Research On Anycast Routing Protocol For Wireless Sensor Networks

Recently, wireless sensor networks (WSN) have attracted a great deal of research attention due to their wide-range potential applications. Because a microsensor node operates on a tiny battery with limited energy and it is hard to recharge or replace battery, the foremost design consideration for WSN is how to save limited battery resource with high efficiency and improve the network lifetime.Anycast is one of three communication models defined by IPv6, and it plays an important role in many network applications. Anycast routing technology is a communication service which is to deliver a packet to any one member in a group of designated recipients. So, applying anycast routing in WSN, it would balance network energy and avoid the fast drain of sensor nodes with high energy consumption, thus improving the network lifetime. But at the present time, the study of the anycast communication technique for WSN is still a new research field, so inevitably there are some problems in performing it. These problems need further study and analysis.The existing problems of the anycast routing technique for WSN are deeply analyzed and studied in this dissertation, and some effective solutions and routing protocols are also proposed in this dissertation. The originalities and contributions of this dissertation are as follows:(1) For autonomy, intelligence and mobility properties of swarm intelligence are applicable in route searching for WSN, an anycast routing protocol for wireless sensor networks based on artificial bee colony (ABCARP) is proposed. Two-tier scouter architecture is adopted in ABCARP, short-distance scouters and long-distance scouters are respectively responsible for searching for node’s foraging zone and base stations, and foragers are responsible for carrying data packets to base stations. In comparison with AODV and Ant-AODV, for mobile object monitoring events, simulation experiments results show that the performance of ABCARP is better in term of both control overhead and energy efficiency. An anycast routing protocol based on mobile agent for WSN (MAARP) is proposed. In the protocol, only one-hop neighbor nodes’routing information is required by each sensor node, and their monitoring data needs to report to any one base station by their cluster-head. By mobile agents, cluster heads search anycast members routing information and establish a local anycast routing table. In this dissertation, energy-consuming models of on-demand anycast routing protocol for both WSN and our protocol are also proposed, experiments show that in contrast to on-demand anycast routing protocols for WSN, our protocol has a good performance of system energy-consuming while network’s scale or density increases.(2) Path-disjoint or node-disjoint scheme is often adopted to search routing paths in most routing protocols for WSN, but it cannot solve the transmission collision problem which is caused by the hidden terminal problem, thus bring much extra retransmission energy consumption. For this problem, an anycast routing algorithm for WSN based on zone-disjoint scheme (ZDARA) is proposed. In the algorithm, anycast paths is selected as the object path one by one in order of their neighbor nodes’ impact factors(the degree that the neighbor nodes participate in other anycast paths) and those paths also should be not collide with each other, in contrast to the zone-disjoint routing algorithm based on shortest path first (SPF) scheme, it can find more zone-disjoint anycast paths in our protocol. In comparison with node-disjoint scheme and zone-disjoint routing scheme based on SPF, experiments results show that the performance of our algorithm is better in term of the composite indicator about total energy consumption and energy balance, thus improving the network lifetime. For the hidden terminal problem, a geographic anycast routing algorithm for WSN based on zone-disjoint scheme (ZDGAR) is also proposed. For reducing the probability of transmission collision, both fan-shaped pipeline technique and the zone-disjoint transmission scheme based on different time slot are adopted in the algorithm. In comparison with node-disjoint scheme and zone-disjoint routing scheme based on SPF, experiments results show that the performances of ZDGAR are better in terms of both packet delivery fraction and node’s average energy consumption, thus improving the network lifetime. Although the two algorithms above can prevent or reduce transmission problem probability caused by the hidden-terminal problem, it bring the problem of sharp reduction of usable anycast paths, thus it reduce the capacity of anycast technique for balancing energy consuming which might result in a bad energy efficiency. For the new problem, an anycast routing protocol based on transmission collision probability for WSN (TCARP) is proposed. In the protocol, how to calculate the probability of transmission collision is detailed discussed. The routing criterion of the protocol is based on the composite indicator about both load balancing and minimum overall energy cost, and the two factors are balanced by parameter a. Compared with non-collision routing protocols and traditional routing protocols based only on routing distance but without transmission collision probability anylysis, experiment results show that performance of our protocol is improved in terms of both energy balancing and network energy consuming, hence the network lifetime is optimized.(3) The influence to the anycast routing technique by client’s QoS constraints is deeply analyzed and studied in this dissertation. In WSN with sleep-wakeup scheduling, anycast technique can significantly reduce time delay. However, previous research work only attend to optimizing the time delay during each hop, this scheme is not optimal, sometimes even bad for end-to-end time delay. In this dissertation, an optimal anycast routing algorithm for WSN (DCARA) is proposed to solve the problem of end-to-end time delay. In the algorithm, AODV-based multipath routing scheme is applied to acquire anycast routing information, then genetic algorithm is applied to search the optimal anycast paths between each nodes and bases stations. For our algorithm has features of both global optimization and self-adaptivity of switching anycast paths, in contrast to previous algorithms, experiment results show that our algorithm can efficiently reduce end-to-end time delay. Client’s QoS constraints may be not just only one. For the problem of anycast routing protocol with both bandwidth constraint and time delay constraint, a polynomial time approximation optimization algorithm (ARAS) is proposed. Since this optimization question is a NP-hardness problem, our optimization algorithm is a near-optimal solution. Simulation results show that the performances obtained by our protocol are better than by our proposed heuristic algorithms (shortest path first anycast routing algorithm and maximum bandwidth path first anycast algorithm), especially in a bandwidth-limited large-size network.(4) In traditional routing protocols for mobile sinks in wireless sensor networks, sinks should repeatedly broadcast their current location information to all sensor nodes while sinks are moving, but this process consumes a large amount of energy. For the problem above, an anycast routing protocol for mobile sinks in wireless sensor networks (ARPMS) based on predictive strategy is proposed. In ARPMS, sinks spread their moving information only when sinks change their moving direction or moving speed, by those moving information, sensor nodes could calculate(predict) sinks current and future location and select the path to a sink having the highest energy efficiency as the anycast path. For sinks need not spread their moving information at all times, ARPMS saves much energy cost hence the performance of ARPMS of energy efficiency is improved.