REAL TIME COMMUNICATION IN WIRELESS SENSOR NETWORKS BY ZILLE HUMA KAMAL.

REAL TIME COMMUNICATION IN WIRELESS SENSOR

NETWORKS

BY

ZILLE HUMA KAMAL

CS 691 - WMU - WSNR

WHAT IS A REAL TIME SYSTEM (RTS)

“A real time system is one in which the correctness of the computations not only depends on their logical correctness, but

also on the time at which the result is produced” [S]

CS 691 - WMU - WSNR

CLASSIFICATION OF RTS

• 2 Categories of RTS:– A Hard RTS is one in which one or more

activities must never miss a deadline or timing constraints, otherwise the system fails or results in catastrophe. [S]

– A Soft RTS is one that has timing constraints, but occasionally missing them has negligible effects, as application requirements as a whole continue to be met. [S]

TERM AND DEFINITIONS

• Task – executable entity• Job – instance of a task• Release Time – time at which task becomes ready to run

and job is released • Period – time between releases of two instances of the

same task• Deadline – relative time at which a job should complete

execution• Execution Time/ Run Time – time taken to complete

execution without interruption • Frame – discrete unit of time [CZSB]

CS 691 - WMU - WSNR

WIRELESS SENSOR NETWORKS

• CHARACHTERISTICS– An instance of MANET– Resource constraint – energy and storage

capacity– Limited range for communication and sensing– Frequent network topology changes– Individual entities are not critical, aggregation

of results is necessary for effectiveness and accuracy

CS 691 - WMU - WSNR

RTS IN WSN

• Two types of communication groups are inherently formed– Local Coordination – to aggregate results

– Sensor-Base Communication – to send results to base station

• This introduces contention on the communication channel, thus the main schedulable resource is the communication channel

CS 691 - WMU - WSNR

RAP

• A Real-Time communication architecture

Sensing/Control Application

Coordination Service

Query/Event Service

Location Addressed ProtocolGeographic Forwarding

Velocity Monotonic SchedulingPrioritize MAC

Query/Event Service APIs

RAP

CS 691 - WMU - WSNR

APIs

• Issue Query

- query name

- attribute list

- area

- timing constraints, e.g. period, deadline

- querier location

CS 691 - WMU - WSNR

APIs

• Event Registration

- event name

- area

- query

CS 691 - WMU - WSNR

Example

register_event{virus_found(0,0,100,100),query{

virus.count,area=(Xevent-1 ,Yevent-1,Xevent+1,Yevent+1),period=1.5, deadline=5,base=(100,100)}

};

CS 691 - WMU - WSNR

LAP

• Location Addressed Protocol- transport layer- connectionless- no IP/ID addressing, location based

addressing- three types of communication

» unicast » area multicast» area anycast

CS 691 - WMU - WSNR

LAP

• Unicast• Message is delivered to node closest to destination, e.g when

sensors send query results back to base station

• Area Multicast• Message is delivered to every node in a specified area, e.g

when base station sends query to an area, or for local coordination

• Area Anycast• Message is delivered to at least one node in the specified

area, e.g when base station wants to send a query to an area, the node which receives it can start the initiation process

CS 691 - WMU - WSNR

GF

• Greedy algorithmA packet is forwarded to a neighbor only if:(1) the neighbor node has the shortest distance to the

packet’s destination among all immediate neighbors AND

(2) the neighbor node is closer to the destination than the forwarding node

• If these conditions not satisfied, GPSR is used instead of GF

CS 691 - WMU - WSNR

VMS Deadline aware Distance aware

• Deadline aware

• Distance aware

• Packet scheduling policy– 2 types of packet scheduling policies

– Static Velocity Monotonic

– Dynamic Velocity Monotonic

CS 691 - WMU - WSNR

VMS

• SVM– Requested velocity is fixed at each hop

V = dis(x0, y0, xd, yd)/D

• DVM– Requested velocity changes at each hop and reflects the

time the packet has spent in the network

vi = dis(x0, y0, xd, yd)/(D-Ti)v0 = dis(x0, y0, xd, yd)/D

CS 691 - WMU - WSNR

Priority Queues

• various FIFO queues, one for each priority• Advantage – per packet overhead decreases,

ordering of each packet is not required

• Disadvantage – more storage capacity required

• single FIFO queue, with priority ordering • Advantage – reflects order of packets requested

• Disadvantage – greater number of packets lost

CS 691 - WMU - WSNR

MAC PRIORITIZATION

• Extensions to 802.11– Initial wait time after idle– Backoff Increase Function

• Initial wait time after idleDIFS = BASE_DIFS * PRIORITY

• Backoff Increase FunctionCW = CW * (2+(PRIORITY-1)/MAX_PRIORITY)

CS 691 - WMU - WSNR

EXPERIMENTATION

Overall deadline miss ratio of DSR and GF with deadlines (5,10)

CS 691 - WMU - WSNR

EXPERMENTATION

Overall deadline miss ratio

CS 691 - WMU - WSNR

EXPERIMENTATION

Miss ratio vs distance between source and destination (Deadline: (5:10) s; Rates: (0.8, 0.36)/s)

CS 691 - WMU - WSNR

REFERENCES

• [CZSB] M Caccamo, L.Y Zhang, L Sha, G Buttazzo, “An Implicit Access Protocol for Wireless Sensor Networks,”Proceedings of IEEE Real-Time Systems Symposium, Austin, TX , Dec 2002.

• [LBASH] C Lu, B.M Blum, T.F Abdelzaher, J.A Stankovic, T He, “RAP: A Real-Time Communication Architecture For Large-Scale Wireless Sensor Networks,” Department of Computer Science, University of Virginia

www.cs.virginia.edu/~stankovic/psfiles/rtas02-rap.pdf

CS 691 - WMU - WSNR

REFERENCES

• [P] T. F Piatkowski, “Citation and acknowledgment guide,” Department of Computer Science, Western Michigan University, Aug, 2000

www.cs.wmich.edu/~piat/citationAckGuide.pdf

• [S] D.B Stewart, “Introduction to Real Time,” Embedded.com, Nov 1, 2001.

www.embedded.com/story/OEG20011016S0120