Physical Layer Practice Questions

Fiber optic cables transmit data as light pulses, offering the highest bandwidth (terabits/sec), very low attenuation, and complete immunity to electromagnetic interference (EMI).

Physical Layer functions: A: TRUE - represents binary 0s and 1s as physical signals (voltage levels for copper, light pulses for fiber, electromagnetic waves for wireless). B: TRUE - defines the data rate (number of bits transmitted per second). C: TRUE - bit synchronization ensures the receiver samples bits at the correct time using clock synchronization techniques (self-clocking codes like Manchester encoding, or separate clock signals). D: FALSE - framing is a Data Link Layer function; the Physical Layer transmits raw bits without frame structure awareness.

Manchester encoding has a transition in the middle of every bit period: low-to-high = 1, high-to-low = 0. It provides self-clocking (synchronization), used in classic Ethernet (10BASE-T).

Shannon's channel capacity: C = B × log₂(1 + S/N) = 3100 × log₂(1 + 1023) = 3100 × log₂(1024) = 3100 × 10 = 31,000 bps = 31 kbps. Note: log₂(1024) = log₂(2^10) = 10. This is the theoretical maximum - no error-correcting code can achieve higher throughput on this channel regardless of the modulation scheme. SNR is often given in decibels: SNR_dB = 10 log₁₀(S/N). SNR_dB = 10 × log₁₀(1023) ≈ 30 dB.

FDM (Frequency Division Multiplexing): the available bandwidth is divided into non-overlapping frequency bands (subchannels). Each source signal modulates a different carrier frequency (subcarrier), and all modulated signals are transmitted simultaneously on the physical medium. Guard bands separate adjacent channels to prevent interference. Examples: AM/FM radio (each station uses a different frequency), cable TV (each channel uses a different frequency range), traditional telephone (30 calls per 2 MHz band using 64 kHz per channel). FDM is used in both analog and digital systems. TDM (Time Division Multiplexing) matches option B. CDMA matches option C.

Nyquist's theorem for a noiseless, ideal channel: Maximum data rate = 2B × log₂M bps, where B is the channel bandwidth in Hz and M is the number of discrete signal levels. The factor 2B represents the Nyquist rate - the maximum number of signal samples per second without intersymbol interference (ISI). Each sample carries log₂M bits. Example: B = 3 kHz, M = 8 levels: max rate = 2 × 3000 × log₂8 = 6000 × 3 = 18,000 bps = 18 kbps. Nyquist applies to noiseless channels; Shannon's theorem extends this to noisy channels.

Frequency band: a contiguous range of frequencies allocated to a radio service (e.g., the 2.4 GHz ISM band spans 2.400-2.4835 GHz, total 83.5 MHz). Channel: a specific frequency subdivision within a band allocated for one communication link. In 802.11b/g WiFi at 2.4 GHz: 14 channels defined (varies by country), each 22 MHz wide with 5 MHz spacing - channels overlap. Non-overlapping channels in 2.4 GHz: 1, 6, 11 (in North America). In 5 GHz band: more channels, less overlap. In IEEE 802.11ac/ax (WiFi 5/6), channels can be 20/40/80/160 MHz wide (channel bonding for higher throughput).

Manchester encoding (IEEE 802.3 definition): each bit period has a transition in the middle. Bit 1: low-to-high transition (rising edge) at the midpoint. Bit 0: high-to-low transition (falling edge) at the midpoint. Advantages: self-clocking - the mid-bit transition provides continuous synchronization signal; no DC component (equal positive and negative time). Disadvantages: requires twice the bandwidth of NRZ (signal changes twice per bit period). Used in: 10BASE-T Ethernet, token ring. Differential Manchester (used in token ring): transition at start of period encodes 0; no transition encodes 1; mid-bit transition always occurs for clocking.

NRZ-L (Non-Return-to-Zero Level): the signal level directly represents the bit value throughout the entire bit period, without returning to zero. Typically: positive voltage = 1, zero/negative voltage = 0 (or vice versa). Advantages: simple, efficient (1 bit per baud). Disadvantages: long strings of identical bits create synchronization problems (no transitions = receiver loses clock sync) and a DC component. Problem with NRZ-L: consecutive 1s or 0s produce a flat line making it hard to detect bit boundaries. NRZ-I (NRZ-Inverted) differs by encoding 1 as a transition and 0 as no transition - better for long runs of 1s but still problematic for long 0 runs.

Bandwidth-delay product (BDP) = bandwidth × round-trip propagation delay, or bandwidth × one-way delay depending on context. Using one-way delay as specified: BDP = 10 × 10⁶ bps × 50 × 10⁻³ s = 500,000 bits = 500 Kbits. The BDP represents the amount of data 'in flight' in the network pipe - the number of bits that can be transmitted before the first bit reaches the receiver. It determines the required buffer size and window size for efficient use of the link. For a 50ms RTT: BDP = 10 × 10⁶ × 0.05 = 500,000 bits for one-way, or 1,000,000 bits for round-trip. The pipe can hold 500,000 bits before the sender receives any acknowledgment.

Three main transmission impairments: (1) Attenuation: signal energy decreases as it travels through the medium (due to resistance in copper, absorption in fiber). Measured in decibels (dB). Remedy: amplifiers (analog) or repeaters (digital). (2) Distortion: signal shape changes because different frequency components travel at different speeds through the medium (frequency-dependent propagation velocity). Especially affects composite analog signals. Remedy: equalization. (3) Noise: unwanted electrical energy added to the signal. Types: thermal noise (random electron motion), intermodulation noise (sum/difference frequencies), crosstalk (signal from adjacent wire), impulse noise (lightning, switching). Noise is quantified by SNR (Signal-to-Noise Ratio). Remedy: shielding, filtering, error correction.

Twisted pair cables: each pair of wires is twisted together at a specific twist rate (twists per meter). The physics: external electromagnetic interference (EMI) induces approximately the same noise in both wires of a pair (common-mode noise). Since the wires carry opposite polarity signals, the differential signal (difference between the two wires) cancels out the common-mode noise - this is differential signaling. The twisting also reduces crosstalk between adjacent pairs by ensuring that each wire alternates between being closer to and farther from the interfering pair. More twists per meter = better noise rejection. Categories: Cat3 (10 Mbps), Cat5 (100 Mbps), Cat5e (1 Gbps), Cat6 (10 Gbps up to 55m), Cat6a (10 Gbps up to 100m).

A T1 line (North American standard) carries 24 DS0 channels (each 64 Kbps), plus framing bits: $24 \times 64 + 8 = 1544$ Kbps = 1.544 Mbps. E1 (European) carries 30 voice channels at 2.048 Mbps.

Attenuation is the reduction in signal strength as it travels over a medium, due to resistance, absorption, etc. It is measured in decibels (dB) and increases with distance and frequency.

Optical fiber: transmits data as pulses of light through a glass or plastic core. Advantages: extremely high bandwidth (terabits per second over long distances), immune to electromagnetic interference (EMI) and radio frequency interference (RFI), low signal attenuation (signal travels farther without amplification), no electrical crosstalk, secure (hard to tap without detection), lightweight. Disadvantages: more expensive than copper, requires specialized connectors and splicing equipment, fragile. Types: single-mode fiber (SMF) for long-distance, smaller core (9μm); multi-mode fiber (MMF) for shorter distances, larger core (50/62.5μm). Used in: backbone networks, undersea cables, FTTH (fiber to the home).

Propagation delay = Distance / Speed = $10^6$ m / $2 \times 10^8$ m/s = $5 \times 10^{-3}$ s = 5 ms.

OFDM (Orthogonal Frequency Division Multiplexing): divides the channel bandwidth into N narrow orthogonal subcarriers (e.g., 802.11a uses 52 subcarriers over 20 MHz). Each subcarrier is independently modulated (BPSK, QPSK, 16-QAM, 64-QAM, 256-QAM) based on channel conditions (adaptive modulation). Orthogonality: subcarriers are spaced such that the peak of each coincides with the zeros of all others - no inter-carrier interference (ICI) without guard bands. Advantages: robust against multipath fading and frequency-selective fading (each narrow subcarrier experiences flat fading), efficient spectrum use. Used in: WiFi (802.11a/g/n/ac/ax), LTE/5G, DVB-T (digital TV), DSL (DMT). OFDMA (multi-user OFDM) allocates different subcarriers to different users.

Single-Mode Fiber (SMF): core diameter ~9 μm (very small). Supports only ONE propagation mode (the fundamental mode). Light travels in essentially a straight line along the fiber axis. No modal dispersion - very high bandwidth over very long distances (100+ km without amplification). Uses laser light source. Used in: long-haul telecommunications, undersea cables, metro networks. Multi-Mode Fiber (MMF): core diameter 50 or 62.5 μm (larger). Supports MANY propagation modes - light rays enter at different angles and travel different paths. Modal dispersion occurs (different modes arrive at different times, limiting bandwidth). Uses LEDs or VCSEL lasers. Maximum distance ~2 km at 1 Gbps. Used in: building/campus networks, data centers (short runs). B and C are reversed from reality; D is also reversed.

QAM-64 has 64 constellation points. Each symbol represents $\log_2(64) = 6$ bits. Higher QAM orders (256-QAM, 1024-QAM) encode more bits per symbol but require better SNR.