Computing devices play a pivotal role in society. Modern computers involve digital electronic circuits operating on discrete values (binary) to encode and manipulate data. Digital circuits and signals are widespread today because of their numerous benefits over their analog counterparts, such as versatility, precision and noise robustness, flexibility in storing and processing data, security, miniaturization and performance at-scale. Digital signal processing, implemented using digital computation, is widely used in communication systems to generate and process sampled signals. In fact, digital communication is performed with the bulk of the processing using digital computation of sampled analog signals.
Analog computing is making a comeback thanks to its ability to perform energy-efficient and massively parallelized computations. Digital computers face two key limitations that may hinder their ability to meet the demand for computational power in the context of communication networks. First, the mixed-signal stages have high power consumption, due to the presence of power-hungry analog-to-digital converters (ADCs) and digital-to-analog converters (DACs). Second, their speed is limited by the clock of the digital processors and by the rate of ADCs and DACs. Analog computers manipulate signals directly in the analog domain to perform specific signal processing operations. Modern analog computing devices have been proposed in recent years based on optical systems, metamaterials and metasurfaces, resistive memory arrays, and microwave networks.
Several communities have developed modern analog computing solutions to accelerate signal processing operations as well as to foster wireless communications, such as through over-the-air computing. However, the efforts of these communities are often siloed, with limited interdisciplinarity, and this fragmentation has led to a lack of a unified model and shared theoretical understanding.
In this SIG on Analog Computing for Signal Processing and Communications, we aim at breaking those silos and provide a platform to bring together Ph.D. students, researchers, scientists and engineers in academia and industry interested in analog computing and its applications in signal processing and communications systems, such as in the lower layers of wireless systems and in particular the physical layer, and to share their ideas and discuss the major technical challenges, recent breakthroughs, new applications, open problems and challenges.
Topics of interest of this SIG include, but are not limited to:
Analog computing with metamaterials for signal processing and communications.
Analog computing with resistive memory arrays for signal processing and communications.
Analog computing with microwave networks for signal processing and communications.
Other types of analog computing for signal processing and communications.
Fundamental limits of analog computing for communications, signal processing, sensing, and imaging.
Analog computing for physical/analog neural networks.
Analog computing characterization and modeling.
Development of novel analog computing architectures with a favorable performance-complexity trade-off.
Analog computing for future beamforming and MIMO communications.
Cross-layer design, optimization, and performance analysis of analog computing for communications.
Analog computing in 6G and beyond services such as enhanced MBB, URLLC, massive MTC, massive IoT, V2X, cellular, integrated sensing and communications, and mobile edge computing.
Channel estimation for analog computing-aided wireless networks.
Artificial intelligence-empowered design for analog computing-aided wireless networks.
Analog computing-empowered artificial intelligence for signal processing and communications.
Modeling of RF impairments in analog computing.
Hardware implementation of analog computing.
Analog computing for RF, mmw, THz to optical.
Analog computing prototyping and experimentation.
Applications of analog computing in signal processing and communications.
Programmable over-the-air computing for communications.