A reader Sanket K. requested ,
” Can you publish a blog post on TDD/ FDD concepts and which companies use what and their take as to why they choose it from business/ technological point of view.”
The concept of TDD and FDD can be understood simply by watching this short video.
Now second part of the question asks why would an operator makes a decision to go for FDD over TDD and vice versa?
In simple words, the demand for capacity is so huge that operators would use all of the spectrum available. If TDD or FDD spectrum is available. Keeping this information in mind LTE as a global standard was introduced with the possibility of TDD deployment as well.
Asian markets such as China and India have been great markets for TDD LTE deployment. The rest of the world comes to learn that more than 2.6 billion people of the world are using TDD.
Following this trend a number of regions are planning to deploy TDD and make use of the available spectrum to meet the capacity requirements.
Sprint in USA is already using LTE TDD in addition to LTE FDD market.
Qualcomm has a full presentation on how much LTE TDD is being deployed currently all over the globe.
Having said all that, in addition, there are a number of factors which can affect the decision of an operator to choose TDD over FDD. These decisions involve
If we look at it from an overall evolved packet system (EPS) architecture perspective. In order to deploy TDD there are some changes needed
not exactly at the architecture level, instead at various levels of the overall system to accommodate TDD aspects. A diagram explaining those
changes which give major and minor impact for FDD LTE system in order to use it as TDD is shown below
As evident from the diagram above. For TDD LTE , major important impacts are at MAC level and physical layer level. Other parts of the
system are not impacted at a larger scale. So from deployment perspective operators have different configuration at major impacted areas.
Remaining deployment remains as it is.
There are a number of technical factors which can be taken into consideration when it comes to TDD and FDD technology utilization. I will
discuss each factor briefly below.
FDD has inherently low latency compared to TDD. In case of FDD, both UL and DL spectrum are transmitting and receiving simultaneously.
Whereas in case of TDD, simultaneous transmissions are not possible. In case of TDD, when DL is transmitting , UL transmission cannot be
initiated until UL sub-frame timing is available. This will cause TDD to have increase in latency compared to FDD. Therefore, FDD has an
advantage to be deployed in those markets, where the requirements for low latency is crucial compared to TDD.
FDD is the winner in this case.
In case of FDD, the end user devices (your smartphones in this case) have to switch from DL to UL and UL to DL respectively. This requires a
diplexer to be installed in end users’ devices. Installation of Diplexer increases the cost of the end user device.
Whereas for TDD scenario, there is no need for a diplexer. This results in reducing the price of the device for TDD. So you can expect TDD
end user devices to be comparatively affordable compared to FDD.
TDD is the winner in this case.
However, vendors now manufacture devices which include both FDD and TDD capabilities, so in some case you may not expect the price
difference.
As the distance from the base station increases. The size of the guard period in TDD has to increase. As point to point signal propagation
delay increase with increase in distances , a larger guard period is needed in LTE TDD to avoid collision between UL and DL. You can read
more details about why a gurad period is used in LTE TDD Downlink only over here.
FDD is the winner in this case. As there is no need for guard period and FDD has more capacity than TDD on account of no guard periods.
TDD can have different flavors and different configurations of LTE radio frame. There are seven different configurations for LTE radio frame
in case of TDD. If majority of the traffic being requested is in the downlink and very minimal UL traffic. In that case , a specific TDD radio
frame can be used.
As FDD radio frame has the same number of DL and UL subframes. It will not make a difference if majority of traffic is being requested in DL or UL.
Different UL/DL configurations in TDD give option and flexibility to focus on imbalanced services.
TDD is clearly the winner in this case.
The spectral efficiency of TDD is better than FDD. It is clear that TDD uses unpaired spectrum, i.e., it uses one frequency while FDD uses a
paired spectrum ,i.e., two frequencies. Therefore, the spectral efficiency of TDD is better than FDD.
TDD merely requires an unpaired spectrum for deployment. What this means is, you can deploy LTE on TDD using unpaired spectrum or
even small availability of spectrum. For the case of FDD, the requirement is to a paired spectrum to be available.
In this case TDD takes lead over FDD .
Primary and Secondary synchronization signals (PSS and SSS) are generated from the same sequence as for FDD. However, only the
location of P-SS and S-SS is different in case of TDD.
For FDD, PSS is mapped to the last OFDM symbol on slot 0 and slot 10. While SSS is mapped to the 2nd last OFDM symbol on slot 0 and slot 10.
For TDD , PSS is mapped to the 3rd OFDM symbol at slot 2 and slot 12, while SSS is mapped to the last symbol on slot 1 and slot 11.
PSS and SSS location in case of TDD
Due to less uplink subframe opportunities in TDD compared to FDD. 3GPP allows maximum of 6 PRACH can be allocated for TDD in one
subframe.
In FDD, maximum HARQ process number in Downlink is 8 and HARQ retransmission timer is fixed to 8 msec.
In TDD , max HARQ process number varies for each radio frame configuration i.e., for each each UL/DL Configuration, and HARQ RTT
timer varies for each DL subframe.
In FDD, maximum HARQ process number in Uplink is 8 and HARQ retransmission timer is fixed to 8 msec.
In TDD, UL HARQ process number varies per UL/DL configuration and HARQ RTT varies for each UL subframe.
For transmission of the uplink radio frame number from the UE. The UE adjusts timing adjustment for
In case of paging, the parameters remain same for both TDD and FDD. The only difference between FDD and TDD for paging takes place in
case of subframes (paging occasions)—- the subframe in which paging message arrives.
Theoretical peak throughput in case of FDD is clearly better than TDD.
TDD has less capacity to accommodate user data and traffic in radio frame compared to FDD.
The below measurements are done in lab environment with the following assumptions.
UL/DL Configuration 6, 1:9
DwPTS: GP:UpPTS, 11:2:2
Both for TDD and FDD , assume
UL/DL Configuration 0, 6:4
Both TDD and FDD assume
In case if you would like to provide the same services (bitrate and user experience) using TDD . It will require more number of sites
compared to FDD for the same services.
In conclusion we can say TDD has more flexibility and it is more suitable where we have Downlink and Uplink unbalanced services. As we
can compensate it using different radio frame configurations of TDD.
TDD has a lot in common with FDD. However, there are few impacts in TDD such as:
In order to provide same services as FDD, more sites and more bandwidths is needed.
Credits: 3GPP 36.847
Interested in getting more information like this. Give me your autograph below.