Next Century Challenges: Mobile Networking for “Smart Dust”

1. Next Century Challenges:Mobile Networking for“Smart Dust”

2. Abstract: • Advances in hardware technology has enabled very compact ,autonomous and mobile nodes, each having one or more sensors, computation and communication capabilities, and a power supply. • There is a need for networking all these devices.

3. Main aim of the paper: • To review the key elements of “Smart Dust”. • To outline the research challenges they present to the mobile networking and systems community. • The paper also explores the limit on size and power consumption of sensor nodes.

4. Introduction: • Networking of wireless sensors has become possible due to the following three technologies: 1. Digital Circuitry 2. Wireless Communication 3. Micro ElectroMechanical Systems (MEMS) In all these areas the main focus has been on reduction in size, power consumption and cost.

5. What is the Smart Dust project: • Being developed at UC Berkeley by Prof. Pister and Prof Kahn. • It explores the limits on size and power consumption in autonomous sensor nodes. • Size reduction is important in order to make the nodes as inexpensive as possible and also easy to deploy. • The researchers believe that these nodes will be of the order of a few cubic millimeters and are called “Smart Dust”.

6. Smart Dust Technology: A Smart Dust mote consists of: • MEMS sensors. • A semi-conductor laser diode and MEMS beam- steering mirror for active optical transmission. • A MEMS corner-cube retroreflector (CCR) for passive optical transmission. • An Optical receiver. • Signal Processing and Control Circuitry. • A power source based on thick-film batteries and solar cells.

7. Major Challenges: • To incorporate all these functions while maintaining a low power consumption. • Maximizing operating life given the limited volume of energy storage. • The functionality can be achieved only if the total power consumption of a dust mode is limited to microwatt levels, and if careful power management strategies are utilized. A Smart Dust mote is shown in next figure.

9. Developing a communications architecture for ultra low-power is a more critical challenge. There are primarily two communication technologies: • Radio Frequency (RF) Technique • Optical Transmission Technique Each technique has certain advantages and disadvantages.

10. RF Technique: • Presents problem because dust motes offer very limited space for antennas, thereby demanding extremely short-wavelength( or high frequency) transmission. • Communication not compatible with low power transmission. • The radio receivers are complex circuits making it difficult to reduce their power consumption to the required microwatt level.

11. Require: • Modulation • Bandpass filtering • Demodulation circuitry • Additional circuitry is required if the transmissions of a large number of motes are to be multiplexed using time, frequency or code division multiple access.

12. Optical Transmission Technique: • Kahn and Pistor showed that when a line-of sight path is available, well designed free-space optical links require significantly lower energy per bit than their RF counterparts. • There are several reasons for the power advantage of the optical links: • Optical transceivers require only simple baseband analog and digital circuitry. • They don’t require modulators, active bandpass filters or demodulators.

13. 3. The short wavelength of the visible or near infrared light makes it possible for a millimeter scale device to emit a narrow beam. 4. A base station transceiver (BTS) equipped with a compact imaging receiver can decode the simultaneous transmissions from a large number of dust motes at different locations within the receiver field of view.

14. There are two requirements for successful decoding of these simultaneous transmissions: • Dust motes should not block one another’s line of sight to the BTS. Such blockage is unlikely due to the small size of dust motes. • Images of different dust motes be formed on different pixels in the BTS imaging receiver. • Another advantage is that the dust motes use passive optical transmission techniques i.e to transmit modulated optical signals without supplying any optical power. This structure is a corner-cube retroreflector or CCR.

15. The CCR is shown here:

16. CCR • Consists of three mutually perpendicular mirrors of gold coated polysilicon. • Has the property that any incident ray of light is reflected back to the source ( provided it is incident with a certain range of angles centered about the cube’s body diagonal). • CCR based passive optical links require uninterrupted line-of-the sight path. • A CCR can transmit to the BTS only when the CCR body diagonal happens to point directly toward the BTS.

17. Following figure shows a free-space optical network utilizing the CCR based passive link.

19. The BTS contains a laser whose beam illuminates an area containing dust motes. • This beam can be modulated with downlink data. • When the illuminating beam is not modulated the dust motes can use their CCRs to transmit uplink data back to the base station. • A high frame rate CCD video camera at the BTS decodes these CCR signals to yield the uplink data. • It uses space division multiplexing.

20. When the application requires dust motes to use active optical transmitters, MEMS technology can be used to assemble a semiconductor laser, a collimating lens and a beam steering micro-mirror. • Active transmitters allows for a peer-to peer communication between dust motes, provided there is a line of sight path between them. • Trade off between bandwidth and range due to power consumption. • The dust motes can communicate over longer ranges at low data rates or higher bit rates over short distances.

21. The active transmitters should be used for short- duration burst-mode communication only.

22. Mobile Networking Challenges There are some limitations in regard to the development of mobile networking protocols: • The free-space optical links requires uninterrupted line-of-sight paths. • The passive and active dust mote transmitters have directional characteristics that must be considered in system design. • There is a huge trade-offs between bit rate, energy per bit, distance and directionality in these energy-limited free-space optical links.

23. Line-of-sight requirement • An unbroken line-of sight path is required for operation of free-space optical links for Smart Dust. • A fixed dust mote without a line-of-sight path to the BTS can communicate with the BTS via multihop routing provided a multihop path exists. • Multihop routing requires dust motes to be equipped with active optical transmitters. • When the dust motes are not fixed a line-of-sight path to the BTS may become available in short periods.

24. In such cases BTS continuously senses the dust motes. • When a line-of-sight path to a mote becomes available, the mote can transmit a packet to the BTS.

25. Link Directionality • The dust mote’s transmitter has different directional characteristics than its receiver. Passive dust mote transmitter • Based on the CCR. • If dust motes use only one CCR each, then any given dust mote, if fixed in a random, upright orientation, has only about a 10% probability of being able to transmit to the BTS. • This probability can be increased significantly by equipping each dust mote with several CCRs each oriented along different direction.

26. Another solution is to distribute randomly an excess number of dust motes with the aim of communicating only with those whose CCRs point toward the BTS. • If the dust motes are not fixed it should delay transmitting until it reach an orientation that enables transmitting to the BTS.

27. Active dust mote transmitter • Based on a laser diode. • Should employ a narrow bandwidth of the order of few degrees or less. • Dust motes should be equipped with beam steering mechanism. • Current research challenge is to develop beam steering algorithms for systems with active dust mote transmitters.

28. Following points are considered in developing beam-steering algorithms:- • Each dust mote should autonomously steer its beam towards the desired direction. • One approach is to make the dust mote receiver directional, and to mount the receiver and transmitter on the same aiming mechanism. • Accordingly, by aiming its receiver so as to maximize the signal received from the BTS or another mote, the dust mote would be aiming its transmitter at that node. • If nodes remain fixed then the direction of the nodes once determined can be stored in the dust mode for future use.

29. In all of the above cases the dust motes transmitter and receiver have different spreads. This leads to non reciprocal link characteristics. What is non-reciprocity?? • A situation in which a dust mote may receive from another node but be unable to transmit it or vice-versa. • May lead to “hidden nodes” which can cause collisions during medium access in free-space optical networks.

30. Collision during active peer-to-peer communication is a potential problem in Smart Dust Networks. • A peer-to-peer collision avoidance scheme must cope with a dynamic network configuration and should not introduce excessive complexity or latency.

31. Trade-offs Between Bit rate, Distance and Energy per Bit • The receiver signal to noise ratio (SNR) is given by

32. Where, • C is constant • Eb average transmitted energy per bit. • Rb is the bit rate • A is the receiver light collection area • N0 is the receiver noise power spectral density • d is angular spread

33. The expression assumes the following:- • phi is small. • transmitter beam is well aimed at the receiver. • SNR must be maintained at a suitably high value to insure reliable link operation. • Eb is proportional to Rb-1/2 i. e. the energy per bit is minimized if packets are transmitted in short bursts at a high bit rate

34. The average transmitted power is Pt = Eb/Rb • Therefore transmission at a high bit rate requires high-power transmitter. • Pt should be chosen as high as possible. The equation can be written in terms of Pt as follows

35. Given a limit on Pt in order to maximize the bit rate Rb and distance d, the receiver area A should be maximized and phi should be minimized. • Since Rb is directly proportional to d-4, it is possible to extend the link distance by drastically lowering the bit rate.

36. Applications • Introduction • Individual dust motes may be attached to the objects that one wishes to monitor or a large number of dust motes may be may dispersed in the environment randomly. • Numerous civil and military applications for Smart Dust. • Smart dust may be deployed over a region to record data for meteorological, geophysical or planetary research. • May be employed at places where wired sensors are unusable or may lead to error

37. eg. Instrumentation of semiconductor processing chambers, wind tunnels, rotating machinery etc. • Smart dust may be used in biological research for eg., to monitor the movements and internal process of insects or other small animals. • Can be used for spying purposes. • To detect the presence of chemical or biological agents on a battlefield.

38. Scenario: Multi-Sensor Emergent Behavior • Sensors should operate in ensembles. • They are typically specialized to detect certain signatures like one may detect motion, another heat and another sound. • When one sensor detects its critical event signature it should make other nearby sensors aware of it, which then act accordingly. • The main optimization challenge is to maximize detection probability and resolution and minimizing power consumption.

39. Technology Approaches for Realizing the Scenario: There are two ways: • Centralized scheme: Motion Sensor BTS nearby heat sensor For passive communication this is the most power efficient technique. 2. Multihop Technique: • The centralized/passive scheme cannot be used if line-of-sight path is blocked or probe revisit rate is too infrequent. • In this case active transmitter must be used.

40. If line of sight path is blocked then mote will need to use ad hoc, multihop techniques to communicate with the BTS or nearby sensor nodes. Detection of blocked path between mote and BTS is easy:- • Assume maximum duty cycle between probe visit. • If sufficient time has passed since last visit, mote can assume that it is blocked.

41. The main challenge is building a multihop route in this environment. • Active transmission in all direction is not feasible. • A possible scheme can be: • A node transmits for short burst. • Wait for an ACK responses from any listening node to determine that its transmission has been received. • Full four phase handshake. • Routing tables can be constructed from such connectivity.

42. 3. Standard routing algorithm may not work as they assume bi-directional transmission, which may not always be the case for smart dust. 4. New routing algorithms may be developed to deal with the general case of links that are unidirectional and/or asymmetric in their performance.

43. Related Projects • Factoid Project at Compaq Palo Alto Western Research laboratory. • WINS Project at UCLA. • Ultralow Power Wireless Sensor Project at MIT.

44. Summary and Conclusion • Wireless sensor networks is an area that has very important application and provides challenges in networking field. • Smart Dust provide an inexpensive way to probe a sensor. • Active optical transmission is possible but consumes more power. It will be used when passive technique cannot be used for eg., blocked line-of-sight path.