Opportunistic and mobile networks/delay tolerant networks

Mobile Delay Tolerant Networks (MDTNs) also called Opportunistic Networs (OppNets) are wireless mobile networks where there does not exist a complete path from the source to the destination. A number of networking scenarios have been categorized as MDTNs, such as wildlife tracking sensor networks, vehicular ad hoc networks, military networks, pocket switched networks, etc. The network asynchrony, coupled with the limited resources of mobile devices (e.g., sensors, smart phones, tablets) make message dissemination (also called routing) one of the fundamental challenges in MDTNs. Data delivery in such networks is more challenging for big-scale data like image and video, especially for videos as it is a continous content. The objectives in this case is twice : how to route the video such that we can play it as soon as possible, with the better quality as possible

Challenges and Contributions

• Video dissemination in OppNets (ongoing: as part of Merza Klaghstan Phd thesis) We choose to consider video transmission that are encoded in SVC (Scalable Video Coding), an extension to H.264 MPEG-4 AVC that subdivides the video into one base layer, l0, and one or more enhancement (spatial or temporal) layers, l1 ... ln.). As the base layer is the most important for begining playing the video, we have proposed an adapted version of the spray-and-wait protocol [1] where the number of layer copies is proportional to their ability to enhance the video reading. This first work reveals a high rate of loss due to the video transmission grain inadequacy. The second proposed protocol [2], uses as data transmission units smaller units called NAL(Network Abstraction Layer). It is an adaptive transmission protocol, where the amount of data transmitted depends on peers inter-contact time. This protocol enhances considerably the video delivery and the video quality of play.
  We are now working on a backward loss concealment mechanism. This method is hybrid as it consists in asking for the minimal amount of lost chunks of data from the network so that the well-known Multimedia video cancealment methods could be applied by the .

• Efficient routing in MDTns (Past: as part of Jingwei Miao Phd thesis)

Contributors

Current contributors

• Nadia Bennani
• Merza Klaghstan
• Sonia Ben Mokhtar
• Zeina Torbey

External collaborators

• Gabriele Gianini
• David Coquil
• Harald Kosh

Past contributors

• Jingwei Miao

Grants

A franco-german grant for the thesis of Merza Klaghstan in the context of the MDPS doctoral college

Selected publications

[1] Merza Klaghstan, Nadia Bennani, David Coquil, Harald Kosch, Lionel Brunie, Contact-Based Adaptive Granularity for Scalable Video Transmission in Opportunistic Networks, IWCMC(10th International Wireless Communications and Mobile Computing Conference), ACM ed. Nicosi- Cyprus. IEEE . 2014.
[2] Merza Klaghstan, David Coquil, Nadia Bennani, Harald Kosch, Lionel Brunie, Enhancing Video Viewing-Experience in Opportunistic Networks Based on SVC, an Experimental Study. PIRMC 2013(IEEE Internatioanl Symposium on Personal, Indoor, and Mobile Radio Communications), IEEE ed. London. 2013.