08-06-2013, 03:17 PM
An Integrated Sensor Microsystem for Industrial and Biomedical Applications
[attachment=55027]
Abstract :
There is considerable interest in the development of
ultra-miniature and low-power sensor microsystems for use in
applications such as medical diagnostics, environmental
monitoring and other industrial applications. Such ultraminiature
sensor microsystems must contain a large diversity of
complex electronics, including sensor interfaces, signal
conditioning, a microprocessor core, digital signal processing, and
wireless transmission technology. In this paper, we will describe
the first steps towards the development of a System on Chip for
such a sensor microsystem and the methodology employed to build
such a microsystem.
INTRODUCTION
There is considerable interest in the development of ultraminiature
and low-power sensor microsystems for use in
applications such as medical diagnostics [1], environmental
monitoring [2] and other industrial applications. Such
systems are required to have many attributes, such as low
cost, robustness and real-time data processing. In many
applications, it is also required that the device be disposable
after a single usage.
An ultra-miniature sensor microsystem must contain a
large diversity of complex electronics, including sensor
interfaces, signal conditioning, a microprocessor core, digital
signal processing (DSP), and wireless transmission
technology. It is therefore desirable to use a design and
implementation methodology that lends itself to low cost and
can achieve a low form-factor and low-power consumption.
Such a methodology is that of system-on-chip (SoC)
technology [3], whereby a system containing many
intellectual property (IP) blocks can be rapidly designed. In
practice, the development of such a microsystem is a
multidisciplinary activity involving micro-electro-mechanics
[4], laboratory-on-a-chip [5], microfluidics [6] and
biochemical sensor technology.
SYSTEM SPECIFICATION
The sensor microsystem we have developed comprises
sensors, an application specific integrated circuit (ASIC) with
sensor interfaces, analogue and digital systems, a radio uplink
to a base station, and a power source. The component
count is therefore only four parts: the sensors, a mixed-signal
ASIC, the radio frequency (RF) transmitter and the power
cells. In future implementations a significant portion of the
RF section will be integrated on a SoC. The microsystem, or
capsule, has a simplex communication link to a base-station
that can handle data from several capsules. The base-station
communicates in turn via standard communication protocols
(TCP/IP or wireless) to a remote site that has an automated
database or an operator. In future implementations the
capsule to base-station link will be a duplex or half-duplex
channel enabling dynamic reconfiguration. Figure 1 shows
the function blocks of the present prototypical microsystem.
DESIGN METHODOLOGY
The design methodology we have employed in order to
achieve the ASIC uses state-of-the-art EDA solutions that
have been selected to provide a relatively straightforward
design flow. Cadence® tools are used for analogue
simulation, digital simulation and back-end design tasks.
Synopsys® tools are used for digital synthesis of
VHDL/Verilog behavioural descriptions [7]. Foundry
services are provided via Europractice. The foundry service
provider is Austria Mikro Systeme (AMS), and the prototype
SoC discussed in this paper has been implemented on a 3 V,
2-poly, 3-metal 0.6 micrometer CMOS process. A key
advantage of using the AMS service is the availability of well
specified analogue and digital IP blocks, such as ADCs and
DACs that have been used to build our design.
RESULTS
The ASIC has been successfully fabricated and chips have
been returned both as unpackaged die and in packages for
test purposes (see Figure. 4). The ASIC is pad-limited since it
has been designed with many contact pads for test purposes.
The total area is 20 mm2. There are 16,000 gates, 60 % of
which are digital, the remaining portion being part of the
analogue system. Of the 64 pads, 20 are power pads and 44
are I/O pads, more than half of which are used for
preliminary chip test. Current consumption of the ASIC from
a 3 V source is 1.1 mA. In addition, the sensors require an
average current of 0.5 mA and the transmitter requires an
average current of 1 mA when operating at a duty cycle of
15%. The microsystem can operate for up to 10 hours under
these conditions.
CONCLUSION
We have designed and implemented an application
specific integrated circuit for a sensor microsystem for
industrial and biomedical applications. Test results
demonstrate the ASIC operates as intended. A system-onchip
design methodology has also been successfully pipecleaned
and the present design contains modules that will be
re-used in future SoC designs. Future implementations will
lead to further miniaturisation and more sophisticated system
specification including greater sensor diversity, sensor fusion
and dynamic reconfigurability, we will also extend the work
to use a more sophisticated SoC methodology to enable the
rapid prototyping of such sensor microsystems.
[attachment=55027]
Abstract :
There is considerable interest in the development of
ultra-miniature and low-power sensor microsystems for use in
applications such as medical diagnostics, environmental
monitoring and other industrial applications. Such ultraminiature
sensor microsystems must contain a large diversity of
complex electronics, including sensor interfaces, signal
conditioning, a microprocessor core, digital signal processing, and
wireless transmission technology. In this paper, we will describe
the first steps towards the development of a System on Chip for
such a sensor microsystem and the methodology employed to build
such a microsystem.
INTRODUCTION
There is considerable interest in the development of ultraminiature
and low-power sensor microsystems for use in
applications such as medical diagnostics [1], environmental
monitoring [2] and other industrial applications. Such
systems are required to have many attributes, such as low
cost, robustness and real-time data processing. In many
applications, it is also required that the device be disposable
after a single usage.
An ultra-miniature sensor microsystem must contain a
large diversity of complex electronics, including sensor
interfaces, signal conditioning, a microprocessor core, digital
signal processing (DSP), and wireless transmission
technology. It is therefore desirable to use a design and
implementation methodology that lends itself to low cost and
can achieve a low form-factor and low-power consumption.
Such a methodology is that of system-on-chip (SoC)
technology [3], whereby a system containing many
intellectual property (IP) blocks can be rapidly designed. In
practice, the development of such a microsystem is a
multidisciplinary activity involving micro-electro-mechanics
[4], laboratory-on-a-chip [5], microfluidics [6] and
biochemical sensor technology.
SYSTEM SPECIFICATION
The sensor microsystem we have developed comprises
sensors, an application specific integrated circuit (ASIC) with
sensor interfaces, analogue and digital systems, a radio uplink
to a base station, and a power source. The component
count is therefore only four parts: the sensors, a mixed-signal
ASIC, the radio frequency (RF) transmitter and the power
cells. In future implementations a significant portion of the
RF section will be integrated on a SoC. The microsystem, or
capsule, has a simplex communication link to a base-station
that can handle data from several capsules. The base-station
communicates in turn via standard communication protocols
(TCP/IP or wireless) to a remote site that has an automated
database or an operator. In future implementations the
capsule to base-station link will be a duplex or half-duplex
channel enabling dynamic reconfiguration. Figure 1 shows
the function blocks of the present prototypical microsystem.
DESIGN METHODOLOGY
The design methodology we have employed in order to
achieve the ASIC uses state-of-the-art EDA solutions that
have been selected to provide a relatively straightforward
design flow. Cadence® tools are used for analogue
simulation, digital simulation and back-end design tasks.
Synopsys® tools are used for digital synthesis of
VHDL/Verilog behavioural descriptions [7]. Foundry
services are provided via Europractice. The foundry service
provider is Austria Mikro Systeme (AMS), and the prototype
SoC discussed in this paper has been implemented on a 3 V,
2-poly, 3-metal 0.6 micrometer CMOS process. A key
advantage of using the AMS service is the availability of well
specified analogue and digital IP blocks, such as ADCs and
DACs that have been used to build our design.
RESULTS
The ASIC has been successfully fabricated and chips have
been returned both as unpackaged die and in packages for
test purposes (see Figure. 4). The ASIC is pad-limited since it
has been designed with many contact pads for test purposes.
The total area is 20 mm2. There are 16,000 gates, 60 % of
which are digital, the remaining portion being part of the
analogue system. Of the 64 pads, 20 are power pads and 44
are I/O pads, more than half of which are used for
preliminary chip test. Current consumption of the ASIC from
a 3 V source is 1.1 mA. In addition, the sensors require an
average current of 0.5 mA and the transmitter requires an
average current of 1 mA when operating at a duty cycle of
15%. The microsystem can operate for up to 10 hours under
these conditions.
CONCLUSION
We have designed and implemented an application
specific integrated circuit for a sensor microsystem for
industrial and biomedical applications. Test results
demonstrate the ASIC operates as intended. A system-onchip
design methodology has also been successfully pipecleaned
and the present design contains modules that will be
re-used in future SoC designs. Future implementations will
lead to further miniaturisation and more sophisticated system
specification including greater sensor diversity, sensor fusion
and dynamic reconfigurability, we will also extend the work
to use a more sophisticated SoC methodology to enable the
rapid prototyping of such sensor microsystems.