Architecture of a novel configurable

International Journal of VLSI design & Communication Systems (VLSICS) Vol.6, No.4, August 2015
DOI : 10.5121/vlsic.2015.6404 35
ARCHITECTURE OF A NOVEL CONFIGURABLE
COMMUNICATION PROCESSOR FOR SDR
Amiya Karmakar1
, Amitabha Sinha2
, Pratik Kumar Sinha3
& Pijush Biswas4
1
Department of Computer Science & Engineering, West Bengal University of
Technology, Salt Lake City, Kolkata, West Bengal, India
2
B.C.Roy Engineering College, Durgapur, West Bengal, India
3
Department of Computer Science & Engineering, N.I.T., Durgapur, West Bengal, India
4
Department of Computer Science & Engineering, N.S.H.M., Kolkata,
West Bengal, India
ABSTRACT
The design of high performance Digital Signal Processing (DSP) Processors for Software Defined Radio
(SDR) with high degree of flexibility and low power consumption has been a major challenge to the
scientific community ever since its conception. The basic philosophy of SDR is to implement different
modulation or demodulation schemes on the same underlying hardware. Currently available high
performance DSP processors, optimized with ‘Very Large Instruction Word (VLIW)’ architecture and
multiply and accumulate (MAC) units, are unable to meet the near real time speed requirements of
Software Defined Radios (SDR) due to their inherent sequential execution of compute intensive signal
processing algorithms. Moreover, their power dissipation is considerably high. Even though, Application
Specific Integrated Circuits (ASIC) exhibit high performance, they are also not suitable because of their
lack of flexibility. Various references on FPGA based implementations of reconfigurable architectures for
SDRs are also available. However, the Look-up Table (LUT) based implementations of FPGAs are not
optimum and therefore, cannot offer highest performance at low silicon cost. Keeping this view, this paper
presents the design of a configurable communication processor for Software Defined Radio. The proposed
scheme features the performance of an ASIC based design combined with the flexibility of software.
Experimental results reveal that the proposed architecture has minimum hardware requirement, improved
silicon area utilization and low power dissipation.
KEYWORDS
Field Programming Gate Array (FPGA), Software Defined Radio (SDR), Digital Signal Processing (DSP)
Processor, Hardware Description language (HDL), Signal Flow Graph (SFG), Application Specific
Integrated Circuit (ASIC), Union of graph, Reconfigurable DSP Processor
1. INTRODUCTION
The communication industry has witnessed substantial changes in digital communication
standards. The journey of digital communication commenced with GSM, D-AMPS and PDC
systems and gradually, a large portion of digital communication solutions were replaced by
WCDMA, HSPA and CDMA 2x standards. Though these 3G standards offer much higher data
rate and reliability, customers have now begun leaning towards the services offered by 4G
standards with data rates reaching 100 Mbits/s, if not more. To cope up with this ever changing
scenario of communication schemes, it is felt that the current developments in communication
systems and protocols should not only be of low cost and high efficiency, but also adaptive and
reusable. In conventional radio communications, the hardware components used for modulations/
demodulations are fixed for a particular communication scheme. For large varieties of

36
applications, particularly for military and cell phone services, radio protocols need to be changed
in real time with very high speed. This necessitates the design of a high performance radio system
with high degree of flexibility. The “Software Defined Radio(SDR)” [1][2][3][4][5] offers
flexible, high performance radio systems where hardware components of modulators/
demodulators are implemented by means of software on different hardware platforms like DSP
processors or FPGAs.
Software Defined Radio (SDR) systems [6][7][8][9] are flexible communication systems where
components of classical radio systems are replaced by software modules. Implementing different
modules (modulators, demodulators, filters etc.) in software has the distinct advantage that SDR
systems can be modified on the fly to deploy different communication schemes [10][11] without
changing the hardware. Though the concept of SDR has been in existence for more than a decade,
introduction of latest Digital Signal processing Processors [12][13] and FPGAs have made newer
types of design of SDR possible. The block diagram of a SDR system is presented in Fig.1.
Figure 1. Block Diagram of SDR
Evaluation history and literature survey of SDR reveals that there are various types SDR systems,
projects and platforms. Joe Mitola[14] proposed software radio which was a software-based
transceiver. He used GSM base station- combination of Ferdensi's digital receiver and E-Systems
Melpar's. E-Systems Melpar sold the software radio idea to the US Air Force and Melpar built a
prototype commanders' tactical terminal in 1990-91 that employed Texas Instruments
TMS320C30 Processors. SPEAKEASY [15] is an open architecture and proves the concept of
multi-band, multi-mode software programmable radio operating from 2 MHz to 2 GHz. Tactical
radio systems as well as voice and data communications to aircraft and onto battlefield has been
used as Military applications and civilian applications. Academic projects, the CHARIOT and
SpectrumWare, were funded by DARPA. GNU Radios as well as some European projects has
also attracted a large number of SDR developers.

37
Flexible radio systems are generally implemented on programmable platforms like Digital Signal
processing Processors and FPGAs. In the case of DSP based implementation [16][17], programs
for the proposed architecture are executed on the DSP processor. Flexibility can be achieved as
the programs are easily modifiable. The major problem is that the power consumption in this case
is considerably high. Moreover the efficiency of the design will also be limited by the efficiency
of the instruction set of that particular digital signal processor. On the other hand FPGA
[18][19][20][21][22][23][24] is configured for a particular design by downloading the
corresponding bit streams onto FPGA. These bit streams are generated after synthesis of the
description of the circuit in some Hardware Description Language (HDL)[25][26] like VHDL,
Verilog etc. Changes in these HDL programs will thus be reflected in the final configuration of
the FPGA, thereby offering flexibility. However the drawback of FPGA based design lie within
the issues of un-optimized Look Up Table (LUT) based implementation, large silicon area usage,
high power dissipation, communication delay and considerable configuration latency. On the
contrary, Application Specific Integrated Circuits (ASICs) [27] provide lower power dissipation,
minimum latency time and higher silicon area utilization factor. The major issue with this type of
VLSI [28] based design is lack of flexibility in the architecture.
Considerable effort has been given by designers and researchers to implement software define
radio (SDR) system in various platforms, including recent works related to FPGA based
implementation[20][21][23][29] of SDR. These designs, though aim to leverage flexibility of
FPGAs, are not also free from the drawbacks discussed above. Jignesh Oza et.al;[29] in their
work have combine different modulation schemes into a single circuit and have implemented on
FPGA but no efforts were made to minimize the hardware requirement. Only selection lines were
used for choosing available schemes at a particular instant of time. On the other hand Tadayuki
Kamisaka, et. al; [30] proposed modulators where different modulation schemes can be chosen by
simply changing the ROM contents. Although it is useful for various data transmission speeds but
the operating speed was bottlenecked due to use of non-optimized LUT-based ROMs. FPGA
based design of composite of modulation schemes either by selection line or by union method
[23](graph theory method) is free from the inherent drawback like high latency time, un-
optimized LUT (i.e. poor silicon utilization factor) and high power consumption. On the other
hand ASIC based design is highly optimized, low power and high speed without or little
flexibility. Hence in this paper, ASIC design of composite modulators and their various level of
hardware optimization using graph theory methods (nomenclature, labelling, precedence
relationship and union of graph) have been presented in details. This design was also converted
into Verilog codes (HDL) and was synthesized by ASIC synthesis tool Synopsis tool sets-Design
Vision/design Compiler.
Keeping these issues in view, this paper aims to present a new reconfigurable architecture for an
ASIC based SDR architecture which offers high performance at the same time adding flexibility
features to overcome the drawback of legacy ASIC based systems.
2. PROPOSED ARCHITECTURE OF RECONFIGURABLE MODULATORS
2.1. Architecture of Reconfigurable Modulators
In the proposed architecture, for case study, a number of different modulation schemes [10] [11]
[31] [32] like amplitude shift keying(ASK), frequency shift keying (FSK), binary phase shift
keying (BPSK), Quadrature phase shift keying (QPSK) have been considered and the architecture
can be configured to work according to any one of the schemes at a given time. The main
intention is to perform all these above mentioned modulation operations using the same
underlying hardware platform while the software above the hardware layer controls the
configuration. The different modulation/demodulation schemes can be visualized as
interconnected building blocks where such building blocks are Adder, Multiplier, Comparator,

38
Multiplexor etc. These building blocks primitives have to be chosen in such a fashion that they
are basic enough so that a particular modulation scheme can be made by connecting these
primitives in appropriate order. Obviously a number of switches are required to maintain the
interconnections among the primitives for implementing a scheme. So, the architecture consists
of:
1. A number of primitive building blocks for performing the signal processing functions for
different modulation schemes. These blocks include Adder, Multiplier, Comparator,
Multiplexor etc.
2. A number of switches for implementing different routing functions, so that by interconnecting
different building blocks different modulation schemes can be implemented. These switches
can be implemented using Multiplexers.
3. A control unit or Logical unit is required to generate the control signals for configuring these
switches. This unit consists of a number of hardware switches and a Decoder. For a particular
modulation scheme to be implemented one particular hardware switch is to be on which sends
some particular bit fashion to the input of the decoder. The decoder in turn generates the
corresponding configuration bit stream for the routing switches to configure that particular
modulation scheme.
The main advantage of such architecture is that it can be configured to any one of the large
number of modulation schemes by generating different control signals to the Multiplexers /
switches resulting the speed of the hardware with the flexibility of the software.
The efficiency of such a scheme depends on the trade-off among three important factors. They
are: 1) minimum number of basic building blocks required for configuring to a desired
modulation scheme, 2) minimum number of switches required establishing a given modulation
scheme, 3) Minimum Routing delay.
Keeping these factors in focus, a graph theoretic approach for hardware optimization has been
considered in this paper.
2.2. Graph theoretic approach for minimization of hardware in composite
architecture
At first four the modulation schemes were visualized as a connection of primitive building blocks.
The same blocks existing in different modulations schemes are then identified and marked with
identical names. With these primitive blocks different Signal Flow Graphs (SFG) are drawn for
different modulation schemes. By combining the SFGs common blocks are eliminated and hence
there will be sufficient hardware reduction in composite circuit. Such practices are discussed in
the following subsection.
2.2.1 Nomenclature of primitive building blocks
Same functionality primitive building blocks have given same nomenclatures are as follows:
Random Number Generator (S1), comparator (B1), Multiplier (B2), Constant1 (B3), Cosine
Generator (S2), Constant2 (B4), Switch1 (B5), Gain(-1) (B6), Switch2 (B7), Adder (B8), Sine
wave generator (S3), Switch3 (B9). Using these primitive blocks as nodes and connecting them
accordingly, Signal Flow Graph (SFG)s of four modulation schemes have been drawn.

39
2.2.2 Signal Flow Graph
The Signal Flow Graphs (SFG) of four modulation schemes and their labelling are shown in the
following subsection.
2.2.2.a Labelling of SFGs
The labelling procedure has been so adapted that the nodes performing the same function has
been labelled by the same symbol in all the signal flow graphs. Further, the same nomenclatures
across the all SFGs are also used for the directed branches joining the same pair of nodes. This
avoids duplication of nodes and branches. The labels are listed in Table 1.
Table 1. Labels of directed branches in SFGs
Source Node Destination Node Label of the directed branch
S1 B1 e1
S1 B8 e2
S2 B2 e3
S2 B5 e4
S2 B6 e5
S2 B7 e6
S2 B9 e7
S3 B6 e8
S3 B7 e9
S3 B9 e10
B1 B2 e11
B1 B5 e12
B1 B9 e13
B3 B1 e14
B4 B8 e15
B6 B5 e16
B6 B7 e17
B8 B7 e18
2.2.2.b SFG representation of four individual modulation schemes
SFG of four individual modulation Schemes following the convention discussed above are shown
in Fig. 2, Fig. 3, Fig. 4 and Fig. 5.
Figure 2. Graph-I (SFG of ASK modulation) Figure 3. Graph-II (SFG of BPSK modulation)

40
Figure 4. Graph-III (SFG of QPSK modulation) Figure 5. Graph-IV (SFG of FSK modulation)
2.2.3 Construction of a Combined Architecture from the SFGs
In the previous section, the four SFGs for four modulation schemes have been presented. Now
using these SFGs and applying the graph theoretic operations [33] like “precedence relationship”
and “union of graph” to these SFGs, a new SFG can be developed for proposed architecture.
Union of graph property is used to eliminate the common portion of the different graphs.
However a combined signal flow graph resulting merely from union of graph would produce
erroneous results unless the order of data processing i.e. signal dependency in individual SFGs
remains preserved in the combined SFG.
2.2.3.a Precedence Relationship
In constructing network diagrams which represents any project comprising of several activities,
preserving precedence relationship is important so that the order of activities be properly
represented. In this work two standard operation research methods i.e. PERT and CPM methods
are used. Assuming that the time required by the signal to flow across all the branches are same
the weight of all branches are considered as 1. Now the precedence relationship for the four
modulation schemes is as follows:
For ASK (Graph-I): e1<e11, e14<e11 i.e. e1, e14<e11;
For BPSK (Graph-II): e1, e14<e12, e5<e16;
For QPSK (Graph-III): e2, e15<e18, e5, e8<e17;
For FSK (Graph-IV): e1, e14< e13;
2.2.3.b Union of Graphs
In graph theory, the union of two graph is defined as follows: IF G1 is a graph with vertex set v1
& edge set E1 and G2 be another graph with vertex set V2 & edge set E2 then their union denoted
by G1 U G2 is a graph with vertex set is V1 U V2 & edge set is E1U E2. Since the union
operation excludes duplicate entry of set elements, the union of two signal flow graphs union will
provide the minimum possible number of vertex and edges without affecting the desired
functionality of the combined network or combined system provided the precedence of operations
are not disrupted. Fig.6 and Fig.7 depicts the two individual directed graphs G1 and G2. Fig. 8
represents the union of two graphs G1 and G2. Now the pair of nodes (V1, V2) and their
connecting branch (e1) is same in both G1 and G2. But the resultant graph from the union
operation of G1 and G2 will have no duplication of nodes or branches.

41
Figure 6. Directed Graph G1 Figure 7. Directed Graph G2
Figure 8. Result of Union Operation G1U G2
In first step of union operation to construct the combined architecture, Graph-I and Graph-II are
combined using the Union and preserving the precedence relationship. The nodes S1, B1, S2 and
B3 are the common nodes in the both graph. These common nodes are combined strictly adhering
to the precedence relationship. The principle being the same pair of nodes and their connecting
branches having same directed path are merged. The node pair (S1, B1) and connecting branch e1
being present in both the graphs (Graph-I and Graph-II). The same thing is applicable for the
node pair (B3, B1) and connecting branch e14. The resultant graph from the union of graph
operation is shown in Fig.9.
Figure 9. Graph Showing Graph-I U Graph-II
In this work the union of graph operation has been applied in the following fashion to obtain the
combined SFG:

42
Step 1: (Graph-I) U (Graph-II)
Step 2: (Graph-I) U (Graph-II) U (Graph-III)
Step 3: (Graph-I) U (Graph-II) U (Graph-III) U (Graph-IV)
During the operation the following points were observed. In the above case (B1, B1, B1), (B6,
B6, B6) are the repeated internal nodes. B1 follows the same pair of branches e1, e14 in all the
three graphs I, II & IV. Hence the sum of the incoming signals is same in the all three graphs. The
outgoing signals, although different, do not affect the output of the combined system. Hence node
B1 will appear once. The repeated node B6 is represented in the Fig.10. The precedence
relationship of this repeated node are e16>e5, e17>e5, e17>e8 i.e. e16>e5, e17>e5, e8. Hence the
combined digraph will need a “Dummy branch” shown in dotted lines depicted in Fig. 12. The
final combined SFG is shown in Fig. 13 which is the union of all four graphs.
Figure 10. Same node B6 but with different input-output branches
Figure 11. Graph for Precedence Relationship e16 > e5, e17> e5,e8
Figure 12. Graph for dummy branch (Shown in dotted line)

43
Figure 13. The Composite Graph
3. ASIC IMPLEMENTATION OF COMPOSITE MODULATORS
3.1. Synthesis results for ASIC implementation of individual modulation schemes
The Synthesis report of VERILOG codes of four individual modulation schemes and combined
modulation scheme using Design Vision / Design Compiler [34]–a Synopsys tool has been
presented and also graphically represented in Fig. 14, Fig. 15 Fig. 16, Fig. 17 and Fig. 18.
Library(s) Used for both Area and Power calculation:
vtvt_tsmc180(File:home/ue/my_cadence_directory/vtvt_tsmc180_release/Synopsys_Libraries/lib
s/vtvt_tsmc180.db)
Power calculation parameters:
Operating Conditions: nom_pvt
Wire Load Model Mode: top
Global Operating Voltage = 1.8
Power-specific unit information:
Voltage Units = 1V , Capacitance Units = 1.000000ff
Time Units = 1ps , Dynamic Power Units = 1mW (derived from V, C, T units), Leakage Power
Units = 1mW

44
HDL Synthesis Reports of all four Individual Modulation Schemes:
Area Report:
ASK:
Number of ports: 40
Number of nets: 77
Number of cells: 45
Total cell area: 1965.675575
FSK:
Number of ports: 40
Number of nets: 47
Number of cells: 15
QPSK:
Number of ports: 40
Number of nets: 131
Number of cells: 99
BPSK:
Number of ports: 40
Number of nets: 91
Number of cells: 59
Power Report:
ASK:
Cell Internal Power = 84.2933 mW (64%)
Net Switching Power = 47.8766 mW (36%)
Total Dynamic Power = 132.1700 mW (100%)
Cell Leakage Power = 986.6118 pW
FSK:
Cell Leakage Power = 591.1196 pW
QPSK:
Cell Leakage Power = 2.4679 nW
BPSK:

45
HDL Synthesis Reports of the Proposed Configurable Combined Architecture:
Area:
Number of ports: 56
Number of nets: 181
Number of cells: 133
Number of buf/inv: 17
Power:
3.2. Analysis of the Synthesis Report
References are available in literature featuring different modulation/demodulation schemes in a
single architecture. However in the reported works combined circuits where implemented mainly
aggregating the individual circuits and the selection lines to choose a particular modulation
/demodulation scheme at any particular instance. On the contrary here in this work, the ASIC
based design, the hardware requirements of the combined circuits have been minimized using
union of graphs and ASIC synthesis tool –Synopsis tool like Design Vision/ Design Compiler.
Although FPGA implementation of different modulation schemes using union of graph shown in
literature but due to inherent drawback of FPGA like configuration latency, un-optimized LUT
and high power consumption this paper suggest ASIC implementation of four composited
modulation scheme with optimal union of graph. Various Graphs have been plotted form the
ASIC synthesis results. From the graphical analysis of the cell area usage experimental data
shown in Fig. 14, it can be clearly derived that the total cell area required to implement the
combined architecture is significantly less than the sum total of the all individual implementation.
The power analysis shown in Fig. 15 also reveals that the combined scheme requires substantially
less power compared to the total power requirement to implement the modulation schemes
individually. The port, net and cell analysis, switching and dynamic power analysis as well as cell
leakage power analysis depicted in Fig. 16, Fig. 17 and Fig. 18 also show that the implementation
of the modulation schemes in the combined architecture is better than implementing the
modulation schemes individually.
Figure 14. Area Analysis of individual and combined schemes

46
Figure 15. Cell Internal Power of individual and Combined Schemes
Figure 16. Number of Ports, Nets and Cells Versus individual and Combined Schemes
Figure 17. Switching and dynamic power analysis of individual and Combined Schemes

47
Figure 18. Cell Leakage Power Analysis of Individual and Combined Schemes
4. CONCLUSIONS
In this paper feasibility of ASIC implementation of a novel architecture of Reconfigurable DSP
Processor for Different Modulation Schemes with optimum hardware requirements has been
explored. The synthesis reports derived from the Synopsis tool have also been presented. It can be
concluded from the synthesis results that the ASIC implementation has substantially reduced area
and power requirements. The architecture is flexible because run time configuration for any
particular modulation scheme can be achieved by activating few switches only. Thus the
proposed architecture provides high performance of while retaining flexibility reduced power
requirements and improved silicon utilization factor.
To achieve enhanced performance, unions of the set of building blocks for different modulation /
demodulation schemes have been combined. Since the performance of most of the building
blocks depend on the speed of adders ,multipliers , exploring the possibilities of implementing
these arithmetic units using non-binary systems can be thought of . The choice of number system
can influence the execution speed of the arithmetic computation to a great extent. Recent studies
in computer arithmetic have unleashed that use of number systems like Double Base number
System[35], Ternary Number System[36], Logarithmic Number system [37], Single Digit Triple
Base Number System[38] etc. can improve the performance of digital signal processing to a great
extent. Thus, the performance of the proposed configurable communication Processor for SDR
may further be improved by leveraging the advantages of the non-conventional number systems.
This study is a potential future research issue related to the current work.
REFERENCES
[1] V. Bose, (1999) “The Impact of Software Radio on Wireless Networking”, Mobile Computing and
Communications Review, Vol. 3, Issue 1, pp30-37.
[2] V. Bose, M. Ismert, M. Welborn, & J. Guttag,(1999) “Virtual Radios”, IEEE/JSAC, Special Issue on
Software Radios, Vol. 17, No. 4, pp591-602.
[3] A. F. B. Selva, A. L. G. Reis, K. G. Lenzi, L. G. P. Meloni, (2012) “Introduction to the Sftware-
defined Radio Approach”, IEEE Latin America Transactions, Vol. 10, No.1.
[4] Vanu Bose, Roger Hu & Robert Morris,(2001) “Dynamic Physical Layers for Wireless Networks
Using Software Radio", Proc. IEEE International Conference on Acoustics, Speech, and Signal
Processing (ICASSP’01), Vol. 4, pp2045-2048.
[5] Chris H. Dick, H. M. Pedersen, (2001) “Design and Implementation of High-Performance FPGA
Signal Processing Data paths for Software Defined Radios”, Embedded System Conference.
[6] A. Saha & A. Sinha,(2008) “Radio Processor – A New Reconfigurable Architecture for Software
Defined Radio”, Proc. ICCSIT, IEEE Conference, Singapore.

48
[7] Qiyue Zou, Mohyee Mikhemar, Ali H. Sayed, (2009) “Digital Compensation of Cross-
ModulationDistortion in Software-Defined Radios”, IEEE journal of selected topics in signal
processing, Vol. 3, No. 3.
[8] Mitola, J., III; Maguire, G.Q., Jr.,(1999) “Cognitive radio: making software radios more personal”,
Personal Communications, IEEE, Vol. 6, Issue 4, pp13-18.
[9] Jefferson L. Xu, Wei Su, Meng Chu Zhou,(2010) “Software-Defined Radio Equipped with Rapid
Modulation Recognition”, IEEE Transactions on Vehicular Technology, Vol. 59, No. 4.
[10] Leon W. Couch II,(2005) Digital and Analog Communication systems, Pearson Education, Sixth
edition.
[11] Simon Haykin, (2005) Communication Systems, John Wiley & Sons, 4th Edition.
[12] Buyer’s guide to DSP Processors, 2004 Edition: Barkley Design Technology Inc.
[13] Pavel Sinha, Amitabha Sinha & Dhruba Basu,(2005) “A Novel Architecture for a Re-configurable
Parallel DSP Processor”, IEEE- NEWCAS Conference, Canada, pp71-74..
[14] Mitola, J., III, (1992) “Software radios-survey, critical evaluation and future directions”, Telesystems
Conference, NTC-92, pp13/15-13/23
[15] R. L. Lackey & D. W. Upmal, 1995 “Speakeasy: The Military Software Radio”, IEEE
communications Magazine, Vol. 33, Issue 5, pp56–61.
[16] Rulph Chassaing, (2005) Digital Signal Processing and Application with the c 6713 and c 6416 DSK,
NJ: John Wiley and Sons.
[17] Kwadwo Boateng and Charles Badu, (2006) EE452 Senior Capstone Project: Integration of Matlab
Tools for DSP Code Generation, Bradley university.
[18] Xilinx, (2004) Introduction and overview, Virtex-II Pro Platform FPGAs.
[19] Joanne De Groat, Gursharam Reehal & S. Nagarjuna, (2008) “Synthesizing FPGA Digital Modules
for Software Defined Radio”, IEEE Conference, NAECON, pp358-362.
[20] A. Saha, & A. Sinha, (2009) “An FPGA based Architecture of a Novel Reconfigurable Radio
Processor for Software Defined Radio”, International Conference on Education Technology and
Computer (ICETE), pp. 45-49.
[21] I. Hatai & I. Chakrabarti , (2011) “ A New High-Performance Digital FM Modulator and
Demodulator for Software Defined Radio and its FPGA Implementation”, International Journal of
Reconfigurable Computing, Vol. 2011, 10 pages, Article ID 342532.
[22] “Introduction and overview” (2012), Virtex-V Platform FPGAs, http://www.XILINX.com.
[23] A. Karmakar & A. Sinha, (2011) “A Novel Architecture of a Reconfigurable Radio Processor for
Implementing Different Modulation Schemes”, 3rd International Conference, ICCRD, IEEE, Vol. 1,
pp115-119.
[24] Wayne Wolf, (2005) FPGA Based System Design, Pearson Education (Singapore) pte. Ltd India, ch-
3, ch-4, pp121-319.
[25] M. D. Ciletti, (2005) Advanced Digital Design with the Verilog HDL, Prentice-Hall of India Private
Ltd., India, ch6 and ch8.
[26] D. E. Thomas & P.R. Moorby, (2002) The Verilog Hardware Description Language, 5th ed., Kluwer
Academic Publications.
[27] R. Singh, (2001) Signal Integraty Effects in Custom IC and ASIC Designs, Wiley-IEEE Press, ISBN:
0-471-15042-8.
[28] K, K. Parhi, (1999) VLSI Digital signal Processing Systems, A Wiley-Inter science Publication.
[29] Jignesh Oza1, Yogesh Patel2, Pratik Trivedi3, Nilesh Ranpura4, Zuber Patel5, Upena Dalal6, Rachna
Jani7, Vijay S.R., (2010) “Optimized configurable architecture of modulation techniques for SDR
applications”, International Conference on Computer and Communication Engineering (ICCCE’10),
Kuala Lumpur, Malaysia, IEEE.
[30] Tadayuki Kamisaka, Yasufumi Takahashi, Shigeki Nakamura, & Kazuo Yagi, “A Digital Modulation
VLSI Covering Various Modulation Techniques and Wide Range Data Speeds”, Global
Telecommunications Conference, GLOBECOM’88, IEEE, Vol. 1, pp141-147.
[31] J. C. Pliatsikas, C. S. Koukourlis & J. N. Sahalos, (2005) “On the Combining of the Amplitude and
Phase Modulation in the same Signal”, IEEE Transactions on the Broadcasting, Vol. 51, No. 2,
pp230-236.
[32] T. D. Memon, W. Ghangro, Dr. B. S. Chowdhry & A. A. Shaikh,(2009) “ Quadrature phase Shift
Keying Modulator & Demodulator for Wireless Modem”, 2nd IEEE Conf. on Computer, Control and
Communication, Karachi, pp1-6.
[33] Frank Harary, (1969) Graph Theory, Addison-Wesley Publishing Co. Inc.
[34] Synopsys Inc., (2005) Design Vision TM User Guide, Version X-2005.09.

49
[35] Dimitrov, V.S.; Jullien, G. A.; Miller, W. C.,(1999) “Theory and applications of the double-based
number system,” IEEE Transactions on Computers, Vol. 48, Issue 10, pp1098-1106.
[36] Brain Hayes, (2001) “Third Base”, American scientist, Vol. 89, No. 6, pp490-494.
[37] Swartzlander, E.E., Alexopoulos, A. G., (1975) “The Sign/ Logarithm Number System”, IEEE
Transactions on Computers, Vol. C-24, Issue 12, pp1238-1242.
[38] S. Maitra & A. Sinha, (2007) “A single digit triple base number system- a new concept for
implementing high performance multiplier unit for DSP applications”, 6th International Conference
on Information, Communications & Signal Processing, IEEE, pp1-5.
AUTHORS
Amiya Karmakar was born in Kolkata, India and obtained his Master of Engineering
(M. E.) in Computer Sc. & Engineering with specialization in Embedded Systems
from West Bengal University of Technology in 2005.Mr. Karmakar has been engaged
in teaching and research for the last nearly twelve years. He has been working with
West Bengal University of Technology as Project Officer for last nine years. Prior to
this he was faculty in a Polytechnic College for two years. His areas of research
include Embedded System Design, Re-configurable Architecture using FPGAs,
Software Defined Radio, VLSI design, Digital Signal Processing etc..
Prof.Amitabha Sinha was born in Kolkata and obtained Ph.D degree in Computer
Sc.& Engg. from Indian institute of Technology ( IIT), Delhi in 1984. He has been
working with Neotia Institute of Technology, an Institution affiliated to West Bengal
University of Technology as a Principal. Prior to this assignment, he was with West
Bengal University of Technology, Kolkata, India as a Professor and Director, school of
Engg. & Technology since August 2005/2006. He has more than 25 years of
experience in industry, premier academic institutes, R&D centers and IT/Telecom
organizations in India & abroad and his areas of research include Embedded System
Design, VLSI design, Digital Signal Processing, Re-configurable Architecture using FPGAs, Software
Defined Radio, Processor Architecture and System On-chip Design, Advanced Computer Architecture
etc..He was Assistant Professor of the Dept.of Comp.Sc.& Engg., Bengal Engineerig College ( Now Bengal
Engineering & Science University), West Bengal, India during 1985-1991 and Associate Professor of the
Dept.of Comp.Sc.&Engg, BITS Pilani, during 1997-1998. He was visiting faculty, Dept.of Computer
Sc.&Engg.,Oakland University, U.S.A. during 1994 and 1999. He was an expatriate faculty member of a
premier Malaysian University Mara Institute of Technology during 1994-1995. Prof. Sinha also served R
&D centres of different premier industries in the country at senior levels.
Mr. Pijush Biswas is a PhD student in Institute of Radiophysics and Electronics,
University of Calcutta, India. He has received his B.Tech and M.Tech in Information
Technology from University of Calcutta. His research interest includes Reconfigurable
Computing, VLSI Design and Computer Architecture. He is a student member of
ACM. He is a professional Fashion Photographer and enjoys Bird watching and
Nature conservation activities.

Architecture of a novel configurable

More Related Content

What's hot (20)

Viewers also liked (19)

Similar to Architecture of a novel configurable (20)

Recently uploaded (20)

Architecture of a novel configurable