16.546 Computer Telecommunications:
Modulation and Data Encoding
Professor Jay Weitzen
Electrical & Computer Engineering Department
The University of Massachusetts Lowell
1
Data Encoding at the PL
Source node
Destination node
Application
Application
Presentation
Presentation
Session
Session
Intermediate node
transport
Network
Packets
transport
Network
Network
Data link
Data link
Physical
Physical
Frames
Data link
Physical
Bits
Signals
2
Network A Node
7
6
5
We Need to Encode PL Frame
Application
AL-Hdr
Presentation
PL-Hdr
Session
4
Transport
3
Network
2
Data Link
1
Physical
SL-Hdr
DLL-Hdr
PL-Hdr
Presentation Layer Msg
Session Layer Msg
TL-Hdr
NL-Hdr
Application Layer Msg
Transport Layer Msg
Network Layer Msg
Data Link Layer Msg
Physical Layer Msg
3
Encoding Techniques
 Digital data, digital signal
 Analog data, digital signal
 Digital data, analog signal
 Analog data, analog signal
4
Digital Data, Digital Signal
 Digital signal
– Discrete, discontinuous voltage pulses
– Each pulse is a signal element
– Binary data encoded into signal elements
5
Terminology
 Unipolar
– All signal elements have same sign
 Polar
– One logic state represented by positive voltage the other by
negative voltage
 Data rate
– Rate of data transmission in bits per second
 Duration or length of a bit
– Time taken for transmitter to emit the bit
 Modulation rate
– Rate at which the signal level changes
– Measured in baud = signal elements per second
 Mark and Space
– Binary 1 and Binary 0 respectively
6
Interpreting Signals
 Need to know
– Timing of bits - when they start and end
– Signal levels
 Factors affecting successful interpreting of signals
– Signal to noise ratio
– Data rate
– Bandwidth
7
Comparison of Encoding Schemes (1)
 Signal Spectrum
– Lack of high frequencies reduces required bandwidth
– Lack of dc component allows ac coupling via
transformer, providing isolation
– Concentrate power in the middle of the bandwidth
 Clocking
– Synchronizing transmitter and receiver
– External clock
– Sync mechanism based on signal
8
Comparison of Encoding Schemes (2)
 Error detection
– Can be built in to signal encoding
 Signal interference and noise immunity
– Some codes are better than others
 Cost and complexity
– Higher signal rate (& thus data rate) lead to higher
costs
– Some codes require signal rate greater than data rate
9
Encoding Schemes









Nonreturn to Zero-Level (NRZ-L)
Nonreturn to Zero Inverted (NRZI)
Bipolar -AMI
Pseudoternary
Manchester
Differential Manchester
B8ZS
HDB3
4B/5B, MLT-3, 8B/10 Schemes
10
Nonreturn to Zero-Level (NRZ-L)
 Two different voltages for 0 and 1 bits
 Voltage constant during bit interval
– no transition, i.e., no return to zero voltage
 Absence of voltage for zero, constant positive
voltage for one
 More often, negative voltage for one value and
positive for the other
 This is NRZ-L
11
Nonreturn to Zero Inverted
 Nonreturn to zero inverted on ones
 Constant voltage pulse for duration of bit
 Data encoded as presence or absence of signal
transition at beginning of bit time
 Transition (low to high or high to low) denotes a
binary 1
 No transition denotes binary 0
 An example of differential encoding
12
NRZ
13
Differential Encoding
 Data represented by changes rather than levels
 More reliable detection of transition rather than
level
 In complex transmission layouts it is easy to lose
sense of polarity
14
NRZ pros and cons
 Pros
– Easy to engineer
– Make good use of bandwidth
 Cons
– dc component
– Lack of synchronization capability
 Used for magnetic recording
 Not often used for signal transmission
15
Multilevel Binary
 Use more than two levels
 Bipolar-AMI (Alternate Mark Inversion)
–
–
–
–
zero represented by no line signal
one represented by positive or negative pulse
one pulses alternate in polarity
No loss of sync if a long string of ones (zeros still a
problem)
– No net dc component
– Lower bandwidth
– Easy error detection
16
Pseudoternary
 One represented by absence of line signal
 Zero represented by alternating positive and
negative
 No advantage or disadvantage over bipolar-AMI
17
Bipolar-AMI and Pseudoternary
18
Trade Off for Multilevel Binary
 Not as efficient as NRZ
– Each signal element only represents one bit
– In a 3 level system could represent log23 = 1.58 bits
– Receiver must distinguish between three levels
(+A, -A, 0)
– Requires approx. 3dB more signal power for same
probability of bit error
19
Biphase
 Manchester
–
–
–
–
–
Transition in middle of each bit period
Transition serves as clock and data
Low to high represents one
High to low represents zero
Used by IEEE 802.3
 Differential Manchester
–
–
–
–
–
Midbit transition is clocking only
Transition at start of a bit period represents zero
No transition at start of a bit period represents one
Note: this is a differential encoding scheme
Used by IEEE 802.5
20
Biphase Pros and Cons
 Con
– At least one transition per bit time and possibly two
– Maximum modulation rate is twice NRZ
– Requires more bandwidth
 Pros
– Synchronization on mid bit transition (self clocking)
– No dc component
– Error detection
• Absence of expected transition
21
Modulation Rate
22
Scrambling
 Use scrambling to replace sequences that would produce
constant voltage
 Filling sequence
– Must produce enough transitions to sync
– Must be recognized by receiver and replace with original
– Same length as original




No dc component
No long sequences of zero level line signal
No reduction in data rate
Error detection capability
23
B8ZS
 Bipolar With 8 Zeros Substitution
 Based on bipolar-AMI
 If octet of all zeros and last voltage pulse preceding was
positive encode as 000+-0-+
 If octet of all zeros and last voltage pulse preceding was
negative encode as 000-+0+ Causes two violations of AMI code
 Unlikely to occur as a result of noise
 Receiver detects and interprets as octet of all zeros
24
HDB3
 High Density Bipolar 3 Zeros
 Based on bipolar-AMI
 String of four zeros replaced with one or two
pulses
25
B8ZS and HDB3
26
Digital Signal Encoding For LANs
 4B/5B-NRZI
– Used for 100BASE-X and FDDI LANs
– Four Data Bits Encoded into Five Code Bits, 80%
 MLT-3
– 100BASE-TX & FDDI Over Twisted Pair
 8B/6T
– Uses Ternary Signaling (Pos, Neg, Zero Voltages)
– Eight Data Bits Encoded into 6 Ternary Symbols
 8B/10B
– Used for Fibre Channel & Gigabit Ethernet
27
10 Gigabit Ethernet (1 of 2)
• IEEE 802.3ae
• MAC: it’s just Ethernet
– Maintains 802.3 frame format and size
– Full duplex operation only
– Throttled to 10.0 for LAN PHY or 9.58464 Gb/s for WAN PHY
• PHY: LAN and WAN phys
– LAN PHY uses simple encoding mechanisms to transmit data on dark fiber
and dark wavelengths
– WAN PHY adds a SONET framing sublayer to utilize SONET/SDH as layer 1
transport
• PMD: optical media only
–
–
–
–
850 nm on MMF to 65m
1310 nm, 4 lambda, WDM to 300 m on MMF; 10 km on SMF
1310 nm on SMF to 10 km
1550 nm on SMF to 40 km
28
10 Gigabit Ethernet (2 of 2)
• Supports dark wavelength and SONET/TDM with
unlimited reach
• Several Coding Schemes (64b/66b; 8B/10B;
Scramblers)
• Three optional interfaces: XGMII; XAUI; XSBI
• Extension of MDIO interface
• Continues Ethernet’s reputation for cost effectiveness
and simplicity (goal 10X performance for 3X cost)
• Expected target for ratification in Spring 2002
29
802.3ae to 802.3z Comparison
10 Gigabit Ethernet
1 Gigabit Ethernet
• CSMA/CD + Full
Duplex
• Carrier Extension
• Optical/Copper Media
• Leverage Fibre Channel
PMD’s
• Reuse 8B/10B Coding
• Support LAN to 5 km
•
•
•
•
Full Duplex Only
Throttle MAC Speed
Optical Media Only
Create New Optical
PMD’s From Scratch
• New Coding Schemes
• Support LAN to 40
km; Use SONET/SDH
as Layer 1 Transport
30
Converting From Analog To Digital
31
Pulse Code Modulation: a digital
encoding scheme used in TDM
 In this modulation technique, an analog signal is
digitized, and interleaved with other digitized
voice signal to create a single bit stream
 At the receiving end, the bit stream is decomposed
into separate digital streams of lower frequencies,
each stream is then converted back into what
resembles the original voice signal.
32
Steps Required to Generate PCM
Streams
 Sampling: periodic measurement of the analog
signals at regular intervals
 Quantizing: assigning discrete values to samples
 Coding: assigned binary codes to samples using
what is known as the PCM code word
33
Sampling
(a)
(b)
(c)
Figure 2.2 : creating a PAM wave for a single sinusoid.
(a) is a sinusoid signal, (b) a pulse train, (c) the result of
passing (a) and (b) through a point by point multiplier.
34
Sampling
 Sampling rate: how often should we take
measurements of the analog signal
 at least at twice the rate of its highest frequency
component
 For a voice channel with a frequency range
between 300 Hz and 3400 Hz (bandwidth of 3100
Hz) we need to take a sample at least at a rate of 2
X 3100 = 6200 Hz or every 1/6200 second
35
Sampling
 In practical system, we sample multiple channel,
we combine the samples of all channels into a
single signal called the PAM signal (Pulse
Amplitude Modulation signal)
 In American systems we sample 24 channels
 In the European systems 30 channels are sampled
36
Quantization
 To represent samples by a fixed number of bits
 For example if the amplitude of the PAM signal
range between -1 and +1 there can be infinite
number of values. For instance one value can be 0.2768987653598364834634
 For practicality, we may use 20 different discrete
values between -1 and +1 volts
 Each value at a 0.1 increment
37
Quantization: the binary world
 Because we live in a binary world, we select the
total number of discrete values to be binary
number multiple (i.e., 2, 4, 8, 16, 32, 64, 128, 256,
and so on)
 This facilitate binary coding
 For instance, if there were 4 values they would be
as follows: 00, 01, 10, 11
 This is a 2-bit code
38
Quantization:
16 coded quantum steps
 Between -1 and + 1 volts signal
 16 discrete steps
 each step at 0.125 volts increment or decrement
from the adjacent step
 0 0000
0v 3
0011 0.375v
 1 0001 0.125v
4
0100 0.500v
 2 0010 0.25v
5
0101 0.625v
39
Quantization: 16 quantum steps
(-1 to + 1 volts)
+1
Range of standard
values (V)
0
-1
15 : 1111
14: 1110
13: 1101
12: 1100
11: 1011
10: 1010
9: 1001
8: 1000
7: 0111
6: 0110
5: 0101
4: 0100
3: 0011
2: 0010
1: 0001
0: 0000
8 9 10 1112 13 12 11 10.. 6 .........
Coded values
40
Quantization Distortion
 Quantization error is the different between the quantum
value and the true value
 More steps reduce quantizing distortion in linear
quantization
 This will require higher bandwidth, since we need more
bits for each code word
 Voice represent a problem because of the wide dynamic
range, the level from the loudest syllable of the loudest
talker to the lowest syllable of the quietest talker
 S/D = 6n + 1.8 dB EX: 7 bit PCM cod 6.7 + 1.8 = 43.8
 practical system S/D = 30 - 33 dB
41
Companding
 Compression/Expanding
 Non-linear
 The voltage level between the loudest and the
lowest is segmented in non-linear manor
 The voltage range of each segment varies
according to the level of the voltage
42
Non-linear Quantization
Segment #
Voltage levels
5.0
1
3.0
2
1.5
3
0.5
4
5
0
6
7
8
43
Non-linear Quantization
Compressed Output
Voltage
Segment 2 has 3 steps like all
of the other segments
Segment 4
Segment 3
Segment 2
Segment 1
-5.0
0.5
1.0 1.5 2.0 2.5 3.0 3.5 4.0
Input Voltage
4.5 5.0
44
Coding for Modern PCM systems
 Non-linear
 Logarithmic
 A-Law
 u-Law
45
A-Law
AX
Y
1  log A
Y
1 _ log( AX
1 _ log A
for
for
V
0v
A
V
 v V
A
46
U-Law
log( 1  u | X |)
| Y |
log( 1  u)
47
Coding for Modern PCM systems


i
v
Y
X
Where  = instantaneous input voltage
V
B
V = maximum input voltage for which peak
limitation is
absent

i = number of quantization steps starting from the center
of the range

B = number of quantization steps on each side of the
center of the range.
48
13-segment A-Law Curve
Segment
(Chord)
Code
6
112
5
96
4
80
1101XXX
3
64
1100XXX
2
48
1011XXX
1111XXX
1/2
1110XXX
1/4
1/8
POSITIVE
1/16
1/32
1
32
1010XXX
1/64
1001XXX
0
1000XXX
0
1/4
2/4
3/4
1
(V)
0000XXX
0001XXX
0010XXX
0011XXX
NEGATIVE
0100XXX
0101XXX
0110XXX
0111XXX
Figure 2.7: 13-segment approximation of the A-law
curve used with E1 PCM equipment
49
PCM Code Word
Sign
S
Segment
Number
Level
Value
A B C
D
Figure 2.8: PCM Code Example
50
S/D for A-law & u-Law
 For A = 87.6: S/D = 37.5 dB
 u = 255: S/D = 37
51
Modems: Modulator/Demodulator
 Used to Package bits for transport over
broadband media
– 3 ways to encode information on a carrier
 - Phase
 - Frequency
 - Amplitude
52
Definition of Modulation
 Let m(t) be an arbitrary modulating (information)
waveform. (could be either analog or digital)
 Let c(t)=cos(wct +f(t)) be the carrier
 The argument of the sinusoid is the instantaneous phase
(wct +f( t ))
 The instantaneous frequency (2pfi)is given by d/dt (wct
+f( t )) = wc +d/dt(f(t))2pfi
53
Types of Modulation
 If c(t)=m(t) cos(wct +f), the information is
transported in the amplitude of the carrier. We
call this Amplitude Modulation (AM)
 If fi(t)=km(t), the information is transported in
the instantaneous frequency. We call this
frequency modulation (FM).
 If f( t )=km(t) the information is carried in the
instantaneous phase, and we call this phase
modulation (PM).
54
Modulation Techniques
55
Amplitude Shift Keying
 Values represented by different amplitudes of
carrier
 Usually, one amplitude is zero
– i.e. presence and absence of carrier is used
 Susceptible to sudden gain changes
 Inefficient
 Up to 1200bps on voice grade lines
 Used over optical fiber
56
Frequency Shift Keying
 Values represented by different frequencies (near
carrier)
 Less susceptible to error than ASK
 Up to 1200bps on voice grade lines
 High frequency radio
 Even higher frequency on LANs using co-ax
57
Frequency Modulation
FM Used for high fidelity audio broadcast and
digital transmission.
Uses Shannon concept of bandwidth expansion.
58
FSK on Voice Grade Line
59
Phase Shift Keying
 Phase of carrier signal is shifted to represent data
 Differential PSK
– Phase shifted relative to previous transmission rather
than some reference signal
60
Phase Modulation
Generally used for digital modulation
61
Quadrature PSK
 More efficient use by each signal element
representing more than one bit
– e.g. shifts of p/2 (90o)
– Each element represents two bits
– Can use 8 phase angles and have more than one
amplitude
– 9600bps modem use 12 angles , four of which have two
amplitudes
62
Constellation Space
 Create 2-axis (e.g. sine and cosine) actually it could be a ndimensional hyper-plane
 Express digital modulation alphabet as points in the
hyper-plane. The farther apart the points are in the space,
the more immunity there is against noise and interference.
 More distance, better error performance. Keep this in
mind.
 The maximum power is the length of the longest vector.
The average transmitter power is the average distance
squared of all the points.
63
Case Study 1: ASK
• If m(t) = {0,1} and we amplitude modulate a carrier with m(t)
then the modulation is called on/off keying (OOK) or 2-amplitude
shift keying (2-ASK)
• 2-ASK, (points are at (0,0), and (0,1), in the 2 dimensional (sine,
cosine plane). Minimum distance between points is 1 for 1 unit of
power, and 1 bit per symbol.
• Distance between points corresponds to error performance
64
Case Study 2: Multi-Level ASK
•If maximum power is normalize to 1 then points are at
(0,0), (0,1/3), (0,2/3), (0,1). Distance is reduced from
2-ASK and performance is worse. Requires 3x or 9x
power to maintain 1 unit of distance.
• From Shannon, as we add more information in a fixed
bandwidth, it becomes increasingly expensive in terms
of SNR to add more data.
65
Case 3: Orthogonal FSK
•Frequencies are chosen so that the waveforms are orthogonal
over the period of the bit T.
• Points are at (0,1) and (1,0) for 2-FSK. Distance is sqrt(2).
Error performance better than 2-ASK but not as good as others.
66
Case 4: QPSK and PSK
y(t)
y(t)
A
-A
-A
A
A
x(t)
x(t)
-A
Example signal constellation
diagram for BPSK signal.
67
Higher Order Modulations Very
Inefficient in terms of Power
68
Case 6: QAM
Beyond 3 bits/symbol, PSK too power inefficient. Must use
hybrid amplitude and phase modulation called QAM
69
Example V.32 Constellation
70
Performance of Digital to Analog
Modulation Schemes
 Bandwidth
– ASK and PSK bandwidth directly related to bit rate
– FSK bandwidth related to data rate for lower
frequencies, but to offset of modulated frequency from
carrier at high frequencies
– (See Stallings for math)
 In the presence of noise, bit error rate of PSK and
QPSK are about 3dB superior to ASK and FSK
71
Coherent vs. Non-Coherent
Detection
 Coherent detection requires a copy of the carrier
to be recovered from the received signal for use in
the detection process. It is more efficient because
it uses all phase information, but requires added
complexity
 Non-coherent detection using an envelope
detector is much easier to implement, but less
efficient because it uses only the envelope
information and not the phase information.
72
Digital Data, Analog Signal
 Public telephone system
– 300Hz to 3400Hz
– Use modem (modulator-demodulator)
 Amplitude shift keying (ASK)
 Frequency shift keying (FSK)
 Phase shift keying (PK)
73
Analog Data, Digital Signal
 Digitization
– Conversion of analog data into digital data
– Digital data can then be transmitted using NRZ-L
– Digital data can then be transmitted using code other
than NRZ-L
– Digital data can then be converted to analog signal
– Analog to digital conversion done using a codec
– Pulse code modulation
– Delta modulation
74
Pulse Code Modulation(PCM) (1)
 If a signal is sampled at regular intervals at a rate higher
than twice the highest signal frequency, the samples
contain all the information of the original signal
– (Proof - Stallings appendix 4A)




Voice data limited to below 4000Hz
Require 8000 sample per second
Analog samples (Pulse Amplitude Modulation, PAM)
Each sample assigned digital value
75
Pulse Code Modulation(PCM) (2)
 4 bit system gives 16 levels
 Quantized
– Quantizing error or noise
– Approximations mean it is impossible to recover
original exactly
 8 bit sample gives 256 levels
 Quality comparable with analog transmission
 8000 samples per second of 8 bits each gives
64kbps
76
Nonlinear Encoding
 Quantization levels not evenly spaced
 Reduces overall signal distortion
 Can also be done by companding
77
Delta Modulation
 Analog input is approximated by a staircase
function
 Move up or down one level () at each sample
interval
 Binary behavior
– Function moves up or down at each sample interval
78
Delta Modulation - example
79
Delta Modulation - Operation
80
Delta Modulation - Performance
 Good voice reproduction
– PCM - 128 levels (7 bit)
– Voice bandwidth 4khz
– Should be 8000 x 7 = 56kbps for PCM
 Data compression can improve on this
– e.g. Interframe coding techniques for video
81
Analog Data, Analog Signals
 Why modulate analog signals?
– Higher frequency can give more efficient transmission
– Permits frequency division multiplexing (chapter 8)
 Types of modulation
– Amplitude
– Frequency
– Phase
82
Analog
Modulation
83
Spread Spectrum





Analog or digital data
Analog signal
Spread data over wide bandwidth
Makes jamming and interception harder
Frequency hoping
– Signal broadcast over seemingly random series of frequencies
 Direct Sequence
– Each bit is represented by multiple bits in transmitted signal
– Chipping code
84
Encoding Schemes - WAN Techniques
1 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0
AMI
0 0 0 0 V B 0 V B
B8ZS
0 0 0 V B 0 0 V
B 0 0 V
HDB3
Both are well suited to characteristics of WAN channels
85
Encoding Schemes - Spectral Density
B8ZS,
HDB3
1.2
1.0
Mean
Square
Voltage .8
per Unit
Bandwidth.6
NRZ-L
NRZI
AMI, Pseudoternary
Manchester,
Diff. Manchester
.4
.2
.2
.4
.6
.8
1.0 1.2 1.4 1.6 1.8 2.0
Normalized Frequency (f/R)
86
Communications Interface
Destination WS
SourceWS
Information Exchange
•Content Material
•Acquisition
•Conversion
•Compression
•Buffering
•Media Access
•Protocol
•Segmentation
•Streaming
Transmission
Or
Network
•Packet Routing
•Node Switching
•Buffering
(Network Delay & Transmission Jitter)
•Content Material
•Acquisition
•Conversion
•Compression
•Buffering
•Media Access
•Protocol
•Reassembly
•Synchronization
87
Asynchronous and Synchronous
Transmission
 Timing problems require a mechanism to
synchronize the transmitter and receiver
 Two solutions
– Asynchronous
– Synchronous
88
Asynchronous
 Data transmitted one character at a time
– 5 to 8 bits
 Timing only needs maintaining within each
character
 Resync with each character
89
Asynchronous (diagram)
90
Asynchronous - Behavior
 In a steady stream, interval between characters is uniform
(length of stop element)
 In idle state, receiver looks for transition 1 to 0
 Then samples next seven intervals (char length)
 Then looks for next 1 to 0 for next char




Simple
Cheap
Overhead of 2 or 3 bits per char (~20%)
Good for data with large gaps (keyboard)
91
Synchronous - Bit Level
 Block of data transmitted without start or stop
bits
 Clocks must be synchronized
 Can use separate clock line
– Good over short distances
– Subject to impairments
 Embed clock signal in data
– Manchester encoding
– Carrier frequency (analog)
92
Synchronous - Block Level
 Need to indicate start and end of block
 Use preamble and postamble
– e.g. series of SYN (hex 16) characters
– e.g. block of 11111111 patterns ending in 11111110
 More efficient (lower overhead) than async
93
Synchronous (diagram)
94
Line Configuration
 Topology
– Physical arrangement of stations on medium
– Point to point
– Multi point
• Computer and terminals, local area network
 Half duplex
– Only one station may transmit at a time
– Requires one data path
 Full duplex
– Simultaneous transmission and reception between two stations
– Requires two data paths (or echo canceling)
95
Traditional Configurations
96
Interfacing
 Data processing devices (or data terminal equipment,
DTE) do not (usually) include data transmission facilities
 Need an interface called data circuit terminating
equipment (DCE)
– e.g. modem, NIC
 DCE transmits bits on medium
 DCE communicates data and control info with DTE
– Done over interchange circuits
– Clear interface standards required
97
Characteristics of Interface
 Mechanical
– Connection plugs
 Electrical
– Voltage, timing, encoding
 Functional
– Data, control, timing, grounding
 Procedural
– Sequence of events
98
V.24/EIA-232-F
 ITU-T v.24
 Only specifies functional and procedural
– References other standards for electrical and mechanical
 EIA-232-F (USA)
–
–
–
–
–
RS-232
Mechanical ISO 2110
Electrical v.28
Functional v.24
Procedural v.24
99
Mechanical Specification
100
Electrical Specification
 Digital signals
 Values interpreted as data or control, depending
on circuit
 More than -3v is binary 1, more than +3v is
binary 0 (NRZ-L)
 Signal rate < 20kbps
 Distance <15m
 For control, more than-3v is off, +3v is on
101
Local and Remote Loopback
102
Procedural Specification
 E.g. Asynchronous private line modem
 When turned on and ready, modem (DCE) asserts DCE
ready
 When DTE ready to send data, it asserts Request to Send
– Also inhibits receive mode in half duplex
 Modem responds when ready by asserting Clear to send
 DTE sends data
 When data arrives, local modem asserts Receive Line
Signal Detector and delivers data
103
Dial Up Operation (1)
104
Dial Up Operation (2)
105
Dial Up Operation (3)
106
Null Modem
107
ISDN Physical Interface Diagram
108
ISDN Physical Interface
 Connection between terminal equipment (c.f.
DTE) and network terminating equipment (c.f.
DCE)
 ISO 8877
 Cables terminate in matching connectors with 8
contacts
 Transmit/receive carry both data and control
109