Digital Communications Overview, ASK, FSK Prepared by: Keyur Desai Department of Electrical Engineering Michigan State University ECE458
Why Digital Communications? How do you place a call from Lansing to Fiji? Your call travels through too many telephone networks and towers, underwater cables, microwave links... If signal analog then once noise gets in its hard to get rid off. Why?
Why Digital Communications? Analog -->Sample-->Quantized --> Digitized 10101010101010000011000000 All we have to make sure that the bits get through well. Its lot easier than trying to get a continuous time continous value waveform get through well. During the initial digitization process we take some hit in terms of quality, but unnoticable for most of the applications. Three biggest advantages of Digital 1. You can store easily 2. You can compress 3. You can add extra bits to protect the bit string from errors
Why Digital Communications? Here comes Shannon Shannon finds: based on the statistical property of a bit stream how much one can compress Shannon finds: based on the statistical property of the channel how many extra bits are needed to get the message across reliably
What is digital modulation Small part of the big set-up Imagine: on your bit stream all the compression and redundancy addition are done and now the bit stream is ready to be transmitted 100000100000111111111110000000000 Compress 100100110000 Add redundancy 100100110000101010101010101 Transmit Digital Modulation is all about getting this bit stream to the other point as safe and sound as possible. There will be errors: success of a digital modulation scheme in terms of probability of error and bandwidth efficiency Invent a scheme that minimizes the number of errors and then shannon redundancy will take care of the remaining errors Why sinusoids in digital modulation?
Why Digital Communications? How do we handle the binary messages Map 10110111101 to voltage levels Remember why modulation was needed 1. To make message suitable for transmission 2. To share the spectrum (use the same medium for multiple users) Does the message have DC level? Then message can be seen as DC + bipolar (+1, -1) An AM signal where the message is binary leveled
ASK What is the bandwidth of?
ASK Bandlimiting of ASK signal Demodulation is a two step process 1 Strip off the bandlimited digital message 2 Reconstruct the binary leveled message
ASK Generation
ASK Recovery ASK signal is similar to AM signal and what we use to recover an AM signal?
ASK Decision Convert envelope detector output to bipolar Use decision maker
What Graphs? ASK waveform and the message in time domain ASK waveform in the frequency domain Calculate the bandwidth of ASK signal Bandlimited ASK signal in time domain Recovered message and the message after decision maker in time domain Try adding noise and see how many errors.
FSK Frequency shift keying Keying suggest that message is of ON-OFF variety f1 for 0 and f2 for 1
FSK demodulation Visualize FSK as two ASKs
FSK generation Continuous phase FSK. Why?
FSK Demodulation Asynchronous. Two ASKs.
FSK What Graphs 1. Using VCO FSK signal and the message signal time domain FSK signal and the message signal frequency domain 2. Using Analog Switch FSK signal and the message signal time domain FSK signal and the message signal frequency domain Calculate bandwidth in both the cases Asyncrhonous demodulation 1. Recovered Mark and Space in time domain PLL demodulation 1. Message output from the PLL set up and the original message: time domain
What Questions to ask yourself? Is ASK similar to AM? How is power efficiency of ASK? How is bandwidth efficiency of ASK? What is the probability of error in ASK? Has FSK any advantages over ASK? Is FSK bandwidth efficient? Is FSK power efficient? Has FSK smaller probability of error over ASK? If yes, then why? What is the immidiate improvement over ASK? What do you know about BPSK?