EC 6014 COGNITIVE RADIO UNIT I - SOFTWARE DEFINED RADIOpit.ac.in/file/jUNIT1EC6014Cognitive...
Transcript of EC 6014 COGNITIVE RADIO UNIT I - SOFTWARE DEFINED RADIOpit.ac.in/file/jUNIT1EC6014Cognitive...
1 | P a g e
EC 6014 – COGNITIVE RADIO
UNIT I - SOFTWARE DEFINED RADIO
With the rapid evolution of microelectronics, wireless transceivers are becoming more
versatile, powerful, and portable.
This has enabled the development of software-defined radio (SDR) technology, where the
radio transceivers perform the baseband processing entirely in software:
modulation/demodulation, error correction coding, and compression.
Definition of SDR
A software radio is a radio whose channel modulation waveforms are defined in software.
All wireless telephones are controlled by this software.
A software-defined radio (SDR) is a radio that can accommodate a significant range of RF
bands and air interface modes through software.
SDR introduced in 1991 is defined as a radio platform of which the functionality is at least
partially controlled or implemented in software.
Software-defined radio (SDR) is a practical reality due to convergence of two key
technologies: digital radio, and computer software.
Software-defined radio refers to technologies wherein these functionalities are performed
by software modules running on field programmable gate arrays (FPGAs), digital signal
processors (DSP), general-purpose processors (GPP), or a combination thereof. This
enables programmability of both Digital Up Conversion/ Digital Down Conversion
(DDC/DUC) and baseband processing blocks. Hence, operation characteristics of the
radio, such as coding, modulation type, and frequency band, can be changed simply by
loading a new software. Also multiple radio devices using different modulations can be
replaced by a single radio device that can perform the same task.
Any waveform defined in the memory of the SDR platform can be employed on any
frequency. Initially it is constrained by the conversion process between analog and digital
signaling domains, emergence of cheap highspeed Digital-to-Analog converters (DACs)
and Analog-to-Digital converters (ADCs) have brought ideal SDR concept of an entirely
software communication system implementation (including radio frequency functionality)
closer to a reality.
Evolution of Software-Defined Radio(SDR) (Nov/Dec 2016, Nov/Dec 2017)
Two decades ago most radios had no software at all, and those that had it didn’t do much
with it.
2 | P a g e
In 1993, Joseph Mitola III envisioned a very different kind of radio: A mostly digital radio
that could be reconfigured in fundamental ways just by changing the software code
running on it. He dubbed this software-defined radio.
Mitola’s vision started to become reality and in mid-1990s military radio systems were
invented in which software controlled most of the signal processing digitally, enabling one
set of hardware to work on many different frequencies and communication protocols.
Fig. 1.1 Evolution of Software-Defined Radio (SDR)
Fig.1.1 shows evolution of software defined radio from 1G, 2G, 3G and 4G.
3 | P a g e
Fig.1.2 Key Elements of Architecture Evolution
Fig.1.2 shows key elements of Architecture Evolution.
SDR Evolution in the United States
SDR concept started in the late 1970s with the introduction of multimode radios operating
in VHF band.
U.S. Air Force Avionics Laboratory initiated the Integrated Communication, Navigation,
Identification and Avionics (ICNIA) program in the late 1970s – Developed an
architecture to support multifunctional, multiband airborne radios in the 30 MHz -1600
MHz band – Successful flight test and final report delivery in 1992 – ICNIA radio was the
first programmable radio.
4 | P a g e
Fig. 1.3 SDR different types
In the late 1980s, the Air Force Research Laboratory initiated the Tactical Anti-Jam
Programmable Signal processor (TAJPSP) – a processor capable of simultaneous
waveform operations using modular approach – TAJPSP later evolved into the
SPEAKeasy program. Fig.1.3 shows SDR different types.
SPEAKeasy was a joint U.S. Government program to develop the architecture and
technology to meet future military requirements for multimedia networking operations –
The first significant military investment to integrate various existing radio families into
one family was COTS-based architecture which demonstrated multiband, multimode radio
capabilities in 1998. SPEAKeasy evolved into the Joint Tactical Radio System (JTRS).
JTRS Joint Program Office was established in 1999 – which was envisioned to be the next
generation tactical radio for future advanced military operations
Mission is to acquire a family of affordable, high-capacity tactical radios to provide
interoperable LOS/BLOS C4I capabilities to the war fighters.
SDR Forum provides expertise in software radio technology for the JTRS program. The
Object Management Group is working toward building an international commercial
standard on the Software Communications Architecture (SCA).
SPEAKeasy I and SPEAKeasy II radios
5 | P a g e
The first (known) example of this type of radio was the U.S. military’s SPEAKeasy I and
SPEAKeasy II radios, allowed units from different branches of armed forces to
communicate for the first time. However, the technology was costly and the first design
took up racks that had to be carried around in a large vehicle.
SPEAKeasy II was a much more compact radio, the size of two stacked pizza boxes, and
was the first SDR with sufficient DSP resources to handle many different kinds of
waveforms. SPEAKeasy II subsequently made its way into the U.S. Navy’s Digital
Modulator Radio (DMR) with many waveforms and modes, able to be remotely
controlled with an Ethernet interface.
SPEAKeasy II and DMR products evolved not only to define radio waveform features in
software, but also to develop an appropriate software architecture to enable porting the
software to an arbitrary hardware platform, thus achieving independence of the waveform
software specification and design from the underlying hardware.
Companies such as Vanu, AirSpan, and Etherstack currently offer SDR products for
cellular base stations. Vanu Inc., a U.S.-based company, has been focusing on the
commercial development of SDR business since 1998.
In 2001 the 3GNewsroom reported SDR base stations as the key solution to the 3G rollout
problem. The ability of SDR base stations to reconfigure and support multiple protocols
was thought to be the safest option for rolling out 3G.
In reality, SDR didn’t play the key role that was anticipated. However, a closer look at the
operator’s infrastructure shows that programmable devices have become a key component
of current 3G base stations.
In 2005 AnywaveTM GSM base station, became the first SDR product to receive approval
under the newly established software radio regulation.
Anywave base station runs on a general-purpose processing platform and provides a
software implementation of the base transceiver station (BTS), base station controller
(BSC), and transcoder and rate adaptation unit (TRAU) modules of the base sation
subsystem (BSS). It supports GSM and can be upgraded to GPRS and Edge. The product
was first deployed in rural Texas by Mid Tex Cellular in a trial, where Vanu base station
showed successfully howit could concurrently run a time division multiple access
(TDMA) and a GSM network, as well as remotely upgrade and fix bugs on the base station
via an Internet link.
AT&T and Nextel, expressed interest in the Anywave base station.
In March 2005 Airspan released the first commercially available SDRbased IEEE 802.16
base station. The AS.MAX base station uses picoarraysTM and a reference software
implementation of the IEEE 802.16d standard. The picoarray is a reconfigurable platform
that is 10 times faster in processing power than today’s DSPs. The AS.MAX base station
6 | P a g e
promises to be upgradeable to the next generation mobile 802.16e standard and so has the
potential to offer a future-proof route to operators looking to rolling out WiMAX services.
In addition to the preceding proprietary SDR platforms developed for the military and
commercial sectors, there is also significant progress in the SDR development in the open-
source research and university communities.
GNU Radio is an open-source architecture designed to run on general-purpose computers.
It is essentially a collection of DSP components and supports RF interface to the universal
software radio peripheral (USRP), an up- and down-convertor board coupled with ADC
and DAC capabilities, which can be coupled to a daughter RF board.
GNU Radio is extensively used as an entry-level SDR within the research community.
SDR Evolution in Europe
R&D in Advanced Communications in Europe (RACE) and Advanced Communications
Technology and Services (ACTS) programs – ACTS projects, FIRST and FRAMES, used
software radios to investigate next-generation air-interfaces.
RACE and ACTS focus on incorporating 3G and potentially 4G standards into its Global
System for Mobile (GSM) Communications network – Pave the way for more capable and
more powerful products and flexible services – Key research areas include receiver
arcchitecture, baseband DSP architecture, enabling technologies.
FIRST: Flexible Integrated Radio System and Technology FRAMES: Future Radio
Wideband Multiple Access System.
SDR Evolution in ASIA
In 1999, Japanese Institute of Electronics, Information and Communication Engineers
(IEICE) software radio group was formed with SDR Forum Radio.
In 2000, Korea Electromagnetic Engineering Society (KEES) monitor software radio
activities in Korea, Japan and Taiwan.
IEICE and KEES mission: – Promote R&D in SDR – Allow protocol, software, hardware
to be easily integrated for future radio system – Foster cross-organization and
collaboration among academia, industries and governments .
Digital Radio (DR)
7 | P a g e
Fig.1.4 Block diagram of a Generic Digital Radio.
As shown in Fig.1.4 shows Block Diagram of Generic Digital Radio, it has five sections:
Antenna section, receives (or transmits) information encoded in radio waves.
RF front-end section, which is responsible for transmitting/receiving radio frequency
signals from the antenna and converting them to an intermediate frequency (IF).
ADC/DAC section, which performs analog-to-digital/digital-to-analog conversion.
Digital up-Conversion (DUC) and Digital Down-Conversion (DDC) blocks, which
essentially perform modulations of the signal on the transmitting path and demodulation of
the signal on the receiving path.
Baseband section, which performs operations such as connection setup, equalization,
frequency hopping, coding/decoding, and correlation, while also implementing the link
layer protocol.
DDC/DUC and baseband processing operations require large computing power, and in a
conventional digital radio are implemented in dedicated hardware.
In programmable digital radio (PDR) systems baseband operations and link layer protocols
are implemented in software while the DDC/DUC functionality is performed using
application-specific integrated circuits (ASICs).
If the AD/DA conversion can be pushed further into the RF block, the programmability
can be extended to the RF front end and an ideal software radio can be implemented.
Architecture
8 | P a g e
Architecture is defined as the comprehensive, consistent set of functions, components and
design rules according to which systems of interest are organized, designed, and
constructed. A specific architecture entails partitioning of functions and components such
that functions are assigned to components and interfaces among components which
correspond to interfaces among functions.
When such functions and interfaces are defined in formal design rules via a public forum,
the resulting architectures are called open. The full economic benefits of open
architectures require the existence of a large commercial base which sometimes fails to
emerge in spite of openness. As system complexity increases, architecture becomes more
critical because of its power to either simplify and facilitate system development (a
“powerful” architecture) or to complicate development and impede progress (a “weak’
architecture).
An architecture is a framework in which a specified class of components is used to achieve
a specified family of functions (eg: communication services) within specific constraints
and design rules.
Fig 1.5 Radio architecture evolving towards a high level of complexity
Radio architectures may be plotted in the phase space of network organization versus channel data
rate, as shown in Fig. 1.5. These architectures have evolved from early point-to-point and
9 | P a g e
relatively chaotic peer networks (e.g., citizens band and push-to-talk mobile military radio
networks) toward more hierarchical structures with improved service quality. In addition, channel
data rates continue to increase through multiplexing and spectrum spreading.
In addition, channel data rates continue to increase through multiplexing and spectrum spreading.
In a multiple-hierarchy application, a single radio unit, typically a mobile terminal, participates in
more than one network hierarchy. A software radio terminal, for example, could operate in a GSM
network, an AMPS network, and a future satellite mobile network. Multiband multimode military
radios and future Personal Communications Systems (PCS) that seamlessly integrate multimedia
services across such diverse access modes represent the high end of that evolution and the focus of
this discussion. . The components of the canonical software radio consist of a power supply, an antenna, a
multiband RF converter, and a single chip containing A/D and D/A converters with an on-chip
general purpose processor and memory that perform the ,radio functions and required interfaces
illustrated in Fig. 1.6. The canonical mobile software radio terminal interfaces directly to the user
(e.g., via voice, data, fax. and/or multimedia). The canonical base station interfaces to the public
switched telephone network (PSTN).
Fully instrumented base stations support operations and maintenance, developers and researchers
via services development workstation(s). The placement of the A/D and D/A converters as close
to the antenna as possible and the definition of radio functions in software are the hallmarks of the
software radio. lthough software radios use digital techniques, software-controlled digital radios
are generally not software radios. The key difference is the total programmability of software
radios, includingprogrammable RF bands channel access modes, and channel modulation
Contemporary radio designs mix analog hard- ware, digital hardware, and software technologies.
It is instructive to consider software radios to better understand benefits, pitfalls and relationships
to other technologies. Software radios have become practical as costs per millions of instructions
per second (MIPS) of digital signal processors (DSPs) and general purpose central processor units
(CPUs) have dropped below U.S. $10 per MIPS. The economics of software radios become
increasingly compelling as demands for flexibility increase while these costs continue to drop by a
factor of two every few years. At the same time, absolute capacities continue to climb into the
hundreds of millions of floating-point operations per second (MFLOPS) per chip, at which point
software radios are compatible with commercial TDMA and CDMA applications.
In addition, A/D/A converters available in affordable single-board open architecture
configurations offer bandwidths of tens of MHz with the dynamic range required for software
radio applications. Multimedia requirements for the desktop and palmtop continue to exert
downward pressure on parts count and on power consumption of such chip sets, pushing the
software radio technology from the base station to the mobile terminal. Although the tradeoffs
among analog devices, low-power ASICs, DSP cores and embedded microprocessors in handsets
remain fluid, cutting-edge base stations employ, software radio architectures. Finally, the
multiband, multimode flexibility of software radios appears central to the goal of seamless
integration of PCS, land mobile and satellite mobile services (including truly nomadic computing)
toward which many of us aspire.
10 | P a g e
Fig.1.6 Canonical Hardware and Software Defined Radio
The software radio architecture is widely applicable to trunk radios, peer networks, air and
sea traffic management, mobile military communications and satellite mobile systems.
For simplicity, this overview in Fig.1.6 describes the software radio architecture in a
mobile cellular/PCS setting.
In an advanced application, a software radio does not just transmit: it characterizes the
available transmission channels, probes the propagation path, constructs an appropriate
channel modulation, electronically steers its transmit beam in the right direction, selects
the appropriate power level, and then transmits.
Again, in an advanced application, a software radio does not just receive: it characterizes
the energy distribution in the channel and in adjacent channels, recognizes the mode of the
incoming transmission, adaptively nulls interferers, estimates the dynamic properties of
desired-signal multipath, coherently combines desired-signal multipath. adaptively
equalizes this ensemble, trellis decodes the channel modulation, and then corrects residual
errors via forward error control (FEC) decoding to receive the signal with lowest possible
BER as shown in Fig. 1.7.
11 | P a g e
Fig.1.7 Software Radio Functional Architecture Mobile Cellular Base Station
The Real-Time Channel Processing Stream
The canonical software radio architecture includes the channel processing stream. the
environment management stream and associated software tools illustrated in Fig. 1.7. The
real-time channel processing stream incorporates channel coding and radio access
protocols. Channel processing is char act e r i ze d by discrete time poi n t operations such
as the translation of a baseband signal to an intermediate frequency (IF) by multiplying a
discrete time-domain baseband waveform by a discrete reference carrier to yield a sampled
IF signal.
The time between samples is on the order of tens of microseconds to hundreds of
nanoseconds. Such point-operations require hundreds of MIPS and/or MFLOPS to Giga-
FLOPS with strictly isochronous performance. That is. Sampled data values must be
computationally produced and consumed within timing windows on the order of the time
between samples in order to maintain the integrity of the signals represented therein.
Input/output (I/O) data rates of this stream approach a gigabit per sec and per AID
converter.
Although these data rates are decimated through processing, it is challenging to sustain
isochronism through I/O interfaces and hard real-time embedded software in this stream.
Multiprocessing is therefore best organized as a pipeline with sequential functions of the
12 | P a g e
stream assigned to serially interconnected processors, i.e., a multiple instruction multiple
data-stream (MIMD) multiprocessing architecture.
Environment Management Stream
The near-real-time environment management stream continuously characterizes radio
environment usage in frequency, time and space. This characterization includes channel
identification and the estimation of other parameters such as channel interference levels
(depending on the specific signaling and multiple access scheme) and subscriber locations.
The environment management stream employs block operations such as fast Fourier
transforms (FFTs), wavelet transforms, and matrix multiplies for beam forming. Channel
identification results are needed in times on the order of hundreds of microseconds to
hundreds of milliseconds, while power levels may be updated in milliseconds and
subscriber locations may be updated less frequently.
The block structure of such operations is readily accommodated by a MIMD parallel
processor. The interface between this highly parallel environment management stream and
the pipelined channel processing streams must synchronize the environment management
parameters to the channel processing streams
On-Line and Off-Line Software Tools
On-line and off-line systems analysis, signal processing, and rehosting tools illustrated in
Fig. 1.7 allow one to define incremental service enhancements. For example, an enhanced
beamformer, equalizer and trellis decoder may be needed to increase subscriber density.
These enhancements may be prototyped and linked into the channel processing stream,
allowing one to debug the algorithm(s), to experiment with parameter settings, and to
determine the service value (e.g., in improved subscriber density) and resources impact
(e.g., on processing resources, I/O bandwidth, and time delays).
Software-based enhancements may be organized around managed objects, collections of
data and associated executable procedures that work with object resource brokers and
conform to related open architecture software interface standards such as the Common
Object Resource Broker (CORBA). Enhancements may then be delivered over the air to
other software radio nodes, as contemplated in the future software-defined
telecommunications architectures being considered by ITU-T and embraced by NTT and
others.
A well integrated set of analysis and rehosting tools leads to the creation of incremental
software enhancements relatively quickly, with service upgrades provided over-the-air as
software-defined networks proliferate. Technology limitations that require hardware-based
delivery are overcome by mapping critical elements of the service enhancement to
hardware via VHDL.
Partitioning of The Channel Processing Stream
13 | P a g e
The classical canonical model of communications concatenates source encoder, channel
encoder, channel (which adds noise, interference, and distortion channel decoder, and
source decoder. The channel encoderidecoder and related radio access functionsconstitute
the real-time stream.The canonical software radio architecture partitions classical channel
coding and decoding into the channel access segments of Fig. 1.8.
These segments are: antennas, RF conversion, IF processing, baseband processing and
bitstream processing. This canonical partitioning is useful because of the significant
differences in functionality between segments, because of the strong cohesion among
functions within a segment; because of large changes in bandwidth due to decimation
within a segment; and because of the ease with which these particular segments are
mapped to affordable open-architecture hardware. Further rationale for this partitioning is
based on set-theoretic considerations. This partitioning also structures the estimation of
first-order resource requirements so that they may be combined in ways that accurately
predict system performance.
Fig.1.8 Canonical Software Radio Functional Architecture
Challenges of SDR
There are a number of challenges in the transition from hardware radio to software (-
defined) radio.
First, transition from hardware to software processing results in a substantial increase in
computation, which in turn results in increased power consumption.
14 | P a g e
This reduces battery life and is one of the key reasons why software-defined radios have
not been deployed yet in end-user.
Issues relating to SDR
The main issue with introducing SDR into portable devices is that it requires the use of
programmable platforms, which are generally power hungry and hence lead to reduced
battery life and large devices.
Advantages of SDR-Potential Benefits of SDR (Nov/Dec 2016)
SDR provides the ability to support multiple waveforms on a single device, and so
ultimately could give an end user increased choice of services if incorporated into a
portable device, such as a handset.
SDR also assist seamless roaming at the national and international levels.
In late 1990s SDR started to spread from the military domain to the commercial sector,
with the pace of penetration into this market considerably accelerating in the new
millennium. Cellular networks are considered potentially most lucrative market for SDR
could penetrate.
Interoperability through the ability to support multiple standards.
Benefits for industry include a general-purpose and economic hardware platform, future-
proofing and easier bug fixes through software upgrades, increased functionality.
SDR can provide potential benefits for aviation community.
Accommodate multiple air-interface standards
Facilitate transition by bridging legacy and future technologies
Allow multiple services – incentives for equipage
Implement future-proof concept
Capable for insertions of future technologies and allow easy upgrades
Implement open-architecture to allow multiple vendors to supply or participate, offer
declining prices
Reduce product development time
Enable other advanced commercial technologies to be adapted to offer user’s services and
benefits
Drawbacks of SDR
Due to its high demand on computation and processing, SDR technology works only in
devices that have less constraint in size and power consumption, such as base stations and
moving vehicles.
15 | P a g e
SDR Applications
SDR is currently used mostly in military applications, where cost is less of a constraint.
SDR is used in commercial sector.
SDR is currently used to build radios that support multiple interface technologies (e.g.,
CDMA, GSM, and WiFi) with a single modem by reconfiguring it in software.
However, SDR modems are expensive, since they typically entail programmable devices
like FPGAs, as opposed to the mass-produced, single-purpose Application Specific
Integrated Circuit (ASIC)s used in most consumer devices today (and are key enablers for
low-cost handsets). Today’s multimode devices tend to just have multiple ASICs (or
multiple cores on a single ASIC).
SDR is a modem technology and it ignores RF design issues. In particular, the RF design
of a wireless device is typically closely coupled with the underlying access technology and
modem design. For example, different air interface technologies have different spectral
mask requirements and different degrees of vulnerability to cochannel interference and
strong adjacent channel power. A device that must work over a wide bandwidth or over a
wide range of RF signal scenarios (i.e., what other devices are operating in the nearby
spectrum neighborhood) will be more complex and expensive than a single-purpose
device.
But there is an increasing demand for SDR to enter portable and handheld devices
in the future.
However, as new processing platforms emerge that overcome power and size
constraints, it is very likely that SDR will make its way into portable devices.
Technology Tradeoff’s (Nov/Dec 2016, April/May 2017, Nov/Dec 2017)
Segment Design Tradeoffs
Segment design tradeoff’s are Antenna Tradeoffs, RF and IF Processing Tradeoffs, ADC
Tradeoffs, Digital Architecture Tradeoffs, Software Architecture Tradeoffs, Performance
Management Tradeoffs, End-to-End Tradeoffs.
Antenna Tradeoff’s
The antennas of the software radio span multiple bands, up to multiple octaves per band
with uniform shape and low losses to provide access to available service bands. In military
16 | P a g e
applications, for example, a mobile terminal may need to employ VHF/UHF line of sight
frequencies, UHF satellite communications, and HF as a backup mode.
As a result, the competing demands for directional selectivity, multipath compensation and
interference suppression versus wideband low-loss antennas versus affordability define the
tradeoffs of the antenna segment
Antenna architecture determines the number and bandwidth of RF channels. This
constrains the number and bandwidth of Analog to Digital to Converters (ADC). Some
waveforms require dedicated Application Specific Integrated Circuits(ASIC) instead of
ADC. Digital streams connect FPGAs, DSPs and general purpose process yielding a multi-
threaded, multi-tasking, multi-processing operating environment.
Flexible antennas, RF hardware and IF processing is a major technology challenge for
software radio. Optimum analog performance requires resonant narrowband antennas.
Fig.1.9 shows results in multiple parallel antenna/RF-conversion channels. In the example
A Personal Digital Assistant(PDA) access 1G cellular AMPS, 2G digital cellular Personal
Communication Systems (PCS) or 3G waveforms in the 1G or 2G bands. For location
aware services it has a GPS receiver. It also uses corporate wireless LAN (WLAN).
Fig. 1.9 Antenna Tradeoff
Broadband approach of above Fig.1.9 simplifies antenna and RF to two parallel channels,
reducing parts count. Next Fig.1.10 shows a unitary wideband channel. The antenna
response is not uniform across such a broad RF range. High performance in multiple RF
bands drives towards parallel narrowband channels. This can be an effective approach if
cost is not an issue. Transmission efficiency and impedance matching are more
challenging as bandwidth increases. Since antennas, RF conversion, IF Processing and the
ADC can account for over 60% of the manufacturing cost of an SDR, reducing the number
of RF channels is a significant design goal.
17 | P a g e
Fig.1.10 Antenna Tradeoff
Anticipating the Joint Tactical Radio System (JTRS) program of the US Department of
Defense (DoD), SPEAKeasy RF range extended from 2 MHz and 2 GHz, a ratio of
through, since the maximum relative-bandwidths of well-established designs are atmost
10:1, one decade. Through in-depth technology tradeoffs it was determined that atleast 3
bands are needed. SPEAKeasy bands were 1)2-30 MHz 2) 30-400 MHz and 3) 0.4 to 2
GHz. Band 2 was implemented in SPEAKeasy I. Bands 1 and 2 are implemented in
SPEAKeasy II. Currently affordable RF access is limited to less than one decade per
channel.
It includes the following parameters RF Access, Parameter Control, Linearity and Phase
Noise, Parameters for Emitter Locations, Packaging, Installation, and Operational
Challenges, Gain versus Packaging, Bandwidth versus Packaging, Antenna Calibration,
Antenna Separation, Human Body Interactions, Antenna Diversity, Spatial Coherence
Analysis Potential Benefits of Spatial Diversity, Spatial and Spectral Diversity, Diversity
Architecture Tradeoffs, Programmable Antennas, Cost Tradeoffs
RF and IF Processing Tradeoffs
RF/IF Conversion Segment Tradeoffs includes
RF Conversion Architectures
Receiver Architectures
The Superheterodyne Receiver
Direct Conversion Receiver
Digital-RF Receivers
Interference Suppression
RF Component Technology
RF MEMS
18 | P a g e
Superconducting Filters
Dual-Mode Amplifiers
Electronically Programmable Analog
Components
RF Subsystem Performance
RF/IF Conversion Issues
The second tradeoff concerns RF and IF conversion. The transmitter requires both linear
operation (eg: for QAM waveforms) and non-linear operation (eg:class-C amplifier for
high power efficiency with FSK or PSK waveforms). Fig. 1.11 shows IF Tradeoff where
desired response is obtained after 70 MHz and spurious response is obtained at 37 MHz in
IF spectrum. Also thermal noise and LO leakage are other side effects of IF processing.
Fig.1.11 IF processing tradeoff
Single channel receivers may non-linearly distort the waveform eg: in a direct conversion
architecture ever, must provide linear response for the strongest and weakest subscriber
signals(near-far ratio, typically 90 dB). The RF and IF conversion linearity and dynamic
range must match the ADC and Automatic Gain Control (AGC) and support digital
filtering and signal enhancement algorithms. The goal of this tradeoff is to balance the
noise, spurious components, intermodulation products and artifacts. The noise floor is
determined by the total bandwidth or by the low noise amplifier (LNA) eg: microwave
bands. Spurious responses and Local Oscillator (LO) leakage sometimes can mask
subscriber signals. LO leakage is problematic in homodyne receivers. A conservative
design keeps the peak energy of all noise, spurs and artifacts at about half of the least
significant bit (LSB) of the wideband ADC.
RF conversion includes output power generation, preamplification, and conversion of RF
signals to and from standard intermediate frequencies (IFS) suitable for wideband AIDIA
19 | P a g e
conversion. In most radio bands, RF conversion will be analog. Certain critical RF
problems are exacerbated in the software radio. These include the need for amplifier,
linearity and efficiency across the access band. Fig.1.12 shows the diagram for RF
conversion and tradeoff
Fig.1.12 RF processing tradeoff with interference suppression
RF shielding of processors may also be necessary to avoid the introduction of processor
clock harmonics into the analog RF,’IF circuits. Consisting of multiple transmitters also
creates electromagnetic interference (EMI) problems. but these are about the same for
software radios as for cosite collections of multiple discrete hardware radios.
The IF processing segment maps the transmit and receive signals between modulated
baseband and IF. The IF receiver processing segment including wideband digital filtering
to select a service band from among those available. Furthermore. IF filtering recovers
medium band channels (e.g.. J 700 kHz TDMA channel in GSM) and/or wideband
subscriber channels (e.g., a 2 MHz CDMA channel) and converts the signal to baseband.
The complexity of frequency conversion and filtering is the first order determinant of the
processing demand of the IF segment.
In a typical application, a 12.5 MHz mobile cellular band is sampled at 30.72 MHz (M
samples per second). Frequency translation, filtering and decimation requiring I/O
operations per sample equates to more than 3000 MIPS of processing demand. Although
such microprocessors are on the horizon, contemporary implementations offload this
20 | P a g e
computationally intensive demand to dedicated chips such as the Harris Decimating Down
Converter (DDC) or Gray digital receiver chip. Spreading and de-spreading of CDMA, is
also an IF processing function and creates demand that is proportional to the bandwidth of
the spreading waveform (typically the chip rate) times the baseband signal bandwidth.
This is so computationally intensive that with current technology limitations, it is typically
assigned to dedicated chips.Fig.1.12 shows Interference Suppression Tradeoff where RF
input is passed into a band of preselectors which selects the predefined frequency and
further passed into a low pass filter for removing the high frequency components. The
output of the filter is passed into step attenuator and the mixer and local oscillator
generates the IF signal with suppression for passing into the subsequent stages.Fig.1.13
shows the roofing filter response for suppressing the interference before use of roof and
after roofing filter.
Fig. 1.13 Interference Suppression Tradeoff
The Baseband Processing Segment
The baseband segment imparts the first level of channel modulation onto the signal (and
conversely demodulates the signal in the receiver). Predistortion for nonlinear channels
would be included in baseband processing. Trellis coding and soft decision parameter
estimation also occur in the baseband processing segment. The complexity of this segment
therefore depends on the bandwidth at baseband wb, the complexity of the channel
waveform, and the complexity of related processing (e.g., soft decision support). For
typical digitally encoded baseband waveforms such as binary phase shift keying (BPSK),
quadrature phase shift keying (QPSK), Gaussian minimal shift keying (GMSK).
21 | P a g e
In the transmission side of the baseband segment, such waveforms are generated one
sample at a time (a “point operation”). If three samples are generated for the highest
frequency component, demand falls between Rb and 6*Rb. Greater oversampling
decreases the transmitted power of spectral artifacts, but also increases transmit power and
processing demand. In the receiver, digital baseband modulations require timing recovery
which typically includes the integration of baud intervals over time. If baud interval is
measured in transitions of a high-speed clock, some timing-sensitive signal structures (e.g.,
TDMA) and some synchronization algorithms require up to 96 b precision integer
arithmetic in the clock recovery loop(s), and such extended precision arithmetic may not
be readily available, particularly on newer chips.
Bitstream Segment
The bitstream segment digitally multiplexes sourcecoded bitstreamsfrom multiple users
(and, conversely, frames and demultiplexes them). The bitstream segment imparts forward
error control (FEC) onto the bitstream, including bit interleaving and block and/or
convolutional coding and/or automatic repeat request (ARQ) detection and response.
Frame alignment, bit-stuffing, and radio link encryption occur in the bitstream segment.
Encryption requires the isolationof encrypted bits from clearbits, resulting in the
requirement to partition and isolate bitstream hardware accordingly.
Source Segment
The source segment differs between the mobile terminal and the base station. In the mobile
terminal, the source segment consists of the user and the source encoders and decoders.
Here, the relatively narrowband voice and fax A/D/A converters are typically located in
the handset, palmtop, or workstation. In the base station, on the other hand, the source
segment consists of the interface to the PSTN for access to remote source coding.
Conversion of protocols required for interoperabilitywith the PSTN creates processing
demand in the base station’s source segment. Conversion of DSO 64 kb/s PCM to RPE-
LTP (GSM [ll])f,o r example, would create 1 to 2 MIPS of demand per subscriber.
End-to-End Timing Budgets
Time delays are introduced in the IF, baseband, bitstream and source segments due to
finite I/0 and processing resources that empty and fill buffers in finite but sometimes
random amounts of time. The end-to-end accumulation of these delays must be kept within
bounds of isochronism at each segment-to-segment interface. These bounds depend on
signal type and the larger network architecture. Thus, for example, end-to-end voice delay
should be less than 150 ms, but the external network may consume 100 ms of this timing
budget, leaving only 50 ms for the software radio. Maintaining such budgets in software
radios is complicated (compared to digital radios) by queuing delays internal to and
between processors.
Interference Suppression
22 | P a g e
Antenna separation, frequency separation, programmable analog notch filters and active
cancellation suppress the interference at the RF stage. A programmable interference
suppression filter is called a roofing filter due to fact that the interference sets the
maximum linearly processable signal level(“roof”) while the dynamic range sets the
minimum floor.
Without roofing filter, the roof of the dynamic range is so high that weak signals fall
below the floor resulting in the dropped calls.
Active cancellation is the process of introducing a replica of the transmitted signal into the
receiver so that it may be coherently subtracted from the input signal.
Wideband antennas and RF exacerbate interference. SDR algorithms contribute to
interference suppression. SDR monitors mutual constraint satisfaction to minimize self-
generated interference.
RF MEMS
• Most RF integrated circuits require off-chip resonators, inductors and capacitors. Each
discrete device increases the cost of production manufacturing which is nearly a linear
function of the number of parts. RF MEMS replaces these with on-chip 3D structures.
MEMS RF switches are an electromechanical alternative to PIN diode switching circuits
substantially reducing size, weight and power while improving performance. MEMS switches
and tunable capacitors operate upto 40 GHZ. Fig.1.14 and Fig. 1.15 shows RF MEMS and its
fabric.
Fig.1.14 RF MEMS
23 | P a g e
Fig. 1.15 RF MEMS SWITCH FABRIC
Architecture Tradeoffs
Software radios ideally place all IF, baseband, bitstream and source processing in a single
processor. The assessment of the feasibility of the software radio centers on comparing
estimated demand to the capacities of the available processors. Implementations back off
from the ideal single- CPU implementation where driven to do so by processor, memory,
or interconnect technology limitations, or to achieve cost advantages (e.g., of off-loading
filtering to a Harris chip so that the DSP is less expensive). Software tasks are then
structured into managed objects designed to run on any DSP or CPU with access to the
data and sufficient processing capacity
ADC and DAC Tradeoffs
It includes the following parameters
Dynamic Range (DNR) Budget
Anti-aliasing Filters
Clipping Distortion
Aperture Jitter
Quantization and Dynamic Range
Technology Limits
24 | P a g e
Quadrature Techniques
Bandpass Sampling (Digital Down Conversion) DAC Tradeoffs
Conversion Rate, Dynamic Range
Applications
ADC Product Evolution
Low-Power Wireless Applications
Digital RF
ADC Design Rules
Linearity
Measuring SNR
Noise Floor Matching
Figure of Merit
Technology Insertion
Wa th e bandwidth of the IF to be digitized, determines what kinds of AID techniques are
feasible. According to the Nyquist criterion for band limited signals f 5 , the sampling rate
of the AID converter, must be at least twice Wa. Practical systems typically require modest
oversampling: f, > 2.5 Wa .
Wideband A/D and D/Aconverters access broad instantaneous segments of spectrum.
typically IO to 50 MHz. Such wide access may also be achieved in parallel subbands of
more modest 1 to 10 MHz bandwidths each. The dynamic range of each parallel subband
depends on the dynamic range of the ND/A converters. Since the product of dynamic
range times sampling rate is approximately constant for a given NDIA technology.
narrower subbands generally increase the useful dynamic range, albeit at the cost of
increased system complexity.
The placement of wideband A/D/A conversion before the final IF and channel isolation
filters achieves three key architectural objectives: It enables digital signal processing
before detcction
I t reduces the cost of mixed channel access modes by consolidating IF and baseband
processing into programmable hardware. It focuses the component tradeoffs to a single
central issue: providing the computational resources bandwidth, memory. and processing
capacity) critical toeach architecture segment, power and cost constraints of the application
subject to the size. weight.
Digital Processing Tradeoffs
Metrics
Heterogeneous Multiprocessing Hardware
Hardware Classes
Digital Interconnect
25 | P a g e
Applications-Specific Integrated Circuits (ASICs)
Digital Filter ASICs
Forward Error Control (FEC) ASICs
Transceiver ASICs
Field-Programmable Gate Arrays (FPGAs)
Reconfigurable Hardware Platforms
FPGA-DSP Architecture Tradeoffs
Table-Driven Signal Generation
Evolutionary Design of FPGA Functions
DSP Architectures
DSP Cores for Wireless Basic DSP: The TMS320C30
Increasing Interconnect Capacity: The C40 and SHARC
Size–Power Tradeoffs: The C54x, and Motorola Chips
Toward Greater Parallelism: The C80 and
C6xx
Chips
Potential Technology Limits
INFOSEC Processor Architectures
The Clipper Chip—Key Escrow Approach
Programmable INFOSEC Modules
Host Processors
Software Architecture Tradeoffs
Software Design Process
Top-Down, Object-Oriented Design
Object-Oriented Design for SDR
Defining Software Objects
Software Architecture Analysis
SDR Software Architecture
SPEAKeasy I Software Architecture
Characteristics of Top-Level Objects
Specialized Tasks
SPEAKeasy II Code
Architecture Implications (Nov/Dec 2017)
Signal Digitization Extensive development of A/D capability in recent years by the
semiconductor industry has been stimulated by the prospects of digital wireless
applications. This has resulted in improvements in accuracy, linearity, sampling rates, and
resolution; however, the trade-off between A/D performance and sampling rate continues
to be a limitation. While the use of multiple A/D channels may be a short-term solution,
power consumption limits this approach to base station products.
26 | P a g e
Hardware Architectures European research into software radio architectures and
technology, as it relates to handset terminals and base stations, was recently reviewed.
Such work is focused mainly on 3G standards. Table 1 summarizes the relevant
collaborative projects, mainly undertaken within the European ACTS and ESPRIT
research programs, and gives sources for additional information.
The main issues explored within these public domain research projects include:
• RF hardware architecture innovation and RF semiconductor process technology evolution
• Baseband DSP architecture evolution, including specific accelerators and software-
programmable hardware reconfigurability
• Architectural and algorithmic implications of software radio, including combinations with other
advanced processing concepts (e.g., beamforming and multiple signal extraction)
• Network implications of software radio — developing a nonconstraining, open evolutionary
route from where we are today In addition, work is beginning on the development of tools,
libraries, and environments to support product development and software portability.
In practice, reflecting the wide gulf between the purist view of the software radio ideal and the
pragmatic reality of commercially available technology today, some handset manufacturers
choose to adopt an understandably cautious and pragmatic approach to the implementation of
software-defined handsets, evolving from today’s products. Fig.1.15 shows architecture
implications at a transceiver sections of software radio architecture (SRA).
Fig. 1.15 Architure Implications
27 | P a g e