5 An Example Implementation and Evaluation Result
This section describes the implementation example and the evaluation result of GPE
architecture using an atomic switch FPGA with CAS configuration [20–23]. We
selected an infrared sensor system for the implementation example [26], because
infrared sensor system is a major concern for inter-planetary missions, as well as for
the Earth observation missions. Infrared sensors are often used to investigate the
nature and the formation processes of planets and asteroids. The complementary
atomic switches (CAS) enable hardware programming without configuration memories, and almost one-tenth of lower power consumption is expected compared to
conventional re-writable FPGAs because of the elimination of configuration memories used in conventional re-writable FPGAs. This enables us to embed a processor
element on each infrared signal detector output channel.
5.1 Infrared Sensor System Implementation Using an Atomic
Switch FPGA
We selected an infrared sensor system as the motif of evaluation [26]. Infrared sensor
systems are high priority payloads for inter-planetary missions like AKATSUKI
[14], the Venus Climate Orbiter, and HAYABUSA2 [15], the asteroid probe, to
investigate the nature and the formation processes of planets and asteroids. Fast and
compact circuitry is required for onboard signal processing equipment to transmit
infrared sensor data from a satellite to ground stations due to the following factors.
Firstly, output signal calibration of infrared detectors is required to compensate for
the intensity of infrared image sensors of satellites and to get high quality data
through the limited capacity of transmission channels towards ground stations.
Secondly, principal investigators of infrared image sensor data require no degradations on the image data quality, even though the resources of the sensor system are
limited. Therefore, lossless image data compression function should be realized
without adding remarkable size, mass, and power consumption for onboard electronics of satellites.
The combination of FPGAs and MCUs are employed by AKATSUKI and
HAYABUSA2 for signal processing, whereas much smaller size and lower power
consumption are demanded for future missions to accommodate larger numbers of
sensors. We implemented a processor element with GPE architecture for the requirement, which consists of reconfigurable cluster cores and programmable-logic cells
with complementary atomic switches. Extension of operation amount has been
achieved by using the reconfigurable function with the support of high level language programming capability. A matured high level and behavioral synthesis tool
enabled the architecture design. In addition to that, the programmability of hardware
circuitry was achieved without configuration memories by adopting complementary
atomic switches. This is a noteworthy advantage for space applications which cannot
Atomic Switch FPGA: Application for IoT Sensing Systems in Space
49
This section describes the implementation example and the evaluation result of GPE
architecture using an atomic switch FPGA with CAS configuration [20–23]. We
selected an infrared sensor system for the implementation example [26], because
infrared sensor system is a major concern for inter-planetary missions, as well as for
the Earth observation missions. Infrared sensors are often used to investigate the
nature and the formation processes of planets and asteroids. The complementary
atomic switches (CAS) enable hardware programming without configuration memories, and almost one-tenth of lower power consumption is expected compared to
conventional re-writable FPGAs because of the elimination of configuration memories used in conventional re-writable FPGAs. This enables us to embed a processor
element on each infrared signal detector output channel.
5.1 Infrared Sensor System Implementation Using an Atomic
Switch FPGA
We selected an infrared sensor system as the motif of evaluation [26]. Infrared sensor
systems are high priority payloads for inter-planetary missions like AKATSUKI
[14], the Venus Climate Orbiter, and HAYABUSA2 [15], the asteroid probe, to
investigate the nature and the formation processes of planets and asteroids. Fast and
compact circuitry is required for onboard signal processing equipment to transmit
infrared sensor data from a satellite to ground stations due to the following factors.
Firstly, output signal calibration of infrared detectors is required to compensate for
the intensity of infrared image sensors of satellites and to get high quality data
through the limited capacity of transmission channels towards ground stations.
Secondly, principal investigators of infrared image sensor data require no degradations on the image data quality, even though the resources of the sensor system are
limited. Therefore, lossless image data compression function should be realized
without adding remarkable size, mass, and power consumption for onboard electronics of satellites.
The combination of FPGAs and MCUs are employed by AKATSUKI and
HAYABUSA2 for signal processing, whereas much smaller size and lower power
consumption are demanded for future missions to accommodate larger numbers of
sensors. We implemented a processor element with GPE architecture for the requirement, which consists of reconfigurable cluster cores and programmable-logic cells
with complementary atomic switches. Extension of operation amount has been
achieved by using the reconfigurable function with the support of high level language programming capability. A matured high level and behavioral synthesis tool
enabled the architecture design. In addition to that, the programmability of hardware
circuitry was achieved without configuration memories by adopting complementary
atomic switches. This is a noteworthy advantage for space applications which cannot
Atomic Switch FPGA: Application for IoT Sensing Systems in Space
49