5G Spectrum :Mmwave and more

5G Spectrum

MmWave :Used to Attain Capacity

1. This band has a very short wavelength which means less coverage. Mmwave range offers around coverage of upto 100-150 meters
2. The objective with mmWave is to increase the data bandwidth available over smaller, densely populated areas.Its expected to provide extremely high data rates.
3. Reflections in Mmwave band provide alternative path when line of sight is blocked .
4. Small objects affect Mmwave propagation.
5. The maximum bandwidth is 400 Mhz.

Sub 6GHz: Used to attain Coverage and Capacity 

1. Its a mid-band group, which is everything between 1 and 6GHz
2. This spectrum is congested with the existing 2.4GHz and 5GHz Wi-Fi and  we also have the LTE network and Bluetooth frequencies in the 2-3 GHz range .
3. The new 5G frequencies will appear in the 3-4GHz range so it will be in between the WiFi and LTE networks to avoid any congestion.
4. The maximum bandwidth is 100 Mhz

Low Band:Used to attain mainly Coverage

1. Sub 6 and mm wave don't work as well over long distances ,so the low band is a very good option for wide area coverage and deep indoor coverage
2. The low band is already occupied with the existing 2G ,3G and partially also used for 4G networks  along with the TV broadcast frequencies.
3. The 700 MHz /800 MHz band if available can form the backbone of wide area coverage for  5G networks and help improve cell edge connectivity along with indoor coverage.

MIMO FOR BETTER AND FASTER MOBILE DATA

Why MIMO?

More antennas sending same data will mean better signal quality ,and more reliability

More antennas sending different data will mean better throughput or you can say improved data speed.

One of the biggest factors limiting data speeds is bandwidth.


More than one antenna can be used to send the signal on the same frequency thus solving the bandwidth limitation.

Speed/Throughput is also limited by signal-to-noise ratio (SNR), to which we can increase the power (or loudness) of the transmission so the phone tower can 'hear' us better.

But this could result in interference. Once we've squeezed all the performance we can from antenna-to-antenna transmission, we have to approach the problem differently.

This is where MIMO comes in to play - if we're unable to improve air transmission, why not increase the number of antennas?

What is MIMO?

A device with multiple antenna's would be called a  MIMO device because it has more than one antenna and can support more than one data stream at once.

                                   Figure above shows 2X2 MIMO

A 2×2 MIMO device has two antennas for two simultaneous data streams and a 4×4 MIMO device has four antennas for four data streams.Where the numbers 2x2,4x2,4x4  indicates the number of Tx antenna and the number of Rx antenna.

The figure below shows the multiple antenna strips on your phone (4X4 MIMO)  

The network provider is also using 4 transmit antennas thus helping your mobile to take advantage of the MIMO feature available on your individual devices

                                                 

The number of antennas helps to increase capacity. 

Theoretically the more antennas, the faster you can transmit/receive or the more users you can serve.The more antennas your device has, the more data it can transfer at once and that means faster wireless download and upload speeds.

                       
                       The figure above shows 2 X 2 MIMO  on a handset
                                    2 antenna strips done

MIMO Technologies Introduced

• Spatial Multiplexing
• Transmit Diversity/Receiver Diversity
• Beamforming

What is Spatial Multiplexing

Spatial multiplexing allows to transmit different streams of data simultaneously on the same downlink resource block(s).

These data streams can belong to one single user (single user MIMO / SU-MIMO) or to different users (multi user MIMO / MU-MIMO). 

While SU-MIMO increases the data rate of one user, MUMIMO allows to increase the overall capacity

Figure below shows an example of a 2X2 MIMO in SU MIMO mode and MU-MIMO mode

Theoretically for 2x2 MIMO, for example, two independent streams can be generated , while for 4x4 (four transmit antennas and four receive antennas), it can generate four streams, potentially quadrupling the throughput.

However practically the actual throughput will depend on Signal to noise Ratio that can be achieved in each of the streams.

MIMO connections use advanced signal processing algorithms.

To take advantage of the additional throughput offered, MIMO wireless systems utilize a matrix mathematical approach. 

Data streams A1, A2 A3 can be transmitted from antennas 1, 2 and 3.
Where B1 = signal received at antenna 1, B2 is the signal received at antenna 2 and so forth.Then there are a variety of paths that can be used with each path having different channel properties from each of these antennas.These can be represented by the properties h12 , travelling from transmit antenna 1 to receive antenna 2 and so forth. 

In this way for a three transmit, three receive antenna system a matrix can be set up:

B1 = h11 A1 + h21 A2 + h31 A3

B2 = h12 A1 + h22 A2 + h32 A3

B3 = h13 A1 + h23 A2 + h33 A3

In matrix format this can be represented as:

                          

To recover the transmitted data-stream at the receiver it is necessary to perform a considerable amount of signal processing. 

First the MIMO system decoder must estimate the individual channel transfer characteristic hij to determine the channel transfer matrix. 

Once all of this has been estimated, then the matrix [H] has been produced and the transmitted data streams can be reconstructed by multiplying the received vector with the inverse of the transfer matrix.

C1,C2 and C3 is the final reconstructed signal at the mobile receiver end.

In any case for MIMO spatial multiplexing the number of receive antennas must be equal to or greater than the number of transmit antennas.

What is Receive/Transmit Diversity

Transmit diversity is the default MIMO mode.
When multiple copies of the same data are transmitted by the Enode B the handset receives those multiple copies. 

The receiver can select the best one or combine them all together in such a way to improve data quality.

So we get less errors while receiving data and helps improving reliability.

This is called Receive /transmit diversity.

Transmit diversity scheme is applicable to all the physical channels such as PDSCH ,Physical Broadcast channel (PBCH),Physical control Format Indicator channel(PCFICH),Physical down link control channel (PDCCH)and Physical Hybrid ARQ indicator channel (PHICH).

The other MIMO schemes are applicable to PDSCH (Physical downlink shared channel)


Note:No transmit diversity scheme is applied to the primary and secondary synchronization signals 

UE can recognize the number of transmit antennas at Enode B among {1,2,4}by decoding PBCH 

Once the number of transmit antennas at eNodeB is detected by the UE it
gives feedback to the EnodeB  using the Rank Indicator parameter
RI value is very closely related to the number of Antenna.It determines the number of data streams detected .For example in 2x 2 MIMO

When the value is 2  it means that there is no interference between the antennas

If the value is 1, it implies that the signal from the two Transmission antenna is perceived by UE (Mobile) to be like single signal from single Antenna.

If transmit diversity scheme will be  applicable to the other physical downlink channels will thus be determined using the Rank Indicator.

The eNodeB may decide the transmission mode , taking into account the RI reported by the UE.

Transmission mode is a combination of the MIMO schemes used by the eNode B depending on the channel state condition experience by the user equipment's.

Some telecom operators use only 2 modes in case of release 8 one is TM1(Transmission Mode 1) that uses transmit diversity in cases where interference is more 

TM4 is used when the signal conditions are good (uses spatial multiplexing Technnology).

Appendix



How is the MIMO Transmission mode selected for a particular user equipment?


Even though the system is configured in transmission mode 4  for a particular UE and if the same UE reports the Rank Indication value 1 to eNB, eNB will start sending the data in Transmit diversity mode to UE . If UE report Rank Indication 2 , eNB will start sending the downlink data in spatial diversity MIMO mode .


How does Category of Handset  determine the throughput and MIMO functionalities?

A UE signals its category (or categories) within UE Capability Information message. So depending upon the handsets category eNode B will support only those features that can be handled by the UE.User Equipment/Handset Category determines if MIMO mode can be used 

What is the use of CQI[Channel Quality Indicator]

The CQI( Channel Quality Indicator) feedback indicates a combination of modulation scheme (ex.QPSK,QAM)and channel coding rate (the redundancy bits/duplication of data/same copies of the data to be sent)that the eNodeB should use to ensure that the block error probability experienced at the UE will not exceed 10%.

LTE DATA TRANSMISSION SCHEME -OFDMA/SCFDMA

How data is transmitted ?

In LTE data is transmitted using OFDMA in the downlink and SCFDMA in the uplink .

High power is consumed in an OFDM signal generation which is feasible for the eNode B but communication from the user equipment has power constraints so SCFDMA is used for signal generation from UE to eNodeB.

OFDMA

Its a special case of FDMA where users are provided a set of sub carriers overlapping in frequency domain.These subcarriers are designed to be orthogonal to each other which allows them to occupy the same bandwidth without any interference ,this inturn negates the use of guard bands as a result the subcarriers can be closely packed to improve channel efficiency

How it works?

In OFDM high-speed data streams of large bandwidth are split into parallel sub streams of lower bandwidth called sub carriers these sub carriers are centered around frequencies in multiples of 15 kHz on both sides of DC as the lowest subcarrier is of 15 kHz

TS =1/ 15 kHz or 66.7 microseconds

for example LTE bandwidth of 20 megahertz which has 1200 sub carriers 
Subcarrier bandwidth is  of 18 MHz because approximately 2 MHz is used as guard band here we have 600 sub carriers on both side of the DC frequency all these carrier frequencies are harmonics of 15kilo Hertz varying from minus 9 megaHertz to 9 mega Hertz in time domain these sub carriers will be represented as everlasting sinusoids at these carrier frequencies as shown below

In order to transmit data over these subcarriers they are loaded with modulation symbols that represent the constellation points of digital modulation schemes like QPSK, 16 QAM ,64 QAM  .

The symbol duration  for each of these subcarriers is always equal to 66.7 microseconds which means that all these sub carriers have a whole number of cycles in one symbol duration 


                              Figure below shows the subcarriers near the DC subcarrier

                                                                                 

 

Figure above shows all these sub carriers have a whole number of cycles in one symbol duration 

As we know that a rectangular function can be represented in frequency domain as a sinc function which is centered around DC when multiplying a signal to a carrier frequency and time domain signal will be shifted in frequency domain by the same amount of carrier frequency thus we can represent these modulated sub carriers in frequency domain as a series of sync waves centered around the carrier frequencies.

Summarizing it basically we are having 1200 such sync waves .The sub carriers are overlapping in frequency domain as we can see the subcarriers are placed in a manner that all the other sub carriers have a zero component at the peak of one sub carrier such sub carriers are called orthogonal as a result a mobile can sample the frequency and phase without any interference from neighboring subcarriers orthogonality is achieved by ensuring that all the sub carriers have same symbol duration TS and the subcarrier spacing is maintained at Delta F equal to 1/TS 

   

Data Transmission in the DL

Data is transmitted in OFDMA over parallel subcarriers of 15kHz.The subcarriers are further divided on the time axis as blocks of 1 symbol duration or 66.7 microseconds.This basic unit is known as resource element.Each resource element carries one symbol.Now in order to transmit data over these resource elements first data is modulated with specific modulation scheme this scheme depends on the physical channels mapped on the resource grid.

Suppose we have 8 bits of data to be transmitted using QPSK modulation the data will be divided into four parallel streams of 2 bits.Phase and amplitude assignment is done according to the QPSK constellation

                               

Finally data is placed over the resource elements by adjusting the phase and amplitude of subcarrier to those derived for the data stream.Mathematically it means multiplying the complex modulation symbol to the corresponding subcarrier frequency 

Thus in a 20 megahertz channel all the 1200 resource elements carrying 1200 symbols over 1200 sub carriers are modulated with appropriate modulation scheme.Now since the data is modulated over 1200 sub carriers a transmitter will require 1200 oscillators for its generation and another 1200 will be required by the receiver for proper demodulation The hardware complexity and sheer amount of power consumption would  have left OFDM to theoretical idea and far from implementation 

                                   

OFDM was made reality with the advent of digital signal processing techniques.So instead of using 1200 oscillators IFFT solves this problem by converting the parallel frequency domain signals into samples of a composite time domain signal which are much easier to generate at the  transmitter side.

                                     


All we need to do is to send these time domain samples at radio frequencies.As the data is being sampled by IFFT samples,it must be taken above the Nyquist rate for faithful reproduction at the receiver .For an LTE spectrum of 20 megahertz the highest frequency component is 9 megahertz which means the sampling rate should be greater than 18mega samples per second.

In other words 1200 samples per OFDM symbol of 66.7microseconds. Since UMTS had a sampling rate of 3.84 mega samples per second.To achieve backwards compatibility sampling rate for LTE is taken as multiples of 3.84 mega samples per second. Thus for 20 megahertz spectrum which has a sampling rate of 30.72 mega samples per second there are 2048 samples per symbol in OFDM 

The FFT size of the IFFT processor thus depends on the LTE bandwidth as shown above.


Guard Period

Abrupt transition in time domain generate high frequency components and disturb the orthogonality in frequency domain which introduces inter carrier interference

                               
                                    Figure above shows a Time Domain Representation 

                                    Figure above shows a Frequency Domain Representation 

 So in OFDMA we employ a more complex kind of guard period called cyclic prefix in which 
end part of a symbol is transmitted in the preceding guard period. It also ensures orthogonality between the subcarriers by keeping the OFDM symbol periodic over the extended symbol duration and therefore avoiding intercarrier and inter channel interference simultaneously.Since OFDMA uses composite IFFT samples,cyclic prefix is added by taking some samples from the end of a symbol period and placing them at the beginning.

Now this time sampled signal is converted into analog wave by a digital to analog converter further composite waveform is modulated at the desired radio frequency for transmission

for an example an operator having license for 2320-2340MHz band the 18 MHz band will be mapped from 2321-2339 MHz using analog modulation and finally the EnodeB transmits the signal over air .Enode B informs the user about the allocated subcarriers and the corresponding modulation scheme so when the RF signal reaches at the user's terminal all these processes are reversed and finally user is able to receive the data intended for it

                                       

Data Transmission in the UL

SCFDMA

Although there are many positives in OFDMA but IFFT summation of multiple
parallel self carrier results in high peak to average power ratio(PAPR).High PAPR  results in high power consumption for signal generation but handheld devices have limited power capacity.This makes OFDMA unfavorable for uplink transmission.In order to overcome this LTE uses SCFDMA in uplink direction.In OFDMA we have one-to-one mapping between symbol and subcarrier but as SCFDMA allows a symbol to be transmitted in parts over multiple sub carriers.For example in OFDMA one symbol occupies one sub carrier of 15 kilo Hertz but in FDMA same symbol is distributed among multiple sub carriers of 15 kilo Hertz

In short SCFDMA behaves like a single carrier system with short symbol duration compared to OFDMA.To achieve this SCFDMA introduces an N point FFT block right after the serial to parallel converter in the OFDMA structure the FFT block converts parallel sequence of symbols in time domain to different frequency points 

Now peak to average power ratio is proportional to the square of number of carriers involved SCFDMA reduces PAPR by reducing the number of carriers .The side effects are reduced by ensuring that the total bandwidth over which a symbol is transmitted is still not too high apart from this remaining block in the uplink direction is same as used in downlink

    



                                                                             

RACH Configurations in LTE

OVERVIEW 

Why RACH Process?

In communication the most important step is to time synchronize between the receiver and the transmitter. 

1>In order to achieve the uplink synch between the UE and the eNodeB
    a RACH process is initiated.
2>Along with the  uplink synchronization user equipment gets the required            radio resources in the uplink direction to send L3/L2 message.

UE gets all the  random access  related parameter of the cell  by reading SIB2 .

When is the RACH initiated?

1. Transition from RRC Idle to RRC Connected 
2. Handover( for UL synch )
3. DL data arrival when in non synch RRC mode
4. UL Data arrival when in non synch RRC mode
5. Re-establishing an RRC connection 
6. For positioning  purpose when Timing advance is needed for UE's positioning.

The Random Access procedure has two types:

Contention based

1. UE selects a random access resource .

2. Used for initial access, the arrival of uplink data and re-establishment 

 Non-contention based:

1. eNB allocates a random access resource to UE 

2. Used for the connection to the target cell during UE's handover.
3. Used when RACH procedure is expected to be faster because there are               timing restrictions

RACH Procedure in Brief

UE triggers RACH procedure by sending RACH preamble to the eNB on the PRACH channel(Physical Random access Channel).

About the preambles-There are 64 preambles available in each cell.10 root sequences are used to generate the 64 preambles.Each root sequence can generate 7 preambles .By reading SIB2 parameter user comes to know about the preamble format and timing.User equipment randomly selects one of the preambles .

Incase of failure by the ENB in reading the Random Access Request the RACH preamble is transmitted again.The number of times user equipment should transmit the RACH preamble is determined by a parameter called as PreambletransMax which is a SIB 2 parameter otherwise UE will loose all its  battery if it continuously transmits the RACH preamble

During retransmission of RACH preamble user equipment will have to increase its power because ENB was not able to detect the RACH request previously.
When ENB is not able to detect the RACH request then SIB2 parameter power ramping step factor is used during retransmission of the RACH preamble if failed to detect at eNodeB.


eNodeB calculates  UE identifier called the RA RNTI(Random Access-Radio Network Temporary Identifier) by the timing of preamble transmission. ENB after receiving RACH Preamble derives RA RNTI from the time slot number in which preamble is received.  Calculates TC RNTI  for this UE .TC RNTI is the temporary cell radio network temporary identifier .It is used for further communication between the UE and ENB

Calculates the timing advance which is transmitted to the UE as part of response message .Resource block to be used for uplink transmission,MCS(Modulation and Coding Scheme)hopping flag,CSI(Channel State Information)field,power to be used by the UE for PUSCH, UL delay .

ENB includes all this information in the random access response and sends this to the UE.If RACH preamble is sent at time x then UE should expect RA response to be received  within the time gap where value of y lies between x+3<=y<=x+"random access response window" where RA response window is determined by RA response window size found in SIB2

After getting the RA response UE  saves the temporary C-RNTI from RA response applies  the received timing correction from the timing advance information received in the randoma access response , uses uplink resource information present in the RAR and  transmits RRC Connection request  to the ENB

UE is synched in the UL direction.UE does  not have an identity it picks a random number as an identity and sends it in the RRC connection request 
After sending RRC connection request UE starts the RRC Connection timer T300 which is broadcasted in SIB2  and waits for RRC connection setup message


The ENB accepts the transmission from the UE and sends RRC Connection setup in the downlink.This message contains the random number and is addressed by TC RNTI.

CONTENTION RESOLUTION

Since the UE randomly selects the preamble there are chances that two user equipment's may pick the same RACH Preamble in the UL at the same time 

a>There is collision and ENB Is not able to decode preamble sent by any UE .

ENB is not able to decode both the user equipments and  will run a back off timer with some random value and initiate the random procedure  again

B>There is collision and ENB is able to decode preamble only for UE A

UE A and B sends the same RACH preamble at the same time and ENB is able to detect preamble  from UE A only.If 2 different UE transmits preamble at the same time,ENB gives same RA-RNTI number for both UE in the Random Access Response(RAR)

RA RNTI will be same for both UE .Random access response(RAR) is intended to UE A.Both UE will decode RAR ,both UE will acquire the same TC-RNTI present in RAR UE B still does not know that eNodeB was not able to decode its preamble.

Both UE will choose some random number as initial identity and send RRC Connection request to ENB and start the timer T300.But ENB will not be able to detect message from UE B as UE B is using the timing advance value intended for UE A

Now ENB will send RRC Connection setup in the  downlink,both UE's will decode this message as its addressed by TC RNTI .eNodeB will include the random number that was sent by UE A in this message.Both UE will decode this message but random number sent and received by UE B will mismatch
At this stage it will understand it has lost out to some other UE in contention resolution.

 TYPICAL RACH PARAMETER CONFIGURATIONS IN AN LTE NETWORK

Preamble Sequence Group 

The set of 64 preamble sequences are divided into two groups: 

Contention based random access 
Group A: when UE has a relatively small quantity of uplink data, or is in poor coverage
Group B: when UE has a relatively large quantity of uplink data, or in good coverage

Non-contention based random access 

The UE selects a group B sequence if both of the following conditions are met: 
 Message Size > messageSizeGroupA  (56bits) AND 
Path Loss < Pmax (23) - preambleInitialReceivedTargetPower  ( - 108dB) - deltaPreambleMsg3  (4dB ) -messagePowerOffsetGroupB (0dB)  i.e measured RSRP>-112

Preamble Transmit Power 

PRACH Preamble Transmit Power=min{Pmax,PL+PreambleRXTargetPower}

Pmax :is the UE maximum transmit power according to the UE Power class ,eg 23 dBm for power class 3
PL(path Loss):Reference signal Transmit Power (15 dBm or 21 dBm )-measured RSRP
Preamble RX Target Power:    PreambleIntialReceivedTargetPower(-108)+DELTA PREAMBLE+(PREAMBLE TRANSMISSION COUNTER-1)*powerRampingstep(2dB)
DELTA PREAMBLE :Defines a power offset dependent upon the random access preamble format:0 dB for format 0-1 and -3 dB for format 2-3
PREAMBLE TRANSMISSION COUNTER: Counter by the UE .Its value starts from 1 ,incremented by 1 if no response is received.

                   Power Ramping in Random Access Procedure

    

Random Access Response Window

After transmitting the PRACH pramble ,the UE searches for a response during the time domain window defined by the Random Access Reponse Window(sf10)

The RA response window starts during the third subframe after the preamble

If the UE does not receive its random access response within the window

It increment the counter of the PREAMBLE TRANSMISSION COUNTER by 1

The UE exits the random access procedure if the maximum number of transmissions has been reached

mac-Contention Resolution Timer

The UE starts a contention resolution timer after transmitting the initial layer 3 message .If the UE does not receive a response until the maccontentionResolutionTimer(sf64) expires then the UE returns to transmitting  PRACH Preambles 

Appendix


UE MAC RACH TRIGGER MESSAGE

UE RANDOM ACCESS MESSAGE PROCEDURE MESSAGE

ROOT_SEQUENCE_INDEX 

The first logical root sequence index used to create a random preamble. Different values should be assigned to neighboring cells. Be careful to change because it can cause inter-cell interference among the cells with same physical root sequence.

Typically  LTE RACH optimization operates in eNB's SON agent and EMS's SON manager.The enB SON Agent has RSI Collision Detection Function and the EMS SON Manager performs the RSI reallocation upon receiving the RSI collision /confusion notify information message