MMSN MultiFrequency Media Access Control for Wireless Sensor

MMSN: Multi-Frequency Media Access Control for Wireless Sensor Networks Cheoleun Moon Computer Science Div. at KAIST /

Contents o o o Motivation Overhead Analysis New Protocol Framework n n o o Frequency Assignment Media Access Design Performance Evaluation Conclusions 2

Ad-hoc Wireless Sensor Networks n Self-organize n n Sensors & Actuators Limited CPU and memorys Limited radio bandwidth 3

Motivation Limited single-channel bandwidth in WSN o n 19. 2 kbps in MICA 2, 250 kbps in MICAz/Telos The bandwidth requirement is increasing o n Support audio/video streams (assisted living, …) Multi-channel design needed Hardware appearing Multi-channel support in MICAz/Telos n More frequencies available in the future n Software still lags behind Collision-based: B-MAC n Scheduling-based: TRAMA n Hybrid: Z-MAC n 4

Multi-Channel MAC in MANET o Require more powerful hardware/multiple transceivers n o Listen to multiple channels simultaneously Frequent Use of RTS/CTS Controls n n For frequency negotiation Due to using 802. 11 5

Basic Problems for WSN o Don’t use multiple transceivers n n o Packet Size n o Energy Cost 30 bytes versus 512 bytes in MANET RTS/CTS n Costly overhead 6

RTS/CTS Overhead Analysis o RTS/CTS are too heavyweight for WSN: n n n Mainly due to small packet size: 30~50 bytes in WSN vs. 512+ bytes in MANET From 802. 11: RTS-CTS-DATA-ACK From frequency negotiation: case study with MMAC o MMAC n n RTS/CTS frequency negotiation 802. 11 for data communication 7

Contributions o First multi-frequency MAC, specially designed for WSN o Developed four frequency assignment schemes n o Supports various tradeoffs New toggle transmission and toggle snooping for media access control 8

Frequency Assignment F 8 Reception Frequency F 7 F 6 F 5 F 1 F 4 Complications - Not enough frequencies - Broadcast F 2 F 3 9

Frequency Assignment Schemes When #frequencies >= #nodes within two hops Exclusive Frequency Assignment o o o When #frequencies < #nodes within two hops Implicit-Consensus Both guarantee that nodes within two hops get different frequencies The left scheme needs smaller #frequencies The right one has less communication overhead Even Selection o o o Eavesdropping Balance available frequencies within two hops The left scheme has fewer potential conflicts The right one has less communication overhead 10

Media Access Design (1/4) o o o Different frequencies for unicast reception The same frequency for broadcast reception Time is divided into slots, each of which consists of a broadcast contention period and a transmission period Tbc Ttran …. . . Tb c Ttran 11

Media Access Design (2/4) o Case 1 n When a node has no packet to transmit 12

Media Access Design (3/4) o Case 2 n When a node has a broadcast packet to transmit 13

Media Access Design (4/4) o Case 3 n When a node has a unicast packet to transmit 14

Toggle Snooping o During “back off (fself, fdest)”, toggle snooping is used 15

Toggle Transmission o When a node has unicast packet to send transmits a preamble n n fself so that no node sends to me fdest so that no node sends to destination 16

Simulation Configuration Components Setting Simulator Glo. Mo. Sim Terrain (200 m X 200 m) Square Node Number 289 (17 x 17) Node Placement Uniform Payload Size 32 Bytes Application Many-to-Many/Gossip CBR Streams Routing Layer GF MAC Layer CSMA/MMSN Radio Layer RADIO-ACCNOISE Radio Bandwidth 250 Kbps Radio Range 20 m~45 m Confidence Intervals The 90% confidence intervals are shown in each figure 17

Performance Metrics o Aggregate MAC throughput n o Packet delivery ratio n o (Total # of data packets delivered by MAC layer) (Total # of data packet the network layer requests MAC) Channel access delay n o Total amount of data successfully delivered in MAC per unit time Delay data packet from the network layer waits for the channel Energy consumption 18

Performance with Different #Physical Frequencies – With Light Load ① ② ③ ④ Performance when delivery ratio > 93% Scalable performance improvement Overhead observed when #frequency is small More scalable performance with Gossip than many-to-many traffic 19

Performance with Different # Physical Frequencies – With Higher Load ① When load is heavy, CSMA has 77% delivery ratio, while MMSN performs much better ② MMSN needs less channels to beat CSMA, when the load is heavier 20

Performance with Different System Load Observation: CSMA has a sharp decrease of packet delivery ratio, while MMSN does not. Reason: The non-uniform backoff in time-slotted MMSN is tolerant to system load variation, while the uniform backoff in CSMA is not. 21

Conclusions o First multi-frequency MAC, specially designed for WSN, where singletransceiver devices are used n n n Explore tradeoffs in frequency assignment Design toggle transmission and toggle snooping MMSN demonstrated scalable performance in simulation 22