WSN (Wireless Sensor Network) is composed of hundreds of low-cost, low-power, small-size, multi-functional sensor nodes deployed in the detection area, and self-organized by single or multiple hops formed by wireless communication. A network system whose purpose is to sense, collect, and process information about a perceived object in a network coverage area and send it to an observer. WSN is widely used in military, commercial, medical care and environmental monitoring.
According to the topology of the node, it can be divided into a planar routing protocol and a hierarchical routing protocol [1].
The flat routing protocol is simple and robust, but its scalability is poor. Hierarchical routing protocols are generally divided into an initialization phase and a data transfer phase. The algorithm is different, and the elected cluster heads may be different, and the data transmission process is basically the same.
1 Uniform clustering routing protocol - LEACH protocol
In the initialization phase [2-3], each node generates a random number between 0 and 1. If it is less than the threshold [2-3], then this node is the cluster head, and it broadcasts to the surrounding nodes that it is a cluster. Head message. According to the strength of the received signal, the ordinary node selects the cluster to be added and informs the corresponding cluster head. At this time, all the cluster heads must be in the receiving state. When the cluster head receives all the join information, it generates a TDMA message to inform the working hours of all nodes in the cluster.
In the data transmission phase [2], the normal node transmits data to the cluster head according to the TDMA [4] time slot. The cluster head fuses the received data and then forwards it to the sink. After a while, reselect the cluster head.
The random election of the cluster head avoids the premature consumption of the cluster head energy and prolongs the survival time of the network, but the data transmission adopts a single-hop method, so that the cluster head farther from the sink spends a lot of energy, resulting in a change in the survival time. Short; frequently electing cluster heads also consumes energy. In order to save resource overhead, the data transfer phase takes longer than the initialization phase.
2 Non-uniform clustering routing protocol
2.1 EEUC Agreement
In the initialization phase, the sink broadcasts a signal to the entire network, and the node calculates its distance to the sink based on the strength of the received signal. According to the preset probability threshold [5], some nodes are selected to become candidate cluster heads to participate in the competition, and the nodes not participating in the competition enter a sleep state until the end of the campaign process. Si is any candidate cluster head, and its distance from the sink is its competition radius [6]. If Si wins, all candidate cluster heads in the competition radius will exit the election. The competition radius of the candidate cluster head decreases as the cluster head to sink distance decreases.
In the data transmission phase, the ordinary node transfers the collected data to the cluster head, and the cluster head performs processing to transfer the data to the sink in a multi-hop manner.
2.2 DEBUC Agreement
The protocol uses a time-based cluster head competition algorithm. The broadcast time depends on the remaining energy of the candidate cluster head and the remaining energy of its neighbor nodes. The candidate cluster heads closer to the sink have a smaller competition range, so that these cluster heads consume less energy in intra-cluster communication, and the saved energy is used for data forwarding between clusters. In the data transmission phase, an inter-cluster multi-hop routing protocol is adopted.
In the initialization phase, the normal node calculates the approximate distance from the sink based on the strength of the signal sent by the sink. First, a threshold is set to control the proportion of candidate cluster heads, and a competition radius [7] is also set for each candidate cluster head. The competition radius of the candidate cluster head is proportional to its distance from the sink.
The candidate cluster head broadcasts the message, and the normal node sleeps. The candidate cluster head that receives the message updates its neighbor node information table, and the candidate cluster head broadcasts the FINAL_HEAD_MSG[7] message according to its own time schedule, and declares itself to be a cluster head. After the cluster head selection is completed, the ordinary node exits the sleep, the cluster head broadcasts the message, and the ordinary node joins the nearest cluster head according to the strength of the received information, and notifies the cluster head that the relay node does not have the capability of data fusion. First, the cluster head broadcasts a message. If the distance from the neighbor cluster head to the sink is small, the cluster head calculates the approximate distance from the neighbor cluster head and establishes a neighbor cluster head information table; the cluster head uses the greedy algorithm in its neighbor cluster head set. Select its relay node, if the relay node of the cluster head is itself, send the data directly to the sink, otherwise the cluster head sends data to the relay node; when each cluster head finds the relay node, the inter-cluster multi-hop routing set up.
In the data transmission phase, the cluster head first fuses the received data and then sends the processing result to the sink.
As the energy of the cluster head decreases, the competition radius of the non-uniform clustering routing protocol decreases gradually, which requires re-clustering. The more energy is reduced, the more clusters are clustered, so in the process of clustering It needs to consume more energy, and some nodes will consume the remaining energy in the process of clustering.
3 PEGASIS Agreement
The PEGASIS protocol assumes that all nodes have global knowledge of the network topology. In the chain-building phase [8-10], the chain is first built from the node farthest from the sink. This node searches for the nearest node to join the chain according to the greedy algorithm. In this way, all nodes are added to the chain in this way. In the data communication phase [8-9], each node in the chain only communicates with its own neighbor node, fuses the received data with its own data, and transmits it to the neighbor node of the next hop, and transmits it to the chain head node. Finally, the data is transferred to the sink by the chain head node.
Through the analysis of the above typical routing algorithm, we can find that the following problems still exist:
(1) In the clustering phase, energy is still wasted to build clusters.
(2) Many protocols assume that sensor nodes and sinks do not move. Once the sensor nodes move, these protocols are likely to no longer hold.
(3) The non-uniform clustering routing protocol alleviates the "hot zone", but as the cluster head energy decreases and the competition radius decreases, the network topology is required to dynamically update the network topology, network topology. The update of the structure consumes more energy to achieve.
(4) The non-uniform clustering algorithm requires that the sensor nodes in the network are preferably evenly distributed. If the density of the sensor nodes is large in the area close to the sink, and the distribution density of the sensor nodes is small in the area far from the sink, Then the cluster head close to the sink will still form a "hot zone". This requires a better agreement to solve such problems.
(5) Most protocols adopt the method of re-splitting when considering that the sensor node fails to exit the network or a new node joins the network. If there are few nodes joining the network, the energy wasteed by re-clustering will be large, which requires the protocol to have high fault tolerance to cope with the topology changes of the network.
(6) As the network scale becomes larger and larger, the current algorithm cannot meet the requirements of hyperscale networks at all, and a multi-layer clustering algorithm needs to be proposed. In the multi-layer clustering algorithm, if the number of layers is large, there may be some nodes that have run out of energy in the initialization phase; if the number of layers is small, the superiority of the multi-layer clustering algorithm cannot be realized at all. Therefore, when using a layered algorithm, it is most appropriate to consider the number of layers.
With the development of WSN routing technology, more and more new algorithms will be proposed. The new algorithm should better deal with the load balancing of the cluster head and minimize the energy waste caused by the topology in the cluster formation stage. In short, the research of WSN routing technology is inseparable from hot issues such as load balancing, energy efficiency, and network lifetime.
references
[1] Ren Fengyuan, Huang Haining, Lin Biao. Wireless sensor network [J]. Journal of Software, 2003, 14 (7): 1282-1291.
[2] Guo Qiangang, Zhou Dexiang, Zhou Xifeng. The optimal cluster head number calculation method for LEACH routing protocol [J]. Microcomputers and Applications, 2013, 32(3): 61-66.
[3] HEINZELMAN WR, CHANDRAKASAN A, BALAKRISH-NAN H [C]. Energy-Efficient CommunicaTIon Protocol for Wireless Microsensor Networks, 2000: 3005-3014.
[4] Liu Jun, Li Yan, Qi Hua. Improvement and simulation of LEACH protocol based on NS2 wireless sensor network [J]. Electronic Technology Applications, 2012, 38(2): 21-27.
[5] Li Chengfa, Ye Mao, Chen Guihai, et al. An energy-efficientunequal clustering mechanism for wireless sensor networks [C]. IEEE InternaTIonal Conference on Mobille Adhoc and Sen-sor Systems Conference, 2005: 597-604.
[6] Li Chengfa, Chen Guihai, Ye Wei, et al. A wireless sensor network routing protocol based on non-uniform clustering [J]. Journal of Computers, 2007, 30 (1): 27-36.
[7] Jiang Changjiang, Shi Weiren, Tang Xianlun, and so on. Energy-equalized wireless sensor network non-uniform clustering routing protocol [J]. Journal of Software, 2012, 23(5): 1222-1232.
Our company is specialized in supplying Manifold Gauge.We have gauges for all common refrigerant like R22,R134A,R410A,R404A,R407C,R32,R1234YF.It applies to wide range of applications, like installation and maintenance of household air conditioner,automobile air conditioner,refrigerator,cold room and other industries .OEM and customization is welcomed.Our parts have been exported to over 50 countries all over the world and are always got good comment by customers.
Manifold Gauge
Manifold Gauge,Manifold Gauge Set,Ac Manifold Gauge Set,Ac Manifold Gauges
ZHEJIANG ICE LOONG ENVIRONMENTAL SCI-TECH CO.,LTD. , https://www.china-refrigerantgas.com