The idea is to dynamically copy the message to the sensor node that is more likely to communicate with the sink node. SRAD consists of two main parts: data transmission and queue management: the former is based on the Random Waypoint random movement model at different times. The transmission probability of each sensor node is used to transmit data messages; the latter determines the priority and discarding principle of message delivery in the queue through the survival time ST (survival time) value of the message to further reduce network transmission energy consumption. The simulation experiment results show that ,Compared with several existing DTMSN (delay tolerant mobile sensor networks) data transmission algorithms, SRAD has a relatively long network life, and it can achieve higher data transmission success with lower data transmission energy consumption and transmission delay. Rate.
With the development of sensor technology, embedded technology and low-power wireless communication technology, it has become possible to produce miniature wireless sensors with sensing, wireless communication and information processing capabilities. These cheap, low-power, highly flexible sensor nodes Cooperate with each other and organize into wireless sensor networks. Data collection is the basic function of most wireless sensor network applications [1]. At present, the common method of collecting data information is to place a large number of small, low-price, and Sensor nodes with low battery energy and low power are used to form an interconnected wireless mesh network, in which each node can cooperate with one or more other nodes to transmit measured environmental parameters or collected data to a sink for processing. [2]. However, this data collection method is not suitable in some situations, such as the collection process of recording data on the living habits of wild animals in biological research, and the recording and collection of harmful substances inhaled by each person on average every day in air quality monitoring. The process of gas quantity, or influenza virus tracking, in order to prevent the spread of influenza viruses, it is necessary to regularly collect influenza virus information in high-density crowd areas. Compared with general applications, these applications have some unique properties: First, the collection of data The process is oriented to moving objects (humans or animals). Although data can be collected by placing some sensor nodes at specific locations, in order to ensure the validity and accuracy of the data, the method of obtaining data directly from moving objects is usually adopted. Therefore , it becomes a natural premise to configure a sensor unit for each moving object. Obviously, the random movement of objects leads to the non-connectivity of some sensor nodes; secondly, these applications allow a certain data delay. Due to the intermittent connectivity of the nodes Sexually, the transmission delay of data in DTMSN (delay tolerant mobile sensor networks) is often high. In addition, the collection process of data information should be transparent and not affect the daily life of humans or animals. For example, we cannot command certain people Perform some special actions or move to a specific location to facilitate the collection and transmission of information.
In order to meet the needs of the above applications, the delay-tolerant mobile wireless sensor network (DTMSN) has emerged [3]. DTMSN consists of two types of nodes: randomly moving sensor nodes and convergence points. The former is bound to moving objects and is used to collect data. data information and form a discontinuously connected mobile sensor network (as shown in Figure 1, this network consists of 9 randomly distributed mobile sensor nodes S1~S9 and 2 convergence points H1, H2. At this moment, only S1, S3, S6 are connected to S7 and S8, S5 and H1). Based on the short transmission distance limitation of the sensor node, it is impossible to directly transmit the collected data to the destination. In addition, some parts are placed in a specific location or carried by some moving objects. High-end nodes that move with the movement of objects are used as convergence points (H1 and H2 in Figure 1) to collect data from sensors and transfer these data to the entry point of the backbone network.
Due to the poor connectivity between mobile sensors in the DTMSN network, it is very difficult to form an interconnected mesh network between each sensor node, that is, there may not be an end-to-end connectivity path between each node. It can be seen that traditional sensor networks The data transmission algorithm is not applicable in the DTMSN environment. In the intermittent connectivity environment of DTMSN, in order to achieve a certain data transmission success rate, data replication is necessary, and replication will also increase the transmission energy consumption of the system. Then ,How to effectively transfer the data collected by sensor nodes to the convergence point to achieve a balance between data transmission success rate, transmission energy consumption and transmission delay has become the primary problem to be solved by DTMSN. In view of the above factors, this paper proposes A dynamic data transmission strategy SRAD (selective replication-based adaptive data delivery scheme) based on selective replication is proposed. The basic idea is to dynamically copy messages to sensor nodes that are more likely to communicate with the convergence point, so as to maximize the The purpose of transmission success rate and reducing transmission energy consumption. SRAD consists of two main parts: data transmission and queue management. The former first calculates the transmission probability of each sensor node at different times through the Random Waypoint random motion model [4], that is, the sensor node The probability of the message being delivered to the convergence point, and then selecting the next hop node based on the transmission probability and copying and transmitting the data message. Queue management uses the survival time ST (survival time) value of the message to determine the importance of the message in the queue. and discarding principles. The simulation experiment results show that compared with the existing flooding algorithm, direct transmission algorithm, and FAD (fault tolerance delivery scheme) strategy, SRAD has a relatively longer network life, and it It can achieve higher data transmission success rate with lower data transmission energy consumption and transmission delay.
Section 1 of this article describes related work. Section 2 describes the motivation for this article and the motion model used. Section 3 gives the detailed design of SRAD. Section 4 conducts simulation verification. Finally, the full text is summarized.