Cambridge International AS Level Computer Science (9618) –

Revision Notes

Chapter 1: Information Representation

• Binary and Number Systems: Data is stored in binary (base-2);

typical units include bits (0/1) and bytes (8 bits). Understand number
bases: binary, decimal (denary), hexadecimal, and BCD (Binary-Coded
Decimal) . Know how to convert between bases and perform binary
addition/subtraction (watch for overflow when results exceed fixed bit
size) .

• Binary Prefixes: Binary prefixes (kibi = 2^10, mebi = 2^20,

gibi = 2^30, tebi = 2^40, etc.) differ from decimal prefixes (kilo = 10^3,
mega = 10^6, etc.)  . For example, 1 kibibyte = 1024 bytes, whereas 1
kilobyte (decimal) = 1000 bytes. These IEC prefixes were introduced to
avoid confusion between binary and decimal quantities .

• Signed Number Representation: Understand sign

representation: in two’s complement, negative numbers are encoded by
inverting bits and adding 1, whereas in one’s complement negatives invert
bits. Two’s complement is widely used because it has a single zero and
simpler arithmetic.

• Character Encoding: Text characters are stored using character

sets. ASCII (7-bit) and extended ASCII encode basic letters/symbols;
Unicode (e.g. UTF-8) covers a much larger set of world characters .
Example: ‘A’ is 65 in ASCII. (No need to memorize codes, but know the
concept of character-to-binary mapping).

• Multimedia – Graphics: Images can be bitmap (raster) or

vector. A bitmap image is defined by pixels; important terms include pixel,
file header, resolution (pixels wide × high), and colour depth/bit depth (bits
per pixel) . File size ≈ width×height×colour depth. A vector graphic uses
geometric shapes (drawing objects) with properties in a drawing list,
typically smaller for simple images. Bitmap images suit photographs;
vector is ideal for logos/diagrams (scales without quality loss).
 • Multimedia – Sound: Sound is sampled to digitize it. Sampling
rate (e.g. 44.1 kHz) and sampling resolution (bits per sample) determine
quality and file size. Higher sampling rate or bit depth → larger files but
better accuracy . E.g. CD audio uses 44100 samples/sec, 16-bit.

• Compression: Compression reduces file size. Lossless

compression (e.g. ZIP, PNG) preserves all original data, so it can be fully
restored . Lossy compression (e.g. JPEG, MP3) discards some data
(often imperceptible) to save more space . Use lossless for text/data (e.g.
RLE or LZW algorithms) and lossy for images/sound where slight quality
loss is acceptable  . Example: Run-Length Encoding (RLE)
compresses sequences of repeated characters. Always justify the choice
(e.g. PNG for logos to avoid loss, JPEG for photos to reduce size).

Chapter 2: Communication

• Network Types: A LAN (Local Area Network) connects devices

in a limited area (home, school); a WAN (Wide Area Network) spans large
distances (often by linking multiple LANs)  . LANs typically have
fast transmission and share one Internet connection; WANs connect LANs
(e.g. via leased lines or satellites) .

• Models – Client-Server vs Peer-to-Peer: In a client-server

model, dedicated servers provide resources/services to clients (e.g. web
server serving pages). In peer-to-peer, all nodes (peers) are equal and share
resources directly (e.g. file-sharing networks) . Client-server is easy to
manage but a single server can be a bottleneck; peer-to-peer is robust (no
single point of failure) but harder to secure and scale. Choose model based
on application needs .

• Thin vs Thick Clients: A thin client is a minimal computer that

relies on a server for processing/storage (common in virtual desktop
setups). A thick (fat) client can perform processing locally (like a typical
PC) and may not need a constant server connection. Thick clients can work
offline but require more local resources; thin clients simplify maintenance
but depend on network/server.
 • Network Topologies: Common layouts are star (all nodes
connected to a central switch/hub), bus (one shared backbone cable with
devices tapped on), mesh (every node connected to many others), and
hybrid. For example, star is easy to manage and failure of one cable only
affects one node; bus is cheaper but if the main cable fails the whole
network goes down; mesh is highly reliable (lots of redundancy) but
expensive to implement . Always pick topology based on scale and
reliability requirements.

• Cloud Computing: Cloud systems deliver computing over the

Internet. Public clouds (e.g. AWS, Google Cloud) provide services to
many users; private clouds are dedicated to one organization. Benefits
include scalability and no need for local hardware; drawbacks include
reliance on Internet access and data security concerns .

• Transmission Media: Wired: Copper cables (e.g. Ethernet) are

common; easy but limited bandwidth and range. Fibre-optic cables use
light for very high-speed, long-distance links. Wireless: WiFi uses radio
waves for local networking; other wireless links use microwaves or
satellites for long-range connectivity . Copper is cheap but prone to
interference; fibre is expensive but high-capacity; wireless gives mobility
but can suffer interference and security issues.

• Network Hardware: Key devices include Network Interface

Cards (NICs) for wired, Wireless NICs (WNICs) for wireless, and
Wireless Access Points (WAPs) to connect WiFi devices. Switches connect
devices within a LAN and forward frames (packets) to the correct port .
Routers connect different networks (e.g. LAN to Internet) and route
packets based on IP addresses  . Bridges link two similar networks.
Repeaters regenerate signals to extend range.

• Ethernet and Collisions: Traditional Ethernet (bus topology)

uses CSMA/CD (Carrier Sense Multiple Access/Collision Detection) to
avoid collisions: devices listen before sending and detect collisions to retry
. Modern switched Ethernet (star topology) uses full-duplex links, so
collisions are rare.

• Bit Streaming: Data (audio/video) can be streamed real-time

(live) or on-demand. Throughput (bit rate) is critical: higher bit rates give
better quality but require more bandwidth . E.g. a 1080p video might
need tens of Mbps; if bandwidth is insufficient, the stream buffers or
degrades.

• WWW vs Internet: The Internet is the global network of

networks (hardware); the World Wide Web is a collection of interlinked
documents and resources accessed via HTTP on the Internet. WWW uses
Internet infrastructure, but the internet also carries email, VoIP, etc.

• IP Addressing: An IP address uniquely identifies a device on the

internet. IPv4 uses 32-bit addresses (e.g. [Link]), IPv6 uses 128-bit
(written in hexadecimal). Subnetting divides IP ranges into sub-networks
for organization. Public IPs are routable on the Internet; private IPs are for
local networks (e.g. 10.x.x.x) . A static IP is fixed; a dynamic IP is
assigned via DHCP and can change .

• URLs and DNS: A URL (Uniform Resource Locator) specifies

a resource on the Web (e.g. [Link] The Domain Name
Service (DNS) translates human-readable domain names (like
[Link]) into IP addresses .

Chapter 3: Hardware

• System Components: Computers need input devices (keyboard,

mouse, microphone), output devices (monitor, printer, speakers), primary
memory (RAM, very fast and volatile) and secondary storage (HDD/SSD,
removable USB drives, optical discs, slower but non-volatile).

• Embedded Systems: A small computer built into appliances

(e.g. microwave controller, digital thermostat) performing dedicated tasks.
Advantages: optimized performance, reliability, low cost. Disadvantages:
inflexible, limited upgradeability.

• Hardware Devices: Know examples and functions: e.g.

laser/inkjet printers, 3D printers, scanners, hard disks (magnetic), flash
memory (SSD), optical drives, touchscreens, VR headsets, etc. .
 • Buffers: Small temporary storage areas (buffers) are used when
sending data between devices of different speeds (e.g. spooling print jobs)
to smooth the flow.

• RAM vs ROM: RAM (Random Access Memory) is volatile and

used for running programs; ROM (Read Only Memory) is non-volatile
firmware. RAM contents are lost on power-off; ROM (or Flash) retains
firmware. There are different ROM types: PROM (Programmable once by
manufacturer), EPROM (Erasable by UV light), EEPROM/Flash
(Electrically erasable) .

• SRAM vs DRAM: SRAM is fast but expensive and used for

CPU cache; DRAM is slower (needs refresh) and used for main memory
. DRAM is denser and cheaper but must be refreshed many times per
second.

• Memory Hierarchy: Faster memory (cache, RAM) close to

CPU; slower (SSD, HDD) further away. Wider bus width and more cache
improve performance .

• Monitoring & Control Systems: A sensor converts physical

stimuli to electrical signals (e.g. temperature sensor), and an actuator
converts signals to physical action (e.g. motor, heater). Monitoring systems
only observe (e.g. thermostats measuring temperature); control systems act
on data (e.g. cruise control adjusting speed). Feedback loops use sensors to
adjust actuators, maintaining system stability .

Chapter 3.2: Logic Gates and Logic Circuits

• Logic Gates: Fundamental digital circuits with Boolean

inputs/outputs. AND: output=1 only if all inputs=1 . OR: output=1 if
any input=1 . NOT: output is the inverse of input (0↔1). NAND:
output=NOT(AND) – 1 if NOT all inputs are 1. NOR: output=NOT(OR) –
1 only if all inputs=0. XOR (exclusive OR): output=1 if exactly one input
is 1 . (Use truth tables to verify: list all input combinations and resulting
output.)

• Truth Tables: For each 2-input gate, list input combinations

(00,01,10,11) and outputs. Useful for designing and checking logic.
 • Circuit Construction: Given a logic expression or truth table,
draw a circuit using gate symbols (NOT, AND, OR, NAND, NOR, XOR).
Conversely, derive an expression from a circuit or table . Practice: write
an expression for a circuit, or design a gate network from a verbal
problem.

• Example: If a system should light an LED only when switch A

or switch B (but not both) are on, implement an XOR gate.

Chapter 4: Processor Fundamentals

• Von Neumann Architecture: CPU and memory share a common

bus. The stored-program concept means instructions and data are both in
memory . The CPU fetches instructions
from memory one by one.

• CPU Registers: Special fast storage in the CPU:

• Program Counter (PC): holds address of next instruction .

• Memory Address Register (MAR): holds address for memory

access.

• Memory Data Register (MDR): holds data read from or written

to memory.

• Current Instruction Register (CIR): holds the current instruction

being executed .

• Accumulator (ACC): general-purpose register for

arithmetic/logic results .

• Index Register (IX): used for indexed addressing.

• Status Register: holds condition codes/flags (Zero, Carry, etc.).

• CPU Components:
 • Arithmetic Logic Unit (ALU): performs arithmetic and logic
operations.

• Control Unit (CU): orchestrates execution

(fetch/decode/execute) and controls buses.

• System Clock: generates timing pulses that synchronize CPU

operations .

• Immediate Access Store (IAS): another term for main memory.

• Buses:

• Address Bus: unidirectional, carries address from CPU to

memory or IO devices.

• Data Bus: bidirectional, carries data between CPU, memory,

and IO.

• Control Bus: carries control signals (read/write, interrupt, etc.).

• Performance Factors: CPU speed depends on number of cores

(parallel processing),
clock speed (MHz/GHz), bus width (how many bits moved at once), and
cache size . E.g. a 64-bit bus can transfer 64 bits per cycle.

• Ports/Interfaces: Common ports connect peripherals: USB

(general I/O devices),
HDMI/VGA (video output), Ethernet port (network), serial/parallel (older
devices) .

• Fetch–Execute Cycle: The CPU repeats: fetch the instruction

pointed by PC into CIR, increment PC, decode the instruction, execute it
(perform operation), then repeat . Use register-transfer notation for
detailed cycles (e.g. MAR ← PC; PC ← PC+1; MDR ← [MAR]; CIR ←
MDR; decode CIR).
 • Interrupts: An interrupt is an external/internal signal to pause
the current F-E cycle and handle an event (e.g. I/O complete, timer).
Causes: hardware signals or software. The CPU executes an Interrupt
Service Routine (ISR), then returns to the interrupted program .
Understand interrupt priorities and that interrupts are checked at specific
points (usually between instructions).

Chapter 4.2: Assembly Language

• Assembly vs Machine Code: Each assembly instruction

corresponds to one machine operation. Assemblers translate mnemonic
code to numeric machine code .

• Two-pass Assembler: First pass builds a symbol table of labels

and addresses; second pass generates machine code using that table .
This handles forward references to labels.

• Instruction Set Groups: Typical groups include Data Movement

(e.g. LDM, MOV,
STO for load and store), Input/Output (e.g. IN, OUT), Arithmetic/Logic
(ADD, SUB, INC, DEC, AND, OR, XOR), Control/Branch (JMP, JPE,
JPN), and Compare (CMP). Recognize each group’s purpose .

• Addressing Modes:

• Immediate (e.g. LDM #n): operand is a constant value .

• Direct (e.g. LDD <addr>): operand is a memory address whose

contents are used .

• Indirect (e.g. LDI <addr>): operand is an address that contains

the actual address of the data .

• Indexed (e.g. LDX <addr>): effective address = base +

contents of Index register .

• Relative: branch addresses given relative to PC (not explicitly

shown, but important concept).
 • Example Instructions: (Using ACC as accumulator register) e.g.
ADD <addr> (ACC←ACC + [addr]), ADD #n (ACC←ACC+n), SUB
<addr>, INC ACC, DEC ACC, JMP label, CMP <addr> (sets flags), JPE
label (jump if equal), IN/OUT (I/O), END (halt) . Know how to interpret
and trace simple assembly programs.

Chapter 4.3: Bit Manipulation

• Shifts: Shifting moves bits left or right. Logical Left Shift

(LSL): all bits shift left; zeros enter on right . Logical Right Shift (LSR):
bits shift right; zeros enter on left. Arithmetic Right Shift: similar to logical
right but replicates sign bit (for signed numbers). Rotate (Circular Shift):
bits shifted out one end re-enter at the other (preserves all bits).

• Bitwise Operations:

• AND: bitwise AND (e.g. AND #n performs ACC←ACC AND

n) results in a 1 only
if both bits are 1 . Useful for masking (testing specific bits).

• OR: sets a bit to 1 if either bit is 1 (ACC←ACC OR n) .

• XOR: sets a bit to 1 if bits differ (ACC←ACC XOR n) .

• Masking: To test a bit, AND with a mask (e.g. ACC AND
00010000b isolates one bit). To set a bit, OR with a mask. To toggle, XOR
with a mask.

• Example from ISA: Instructions like LSL #n (logical shift),

AND Bn/&n (bitwise AND with a number), XOR and OR instructions
operate on ACC and either an immediate or memory. Understand how
these affect binary values .
Chapter 5: System Software

Chapter 5.1: Operating Systems

• Purpose of OS: Manages hardware and provides common

services to applications
(file management, memory allocation, I/O control, user interface) .
Without an OS, each program would need its own drivers and hardware
handling.

• Key OS Functions:

• Memory management: allocating RAM to programs, virtual

memory.

• File management: creating, organizing, and securing

files/folders.

• Device management: interfacing with peripherals via drivers

(input/output management).

• Process management: scheduling CPU time, multitasking.

• Security management: enforcing user permissions, process

isolation .

• Utility Software: Built-in tools in an OS for maintenance: e.g.

disk formatting utilities, antivirus/anti-spyware, disk defragmenters, file
compression, backup software . Useful for reliability and performance.

• Program Libraries: Reusable code collections (e.g. DLLs).

Programmers use libraries to avoid rewriting common functions. Benefits:
faster development, consistency, smaller code size. Dynamic linking
(DLL) allows sharing a single library in memory .
Chapter 5.2: Language Translators

• Assembler: Converts assembly language to machine code one-

to-one. Needed because CPU only understands binary opcodes.

• Compiler: Translates an entire high-level source program into

machine code (object code) . Advantage: resulting executable runs
quickly. Drawback: compilation can be slow and less flexible (needs
recompilation on different systems).

• Interpreter: Reads and executes high-level code line-by-line .

Advantage: easier debugging and platform independence (no compilation
step). Drawback: slower execution and requires source code at runtime.
Java is partly compiled (to bytecode) and then interpreted by the JVM
(console mode) .

• IDE (Integrated Development Environment) Features: Tools

that assist coding:

• Coding aids: syntax highlighting, auto-completion (context-

sensitive prompts), automatic indenting.

• Error detection: on-the-fly syntax checking, compile-time error

reports.

• Code presentation: pretty-printing, code folding

(expand/collapse blocks).
• Debugging tools: setting breakpoints, single-step execution,
watching variables/expressions, output windows .
Chapter 6: Security, Privacy and Data Integrity

Chapter 6.1: Data Security

• Security vs Privacy vs Integrity: Security means protecting

systems and data from unauthorized access or damage. Privacy refers to
control over personal or sensitive information. Integrity means data
remains accurate and unaltered . All three are important in computer
systems.

• Need for Security: Both data (files, databases) and systems

(PCs, servers) must be secured. Unauthorized access or malware can
compromise confidentiality, availability, and integrity.

• Security Measures:

• Authentication: user accounts with strong passwords or

biometrics (fingerprint, iris scan).

• Access controls: permissions and rights to limit user actions.

• Encryption: transforms data into ciphertext to protect

confidentiality during storage/transmission.

• Digital signatures: validate sender identity and data integrity.

• Firewalls: block unauthorized network traffic.

• Antivirus/Anti-malware: detect and remove malware (viruses,

worms, spyware) .

• Threats:

• Malware: viruses, worms, trojans, ransomware.

• Hackers: unauthorized users exploiting vulnerabilities.

 • Social engineering: phishing (fake emails) and pharming
(redirecting to fake websites). These target users to steal credentials .

• Mitigations: Regular software updates, use of security suites,

network segmentation, employee training, and backups help restrict risks.
Encryption and access controls protect data even if systems are breached
.

Chapter 6.2: Data Integrity

• Validation vs Verification: Data validation checks data is

reasonable on entry (e.g. age ≥0, no letters in a phone number).
Verification checks data accuracy after entry or during transfer (e.g. re-
enter password to verify no typo) .

• Validation Methods: Common checks include range check (e.g.

1–31 for a day), format check (e.g. email pattern), length check, presence
check (ensure field not blank), existence check (e.g. user ID exists), and
check digit (last digit computed from others) .
• Verification Methods:

• During data entry: visual checks (display back input), double

entry (type twice to compare) .

• During data transfer: parity checks (add a parity bit to each

byte), checksums and CRCs on blocks of data to detect errors . These
ensure data integrity in storage/transmission.
Chapter 7: Ethics and Ownership

• Professional Ethics: Computing professionals should follow

ethical codes (e.g. BCS, IEEE) . This means respecting privacy,
avoiding harm, and reporting issues. Ethical behavior builds trust and
standards.

• Acting Ethically: Consider consequences: using software

responsibly, not exploiting data. Violations (e.g. malware creation, data
theft) have social and legal implications.

• Copyright: Legal protection for creators of software and digital

content. Unauthorized copying/distribution is illegal. Respect licenses and
intellectual property.

• Software Licenses: Common types include Free Software

(FSF/GNU GPL – source must remain open), Open Source (OSI-
approved), Shareware (try-before-buy), and Commercial (closed-source,
paid) . Choose based on needs: open source for transparency and
modifiability; commercial for professional support or unique features.

• Artificial Intelligence (AI) Ethics: AI applications raise issues

(bias in data, job displacement, privacy). Professionals must consider
social, economic, and environmental impacts of AI systems . For
example, ensure facial recognition respects consent; minimize AI’s energy
usage.
Chapter 8: Databases

Chapter 8.1: Database Concepts

• File vs Database: A simple file system (CSV or text files) has

limitations: data redundancy, difficulty maintaining consistency, limited
query options. A relational database overcomes these by storing data in
structured tables with defined relationships .

• Relational Model: Data is organized in tables (relations). Each

table represents an entity (e.g. Students), with rows (records/tuples) and
columns (fields/attributes). Keys: A primary key uniquely identifies each
record. A foreign key in one table references a primary key in another,
defining relationships.

• Terminology:

• Entity/Table: a thing or concept (e.g. Book, Customer).

• Attribute/Field: a column in a table (e.g. Title, Price).

• Tuple/Record: a row (one instance of entity).

• One-to-many, one-to-one, many-to-many: types of relationships

between tables.

• Referential integrity: ensures foreign keys match existing

primary keys.

• Indexing: speeds up searches on fields.

• ER Diagrams: Use Entity-Relationship diagrams to model

database design. Entities are rectangles, relationships are lines (e.g. one-to-
many symbol) . ER diagrams help visualize how tables link.

• Normalization: Process to reduce redundancy and improve

integrity:
 • 1NF: eliminate repeating groups (each field contains atomic
value).

• 2NF: remove partial dependencies (non-key attributes depend

on whole composite key).

• 3NF: remove transitive dependencies (non-key attributes

depend only on primary key) .

• Example: If a table lists (StudentID, CourseID, InstructorName)

and Instructor depends only on Course, break into two tables.

• Assessment: Be ready to analyze a set of tables or data,

determine if they are in 3NF, and redesign tables to eliminate redundancies
.

Chapter 8.2: Database Management Systems (DBMS)

• DBMS Functions: Software (e.g. MySQL, SQLite) that

manages databases. It provides a data dictionary (metadata about
tables/fields) and supports data modelling and logical schema management
. It handles all data operations securely and efficiently.

• Features:

• Data management: enforcing constraints, maintaining data

dictionary of table/field definitions.

• Schema definition: logical design (tables, relations).

• Integrity: built-in support for keys and constraints.

• Security: user authentication, access rights, backup/recovery

procedures .
 • DBMS Tools:

• Developer Interface: tools (GUI) for database design and

administration.

• Query Processor: engine that executes SQL queries and returns

results .

• Utility Tools: backup/restore, index maintenance, and others to

maintain the database.

Chapter 8.3: DDL and DML

• Data Definition Language (DDL): SQL commands that

define/alter database structure. Standard commands include:

• CREATE DATABASE dbName; – create new database.

• CREATE TABLE tableName ( field1 TYPE, field2 TYPE, ...,

PRIMARY KEY(...), FOREIGN KEY(...)); – define tables with data types
and keys .

• ALTER TABLE tableName ADD/DROP COLUMN – modify

existing table schema.

• DROP TABLE tableName; – delete a table.

• Include data types like INT, CHAR(n), VARCHAR(n),

BOOLEAN, REAL, DATE, TIME .

• Data Manipulation Language (DML): SQL commands to query

and change data:

• SELECT columns FROM table WHERE condition [ORDER

BY ... GROUP BY ...] – retrieve data (with optional sorting, grouping,
aggregation such as SUM, COUNT, AVG) . Can involve joins (e.g.
INNER JOIN) across up to two tables in AS Level.
 • INSERT INTO table (fields) VALUES (...); – add new records.

• UPDATE table SET field=value WHERE condition; – modify

existing records.

• DELETE FROM table WHERE condition; – remove records.

• SQL Standard: SQL is the industry-standard language for both

DDL and DML . Practice writing simple SQL statements for given table
structures and queries.

• Example: Given a table of Employees(Name, DeptID) and

Departments(DeptID, DeptName), you might:

• DDL: CREATE TABLE Employees(Name VARCHAR(50),

DeptID INT, FOREIGN KEY (DeptID) REFERENCES
Departments(DeptID)); .

• DML: SELECT DeptName, COUNT(*) FROM Employees E

INNER JOIN Departments D ON [Link]=[Link] GROUP BY
DeptName; to count employees per department .