4. Video surveillance continuously transmits two real-time video streams from each of the trains to the control center. Video surveillance is based on Closed
2.3 Traffic Control: Traffic Conditions Video Surveillance
For traffic control purposes it might be necessary to ascertain the current situ- ation at a given road section, intersection or even lane. Thanks to the benefits of global positioning systems, traffic management center can activate the ex- ternal cameras of vehicles located in the geographical area of interest. Video information with the current road views is then compressed at the vehicle side and transmitted back to the management center. Real-time reaction to traffic jams caused by accidents can be achieved if the video surveillance system of traffic conditions is combined with the eCall [12] or a similar system, which automatically notifies the emergency services of the crash.
3 Performance Evaluation
In future practical scenarios there are likely to be numerous closely located vehicles executing one or several of the above applications, which transmit video over the IEEE 802.11/WAVE service channels. The required video bit rate at each vehicle can be achieved by varying the video compression parameters, that define the trade-off between the visual quality and compression ratio. In order to minimize the video packet drops, the selected video bit rate of a vehicle should be consistent with the multiple access channel throughput.
One way to solve the above problem was introduced in our paper [10], where we propose that each vehicle selects its video bit rate in such a way that the service channel resources are allocated equally among all the neighboring trans- mitting vehicles. The key factor in this approach is proper estimation of channel throughput, that can be allocated to one vehicle, depending on the number of other transmitting vehicles in the system.
The following is the simplest estimate of the IEEE 802.11/WAVE service channel throughput per user:
SˆSCH(N) =μSCHãR
N , (1)
where μSCH is the percentage of time allocated to the service channel in the control/service channels alternating scheme [2], R is the service channel data rate andN is the number of neighboring vehicles simultaneously transmitting video information on the service channel, estimated on the basis of the informa- tion from cooperative awareness messages broadcast frequently on the control channel [3].
2 4 6 8 10 12 0
500 1000 1500 2000 2500 3000
Number of vehicles
Service channel bitrate per vehicle, kbps
IEEE 802.11p/WAVE Upper bound
Fig. 1.IEEE 802.11p/WAVE service channel throughput per vehicle
2 4 6 8 10 12
28 30 32 34 36 38 40 42 44
Number of vehicles
Y−PSNR, dB
container coastguard foreman highway hall
mother−daughter
Acceptable visual quality
Low visual quality High visual quality
Fig. 2.Visual quality per vehicle
1 )N
a b)N 6 c)N 12
Fig. 3.Received video frame of test video sequence “highway“ for different number of simultaneously transmitting vehicles
1 )N
a b)N 6 c)N 12
Fig. 4. Received video frame of test video sequence “hall“ for different number of simultaneously transmitting vehicles
1 )N
a b)N 6 c)N 12
Fig. 5.Received video frame of test video sequence “coastguard“ for different number of simultaneously transmitting vehicles
In fact, ˆSSCH(N) is an upper bound of the real throughput value, because it neglects the overhead caused by the backoff algorithm, collisions and availability of the service channel. A more precise estimate can be calculated as:
SSCH(N) = μSCHãρ(N)ãS802.11(N)ãR
N , (2)
whereρ(N) is the probability of service channel availability due to the successful reception of the corresponding wireless service advertisement [16],S802.11(N) is the throughput of legacy 802.11 basic access scheme withN saturated stations.
The values of SSCH for R = 6 Mbit/s, μSCH = 0.46 [2] and the different number of transmitting vehicles N are shown in Fig. 1. On the basis of these data we demonstrate the corresponding maximum achievable visual quality, as shown in Fig. 2. For visual quality measurements we use x.264 codec [13], which is the real-time software implementation of the H.264/AVC standard [14]. The test video sequences “container“, “coastguard“, “foreman“, “highway“, “hall“,
“mother-daughter“ [15] with a frame resolution 352×288, 30 frames per second were used. These sequences correspond to typical scenarios for the considered applications, e.g. indoor video surveillance with high or low level of objects mobility, view from the moving vehicle, etc.
The well-known peak signal-to-noise ratio (PSNR) was used as a visual quality metric. Typically, PSNR greater than 36 dB corresponds to high visual quality (received video either cannot be distinguished visually from the captured one).
PSNR less than 30 dB corresponds to low visual quality (received video contains a lot of distortions). PSNR from 30 to 36 dB corresponds to acceptable visual quality. For video quality estimation each test video sequence is compressed by x.264 encoder for given bit rate per user. Then video stream is decompressed and PSNR values between original and reconstructed video sequences are computed.
From Fig. 2, it follows that high and acceptable video quality can be achieved if the number of closely located vehicles broadcasting video simultaneously on a selected service channel does not exceed 5–12 in most of the cases. Taking into account that six service channels are available, according to [2], the above maxi- mum number of vehicles can be increased up to 40–70 depending on application.
Examples of received video frame of test video sequences for different number of simultaneously transmitting vehicles are given at Fig. 3–5.
4 Conclusion
We have introduced novel video-based vehicular applications that promote road safety and public security as well as the efficiency of road traffic control. The key feature of the proposed applications is that many moving vehicles perform video transmission to each other and to the roadside infrastructure. We have demonstrated that the emergent IEEE 802.11p/WAVE standard can serve as a basis of communication for the above applications, given that video bit rate is chosen in accordance with the available throughput of the service channels used.
Our analysis shows that 40–70 vehicles in the vicinity can be served simultane- ously with acceptable visual quality, which in our view confirms the practical feasibility of the proposed approach.
References
1. IEEE 802.11p, Wireless Access in Vehicular Environments (July 2010)
2. IEEE 1609.4-2010, IEEE Standard for Wireless Access in Vehicular Environments (WAVE) - Multi-channel Operation (February 2011)
3. Karagiannis, G., Altintas, O., Ekici, E., Heijenk, G.J., Jarupan, B., Lin, K., Weil, T.: Vehicular networking: A survey and tutorial on requirements, architectures, challenges, standards and solutions. IEEE Communications Surveys and Tutori- als 13(4), 584–616 (2011)
4. Tsai, M.F., Chilamkurti, N., Zeadally, S., Vinel, A.: Concurrent multipath trans- mission combining forward error correction and path interleaving for video stream- ing. Computer Communications 34(9), 1125–1136 (2011)
5. Qadri, N.N., Fleury, M., Altaf, M., Ghanbari, M.: Multi-source video streaming in a wireless vehicular ad hoc network. IET Communications 4(11), 1300–1311 (2010) 6. Asefi, M., Mark, J.W., Shen, X.: An Application-Centric Inter-Vehicle Routing Protocol for Video Streaming over Multi-Hop Urban VANETs. In: IEEE Interna- tional Conference on Communications (ICC 2011), Kyoto (June 2011)
7. Zhou, L., Zhang, Y., Song, K., Jing, W., Vasilakos, A.: Distributed Media Ser- vices in P2P-Based Vehicular Networks. IEEE Transactions on Vehicular Technol- ogy 60(2), 692–703 (2011)
8. Soldo, F., Casetti, C., Chiasserini, C.-F., Chaparro, P.A.: Video Streaming Distri- bution in VANETs. IEEE Transactions on Parallel and Distributed Systems 22(7), 1085–1091 (2011)
9. Gomes, P., Olaverri-Monreal, C., Ferreira, M.: Making Vehicles Transparent through V2V Video Streaming. IEEE Transactions on Intelligent Transportation Systems 13(2), 930–938 (2012)
10. Vinel, A., Belyaev, E., Egiazarian, K., Koucheryavy, Y.: An overtaking assistance system based on joint beaconing and real-time video transmission. IEEE Transac- tions on Vehicular Technology 61(5), 2319–2329 (2012)
11. Liu, H., Chen, S., Kubota, N.: Special Section on Intelligent Video Systems and Analytics. IEEE Transactions on Industrial Informatics 8(1) (2012)
12. eCall:http://ec.europa.eu/information society/activities/
esafety/ecall/index en.htm 13. x.264 video codec:http://x264.nl/
14. Advanced video coding for generic audiovisual services, ITU-T Recommendation H.264 and ISO/IEC 14496-10, AVC (2009)
15. Xiph.org test media:http://media.xiph.org/video/derf/
16. Campolo, C., Molinaro, A., Vinel, A., Zhang, Y.: Modeling Prioritized Broadcast- ing in Multichannel Vehicular Networks. IEEE Transactions on Vehicular Technol- ogy 61(2), 687–701 (2012)