Underwater communication is a challenging topic due to its singular channel characteristics. Most protocols used in terrestrial wireless communication can not be directly applied in the underwater world. In this paper, we focus on the issue of energy efficient transmission in underwater sensor networks (UWSNs) and analyze this problem in a rigorous and theoretical way. We formalize an optimization problem which aims to minimize energy consumption and simultaneously accounts for other performance metrics such as the data reliability and the communication delay. With the help of Karush-Kuhn-Tucker conditions (KKT conditions), we derive a simple and explicit, but nevertheless accurate, approximate solution under reasonable assumptions. This approximate solution provides theoretical guidelines for designing durable and reliable UWSNs. Our result also shows that reliability and communication delay are crucial factors to the energy consumption for transmission.

Considering the constraint brought by mobility and resources, it is important for routing protocols to efficiently deliver data in Intermittently Connected Mobile Network (ICMN). Different from previous works that use the knowledge of previous encounters to predict the future contact, we propose a storagefriendly REgioN-bAsed protocol, namely, RENA, in this paper. Instead of using temporal information, RENA builds routing tables based on regional movement history, which avoids excessive storage for tracking encounter history. We validate the generality of RENA through time-variant community mobility model with parameters extracted from the MIT WLAN trace, and the vehicular network based on 8 bus routes of the city of Helsinki. The comprehensive simulation results show that RENA is not only storage-friendly but also more efficient than the epidemic routing, the restricted replication protocol SNW and the encounter-based protocol RAPID under various conditions.

The mobility of underwater sensor nodes makes the network topology inconveniently controlled and slowly changed. Thus, in order to enable underwater sensor networks to work more effectively, it is necessary for us to periodically detect the coverage rate and redeploy nodes to non-coverage areas. In this paper, we take the lead in introducing the redeployment issue in underwater sensor networks. In our opinion, the key point of the redeployment issue is coverage. For this special coverage topic, we first propose a coverage rate definition scheme. Then along with the definitions, two redeployment algorithms are introduced, of which one is based on adding new nodes while the other one is by the means of moving redundant ones. By modeling the mobility behavior of underwater nodes with three-dimensional random walks, we employ simulation experiments to verify our ideas, the results of which show the importance of redeployment in the underwater environment.

Most previous models about wireless channels are mainly adopted to obtain average packet error rates or to generate artificial network traces. However, when we combine packet group transmission with error correcting mechanisms, the steady packet error rate (PER) is not accurate enough to depict the short-term error event. In this paper, we propose a Markov chain model for group transmission to indicate influences of group length N and initial channel state on transmission reliability. Based on the model, we prove that the difference between steady PER and packet error rate within a group (PERG) is O(1/N), which can be used to estimate group length under a certain error rate constraint. Finally, we apply the model to compare the multipath transmission with the single-path transmission, and investigate an extreme case where the multipath way is less reliable than the single-path way.

As a potentially competitive technique, erasure coding has been employed in wireless sensor networks (WSNs) to enhance transmission reliability. In this paper, to design a practical and efficient redundancy mechanism in WSNs, we firstly provide a theoretical study of packet delivery probability and average energy consumption for retransmission and redundancy mechanisms. The theoretical results indicate that in WSNs, the adaptive redundancy coding mechanism is more energy efficient than retransmission while keeping the same level of reliability in most scenarios. Then based on the mapping table and two basic design principles obtained from the theoretical analysis, we propose a reliable transport protocol ARRTP which combines the adaptive redundancy mechanism and partial retransmission. The simulation and trace-driven results both verify that when the loss probability varies from low to high, our protocol provides reliability with comparably lower energy consumption. Furthermore, compared with fixed redundancy degrees, the robustness of the adaptive mechanism are also evaluated.

In underwater sensor networks (UWSNs), high-delay acoustic channels are used for communications, which introduces new challenges for analyzing and choosing proper reliable transport mechanisms. In this paper, we investigate four key parameters (node distance, communication frequency, packet length and SNR) influencing the transport of UWSNs. Since treating all the four factors equally will make the analysis almost intractable, we propose a four-step analytical framework. In this framework, by optimizing both the communication frequency and packet length, we firstly show that the average energy consumption and the average communication delay of different transport schemes are only related to node distance and SNR. Then when considering practical scenarios, we further demonstrate that the scheme performances are mainly determined by SNR, which rather simplifies the analysis process. In addition, an important metric PEDP is proposed, which is only related to SNR but has the capability of evaluating the performance of transport schemes roundly and effectively. Finally, the numeric results offer us some profound insights for choosing proper transport schemes in UWSNs.

More attention has been paid to congestion control in the emerging area of wireless sensor network (WSN). However, most research works in the past stayed at the level of algorithm design or modification, and seldom sought solutions on the viewpoint of architecture. In this paper, Active Networking (AN) technology is used to make congestion control more responsive to detect/recover congestion in WSN. We design a simple Active Backpressure (BP) mechanism to allocate bandwidth Proportional to the Size of tree (ABPS). ABPS introduces programs in each data packet that tell nodes how to react to congestion, and quickly converges to a fair and efficient rate. Finally, we evaluate ABPS extensively on a 50-node wireless sensor network. Simulation results validate the effectiveness of our ABPS.

As an application-driven network, wireless sensor network generally requires high data reliability to maintain detection and response capabilities. Although two approaches, which are retransmission and redundancy, have been proposed to enhance data reliability, the theoretical work is required to evaluate their impact on transmission reliability and energy efficiency. In this paper, we offer a comprehensive theoretical study on the packet arrival probability and average energy consumption for both approaches. Our analysis indicates that when loss probability remains low or moderate, erasure coding, a scheme based on redundancy, is more reliable and energy efficient than retransmission. However, the performance of erasure coding would largely deteriorate under high packet loss condition. We also demonstrate that its resistance capability against packet loss weakens as hop number increases. Furthermore, with the increase in redundancy, erasure coding has to sacrifice the advantage of energy efficiency for reliability.