📚 A-Level Edexcel Computer Science: Ultimate Revision Checklist | Edexcel A-Level 计算机：期末复习提纲
This end-of-term revision checklist covers all essential topics for the Edexcel A-Level Computer Science specification. Use it to structure your final review, identify knowledge gaps, and build confidence for both Paper 1 (Principles of Computer Science) and Paper 2 (Application of Computational Thinking). Each section pairs concise English explanations with their Chinese equivalents, ensuring you can articulate concepts clearly in any context.
这份期末复习提纲涵盖 Edexcel A-Level 计算机科学考试大纲的全部核心主题。用它来规划期末总复习、定位知识盲区，并为 Paper 1（计算机科学原理）和 Paper 2（计算思维应用）建立信心。每个小节都以简洁的英文解释搭配对应的中文表述，确保你能在任何语境下清晰阐述概念。
1. Computational Thinking | 计算思维
Decomposition means breaking a complex problem into smaller, more manageable parts that can be solved independently and then combined.
分解是指将一个复杂问题拆分成更小、更易管理的部分，这些部分可以独立求解，最后再整合起来。
Pattern recognition involves identifying similarities, trends, or recurring elements within a problem or between different problems, which allows the reuse of known solutions.
模式识别是指在问题内部或不同问题之间发现相似性、趋势或重复出现的元素，从而能够复用已知的解决方案。
Abstraction is the process of filtering out unnecessary details to focus on the essential characteristics that define the problem, creating a simplified model.
抽象是指滤除不必要的细节，聚焦于定义问题的核心特征，从而建立一个简化模型。
Algorithmic thinking is the ability to design a step-by-step solution or set of rules that can be executed by a computer to solve a specific problem.
算法思维是指设计分步骤的解决方案或规则集的能力，该方案可由计算机执行以解决特定问题。
2. Programming Concepts | 编程概念
Sequence, selection, and iteration are the three fundamental control structures that underpin all procedural programming languages.
顺序、选择和迭代是支撑所有过程式编程语言的三种基本控制结构。
Variables store data values that can change during program execution, whereas constants hold values that remain fixed.
变量存储在程序执行期间可以改变的数据值，而常量保存的值保持不变。
Local variables are declared inside a subroutine and exist only during its execution, while global variables are accessible throughout the entire program.
局部变量在子程序内部声明，仅在其执行期间存在；全局变量则可在整个程序中访问。
Parameter passing can be by value, where a copy of the data is sent, or by reference, where the actual memory address is passed and the original data can be modified.
参数传递可以是传值，即发送数据的副本，也可以是传引用，即传递实际的内存地址，从而可能修改原始数据。
Recursion is a technique where a subroutine calls itself with a modified argument, progressing towards a base case that stops the chain.
递归是一种技术，子程序用经过修改的参数调用自身，逐步向停止调用链的基准情形推进。
3. Data Structures | 数据结构
An array is a static, ordered collection of elements, all of the same data type, accessed via an index.
数组是静态的有序元素集合，所有元素具有相同数据类型，通过索引访问。
A linked list consists of nodes, each containing data and a pointer to the next node, enabling dynamic memory usage and efficient insertion/deletion.
链表由节点组成，每个节点包含数据和指向下一节点的指针，支持动态内存使用以及高效的插入和删除。
A stack is a LIFO (Last In, First Out) abstract data type with operations push (add to top) and pop (remove from top).
栈是一种后进先出的抽象数据类型，主要操作包括压入（向顶部添加）和弹出（从顶部移除）。
A queue is a FIFO (First In, First Out) structure with enqueue (add to rear) and dequeue (remove from front).
队列是一种先进先出的结构，拥有入队（向尾部添加）和出队（从头部移除）操作。
A binary tree is a hierarchical structure where each node has at most two children; traversal includes pre-order, in-order, and post-order.
二叉树是一种层次结构，每个节点最多有两个子节点；遍历方式包括前序、中序和后序。
A hash table maps keys to values using a hash function, offering average-time O(1) lookup; collisions are handled by techniques like chaining or open addressing.
哈希表使用哈希函数将键映射到值，平均查找时间复杂度为 O(1)；碰撞通过链表法或开放寻址等技术处理。
4. Algorithms and Complexity | 算法与复杂度
Linear search checks each element sequentially until the target is found, with a time complexity of O(n).
线性搜索依次检查每个元素直到找到目标，时间复杂度为 O(n)。
Binary search requires a sorted list, repeatedly dividing the search interval in half, achieving O(log n) time complexity.
二分搜索需要一个已排序的列表，反复将搜索区间一分为二，时间复杂度为 O(log n)。
Bubble sort repeatedly swaps adjacent out-of-order elements, bubble sort has O(n²) worst-case complexity.
冒泡排序反复交换相邻的未排序元素，最坏情况复杂度为 O(n²)。
Merge sort uses a divide-and-conquer strategy, splitting the list and merging sorted halves, with O(n log n) time complexity.
归并排序采用分治策略，分割列表并合并已排序的半部分，时间复杂度为 O(n log n)。
Dijkstra’s algorithm finds the shortest path in a weighted graph, provided all edge weights are non-negative.
Dijkstra 算法在所有边权重非负的情况下找到加权图中的最短路径。
The A* algorithm improves on Dijkstra by using a heuristic to guide the search towards the goal, often making it more efficient for pathfinding.
A* 算法通过使用启发式函数引导搜索朝向目标来改进 Dijkstra 算法，通常在寻路中更高效。
5. Data Representation | 数据表示
Unsigned binary uses positional notation (bits represent powers of 2) to represent non-negative integers.
无符号二进制使用位置记数法（位表示2的幂）来表示非负整数。
Two’s complement is the standard method for representing signed integers; the most significant bit carries a negative weight.
补码（二进制补码）是表示有符号整数的标准方法；最高位携带负权重。
Floating-point representation stores a number in the form mantissa x 2^exponent, following the mantissa and exponent format to balance range and precision.
浮点数表示法以“尾数×2^指数”的形式存储数字，遵循尾数和指数的格式来平衡范围和精度。
ASCII uses 7 bits to represent 128 characters, while Unicode (e.g., UTF-8) can encode characters from virtually all writing systems.
ASCII 使用7位表示128个字符，而 Unicode（如 UTF-8）可以编码几乎所有书写系统的字符。
Bitmap images are stored as a grid of pixels, with colour depth determining the bits per pixel; higher resolution and depth increase file size.
位图图像以像素网格存储，色彩深度决定每像素的位数；分辨率和深度越高，文件体积越大。
Sound is sampled at a given rate and bit depth; the Nyquist theorem states the sampling rate must be at least twice the highest frequency to avoid aliasing.
声音以给定的采样率和位深度进行采样；奈奎斯特定理指出，采样率必须至少是最高频率的两倍以避免混叠。
6. Computer Systems | 计算机系统
A system bus consists of the data bus (transfers actual data), the address bus (specifies memory location), and the control bus (carries command signals).
系统总线由数据总线（传输实际数据）、地址总线（指定内存位置）和控制总线（传送命令信号）组成。
The CPU fetches instructions from memory, decodes them to determine required actions, and executes them, following the fetch-decode-execute cycle.
CPU 从内存中取出指令，进行译码以确定所需操作，然后执行，遵循取指-译码-执行周期。
Cache memory is a small, high-speed memory placed near the CPU to store frequently accessed data, significantly reducing average access time.
高速缓存是靠近 CPU 的小容量高速存储器，用于存储频繁访问的数据，可显著降低平均访问时间。
Von Neumann architecture uses a single shared memory for both instructions and data, typically leading to a bottleneck, whereas Harvard architecture separates them into distinct memories.
冯·诺依曼体系结构使用单一的共享内存存放指令和数据，通常会造成瓶颈；而哈佛体系结构将指令和数据存放在独立的内存中。
Pipelining allows the CPU to begin executing the next instruction before the current one finishes, improving throughput but introducing hazards.
流水线技术允许 CPU 在当前指令完成前就开始执行下一条指令，从而提高了吞吐量，但会引入流水线冲突。
7. Computer Architecture | 计算机体系结构
The ALU (Arithmetic Logic Unit) performs arithmetic and logical operations; the CU (Control Unit) directs the operation of the processor.
算术逻辑单元（ALU）执行算术和逻辑运算；控制单元（CU）指挥处理器的操作。
Registers such as PC (Program Counter), MAR (Memory Address Register), MDR (Memory Data Register), and CIR (Current Instruction Register) hold small amounts of data needed during instruction execution.
如程序计数器（PC）、内存地址寄存器（MAR）、内存数据寄存器（MDR）和当前指令寄存器（CIR）等寄存器用于存放指令执行期间需要的少量数据。
RISC processors use a small, highly optimised set of simple instructions, often executing one instruction per clock cycle; CISC processors have a larger and more complex instruction set capable of multi-cycle tasks.
RISC 处理器使用一组小而高度优化的简单指令，通常每个时钟周期执行一条指令；CISC 处理器拥有更大、更复杂的指令集，能完成多周期任务。
GPU architectures contain thousands of small cores designed for parallel processing, making them suitable for graphics rendering and data-intensive calculations.
GPU 架构包含成千上万个为并行处理设计的小核心，这使它们非常适合图形渲染和数据密集型计算。
Input and output devices connect via I/O controllers; methods include memory-mapped I/O and interrupt-driven I/O to avoid wasting CPU cycles.
输入和输出设备通过 I/O 控制器连接；方法包括内存映射 I/O 和中断驱动 I/O，以避免浪费 CPU 周期。
8. Networks and Internet | 网络与互联网
A LAN (Local Area Network) covers a small geographical area and typically uses Ethernet or Wi-Fi, while a WAN (Wide Area Network) spans large distances using leased lines or satellite links.
LAN（局域网）覆盖小范围地理区域，通常使用以太网或 Wi-Fi；WAN（广域网）跨越长距离，使用专线或卫星链路。
The TCP/IP protocol stack consists of the Application, Transport, Internet, and Link layers, each handling different tasks from data formatting to physical transmission.
TCP/IP 协议栈由应用层、传输层、互联网层和链路层组成，每一层处理从数据格式化到物理传输的不同任务。
Packet switching breaks data into packets that are routed independently over a shared network; routers use destination IP addresses to forward packets.
分组交换将数据拆分为数据包，这些数据包在共享网络上独立选路；路由器使用目的 IP 地址转发数据包。
HTTP/HTTPS are application-layer protocols for web transfer; HTTPS adds SSL/TLS encryption to secure the communication.
HTTP/HTTPS 是用于网页传输的应用层协议；HTTPS 增加了 SSL/TLS 加密以保护通信安全。
Network security threats include malware, phishing, DoS attacks, and man-in-the-middle attacks; defences include firewalls, encryption, and user awareness training.
网络安全威胁包括恶意软件、网络钓鱼、拒绝服务攻击和中间人攻击；防御措施包括防火墙、加密和用户意识培训。
9. Databases and SQL | 数据库与 SQL
A relational database organises data into tables (relations) with rows (tuples) and columns (attributes), linked by primary and foreign keys to minimise redundancy.
关系型数据库将数据组织成具有行（元组）和列（属性）的表（关系），通过主键和外键关联以减少冗余。
Normalisation is the process of structuring data to reduce data anomalies; it involves progressing through normal forms (1NF, 2NF, 3NF) by removing partial and transitive dependencies.
规范化是构建数据结构以减少数据异常的过程；它通过消除部分依赖和传递依赖，依次达到第一、第二和第三范式。
SQL (Structured Query Language) includes DDL for defining schema, DML for manipulating data, and DCL for controlling access.
SQL（结构化查询语言）包括用于定义模式的 DDL、用于操作数据的 DML 和用于控制访问的 DCL。
A simple SELECT statement retrieves data using clauses like FROM, WHERE, ORDER BY, and GROUP BY; JOIN combines rows from multiple tables based on a related column.
简单的 SELECT 语句使用 FROM、WHERE、ORDER BY 和 GROUP BY 等子句检索数据；JOIN 基于相关列合并多个表中的行。
10. Ethical, Legal, and Environmental Issues | 伦理、法律与环境问题
The Data Protection Act governs the collection, processing, and storage of personal data, giving individuals rights over how their data is used.
数据保护法规范个人数据的收集、处理和存储，赋予个人对其数据使用方式的权利。
The Computer Misuse Act criminalises unauthorised access to computer material, unauthorised modification, and creating/distributing malware.
计算机滥用法将未经授权访问计算机材料、未经授权修改以及制造/传播恶意软件定为犯罪行为。
Copyright law protects original works of authorship, including software and digital content; plagiarism is presenting someone else’s work as your own.
版权法保护原创作品，包括软件和数字内容；抄袭是指将他人的作品当作自己的展示。
Environmental concerns include energy consumption of data centres and e-waste disposal; sustainable computing promotes energy-efficient hardware and responsible recycling.
环境问题包括数据中心的能耗和电子垃圾处理；可持续计算提倡节能硬件和负责任的回收。
Artificial intelligence raises ethical questions around bias, accountability, job displacement, and the need for transparent decision-making systems.
人工智能引发了有关偏见、问责、就业替代以及透明决策系统需求的伦理问题。
11. Problem Solving and Programming Project (NEA) Recap | 问题解决与编程项目回顾
The Non-Exam Assessment requires you to analyse a problem, design a solution, develop a complex program, and evaluate it against success criteria.
非考试评估要求你分析一个问题，设计解决方案，开发一个复杂的程序，并根据成功标准对程序进行评价。
Thorough analysis includes identifying stakeholders, defining requirements, and modelling the system using tools like data flow diagrams or UML.
彻底的分析包括识别利益相关者、定义需求，以及使用数据流图或 UML 等工具对系统建模。
Testing must cover normal, boundary, and erroneous data; alpha and beta testing ensure the program meets user needs in real environments.
测试必须覆盖正常、边界和错误数据；alpha 和 beta 测试确保程序在真实环境中满足用户需求。
An evaluation critically reflects on the solution’s effectiveness, highlighting what went well and what could be improved, supported by evidence.
评价应批判性地反思解决方案的有效性，借助证据指出哪些方面成功、哪些可以改进。
12. Exam Tips and Common Pitfalls | 考试技巧与常见陷阱
Read the command words carefully: ‘state’, ‘describe’, ‘explain’, and ‘evaluate’ require increasingly detailed responses.
仔细阅读指令词：’state’（陈述）、’describe’（描述）、’explain’（解释）和 ‘evaluate’（评价）要求越来越详细的回答。
When tracing algorithms, work through step by step and show values of all variables at each stage to secure full marks.
在追踪算法时，要逐步执行，并在每个阶段显示所有变量的值，以确保拿到满分。
For extended writing questions, structure your answer using the point–example–explanation pattern and relate it back to the scenario.
对于扩展写作题，使用“观点-示例-解释”的模式组织答案，并联系题目情境。
Manage your time: Paper 1 is 2 hours for 75 marks; spend about 1 minute per mark and reserve time to check calculations and logical errors.
合理安排时间：Paper 1 为 75 分 2 小时，大约每分钟完成 1 分的题目，并留出时间检查计算和逻辑错误。
Ensure programming syntax used in pseudocode is consistent; declare variables, use indentation, and include meaningful comments where required.
确保伪代码中使用的编程语法保持一致；声明变量，使用缩进，并在需要时添加有意义的注释。
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