Tutorial 14: Comprehensive Unit 7 Review and Integration

OFDMA allows sub‑channel allocation to multiple users simultaneously, improving spectral efficiency and reducing latency in dense environments.

Frequency reuse uses the same frequencies in cells separated by sufficient distance to avoid interference. The cluster size N determines the number of cells before frequencies are reused; a smaller N increases capacity but also increases co‑channel interference.

CSMA/CD (used in wired Ethernet) detects collisions after transmission. CSMA/CA (used in wireless) avoids collisions by sensing the channel and using random backoff; collision detection is not feasible due to the hidden terminal problem and the inability to listen while transmitting.

The Home Agent is a router on the mobile node's home network that maintains a binding between the node's home address and its current care‑of address, and tunnels packets to the mobile node when it is away.

Network slicing creates isolated logical networks with dedicated resources and QoS parameters. URLLC slices have low latency and high reliability, while eMBB slices provide high throughput.

The hidden terminal problem occurs when two stations are out of range of each other but both can communicate with an access point, causing collisions. Mitigation: RTS/CTS handshake reserves the medium.

Soft handoff (make‑before‑break) connects to the new cell before disconnecting from the old, providing seamless continuity. Hard handoff (break‑before‑make) disconnects first, causing a brief interruption.

WPA3 introduces SAE (resistant to offline dictionary attacks), mandatory Management Frame Protection, and OWE for open networks.

ZigBee supports mesh networks for home/industrial automation; BLE is a star topology with low‑power peripherals for wearables and beacons.

The cyclic prefix prevents intersymbol interference (ISI) by absorbing multipath delay spread and preserving orthogonality between subcarriers.

Erlang B calculates the probability of call blocking given the offered traffic (Erlangs) and the number of traffic channels, enabling capacity planning.

Carrier aggregation (up to 100 MHz), enhanced MIMO (8×8 DL, 4×4 UL), and Coordinated Multi‑Point (CoMP).

LAs are used in 2G/3G; TAs are used in 4G/5G. Both are groups of cells for location management, but TAs support more flexible configurations and are optimized for packet‑switched networks.

It uses energy detection, cyclostationary detection, or matched filtering to detect the presence of primary users in a frequency band.

The SGW routes user‑plane packets, acts as a local mobility anchor, and buffers downlink data for idle UEs.

MU‑MIMO allows multiple users to be served simultaneously on the same channel using spatial multiplexing, increasing overall throughput and capacity.

Cell splitting reduces the cell radius, creating more cells in the same area. Since frequencies are reused more densely, capacity increases approximately as (R_old / R_new)².

PDCP provides header compression (ROHC), ciphering, integrity protection, and in‑order delivery of packets.

LoRa is unlicensed, long‑range, low‑power, and cheaper; NB‑IoT uses licensed spectrum, offers better reliability, deeper indoor penetration, and is integrated with cellular networks.

BSS coloring allows spatial reuse by enabling stations to differentiate between transmissions from overlapping BSSs and transmit concurrently if the signal is below a threshold.

MPTCP allows a TCP connection to use multiple IP addresses (paths) simultaneously. If one path fails, the connection continues over another, providing seamless mobility.

Air‑interface latency < 1 ms, end‑to‑end latency < 5 ms, reliability 99.999%, and availability.

S‑MAC uses a fixed duty cycle; T‑MAC adapts the active period based on traffic load to reduce energy waste in low‑traffic conditions.

AMF handles registration, connection management, mobility, and paging.

DCI carries scheduling information for downlink and uplink data, including resource allocation, modulation, and HARQ parameters.

SDR implements signal processing in software, allowing reconfigurability and support for multiple protocols on the same hardware.

RIS can reflect and steer signals to improve coverage, reduce blockage, and enhance spectral efficiency by intelligently shaping the propagation environment.

High path loss, atmospheric absorption, and lack of high‑power, low‑cost components.

NTN integrates satellite, HAPS, and drone communications with terrestrial 5G to extend coverage to remote, maritime, and aerial areas.

Licensed spectrum offers guaranteed QoS and interference protection but is expensive and scarce. Unlicensed spectrum is free but subject to interference, limited power, and no QoS guarantees.

For LTE, 20 MHz gives ~150 Mbps with 2×2 MIMO and 64‑QAM. 30 MHz can be split into carriers (e.g., 20+10 MHz with carrier aggregation) or use 20 MHz. Assuming 20 MHz, capacity per cell ≈ 150 Mbps. With 3‑cell reuse (N=3), each cell gets 30/3=10 MHz? Actually, reuse pattern affects bandwidth per cell. Typically, bandwidth per cell = total bandwidth / N = 30/3 = 10 MHz. With 10 MHz, capacity ≈ 75 Mbps. For 10 cells, total capacity = 10 × 75 = 750 Mbps.

Wi‑Fi: 1% duty cycle: I = 0.01×30 + 0.01×20 + 0.98×0.01 = 0.3+0.2+0.0098 = 0.5098 mA. Battery life = 1000 / 0.5098 ≈ 1961 hours ≈ 82 days.
LoRa: transmission 1 s per hour = 1/3600 = 0.000278 duty cycle. I = 0.000278×100 + 0.000278×10 + 0.99944×0.005 ≈ 0.0278+0.00278+0.004997 = 0.0356 mA. Battery life = 1000 / 0.0356 ≈ 28090 hours ≈ 3.2 years. LoRa is more energy‑efficient for very low data rate.

For VoLTE to Wi‑Fi (VoWi‑Fi), the device uses IMS. The handover is controlled by the IMS core. Steps: 1) UE measures Wi‑Fi signal; 2) UE sends SIP re‑INVITE to IMS with new contact (Wi‑Fi IP); 3) IMS updates session; 4) Media path switches to Wi‑Fi. For LTE to 2G/3G (eSRVCC), the MME and MSC coordinate the handover with SRVCC.

Slice 1 (mMTC): 5QI for low power, massive connectivity, non‑GBR, low priority. Use NB‑IoT or LTE‑M within the 5G system. Slice 2 (URLLC): 5QI 1 or 2, GBR, low latency, high reliability. Dedicated resources (mini‑slots), priority scheduling. Isolation: radio resource reservation, separate UPF and core NFs, and dedicated transport slices.

Interference between BLE and ZigBee in the same band (2.4 GHz) can cause packet collisions and retransmissions. Mitigations: 1) Use channel hopping (BLE adaptive hopping, ZigBee can also hop), 2) Use different frequency channels (e.g., ZigBee on channels 15-26, BLE on advertising channels 37-39), 3) Power control to reduce interference, 4) Time‑division multiplexing (e.g., ZigBee schedule).

Prevent by: 1) Enforce server certificate validation on clients (trusted CA, check CN), 2) Use EAP‑TLS with mutual certificates (stronger), 3) Deploy WIDS to detect rogue APs, 4) Use 802.11w for management frame protection.

Reuse distance D = √(3N) × R = √(21) × 2 = 9.17 km. C/I = 1 / (6 × (D/R)^(-γ)) = 1 / (6 × (4.58)^(-3.5)) = 1 / (6 × (1/4.58^3.5)). 4.58^3.5 ≈ 4.58^3 * √4.58 = 96.1 * 2.14 = 205.7. So C/I = 1 / (6/205.7) = 205.7/6 = 34.3 linear = 15.4 dB. Typical requirement is 10‑12 dB, so it's acceptable.

Architecture: Controller with global view, APs with OpenFlow‑like extensions. Messages: APs send measurement reports (channel quality, interference). Controller computes optimal channel assignment using an algorithm (e.g., graph coloring). Controller sends configuration commands (channel change, power adjustment) to APs. The controller can also enforce policies and load balance.

Rain attenuation at Ku/Ka bands (used by Starlink) causes signal loss, reducing SNR. AMC adjusts modulation and coding: at lower SNR, it switches to lower‑order modulation (QPSK) and more coding, reducing data rate but maintaining the link. This prevents complete loss of connectivity.

Integration: LoRa gateways send data to a LoRaWAN network server, which pushes data to a cloud platform. 5G data (video streams) is sent via 5G to the same cloud. The cloud platform (with APIs) provides a unified dashboard and analytics. Edge computing can be used for real‑time processing of 5G video before sending to cloud.

MPTCP maintains two subflows (one per interface). If Wi‑Fi drops, the subflow fails, but the LTE subflow continues. The MPTCP scheduler can prioritise data on the best available path (e.g., higher throughput). The connection stays alive and the video continues with minimal disruption.

5G URLLC offers <1 ms latency, 99.999% reliability, and wide‑area mobility, making it suitable for large factories with moving vehicles. Wi‑Fi 7 can achieve sub‑5 ms latency with high reliability but is limited to shorter range and may have more interference in dense environments. For indoor, small‑scale AGVs, Wi‑Fi 7 might be cost‑effective; for large or outdoor, 5G is better.

NSA: add 5G NR to existing LTE eNodeBs with EPC. Use LTE for control and 5G for data. Faster rollout, lower capex, but no network slicing, higher latency. SA: deploy 5GC and NR for both control and data, enabling all 5G features (slicing, low latency). Steps for NSA: upgrade eNodeBs, add 5G NR, and ensure inter‑working.

This is a simplified analysis. Probability of detection depends on SNR = 10 dB and threshold. Using a standard formula: P_d = Q((λ - μ)/σ) where μ = received power. For false alarm, noise only. Trade‑off: lowering threshold increases detection but also increases false alarms. In practice, cooperative sensing and adaptive thresholding are used.

Wi‑Fi 6: for laptops and phones indoors (high throughput, dense environment). 5G small cells: for outdoor areas and high‑mobility users (phones in campus). LoRa: for IoT sensors (environmental monitoring, parking). Backhaul: fiber to APs and small cells. Use a unified management system (SDN) for visibility and policy enforcement.

Wi‑Fi 6: 15 kHz SCS, up to 1024‑QAM, MU‑MIMO, OFDMA. 5G NR: flexible SCS (15,30,60,120,240 kHz), higher QAM (256/1024), massive MIMO. Wi‑Fi is optimized for indoor, high‑density; 5G for both indoor/outdoor with mobility. Spectral efficiency: 5G can achieve >10 bps/Hz, Wi‑Fi ~5 bps/Hz.

Threats: side‑channel attacks (e.g., cache timing), DoS on one slice affecting others, unauthorised access across slices. Mitigations: isolation of resources (CPU, memory, bandwidth), strict authentication and authorization, encryption, and monitoring of slice performance for anomalies.

Capacity C = Σ log₂(1 + (λ_i * P_i)/N₀) where λ_i are eigenvalues of H H^H. Water‑filling allocates power to the strongest eigenmodes. Capacity scales linearly with min(N_t, N_r) in the high‑SNR regime with full CSI at the transmitter.

LTE: MME for control, SGW/PGW for user‑plane, with fixed interfaces. 5G: AMF for access/mobility, SMF for sessions, UPF for user‑plane; service‑based interfaces allow modularity, dynamic scaling, and network slicing. This enables faster handover and support for diverse services.

NR‑U uses LBT (listen‑before‑talk) similar to Wi‑Fi, but with different parameters. Coexistence challenges: unfairness if one has higher backoff, hidden nodes. Mechanisms: use of DFS, channel occupancy times, and coordination via 5G core. Studies show fair coexistence with proper parameter tuning.

The Bianchi model gives throughput S = (P_s P_tr L) / ((1-P_tr)σ + P_tr P_s T_s + P_tr(1-P_s) T_c). As n increases, collisions increase, reducing throughput. Optimizing CW_min and CW_max can improve performance.

LEO provides global coverage but cannot match the capacity (Gbps per km²) and ultra‑low latency of terrestrial 5G, especially in dense urban areas. They are complementary: LEO for rural/remote, 5G for urban. Economic: LEO is costly to deploy and maintain. Regulatory: spectrum coordination is complex. So, they will coexist.

ISAC uses the same waveform for communication and sensing, e.g., using OFDM signals for radar. Challenges: high‑resolution sensing requires large bandwidth; interference between sensing and communication; and processing complexity. Benefits: enables new applications like health monitoring, autonomous driving.

Spectrum: CBRS (3.5 GHz) or licensed. Architecture: gNBs on towers, 5GC with local UPF and MEC for low latency. QoS: URLLC for crane control (latency <10 ms), eMBB for video surveillance. Security: secure authentication, encryption, network slicing for isolation, and WIDS for rogue devices.

Energy consumption depends on transmit power, data rate, duty cycle, and receiver sensitivity. LoRa and NB‑IoT are energy‑efficient for low data rate and long range. Wi‑Fi is less efficient for IoT but suitable for high‑throughput. ZigBee is a middle ground. Key factors: duty cycle (sleep time), data rate (shorter transmission), and power amplifier efficiency.

AI/ML can predict traffic patterns for resource allocation; reinforcement learning for beamforming and handover decisions; and anomaly detection for predictive maintenance of base stations. The NWDAF in 5G provides data for such analytics.

5G‑AKA provides mutual authentication and key derivation with forward secrecy. SUCI encrypts IMSI. Vulnerabilities: false base station attacks (though mitigated by SUCI and mutual auth), side‑channel attacks, and denial‑of‑service. Countermeasures: use of strong crypto, secure hardware, and regular updates.

RIS can improve coverage and energy efficiency. Challenges: real‑time optimization of thousands of elements, channel estimation (since RIS introduces new paths), and hardware cost. Research: using deep learning for control, simplified models, and low‑cost designs.

LTE: X2 handover is faster, S1 goes through MME. 5G: uses conditional handover (CHO) and fast switching with RRC; also, the service‑based core allows quicker path updates. 5G handover latency is < 50 ms, often < 20 ms.

Proposal: Investigate the feasibility of THz communications for short‑range high‑data‑rate links, combined with RIS for coverage enhancement. Develop post‑quantum crypto for security. Methodologies: channel measurement and modeling, RIS optimization algorithms, and implementation on SDR platforms. Expected outcomes: 100 Gbps data rates and ultra‑secure communications.