📚 Year 13 Cambridge Computer Science: Unit Test Mock Paper Walkthrough | 剑桥计算机单元测试模拟卷解析
This walkthrough breaks down a typical Year 13 Cambridge Computer Science unit test, revealing the key concepts and step-by-step reasoning required for top marks. The mock paper is designed to mirror the depth and style of real 9618 A2 assessments, covering data representation, processor architecture, networking, databases, algorithms, OOP, assembly and security. Each question is unpacked with examiner-style commentary to help you master both theory and exam technique.
本解析详细拆解了一套典型的 Year 13 剑桥计算机单元测试模拟卷，揭示了获取高分的核心概念与逐步推理过程。该模拟卷紧密对标 9618 A2 真题的深度与风格，涵盖数据表示、处理器架构、网络、数据库、算法、面向对象编程、汇编与安全。每道题目都配有考官式点评，帮助你同时掌握理论知识与应试技巧。
1. Overview of the Mock Paper | 模拟试卷概览
This mock paper contains eight structured questions and is designed to be completed in 90 minutes. It mixes short-answer, calculation, explanation and design items, mirroring the balance of recall and application found in Cambridge exams. The topics span both theoretical foundations and computational thinking, ensuring full coverage of Year 13 learning outcomes. A common emphasis is placed on precision, use of technical vocabulary and the ability to justify design decisions.
本模拟卷包含八道结构题，设计答题时间为 90 分钟。试卷融合了简答、计算、解释与设计题，再现了剑桥考试中记忆与应用并重的平衡性。主题兼顾理论基础与计算思维，全面覆盖 Year 13 学习目标。试卷格外看重精确性、专业术语的运用以及论证设计决策的能力。
2. Q1 – Data Representation & Binary Arithmetic | 数据表示与二进制运算
The first question asks: “Represent the decimal integer -73 using an 8-bit two’s complement representation. Show all steps.” To solve this, begin by converting the positive magnitude 73 to binary: 73 = 64 + 8 + 1, which yields 0100 1001 in 8 bits. Two’s complement negation requires inverting the bits to get 1011 0110 and then adding 1, producing the final result 1011 0111. A common error is to omit leading zeros or to mishandle the addition of 1 when the inverted bits end in 0. Another part of the question may extend to normalised floating-point representation, where the mantissa and exponent must be carefully aligned.
第一题要求：「用 8 位补码表示十进制整数 -73，展示所有步骤。」解答时首先将正数 73 转为二进制：73 = 64 + 8 + 1，得到 8 位的 0100 1001。补码求负需将各位取反为 1011 0110，再加 1，最终结果为 1011 0111。常见错误是遗漏前导零，或在取反后末位为 0 时进位处理不当。题目可能延伸至规格化浮点表示，需仔细对准尾数和阶码。
3. Q2 – Processor Architecture & The FDE Cycle | 处理器架构与取指-解码-执行周期
Question 2 typically provides a diagram of Von Neumann architecture and asks “Describe the Fetch–Decode–Execute (FDE) cycle, naming the registers involved.” A full answer must trace the sequence: the Program Counter (PC) holds the address of the next instruction, which is copied to the Memory Address Register (MAR); the instruction is fetched into the Memory Data Register (MDR) and then the Current Instruction Register (CIR). The Control Unit decodes the opcode, and if data is needed, the MAR/MDR are reused to fetch or store operands. The ALU executes the operation, and the PC is incremented unless a branch occurs. The Accumulator and Status Register may also be referenced for results.
第二题通常会给出冯·诺依曼架构图，要求「描述取指-解码-执行周期，并说出涉及哪些寄存器」。完整答案需追踪流程：程序计数器 (PC) 存放下一条指令地址，该地址被复制到内存地址寄存器 (MAR)；指令取出至内存数据寄存器 (MDR) 再进入当前指令寄存器 (CIR)。控制单元对操作码译码，若需数据，则复用 MAR/MDR 取操作数或存结果。ALU 执行运算，PC 自增（除非发生分支）。累加器和状态寄存器也可提及。
4. Q3 – Networking & The TCP/IP Protocol Suite | 网络与 TCP/IP 协议栈
This question challenges students to “Explain how the TCP/IP protocol suite supports reliable data transmission between two hosts on different networks.” A model answer describes the four-layer model: Application layer produces data, Transport layer (TCP) segments it, adds sequence numbers and implements three-way handshakes, acknowledgements and retransmissions for reliability. The Internet layer (IP) attaches source and destination IP addresses and routes packets across routers, while the Network Access layer frames data for transmission over the physical medium. Port numbers allow multiplexing; checksums detect corruption. Students often confuse the roles of TCP and IP or forget to mention that IP itself is connectionless and unreliable, relying on TCP for reliability.
本题要求学生「解释 TCP/IP 协议栈如何支持两台位于不同网络的主机进行可靠数据传输」。参考答案会描述四层模型：应用层生成数据，传输层 (TCP) 对数据分段，添加序列号，并通过三次握手、确认与重传实现可靠性。网络层 (IP) 附加源与目的 IP 地址，在路由器间转发数据包；网络访问层将数据成帧以便在物理介质上传输。端口号用于多路复用，校验和用于检测错误。学生常混淆 TCP 与 IP 的职责，或忘记提及 IP 本身是无连接且不可靠的，可靠传输依赖 TCP 实现。
5. Q4 – Relational Databases & Normalisation | 关系数据库与规范化
In Question 4, a table of student enrolments is given with columns StudentID, StudentName, CourseID, CourseTitle, TutorID, TutorName. The task is to normalise this unnormalised form (UNF) to Third Normal Form (3NF), justifying each step. The solution first moves to 1NF by ensuring atomic values and removing repeating groups. 2NF requires the removal of partial dependencies: non-key attributes must depend on the whole primary key; here, StudentName depends only on StudentID, and CourseTitle/TutorID only on CourseID. The table is decomposed into Student, Course, and Enrolment relations. For 3NF, transitive dependencies are eliminated: TutorName depends on TutorID, not directly on CourseID, so a separate Tutor table is created. Explicitly identifying candidate keys and functional dependencies earns marks.
第四题给出学生选课表，包含 StudentID、StudentName、CourseID、CourseTitle、TutorID、TutorName。要求将该非规范化形式 (UNF) 规范化至第三范式 (3NF)，并说明每步理由。解法先将数据转为 1NF：确保值为原子且无重复组。2NF 需消除部分依赖：非键属性必须依赖于整个主键；此处 StudentName 仅依赖于 StudentID，CourseTitle/TutorID 仅依赖于 CourseID，分解为学生、课程和选课关系。3NF 消除传递依赖：TutorName 依赖于 TutorID 而非直接依赖 CourseID，故需独立的导师表。明确识别候选键与函数依赖是得分关键。
6. Q5 – Algorithm Analysis & Big O Notation | 算法分析与大 O 表示法
Question 5 presents pseudocode for a linear search and a binary search. Students must “determine and justify the worst-case time complexity in Big O notation for each algorithm, comparing their suitability for a dataset of 1 000 000 sorted elements.” Linear search has O(n) because, in the worst case, every element must be examined. Binary search halves the search space each iteration, leading to O(log n). Therefore, for n = 1 000 000, linear search requires up to 1 000 000 comparisons, while binary search requires at most about 20 comparisons. Answer should also note that binary search requires a sorted array and random-access memory. Often, candidates lose marks by confusing Big O with exact counts or by failing to mention the assumptions (e.g., sort overhead).
第五题给出线性搜索与二分搜索的伪代码。学生需「确定并论证每种算法的最坏情况时间复杂度（用大 O 表示法），并比较它们对 1 000 000 个已排序元素的适用性」。线性搜索为 O(n)，因为最坏情况下需检查每个元素。二分搜索每次迭代将搜索空间减半，故为 O(log n)。因此对于 n = 1 000 000，线性搜索最多需 1 000 000 次比较，而二分搜索最多约 20 次。答案还应指出二分搜索要求数组已排序且内存可随机访问。考生常因混淆大 O 与精确次数，或未说明假设条件（如排序开销）而失分。
7. Q6 – Object-Oriented Programming & Polymorphism | 面向对象编程与多态
This question provides a UML class diagram with a superclass Vehicle and subclasses Car, Bicycle. It asks “Explain how polymorphism can be implemented using method overriding and illustrate with a code snippet in your chosen OOP language.” Polymorphism means ‘many forms’ and allows objects of different subclasses to be treated uniformly through a common interface. Overriding lets a subclass provide a specific implementation of a method already defined in its superclass, e.g., move(). When a Vehicle reference calls move(), the actual method executed is determined at runtime by the object’s type. This supports extensibility and maintainability. A concise example in pseudocode or Python/Java style should show virtual or inherited methods, dynamically bound.
本题提供 UML 类图，包含超类 Vehicle 与子类 Car、Bicycle，要求「解释如何通过方法重写实现多态，并用你选择的 OOP 语言给出代码片段」。多态意为「多种形态」，允许不同子类的对象通过公共接口被统一处理。重写允许子类提供超类已定义方法的具体实现，例如 move()。当 Vehicle 引用调用 move() 时，实际执行的方法在运行时根据对象类型动态绑定。这提高了可扩展性和可维护性。简明的伪代码或 Python/Java 示例应展示虚方法或继承方法的动态绑定。
8. Q7 – Assembly Language & Addressing Modes | 汇编语言与寻址模式
In Question 7, students are given a hypothetical instruction set with immediate, direct and indirect addressing modes. The task: “Write an assembly program using the available instructions to load two numbers from memory locations 50 and 51, add them, and store the result in location 52.” A solution sequence might be: LDA 50 (direct), MOV B, A, LDA 51, ADD B, STA 52. An accompanying sub-question asks to “compare the advantages of indirect addressing over direct addressing.” Indirect addressing allows the address to be changed dynamically at runtime without modifying the instruction itself, supporting data structures like arrays and pointers. Candidates must label each mnemonic and trace register values to earn full marks.
第七题给出一个含有立即、直接和间接寻址模式的假想指令集。任务：「用可用指令写一段汇编程序，从内存地址 50 和 51 中加载两个数，相加后将结果存入地址 52」。解答序列可为：LDA 50（直接）、MOV B, A、LDA 51、ADD B、STA 52。附加问题要求「比较间接寻址相较直接寻址的优势」。间接寻址允许地址在运行时动态改变，无需修改指令本身，从而支持数组和指针等数据结构。考生需标注每条助记符并追踪寄存器值，方可得到满分。
9. Q8 – Security, Encryption & Hashing | 安全、加密与哈希
The final question examines cryptography fundamentals: “Distinguish between symmetric and asymmetric encryption, and explain how a digital signature uses hashing and asymmetric encryption to provide authentication and integrity.” Symmetric encryption (e.g., AES) uses a single shared key, making it fast but requiring secure key distribution. Asymmetric encryption (e.g., RSA) uses a key pair: the public key encrypts, the private key decrypts. A digital signature is created by hashing the plaintext (e.g., SHA-256) and then encrypting the hash with the sender’s private key. The recipient decrypts this signature with the sender’s public key and compares the hash with a newly computed hash; a match confirms the message was not altered (integrity) and originated from the owner of the private key (authentication). Weaknesses related to key length and man-in-the-middle attacks can also be discussed.
最后一道题考察密码学基础：「区分对称与非对称加密，并解释数字签名如何利用哈希和非对称加密实现身份验证和完整性。」对称加密（如 AES）使用单个共享密钥，速度快但需要安全的密钥分发。非对称加密（如 RSA）使用密钥对：公钥加密，私钥解密。数字签名制作时先将明文哈希（如 SHA-256），再用发送方私钥加密该哈希。接收方用发送方公钥解密签名，并与重新计算的哈希比对；若一致，则可确认消息未被篡改（完整性）且确实来自私钥持有者（身份验证）。还可讨论密钥长度和中间人攻击等弱点。
10. Common Pitfalls & Tips | 常见错误与技巧
Across the mock, common errors include forgetting leading zeros in binary representations, confusing the roles of MAR and MDR, failing to state the distinction between a protocol and an interface, mixing up functional dependency with natural language “dependency”, overusing absolute counts in Big O explanations, and treating ‘overloading’ as ‘overriding’. To avoid these, always double-check bit-width requirements, label every register in cycle traces, learn official definitions, and practise functional dependency identification on sample tables. For code topics, draw tiny class diagrams and trace method calls to visualise dynamic binding.
整张模拟卷中，常见错误包括：二进制表示中遗忘前导零，混淆 MAR 与 MDR 的职能，未区分协议与接口，将函数依赖误作自然语义上的依赖，大 O 解释中滥用绝对次数，以及将重载与重写混为一谈。为避免这些问题，务必检查位宽要求、标注周期追踪中的每个寄存器、牢记正式定义，并在示例表上练习识别函数依赖。对于代码主题，可绘制微型类图并追踪方法调用，以可视化动态绑定。
11. Marking Scheme Insights | 评分方案详解
Examiners award marks for explicit keywords (“Program Counter”, “register”, “increment”), correct step-by-step calculations, and logical justification. For a 6-mark explanation question, typically 2 marks are allocated to recall of factual correct terms, 2 to sequence or cause-and-effect reasoning, and 2 to application or example. If a question says “describe”, giving only bullet points may lose a coherence mark; a flowing narrative is safer. In calculation questions, even if the final answer is wrong, method marks are earned by showing intermediate steps. Always align your answer length with the mark tariff — a 1-mark ‘state’ question needs a short, precise answer.
考官根据关键词（如“程序计数器”“寄存器”“自增”）、正确的分步计算以及逻辑论证给分。一道 6 分的解释题，通常 2 分给准确的事实术语记忆，2 分给顺序或因果推理，2 分给应用或举例。若题目要求 “describe”，仅用要点列表可能丢失连贯性分数；流畅的叙述更为稳妥。在计算题中，即使最终答案错误，展示中间步骤也能获得方法分。答案篇幅务必与分值匹配——1 分的 “state” 题只需简短精准的回答。
12. Final Revision Advice | 最终复习建议
Focus revision on high-weight topics: processor fundamentals, data representation, networking layers, normalisation, Big O analysis, OOP principles and cryptography. Use past paper walkthroughs to internalise command words like ‘explain’, ‘describe’, ‘compare’. Create flashcards for register roles, protocol layers and normal forms. Time yourself writing full answers under exam conditions, and review model answers to see where marks are gained. Finally, maintain a concise glossary of technical terms in both English and your study language, as bilingual clarity strengthens recall in Cambridge assessments that demand precise definitions.
复习应聚焦于权重较高的主题：处理器基础、数据表示、网络分层、规范化、大 O 分析、OOP 原则与密码学。利用真题解析内化 “explain”“describe”“compare” 等指令词。为寄存器角色、协议层和范式制作闪卡。在考试条件下限时书写完整答案，并对照参考答案找出得分点。最后，维护一份英汉双语的简洁术语表，因为双语清晰度能增强记忆，满足剑桥考试对精确定义的要求。
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Cancel reply