Computer Engineering is coming soon

We're still writing and reviewing the notes for every subject, and we'll open it when every subject is ready. Meanwhile you can practise aptitude, ask Aria about any subject, or see the branches that are ready.

Mobile security and device management

Mobile security and device management covers protecting mobile devices, apps and data (confidentiality, integrity, availability) and managing devices at scale using MDM/EMM/UEM, secure boot, app controls, VPNs, and threat mitigation.

Drafted with Aria, reviewed by the AiCanCode.org team · Updated . How we write and check our notes. Spotted an error? Use Give Feedback at the bottom of the page.

Why it matters

Mobile phones and tablets hold sensitive user and enterprise data. Engineers use mobile security and device management to keep data safe, enforce corporate policies, remove threats remotely, and ensure devices continue to work reliably in a fleet (for example, company phones, BYOD across employees, or field IoT/mobile gateways).

Key ideas

Start by defining terms simply and exactly.

  • Mobile device: an electronic computing device that is portable and includes wireless communications (smartphone, tablet, or mobile IoT gateway).
  • Security: the practice of protecting systems, networks and data from unauthorized access, disclosure, disruption, modification, or destruction.
  • Confidentiality: only authorized parties can read data.
  • Integrity: data is not modified by unauthorized parties; modifications are detectable.
  • Availability: authorized users can access resources when needed.
  • Threat: a potential cause of an unwanted impact on a system (e.g., malware, network eavesdropping, lost device).
  • Vulnerability: a weakness in a system that can be exploited by a threat.
  • Risk: the expected impact from threats exploiting vulnerabilities (often Risk = Likelihood × Impact, used qualitatively or quantitatively).
  • Authentication: proving identity (passwords, biometrics, tokens).
  • Authorization: deciding what an authenticated user/device may do.
  • Mobile Device Management (MDM): a system that lets administrators enroll, configure, monitor and control mobile devices remotely.
  • Enterprise Mobility Management (EMM): MDM plus mobile application management, content management and often identity management.
  • Unified Endpoint Management (UEM): converged management of all endpoints (PCs, mobile, IoT) from a single console.
  • BYOD (Bring Your Own Device): using personal devices for work; raises privacy and management trade-offs.
  • Containerization / App Container: separating corporate apps/data from personal apps on the same device using software fences.
  • Sandboxing: running an app in a restricted environment so it cannot access other apps' data.
  • Secure Boot: a hardware/firmware process that ensures only trusted software loads during device start-up.
  • Trusted Execution Environment (TEE): a secure area of the processor to run sensitive code and store keys (e.g., ARM TrustZone).
  • Secure Enclave: Apple’s dedicated secure coprocessor that protects keys and biometrics; distinct from ARM TrustZone/TEE on the main CPU.
  • Jailbreak / Rooting: removing OS vendor restrictions to gain privileged access; increases risk.
  • Mobile Threat Defense (MTD): techniques and products (often with behavioral analysis) to detect on-device threats like malware and network attacks.

Assumptions behind common university derivations:

  • For brute-force security calculations we assume a fixed number of attempts per second and independent guesses.
  • For risk scoring we assume risk can be approximated as a product or mapped on a table of likelihood vs impact.

MDM architecture — what to draw and label:

  • Draw a cloud (MDM server) labelled MDM server / console. Draw three arrows from the server to three boxes labelled Device A (Android), Device B (iOS), Device C (Windows). Between server and devices, label the arrows "Enrollment, Policies, Commands (remote wipe, lock, config)". Next to MDM server, draw boxes for App catalog, Policy engine, Inventory DB, Certificate authority and an arrow to Enterprise network / VPN Gateway. This shows control plane, app distribution, inventory and connection to enterprise resources.

Policy enforcement flow (describe to draw):

  1. Draw user with mobile device -> arrow -> MDM Enrollment API (label: device sends identity and certificate request).
  2. Draw MDM -> Certificate Authority (CA) (label: signs device certificate).
  3. Draw device -> Enterprise resources via VPN using device certificate. Show remote actions (wipe, lock, push-config) from MDM console to device.

Security controls to explain and exact definitions:

  • Device enrollment: process that registers a device with the MDM and issues identity credentials. Should include device integrity attestation when supported (e.g., Android Play Integrity/SafetyNet, Apple Device Attestation).
  • Policy enforcement: rules MDM applies (password complexity, OS patch level, allowed apps). Enforcement can be soft (notify) or hard (block network access).
  • App management: distributing, updating and controlling apps (app whitelisting, blacklisting, managed app configuration).
  • Certificate-based authentication: issuing device certificates so access decisions use cryptographic identity rather than passwords.
  • Network protections: enterprise VPNs, per-app VPNs, and always-on VPNs.
  • Data protections: disk encryption (device-level), app-level encryption, containerization and DLP (data loss prevention).

Differences between platforms (brief, exact):

  • iOS: strong sandboxing, managed profiles via Apple MDM (Apple Push Notification Service — APNs — used as a notification/wake channel; the device pulls MDM commands over the check-in connection), less permissive for background access; jailbreaking is uncommon but increases risk.
  • Android: wider device variety and vendor-modified OS images; offers Android Enterprise (Work Profile) for containerization; rooting increases capability for attackers.

Cryptographic principles to remember: use vetted algorithms (AES, RSA/ECC with recommended key sizes) and protect keys in TEE/secure element. Recommended sizes/protocols: AES-128/256; RSA ≥ 2048–3072 bits; ECC ≥ 224/256-bit curves (e.g., NIST P-256 or Curve25519); use SHA-256/384; prefer TLS 1.2/1.3.

Formulas

Keyspace = 2^k — k is key length in bits. Unit: dimensionless. Applies to symmetric keys (e.g., AES-128 has k=128).

Expected_time_to_bruteforce ≈ Keyspace / (2 × Attempts_per_second) — average-case; worst-case = Keyspace / Attempts_per_second. Keyspace is number of keys; Attempts_per_second in attempts·s^-1; Time in seconds. Assumes independent, uniformly random guesses.

Entropy = -Σ p_i · log2(p_i) — p_i are probabilities of each symbol; entropy unit: bits. Applies to password/passphrase randomness estimation.

Risk ≈ Likelihood × Impact — both are scored on a compatible numeric scale (e.g., 1–5); units: arbitrary risk units. Applies to qualitative/quantitative risk assessments.

Password_strength_bits ≈ log2(N^L) = L · log2(N) — L is length, N is number of possible symbols per position. Unit: bits of entropy. Applies to estimating brute-force resistance for uniformly random passwords.

Worked examples

  1. Brute-force time for AES-128
  • Problem: Calculate time to brute-force a 128-bit symmetric key if an attacker can try 10^12 keys per second (one trillion per second, optimistic with custom hardware). Use formulas and show units.
  • Formula: Time = 2^128 / 10^12 s.
  • Compute: 2^128 ≈ 3.4028237 × 10^38. So Time ≈ 3.4028×10^38 / 1×10^12 s = 3.4028×10^26 s.
  • Convert to years: 1 year ≈ 3.1536×10^7 s. Years ≈ 3.4028×10^26 / 3.1536×10^7 ≈ 1.079×10^19 years.
  • Final answer (bold): ~1.08 × 10^19 years. This shows AES-128 is effectively safe from brute force with realistic compute. Average-case (expected) time is half of this (~5.4 × 10^18 years).
  1. Risk scoring for a lost device with unencrypted storage
  • Problem: Company assigns Likelihood = 3 (possible) and Impact = 5 (severe) on a 1–5 scale for a lost unencrypted device. Compute risk and mitigation.
  • Formula: Risk = Likelihood × Impact = 3 × 5 = 15 (on a 1–25 scale).
  • Interpretation: High risk (commonly categorized as High on 5×5 risk matrices). Mitigations: enable full-disk encryption, enforce PIN/biometric, remote wipe, MDM policy to require data-only containers, use DLP and remote lock. After controls applied, Likelihood reduces to 1, Impact to 2 => Risk = 2, acceptable.
  • Full-marks exam answer should state calculation and the concrete controls that reduce each factor.
  1. Step-by-step MDM enrollment and remote wipe (algorithmic trace)
  • Setup: User enrolls Device X in MDM.
  • Steps: 1) User opens MDM enrollment URL and authenticates (username/password + 2FA). 2) Device generates key pair and CSR (certificate signing request). 3) MDM validates identity, forwards CSR to CA; CA signs certificate. 4) Device receives certificate and profile. 5) MDM pushes policies (password rules, VPN config, app install). 6) If device reported compromised, admin issues remote wipe command. 7) Device receives a push notification (APNs/FCM) and securely checks in to the MDM; it verifies the command from the MDM server over the authenticated check-in channel, then performs a factory reset or selective wipe of the corporate container depending on policy.
  • Show logs and timestamps in an exam answer; mention reliance on push services (APNs for iOS, FCM for Android).

Common mistakes

  • Forgetting to protect keys (storing keys in app code or insecure storage instead of TEE/secure enclave).
  • Using simple arithmetic for brute-force time but forgetting to convert units (seconds → years).
  • Assuming MDM can fully control a rooted/jailbroken device without attestation checks.
  • Treating BYOD as identical to corporate-owned devices (privacy + consent issues change allowed controls).

For your exam

Question types you will see: definitions (define MDM, sandboxing), short notes (compare MDM vs EMM vs UEM), derivations (compute brute-force time, entropy of a password), diagrams (draw MDM architecture and label components), and scenario-based answers (design policy for BYOD or laundry-list of steps to secure app distribution).

What full-marks answers include:

  • Clear definitions of each term used (bold terms in your answer if allowed).
  • A labelled diagram where asked (follow the exact drawing description in Key ideas).
  • For numeric questions: show the formula, substitute numbers with units, show each unit conversion, and give final numeric answer with units.
  • For policy/design questions: list assumptions, objectives, controls (technical and operational), and explain how each control reduces likelihood or impact.

Quick check

  1. What is the main purpose of an MDM server?
  2. Write the formula for time to brute-force a k-bit key at r attempts per second.
  3. Name two platform differences important for mobile security.

Answers: 1) To enroll, configure, monitor and control mobile devices remotely (enforce policies, push apps, take remote actions). 2) Expected time ≈ 2^(k-1) / r (average-case); upper bound 2^k / r. 3) iOS has stricter sandboxing and uses APNs only as the MDM notification channel; Android has more OEM variation and different work-profile/container options (Android Enterprise).

Finished this topic? Mark it so your progress, study plan and readiness keep up.

Stuck on something here?