Real Time Embedded Systems

Document Sample
Real Time Embedded Systems Powered By Docstoc
					                              Real Time
                              Embedded Systems
                           Done by: Lara Quadan & Lina Sawalha
                             Supervised by: Dr. Lo’ai Tawalbeh

                              Computer Engineering Department
CPE-746-JUST-Fall 2006
                         Jordan University of Science and Technology
Quadan & Sawalha
   Computer is entering all facets of life from home
    electronics to production of different products and
   Many of the computers are embedded and thus
    “hidden” for the user.
   So it is necessary to know about the control and
    characteristics of embedded real-time systems to
    implement reliable ones.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Real-time Systems
   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. In other words,
    a late answer is a wrong answer.
   A real-time system consists of tasks under deadline constrains.
   Real-time systems maybe distributed systems, consisting of
    components that are physically distributed. These systems
    consists of multiple processing units and communication links.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Functional Requirements
   Data Collection
     Sensors
     Signals
     Alarm
   Direct Digital Control
     Actuators
   Man-machine interaction

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Temporal Requirements
   Tasks may have deadline
   Minimal latency jitter
   Minimal error detection latency
   Timing requirements due to tight software control
   Human interface timing requirements

CPE-746-JUST-Fall 2006- Quadan & Sawalha
    Application Areas
     Customer electronics:                       Computer/Communication products:
        Cameras                                     Peripherals
        Camcorders                                  Fax machines
     Customer products:                          Protection & security systems :
        Dish washers                                Intruder Alarm
        Microwave ovens                             Smoke/Gas detection
     Cars:                                       Robotics
        Anti-lock braking
        Engine control
        Drive-by-wire
     Planes:
        Stability
        Jet engine
        Fly-by-wire
     Military:
        Weapons
        Satellites
     Industrial process controllers

    CPE-746-JUST-Fall 2006- Quadan & Sawalha
Control Theory
   A control system is an interconnection of components forming a system configuration
    that will provide a desired system response.
   Types of control :
    – Open-loop control: a system that utilizes a device to control the process without
    feedback. Thus the output has no effect upon the signal to the process.

    – Closed-loop control: A system that uses a measurement of the system output
    through sensors and compares it with the desired output (called feedback system).

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Embedded digital controlled
   When a digital controller (microprocessor/Microcontroller) is used in the embedded
    system to control the plant or process an analog/digital converter (A/D) is required as
    well as digital/analog converter (D/A) to interface the input/output controller digital
    signals with the analog signals from the other component in the system.

   it is very important to consider the timing for the real-time digital controlled system.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Cruise Control

   Regulate the speed of a car by adjusting the
     Input by the driver  sets a speed and car
    maintains it.
   Measures the speed through device connected
    to the driver shaft.
   Hard real-time: driver shaft revolution event.
   Soft real-time: driver inputs, throttle adjustments.
CPE-746-JUST-Fall 2006- Quadan & Sawalha

   Disturbances: road surface and grade, wind and obstacles.
   Input by the driver  sets a speed and car maintains it.
   If the driver give a brake command to the car the cruise control system will
    be interrupted and the car speed will slow down or be stopped.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Other car embedded systems

CPE-746-JUST-Fall 2006- Quadan & Sawalha

 Objective function: to control the direction
and speed of the plane.
 Outputs: actual direction and speed of the plane
 Control inputs: path markings and speed.
 Disturbances: wind, obstacles.
 subsystems: power system,
engines, steering system,
braking system, . . .
CPE-746-JUST-Fall 2006- Quadan & Sawalha

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Robot Vacuum cleaner

CPE-746-JUST-Fall 2006- Quadan & Sawalha
    Operating environment:
      closed indoor environments
     such as rooms.

    Controlling System
     – Human : turn on and off the robot
     – Computer: Sensors: stairs and wall detection (IR
    Controls: robot location through controlling the robot
     speed and steering to follow the walls.
    Actuators: motors and wheels.
    CPE-746-JUST-Fall 2006- Quadan & Sawalha
    Hard Disk Drive Control
    The disk rotates at a speed
     between 1800 and 7200 rpm.
    The head flies above the disk at
     a distance of less than 100 nm.
    Position accuracy is 1 micro
    Move the head from track a to
     track b within 50 ms.
    Motor to actuate the arm to the
     desired location on the disk.

    CPE-746-JUST-Fall 2006- Quadan & Sawalha

  For DVD

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Wireless Sensor Network
Embedded System

                    Sensor nodes scattered in a sensor field

CPE-746-JUST-Fall 2006- Quadan & Sawalha
   Collection of sensor nodes in a
   Each node collects data and send it
    to a sink.
   The sink aggregates the data and
    send it to a main processing unit.
   Each sensor node can manipulate
    data using its own embedded
    microcontroller (microprocessor),
    transceiver and communication

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Keck Astronomical
   World largest telescope
   The main objective is to collect and focus starlight
    using a large concave mirror.
   The shape of the mirror determines the quality of the
    observed image.
   The diameter of the mirror is 10 m.
   The mirror is a mosaic of 36 hexagonal small mirrors.
   The 36 segments must be aligned so that the resulting
    mirror has the desired shape

CPE-746-JUST-Fall 2006- Quadan & Sawalha
   Behind each segment are three actuators applying
    forces at three points on the segment affecting its
   In the gap between every two adjacent segments a
    capacitor type sensors measuring the local displacement
    between segments.
   These displacements should be controlled to give the
    desired shape and direction for the mirror.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
   Auto-washing machines

 The user will select the required washing program.
 A set of sensor will monitor the washing process variables such as water level,
temperature, and the rotation speed. Where the embedded controller will keep these
values agree with the selected washing program. through out the washing time.
   CPE-746-JUST-Fall 2006- Quadan & Sawalha
Classifications of RT systems
      Hard vs.  Soft systems
      Fail-safe vs. Fail operational
      Guaranteed-response vs. Best-effort
      Resource-adequate vs. Resource-inadequate
      Event-triggered vs. Time-triggered

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Hard vs. Soft Real Time Systems
   Hard RTS
     A RTS where the tasks have to be
      performed not only correctly but also
      on time. i.e. a catastrophic system failure can
      occur if the system is not responding on time.
     E.g., the aircraft turbulence (instability in the atmosphere)
      controller should respond on time, if not there will be a definite
     Other examples
         control systems for space probes and nuclear reactors.

         refresh rates for video, or DRAM.

         collision alert.
CPE-746-JUST-Fall 2006- Quadan & Sawalha
   Soft RTS
      A RTS where the tasks have to be
       performed as fast as possible, but the tasks
       need not be finished within deadlines.

      I.e. It  may go beyond specified deadlines without catastrophic
         failures (For example, if the video game is not responding on
         time, it’s only a delay where the user may not want to wait,
         but even then nothing awful is going to happen).

CPE-746-JUST-Fall 2006- Quadan & Sawalha
        Another e.g.: Laser printer: rated by pages-per-minute, but can take
         differing times to print a page (depending on the "complexity" of
         the page) without harming the machine or the customer.

        Task execution may also be timed-out: display updates, connection

        Other examples of Soft real time systems:
           videoconferencing
           stock price quotation systems
           airline reservation systems
           automatic teller

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Characteristics of RTS
   Response time: - hard real-time applications, milliseconds or less,
    preclude direct human intervention during normal operation and in critical
    situations. - Soft real-time and on-line systems are often in the order of
   Peak-load performance: - hard real-time system: the peak-load scenario
    must be well-defined, guaranteed to meet the specified deadlines in all
    situations.- soft-real time system, the average performance is important,
    and a degraded operation in a rarely occurring peak load case is tolerated
   Control of Pace: A hard real-time system must remain synchronous with
    all the state of the environment in all cases. This is in contrast to an on-
    line system, which can exercise some control over the environment in case
    it cannot process the offered load.
   Safety: Hard real-time systems are often safety critical.
   Size of Data files: Hard real-time systems have small data files and real-
    time databases. Temporal accuracy is often the concern here. Soft real
    time systems has larger databases so long-term integrity of large data files is
    the key issues.
CPE-746-JUST-Fall 2006- Quadan & Sawalha
Characteristics of RTS
   Redundancy Type: If an error occurs in a soft real-time system,
    the computation is rolled back to a previously established
    checkpoint to initiate a recovery action. In hard real-time
    systems, recovery is of limited use.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Common Misconceptions
 Misconceptions about Real-Time Systems
 faster hardware implies all deadlines will be met
 real-time computing is merely fast computing
 RT systems are low-level coding done using ad-hoc methods.
 Fast is relative. More important that system is “fast enough",
  deterministic and predictable.
 Worst-case response times of interest rather than average-case.
 Scheduling theory, software design, formal methods and RTOS
  are changing things.
 It is important to note that hard versus soft real-time does not
  necessarily relate to the length of time available.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Real Time terminologies
   Synchronous: events occur at predictable times in the flow-of-control.
   Asynchronous: interrupts
   Multitasking: The process of scheduling and switching the CPU between
    several tasks, a single CPU switches its attention between several sequential
    tasks. Multiple is like foreground/background with multiple backgrounds.
    Multitasking maximizes the utilization of the CPU and also provides for
    modular constructions of applications.
   Task: A task is called a Thread, is a simple program which thinks it has the
    CPU all to itself. Each task typically is an infinite loop that can be in any one
    of the three states: RUNNING, READY (-TO-RUN), BLOCKED.
   Resource: A resource is any entity used by a task. A resource can thus be an
    I/O device, such as a printer, a keyboard, or a display, or a variable, a
    structure, or an array.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Real Time terminologies
   Shared Resource: A shared resource is a resource that can be used by more
    than one task. Each task should gain exclusive access to the shared resource
    to prevent data corruption. This is called Mutual Exclusion. For e.g., a global
    variable can be used by many tasks, or printer which could be shared by "n"
   Critical Section of Code: Also called Critical Region, is code that needs to
    be treated indivisibly; once the section of code starts executing, it must not be
    interrupted. To ensure this, interrupts are typically disabled before the critical
    code is executed and enable when the critical code is finished.
   Reentrancy: A reentrant function can be used by more than one task without
    fear of data corruption. a reentrant function can be interrupted at any time
    and resumed at a later time without loss of data. Reentrant functions either
    use local variables (i.e. CPU registers or variables on the stack) or protect data
    when global variables are used.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Operating System
    Real-time operating systems must provide support for:
      Guaranteeing real-time constraints.
      Supporting fault tolerance and distribution.
      Time constraint resource allocations and
      Time constraint end-to-end communications.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Real Time Operating System
   Fundamental requirements for an RTOS
      The OS behavior must be predictable
      The OS must be multithreaded and preemptive.
      The OS must support thread priority.
      The OS must support predictable thread synchronization mechanisms.
   Additional Requirements:
      The maximum time that device drivers use to process an interrupt, and
       specific IRQ information relating to those device drivers, must be known.
      The interrupt latency (the time from interrupt to task run) must be
       predictable and compatible with application requirements

CPE-746-JUST-Fall 2006- Quadan & Sawalha
   Kernel
      Is the core program of the Operating System, and determines its behavior
      Is responsible for System Resources and for “context switch”.
      Also responsible for task scheduling.
      In RTOS since the size of the kernel is small, its also referred to as Micro

     Kernel types
      Microkernel - scheduler
      Kernel - a microkernel with intertask synchronization
      Executive - a kernel that includes privatized memory blocks, I/O services,
       and other complex issues. Most commercial realtime
       kernels are in this category.
      Operating system - an executive that also provides generalized user interface,
       security, file management system, etc.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
   Three groups
      Small, fast, proprietary kernels
      Real-time extensions to commercial operating systems
      Research operating systems
   Small, fast, proprietary kernels
      homegrown
      commercial offerings:
           QNX, PDOS, pSOS, VCOS, VRTX32, VxWorks
      To reduce the run-time overheads incurred by the kernel and to make the
       system fast, the kernel
           has a small size
           responds to external interrupts quickly
           minimizes intervals during which interrupts are disabled
           provides fixed or variable sized partitions for memory
           management as well as the ability to lock code and data in memory
           provides special sequential files that can accumulate data at a fast rate

CPE-746-JUST-Fall 2006- Quadan & Sawalha
        To deal with timing constraints, the kernel
           provides bounded execution time for most primitives
           maintains a real-time clock
           provides primitives to delay processing by a fixed amount of time and
             to suspend/resume execution
        Also, the kernel
           performs multitasking and intertask communication and
             synchronization via standard primitives such as mailboxes, events,
             signals, and semaphores.
        For complex embedded systems, these kernels are inadequate as they are
         designed to be fast rather than to be predictable in every aspect.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Examples of RTOS
   Open source:                              Proprietary:
      eCos                                        Ardence RTX    - BeOS
      Fiasco (L4 clone) [1]                       ChorusOS       - DNIX
                                                   DSOS           - embOS (Segger)
      FreeRTOS
                                                   ITRON
      Linux as of kernel version
                                                   LynxOS         - MicroC/OS-II
                                                   MQX RTOS [5]    - Nucleus
      Phoenix-RTOS
                                                   OS-9            - OSE
      Nut/OS [2]
                                                   OSEK/VDX        - OSEKtime
      Prex                                        PDOS            - Phar Lap ETS
      RTAI                                        PikeOS          - Portos
      RTEMS                                       pSOS            - QNX
      RTLinux                                     RMX             - RSX-11
      SHaRK [3]                                   RT-11           - RTOS-UH
                                                   RTXC            - Salvo RTOS [6]
      TRON Project
                                                   SINTRAN III    - Symbian OS
      Xenomai [4]
                                                   ThreadX        - VRTX
                                                   VxWorks         - Windows CE
CPE-746-JUST-Fall 2006- Quadan & Sawalha
                                                   µnOS           - UNIX-RTR
   Real-Time Linux (now part of FSMLabs Inc.)
   that is suitable for Real-Time applications.
   Its view is that the application system can be split into two parts
      Real-time
      Non real-time
   With this approach, it splits the applications to run on either the
    Linux kernel, or a real-time kernel!

CPE-746-JUST-Fall 2006- Quadan & Sawalha
   Between the real and non-real parts communication is
    performed through FIFOs called RT-FIFOs.
   These FIFOs are:
      Locked to memory in kernel space.
      FIFOs appear as devices to Linux user processes.
      Reads and writes are non-blocking and atomic.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
   RTLinux eliminates the problem of the kernel blocking
    interrupts by replacing HW interrupts by SW emulated
   Thus, the RT kernel intersects all interrupts!
   If an interrupt is to let a RT task execute, then the RT
    kernel pre-empts any general application and runs the
    RT task.
   Flags are used to emulate interrupt disabling.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
    Monolithic Kernel
       Reduced run-time overhead, but increased kernel
        size compared to Microkernel designs
    Supports Real-Time POSIX standards
    Common in industry
       Mars missions
       Honda ASIMO robot
       Switches
       MRI scanners
       Car engine control systems
CPE-746-JUST-Fall 2006- Quadan & Sawalha
Software vs. Hardware
   Hardware
      Fast
      Power-efficient
      Used for performance and security
   Software
      Flexible
      Reusable

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Examples on Software Subsystems
   The software components of an embedded system can be diverse.
      Software running on a microprocessor core.
          Written in a “High level language”

      Cooperative multi-tasking software running on one or more DSP
              Largely written in Assembly code due to a lack of high quality
        An RTOS kernel running on the microprocessor core and/or DSP core
        Control Processes running on a microcontroller
        User interface modules running on the microprocessor
        Devise drivers for interface protocols e.g. TCP/IP, ATM

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Software Languages
   Specifying machine code concisely
   Sequential semantics:
      Perform this operation
      Change system state
   Raising abstraction: symbols, expressions, control-flow,
    functions, objects, templates, garbage collection.
   C/C++/Java
   Real Time Operating Systems: Adds concurrency, timing control

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Hardware Subsystems
   Microprocessor cores
   DSP cores
   Field Programmable Gate Arrays (FPGAs)
   Application specific integrated circuits
   (ASICs)
      standard cell/synthesized deigns
      custom ICS
   Memory
      RAM, ROM, PROM, ...
      Conten addressable memory (CAM)
      Specialized DRAMs, multi-bank memories,
   System Bus structures & interfaces
   I/O interfaces
   Others
CPE-746-JUST-Fall 2006- Quadan & Sawalha
Hardware languages
   Specifying connected gates concisely
   Originally targeted at simulation
   Discrete event semantics skip idle portions
   Mixture of structural and procedural modeling
   Hardware Languages:
      Verilog
           Structural and procedural modeling
           Four-valued vectors
           Gate and transistor primitives
           Less flexible, succinct
      VHDL
           Structural and procedural modeling
           Few built-in types; Powerful type system
           Fewer built-in features for hardware modeling
           More flexible, verbose
      SystemC 1.0, Cynlib: “VHDL/Verilog in C++”
CPE-746-JUST-Fall 2006- Quadan & Sawalha
Hardware/Software Codesign
   Most systems and mixed between hardware and software.
   Traditional Design
      SW and HW partitioning is decided at an early stage, and designs proceed
       separately from then onward.
   CAD today addresses synthesis problems at a purely hardware level:
      efficient techniques for data-path and control synthesis down to silicon.
   ECS use diverse (commodity) components
      uP, DSP cores, network and bus interfaces, etc.
   "New fangled" Codesign
      A flexible design strategy, wherein the HW/SW designs proceed in
       parallel, with feedback and interaction occurring between the two as the
       design progresses.
      Final HW/SW partition/allocation is made after evaluating trade-offs and
       performance of options.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
Examples of mixed hardware-
software embedded systems
   Embedded controllers for reactive real-time applications are
    implemented as mixed software-hardware systems. These
    controllers utilize Micro-processors, Micro-controllers and
    Digital Signal Processors but are neither used nor perceived as
    computers. Generally, software is used for features and
    flexibility, while hardware is used for performance. Some
    examples of applications of embedded controllers are:
      Consumer Electronics: microwave ovens, cameras,
        compact disk players.
      Telecommunications: telephone switches, cellular phones.
      Automotive: engine controllers, anti-lock brake controllers.
      Plant Control: robots, plant monitors.
CPE-746-JUST-Fall 2006- Quadan & Sawalha
Embedded System Design
   Modelling
      the system to be designed, and experimenting with algorithms involved;
   Refining (or “partitioning”)
      the function to be implemented into smaller, interacting pieces;
   HW-SW partitioning: Allocating
      elements in the refined model to either (1) HW units, or (2) SW running
        on custom hardware or a general microprocessor.
   Scheduling
      the times at which the functions are executed.
       This is important when several modules in the partition share a single
        hardware unit.
   Mapping (Implementing)
      a functional description into (1) software that runs on a processor or (2) a
        collection of custom, semi-custom, or commodity HW.

CPE-746-JUST-Fall 2006- Quadan & Sawalha
    Kopetz, H. , Real-Time Systems: Design Principles for Distributed
     Embedded Applications.
    Calton Pu, 1B: RTE Concepts and Examples
    A Framework for Hardware-Software Co-Design of Embedded Systems,
     University of California, Berkeley.

CPE-746-JUST-Fall 2006- Quadan & Sawalha