Understanding Multiplexing Schemes: TDM, FDM, and CDMA in Computer Networks
The Physical Layer – Multiplexing Details
CSC 570 Computer Networks Fall 2016
Motivation
Multiplexing schemes are used to share communication lines among many signals.
In a way, it is similar to people having many conversations in the same room. How?
Multiplexing Schemes
- Time Division Multiplexing (TDM)
- Frequency Division Multiplexing (FDM)
- Code Division Multiplexing (CDM)
Time Division Multiplexing
It’s like people taking turns in speaking. TDM shares a channel over time.
Users take turns on a fixed schedule; each using a fixed timeslot
User streams must be synchronized in time
Guard time to accommodate small timing variations
Problem
What happens if a user has nothing to transmit?
Answer: His time slot is wasted. Other users cannot use the empty slot.
Solution: Statistical Time Division Multiplexing (STDM)
STDM
Individual streams contribute to the multiplexed stream according to their demand. Not on a fixed schedule
Another name for packet switching
Multiplexing Schemes
- Time Division Multiplexing (TDM)
- Frequency Division Multiplexing (FDM)
- Code Division Multiplexing (CDM)
FDM
It’s like pairs of people using different pitches when speaking
Shares the channel by placing users on different frequencies:
Guard Band
Keeps the individual channels well separated to limit interference
It is possible to divide the spectrum without using guard bands. How?
Orthogonal Frequency Division Multiplexing (OFDM)
OFDM
Carriers are packed tightly–no guard bands.
How is this possible? What about interference?
Notes
The data is distributed among a large number of carriers instead of using one carrier.
Knowing Fourier transform of digital data and careful design leads to a frequency response of each carrier s.t. it is 0 at the center of adjacent carriers. no interference at center frequencies.
Error-coding techniques — adding further data to the transmitted signal –enables reconstruction of the corrupted data
Multiplexing Schemes
- Time Division Multiplexing (TDM)
- Frequency Division Multiplexing (FDM)
- Code Division Multiplexing (CDM)
CDM or CDMA
It’s like pairs of people using different languages when speaking
Allows signals from multiple users to share the same frequency band – That’s why it is also know Code Division Multiple Access (CDMA). Widely used in 3G/4G cellular networks. Why is this multiple access important/interesting?
CDMA Advantages (1/3)
Think cellular technology … Soft handoff
CDMA Advantages (2/3)
1. Better channel utilization than TDM and FDM. Why?
In TDM and FDM, A user is assigned time slots or frequency bands. if a user is not actively sending data (pausing when talking over cellular phone), his assigned time slots/frequency bands are wasted – not used by others. Not the case with CDMA
CDMA Advantages (3/3)
- Simplifies time slot/frequency bands planning between the cell phones and towers (base stations) and between the towers.
- Facilitates soft handoff in which the mobile is acquired by the new base station before the previous one signs off. No need to reserve a frequency on the next hop
How is CDMA implemented?
In order to share the channel, each user is assigned a different code (chip sequence). What is that?
Each bit time is subdivided into m short intervals called chips. m is typically 64 or 128
Each station (cell phone) is assigned a unique m-bit code (chip sequence)
Chip Sequences (1/3)
Defined using bipolar notation with +1 and -1
Example: Assume m=8
Station A is assigned a chip sequence of
(-1 -1 -1 +1 +1 -1 +1 +1)
Other stations are assigned different chip sequences
Chip Sequences (2/3)
To transmit a 1 bit: the station sends its chip sequence
(-1 -1 -1 +1 +1 -1 +1 +1)
To transmit a 0 bit: the station sends the negation of its chip sequence
(+1 +1 +1 -1 -1 +1 -1 -1)
It is really signals with these voltage levels that are sent
Chip Sequences (3/3)
Codes are pairwise orthogonal (?); and can be sent at the same time
Two sequences are orthogonal iff their normalized inner product is 0.
Orthogonal codes are generated using a method known as Walsh codes.
Orthogonality (1/4)
Let S be the chip sequence for station A Let S’ be its negation
Let T be the chip sequence for station B Let T’ be its negation
S and T are orthogonal iff
. ≡∑=0
Orthogonality (2/4)
. =0also implies
. ′ =0 . =1
. ′ =−1
Example
S=(-1-1-1+1+1-1+1-1) T=(-1 -1 +1 -1 +1 +1 +1 -1)
Compute S.T, S.T’, S.S, and S.S’
How does CDMA work? (1/2)
Consider 4 stations, with the following codes, transmitting at the same time.
How does CDMA work? (2/2)
If someone (cell tower) wants to decode the transmission from user C, then given his knowledge of C’s code and the received signal, he computes the inner product to decode C’s signal
CDMA Bandwidth requirements
Increasing the amount of information to be sent from B bits/s to mB chips/s increases the bandwidth (Hz) requirements of CDMA m fold. (Nyquist Equation)
We have seen a similar scenario in Manchester line code
In this case, that’s ok since now each user can use the whole frequency spectrum used by all other users.
CDMA Assumptions (1/2)
1. All signals are synchronized on the start time of their chip sequences – Not practical
Solution:
Instead of using perfectly orthogonal codes, rely on codes from pseudo-random sequences. These are known to be close to orthogonal (low cross-correlation, and low auto- correlation). Lets the base stations receive CDMA messages from unsynchronized mobiles
CDMA Assumptions (2/2)
1. The power levels of all mobiles are the same at the receiver. If not, then a powerful signal will interfere with low power signals.
Solution:
The power levels must be controlled to minimize interference. A good heuristic is for each mobile to transmit to the base station at the inverse of the power level it receives from the base station. There are also other solutions
Next Lecture
1. The Physical Layer – Media and Systems