Embedded System Design Process PDF

CPEN 120- EMBEDDED SYSTEMS Topic 3 Embedded System Design Process Intended Leaming Outcomes: At the end if this topic, the student should be able to: 1. Describe stages of the embedded system design process 2. ldentify functi...

CPEN 120- EMBEDDED SYSTEMS Topic 3 Embedded System Design Process Intended Leaming Outcomes: At the end if this topic, the student should be able to: 1. Describe stages of the embedded system design process 2. ldentify functional and non-functional requirements for embedded systems based on user needs and system constraints. 3. Apply established design methodologies to create embedded system solutions. Learning Discussions specialized computing Embedded system design is the process of creating product. This involves larger device or systems that peform dedicated tasks within a like microcontrollers, sensors, and selecting and integrating hardware components optimized software (firmware) to communication interfaces, alongside developing requirements. Embedded systems and power meet specific functional, performance, constraints, requiring fast and reliable responses to often operate under real-time external inputs. Embedded System Design Process developing embedded system design process is a structured approach to devices. It The perform dedicated tasks within larger software specialized computing systems that and defining system requirements, designing hardware system. involves a cohesive selecting components, and integrating them intodevelopment, system architecture, fimware like hardware design, The process includes steps to ensure the system meets functional, performance, integration,and rigorous testing and validated, the system is deployed in its Once tested and power constraints. by ongoing maintenance and updates to enhance target environment, followed process. issues. Figure 3-1 shows the steps in the design performance or address CEIT -DCEEE  CPEN120 - EMBEDDED SYSTEMs Requirement Top down Bottom-up design design Specifications Architecture Components System Integration Steps Figure 3-1: Design Process approaches process, top-down and bottom-up approaches design In the embeddedmethodologies for system development. Both they represent two distinct integrate different parts of a system, but break down and guide how designers points and focus. differ in their starting the overall system's high-level approach begins with broken down intO smaller, The top-down From there, the system is detail as the requirements and architecture. modules. Each module is refined in components and more specific design progresses. overall Advantages top-down approach provides a clear within the System Architecture: The component fits well Clear each system, ensuring that Structure of the as a bigger picture. Issues: By focusing on the system pertormance System-Level intera ction, Early Detection of architectural issues (e.g., module whole, potential identified early in the design process. bottlenecks)can be CET -DCEEE 2  CPEN 120- EMBEDDED SYSTEMS Easler Maintenance: The design is modular, replace individual components without disruptingmaking it easier to update or the entire system. Disadvantages Risk of Over-Planning: Since detailed design is deferred until after high-level architecture is developed, there is a risk that some low-level design aspects may not fit neatly into the system architecture, requiring later adjustments. Requires Comprehensive Initial Knowledge: It assumes that designers fully understand the entire system at the start, which may not always be feasible. Example: In designing an automotive embedded systemlike a smart braking system (ABS), the top-down approach would begin with the high-level requirement of ensuring safety by controllingthe brakes under specific conditions. From there, the designer would break the system into subsystems such as the sensor module (wheel speed sensors), the actuator module (braking mechanism), and the controller module (microcontroller managing decision-making), refining each module as needed. implementing the The bottom-up approach begins with designing and components are then individual smallest, most specific components first. These eventually building up to the gradually combined to form higher-level subsystems, complete system. Advantages Each individual component is tested thoroughly as it's Early Testing: early in the design process. developed, allowing issues to be caught reusable The bottom-up approach promotes the creation of Reusability: development time in future projects. modules, which can reduce work on different components independently and Flexibility: Designers can in complex they are completed, which can save time integrate them as projects. Disadvantages components, Since the focus is on individual leading to Overview: potentially Lack of System sight of the bigger picture, there is a risk of losing integration issues later. architecture, Without a well-defined high-level interactions Integration Challenges: become complex, particularly if the may Integrating components planned out. between modules aren't CET -DCEEE  CPEN 120- EMBEDDED SYSTEMS Risk of Redesign: If early components are not designed with the ful system in mind, they may need to be reworked to fit the final architecture. Example: Inthe case of an loT-based embedded home automation system, using the bottom-up approach would involve first designing individual components such as a temperature sensor module, a relay control module for turning devices on and off, and a Wi-Fi communication module. Once each module works independentiy, they are integrated into a larger controlsystem that communicates with a central server or mobile app. Top-Down vs Bottom-Up Criteria Top-Down Approach Bottom-Up Approach Starting Point High-level system architecture Small, specific components and and design modules Focus System behavior and Building reliable and reusable interaction between subsystems components Development Decomposes system into Combines components into subsystems subsystems Planning Heavily planned from the start More exploratory and flexible Risk Overlooking low-level details Integration issues due to lack of system overview Testing System-level testing perfomed Early component-level testing after major design Best Use Complex systems requiring Systems with reusable Cases clear overall archite cture components or incremental builds Which Approach to Use? Top-Down Approach is better suited for complex, large-scale systems where the interactions between components need to be carefully orchestrated. This method provides a clear understanding of the overall system before delving into details. Bottom-Up Approach is useful in scenarios where components are wel understOod and can be developed independently, or in systems where Components need to be hiahly reusable. It is more flexible and allows for incremental progress. CEIT -DCEEE  CPEN 120- EMBEDDED SYSTEMS In practice, many embedded systems are designed using a hybrid approach that combines elements of both top-down and bottom-up strategies. For example, high-level architecture may be defined early on (top-down), while specific subsystems are developed and tested incrementally (bottom-up). Designing embedded systems involves a structured approach, from gathering requirements to system integration. 1. Requirement Gathering The first step in designing an embedded system is to collect the requirements from the stakeholders like the ciient or end-users. It defines the high-level goals and objectives of the system, such as what the embedded system is supposed to do. and non-functional It involves gathering and understanding the functional is supposed requirements of the system. Youneed to determine what the system constraints, consumption limits, cost to do, its performance expectations, power and any real-time or safety requirements. Functional and Non-functional Requirements are generally split into two types: requirements. should do) - These are the Functional Requirements (What the system demands as basic facilities that the specifically requirements that the end user system should offer. Requirements are: Example:Smart Thermostat Functional the Monitoring: The themostat must continuously monitor Temperature display it on the user interface. room temperature and desired target Temperature Setting: Users must be able to set a Target or a mobile app. temperature through physical buttons themostat must activate the heating byor Heating/Cooling Control: The from the target system when the room temperature deviates Cooling threshold. or weekly more than a specified should allow users to set daily Scheduling: The system temperature adjustments. schedules for automatic themostat must connect to the home's Wi-Fi W-Fi Connectivity: The control via amobile application. network to allow remote are the system should pertom) - Theseproject (How Non-Functional Requirements system must satisfy according to the constraints that the ne quality CEIT-DCEEE  CPEN 120-EMBEDDED SYSTEMS contract. They are also called non-behavioral requirements. Some of the non functional requirements are Portability Security Maintainability Reliability Scalability Performance Reusability Flexibility Example: Smart Thermostat Non-Functional Requirements are: Usability: The user interface (both physical and mobile app) must be less. intuitive, allowing users to set preferences within 3 steps or constant Reliability: The system must maintain 99.9% uptime, ensuring monitoring and controlof the HVAC system. temperature Response Time: The thermostat must react to changes in seconds. app within 5 within 1 second and update the display or mobile consume less than 0.5W in Energy Efficiency: The device itself should extend battery life power to standby mode and must efficiently manage (if battery-operated). Communication protocols (e.g., Security: The system must use secure prevent unauthorized access. HTTPS, WPA2) to protect user data and handle up to 10 connected Scalability: The system should be able to multiple themostats or remote devices in the same household (e.g., sensors). modular and well-documented, Maintainability: The firmware must be downtime. fixes with minimal allowing for easy updates or bug 2. System Speciflcatlons refined into a specification. The specifications Requirements gathered are descriptions to meet the requirements. provide detailed technical Specifications Example: Smart Thermostat accuracy. to 50°C with +0.5°C Temperature Sensor Range: -20°C resolution of 320x240 pixels. with a Display: A2-inch LCD screen(32-bit, 120 MHz). Processor: ARM Cortex-M4 802.11 b/g/n for wireless connectivity. Communication Protocols: Wi-Fi 0.5W in standby mode. Power Consumption: Max buttons and amobile app (using a REST Interface: Capacitive touch API). Operating Voltage: 3.3V. CHII -DCEEE 6  CPEN 120- EMBEDDED SYSTEMS 3. System Architecture The system architecture is the blueprint for how the components of the embedded system will interact with each other. It defines how the hardware and software components interact. It includes communication protocols, interfaces, data flow, and control algorithms. It often involves modular design, where each module performs a specific function, such as sensing, processing, communication, or user interaction. User input temperalure User Set temperature Transmit weather forecast Smart HVAC System Wifi Router Thermostat Send temperatre Temperature Sensor Current lemperatwe diagram for Smart Thermostat Figure 3-2: Example of Data flow Components 4. Hardware and Software need to be architecture is already defined, specific componentshardware and When the design. The component in general includes both (procesSors, selected for the choose appropriate hardware components modules. You must goal is to software modules, etc.) based on the specifications. The sensors, Communication the required performance, reliability, and cost meet from ensure that the hardware can are ready-made. some can be modified constraints. Some componentsdesigned from scratch. existing designs, others must be  CPEN 120- EMBEDDED SYSTEMS Example: Hardware Components for Smart Thermistor Microcontroller: STM32F4 Series (Cortex-M4) due to its processing power, low energy consumption, and peripheral support. Temperature Sensor: DS18B20, a digital temperature sensor known for high accuracy and ease of interfacing. WI-Fi Module: ESP8266, a low-cost, low-power Wi-Fi chip with sufficient support for embedded systems. Display: 2" TFT LCD with SPI interface. input to 3.3V Power Supply: A buck converter circuit that steps down a 5V for the microcontroller and peripherals. Capacitive Touch IC: TTP223 touch sensor for user input buttons. 5. System Integration integrated into a Once the hardware and software are developed, they are connected, and the complete system. During integration, the components are together correctly. software is loaded onto the hardware to ensure they function behavior Communication between modules, sensor data flow, and overall system is tested. Testing and Validation perform as Embedded systems undergo rigorous testing to ensure they performance required in real-world conditions. This includes functional testing, testing, and reliability testing. Testing includes: Unit Testing: Testing individual components or software modules. System Testing: Ensuring that the entire system operates as expected when all components are integrated. Performance Testing: Checking that the system operates efficiently and meets timingor real-time constraints. o Power Testing: Ensuring power consumption is within acceptable limits. Field Testing: Testing the system in real-wortd conditions to ensure its o reliability and robustness. Validation ensures the system meets the end-user's needs and operates reliably over time. CEIT -DCEEE  CPEN 120- EMBEDDEDSYSTEMS Embedded System Design lssues In general, an embedded system design must consider the following : 1. It should handle real time operations and react to external events 2. It should conform to size and weight limits 3. It should be within the power budget and cooling requirements. 4. It should satisfy safety and reliability requirements 5. It must meet tight cost targets. Embedded system design faces several challenges due to the nature of embedded systems and their application-specific requirements. Some of the common embedded design issues: Real-Time Constraints Many embedded systems must respond to inputs or changes in the environment within strict time limits. Missing these deadlines can cause the system to fail or behave unpredictably. Example: In automotive safety systems like airbags, real-time processing is critical for timely deployment during an accident. Power Consumption Embedded systems, especially in portable or battery-operated devices, must consume minimal power to extend battery life. Designers need to balance performance with power efficiency. Example: Wearable health monitors need to operate continuously for extended periods on small batteries. Memory and Resource Constraints Embedded systems typically have limited memory (RAM, ROM, Flash) and Computational resources due to size, cost, and power constraints. This limits the complexity of the software and algorithms that can be implemented. Example: An embedded microcontroller in a smart sensor might only have a few kilobytes of RAM, requiring highly optimized software. Security Embedded systems are increasingly connected to networks (loT), making hem vulnerable to cyber-attacks. Ensuring data security and unauthorized access are crucial challenges. prevening CEIT -DCEEE  CPEN120- EMBEDDED SYSTEMS Example. A connected home security system must attempts, ensuringthat the data and controls are safeguard against hacking protected Reliability and Robustness Many embedded systems are deployed in harsh automotive, industrial, medical) and need to operate environments (e.g. reliably over long periods. They must tolerate temperature fluctuations, vibrations, or humidity without failure Example: A medical device like a pacemaker must operate flawlessly in diverse conditions without failure, as it dire ctly impacts a patient's life. Scalability Embedded systems must be designed to support future enhancements and integrations without requiring a complete redesign. Adding new features or capabilities can be challenging within the constrained environment. Example: Asmart home thermostat that needs to support more devices or sensors over time must scale without significant redesign. Integration with Hardware integration between hardware Embedded system design often involves tight the hardware components ensure that and software. Designers must be actuators) and the software work seamlessly together, which can (sensors, involved. difficult whenmultiple vendors are system Designing firmware for a sensor in an industrial monitoring in data Example: or latency hardware's quirks, such as noise must account for the collection. Cost Constraints produced at low cost, especially for need to be and requires Many embedded systems limits the choice of components Consumner products. This reduce hardware and production costs. optimization of the design to household appliance must embedded system for a processing Example: A loW-cost affordability, leading to compromises on balance performance with power or memory. CET -DCEEE 10  CPEN 120- EMBEDDED SYSTEMS Software Complexity and Testing Embedded softvware can become complex, especially as systems add more features. Ensuring thorough testing and validation is difficult, real-time or critical systems where bugs can lead to particularly in catastrophic failure. Example: Asoftware bug in an embedded flight control system can lead serious consequences, making thorough testing essential. to Upgrade and Maintenance Many embedded systems are deployed in environments where physical access is limited, making upgrading and maintenance difficult. Ensuringover the-air (OTA) updates while maintaining security and system integrity is a challenge. Example: An embedded system in aremote wind turbine must support OTA updates since manualacCcess is Costly and inconvenient. Interfacing and Communication Protocols Embedded systems often need to interface with other systems or sensors using various communication protocols (e.g., 12C, SPI, CAN, Bluetooth). Ensuring compatibility and reliability across different interfaces can be a challenge. Example: An loT sensor network may need to support multiple wireless protocols like Wi-Fi and Zigbee for different devices, requiring careful protocol management.