2-Signalling Flashcards
What does a signal do?
A signal transfers information from one location to another.
What are the 2 ways signals can be converyed?
Through sound or light
What are digitised signals?
Signals of a string of binary digits from the sender to the reciever.
What is the number of bits required?
number of bits per character X number of characters per page X number of pages
What is 1 Byte?
8 Bits
What is an analogue signal?
Signals which vary from one value to the next without fixed values.
For most of the 20th century, what were most long distance communication systems examples of?
Analogue signals
What happens in an analogue telephone?
A sound vibration changes into matching oscillations of P.D.
What is the biggest problem with analogue signals?
The need for amplification as the signal becomes weaker
If the signal becomes distorted or ‘noisy’ what does the amplification do?
It boosts the signal and the noise.
What is the noise in relation to signals?
The random variation (the spiky parts)
What happens if you filter out the noise?
You lose a lot of detail in the signal.
What can transmit information quicker: digital signals or analogue signals?
Digital signals
What is sampling?
The process in which the signal is measured at small time intervals and turned into a digital string of binary numbers.
What is analogue to digital conversion?
When you convert varying analogue to a stream of numbers
What is the quantisisation error?
The difference between the signal value and the quantisation level.
What is the resolution of a sample?
The smallest change in P.D that can be determined.
What is the equation for resolution?
P.D range of signal / number of quantisation levels
How do you calculate the number of quantisation levels possible?
2 to the power of the number of bits
What is the resolution of a sample measured in?
Volts
What does increasing the number of quantisation levels do?
It increases the demands on data storage and transmission.
What is the equation for the maximum useful number of levels?
Total noisy signal variation / noise variation
b=log2( Vtotal/Vnoise )
When working out the maximum number of useful number levels, do you round up or down?
You round up
What is a frame rate?
The time interval between each frame
What is the problem of not taking enough samples per second?
The signal will not be fully accurate
What is the sampling rate?
The number of samples taken each second
For a varying signal, what does the sampling rate have to be?
Sampling than the time interval between important changes in the signal.
What happens if the sampling rate is too small?
The signal will lose detail
What does the minimum sampling rate have to be?
Greater than 2 X the highest frequency component
To ensure that the original signal can be met, what does the frequency have to be below?
A certain maximum
What happens if the signal contains frequencies above a certain maximum?
The signal will not be constructed accurately
What are aliases?
Lower frequecy signals which are not present in the original signal.
As the human ear cannot detect sound above frequencies of 20kHz, what does the sampling frequency of music have to be?
Greater than 40kHz
What is bit rate measured in?
Bit/second or Hz
How do you calculate the bit rate?
Samples per second X bits per sample
How do you calculate the duration of a signal?
Number of bits in signal / Bit rate
As digital signals are numbers, what can happen to them?
They can be changed or scrambled
Why is there no point using a voltage resolution better than the noise resolution of the signal?
You end up digitising noise
Why is the number of bits b to use when digitising a signal is given by 2b = Signal / Noise ratio?
There is no point using a voltage resolution better than the noise variation of the signal, as you just end up digitising noise, thus we can write:
Noise variation = Voltage resolution
So Voltage resolution = Signal variation / 2b where b is the number of ADC bits
So Noise variation = Signal variation / 2b , which rearranges to give Signal variation / Noise variation = 2b
