Practical Bloom filter based epidemic forwarding and congestion control in DTNs: A comparative analysis

Epidemic forwarding has been proposed as a forwarding technique to achieve opportunistic communication in Delay Tolerant Networks (DTNs). Even if this technique is well known and widely referred, one has to address several practical problems before using it. Unfortunately, while the literature on DTNs is full of new techniques, very little has been done in comparing them. In particular, while Bloom filters have been proposed to exchange information about the buffer content prior to sending information in order to avoid redundant retransmissions, up to our knowledge no real evaluation has been provided to study the tradeoffs that exist for using Bloom Filters in practice. A second practical issue in DTNs is buffer management (resulting from finite buffers) and congestion control (resulting from greedy sources). This has also been the topic of several papers that had already uncovered the difficulty to acquire accurate information mandatory to regulate the data transmission rates and buffer space. In this paper, we fill this gap. We have been implementing a simulation of different proposed congestion control schemes for epidemic forwarding in ns-3 environment. We use this simulation to compare Preprint submitted to Computer Communications April 1, 2014 different proposed schemes and to uncover issues that remain in each one of them. Based on this analysis, we proposed some strategies for Bloom filter management based on windowing and describe implementation tradeoffs. Afterwards, we propose a back-pressure rate control as a well as an aging based buffer managing solution to deal with congestion control. By simulating our proposed mechanisms in ns-3 both with random-waypoint mobility and realistic mobility traces coming from San-Francisco taxicabs, we show that the proposed mechanisms alleviate the challenges of using epidemic forwarding in DTNs.