📚 Year 13 OCR Computer Science: Core Knowledge Points Review | A Level计算机核心知识点梳理
As you approach your OCR A Level Computer Science examination, a solid grasp of the entire two‑year specification is essential. This revision guide pulls together the most significant theoretical concepts and practical techniques from both Component 1 (Computer Systems) and Component 2 (Algorithms and Programming), offering a structured refresher designed to reinforce your understanding and fill any gaps before the final assessment.
在备战OCR A Level计算机科学考试之际，全面掌握两年课程大纲中的核心知识至关重要。本复习指南整合了组件一（计算机系统）与组件二（算法与编程）中最关键的理论概念和实践技术，为你提供一份结构清晰的要点梳理，旨在巩固理解、查漏补缺，帮助你在最终测评前做好充分准备。
1. Processor Architecture and Instruction Sets | 处理器架构与指令集
The Central Processing Unit (CPU) is the brain of the computer, organised into the control unit (CU), arithmetic logic unit (ALU), and a set of registers such as the program counter (PC), memory address register (MAR), memory data register (MDR), current instruction register (CIR), and accumulator (ACC). Together they carry out the Fetch‑Decode‑Execute cycle, which is repeated at a frequency determined by the system clock.
中央处理器（CPU）是计算机的大脑，由控制单元（CU）、算术逻辑单元（ALU）及一组寄存器组成，包括程序计数器（PC）、内存地址寄存器（MAR）、内存数据寄存器（MDR）、当前指令寄存器（CIR）和累加器（ACC）。它们共同执行“取指—译码—执行”循环，该循环以系统时钟确定的频率不断重复。
Performance is influenced by clock speed, the number of processor cores, and the size and level of cache memory. Parallel processing exploits multiple cores to execute instructions simultaneously, but its benefit depends on how well a program can be divided across cores. Pipelining is another technique that improves throughput by overlapping the fetch, decode and execute stages of different instructions.
性能受时钟速度、处理器核心数以及缓存的大小与层级影响。并行处理利用多个核心同时执行指令，但其收益取决于程序能否被有效分解到各个核心上。流水线是另一种通过重叠不同指令的取指、译码和执行阶段来提高吞吐量的技术。
Modern CPUs adhere to either Complex Instruction Set Computer (CISC) or Reduced Instruction Set Computer (RISC) philosophies. CISC processors (e.g., x86) employ a large set of variable‑length instructions, often with single instructions that perform multi‑step operations, which can reduce memory usage but complicate pipelining. RISC processors (e.g., ARM) use a smaller set of fixed‑length instructions, typically executing one instruction per clock cycle, relying on compilers to optimise code, which makes pipelining simpler and more efficient.
现代CPU遵循复杂指令集计算机（CISC）或精简指令集计算机（RISC）的设计理念。CISC处理器（如x86）使用大量变长指令，常有单条指令完成多步操作的特点，这降低了内存占用但使流水线设计复杂化。RISC处理器（如ARM）采用一组较小的定长指令，通常每个时钟周期执行一条指令，依赖编译器优化代码，这使得流水线更简单、更高效。
2. Operating Systems and System Software | 操作系统与系统软件
An operating system (OS) acts as an interface between hardware and user applications, managing resources and providing essential services. Key functions include memory management (allocating RAM to processes, using paging and segmentation), processor scheduling (deciding which process runs next using algorithms such as round robin, shortest job first or multi‑level feedback queues), and handling interrupts from hardware or software.
操作系统（OS）充当硬件与用户应用程序之间的接口，管理资源并提供基本服务。其主要功能包括内存管理（通过分页和分段为进程分配内存）、处理器调度（使用轮转调度、最短作业优先或多级反馈队列等算法决定下一个运行的进程），以及处理来自硬件或软件的中断。
The OS also provides a file system for organising storage, manages input/output (I/O) devices through device drivers, and enforces security via user authentication and access rights. Utility software, such as disk defragmenters, backup tools and antivirus programs, sits alongside the OS to help maintain system health and performance.
操作系统还通过文件系统组织存储空间，利用设备驱动程序管理输入/输出（I/O）设备，并借助用户认证和访问权限实施安全策略。工具软件，如磁盘碎片整理程序、备份工具和防病毒软件，与操作系统协同工作，维护系统健康与性能。
3. Data Representation and Compression | 数据表示与压缩
All data inside a computer is stored as binary digits. Integers can be represented using unsigned binary, signed two’s complement (which simplifies subtraction by allowing the same addition circuitry), or fixed‑point representation for fractional values. Floating‑point numbers store very large or very small values using a mantissa and an exponent, but with limited precision that can lead to rounding errors.
计算机内部所有数据均以二进制形式存储。整数可以用无符号二进制、有符号二进制补码（通过相同的加法电路即可实现减法）或用于小数值的定点表示。浮点数采用尾数和指数存储极大或极小的值，但有限的精度可能导致舍入误差。
Character representation relies on ASCII (7 or 8 bits per character) and Unicode (providing a unique code point for characters of virtually all writing systems, with UTF‑8 being a variable‑width encoding that is backward‑compatible with ASCII). Images can be stored as bitmaps (a grid of pixels, each coded with a colour value, determined by colour depth and resolution) or as vectors (mathematical descriptions of shapes, which scale without loss of quality). Sound is sampled at regular intervals, with sample rate and bit depth determining the fidelity; the analogue signal is converted to digital via an ADC.
字符表示依赖于ASCII（每字符7或8位）和Unicode（为几乎所有书写系统的字符提供唯一码点，UTF‑8是一种可变长度编码，向后兼容ASCII）。图像既可以存储为位图（像素网格，每个像素用颜色值编码，由颜色深度和分辨率决定），也可以存储为矢量图（图形的数学描述，缩放无损）。声音以固定间隔采样，采样率和位深决定了保真度；模拟信号通过模数转换器（ADC）转换为数字信号。
To reduce storage and transmission demands, compression is used. Lossless compression (e.g., run‑length encoding, dictionary‑based methods like LZW) preserves every bit of original data; lossy compression (e.g., JPEG for images, MP3 for audio) discards perceptually less important information to achieve much higher compression ratios.
为减少存储和传输需求，常采用压缩技术。无损压缩（如游程编码、基于字典的LZW算法）完整保留原始数据；有损压缩（如JPEG图像、MP3音频）丢弃感知上较不重要的信息，以获得高得多的压缩比。
4. Computer Networks and Protocols | 计算机网络与协议
Networks connect autonomous computers to share resources and data. They are classified by scale: LAN (Local Area Network) covering a small geographic area, often using Ethernet and Wi‑Fi; WAN (Wide Area Network) linking LANs over large distances, typically using leased lines or the internet. Network topologies include star (each node connected to a central switch, failure of one cable does not affect others) and mesh (nodes interconnected, providing redundancy).
网络将自立的计算机连接起来，以共享资源和数据。按规模可分为：覆盖小范围地理区域的局域网（LAN），常使用以太网和Wi‑Fi；广域网（WAN）跨越长距离连接局域网，通常使用专线或互联网。网络拓扑结构包括星型（每个节点连接到中央交换机，单根电缆故障不影响其他节点）和网状（节点互连，提供冗余）。
The TCP/IP protocol stack underpins internet communication. At the application layer, protocols such as HTTP, HTTPS, FTP, SMTP and DNS define how services interact. The transport layer hosts TCP (reliable, connection‑oriented with acknowledgements and sequencing) and UDP (faster, connectionless, suitable for streaming). The internet layer routes packets using IP addresses, while the link layer handles the physical transmission of frames.
TCP/IP协议栈是互联网通信的基础。应用层协议如HTTP、HTTPS、FTP、SMTP和DNS定义了服务如何交互。传输层容纳TCP（可靠、面向连接，有确认和排序）和UDP（快速、无连接，适用于流媒体）。网际层使用IP地址路由数据包，链路层负责帧的物理传输。
Circuit switching establishes a dedicated physical path for the duration of a communication (e.g., traditional telephone networks), whereas packet switching breaks data into packets that may travel different routes and are reassembled at the destination, making more efficient use of network capacity.
电路交换在通信持续期间建立专用物理路径（如传统电话网络），而分组交换将数据分成分组，这些分组可能经由不同路由传输，在目的地重新组合，从而更高效地利用网络容量。
5. Network Security and Encryption | 网络安全与加密
Modern computer systems face a range of threats: malware (viruses, worms, trojans, ransomware), social engineering (phishing, pretexting), brute‑force attacks, denial‑of‑service (DoS) attacks, and SQL injection. Defences include firewalls (filtering traffic based on predefined rules), anti‑malware software, regular patch management, user access controls, and education to reduce human error.
现代计算机系统面临一系列威胁：恶意软件（病毒、蠕虫、特洛伊木马、勒索软件）、社会工程学攻击（钓鱼、伪造身份）、暴力破解、拒绝服务攻击（DoS）以及SQL注入。防御措施包括防火墙（基于预定义规则过滤流量）、反恶意软件、定期补丁管理、用户访问控制以及通过教育减少人为失误。
Encryption transforms plaintext into ciphertext to protect confidentiality. Symmetric encryption (e.g., AES, DES) uses a single shared key for both encryption and decryption; asymmetric encryption (e.g., RSA) employs a public key for encryption and a private key for decryption. This asymmetric model underpins digital signatures, whereby a hash of a message is encrypted with the sender’s private key, allowing recipients to verify both integrity and authenticity. Certificates issued by trusted Certificate Authorities (CAs) bind public keys to identities, forming the backbone of HTTPS.
加密将明文转换为密文以保护机密性。对称加密（如AES、DES）使用单一共享密钥同时进行加密和解密；非对称加密（如RSA）采用公钥加密、私钥解密。后者是数字签名的基础：发送方用自己的私钥加密报文散列值，接收方可验证完整性与真实性。由受信任的证书颁发机构（CA）签发的证书将公钥与身份绑定，构成了HTTPS的核心。
6. Computational Thinking and Problem Solving | 计算思维与问题解决
Computational thinking is a problem‑solving approach built on four cornerstones: decomposition (breaking a complex problem into smaller, manageable parts), pattern recognition (identifying similarities among problems to reuse solutions), abstraction (filtering out unnecessary detail to focus on what is important), and algorithm design (creating a step‑by‑step solution). This mindset is applicable far beyond programming.
计算思维是一种以分解（将复杂问题拆分为较小、易处理的部分）、模式识别（找出问题间的相似之处以复用方案）、抽象（滤除无关细节，聚焦关键要素）和算法设计（规划逐步解决方案）为基石的问题解决方法。这种思维远不止适用于编程领域。
The software development lifecycle provides a structured framework for turning a requirement into a reliable program: analysis (defining the problem and user needs), design (specifying data structures, user interface, and algorithms), implementation (coding, often using an iterative approach), testing (unit, integration, and acceptance testing with normal, boundary and erroneous data), and evaluation (assessing whether the solution meets the original requirements and identifying possible improvements). Good practice includes using version control, meaningful identifiers, and comprehensive commenting.
软件开发生命周期为将需求转变为可靠的程序提供了结构化框架：分析（定义问题与用户需求）、设计（明确数据结构、用户界面和算法）、实现（编码，常采用迭代方式）、测试（使用正常、边界和异常数据进行单元、集成及验收测试）以及评估（判断解决方案是否满足原定需求并识别改进空间）。良好实践包括使用版本控制、有意义的标识符和充分的注释。
7. Data Structures and Recursion | 数据结构与递归
Efficient programs rely on choosing the right data structure. Arrays and records provide contiguous storage; stacks (last‑in, first‑out) support operations like undo, backtracking and expression evaluation; queues (first‑in, first‑out) schedule processes or buffer data. Linked lists allow dynamic memory allocation but require sequential access. Trees, particularly binary search trees, store data hierarchically, enabling fast search, insertion and deletion when balanced. Graphs model relationships such as transport routes or social connections.
高效的程序依赖于选择合适的数据结构。数组和记录提供连续存储；栈（后进先出）支持撤销、回溯和表达式求值等操作；队列（先进先出）用于调度进程或缓冲数据。链表允许动态内存分配，但需顺序访问。树，特别是二叉搜索树，分层存储数据，平衡时能实现快速查找、插入和删除。图则模拟交通路线或社交关系等关联网络。
Many algorithms operate on these structures: tree traversal can be pre‑order, in‑order or post‑order; graph traversal includes depth‑first and breadth‑first search. Standard algorithms for adding, deleting and searching within each structure must be thoroughly understood, including the management of pointers in linked lists.
众多算法在这些结构上运行：树的遍历可为前序、中序或后序；图的遍历包括深度优先和广度优先搜索。必须彻底理解各结构中的添加、删除和搜索等标准算法，包括链表中指针的管理。
Recursion is a technique where a function calls itself to solve a smaller instance of the same problem. Every recursive algorithm must have a base case to terminate and a recursive step that moves towards it. While recursion can produce elegant solutions for problems like calculating factorials, generating Fibonacci sequences, or traversing trees, it may be less memory‑efficient than iteration due to stack frame overhead.
递归是一种函数通过调用自身来解决同类问题更小实例的技术。每个递归算法必须有一个终止的基准情形，以及一个逐步趋近基准情形的递归步骤。尽管递归能为计算阶乘、生成斐波那契数列或遍历树等问题产生简洁的方案，但由于栈帧的开销，其内存效率可能低于迭代。
8. Algorithm Analysis and Big O Notation | 算法分析与大O表示法
Analysing an algorithm requires understanding its time and space complexity – how the resource demands grow as the size of the input increases. Big O notation provides an upper bound on this growth, expressed as a function of n (the input size). Common complexities, from most efficient to least, include:
分析算法需要理解其时间与空间复杂度——即随着输入规模增长，资源需求如何扩大。大O表示法为这一增长趋势提供了一个上界，表示为输入规模 n 的函数。常见复杂度由高到低排序如下：
| Complexity | Notation | Example Algorithm |
|---|---|---|
| Constant | O(1) | Accessing an array element by index |
| Logarithmic | O(log n) | Binary search on a sorted array |
| Linear | O(n) | Simple loop through an array; linear search |
| Linearithmic | O(n log n) | Efficient sorting algorithms (merge sort, quick sort average case) |
| Quadratic | O(n²) | Bubble sort, insertion sort (worst case) |
| Exponential | O(2ⁿ) | Recursive calculation of Fibonacci (naive) |
Searching algorithms like binary search (O(log n)) rely on sorted data, while linear search (O(n)) works on unsorted lists. Sorting algorithms range from simple but inefficient (bubble, insertion) to divide‑and‑conquer methods like merge sort, which offers O(n log n) time but requires O(n) additional space. Algorithmic efficiency is critical when choosing a method for large datasets.
搜索算法如二分查找（O(log n)）依赖于已排序的数据，而线性查找（O(n)）适用于未排序列表。排序算法从简单但低效的（冒泡、插入）到分治法如归并排序，归并排序提供O(n log n)时间但需要O(n)额外空间。处理大规模数据集时，算法的效率至关重要。
9. Boolean Algebra and Logic Circuits | 布尔代数与逻辑电路
Boolean algebra deals with true/false values using operators AND (conjunction, ∧), OR (disjunction, ∨) and NOT (negation, ¬). These operators can be combined to form logical expressions, which are the foundation of digital circuits. Logic gates physically implement these operations: AND gate, OR gate, NOT gate, as well as NAND, NOR, XOR and XNOR. Truth tables systematically list all possible input combinations and their corresponding outputs.
布尔代数处理真/假值，使用 AND（合取，∧）、OR（析取，∨）和 NOT（否定，¬）运算符。这些运算符可组合形成逻辑表达式，它是数字电路的基础。逻辑门物理实现这些运算：与门、或门、非门，以及与非、或非、异或和同或门。真值表系统列出所有可能的输入组合及其对应的输出。
Expressions can be simplified using Boolean identities (e.g., De Morgan’s laws: (A+B)’ = A’·B’, and (A·B)’ = A’+B’; distributive, commutative, associative and absorption laws). Simplification reduces the number of gates needed, lowering cost and power consumption. Karnaugh maps (K‑maps) provide a graphical method for simplifying expressions with up to four variables by grouping adjacent 1s.
布尔表达式可以使用布尔恒等式进行化简（例如德摩根律：(A+B)’ = A’·B’，以及 (A·B)’ = A’+B’；分配律、交换律、结合律和吸收律）。化简可减少所需门数，降低成本与功耗。卡诺图（K‑map）为不超过四个变量的表达式化简提供图形化方法，通过圈定相邻的“1”实现。
Adders demonstrate how logic gates are combined to perform arithmetic. A half adder takes two input bits (A, B) and produces a sum (S = A XOR B) and a carry (C = A AND B). A full adder includes a third carry‑in input, allowing multiple adders to be cascaded for multi‑bit addition, often with a look‑ahead carry unit to improve speed.
加法器展示了如何组合逻辑门来执行算术运算。半加器以两个输入位（A, B）产生和（S = A XOR B）与进位（C = A AND B）。全加器增加第三个进位输入，使得多个加法器可级联实现多位加法，并常采用先行进位单元提升速度。
10. Legal, Ethical and Environmental Impacts | 法律、道德与环境影响
Computer science professionals must operate within a framework of legislation designed to protect individuals and society. The Data Protection Act 2018 (incorporating GDPR) governs how personal data is collected, processed and stored, giving individuals rights over their data. The Computer Misuse Act 1990 criminalises unauthorised access to computer material, including hacking and the creation of malware. The Copyright, Designs and Patents Act 1988 protects the intellectual property rights of software developers and digital content creators. The Regulation of Investigatory Powers Act 2000 (RIPA) sets out the conditions under which public bodies may monitor communications and demand access to encrypted information.
计算机科学从业者必须在旨在保护个人和社会的法律框架内运作。《2018年数据保护法》（包含《通用数据保护条例》GDPR）规定了个人数据的收集、处理和存储方式，赋予个人对其数据的权利。《1990年计算机滥用法》将未经授权访问计算机材料的行为（包括黑客攻击和制作恶意软件）定为刑事犯罪。《1988年版权、设计及专利法》保护软件开发者与数字内容创作者的知识产权。《2000年调查权规范法》（RIPA）明确了公共机构在何种条件下可监视通信并要求获取加密信息。
Ethical considerations extend beyond the law. Topics such as the use of artificial intelligence in decision‑making, algorithmic bias, the digital divide, and the responsibilities of social media platforms require careful evaluation of benefits and harms. Environmental impacts of computing include the energy consumption of data centres, e‑waste from discarded devices, and the carbon footprint of cryptocurrency mining. Exam questions often ask you to weigh these issues and propose responsible solutions.
道德考量远不止于法律。诸如人工智能在决策中的运用、算法偏见、数字鸿沟以及社交媒体平台的责任等议题，都需要审慎评估其利弊。计算机对环境的影响包括数据中心的能源消耗、废弃设备产生的电子垃圾以及加密货币挖矿的碳足迹。考试题目常要求你权衡这些问题并提出负责任的解决方案。
11. Databases and SQL | 数据库与SQL
A relational database organises data into tables (relations) consisting of rows (tuples) and columns (attributes). Each table has a primary key that uniquely identifies a record. Relationships between tables are implemented using foreign keys. Normalisation (1NF, 2NF, 3NF) is a design technique that eliminates data redundancy and dependency problems by ensuring that attributes are dependent on the key, the whole key, and nothing but the key.
关系数据库将数据组织成由行（元组）和列（属性）组成的表（关系）。每个表设有一个唯一标识记录的主键。表之间通过外键实现关联。规范化（1NF、2NF、3NF）是一种设计技术，通过确保属性依赖于键、整个键且只依赖于键，消除数据冗余和依赖问题。
Structured Query Language (SQL) is the standard language for interacting with relational databases. Core operations include: SELECT field1, field2 FROM table WHERE condition ORDER BY field; inserting new data with INSERT INTO table (fields) VALUES (values); updating records using UPDATE table SET field = value WHERE condition; and removing data with DELETE FROM table WHERE condition. Joins (INNER JOIN, LEFT JOIN) combine rows from multiple tables based on a related column.
结构化查询语言（SQL）是与关系数据库交互的标准语言。核心操作包括：SELECT 字段1, 字段2 FROM 表 WHERE 条件 ORDER BY 字段；使用 INSERT INTO 表 (字段) VALUES (值) 插入新数据；使用 UPDATE 表 SET 字段 = 值 WHERE 条件更新记录；以及使用 DELETE FROM 表 WHERE 条件删除数据。连接（INNER JOIN、LEFT JOIN）根据相关列合并来自多个表的行。
Database management systems (DBMS) also handle transactions with ACID properties (Atomicity, Consistency, Isolation, Durability) to ensure data integrity, especially in multi‑user environments.
数据库管理系统（DBMS）还通过具备ACID特性（原子性、一致性、隔离性、持久性）的事务处理来保证数据完整性，这在多用户环境中尤为重要。
Published by TutorHao | Computer Science Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导