Huajun Meng’s research while affiliated with Shanghai Ocean University and other places

In wireless sensor network (WSN), the reliability and energy consumption of nodes are very important parameters to extend the lifetime. This paper proposes a concept that balances reliability and residual energy to control the success rate of message transmission. By formulating the weights of the reliability and remainder energy, the reliability-energy metric is designed to measure routing path and an efficient algorithm is developed based on Dijkstra algorithm to search a reliability-energy disjoint path set. The simulation shows the proposed algorithm is effective and efficient.

We consider the undirected simple connected graph for which edges fail independently of each other with equal probability 1-p and nodes are perfect. The all-terminal reliability of a graph G is the probability that the spanning subgraph of surviving edges is connected, denoted as R(G,p). Graph G∈Ω(n,e) is said to be uniformly least reliable if R(G,p)≤R(G ' ,p) for all G ' ∈Ω(n,e), and for all edge failure probabilities 0<1-p<1. In this paper, we prove the existence of uniformly least reliable graphs in the class Ω(n,e) for e≤n+1 and give their topologies.

The uniformly optimal graph problem with node failures consists of finding the most reliable graph in the class Ω(n,m) of all graphs with n nodes and m edges in which nodes fail independently and edges never fail. The graph G is called uniformly optimal in Ω(n,m) if, for all node-failure probabilities q∈(0,1), the graph G is the most reliable graph in the class of graphs Ω(n,m). This paper proves that the multipartite graphs K(b,b+1,…,b+1,b+2) are uniformly optimal in their classes Ω((k+2)(b+1),(k2+3k+2)(b+1)2/2−1), where k is the number of partite sets of size (b+1), while for i>2, the multipartite graphs K(b,b+1,…,b+1,b+i) are not uniformly optimal in their classes Ω((k+2)b+k+i,(k+2)(k+1)b2/2+(k+1)(k+i)b+k(k+2i−1)/2).

Wireless sensor network (WSN) is a self-organizing network which is composed of a large number of intelligent sensor nodes. Since the energy exhaustion of some nodes, the network topology is also changing with the disappearance of some sensor nodes, and even induces the disconnection. The system reliability of the network is also affected by the change of network topology. Although some sensor nodes are dead and impossible to be reactivated, the connectivity of the network can be repaired by adding some new nodes and continue to work in high reliability, this is so-called the repairability of WSN. This paper proposes an idea to redefine the reliability for repairability of disconnected network. A new reliability is proposed as the probability that the edge-induced subgraph is connected. Different from traditional all-terminal reliability, this new reliability measure focuses on residual edge connectedness and is able to distinguish the reliabilities of different tree topologies. Furthermore, we give the repairability measure by considering a reliability and disconnected network. The examples illustrate the effectiveness of the proposed measure and method.

A wireless sensor network is envisioned as a group of tiny power-constrained devices with functions of sensing and communication, which can be scattered over a region to enable monitoring of the region. One of challenges in developing this technique is the problem of topology control. The main idea of topology control is reducing node energy consumption and/or increasing network capacity. This paper considers an optimal problem, which is to search a minimal node-set such that its sensing area can cover an expected area. To solve the optimal problem, the weighted sensing topology is introduced, and some strategies, such as finding the ‘basic’ maximal independent node-set and estimating the coverage area of a node-set, are proposed to design a heuristic algorithm. Further, it is proved that the proposed algorithm can obtain the exact optimal solutions in the case that the resulting optimal solutions are independent node-sets, and the complexity of the heuristic algorithm is O(n2). The simulation shows its effectiveness and efficiency.

In wireless distributed sensor network, the efficiency of energy consumption is very important in developing the techniques to extend its lifetime. Due to the random failures of sensor nodes, this paper proposes a concept of the expected energy consumption to control the success rate of message transmission, decrease the energy consumption, and measure the network performance. Furthermore, based on the minimum energy reliable path and its two algorithms, restricted shortest path algorithm and "restricted" Dijksta algorithm, an algorithm is developed to compute the expected energy consumption of a sensor network. The simulation shows the effectiveness and efficiency of the proposed algorithm. Finally, the influence of reliability threshold on the expected energy consumption of a network is simulated and analyzed.