Wireless Communication Interview Questions

RF (Radio Frequency) refers to electromagnetic waves in the frequency range of approximately 3 kHz to 300 GHz. This spectrum is used for wireless communication, broadcasting, and radar systems. The RF spectrum is divided into bands: VLF (3-30 kHz), LF (30-300 kHz), MF (300 kHz-3 MHz), HF (3-30 MHz), VHF (30-300 MHz), UHF (300 MHz-3 GHz), SHF (3-30 GHz), and EHF (30-300 GHz). Different applications use different bands based on propagation characteristics and bandwidth requirements.

An antenna is a transducer that converts electrical signals into electromagnetic waves for transmission, or vice versa for reception. It serves as the interface between the transmission line and free space. Key functions include: radiating RF energy efficiently, providing directional characteristics (gain), matching impedance between transmitter/receiver and free space (377 ohms), and operating at specific frequency bands. The antenna's size is typically related to the wavelength of the operating frequency.

WiFi primarily operates in two main frequency bands: 2.4 GHz and 5 GHz. The 2.4 GHz band (2.400-2.4835 GHz) offers better range and wall penetration but is more congested and has limited bandwidth. The 5 GHz band (5.150-5.825 GHz) provides higher data rates and less interference but shorter range. WiFi 6E and WiFi 7 also use the 6 GHz band (5.925-7.125 GHz) for even more bandwidth and channels. The 2.4 GHz band has 14 channels (only 3 non-overlapping), while 5 GHz has many more non-overlapping channels.

Bluetooth is a short-range wireless technology standard for exchanging data over short distances using UHF radio waves in the 2.4 GHz ISM band (2.400-2.4835 GHz). It uses frequency hopping spread spectrum (FHSS), hopping between 79 channels at 1 MHz spacing to minimize interference. Bluetooth is designed for low power consumption and supports data rates from 1 Mbps (Classic) to 2 Mbps (Bluetooth 5.0). It's commonly used for headphones, keyboards, mice, and IoT devices with a typical range of 10-100 meters depending on the class.

Electromagnetic wave propagation refers to how radio waves travel from transmitter to receiver through different media. The main propagation modes are: Ground wave (follows Earth's surface, used for AM radio), Sky wave (reflects off ionosphere, used for HF communication), Line-of-sight (direct path between antennas, used for VHF/UHF and above), and Space wave (combination of direct and ground-reflected waves). Propagation characteristics depend on frequency, terrain, atmospheric conditions, and distance. Higher frequencies generally require line-of-sight paths.

A cellular network is a wireless communication network distributed over land areas called cells, each served by a base station (cell tower). It's called 'cellular' because the coverage area is divided into hexagonal cells, like a honeycomb pattern. Each cell uses a specific set of frequencies that can be reused in non-adjacent cells, enabling frequency reuse and supporting many users. As users move between cells, handoff/handover occurs to maintain connectivity. This architecture enables efficient spectrum utilization and scalable network capacity.

GSM (Global System for Mobile Communications) is a 2G digital cellular standard developed in Europe. Key features include: TDMA (Time Division Multiple Access) for channel access, GMSK modulation, 200 kHz channel bandwidth, support for voice and SMS, SIM card-based authentication, and international roaming capability. GSM operates in various frequency bands (900 MHz, 1800 MHz, 850 MHz, 1900 MHz). It introduced the concept of separating user identity (SIM) from the device, enabling easy phone switching while keeping the same number.

A dipole antenna is a simple and widely used antenna consisting of two conductive elements (usually wires or rods) of equal length, oriented end-to-end with a feed point at the center. A half-wave dipole has a total length of half the wavelength (λ/2). When RF current flows through it, it creates an electromagnetic field that radiates perpendicular to the antenna axis. The dipole has an omnidirectional pattern in the horizontal plane and a figure-8 pattern in the vertical plane, with a gain of about 2.15 dBi and characteristic impedance of 73 ohms.

Free Space Path Loss (FSPL) is the loss in signal strength that occurs when electromagnetic waves travel through free space without any obstructions. It's calculated using the formula: FSPL(dB) = 20log10(d) + 20log10(f) + 20log10(4π/c), where d is distance and f is frequency. Simplified: FSPL(dB) ≈ 32.45 + 20log10(f_MHz) + 20log10(d_km). FSPL increases with both distance (6 dB per doubling of distance) and frequency (6 dB per doubling of frequency). This is why higher frequency signals have shorter range for the same transmit power.

Signal-to-Noise Ratio (SNR) is the ratio of the power of the desired signal to the power of background noise, typically expressed in decibels (dB). SNR(dB) = 10log10(Psignal/Pnoise). A higher SNR indicates a cleaner signal with less noise corruption. In wireless communications, SNR determines the achievable data rate (Shannon-Hartley theorem: C = B × log2(1 + SNR)) and bit error rate. Typical requirements: voice calls need ~10 dB SNR, while high-speed data needs >20 dB SNR for low error rates.

FDMA (Frequency Division Multiple Access): Divides the spectrum into frequency channels; each user gets a dedicated frequency band. TDMA (Time Division Multiple Access): Users share the same frequency but are assigned different time slots in a repeating frame. CDMA (Code Division Multiple Access): All users share the same frequency and time but are separated by unique spreading codes. FDMA is simple but inefficient; TDMA improves efficiency by time-sharing; CDMA provides soft capacity and better resistance to interference but requires power control.

Duplexing refers to the method of separating uplink (device to base station) and downlink (base station to device) transmissions in a two-way communication system. The two main types are: FDD (Frequency Division Duplexing) uses separate frequency bands for uplink and downlink, requiring guard bands and duplexer filters. TDD (Time Division Duplexing) uses the same frequency but alternates between uplink and downlink in different time slots. FDD offers constant latency; TDD allows flexible allocation but requires synchronization.

ISM (Industrial, Scientific, and Medical) bands are portions of the radio spectrum reserved internationally for non-commercial use in industrial, scientific, and medical applications. The most commonly used ISM bands are: 2.4-2.5 GHz (WiFi, Bluetooth, microwave ovens), 5.725-5.875 GHz (WiFi), and 902-928 MHz (in Americas). These bands are unlicensed, meaning anyone can use them without a specific license, but devices must meet power and emission limits. The shared nature of ISM bands leads to potential interference between different technologies.

Antenna gain is a measure of how well an antenna focuses or directs radiated power in a particular direction compared to a reference antenna. It's measured in dBi (relative to isotropic radiator) or dBd (relative to dipole antenna), where dBi = dBd + 2.15. Gain is achieved by focusing energy in specific directions rather than creating energy. A higher gain antenna has a narrower beam and longer range in the beam direction. For example, a parabolic dish antenna might have 30+ dBi gain, while a simple dipole has 2.15 dBi.

Fading is the variation in signal strength over time and/or location due to changes in the propagation environment. Main types include: Large-scale fading (path loss and shadowing from obstacles), and Small-scale fading (rapid fluctuations due to multipath propagation). Multipath fading occurs when signals arrive via multiple paths with different delays, causing constructive or destructive interference. Rayleigh fading describes conditions without a dominant line-of-sight path, while Rician fading includes a strong direct component. Fading mitigation techniques include diversity, equalization, and OFDM.

A link budget calculates the received signal power considering all gains and losses: Prx = Ptx + Gtx - Ltx - FSPL - Lmisc + Grx - Lrx, where Ptx is transmit power (dBm), Gtx/Grx are antenna gains (dBi), Ltx/Lrx are cable/connector losses (dB), FSPL is free space path loss, and Lmisc includes fading margins, body loss, and environmental factors. The link margin = Prx - Sensitivity. For reliable communication, the margin should be 10-20 dB to account for fading. Example: 20 dBm Tx + 3 dBi Tx antenna - 2 dB cable loss - 100 dB FSPL + 6 dBi Rx antenna - 1 dB Rx loss = -74 dBm received power.

OFDM (Orthogonal Frequency Division Multiplexing) divides a wideband channel into many narrowband orthogonal subcarriers. Each subcarrier carries a portion of data at a lower rate, making them resistant to frequency-selective fading. The orthogonality (subcarrier spacing = 1/symbol duration) allows overlapping spectra without interference. Implementation uses IFFT at transmitter and FFT at receiver. A cyclic prefix (guard interval) eliminates inter-symbol interference from multipath. Advantages: robust against multipath fading, high spectral efficiency, easy equalization (single-tap per subcarrier), and flexible resource allocation. Used in WiFi, LTE, 5G NR, and DVB.

MIMO (Multiple-Input Multiple-Output) uses multiple antennas at both transmitter and receiver to improve performance. Key MIMO techniques: Spatial Multiplexing transmits different data streams on different antennas, multiplying throughput (up to min(Nt,Nr) streams). Spatial Diversity transmits the same data on multiple paths to combat fading, improving reliability. Beamforming focuses energy toward the receiver using phase/amplitude weighting. MIMO capacity grows linearly with min(Nt,Nr) in rich scattering environments. Modern systems use Massive MIMO with 64-256 antennas for significant capacity gains in 5G networks.

LTE (Long Term Evolution) architecture consists of two main parts: E-UTRAN (Evolved UMTS Terrestrial Radio Access Network) contains eNodeB (evolved Node B) base stations that handle all radio functions and connect directly to the core network, eliminating the RNC layer of 3G. EPC (Evolved Packet Core) includes: MME (Mobility Management Entity) for signaling and mobility; S-GW (Serving Gateway) for user data routing; P-GW (PDN Gateway) for connection to external networks; HSS (Home Subscriber Server) for user authentication. LTE is all-IP with no circuit-switched voice natively (VoLTE uses IMS). The flat architecture reduces latency to ~10-20ms.

An antenna radiation pattern is a graphical representation of antenna radiation properties as a function of direction. Key features: Main lobe indicates the direction of maximum radiation with its width (beamwidth) measured at -3 dB points. Side lobes are secondary radiation peaks; lower side lobes mean less interference. Back lobe shows radiation in opposite direction; front-to-back ratio indicates directivity. Null points show directions of no radiation. Patterns are shown in E-plane (containing E-field and direction of max radiation) and H-plane. Polar plots show 2D cuts; 3D plots show complete pattern. HPBW (Half-Power Beamwidth) is the angular width between -3 dB points.

Spread spectrum techniques spread the signal over a bandwidth much wider than necessary for the data rate. FHSS (Frequency Hopping Spread Spectrum): Signal hops between frequencies in a pseudo-random sequence; used in Bluetooth. Processing gain = number of hopping frequencies. DSSS (Direct Sequence Spread Spectrum): Data is multiplied by a high-rate spreading code (chips); used in GPS, CDMA. Processing gain = chip rate/data rate. Benefits include: resistance to narrowband interference, low probability of interception, resistance to jamming, multiple users can share same spectrum (CDMA), and reduced multipath effects. Trade-off is increased bandwidth requirement.

5G enabling technologies include: Massive MIMO with 64-256 antennas for high spectral efficiency and beamforming. Millimeter wave (mmWave) using 24-100 GHz bands for massive bandwidth but short range. Beamforming and beam tracking to compensate for mmWave path loss. Network slicing to create virtual networks for different use cases (eMBB, URLLC, mMTC). Small cells for dense deployment and capacity. Flexible numerology with scalable subcarrier spacing (15-240 kHz) for different requirements. LDPC and Polar codes for improved error correction. Edge computing for ultra-low latency. D2D (Device-to-Device) communication for direct connections.

Handoff (handover) maintains connectivity as users move between cells. Types include: Hard handoff: Break-before-make; connection to old cell dropped before new cell connected (GSM). Soft handoff: Make-before-break; simultaneous connection to multiple cells during transition (CDMA). Softer handoff: Between sectors of the same cell. Horizontal handoff: Between cells using same technology. Vertical handoff: Between different access technologies (WiFi to LTE). Handoff decisions based on signal strength, quality (BER), interference, and load balancing. Modern LTE uses hard handoff but with preparation to minimize interruption.

802.11b (1999): 2.4 GHz, DSSS, 11 Mbps max. 802.11a (1999): 5 GHz, OFDM, 54 Mbps max. 802.11g (2003): 2.4 GHz, OFDM, 54 Mbps, backward compatible with b. 802.11n/WiFi 4 (2009): 2.4/5 GHz, MIMO (up to 4 streams), 600 Mbps, channel bonding. 802.11ac/WiFi 5 (2013): 5 GHz only, MU-MIMO (downlink), 256-QAM, up to 6.9 Gbps, 80/160 MHz channels. 802.11ax/WiFi 6 (2019): 2.4/5 GHz, OFDMA, MU-MIMO (both directions), 1024-QAM, up to 9.6 Gbps, better for dense environments with BSS coloring and TWT for power saving.

Shannon's channel capacity theorem states: C = B × log2(1 + S/N), where C is maximum achievable data rate (bits/s), B is bandwidth (Hz), and S/N is signal-to-noise ratio (linear, not dB). Implications: Capacity increases linearly with bandwidth but logarithmically with SNR, meaning doubling bandwidth doubles capacity, but doubling SNR only adds about 3 bits/symbol. There exists a theoretical maximum rate below which error-free transmission is possible. Modern coding schemes like LDPC and Turbo codes approach this limit within 1 dB. For 1 MHz bandwidth and 20 dB SNR: C = 1M × log2(1+100) ≈ 6.66 Mbps maximum.

Impedance matching ensures maximum power transfer and minimizes signal reflections between source, transmission line, and load. When impedances are mismatched, part of the signal reflects back (quantified by reflection coefficient Γ = (ZL-Z0)/(ZL+Z0)), causing standing waves and power loss. VSWR (Voltage Standing Wave Ratio) measures mismatch severity; ideal is 1:1, acceptable is typically <2:1. Matching techniques include: L-networks (series-shunt components), stub matching, quarter-wave transformers, and Smith chart-based design. Standard RF impedance is 50Ω (compromise between power handling and loss for coax) or 75Ω (minimum loss for video).

Diversity techniques combat fading by providing multiple independent signal copies. Spatial diversity: Multiple antennas separated by λ/2 or more receive independent fading signals; MIMO exploits this. Time diversity: Same data transmitted at different times separated by more than coherence time; uses interleaving and coding. Frequency diversity: Same data on multiple frequencies separated by more than coherence bandwidth; spread spectrum and OFDM use this. Polarization diversity: Antennas with different polarizations (horizontal/vertical) provide independent channels. Combining methods: Selection combining (choose best), equal gain combining, and maximum ratio combining (MRC - optimal, weights by SNR).

BLE (Bluetooth Low Energy) was designed for IoT and wearables with ultra-low power consumption. Key differences: Power consumption is much lower in BLE (can run for years on coin cell). Connection time is faster (~3ms vs ~100ms). Data rate is lower (1-2 Mbps vs 3 Mbps). Topology supports broadcast and mesh (BLE Mesh) vs point-to-point. BLE uses 40 channels (2 MHz spacing) vs 79 channels (1 MHz) for Classic. Sleep current is ~1µA in BLE. BLE is optimized for small, infrequent data transfers (sensors, beacons), while Classic Bluetooth suits continuous streaming (audio). Both can coexist in dual-mode devices.

RF power amplifier classes differ in efficiency and linearity trade-offs. Class A: Conducts full cycle (360°), most linear but only 25-50% efficient; used for low-power linear applications. Class B: Conducts half cycle (180°), 78.5% max efficiency but has crossover distortion; push-pull configuration common. Class AB: Conducts slightly more than half cycle, compromise between A and B; common in base stations. Class C: Conducts less than half cycle, 80%+ efficient but non-linear; used for FM and switching. Class D/E/F: Switching amplifiers with 90%+ efficiency; Class E popular for RF with ZVS/ZCS. Modern systems use Doherty amplifiers combining classes for efficiency at back-off.

LTE uses OFDMA (downlink) and SC-FDMA (uplink) with a specific frame structure. Radio frame: 10ms duration containing 10 subframes. Subframe: 1ms duration containing 2 slots. Slot: 0.5ms containing 7 OFDM symbols (normal CP) or 6 symbols (extended CP). Resource Block (RB): 12 subcarriers × 1 slot = 84 resource elements (normal CP); smallest schedulable unit. Subcarrier spacing: 15 kHz. Bandwidth options: 1.4, 3, 5, 10, 15, 20 MHz (6 to 100 RBs). Cyclic prefix: Normal (4.7µs) or Extended (16.7µs) for high delay spread. Reference signals (pilots) enable channel estimation. This structure supports both FDD and TDD modes.

ZigBee is a low-power, low-data-rate wireless protocol based on IEEE 802.15.4, designed for IoT and smart home applications. Key features: Operates in 2.4 GHz (global), 915 MHz (Americas), and 868 MHz (Europe) bands. Data rate of 250 kbps (2.4 GHz). Supports mesh networking with up to 65,000 nodes. Low power consumption enabling years of battery life. Range of 10-100 meters. Uses DSSS and O-QPSK modulation. Network roles include Coordinator (one per network), Router (extends range), and End Device (low power). Applications include home automation, smart lighting, industrial sensors, and medical devices. Competes with Z-Wave, Thread, and BLE Mesh.

Polarization describes the orientation of the electric field vector of an electromagnetic wave. Types: Linear polarization has the E-field in one plane (horizontal or vertical). Circular polarization has the E-field rotating as the wave propagates (RHCP or LHCP). Elliptical polarization is general case with unequal axes. Cross-polarization discrimination measures isolation between polarizations. Polarization mismatch between Tx and Rx antennas causes loss (up to 3 dB for 45° mismatch, total loss for orthogonal). Circular polarization is used in satellite communications because it's immune to Faraday rotation and orientation changes. Mobile systems often use dual-polarized antennas (+45° and -45°) for diversity.

Multipath channel models mathematically describe how signals propagate through wireless environments with reflections, diffraction, and scattering. Key parameters: Power delay profile shows power vs delay of multipath components. RMS delay spread indicates channel's time dispersion; if >> symbol period, causes ISI. Coherence bandwidth (~1/delay spread) defines frequency selectivity. Doppler spread indicates time-varying nature from mobility. Coherence time (~1/Doppler spread) defines how fast the channel changes. Standard models include: Rayleigh (no LOS), Rician (with LOS), TDL (Tapped Delay Line), and 3GPP channel models. These models are essential for system simulation, equalizer design, and performance evaluation.

An RF front-end is the interface between the antenna and the baseband processor. Key components: Antenna for radiating/receiving RF signals. Duplexer/switch separates Tx and Rx paths (FDD) or switches between them (TDD). LNA (Low Noise Amplifier) amplifies weak received signals with low noise figure (typically 1-2 dB). PA (Power Amplifier) amplifies transmitted signals to required power level. Filters (bandpass, SAW, BAW) select desired frequency band and reject out-of-band signals. Mixer converts between RF and IF or baseband frequencies. VCO/PLL generates local oscillator frequencies. Modern smartphones integrate many components into FEMs (Front-End Modules) to save space.

QoS in wireless networks ensures different traffic types receive appropriate treatment. Techniques include: Traffic classification marks packets by priority (VoIP highest, best-effort lowest). Scheduling algorithms (round-robin, weighted fair queuing, priority queuing) allocate resources. Admission control limits new connections to protect existing services. Rate limiting prevents bandwidth monopolization. WiFi uses WMM (WiFi Multimedia) with four access categories (Voice, Video, Best Effort, Background) and different AIFS, CWmin, CWmax. LTE provides QCI (QoS Class Identifier) mapping to bearers with guaranteed/non-guaranteed bit rates. 5G introduces network slicing for end-to-end QoS isolation.

Massive MIMO (64-256+ antennas) introduces significant design challenges. Channel estimation overhead scales with users, not antennas, using TDD reciprocity; FDD requires excessive feedback. Pilot contamination from pilot reuse in adjacent cells creates interference floor even with infinite antennas; solutions include pilot coordination and blind estimation. Hardware complexity requires low-cost, low-power components with calibration for reciprocity; hybrid beamforming (analog + digital) reduces RF chains. Computational complexity of precoding/detection (ZF, MMSE) requires O(K³) operations; iterative detectors and machine learning approaches help. Channel hardening and favorable propagation assumptions may not hold in real deployments. Thermal management for dense antenna arrays is critical.

mmWave (24-100 GHz) faces severe propagation challenges. High path loss: FSPL increases with f², requiring ~20-30 dB more link budget at 28 GHz vs 2 GHz; mitigated by high-gain beamforming (256 elements can provide 24 dBi). Atmospheric absorption: O2 absorption peak at 60 GHz (~15 dB/km), rain attenuation (1-10 dB/km); limits outdoor range. Blockage: Human body causes 20-35 dB loss, materials vary widely; requires beam tracking, multi-connectivity, and dense deployment. Limited diffraction: Minimal propagation around corners; reflection-based NLOS paths critical. Foliage attenuation: Severe loss through vegetation. Solutions include beam tracking algorithms, multi-panel antennas, integrated access and backhaul (IAB), and cell densification with seamless handover.

Phased array antennas electronically steer beams by controlling phase (and amplitude) of individual elements. Design principles: Element spacing typically λ/2 to avoid grating lobes; closer spacing for wider scan angles. Phase shift per element: Δφ = (2π/λ) × d × sin(θ), where d is spacing and θ is scan angle. Array factor multiplies element pattern for overall pattern; narrow beam with N elements has beamwidth ≈ λ/(Nd). Implementation uses: phase shifters (3-6 bit resolution for <1 dB loss), variable gain amplifiers, and calibration for manufacturing variations. Analog beamforming is lower cost; digital gives full flexibility; hybrid combines both. Challenges include beam squint with wideband signals, mutual coupling between elements, and scan blindness at certain angles.

5G NR introduces flexible numerology to support diverse use cases. Subcarrier spacing (SCS): 15, 30, 60, 120, 240 kHz (μ = 0 to 4, SCS = 15 × 2^μ kHz). Slot duration scales inversely: 1ms at 15 kHz to 0.0625ms at 240 kHz, reducing latency. Frame structure: 10ms frame = 10 subframes, each 1ms. Slots per subframe = 2^μ. Each slot has 14 OFDM symbols (normal CP). Bandwidth parts (BWP) allow different numerologies within a carrier. Mini-slots (2, 4, 7 symbols) enable ultra-low latency. Cyclic prefix scales with SCS for consistent delay tolerance. FR1 (<7 GHz) supports 15-60 kHz; FR2 (mmWave) supports 60-120 kHz (data), 240 kHz (sync). Flexible slot formats with configurable DL/UL symbols support dynamic TDD.

RF impairments degrade signal quality and must be compensated. Phase noise: VCO/PLL instability causes constellation rotation and ICI in OFDM; improved PLL design, common phase error estimation, and CPE correction. I/Q imbalance: Gain/phase mismatch between I and Q paths creates image interference; compensated by calibration or digital pre/post-compensation. DC offset: Mixer LO leakage creates DC component; removed by DC blocking or digital subtraction. PA nonlinearity: Causes spectral regrowth and EVM degradation; DPD (Digital Pre-Distortion) linearizes PA, memory polynomial models popular. Carrier frequency offset: Due to oscillator mismatch; compensated by frequency synchronization algorithms. Sample timing offset: Causes ISI; compensated by timing recovery and interpolation. ADC/DAC quantization: Causes noise floor increase.

5G NR uses LDPC for data and Polar codes for control channels. LDPC (Low-Density Parity-Check): Sparse parity-check matrix enables parallel decoding. 5G uses quasi-cyclic LDPC with base graphs (BG1 for large blocks, BG2 for small). Iterative belief propagation decoding with excellent performance at moderate-high SNR. Better for larger block sizes (>500 bits). Polar codes: Based on channel polarization; some bit channels become reliable, others unreliable. Frozen bits placed on unreliable channels. Successive cancellation (SC) decoding is basic; SC List (SCL) with CRC improves performance. Better for small block sizes and low latency. Both approach Shannon limit within ~0.5 dB. LDPC has lower decoding latency for high throughput; Polar has more flexible rate matching.

Inter-cell interference is a major capacity limiter in cellular networks. ICIC (Inter-Cell Interference Coordination): LTE Rel-8 coordinates frequency resources across cells; cell-edge users get protected subcarriers. eICIC (enhanced ICIC): LTE Rel-10 uses Almost Blank Subframes (ABS) for time-domain coordination in HetNets; small cells transmit during macro ABS. FeICIC: Adds interference cancellation capability at UE. CoMP (Coordinated Multi-Point): Joint transmission from multiple cells or coordinated scheduling/beamforming; requires backhaul and synchronization. Massive MIMO: Spatial isolation through narrow beams reduces interference naturally. Network MIMO/C-RAN: Centralized processing enables joint precoding. Machine learning approaches optimize resource allocation dynamically based on traffic patterns.

OFDMA resource allocation assigns subcarriers and power to maximize throughput while ensuring fairness. Strategies: Maximum throughput allocates resources to users with best channel conditions; unfair to cell-edge users. Proportional fair (PF) schedules based on instantaneous/average rate ratio; balances throughput and fairness. Round-robin ensures equal resources but ignores channel conditions. Water-filling power allocation assigns more power to better channels (capacity-optimal). Practical constraints include minimum rate requirements, maximum power limits, and discrete resource blocks. Cross-layer design considers queue states, QoS requirements, and interference. Channel-dependent scheduling exploits multiuser diversity; gains increase with users. Modern systems use machine learning for dynamic resource management in complex scenarios.

Full-duplex (simultaneous transmit and receive on same frequency) doubles spectral efficiency but faces severe self-interference. Self-interference cancellation (SIC) required in three domains: Antenna domain: Isolation through antenna separation, polarization, and directional patterns; ~40-50 dB achievable. Analog domain: RF canceller samples Tx signal, adjusts delay/amplitude, and subtracts; handles PA nonlinearity; ~40-50 dB. Digital domain: Adaptive filters cancel residual interference; ~30-40 dB. Total ~110-130 dB cancellation needed (Tx power vs noise floor gap). Challenges include: Tx signal dynamics across bandwidth, PA nonlinearity modeling, ADC dynamic range requirements, phase noise, and real-time adaptation. Applications include relays, small cells, and radar. Full-duplex gains diminished by increased network interference in cellular deployments.

RIS (also called intelligent reflecting surfaces) are planar surfaces with many passive reflecting elements whose properties can be electronically controlled. Each element adjusts phase (and potentially amplitude) of incident waves, enabling intelligent wavefront manipulation. Benefits: Create virtual LoS paths in blocked environments. Enhance coverage without power amplification (passive). Low power consumption (only control circuits). Beamforming and focusing without traditional antenna arrays. Challenges: Large number of elements (hundreds to thousands) for meaningful gain. Channel estimation for many elements with passive operation. Joint optimization of transmitter and RIS configuration. Hardware implementation with sufficient phase resolution and speed. Near-field effects at close range. Integration with existing networks. Research areas include channel modeling, optimization algorithms, and practical deployment strategies for 6G.

WiFi 6 (802.11ax) combines OFDMA and MU-MIMO for efficient multi-user access. OFDMA divides the channel into Resource Units (RUs): smallest 26 tones (~2 MHz), largest 996 tones (~80 MHz) in 80 MHz channel. Multiple users served simultaneously on different RUs, reducing latency for small packets. MU-MIMO serves up to 8 users (up/down) on the same RU using spatial streams. Combined operation: Channel can be divided into RUs, with MU-MIMO within each RU. Scheduler decides OFDMA vs MU-MIMO based on traffic size, user capabilities, and channel conditions. Trigger-based uplink OFDMA coordinates transmissions. BSS coloring reduces interference from neighboring networks. TWT (Target Wake Time) schedules transmissions for power saving. This combination significantly improves efficiency in dense deployments.

Mutual coupling occurs when nearby antenna elements induce currents in each other, affecting array performance. Effects: Changes element input impedance (active impedance depends on excitation of all elements). Alters individual element patterns, affecting array factor. Can cause scan blindness at certain angles where surface waves are excited. Reduces efficiency due to power coupled to adjacent elements. Modifies array radiation pattern and gain. Analysis/mitigation: Full-wave simulation (MoM, FEM) for accurate modeling. Coupling matrix characterization enables compensation in signal processing. Element spacing >λ/2 reduces coupling but creates grating lobes. Decoupling networks can improve isolation. EBG (electromagnetic bandgap) structures suppress surface waves. Proper calibration accounts for coupling in beamforming weights. Modern massive MIMO systems must include coupling in channel estimation and precoding.

Network slicing creates multiple virtual end-to-end networks on shared physical infrastructure. Implementation layers: RAN slicing: Allocates radio resources per slice through scheduling and resource reservation. Transport slicing: Uses SDN/NFV for traffic isolation, QoS differentiation. Core slicing: Dedicated or shared network functions per slice; AMF, SMF, UPF can be slice-specific. Key identifiers: S-NSSAI (Single Network Slice Selection Assistance Information) identifies slices; contains SST (Slice/Service Type) and optional SD (Slice Differentiator). Selection process: UE requests slice(s), AMF validates and selects based on subscription and availability. Slice types: eMBB (high throughput), URLLC (low latency), mMTC (massive IoT). Challenges: Resource isolation guarantees, slice lifecycle management, cross-slice optimization, and SLA enforcement. Orchestration uses NFV MANO frameworks.

Terahertz (0.1-10 THz) communication offers extreme bandwidth for 6G but faces significant challenges. Opportunities: Hundreds of GHz of available spectrum. Tbps data rates possible. High-resolution sensing and imaging. Secure communication due to narrow beams. Challenges: Extreme path loss (spreading + molecular absorption; H2O, O2 peaks at specific frequencies). Very short range (meters to tens of meters indoor). Severe blockage by obstacles and even humidity. Limited Tx power from solid-state devices; novel sources needed. High-gain antennas required (easier due to small λ). Component technology immature (mixers, amplifiers, modulators). Channel modeling poorly understood. Applications: Short-range high-capacity links, kiosk downloading, data centers, nano-networks. Research focuses on transmission windows (around 300 GHz initially), novel materials, and hybrid RF-THz systems.

O-RAN (Open Radio Access Network) disaggregates and virtualizes the RAN with open interfaces. Architecture splits: RU (Radio Unit) handles RF functions; O-DU (Distributed Unit) handles real-time L1/L2; O-CU (Central Unit) handles non-real-time L2/L3. Open fronthaul (7.2x split) between RU and DU enables multi-vendor interoperability. RIC (RAN Intelligent Controller): Near-RT RIC (<10ms) hosts xApps for radio resource management. Non-RT RIC (>1s) hosts rApps for policy, ML model training. SMO (Service Management and Orchestration) manages the network. Advantages: Vendor diversity reduces lock-in and costs. Innovation through xApps/rApps ecosystem. Cloud-native deployment on COTS hardware. AI/ML integration for network optimization. Challenges: Performance of virtualized RAN, fronthaul bandwidth requirements, multi-vendor integration testing, and security of open interfaces.