Data-Driven Data Transmission Mechanism for Wireless Sensor Networks in Harsh Communication Environment

Kenji Yoshigoe Department of Computer Science University of Arkansas at Little Rock

1

Outline • Introduction

• Problem • Solution • Evaluation • Conclusion • Future Works

2

Introduction Some WSN Applications •  Disaster relief operations •  Biodiversity mapping •  Precision agriculture •  Machine surveillance/maintenance •  Medicine and health care •  Logistics Challenge: Many applications have to survive in harsh communication environments 3

Introduction Efficient Resource Sharing of WSN • Node with multiple sensors (already available) • Remote programming/uploading (already available)

Various applications can coexist in a common WSN infrastructure

4

Introduction Challenge: Each application has unique QoS requirement 1. Time-sensitive – emergency, surveillance, reactive systems 2. Reliable data delivery – survivability-focused applications, disaster recovery 3. High bandwidth – applications requiring transfer of relatively high quality media

5

Problem Existing Solutions Redundant packet transfer can improve packet delivery and response time [12, 13] -) Constant duplicate transmission introduces significant power consumption, resulting fast depletion of the WSN. Differentiated service based on Priority Queues (instead of a single First-in-First-Out (FIFO)) to favor high priority packets on IEEE 802.15.4 Slotted CSMA/CA MAC [14].

6

Problem Existing Solutions Adjusting IEEE 802.15.4 MAC parameters: the maximum backoff window, macMinBE, the minimum backoff window, macMaxBE, and the maximum number of backoff stages, macMaxCSMABackoffs [14 -16].

Dynamic MAC exponential backoff • Existing MAC layer time out period is based on a fixed exponential backoff for all frames. • Assignment of less (more) aggressive backoff to critical (non-critical) frames. 7

Problem (Quick background for existing Exponential Backoff:) •  Goal: adapt retransmission attempts to estimated current load –  heavy load: random wait will be longer •  first collision: choose K from {0,1}; delay is K· 512 bit transmission times •  after second collision: choose K from {0,1,2,3}… •  after ten collisions, choose K from {0,1,2,3,4,…, 1023}

8

Problem

All existing solutions address the issue of contention to improve packet delivery rate and energy efficiency, but none directly improve the data survivability and the energy efficiency in a harsh communication environment.

9

Solution 1. Application-level Dynamic Transmission •  Adjust transmission frequency based-on associated message state

//At each node at every sampling cycle While (True) if (CHECK_STATE (state(current_sample)) != prev_state) update_transmission_frequecy (current_freq); // At transmitter with current_freq While (True) TRANSMIT (current_sample);

10

Solution 2. Link-level content-aware routing •  On-demand packet-level prioritization a) Dynamic MAC exponential backoff [14-16] b) Dynamic MAC retransmission count •  No long-term resource reservation needed •  Receiver unaware of packet priority can still receive the packet with the highest priority.

11

Solution 2. b) Dynamic MAC retransmission count • Existing MAC retransmission protocol has predefined # of maximum retransmission count for all frames: macMaxFrameRetries • Dynamic assignment of large (small) macMaxFrameRetries at each node for critical (noncritical) frames If frequent change in macMaxFrameRetries is not desired, a fixed macMaxFrameRetries value along with flexible initial assignment of frame retransmission counter value, FrameRetries, is an alternative solution (e.g., FrameRetries = macMaxFrameRetries means no 12 retransmission is allowed.).

Solution Assumption - Targeted applications require remote device to constantly inform the states of the observed environment/system to a sink or any other part of the deployed network. - Sensor’s sampling frequency is independent of transmission frequency. In case the transmission frequency is greater than the sampling frequency, duplicated data may be transmitted.

13

Evaluation Goals: 1) Effect of transmission frequency on packet delivery rate 2) Effect of multi-hop routing on packet delivery rate 3) Effect of retransmission on energy consumption Assumption - Binary Symmetric Channel (BSC): Each transmitted bit has a certain fixed error probability independent of previous bit error occurrences, and the error probability does not depend on the symbol value.

14

Evaluation PH-PLP – Per hop packet loss probability n – number of transmission of the same packet m – number of hops in a routing path

1) Pr (Successful packet delivery) = 1 - PH-PLP 2) Pr (n consecutive packet loss) = PH-PLP n 3) Pr (At least one packet delivery) = 1 - PH-PLP n 4) Pr (m-hop packet delivery) = (1 - PH-PLP n)m

15

Evaluation Effect of increased transmission frequency on successful packet delivery rate (analysis)

16

Evaluation Effect of increased transmission frequency on successful packet delivery rate (analysis (---) & simulation ( ))

17

Evaluation Effect of multi-hop routing on packet delivery rate (analysis)

18

Evaluation Effect of retransmission on energy consumption (simulation)

19

Conclusions •  Promising methods to route critical data in cogested/unreliable network • Improved data delivery • Reduced energy consumption

20

Future (Current) Works • Adaptive Handshaking - Disable RTS/CTS handshake in harsh communication environment

- Rationale: With high packet loss, handshake frames further consume unnecessary time and bandwidth If delay is due to congestion, sender at least receives CTS frame. • Time sensitivity analysis

21

References (# associated with reference # in the paper) [12] S. Mishra, J. Deng and, R. Han. Countermeasures Against Traffic Analysis Attacks in Wireless Sensor Networks. Technical Report CU-CS-987-04, University of Colorado at Boulder, December 2004. [13] W. K. G. Seah and H.P. Tan, “Multipath Virtual Sink Architecture for Wireless Sensor Networks in Harsh Environments”, Proceedings of 1st International Conference on Integrated Internet, Ad Hoc and Sensor Networks (InterSense), Vol. 138, May, 2006. [14] Koubaa, A., Alves, M., Nefzi, B., & Song, Y.-Q. (2006). Improving the IEEE 802.15.4 Slotted CSMA/CA MAC for time-critical events in wireless sensor networks. Proceedings Workshop of Real-Time Networks (RTN 2006), Satellite Workshop to (ECRTS 2006). [15] D. Rohm, M. Goyal, H. Hosseini, A. Divjak, and Y. Bashir, "Configuring Beaconless IEEE 802.15.4 Networks Under Different Traffic Loads," Proceedings of International Conference on Advanced Information Networking and Applications, pp.921-928, 2009. [16] M. Di Francesco, G. Anastas, M. Cont, S. K. Das, and V. Neri, “An Adaptive Algorithm for Dynamic Tuning of MAC Parameters in IEEE 802.15.4/ZigBee Sensor Networks,” Sixth IEEE International Workshop on Sensor Networks 22 and Systems for Pervasive Computing, pp. 400 – 405, 2010.