An improved transport layer protocol for wireless sensor networks

https://doi.org/10.1016/j.comcom.2010.09.013Get rights and content

Abstract

Wireless sensor networks (WSNs) are a kind of communication networks having independent sensor nodes that form multi-hop ad hoc network to transfer data. In the past few years, various transport control protocols in wireless sensor networks (WSNs) have been developed and proposed in the literature. In this paper, we have analyzed pump slowly, fetch quickly (PSFQ) protocol and presented an improved transport layer protocol for wireless sensor networks. The improved protocol has been analyzed based on various factors such as average latency and average error tolerance and it is found that the proposed protocol is better than PSFQ in terms of these factors.

Introduction

Wireless sensor network (WSN) is a self organized distributed system consisting of one or more base stations, which are usually called sinks, and perhaps tens or thousands of sensor nodes sprinkled in a physical area. Integration of the information sensed through sensors, computations on that, and wireless communication make the sensor node able to gather physical data, process that data to obtain crude information, and reports to sinks. The sink can query the sensor nodes for more information. WSNs have several characteristic features [1]. Firstly, it has an extraordinary network topology. Sensor nodes are generally organized in a multihop star-tree topology. The base station (sink) is at the root of the tree and is responsible for data collection and relaying to external networks, may be wired or wireless networks. This topology varies according to time-varying link condition and node variation. Secondly, WSNs are resource constrained. Sensor nodes have limited resources, including low processing capability, small memory, low wireless communication bandwidth, and a limited, usually nonrechargeable battery. Thirdly, traffic characteristics of WSNs are varying. In WSNs, the primary traffic is in the upstream direction from the sensor nodes to the base station. This is a many-to-one type of communication. Upstream traffic delivery can be continuous, event-driven, query-driven or hybrid delivery. Type of delivery depends on the specific applications. For example, in temperature sensing, event-driven delivery is used while in target tracking continuous delivery should be used. This way, WSNs are used for diverse applications, each of which has different quality of service (QoS) and reliability requirements. Sink generates certain downstream traffic only for the purposes of query and control. Finally, WSNs use messages with very small message size to reduce their processing time. The need of segmentation is also eradicated in WSNs because of small message size.

The major focus in this paper is on the design of transport layer protocol for WSNs. Transport layer protocols are used to mitigate congestion and reduce packet loss, to provide fairness in bandwidth allocation, and to guarantee end-to-end reliability. However, UDP and TCP transport protocols, which are currently used for the Internet, cannot be directly implemented for WSNs. For example, UDP does not provide delivery reliability that is often needed for many sensor applications, nor does it offer flow and congestion control that can lead to packet loss and unnecessary energy consumption. TCP has also several drawbacks while used for WSNs [1]: TCP’s connection establishment is too heavier for event-driven applications, it does not provide fairness in terms of bandwidth allocation and discriminate against sensor node that is far away from the sink, it provides reliability by end-to-end retransmission which consumes more energy and bandwidth than hop-by-hop retransmission, its throughput will be degraded when there is a higher packet loss rate, congestion control in TCP is end-to-end which takes more time to alleviate the congestion as compared to hop-by-hop congestion control and it guarantees 100% packet reliability which is not necessary for event-driven applications.

The rest of this paper is organized as follows. An outline of different transport layer protocols for WSN is presented in Section 2. Analysis of PSFQ (pump slowly, fetch quickly) transport layer protocol for WSNs is provided in Section 3. In Section 4, weaknesses of PSFQ [7] protocol have been described. Section 5 provides solutions for those weaknesses for PSFQ. The performance analysis of both original PSFQ and improved PSFQ has been done through simulation and is presented in Section 6. Finally, Section 7 concludes the paper.

Section snippets

Related work

For WSNs, various transport layer protocols have been designed to address various issues such as congestion control, reliability or both. A brief summary of the related protocols is provided in this section.

STCP (sensor transmission control protocol) [2] is a generic end-to-end upstream transport protocol. It provides both the congestion control and reliability by allocating most of the responsibility at the sink. It provides controlled variable reliability utilizing the diversity in

Analysis of PSFQ protocol

In this section, we have presented the analysis of one of the transport control protocols known as pump slowly, fetch quickly (PSFQ) [7].

PSFQ is a hop-by-hop downstream transport protocol which is reliable, scalable and robust. It uses NACK-based loss detection and notification, and local retransmission for loss recovery and is basically designed for retasking/reprogramming application which requires 100% packet reliability in downstream direction. It consists of three operations: pump, fetch,

Shortcomings of PSFQ

We have already discussed the functionality of PSFQ [7] protocol in previous section. In this section, we have discussed some weaknesses of PSFQ.

Firstly, PSFQ does in-sequence forwarding of data packets. For example, if any intermediate node, running PSFQ as its transport layer protocol, does not have data packet ‘2’ and has all data packet from ‘3’ up to last data packet, then this node will not forward any of the data packet to the next sensor node. Node remains idle waiting for data packet

Improved transport protocol

In this section, we have provided the solutions for weaknesses described in previous section. We have also solved the problem of NACK implosion, which has arisen from our proposal. In the next section, we have done performance analysis of both PSFQ [7] protocol and the improved protocol.

Performance analysis

In this section, we have done performance analysis of original PSFQ [7] protocol and PSFQ protocol with our enhancements. We have used packet level simulation. We have analyzed some of the performance metrics of the proposed protocol.

Conclusion

We have initially presented various protocols available for transport layer in wireless sensor networks. After that, we have presented overview of PSFQ protocol. We provided some alterations to improve PSFQ’s efficiency. We have also shown performance analysis of original PSFQ and improved PSFQ. It has been shown that improved protocol works better than original PSFQ in terms of error tolerance and average latency. We have evaluated both PSFQ and improved PSFQ using simulation. We found that

Acknowledgements

This work has been carried out in wireless sensor network research facility in Computer Science and Engineering department of PEC University of Technology, Chandigarh, India. The lab has been funded from the inhouse grant of PEC University of Technology. We thank the Director Prof. Manoj Datta and the head of the department Prof. Sanjeev Sofat for his co-operation in setting up of this lab and giving us an opportunity to work there.

References (18)

  • Chonggang Wang, Kazem Sohraby, Bo Li, Mahmoud Daneshmand, Yueming Hu, A survey of transport protocols for wireless...
  • Y.G. Iyer, S. Gandham, S. Venkatesan, STCP: a generic transport layer protocol for wireless sensor networks, in:...
  • C.-Y. Wan, S.B. Eisenman, A.T. Campbell, CODA: congestion detection and avoidance in sensor networks, in: Proceedings...
  • C.-T. Ee, R. Bajcsy, Congestion control and fairness for many-to-one routing in sensor networks, in: Proceedings of ACM...
  • C.G. Wang, K. Sohraby, V. Lawrence, B. Li, Y.M. Hu, Priority-based congestion control in wireless sensor networks’, in:...
  • Y. Sankarasubramaniam, O.B. Akan, I.F. Akyilidiz, ESRT: event-to-sink reliable transport in wireless sensor networks,...
  • C.-Y. Wan, A.T. Campbell, L. Krishnamurthy, PSFQ: a reliable transport protocol for wireless sensor, in: Proceedings of...
  • Y. Zhou, M.R. Lyu, PORT: a price-oriented reliable transport protocol for wireless sensor network, in: Proceedings of...
  • H. Zhang, Anish Arora, Young-Ri Choi, Mohamed G. Gouda, Reliable bursty convergecast in wireless sensor networks,...
There are more references available in the full text version of this article.

Cited by (10)

  • A hybrid and dynamic reliable transport protocol for wireless sensor networks

    2015, Computers and Electrical Engineering
    Citation Excerpt :

    The fetch operation requests a retransmission of the lost packet from the neighbouring nodes. Lastly, the report operation provides a feedback status to the users [18]. It has four types of message formats: INJMSG, NAKMSG, REPMSG and REPAIR.

  • A hierarchical energy efficient reliable transport protocol for wireless sensor networks

    2014, Ain Shams Engineering Journal
    Citation Excerpt :

    Many reliable transport protocols [20] have been proposed for reliable data transmission in WSNs. These are Reliable Multi-Segment Transport (RMST) [21], Event to Sink Reliable Transport (ESRT) [22], Asymmetric Reliable Transport (ART) [23], Rate-controlled Reliable Transport protocol (RCRT) [24], Flush [25], Energy-efficient and Reliable Transport Protocol (ERTP) [26], Pump Slowly Fetch Quickly (PSFQ) [27], Improved PSFQ [28] and Data-Reliable Energy Efficient Transport Layer Protocol (DREET) [29] and Distributed Caching for Sensor Network (DTSN) [30,31]. These transport protocols are analysed on the basis of reliability and energy efficiency.

  • Energy efficient and reliable data delivery in urban sensing applications: A performance analysis

    2013, Computer Networks
    Citation Excerpt :

    Relevant applications include, among others, environmental monitoring, surveillance, event detection, intelligent agriculture, health monitoring. Traditionally, a WSN consists of static and resource-constrained sensor nodes, densely deployed in the sensing area, so that sensed data can be relayed through multi-hop communication and collected by a sink node [7–12]. More recently, there has been an increased interest in the research community to introduce mobile elements in WSNs, to increase their performance [13–15].

  • Resource aware reliable transport layer protocol for wireless sensor network

    2015, ICIIECS 2015 - 2015 IEEE International Conference on Innovations in Information, Embedded and Communication Systems
View all citing articles on Scopus
View full text