基于网络编码和压缩感知的无线传感器网络节能算法研究

发布时间：2018-04-14 07:56

本文选题：无线传感器网络 + 网络编码　；参考：《广西大学》2014年硕士论文

【摘要】：节点能量有限已成为制约无线传感器网络性能的主要瓶颈,在保障信息传输质量的前提下,如何充分利用有限的网络资源、减少网络节点能耗、延长网络生命周期等已成为无线传感器网络研究工作中的重点问题。网络编码通过对数据编码后再转发的方式,能够有效提升数据投递率、网络吞吐量及网络能量效率。压缩感知将信号投影到稀疏域,通过部分数据能够以较高概率恢复出原始信息的近似值,从而减少了网络中传输的数据量,节约了网络资源。本文的研究工作以降低网络能耗为目标,从减少冗余传输数据量出发,结合网络编码和压缩感知,对无线传感器网络中传感器节点的节能方式进行深入了研究,主要的工作和创新点如下： 1、提出一种基于随机网络编码的无线传感器网络多路径节能算法(RNC-ESMP)。该算法首先综合考虑网络节点剩余能量和节点间通信能耗的路径选择概率,引入条件传输价值比；接着构建从信宿节点到信源节点的反馈机制；最后提出中继节点编码选择方案。实验结果表明,该算法有效减少了传输时延和编码节点数,从而实现节能,而对比其他多路径和网络编码算法,RNC-ESMP能有效降低网络平均能耗约15%-50%,减少数据传输时延约12%-33%,从而提高了网络性能。 2、提出一种结合随机网络编码和占空比的无线传感器网络节能算法(NCDES)。该算法提出首先根据接收数据的ID信息决定节点处于工作或睡眠状态,以避免数据重复接收；并通过结合随机网络编码,增加在相同传输次数下传输的数据信息量,从而实现节约网络能耗。并且,本文通过理论计算分析了NCDES算法所构建传输模式下的网络能耗最大值,并验证了多跳网络能耗的最优解。实验结果表明,对比其他联合随机占空比网络编码算法和改进型自适应网络编码算法,NCDES能分别延长网络生命周期4.02%和8.51%,并分别提升包投递率14.83%和4.65%,从而有效提升数据包投递率和网络能量效率。 3、提出一种融合压缩感知和网络编码的无线传感器网络节能算法(CS-NCES)。该算法首先运用无线传感器网络数据的时间和空间相关性以及随机网络编码矩阵和压缩感知测量矩阵的相似性,在信源节点对数据进行编码,并应用有限域压缩感知对数据进行压缩,将压缩与编码融为一体,以实现数据的编码-压缩-再编码,从而使得中继节点传输数据量少于原数据量,随之降低网络能耗,而信宿节点通过译码-重构-再译码的方式来提升数据传递率。并且,本文通过对压缩感知和网络编码分别作用于实数域和有限域进行研究分析,构造了改进型多项式确定性矩阵,并对其可行性进行了验证。实验结果表明,CS-NCES对比其他网络编码算法,能有效降低网络能耗25.3%-34.5%,提升数据重构效率1.56%-5.98%,从而提升网络编码在无线传感器网络中的实用性和网络性能。
[Abstract]:Limited node energy has become the main bottleneck of wireless sensor network performance. Under the premise of ensuring the quality of information transmission, how to make full use of the limited network resources and reduce the energy consumption of network nodes.Prolonging the network life cycle has become a key issue in the research of wireless sensor networks (WSN).Network coding can effectively improve the data delivery rate, network throughput and network energy efficiency by the way of data coding and forwarding.Compressed perception projects the signal into sparse domain, which can restore the approximate value of the original information with a higher probability through partial data, thus reducing the amount of data transmitted in the network and saving the network resources.In order to reduce the energy consumption of wireless sensor network, the research work of this paper aims at reducing the redundant transmission data, combining with network coding and compression perception, the energy saving mode of sensor nodes in wireless sensor network is deeply studied.The main areas of work and innovation are as follows:1. A multipath energy saving algorithm for wireless sensor networks based on random network coding is proposed.The algorithm firstly considers the path selection probability of network nodes' residual energy and inter-node communication energy consumption, then introduces conditional transmission value ratio, and then constructs a feedback mechanism from the host node to the source node.Finally, a scheme of coding selection for relay nodes is proposed.Experimental results show that the proposed algorithm can effectively reduce the transmission delay and the number of coding nodes, thus achieving energy saving.Compared with other multipath and network coding algorithms, RNC-ESMP can effectively reduce the average energy consumption of the network by about 15 to 50, and reduce the delay of data transmission by about 12 to 33, thus improving the network performance.2. An energy saving algorithm for wireless sensor networks based on random network coding and duty cycle is proposed.The proposed algorithm firstly determines that the node is in the state of working or sleeping according to the ID information of the received data, so as to avoid the repeated receiving of the data, and increases the amount of data information transmitted under the same number of times by combining the random network coding.Thus, the network energy consumption can be saved.In addition, the maximum energy consumption in the transmission mode constructed by NCDES algorithm is analyzed by theoretical calculation, and the optimal solution of multi-hop network energy consumption is verified.The experimental results show that,Compared with other joint random duty cycle network coding algorithms and improved adaptive network coding algorithm, NCDES can prolong the network life cycle by 4.02% and 8.51%, and increase the packet delivery rate by 14.83% and 4.65%, respectively, thus effectively improving the packet delivery rate and network energy efficiency.3. An energy saving algorithm for wireless sensor networks based on compression sensing and network coding is proposed.The algorithm firstly uses the temporal and spatial correlation of wireless sensor network data and the similarity between random network coding matrix and compressed sensing measurement matrix to code the data at the source node.In order to realize the data encoding, compression and recoding, the relay nodes transmit less data than the original data, and then reduce the network energy consumption.The host node improves the data transfer rate by decoding-reconstructing-redecoding.Furthermore, through the research and analysis of compressed sensing and network coding acting on real and finite fields respectively, the improved polynomial deterministic matrix is constructed and its feasibility is verified.The experimental results show that compared with other network coding algorithms, CS-NCES can effectively reduce the network energy consumption by 25.3- 34.5and improve the efficiency of data reconfiguration 1.56-5.98. thus, the practicability and network performance of network coding in wireless sensor networks can be improved.
【学位授予单位】：广西大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5

【参考文献】