Ad hoc and Sensor Networks Chapter 2: Single node architecture

Transcription

1 Ad hoc and Sensor Networks Chapter 2: Single node architecture

2 Goals of this chapter Survey the main components of the composition of a node for a wireless sensor network Controller, radio modem, sensors, batteries Understand energy consumption aspects for these components Putting into perspective different operational modes and what different energy/power consumption means for protocol design Operating system support for sensor nodes Some example nodes Note: The details of this chapter are quite specific to WSN; energy consumption principles carry over to MANET as well

3 Outline Sensor node architecture Energy supply and consumption Runtime environments for sensor nodes Case study: TinyOS

4 Node Characteristics Wide variety of node characteristics: Volume <1 ccm; weight <100 g; cost < 1 $; power dissipation < 100 µw Sensor Nodes for environment observation Volume < 1 cmm; weight < 1 g; cost < 10 C; power dissipation ~ 0 W Intelligent grains; smart dust

5 Sensor node architecture (minimum configuration) Main components of a WSN node Controller Communication device(s) Sensors/actuators Memory Power supply Memory Communication device Controller Sensor(s)/ actuator(s) Power supply

6 Sensor node architecture Additional components of a WSN node Watchdog Sensor controlled wake-up Time controlled wake-up Communication-controlled wake-up Power Manager Shut-down /reactivate power Memory Communication device Controller Sensor(s)/ actuator(s) Power manager Watchdog devices Power supply

7 Ad hoc node architecture Core: essentially the same But: Much more additional equipment Hard disk, display, keyboard, voice interface, camera, Essentially: a laptop-class device

8 Controller Main options: Microcontroller general purpose processor, optimized for embedded applications, low power consumption Often the general purpose nature leads to inefficiencies wrt. Power consumption DSPs optimized for signal processing tasks, not suitable here In special cases part of the processing can be performed by embedded DSPs or specialized ALU instruction set (e.g. FFT) (IHP proposed a project called Tandem with 2 processors per node) FPGAs may be good for testing ASICs only when peak performance is needed, no flexibility The convergence between an ASIC and a specialized microcontroller for sensor node applications is sliding Example microcontrollers Texas Instruments MSP bit RISC core, up to 4 MHz, versions with 2-10 kbytes RAM, several DACs, RT clock, prices start at 0.49 US$ Atmel ATMega 8-bit controller, larger memory than MSP430, slower

9 MSP430

10 ATMEL-Dual Core

11 BAN: Node Architecture Serial I- SPRAM DMA IRQ UART EJTAG CPU EC Bridge AMBA AHB Bridge APB GPIO GPIO D- SPRAM Protocol Processor Memory Controller AES Flash SRAM Transceiver iram Antenna

12 Memory Different types of memory possible: RAM: to store data and interim results Registers for fast access to results in computations NVR (Non Volatile Ram) to store results also during power down Dynamic RAM to save space in embedded applications ROM: to store programs PROM: For programs and fixed configuration data Flash: For programs and data EEPROM: For programs and fixed configuration data EPROM: see above Research today: Have an All in One approach based on NVR with high speed read time and fast write time. Increase the number of programming cycles to last for the total lifetime of the node Reduce power consumption by changing the power supply voltage dynamically Store all state carrying information in NVM before power down

13 Memory Structure

14 SRAM-Cell

15 DRAM Cell

16

17 Communication device Which transmission medium? Electromagnetic at radio frequencies? Preferred in most conditions (see later slides) Electromagnetic, light? Always Line of Sight required Techniques like diffuse transmission possible under certain conditions Spot diffusion for high data rate with a reflective common exchange spot Ultrasound? Can be used in conditions where radio does not work properly (e.g. under water) Only very long waves propagate under water Ultrasonic has much lower losses Quite good for exact location determination (cm range possible) Radio transceivers transmit a bit- or byte stream as radio wave Receive it, convert it back into bit-/byte stream

18 Signal propagation ranges Transmission range communication possible low error rate Detection range detection of the signal possible no communication possible Interference range signal may not be detected signal adds to the background noise sender transmission detection interference distance

19 Signal propagation Propagation in free space always like light (straight line) Receiving power proportional to 1/d² (d = distance between sender and receiver) Receiving power additionally influenced by fading (frequency dependent; H 2 O resonance at 2.5 GHz; O 2 Resonance at 60 GHz) shadowing reflection at large obstacles refraction depending on the density of a medium scattering at small obstacles diffraction at edges shadowing reflection refraction scattering diffraction

20 Real world example

21 Multipath propagation Signal can take many different paths between sender and receiver due to reflection, scattering, diffraction LOS pulses multipath pulses signal at sender signal at receiver Time dispersion: signal is dispersed over time (delay spread) interference with neighbor symbols, Inter Symbol Interference (ISI) The signal reaches a receiver directly and phase shifted distorted signal depending on the phases of the different parts

22 Transceiver Building Blocks LNA Antenna Interface LO Receive/Transmit resonator PA Amplification stage Down/Up conversion stage

23 AFE Components Antenna: Gain and directivity; Bandwidth; Quality Factor Amplification stage PA (Power Amplifier) Linearity; Power Control; Efficiency LNA (Low Noise Amplifier) Noise Suppression; Bandwidth; Dynamic Range Down/Up conversion stage Local Oscillator Stability; Noise; Power consumption; Stabilization time Mixer Gain; Quality; Intermixing; Lo-suppression; Noise

24 Modulation and demodulation analog baseband digital signal data digital analog modulation modulation radio transmitter radio carrier analog demodulation analog baseband signal synchronization decision digital data radio receiver radio carrier

25 Digital Components Ad/DA Conversion Dynamic Range; Efficiency Modulation/Demodulation Modulation Form Single Carrier/ Multi Carrier Channel Coding Hamming Distance; Code selection; Interleaving Spreading Narrow band distortion suppression Synchronization Synchronization time for locking Synchronization Framing; Synchronization Efficiency

26 Typical simple FSK-Transceiver

27 Analog Modulation: 5.8 GHz Transceiver

28 5 GHz Transceiver Blocks

29 Transceiver characteristics Capabilities Interface: bit, byte, packet level? Supported frequency range? Typically, somewhere in 433 MHz 2.4 GHz, ISM band Upcoming UWB Multiple channels? Data rates? Range? Energy characteristics Power consumption to send/receive data? Time and energy consumption to change between different states? Transmission power control? Power efficiency (which percentage of consumed power is radiated?) Radio performance Modulation? (ASK, FSK,?) Noise figure? NF = SNR I /SNR O Gain? (signal amplification) Receiver sensitivity? (minimum S to achieve a given E b /N 0 ) Blocking performance (achieved BER in presence of frequency-offset interferer) Out of band emissions Carrier sensing & RSSI characteristics Frequency stability (e.g., towards temperature changes) Voltage range

30 Transceiver states Transceivers can be put into different operational states, typically: Transmit Receive Idle ready to receive, but not doing so Some functions in hardware can be switched off, reducing energy consumption a little Leakage is a main source of power dissipation Sleep significant parts of the transceiver are switched off Not able to immediately receive something Recovery time and startup energy to leave sleep state can be significant Research issue: Wakeup receivers can be woken via radio when in sleep state (seeming contradiction!)

31 Simple Wakeup-Device Antenna Interface Wakeup-1 Energy Detector Correlator Wakeup-2 Transceiver Wakeup-1: Simple energy detector after bandpath filter in antenna and antenna interface Wakeup-2: Single correlation to detect characteristics of the sensor network can be simple pattern matching can be complicated pattern analysis including address recognition

32 Example radio transceivers Almost boundless variety available Some examples RFM TR1000 family 916 or 868 MHz 400 khz bandwidth Up to kbps On/off keying or ASK Dynamically tuneable output power Maximum power about 1.4 mw Low power consumption Chipcon CC1000 Range 300 to 1000 MHz, programmable in 250 Hz steps FSK modulation Provides RSSI Chipcon CC 2400 Implements GHz, DSSS modem 250 kbps Higher power consumption than above transceivers Infineon TDA 525x family E.g., 5250: 868 MHz ASK or FSK modulation RSSI, highly efficient power amplifier Intelligent power down, self-polling mechanism Excellent blocking performance

33 Example radio transceivers for ad hoc networks Ad hoc networks: Usually, higher data rates are required Typical: IEEE b/g/a is considered Up to 54 MBit/s Relatively long distance (100s of meters possible, typical 10 s of meters at higher data rates) Works reasonably well (but certainly not perfect) in mobile environments Problem: expensive equipment, quite power hungry

34 Wakeup receivers Major energy problem: RECEIVING Idling and being ready to receive consumes considerable amounts of power When to switch on a receiver is not clear Contention-based MAC protocols: Receiver is always on TDMA-based MAC protocols: Synchronization overhead, inflexible Desirable: Receiver that can (only) check for incoming messages When signal detected, wake up main receiver for actual reception Ideally: Wakeup receiver can already process simple addresses Not clear whether they can be actually built, however

35 Ultra-wideband communication Standard radio transceivers: Modulate a signal onto a carrier wave Requires relatively small amount of bandwidth Alternative approach: Use a large bandwidth, do not modulate, simply emit a burst of power Forms almost rectangular pulses Pulses are very short Information is encoded in the presence/absence of pulses Requires tight time synchronization of receiver Relatively short range (typically) Advantages Pretty resilient to multi-path propagation Very good ranging capabilities Good wall penetration

36 UWB Qualities 1000 UWB Short Distance Fast download 2m 4m UWB is wireless personal area networking (WPAN) technology for transmitting data Quickly Cost-effectively With low power consumption Data Rate (Mbps) UWB Room-range High-definition Quality of service, streaming 10m a/b/g/n Data Networking.11n promises 100m 1 Bluetooth Range (m) Source: Texas Instruments

37 Comparison of Shannon Channel Capacity Theoretical Capacity depends on: z Bandwidth B Signal to Noise Ratio SNR signal power distance ambient noise RX noise figure distance R in m Ultra Wideband short range and high data rate communications

38 What is Ultra Wideband? Radio technology that modulates impulse based waveforms instead of continuous carrier waves Time-domain behavior Frequency-domain behavior Ultrawideband Communication Impulse Modulation time 3 frequency 10 GHz (FCC Min=500Mhz) Narrowband Communication Frequency Modulation GHz

39 Ultra-Wideband (UWB) at a Glance Regulated in the US since February 2002 UWB is available spectrum, not a specific technology 7,500MHz of unlicensed spectrum First regulation ever that allows spectrum sharing: low emission limit (-41.3dBm/MHz EIRP) doesn t cause harmful interference Transmitters need to occupy at least 500MHz all the time UWB devices are NOT defined as impulse radios or by any specific modulation Enough spectrum to reach much higher data rates than in the ISM band (83.5MHz at 2.4GHz) or the U-NII bands (300MHz at 5GHz) Optimized for short-distances applications

40 3.1 to 5.1 GHz UWB Transceiver Reg Rx SR PLL ~ N:1 PN Tx

41 Relevant Standards IEEE : Bluetooth Most probably not very well suited for sensor network applications IEEE : PAN Supports high data rates up to 22 Mb/s MAC protocol suited mainly for data transmission Might be taken for sensor applications but has to be amended by extra energy save states IEEE (ZigBee): Specially developed for sensor network applications Low data rates Multi-hop networks supported a: Special Phy layer for UWB, Chirp, etc. for robustness and location services b: PSSS for high bandwidth efficiency

42

43 Sensors as such Main categories Any energy radiated? Passive vs. active sensors Has measurement method impact on the measured object? Sense of direction? Omnidirectional? Passive, omnidirectional Examples: light, thermometer, microphones, hygrometer, Passive, narrow-beam Example: Camera Active sensors (without impact on object) Example: Radar Active sensor (with impact) Chemical sensors that analyze some molecules Important parameter: Area of coverage Which region is adequately covered by a given sensor?

44 Actuators From the point of view of the sensor node actuators are seen via I/O operations. These output operations may: Output digital signals (on/off, digital value etc.) Output analog values Receive response back from actuator Good design practice: Never trust a actuator -> always pair actuator/sensor such that the actuator action can be independently observed and reported

45 Outline Sensor node architecture Energy supply and consumption Runtime environments for sensor nodes Case study: TinyOS

46 Energy supply of mobile/sensor nodes Goal: provide as much energy as possible at smallest cost/volume/weight/recharge time/longevity In WSN, recharging may or may not be an option Options Primary batteries not rechargeable Secondary batteries rechargeable, only makes sense in combination with some form of energy harvesting (or recharge systems) Requirements include Low self-discharge Long shelf live (similar to above) Capacity under load (the method of loading can make a big difference on the accessible capacity Efficient recharging at low current Good relaxation properties (seeming self-recharging) Voltage stability (to avoid DC-DC conversion)

47 Battery examples Energy per volume (Joule per cubic centimeter): Primary batteries Chemistry Zinc-air Lithium Alkaline Energy (J/cm 3 ) Secondary batteries Chemistry Lithium NiMHd NiCd Energy (J/cm 3 )

48 How to recharge a battery? Energy scavenging A laptop: easy, plug into wall socket in the evening A sensor node? Try to scavenge energy from environment Ambient energy sources Light! solar cells between 10 µw/cm 2 and 15 mw/cm 2 Temperature gradients 80 µ W/cm 1 V from 5K difference (Seebeck-Effect) Vibrations between 0.1 and µ W/cm 3 Pressure variation (piezo-electric) 330 µ W/cm 2 from the heel of a shoe Air/liquid flow (MEMS gas turbines)

49 Energy scavenging overview

50 Energy consumption A back of the envelope estimation Number of instructions Energy per instruction: 1 nj Small battery ( smart dust ): 1 J = 1 Ws Corresponds: 10 9 instructions! Lifetime Or: Require a single day operational lifetime = 24*60*60 = s 1 Ws / 86400s ~ 11.5 µw as max. sustained power consumption! Not feasible!

51 Multiple power consumption modes Way out: Do not run sensor node at full operation all the time If nothing to do, switch to power safe mode Question: When to throttle down? How to wake up again? Typical modes Controller: Active, idle, sleep Radio mode: Turn on/off transmitter/receiver, both Multiple modes possible, deeper sleep modes Strongly depends on hardware TI MSP 430, e.g.: four different sleep modes Atmel ATMega: six different modes

52 Some energy consumption figures Microcontroller TI MSP MHz, 3V): Fully operation 1.2 mw Deepest sleep mode 0.3 µw only woken up by external interrupts (not even timer is running any more) Atmel ATMega Operational mode: 15 mw active, 6 mw idle Sleep mode: 75 µw

53 Design methods Asynchronous design: Activate a unit only if there is activity for this unit requested Question: what is the granularity of such a unit? What is the overhead cause by this method Clock Gating Feed clock only to units that require to be active Based on the fact that the energy consumption of a CMOS device linearly scales with the frequency. The basic (standby) current cannot be avoided by this method Problem: the standby current grows with smaller structure sizes Switch of the power supply if possible Questions: What is the granularity of units What is the additional overhead e.g. to save the state When and for how long to switch off

54 Switching between modes Simplest idea: Greedily switch to lower mode whenever possible Problem: Time and power consumption required to reach higher modes not negligible Introduces overhead Switching only pays off if E saved > E overhead Example: Event-triggered E wake up from saved sleep mode Scheduling problem with uncertainty (exercise) P active P sleep E overhead t 1 τ down t event τ up time

55 Alternative: Dynamic voltage scaling Switching modes complicated by uncertainty how long a sleep time is available Alternative: Low supply voltage & clock Dynamic voltage scaling (DVS) Rationale: Power consumption P depends on Clock frequency Square of supply voltage P ~ f V 2 Lower clock allows lower supply voltage Easy to switch to higher clock But: execution takes longer

56 Example Dynamic Voltage Scaling Transmeta Crusoe processor: Scaling dynamics: 700 V to 200 V 700* /200*1.1 2 = Speed reduction 700/200 = 3.5 Energy required per instruction is reduced by 3.5/7.875 = 0.44 Thus 44 % reduction

57 Memory power consumption Crucial part: FLASH memory Power for RAM almost negligible (only true for sensor node with very small amount of RAM) FLASH writing/erasing is expensive Example: FLASH on Mica motes Reading: 1.1 nah per byte Writing: 83.3 nah per byte

58 Transmitter power/energy consumption for n bits Amplifier power: P amp = α amp + β amp P tx P tx radiated power α amp, β amp constants depending on model Highest efficiency (η = P tx / P amp ) at maximum output power Typical efficiency figures range between 15% and 70% In addition: transmitter electronics needs power P txelec Time to transmit n bits: n / (R * R code ) (e.g. 1000b) R nomial data rate (e.g. 10kb/s, R code coding rate (e.g. ¾) Number of send bits= n/ R code (e.g.=1000/3/4=4000/3=1300 ) To leave sleep mode Time T start, average power P start E tx = T start *P start + n / (R * R code ) (P txelec + α amp + β amp P tx ) Simplification: Modulation not considered

59 Receiver power/energy consumption for n bits Receiver also has startup costs Time T start, average power P start Time for n bits is the same n / (R * R code ) Receiver electronics needs P rxelec Plus: energy to decode n bits E decbits E rx = T start *P start + n / (R * R code ) P rxelec + E decbits ( R ) Remark: This model is very much simplified. It does not contain any energy consumption for all digital operation e.g. the FEC decoder. In reality this can be the domination part.

60 Some transceiver numbers

61 Comparison: GSM base station power consumption Heat 602W Heat 1920W Heat 360W Overview AC power 3802W PS 84% DC power 3200W -48V TRX 2400W TRXs RF power 480W ACE Combining TOC RF 120W BTS CE 800W Central equipm. Heat 800W Total Heat 3682W Details (just to put things into perspective) 220V AC Power supply -48V Rack Com- cabling 300W mon -48V 85% 99% 3802W 3232W 3200W PAs consume dominant part of power (12*140W)/2400W=70% Usable PA efficiency 40W/140W=28% Overall efficiency (12*10W)/3802W=3.1% 2400W Fans (No active cooling) 500W cooling 12 transceivers 200W idle 140W 60W Converter 85% -48V/+27V 119W Erlang efficiency 75% DTX activity 47% 9W Bias Combiner Diplexer TOC 110W PA 40W 15W 10W

62 Controlling transceivers Similar to controller, low duty cycle is necessary Easy to do for transmitter similar problem to controller: when is it worthwhile to switch off Difficult for receiver: Not only time when to wake up not known, it also depends on remote partners! Dependence between MAC protocols and power consumption is strong! Only limited applicability of techniques analogue to DVS Dynamic Modulation Scaling (DMS): Switch to modulation best suited to communication depends on channel gain Dynamic Coding Scaling vary coding rate according to channel gain Combinations

63 Computation vs. communication energy cost Tradeoff Directly comparing computation/communication energy cost not possible But: put them into perspective! Energy ratio of sending one bit vs. computing one instruction : Anything between 220 and 2900 in the literature To communicate (send & receive) one kilobyte = computing three million instructions! Communication is significantly more energy consuming than computing Hence: try to compute instead of communicate whenever possible Key technique in WSN in-network processing! Exploit compression schemes, intelligent coding schemes, Process data in the nodes and only send information rather than data Do as much as possible at the edges of the network

64 Outline Sensor node architecture Energy supply and consumption Runtime environments for sensor nodes Case study: TinyOS

65 Operating system challenges in WSN Usual operating system goals Make access to device resources abstract (virtualization) Protect resources from concurrent access Usual means Protected operation modes of the CPU hardware access only in these modes Process with separate address spaces Support by a memory management unit Problem: These are not available in microcontrollers (only partially true) No separate protection modes, no memory management unit Would make devices more expensive, more power-hungry

66 Operating system challenges in WSN Possible options Try to implement as close to an operating system on WSN nodes In particular, try to provide a known programming interface Namely: support for processes! Sacrifice protection of different processes from each other! Possible, but relatively high overhead Do (more or less) away with operating system After all, there is only a single application running on a WSN node No need to protect malicious software parts from each other Direct hardware control by application might improve efficiency Currently popular verdict: no OS, just a simple run-time environment Enough to abstract away hardware access details Biggest impact: Unusual programming model

67 Main issue: How to support concurrency Simplest option: No concurrency, sequential processing of tasks Not satisfactory: Risk of missing data (e.g., from transceiver) when processing data, etc.! Interrupts/asynchronous operation has to be supported Poll sensor Process sensor data Why concurrency is needed Sensor node s CPU has to service the radio modem, the actual sensors, perform computation for application, execute communication protocol software, etc. Poll transceiver Process received packet

68 Traditional concurrency: Processes Traditional OS: processes/threads Based on interrupts, context switching But: not available memory overhead, execution overhead Handle sensor process Handle packet process But: concurrency mismatch One process per protocol entails too many context switches Many tasks in WSN small with respect to context switching overhead And: protection between processes not needed in WSN Only one application anyway OS-mediated process switching

69 Event-based concurrency Alternative: Switch to event-based programming model Perform regular processing or be idle React to events when they happen immediately Basically: interrupt handler Problem: must not remain in interrupt handler too long Danger of loosing events Only save data, post information that event has happened, then return! Run-to-completion principle Two contexts: one for handlers, one for regular execution Sensor event Sensor event handler Idle / Regular processing Radio event Radio event handler

70 Components instead of processes Need an abstraction to group functionality Replacing processes for this purpose E.g.: individual functions of a networking protocol One option: Components Here: In the sense of TinyOS Typically fulfill only a single, well-defined function Main difference to processes: Component does not have an execution Components access same address space, no protection against each other NOT to be confused with component-based programming!

71 API to an event-based protocol stack Usual networking API: sockets Issue: blocking calls to receive data Ill-matched to event-based OS Also: networking semantics in WSNs not necessarily well matched to/by socket semantics API is therefore also event-based E.g.: Tell some component that some other component wants to be informed if and when data has arrived Component will be posted an event once this condition is met Details: see TinyOS example discussion below

72 Components and Publish/Subscribe Protocol functions can be considered as components in a flexible computing environment Components fit very well for pushed based applications E.g. a value can be delivered to the receiver as soon as it is acquired For cross-layer aspects often some additional information from other layers is required RSSI value can be used for: routing, FEC/ARQ, location determination (guessing) etc. There might be several subscribers to this kind of information Event based processing is not the best model to process these structures Use a Publish/Subscribe paradigm for pull based information Data is put onto a blackboard Each subscriber can access the information at any time independent of the data acquisition

73 From Protocol Stack to Functional Network OS/Appl. Encryption RPC Interface OS/Application RPC Interface Transport Fragm./Reass. Error Check ARQ Conn.-Mgmt Flow Control Flow Control Addressing ARQ Network LLC MAC Addressing Routing Addressing Fragm./Reass. Error Check ARQ Conn.-Mgmt Flow Control Encryption Fragm.- Reass. Conn.- Mgmt Control/ Configuration Module Error Check Modulation Digital BB Coding Modulation... Encryption Analog FE Coding Analog FE Medium Medium

74 Dynamic power management Exploiting multiple operation modes is promising Question: When to switch in power-safe mode? Problem: Time & energy overhead associated with wakeup; greedy sleeping is not beneficial (see exercise) Scheduling approach Question: How to control dynamic voltage scaling? More aggressive; stepping up voltage/frequency is easier Deadlines usually bind the required speed form below Or: Trading off fidelity vs. energy consumption! If more energy is available, compute more accurate results Example: Polynomial approximation Start from high or low exponents depending where the polynomial is to be evaluated

75 Outline Sensor node architecture Energy supply and consumption Runtime environments for sensor nodes Case study: TinyOS

76 Case study embedded OS: TinyOS & nesc TinyOS developed by UC Berkely as runtime environment for their motes nesc as adjunct programming language Goal: Small memory footprint Sacrifices made e.g. in ease of use, portability Portability somewhat improved in newer version Most important design aspects Component-based system Components interact by exchanging asynchronous events Components form a program by wiring them together (akin to VHDL hardware description language)

77 TinyOS components Components Frame state information Tasks normal execution program Command handlers Event handlers Handlers Must run to completion Form a component s interface Understand and emits commands & events Hierarchically arranged Events pass upward from hardware to higher-level components Commands are passed downward init start stop fired Command handlers TimerComponent Tasks setrate fire Frame Event handlers

78 Handlers versus tasks Command handlers and events must run to completion Must not wait an indeterminate amount of time Only a request to perform some action Tasks, on the other hand, can perform arbitrary, long computation Also have to be run to completion since no non-cooperative multi-tasking is implemented But can be interrupted by handlers! No need for stack management, tasks are atomic with respect to each other

79 Split-phase programming Handler/task characteristics and separation has consequences on programming model How to implement a blocking call to another component? Example: Order another component to send a packet Blocking function calls are not an option! Split-phase programming First phase: Issue the command to another component Receiving command handler will only receive the command, post it to a task for actual execution and returns immediately Returning from a command invocation does not mean that the command has been executed! Second phase: Invoked component notifies invoker by event that command has been executed Consequences e.g. for buffer handling Buffers can only be freed when completion event is received

80 Structuring commands/events into interfaces Many commands/events can add up nesc solution: Structure corresponding commands/events into interface types Example: Structure timer into three interfaces StdCtrl Timer Clock init start stop fired Build configurations by wiring together corresponding interfaces StdCtrl Timer TimerComponent Clock setrate fire

81 Building components out of simpler ones Wire together components to form more complex components out of simpler ones StdCtrl StdCtrl Timer Timer New interfaces for the complex component TimerComponent Clock CompleteTimer Clock HWClock

82 Defining modules and components in nesc

83 Wiring components to form a configuration

84 Basic properties of Reflex OS OS platform for deeply embedded systems Consists of: Scheduler FIFO, Fixed Priority; EDF (Earliest Deadline First) Common interrupt handling scheme Mechanisms for event-driven applications Completely written in C++ Basic programming model : event flow

85 Reflex Event Flow Scheme

86 TinyOS versus Reflex 1 Implementation Language TinyOS: NesC; Components do not have state Special compiler and preprocessor needed Reflex: C++; full multi-threaded stateful components Standard C++ tool-chain applicable System Configuration TinyOS: Components; Handlers; Frame Command Handlers; Event Handlers; Tasks Commands run to completion; Handlers cannot be interrupted Connections are done during compile time Reflex: Objects Objects are connected to objects via event channels (an abstract object class) Connections are done during runtime

87 TinyOS versus Reflex 2 Scheduling TinyOS: Simple FIFO scheme; no preemtion possible but via interrupts; No nested interrupts possible to avoid stack management. No Hard Realtime possible Reflex: Flexible and dynamic Scheduling, Three schedulers are available already and can be extended and amended. Preemtion possible but no nested or recursive call (self lock in run method) Event Flow Support TinyOS Simple parameterless event call chain No implicit synchronization Reflex: Event channels via synchronized buffers Buffers are typed and transparent to allow high flexibility

88 TinyOS versus Reflex 3 Codesize comparison:

89 Summary For WSN, the need to build cheap, low-energy, (small) devices has various consequences for system design Radio frontends and controllers are much simpler than in conventional mobile networks Energy supply and scavenging are still (and for the foreseeable future) a premium resource Power management (switching off or throttling down devices) crucial Unique programming challenges of embedded systems Concurrency without support, protection De facto standard: TinyOS TinyOS has some big drawbacks: nesc: Special language; needs precompiler; Portability Other operating systems: Reflex, Kontiki Can we build an efficient OS for WSN applications with the used comfort of ordinary OSs?