Transcript 15-05-0028-02-004a-waveform-modulated-low-rate-uwb-system-proposal-15-4a-alt-phy.ppt

doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs)
Submission Title: [WAVEFORM MODULATED LOW RATE UWB SYSTEM - Proposal for 15.4a alt PHY]
Date Submitted: [Mar., 2005]
Source: [Soo-Young Chang] Company [California State University, Sacramento]
Address [6000 J Street, Dept. EEE, Sacramento, CA 95819-6019 ]
Voice:[916 278 6568], FAX: [916 278 7215], E-Mail:[[email protected]]
Re: [This submission is in response to the IEEE P802.15.4a Alternate PHY Call for Proposal ]
Abstract: [This document describes the waveform modulated UWB proposal for IEEE 802.15 TG4a.]
Purpose: [For discussion by IEEE 802.15 TG4a.]
Notice: This document has been prepared to assist the IEEE P802.15. It is offered as a basis for
discussion and is not binding on the contributing individual(s) or organization(s). The material in this
document is subject to change in form and content after further study. The contributor(s) reserve(s) the
right to add, amend or withdraw material contained herein.
Release: The contributor acknowledges and accepts that this contribution becomes the property of
IEEE and may be made publicly available by P802.15.
Submission
Slide 1
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
WAVE FORM MODULATED LOW
RATE UWB SYSTEM
- Proposal for 15.4a alt PHY-
Soo-Young Chang
California State University, Sacramento
Submission
Slide 2
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
INTRODUCTION
Submission
Slide 3
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
INTRODUCTION
• Use short duration impulses: purely processed in time
domain, not in frequency domain
– Simple concept: only a few components in TX and RX
– Simple digital processing  Low complexity  Low cost
– No components for processing frequency information (e.g. filter,
osc., etc.)
– High locating accuracy and fast ranging with very short duration
pulses
– Stealth mode of operation possible with relatively small RF
signature by coding frequency subbands with orthogonal
waveforms andcodes
– Excellent co-existence capability due to adaptive frequency band
usage – flexible to eliminate forbidden bands (e.g. UNII band)
Submission
Slide 4
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
PHY TASKS (1)
•
802.15.4 PAR – Purpose
[To provide a standard for ultra low complexity, ultra low cost, ultra low power consumption and low
data rate wireless connectivity among inexpensive devices. The raw data rate will be high enough
(maximum of 200kbs) to satisfy a set of simple needs such as interactive toys, but scaleable down to
the needs of sensor and automation needs (10kbps or below) for wireless communications.
•
802.15.4a PAR -- Purpose
[To provide a standard for a low complexity, low cost, low power consumption alternate PHY for
802.15.4 (comparable to the goals for 802.15.4). The precision ranging capability will be accurate
enough, several centimeters or more, and the range, robustness and mobility improved enough, to
satisfy an evolutionary set of industrial and consumer needs for WPAN communications. The project
will address the requirements to support sensor, control, logistic and peripheral networks in multiple
compliant co-located systems and also coexistence (18b).]
•
802.15.4 PAR – Scope
[This project will define the PHY and MAC specifications for low data rate wireless connectivity
with fixed, portable and moving devices with no battery or very limited battery consumption
requirements typically operating in the Personal Operating Space (POS) of 10 meters …
•
802.15.4a PAR – Scope
[This project will define an alternative PHY clause for a data communication standard with precision
ranging, extended range, enhanced robustness and mobility amendment to standard 802.15.4 (18a).]
Submission
Slide 5
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
PHY TASKS (2)
• Specified in existing15.4 standard
–
–
–
–
–
–
Activation and deactivation of the radio transceiver
ED within the current channel
LQI for received packets
CCA for CSMA-CA
Channel frequency selection
Data transmission and reception
• Range
– typical indoor range may be 10 to 30 m
– maximum outdoor range may be several km !!!
* ED: energy detection
* LQI: link quality indication
* CCA: clear channel assessment
Submission
Slide 6
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
PLAUSIBLE MYTHS
•
Myth 1
– ‘Low rate needs less power consumption.’
 With high rates, low power consumption can be achieved.
 Key issue is the amount of information delivered and power consumption is mainly related to transmission time
and processing time.
•
Myth 2
–
‘Digital implementation needs more complexity and is not easily realizable with the state-of-the art
technologies.’
 Digital implementation can be realized with less complexity and simple hardware and provide full flexibility
and adaptivity.
 As the processing power increases and technologies advances, full digital processing is the trend.
•
Myth 3
– ‘Higher frequency is not easy to manage or implement.’
 Unless high power is not considered, digital processing method can be applied for higher frequency band without
using power amplifiers.
•
Myth 4
– ‘Since this technology was not realizable yesterday, today also it is not easy to realize.’
 Since technologies advances rapidly, more sophisticated and conceptual ideas should be realized in the near
future and considered for future applications.
 Moore’s law says that processing power increases double every 18 months: cost amd complexity can be
decreased with the same rate.
Submission
Slide 7
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
CONSIDERATIONS
FOR LOW RATE UWB (1)
•
Frequency band
– Enjoy full frequency band assigned: 3.1 – 10.6 GHz in the US
– Only max power spectral density is limited: Transmitted power is proportional to
the bandwidth: more bandwidth means more transmitted power
– Pulse width is inversely proportional to bandwidth: more accurate ranging possible
for time based ranging
– Large bandwidth entails low fading
 High rate sampling is needed to process higher frequency signal using digital
methods
 To overcome this problem, new processing method should be devised
•
Transmit power
– Enjoy full power transmitted under frequency mask if waveforms have the
spectrum similar to frequency mask
– Max power will be -41.3dBm/MHz*7500MHz = -2.54dBm = 0.5mW
– More transmit power needs more power consumption ???: power consumption is
mainly related to processing time
 New waveform is needed to fit exactly to frequency mask
Submission
Slide 8
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
CONSIDERATIONS
FOR LOW RATE UWB (2)
•
Data rate
– In the technical requirements, “low rate” is suggested with expectation to reduce
power consumption and complexity/cost
– Power consumption is mainly proportional to the time duration of signal
transmission and processing
– No need to reduce data rates if higher rates possible with almost the same
cost/efforts
• With higher data rates, less probability of conflict with other transmissions for random
multiple access methods like CSMA and higher success rate with acknowledgements
– More pulses may be transmitted for the same information with higher rates: higher
robustness and more redundancy can be achieved: more flexibility can be provided
• The amount of information delivered is the key issue for any communication systems
• The higher the data rate is, the less time it takes to deliver.
 More sophisticated signal processing for higher rates and lower cost is inevitable.
Submission
Slide 9
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
CONSIDERATIONS
FOR LOW RATE UWB (3)
•
Full digital processing
– Provide full flexibility for any change in signal environments, system concepts and
requirements
– A variety of complex digital modulation schemes and any complicated system
concepts can be accommodated
– Eliminate the cost and complexity of a down conversion stage at receiver without
using oscillators (or crystals)
 Sophisticated digital signal processing technologies needed including high speed
ADCs and DACs with sampling rate > 1 Gsamples/sec
 Need to devise new signal processing implementations which may need new
technology
Submission
Slide 10
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
SYSTEM DESCRIPTION
Submission
Slide 11
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
KEY CONSIDERATIONS
• Modulation
• Source coding
• Channel coding (FEC)
– ARQ not considered
•
•
•
•
Interleaving
Pulse generation
Antenna
Multiple access
Submission
• Synchronization
• LNA
– accommodate ultra wideband
• Message relaying
• Simultaneously operated
piconets (SOP)
• Localization function
• Transmit only device
• Detection
Slide 12
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
FREQUENCY PLAN
• Flexible enough to satisfy any frequency mask and
to avoid any forbidden bands
 pulse waveforms can be adaptively tailored to any
frequency mask applied with any forbidden bands
• With FCC mask, 3.1GHz to 10.6 GHz full
frequency band can be used to enjoy more
transmitted power
 3.8 dB more power used than Gaussian pulse’s case
with the same frequency band
 3.8 dB more margin for link budget
Submission
Slide 13
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
FREQUENCY SUBBANDS
• Whole frequency band under FCC mask is divided into 4 groups
• Each group has 4 subbands
– BW of a subband = (10.6-3.1) GHz /16 = 469 MHz *
– Each subband has its own waveform: base waveform
group 1
group 2
group 3
group 4
f
3.1 GHz
10.6 GHz
subband 1
subband 2
subband 3
subband 4
f
base waveform
w21
w23
w22
w24
* If some bands should be abandoned, this subbamd should be a little bit smaller – for example the case that UNII band is excluded.
Submission
Slide 14
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
PULSE WAVEFORM
Submission
Slide 15
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
PULSE WAVEFORM OF SUBBAND
•
Pulse waveform shape
– Mathematical derivation/expression
– Shape: duration: 9 ns
– Spectrum: almost flat throughout the whole band
•
How can pulses be generated
– Digital way? Overlapped with various delays
 can be generated with relatively lower sampling rate DACs
• 90 samples/waveform:
• 16 waveforms/group for binary representation
81 waveforms/group for ternary representation
• 1440 or 7290 sample information stored in ROM per group
 1.44 or 7.29 Kbytes ROM needed to store waveform information if 8 bits/sample is
adopted
• Generate waveforms using DACs which have a sampling rate of 1 Gsamples/sec
– Analog way?
• No idea
– 4 digital ways considered in this proposal
•
How can delay devices for TX and RX be implemented?
 Cost/accuracy/step size are the key issues
Submission
Slide 16
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
amplitude in dB
TYPICAL PULSE WAVEFORM
(BASE WAVEFORM)
0.2
amplitude
0.15
0.1
0.05
0
2
0
-2
-4
-6
-8
-0.05
-10
-0.1
-12
-0.15
-14
-0.2
-2
-1.5 -1
-0.5
0
0.5
time
1
1.5
2
-16
-10 -8 -6 -4 -2
0 2 4
frequency
6
8
10
•
The above base waveform for bandwidth of 3.8 GHz, 20 samples/ns
•
For each subband, there is one waveform which has flat spectrum almost throughout the subbnad as
shown in the above.
Group i has four base waveforms: wi1, wi2 , wi3 , and wi4
Group i has 16 waveforms: mi1, mi2, mi3, . . . , mi16
•
•
mij,=a* wi1 +b* wi2 +c* wi3 +d* wi4
where a, b, c, and d are determined by modulation method applied
Submission
Slide 17
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
TYPICAL PULSE WAVEFORM
(BASE WAVEFORM)
•
The above base waveform for bandwidth of 0.469 GHz, 10 samples/ns
•
•
•
For each subband, there is one waveform which has flat spectrum as shown in the above.
Group i has four base waveforms: wi1, wi2 , wi3 , and wi4
Group i has 16 waveforms: mi1, mi2, mi3, . . . , mi16
mij,=a* wi1 +b* wi2 +c* wi3 +d* wi4
where a, b, c, and d are determined by modulation method applied
Submission
Slide 18
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
BASE WAVEFORMS FOR ONE GROUP
• For four subbands – assuming each
subband has 1 GHz BW
+
amplitude
– If smaller BW, larger pulse width
Submission
2
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
00
+
+
1
2
3
4
5 6 7 8 9
Frequency( GHz.)
t (ns)
10
0
Slide 19
4
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
BASE WAVEFORMS FOR ONE GROUP
• For four subbands - for smaller BW, larger pulse width
For BW of a subbnad in Group 1=469 MHz
subband 1
subband 2
subband 3
subband 4
Submission
Slide 20
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
ORTHOGONALITY OF WAVEFORMS
• For each subband, one base waveform exists
– 16 base waveforms throughout whole band (four groups):
w11(t), w12(t), w13(t), w14(t), w21(t), . . . . , w43(t), w44(t)
– Each waveform is almost orthogonal to each other or perfectly orthogonal
after de-emphasis at RX
• Each group has
– 16 waveforms for binary base waveform modulation (OOK or BPSK) or
– 81 waveforms for ternary base waveform modulation (OOK+BPSK)
– These waveforms are orthogonal to each other after de-emphasis at RX
• m1,1=0, m1,2= w1, m1,3= w2, . . . . , m4,16= w13+ w14+ w15+ w16 with OOK
• m1,1= -w1 - w2 – w3 – w4, . . . . , m4,16= w13+ w14+ w15+ w16 with BPSK
Submission
Slide 21
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
CORRELATIONS BETWEEN WAVEFORMS
•
Correlation
N
N
autocorrelation   si (k ) si (k )   si (k ) 2
*
k 1
k 1
where si (k ) : kth sample of ith base waveform of a group for N samples/waveform
N
crosscorre lation   si (k ) s j (k )
*
k 1
•
•
•
Submission
Ratio of correlations = autocorrel/crosscorrel for various N values
Orthogonality holds for sinusoidal waveforms with some conditions (Orthogonality
condition, refer to next slide), but the waveforms used here are not sinusoidal with some
envelope
– At receiver, de-emphasis can be used to make pure sinusoidal for a period
• mij*mij=(a* wi1 +b* wi2 +c* wi3 +d* wi4 )(a* wi1 +b* wi2 +c* wi3 +d* wi4) where
mij is the waveform transmitted and mij is the waveform generated at RX after
de-emphasis
• After integration for a one waveform duration, only autocorrelation terms
remain
• Orthogonality can hold at RX during detection for matched waveforms
– What is the best sampling frequency such that orthogonality can be achievable?
Less than 8 bits/sample will be enough for orthogonality evaluation? – need to verify
– “Power consumption of ADCs goes up exponentially with resolution”, EE times, Jan
17, 2005, pp 49
Slide 22
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
ORTHOGONALITY OF SINUSOIDS
•
A key property of sinusids is that they are orthogonal at different frequencies.
That is,
•
This is true whether they are complex or real, and whatever amplitude and
phase they may have. All that matters is that the frequencies be different. Note,
however, that the sinusoidal durations must be infinity.
For length N sampled sinusoidal signal segments exact orthogonality holds
only for the hamonics of the sampling rate-divided-by-N , i.e., only for the
frequencies
•
•
•
These are the only frequencies that have a whole number of periods in
samples
Ex. N=100 for 4 ns pulse duration, fs=25 GHz
– fk=k*25*10**9/100=2.5*10**8*k=0.25*k GHz
– For any integer k, fk can be determined  center frequencies of each subband can
be determined
http://ccrma.stanford.edu/~jos/r320/Orthogonality_Sinusoids.html
Submission
Slide 23
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
CORRELATIONS
BETWEEN TWO BASE WAVEFORMS
correlation
correlation ratio
w11
w12
w13
w11
0.020984
1/1
w12
0.0012155
17.264/9.7396
0.020984
1/1
w13
2.2562×10-5
930.05/3957.3
6.8651×10-6
305.66/106.69
w14
3.4173×10-6
6140.6/9681.8
0.0012155
2.2562×10-5
930.05/3957.3 17.264/9.7396
w14
0.0012155
2.2562×10-5 3.4173×10-6
17.264/9.7396 930.05/3957.3 6140.6/9681.8
6.8651×10-6 2.2562×10-5
305.66/106.69 930.05/3957.3
0.020984
1/1
0.0012155
17.264/9.7396
0.020984
1/1
– # of samples = 180
# of samples = 90
– Correlation ratio = autocorrelation/crosscorrelation
Submission
Slide 24
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MODULATION
Submission
Slide 25
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
POSSIBLE MODULATIONS
FOR EACH WAVEFORM
•
Each waveform can be modulated by using the following modulation schemes
depending on required data rates, system complexity, detection method, etc
mod
No. of levels
complexity
Data rate
Detection method
OOK
2 (+1, 0)
lowest
low
Non-coherent/coherent
Anti-podal:BPSK
2 (+1, -1)
low
low
Coherent/differential
OOK+anti-podal
3 (+1, 0, -1)
moderate
moderate
Coherent/differential
n level mod
n
high
high
Coherent/differential
nQAM
n
high
high
Coherent/differential
Submission
Slide 26
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MODULATION/MULTIPLE ACCESS (MA)
EFFICIENCY
•
Energy or power efficient?
–
–
Energy=power*time
Power limited by FCC mask
•






•
Pmax=-41.3dBm/MHz*7500MHz=-2.54dBm=0.5mW
to use more energy, more time needed to be transmitted  totally related to transmit time
for UWB, BW>500MHz or fractional BW>20% of fc  short duration pulses
one possibility to increase energy by using multiple pulses for one bit (or symbol)
need to use more power under frequency mask to have higher power
power constrained with frequency mask for LR-WPAN case
new waveform needed to fit the frequency mask to have more transmitted power
Spectrally efficient?
–
•
joule/sec
bit/Hz
Not important for UWB because of plenty of bandwidth
Time efficient?
–
–
Submission
bit/sec
For higher rate, more important: but for lower rate, less important  more room for flexibility
for LR-WPAN
However, as bit duration increases, more power consumption may be required
Slide 27
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
NO OF COMBINATIONS (BINARY MOD)
• For each subband, one base waveform exists
– 16 base waveforms throughout whole band:
w11(t), w12(t), w13(t), w14(t), w21(t), . . . . , w43(t), w44(t)
– Each waveform is almost orthogonal to each other
• For one symbol duration
 16 waveforms per group: Each group has 16 waveforms
• m1,1=0, m1,2= w1, m1,3= w2, . . . . , m4,16= w13+ w14+ w15+ w16 for OOK
• m1,1= -w1- w2- w3- w4, . . . . . , m4,16= w13+ w14+ w15+ w16 for BPSK
16 symbols in frequency domain because of 16 frequency bins
• For n durations in time domain
 to provide MA + FEC
– 16*n, 8*n, 4*n, 2*n, and 1*n symbols
 n can be specified for each type of devices/communications/applications
Submission
Slide 28
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
WAVEFORMS FOR EACH GROUP
(BINARY MOD)
group 1
group 2
group 3
group 4
f
3.1 GHz
Submission
10.6 GHz
m1,1(t)
m2,1
1,1(t)
m3,1
1,1(t)
m4,1(t)
m1,2(t)
m2,2
1,2(t)
m3,2
1,2(t)
m4,2(t)
m1,16(t)
m2,16
1,16(t)
m3,16
1,16(t)
m4,16(t)
Slide 29
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
SUBGROUPS FOR EACH GROUP
(BINARY MOD)
group 1
group 2
group 3
group 4
f
3.1 GHz
10.6 GHz
SG1
m1,1(t) m1,6(t)
m1,11(t) m1,16(t)
m2,1(t) m2,6(t)
m2,11(t) m2,16(t)
m3,1(t) m3,6(t)
m3,11(t) m3,16(t)
m4,1(t) m4,6(t)
m4,11(t) m4,16(t)
SG2
m1,4(t) m1,7(t)
m1,10(t) m1,13(t)
m2,4(t) m2,7(t)
m2,10(t) m2,13(t)
m3,4(t) m3,7(t)
m3,10(t) m3,13(t)
m4,4(t) m4,7(t)
m4,10(t) m4,13(t)
SG3
m1,2(t) m1,8(t)
m1,9(t) m1,15(t)
m2,2(t) m2,8(t)
m2,9(t) m2,15(t)
m3,2(t) m3,8(t)
m3,9(t) m3,15(t)
m4,2(t) m4,8(t)
m4,9(t) m4,15(t)
SG4
m1,3(t) m1,5(t)
m1,12(t) m1,14(t)
m2,3(t) m2,5(t)
m2,12(t) m2,14(t)
m3,3(t) m3,5(t)
m3,12(t) m3,14(t)
m4,3(t) m4,5(t)
m4,12(t) m4,14(t)
Submission
Slide 30
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MODULATION PROPOSED (1)
• 4 waveforms of a subgroup are mapped to 2 bit
(quaternary) information
ex)
m1,1(t)  00 m1,6(t)  01 m1,11(t)  10 m1,16(t)  11
• Each user sends information using one subgroup
of each group
 in one time duration 8 bit information is
delivered for whole band
• Each waveform is modulated by OOK (+1,0)
Submission
Slide 31
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
EXAMPLES OF WAVEFORMS (OOK)
m1,1(t)
Submission
m1,5(t)
Slide 32
m1,15(t)
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MODULATION PROPOSED (2)
• 4 waveforms of a subgroup are mapped to 2 bit
(quaternary) information
ex)
m1,1(t)  00 m1,6(t)  01 m1,11(t)  10 m1,16(t)  11
• Each user sends information using one subgroup
of each group
 in one time duration 8 bit information is
delivered for whole band
• Each waveform is modulated by BPSK (+1, -1)
Submission
Slide 33
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
EXAMPLES OF WAVEFORMS (BPSK)
m1,1(t)
Submission
m1,11(t)
Slide 34
m1,16(t)
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MODULATION PROPOSED (3)
• 16 waveforms of a group are mapped to 4
bit information
ex)
mi,1(t)  0000
mi,6(t)  0101 mi,11(t)  1010
mi,16(t)  1111
• Each user send information using one group
 in one time duration 4 bit information is
delivered
• Each waveform is modulated by OOK
(+1,0)
Submission
Slide 35
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
WAVEFORM FOR DATA STREAM (OOK)
FOR MODULATION PROPOSED (3)
Submission
Slide 36
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
NO OF COMBINATIONS (TERNARY MOD)
• For each subband, one base waveform exists
– 16 base waveforms throughout whole band:
w11(t), w12(t), w13(t), w14(t), w21(t), . . . . , w43(t), w44(t)
– Each waveform is almost orthogonal to each other
• For one symbol duration
 81 waveforms per group: 64 waveforms selected out of 81 waveforms per
group
• m1,1=0, m1,2= w1, m1,3= w2, . . . . , m4,16= w13+ w14+ w15+ w16 for OOK
• m1,1= -w1- w2- w3- w4, . . . . , m4,16= w13+ w14+ w15+ w16 for BPSK
 16 symbols in frequency domain because of 16 frequency bins
• For n durations in time domain
 to provide MA + FEC
– 64*n , 32*n , 16*n, 8*n, 4*n, 2*n, and 1*n symbols
 n can be specified for each type of devices/communications/applications
Submission
Slide 37
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
WAVEFORMS FOR EACH GROUP
(TERNARY MOD)
group 1
group 2
group 3
group 4
f
3.1 GHz
Submission
10.6 GHz
m1,1(t)
m2,1
1,1(t)
m3,1
1,1(t)
m4,1(t)
m1,2(t)
m2,2
1,2(t)
m3,2
1,2(t)
m4,2(t)
m1,64(t)
m2,64(t)
m3,64(t)
m4,64(t)
Slide 38
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
DATA RATES WITH 100% OVERHEAD
• 1 Mbps max with 100% overhead  Tb = 1/(2 Mbps) = 500 ns
• Pulse width = 9 ns  Duty cycle < 2 %
Mod type
# of waveforms
/subgroup
# of bits/symbol
duration
symbol duration
(ns)
Data rate w/100%
overhead
(Mbps)
Mods (1) & (2)
OOK or BPSK
4
8
500
8
Mod (2)
OOK + BPSK
16
16
500
16
500 ns
Submission
500 ns
Slide 39
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MULTIPLE ACCESS (MA)
Submission
Slide 40
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MULTIPLE ACCESS (MA)
• Possible MAs considered
– Frequency hopping (FH)
among subbands/groups
f
• Not efficient because of
uncertainty of FCC’s ruling
on FH so far and less usage
of power
Group 4
– TDMA
• Less time efficient
• More difficult to synchronize
Group 3
– Direct-sequence (DS) CDMA
Group 2
• Less time efficient and more
complex to process
Group 1
• New MA needed?
t1
t2
t3
t4
t5
t
16 frequency bins
time domain bins
Submission
Slide 41
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MULTIPLE ACCESS (A)
•
An orthogonal set of 8 8-bit Walsh codes is used
– Max autocorrelation, min (or zero) crosscorrelation each other
– One code consists of 8 frequency domain bins
– Minimal Hamming distance of this code set is 4
• One frequency bin error can be corrected while three bin errors can be detected; works as
an ECC code; increases robustness
•
8 SOPs case
– For one user, one code is assigned
– One time domain bin is occupied by two codes
• Each code represents one bit; one time domain bin represents two bits; during one time
domain bin two bits are delivered
• Hamming distances between two piconets codes is 4.
•
For each frequency bin waveform, BPSK is applied
Submission
Slide 42
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MULTIPLE ACCESS (B)
•
An orthogonal set of 8 8-bit Walsh codes is used
– Max autocorrelation, min (or zero) crosscorrelation each other
– One code consists of 8 frequency domain bins
– Minimal Hamming distance of this code set is 4
• One frequency bin error can be corrected while three bin errors can be detected; works as
an ECC code; increases robustness
•
64 SOPs case
– For one user, two Walsh codes (16 bits) are assigned
– One time domain bin is occupied by two codes
• two codes represent one bit; one time domain bin represents one bit; one time domain bit
deliver one bit
• Hamming distances between two piconets codes are 4 and 8.
•
For each frequency bin waveform, BPSK is applied
Submission
Slide 43
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
MAPPING FREQUENCY BINS TO WALSH
ENCODED SYMBOLS
Submission
Slide 44
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
IMPLEMENTATION
Submission
Slide 45
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
TRANSMITTER STRUCTURE
• Simple structure with impulse radio concept
–
–
–
–
–
FEC encoder
Interleaver
Pulse generator
Modulator
Antenna
Data in
antenna
This part can be
realized using digital
processing
Data
manipulator
Source coding
Channel coding
interleaving
Submission
Slide 46
modulator
Pulse
generator
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
TRANSMITTER BLOCK DIAGRAM
S/P converter
data
manipulator
input data
encoding
interleaving
encryption
Submission
ROM, group 1
ROM, group 2
DAC
DAC
waveform transformer
waveform transformer
ROM, group 3
DAC
waveform transformer
ROM, group 4
DAC
waveform transformer
Slide 47
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
RECEIVER STRUCTURE
• Simple receiver structure
–
–
–
–
–
–
–
Antenna
LNA
Demodulator
Data detector
De-interleaver
Channel decoder
Synchronizer
- Pulse generator
- Location processor
Synch
Information
retriever
location
Pulse
generator
Data out
antenna
Submission
LNA
demodulator
detector
Slide 48
Data
De-manipulator
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
RECEIVING BLOCK
received
signal
correlation
pulse
generator
Time correlator concept
ROM
LNA
Submission
waveform
conditioner
Slide 49
ADC
correlator correlation
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
LINK BUDGET ANALYSIS
•AWGN and 0 dBi gain at TX/RX antennas assumed. Fc=5.73GHz
Parameter
Value
Value
Value
Information Data Rate
1 Mb/s
2 Mb/s
1 Mb/s
Average TX Power
-2.54 dBm
-2.54 dBm
-2.54 dBm
Total Path Loss
(49.15dB@1m + L2)
77.14 dB
(@ 30 meters)
67.60 dB
(@ 10 meters)
67.60 dB
(@ 10 meters)
Average RX Power
-79.68 dBm
-70.14 dBm
-70.14 dBm
Noise Power Per Bit
-114 dBm
-111 dBm
-114 dBm
RX Noise Figure
8 dB
8 dB
8 dB
Total Noise Power
-106 dBm
-103 dBm
-106 dBm
Required Eb/N0
6.25 dB
6.25 dB
6.25 dB
Implementation Loss
2.5 dB
3.0 dB
2.5 dB
Link Margin
17.57 dB
23.11 dB
22.61 dB
RX Sensitivity Level
-97.25 dBm
-93.25 dBm
-92.75 dBm
Submission
Slide 50
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
CONCLUSIONS
Submission
Slide 51
Soo-Young Chang, CSUS
doc.: IEEE 802.15-05-0028-02-004a
Mar. 2005
WHY THIS PROPOSAL?
•
More transmit power used under frequency mask
–
More margin: at least 3 dB more by using full power under any frequency-power
constraints with waveforms adaptive to frequency mask
 Spectrally efficient / more received signal power
 More chance to intercept signals
•
Very simple architecture
–
–
Directly generated pulse waveforms using ROM
Processing in digital methods
•
No need to have analog devices (e.g., mixer, LO, integrator, etc)
 low cost / low power consumption
•
High location accuracy
– Wider bandwidth for each waveforms  narrower pulse width
 more accurate location information
•
High adaptability to frequency, data rate, transmit power requirements
 high scalability in frequency, data rate, system configuration, waveform, etc.
Submission
Slide 52
Soo-Young Chang, CSUS