COMP90007 Chap.2 Signals, Media and Channel Capacity
Signals, Media and Channel Capacity
The physical layer makes electrical, timing and mechanical choices visible as a link. Bandwidth is a transmission rate, while delay describes time. A frame's transmission delay grows with its bit length and falls with rate; propagation delay grows with distance and falls with signal speed.
Media change attenuation, interference, reach, installation constraints and available spectrum, so a nominal rate is never the only selection criterion. Digital modulation maps bits onto physical signal states. More distinguishable levels can carry more bits per symbol, but noise limits how finely a receiver can discriminate them. Nyquist formalises the noiseless relationship among bandwidth, signal changes and levels.
Shannon instead uses signal-to-noise ratio to cap a noisy channel. The correct maximum is constrained by the assumptions that actually describe the channel. Multiplexing then shares a medium by allocating time, frequency, statistical opportunity or code. These methods solve an allocation problem, not the separate question of how each user's bits are represented.
What this chapter covers
- 01
Transmission delay
- 02
Attenuation
- 03
Multiplexing
- 04
Bandwidth, transmission delay and propagation delay
- 05
Attenuation and physical-media trade-offs
- 06
Signal states and digital modulation
- 07
Nyquist's noiseless-channel ceiling
- 08
Shannon's noisy-channel ceiling
- 09
Duplex modes and multiplexing
Choose the tighter channel ceiling
- 1Four levels carry log base two of four, or two, bits per symbol.
- 1Nyquist gives two times 3,000 times two, which is 12,000 bit/s.
- 1The Shannon ceiling is 9,000 bit/s, lower than the Nyquist ceiling.
- 1The channel cannot exceed the tighter 9 kbit/s bound under the stated assumptions.
Key terms
- Transmission delay
- Frame length in bits divided by transmission rate in bits per second.
- Attenuation
- Reduction in signal amplitude as it travels through a medium.
- Multiplexing
- Sharing a medium among users by separating their transmissions in time, frequency, code or statistical opportunity.
Signals, Media and Channel Capacity FAQ
Can faster bandwidth eliminate every network delay?
No. It shortens the time required to put a message onto a link, but distance still determines propagation time. Processing and queueing may also dominate. Improvement requires identifying the largest term rather than treating latency as another name for rate.
Why does adding signal levels have a limit?
Each extra level can encode more bits per symbol, but adjacent states become harder to distinguish. In a noisy environment, the receiver's uncertainty imposes a ceiling captured by Shannon's relationship, regardless of the nominal number of levels selected.
Exam move
Keep a dimension table beside each physical-layer problem. Mark message quantities in bits, rates in bits per second, distances in kilometres or metres, and speeds in the matching distance per second. For conceptual comparisons, write which impairment or allocation problem each medium or multiplexing method is intended to manage. Maintain a dimensional ledger for every physical-layer calculation.
Write message length in bits, rate in bits per second, distance in kilometres or metres and propagation speed in matching distance units before selecting an equation. Calculate transmission and propagation delay on separate lines, then add only quantities expressed as time. Change one input at a time and state which term responds; this makes it clear why a faster rate cannot shorten travel through a long medium.
Build a media comparison that includes attenuation, interference, usable spectrum, reach, installation, directionality and mobility rather than ranking links by nominal rate alone. For modulation, draw the permitted signal states and convert levels into bits per symbol before discussing capacity.
Work Nyquist and Shannon as rival upper-bound questions: identify whether the premise is a noiseless signalling model or a noisy channel with a stated signal-to-noise ratio, convert decibels only when required, and never add the two limits. If a calculated ceiling exceeds a supplied implementation rate, explain that a bound is not guaranteed throughput.
For multiplexing, sketch how time, frequency, statistical opportunity or code separates users, then identify the condition under which an allocation sits idle or produces interference. Finish each problem with a sentence naming the binding assumption and a sensitivity check. Recalculate after doubling frame size, path length, bandwidth or noise independently, and explain the direction of change before trusting arithmetic.
Keep the written interpretation beside the numbers so an apparently precise result cannot hide a unit error or an inapplicable model. Record significant figures only after the model is chosen, and estimate the expected order of magnitude first. A result far outside that estimate should trigger a unit and assumption audit before interpretation.
Working through Signals, Media and Channel Capacity in COMP90007? Sia is AskSia’s AI Computer Science tutor — ask any COMP90007 Signals, Media and Channel Capacity question and get a clear, step-by-step explanation grounded in how COMP90007 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
COMP30023 Computer Systems · COMP90038 Algorithms and Complexity · COMP20008 Elements of Data Processing
