Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Optimized Hardware Implementation of Linear Transformations

Kollias, Georgios LU and Rizvi, Hammad Ahmad LU (2026) EITM02 20261
Department of Electrical and Information Technology
Abstract
Adders are fundamental building blocks in Application-Specific Integrated Circuits (ASICs) and play a central role in the arithmetic datapaths of modern digital signal processing systems. Operations such as finite impulse response (FIR) filtering, beamforming, and signal combining rely on evaluating weighted sums of multiple inputs. These computations can be expressed as linear combinations and are typically implemented using multiply-add structures followed by multi-operand adder trees. Since such arithmetic structures are replicated extensively throughout Digital Front-End (DFE) ASICs for 5G and emerging 6G systems, even small improvements in their implementation can translate into significant reductions in silicon area and power... (More)
Adders are fundamental building blocks in Application-Specific Integrated Circuits (ASICs) and play a central role in the arithmetic datapaths of modern digital signal processing systems. Operations such as finite impulse response (FIR) filtering, beamforming, and signal combining rely on evaluating weighted sums of multiple inputs. These computations can be expressed as linear combinations and are typically implemented using multiply-add structures followed by multi-operand adder trees. Since such arithmetic structures are replicated extensively throughout Digital Front-End (DFE) ASICs for 5G and emerging 6G systems, even small improvements in their implementation can translate into significant reductions in silicon area and power consumption.
This thesis investigates the optimal pipelining and decomposition of multi-input arithmetic aggregation circuits with the objective of minimizing area, dynamic power, and leakage power. A parameterized library of arithmetic building blocks was developed that covers variations in depth, width, and architecture to serve as the design space for the optimization phase. To evaluate the implementation cost of each building block, an automated RTL-to-gate synthesis and power estimation flow was established, generating a characterization dataset across multiple design points. To further assess design robustness across operating conditions, the exploration was extended to include multiple supply voltages and clock frequencies.
Based on this dataset, a dynamic programming framework was developed to determine cost-efficient hierarchical implementations of large summation structures. The proposed methodology decomposes a target aggregation problem into two stages: an initial reduction stage composed of arithmetic compression blocks, followed by a consolidation stage that combines the intermediate results using a selected adder architecture. The optimization framework evaluates alternative partitions and architectural choices while accounting for synthesis-derived area and power costs. The results demonstrate that the proposed framework can efficiently identify implementation configurations that provide favorable area-power trade-offs while balancing the competing objectives of silicon area, dynamic power, and leakage power. The developed methodology provides a systematic approach for exploring arithmetic architectures in ASIC design and can support the implementation of energy-efficient signal-processing hardware for future communication systems. (Less)
Popular Abstract
Modern communication systems are expected to move enormous amounts of data every second while still remaining energy efficient. As wireless technology advances from 5G toward future 6G systems, the hardware inside radios and base stations must process increasingly complex signals at extremely high speed. Hidden inside these systems are millions of arithmetic operations running continuously in the background. Among the most important of these operations is addition.
Although addition sounds simple, large communication systems rarely add just two numbers at a time. Technologies such as beamforming, MIMO processing, and digital filtering require many signals to be combined simultaneously and repeatedly. Inside a modern ASIC, these arithmetic... (More)
Modern communication systems are expected to move enormous amounts of data every second while still remaining energy efficient. As wireless technology advances from 5G toward future 6G systems, the hardware inside radios and base stations must process increasingly complex signals at extremely high speed. Hidden inside these systems are millions of arithmetic operations running continuously in the background. Among the most important of these operations is addition.
Although addition sounds simple, large communication systems rarely add just two numbers at a time. Technologies such as beamforming, MIMO processing, and digital filtering require many signals to be combined simultaneously and repeatedly. Inside a modern ASIC, these arithmetic networks operate at very high clock frequencies and under strict power constraints. Even a small inefficiency in one adder structure can spread across an entire chip and significantly increase energy consumption or silicon area.
This thesis investigates how these large arithmetic structures can be organized more efficiently. The work focuses on multi-input adder trees used in digital signal-processing hardware, particularly in FIR filters and other datapaths commonly found inside digital front-ends of wireless systems. Different architectures were studied to understand how choices in pipelining, logic organization, and reduction structures influence timing performance, area utilization, and power consumption after hardware synthesis.
To explore these trade-offs systematically, a large library of diverse adder designs was developed together with an automated ASIC synthesis and power-analysis workflow. In stead of manually comparing a few design possibilities, the thesis introduces an optimization framework capable of searching through many possible architectural combinations automatically. Dynamic programming was used to determine how arithmetic operations should be partitioned across stages in order to achieve efficient implementations under different operating conditions.
The methods developed in this thesis can contribute to the design of faster and more energy-efficient ASICs for future wireless infrastructure and other high-performance digital systems where arithmetic operations dominate computational cost. (Less)
Please use this url to cite or link to this publication:
author
Kollias, Georgios LU and Rizvi, Hammad Ahmad LU
supervisor
organization
course
EITM02 20261
year
type
H2 - Master's Degree (Two Years)
subject
keywords
VLSI, ASIC, 5G/6G telecommunications, EDA, Adder Trees, Wallace Compressors, Dynamic Programming
report number
LU/LTH-EIT 2026-1150
language
English
additional info
Done in close collaboration with Ericsson AB.
Company-side supervisors:
Joakim Ahlgren
Olof Troeng
id
9236205
date added to LUP
2026-06-16 13:17:36
date last changed
2026-06-16 13:17:36
@misc{9236205,
  abstract     = {{Adders are fundamental building blocks in Application-Specific Integrated Circuits (ASICs) and play a central role in the arithmetic datapaths of modern digital signal processing systems. Operations such as finite impulse response (FIR) filtering, beamforming, and signal combining rely on evaluating weighted sums of multiple inputs. These computations can be expressed as linear combinations and are typically implemented using multiply-add structures followed by multi-operand adder trees. Since such arithmetic structures are replicated extensively throughout Digital Front-End (DFE) ASICs for 5G and emerging 6G systems, even small improvements in their implementation can translate into significant reductions in silicon area and power consumption.
This thesis investigates the optimal pipelining and decomposition of multi-input arithmetic aggregation circuits with the objective of minimizing area, dynamic power, and leakage power. A parameterized library of arithmetic building blocks was developed that covers variations in depth, width, and architecture to serve as the design space for the optimization phase. To evaluate the implementation cost of each building block, an automated RTL-to-gate synthesis and power estimation flow was established, generating a characterization dataset across multiple design points. To further assess design robustness across operating conditions, the exploration was extended to include multiple supply voltages and clock frequencies.
Based on this dataset, a dynamic programming framework was developed to determine cost-efficient hierarchical implementations of large summation structures. The proposed methodology decomposes a target aggregation problem into two stages: an initial reduction stage composed of arithmetic compression blocks, followed by a consolidation stage that combines the intermediate results using a selected adder architecture. The optimization framework evaluates alternative partitions and architectural choices while accounting for synthesis-derived area and power costs. The results demonstrate that the proposed framework can efficiently identify implementation configurations that provide favorable area-power trade-offs while balancing the competing objectives of silicon area, dynamic power, and leakage power. The developed methodology provides a systematic approach for exploring arithmetic architectures in ASIC design and can support the implementation of energy-efficient signal-processing hardware for future communication systems.}},
  author       = {{Kollias, Georgios and Rizvi, Hammad Ahmad}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Optimized Hardware Implementation of Linear Transformations}},
  year         = {{2026}},
}