction ofmotion, and the NS value. Each vehicle advertises its NVT with a HELLO messageand updates it. This updated NVT information is then used by the vehicles to startto compose a new cluster. The NS values of its own NVT are taken into accountfor the vehicle with the lowest NS value to announce itself as a new CH and thenbroadcast a Ready-to-Clustering (RTC) message to its neighbors. Upon receiving theRTC, the vehicle responds to the CH with a Request-to-Join (RTJ) message. In thecase where multiple RTCs are received, it responds to the CH that has the lowestNS value. In the case where the NSs are the same, it selects the one with the lowestID. Finally, the new CH broadcasts a response to its CM vehicles and terminates theCH election procedure. After the cluster formation is terminated, the CH informs theinfrastructure of the cluster information, including the CH ID and CM IDs. Whenthe infrastructure has received the cluster information from the CH, it updates itsown cluster information. Then, the infrastructure communicates with the CH using Neighbor stability-based VANET clustering for urban vehicular.
Fig. 5 Cluster head (CH) election procedure
Meanwhile, when the Old_ch_ret message is received, the old CH vehicle continues as a CH, and only vehicles that are newly incoming initiate the cluster formationprocedure by sending RTJ to the old CH.
The operational example to elect the CH election is shown in Fig. and Table Both vehicles 1 and 10 are first selected as the CH. Although vehicle 6 has the sameNS value as vehicles 1 and 10, it cannot be a CH because it has a higher ID than vehicle1. The remaining vehicles (i.e., ID 3, 4, and 8) re-broadcast a ‘Hello' message and setin the same manner. Consequently, vehicles 1, 3, and 10 are selected as the CH.
3.3 Classified transmission scheduling
To support fast data delivery for emergency messages, such as an eCall, NSVC usesclassified transmission scheduling. In NSVC, vehicles with emergency data transmittheir data in the HPDP within a beacon interval (as shown in Fig. using polling-basedchannel access. Figure shows the data transmission in the HPDP. At the beginning ofthe HPDP, the CH periodically broadcasts the Emergency Discovery (E-Disc) message,designating a specific moving direction to check whether emergency data exist andneed to be sent. When receiving an E-Disc message, the CM vehicles check the movingdirection of the E-Disc and try to transmit its emergency data there is a match. When HPDP has finished, the GDP period starts for general data transmission.
J.-H. Kwon et al.
Fig. 6 Example of the cluster head (CH) election
Table 1 Example of neighbor
vehicles' mobility value 4 Performance evaluation
We evaluated the performance of NSVC by conducting experimental simulations. Weconsider the various performance metrics of a network running NSVC and comparethem to those of existing schemes, such as, MOBIC ] and LID In this section,we first present the setting and the configuration of the simulation. We then presentthe results that were obtained in further detail.
4.1 Setting and configuration
In the simulation, we consider the urban traffic scenarios on a 4 × 4 grid-type lay-out. Figure shows the overall layout of the simulation, which includes the possiblecoordinates for the initial positions of the vehicles This specifically consistsof sixteen identical square blocks, with five vertical and horizontal traffic roads that,respectively, surround the blocks. Except for the corner of the overall layout, whichis set to a single-lane road, a two-lane road is used in our simulation scenario. Thedirection of the vehicles in the two-lane road is exactly the opposite, and traffic keepsto the right of the two-lane road. To set the initial position of the vehicle, we randomly select the discrete coordinate (x, y) of the traffic without duplication For V2I Neighbor stability-based VANET clustering for urban vehicular.
Fig. 7 Procedure for emergency data transmission
Fig. 8 Overall layout of the urban traffic scenarios
communication, we place the infrastructure on the upper left and in the bottom rightcorners of each block. The simulation parameters are listed in Table and the vehi- cle mobility is applied by assuming that vehicles move at a velocity between 10 and J.-H. Kwon et al.
Table 2 Simulation parameters
10 ∼ 60 km/h Traffic application Number of vehicles Transmission range Control packet size Clustering period Emergency message High-priority data Clustering decision period Table 3 CSMA-specific parameters
60 km/h in a field of 1000 × 1000 m2. If the maximum velocity is 30, the velocitybetween 10 and 30 is randomly assigned to the vehicles. We further assume a 250-mtransmission range, 100 ms beacon interval, 6 Mbps data rate, and constant bit rate(CBR) traffic.
The NSVC is implemented over the IEEE 802.11p PHY layer model. Since IEEE802.11p supports four different access categories, CSMA-specific parameterssuch as CWmin, CWmax and arbitration interframe space (AIFS) are configured, asshown in Table ]. In the simulation, the access categories for the voice traf-fic (i.e., the highest priority) and the best effort traffic are used for emergency datatransmission and general data transmissions, respectively.
4.2 Simulation results
Figure shows the average CH lifetime according to the variations in the maximumvelocity. Overall, the CH lifetime decreases as the maximum velocity increases, anda large deviation in the velocities causes a frequent change of neighbors, which leadsto the re-election of the CH. NSVC exhibits a longer lifetime relative to others since itelects the vehicle with the lowest NS as the CH. Thus, the old CH that has the lowestNS retains its own CH position. On the other hand, the LID and MOBIC elect the CH using the vehicle ID and the aggregated relative mobility as criterion, respectively Neighbor stability-based VANET clustering for urban vehicular.
Fig. 9 Average cluster head lifetime for a variation in the maximum velocity
Fig. 10 Average rate of the cluster head change for a variation in the maximum velocity
Therefore, the CH should be re-elected more frequently. When the maximumvelocity is 30 km/h, the average NSVC lifetime is approximately 30 and 69 % longerthan MOBIC and LID, respectively.
Figure shows the average rate of cluster head changes per second. The average rate for the CH change is proportional to the inverse of the CH lifetime. Thus, theaverage rate of the CH change increases when the maximum velocity increases. In thisfigure, the NSVC can maintain the CH for a longer time when compared to the MOBICand LID, and CH is more frequently re-elected as the maximum velocity increases. Theaverage rates of the CH change for NSVC, MOBIC and LID are 0.10, 0.11 and 0.12,respectively, when the maximum velocity is 10 km/h. When the maximum velocityincreases to 30 km/h, the results for each scheme increase to 0.27, 0.35 and 0.43, J.-H. Kwon et al.
Fig. 11 Overall throughput for a varying number of vehicles (300 bytes packet)
Fig. 12 Average length of clustering period for a variation in the maximum velocity
Figure shows the overall throughput when the number of vehicles varies with a 300 bytes packet. In this experiment, the maximum velocity is set to 30 km/h, andall vehicles transmit a packet every 1.25 ms. The overall throughput increases whenthe number of vehicles increases, and when the number of vehicles increases from40 to 80, the overall throughput of the NSVC, MOBIC and LID increases from 54.6to 115.3 Mbps, from 48.2 to 101.7 Mbps and from 47.2 to 98.7 Mbps, respectively.
As a result, NSVC achieves a higher throughput than MOBIC and LID because theCH running NSVC conducts a CH election procedure only when needed. If the CHelection procedure is not executed during the CP, NSVC can spend more time onthe data transmission. For further clarification, we measure the average CP length, as shown in Fig. The increase in the maximum velocity results in a longer CP length Neighbor stability-based VANET clustering for urban vehicular.
Fig. 13 Overall throughput for various numbers of vehicles (600 bytes packet)
Fig. 14 End-to-end latency for various numbers of vehicles (300 bytes packet)
since NSVC requires a Long-type CC to elect the CH. The results of the measurementshow that the average CP length is between 10.2 and 10.7 ms. Compared to MOBICand LID, which use a fixed CP length of 30 ms, NSVC can maintain a longer datatransmission duration in a beacon interval.
Figure shows a comparison of the overall throughput when the vehicles transmit a 600 bytes packet. The overall throughput is higher than that of the 300 bytes casesince the increase in packet size implies fewer overhead for packet exchange. However,in this case, the number of transmitted packets can slightly decrease due to the limitedlength of BI. Similarly to the 300 bytes case, when the number of vehicles is 50,NSCV shows results that are 11.95 and 13.71 % higher than those of MOBIC and LID, respectively.
J.-H. Kwon et al.
Fig. 15 End-to-end latency for various numbers of vehicles (600 bytes packet)
Figure shows the average end-to-end latency for various numbers of vehicles with a 300 bytes packet. We measure the average packet transmission time betweenCH and CM within a cluster and set the maximum velocity to 50 km/h. In the figure,the end-to-end latency is affected by the number of vehicles since the dense trafficenvironment can cause a long delay in the channel access. The NSVC exhibits betterperformance than the others because it can transmit more packets in a BI. When thenumber of vehicles is 50, the end-to end latency of NSVC is shorter by 24.8 and 37.3 %than those of MOBIC and LID, respectively. Figure shows the average end-to-endlatency when the vehicles transmit 600 bytes packets. Similarly to the 300 bytes case,NSVC exhibits a shorter end-to-end latency than the others. The end-to-end latencyincreases when the packet size is large as a result of the increase in transmission time.
NSCV thus spends 8.58 % longer time for one packet transmission than the averageof the case with 300 bytes.
Figure shows the emergency message latency for various numbers of vehicles.
In this experiment, we set the number of vehicles and the maximum velocity to 50 and50 km/h, respectively. When the number of vehicles is small (i.e., below 8), NSVCwith CSMA shows a slightly shorter end-to-end latency due to the low contention levelwhile NSVC with classified transmission scheduling (CTS) shows a relatively highlatency due to the overhead of HPDP duration used in the CTS mechanism. However,when the number of vehicles with emergency message is more than 10, NSVC withCTS shows a better latency performance since NSVC with CSMA suffers from a longchannel access delay under a high contention.
This paper presents a neighbor stability-based VANET clustering (abbreviated NSVC).
To achieve reliable V2X communications, NSVC provides two mechanisms, i.e., Neighbor stability-based VANET clustering for urban vehicular.
Fig. 16 Latency for various numbers of vehicles with emergency message
neighbor stability-based clustering and classified transmission scheduling mecha-nisms. The former minimizes the number of changes in the topology and the delay inthe cluster formation, for which it takes into account the rate of change for the numberof neighbors for the clustering decision and the CH election. The latter supports a fasttransmission for emergency messages, such as an eCall, using polling-based chan-nel access. The performance was evaluated by conducting simulations with varioustraffic scenarios on a 4 × 4 grid type of a two-lane traffic road. As verified with thesimulations, NSVC achieves a substantial improvement in performance in terms ofthe CH lifetime, the number of topology changes, the overall network throughput, theend-to-end latency, and the emergency message latency.
This research was supported in part by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2014R1A1A2057641), by the ICT R&D program of MSIP/IITP [B0101-14-0059, Human ResourceDevelopment Program for Future Internet], and by the MSIP (Ministry of Science, ICT and Future Planning),Korea, under the ITRC (Information Technology Research Center) support program (IITP-2015-R0992-15-1006) supervised by the IITP (Institute for Information & communications Technology Promotion).
1. Jayavardhana G, Rajkumar B, Slaven M, Marimuthu P (2013) Internet of Things (IoT): a vision, architectural elements, and future directions. Future Gener Comp Sy 29(7):1645–1660. 2. Kim EJ, Youm S (2013) Machine-to-machine platform architecture for horizontal service integration.
EURASIP J Wirel Commun 2013(1):1–9. 3. Atzori L, Iera A, Morabito G (2010) The Internet of things: a survey. Comput Netw 54(15):2787–2805.
4. Task Group P (2009) IEEE draft amendment to standard for information technology— telecommunications and information exchange between systems—local and metropolitan networks— specific requirements—part 11: Wireless LAN medium access control (MAC) and physical layer (PHY)specifications: Amendment: Wireless access in vehicular environments. IEEE 802.11p Draft 9.0 J.-H. Kwon et al.
5. Uzcategui R, Acosta-Marum G (2009) WAVE: a tutorial. IEEE Commun Mag 47(5):126–133. 6. Xu Q, Mak T, Ko J, Sengupta R (2004) Vehicle-to-vehicle safety messaging in DSRC. In: Proceedings of 1st ACM international workshop on Vehicular ad hoc networks, pp 19–28. 7. Bai F, Stancil DD, Krishnan H (2010) Toward understanding characteristics of dedicated short range communications (DSRC) from a perspective of vehicular network engineers. In: Proceedings of 16thannual international conference on mobile computing and networking, pp 329–340. 8. Lee D, Chang SW, Lee SS (2015) Analysis and design on efficient message relay methods in VANET.
Multimed Tools Appl 74(16):6331–6340. 9. Vinel A (2012) 3GPP LTE versus IEEE 802.11 p/WAVE: which technology is able to support cooper- ative vehicular safety applications? IEEE Wirel Commun Lett 1(2):125–128. 10. Jarupan B, Ekici E (2009) Location-and delay-aware cross-layer communication in V2I multihop vehicular networks. IEEE Commun Mag 47(11):112–118. 11. Lin CR, Gerla M (1997) Adaptive clustering for mobile wireless networks. IEEE J Sel Area Commun 15(7):1265–1275. 12. Basu P, Khan N, Little, TD (2001) A mobility based metric for clustering in mobile ad hoc networks.
In: Proceedings of international conference on computing systems workshop, pp 413–418. 13. Li W (2011) Network clustering in vehicular communication networks. Dissertation, University of 14. Hamid RA, Reza EA, Abolfazl D, Atefe P (2015) A cluster-based vehicular cloud architec- ture with learning-based resource management. J Supercomput 71(4):1401–1426. 15. Kwon JH, Kim EJ (2015) Adaptive multi-channel allocation for vehicular infrastructure mesh systems.
Multimed Tools Appl 74(5):1593–1609. 16. Omar HA, Zhuang W, Li L (2013) VeMAC: A TDMA-based MAC protocol for reliable broadcast in VANETs. IEEE T Mobile Comput 12(9):1724–1736. 17. Singh JP, Bali RS (2015) A hybrid backbone based clustering algorithm for vehicular ad-hoc networks.
Procedia Comput Sci 46:1005–1013. 18. Kwon JH, Kwon C, Kim EJ (2015) Neighbor mobility-based clustering scheme for vehicular ad hoc networks. In: Proceedings of international conference on PlatCon-15, pp 31–32. 19. (2003) IEEE Standard for information technology—telecommunications and information exchange between systems—local and metropolitan area networks—specific requirements—Part 11: WirelessLAN medium access control (MAC) and physical layer (PHY) specifications. 20. Yoo J (2015) Receiver-centric physical carrier sensing for vehicular ad hoc networks. Multimed Tools Appl 74(7):2405–2415. 21. Kim EJ, In J, Youm S, Kang CH (2012) Delay attack-resilient clock synchronization for wireless sensor networks. IEICE Trans Inf 95(1):188–191. 22. ETSI TR 102 861 V1.1.1 (2012) Intelligent Transport Systems (ITS); STDMA recommended para- meters and settings for cooperative ITS; access layer part 23. ETSI TR 102 862 V1.1.1 (2011) Intelligent Transport Systems (ITS); Performance Evaluation of Self-Organizing TDMA as Medium Access Control Method Applied to ITS; Access Layer Part 24. Tsai MF, Wang PC, Shieh CK, Hwang WS, Chilamkurti N, Rho S, Lee YS (2015) Improving posi- tioning accuracy for VANET in real city environments. J Supercomput 71(6):1975–1995. 25. Chu IC, Chen PY, Chen WT (2012) An IEEE 802.11p Based distributed channel assignment scheme considering emergency message dissemination. In: Proceedings of vehicular technology conference,pp 1–5.
Impressum: Institut für Neue Kulturtechnologien/t0, Neustiftgasse 17, 1070 Wien, Verlagspostamt 1070 Wien, Sponsoring Post GZ02Z033689 S. t0 World Summit on the Information Society World-Information City TUNIS 16-18 November 2005 BANGALORE 14-20 November 2005 TABLE OF CONTENTS P 02 The Black and White (and Grey) of By Lawrence Liang Options to traditional patents - The