Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Computing on Dirty Paper: Interference-Free Integrated Communication and Computing

Kuranage Roche Rayan Ranasinghe, Giuseppe Thadeu Freitas de Abreu, David González G., Carlo Fischione

TL;DR

The paper addresses efficient integration of communication and computation over shared wireless channels (ICC) by exploiting known interference at the transmitter. It introduces Computing on Dirty Paper, which combines Costa’s dirty paper coding with nested lattice codes to pre-cancel computing symbols at the transmitters in a SIMO uplink, enabling simultaneous data transmission and over-the-air computation of nomographic functions without mutual interference. The approach employs an LMMSE receiver and lattice-based decoding, and simulations show substantial improvements in both BER for data and MSE for function computation compared to state-of-the-art ICC methods, with performance improving as more observation slots are used. This work offers a practical pathway for low-latency, resource-efficient joint communication and computation in IoT and edge-network scenarios.

Abstract

Inspired by Costa's pioneering work on dirty paper coding (DPC), this paper proposes a novel scheme for integrated communication and computing (ICC), named Computing on Dirty Paper, whereby the transmission of discrete data symbols for communication, and over-the-air computation (AirComp) of nomographic functions can be achieved simultaneously over common multiple-access channels. In particular, the proposed scheme allows for the integration of communication and computation in a manner that is asymptotically interference-free, by precanceling the computing symbols at the transmitters (TXs) using DPC principles. A simulation-based assessment of the proposed ICC scheme under a single-input multiple-output (SIMO) setup is also offered, including the evaluation of performance for data detection, and of mean-squared-error (MSE) performance for function computation, over a block of symbols. The results validate the proposed method and demonstrate its ability to significantly outperform state-of-the-art (SotA) ICC schemes in terms of both bit error rate (BER) and MSE.

Computing on Dirty Paper: Interference-Free Integrated Communication and Computing

TL;DR

The paper addresses efficient integration of communication and computation over shared wireless channels (ICC) by exploiting known interference at the transmitter. It introduces Computing on Dirty Paper, which combines Costa’s dirty paper coding with nested lattice codes to pre-cancel computing symbols at the transmitters in a SIMO uplink, enabling simultaneous data transmission and over-the-air computation of nomographic functions without mutual interference. The approach employs an LMMSE receiver and lattice-based decoding, and simulations show substantial improvements in both BER for data and MSE for function computation compared to state-of-the-art ICC methods, with performance improving as more observation slots are used. This work offers a practical pathway for low-latency, resource-efficient joint communication and computation in IoT and edge-network scenarios.

Abstract

Inspired by Costa's pioneering work on dirty paper coding (DPC), this paper proposes a novel scheme for integrated communication and computing (ICC), named Computing on Dirty Paper, whereby the transmission of discrete data symbols for communication, and over-the-air computation (AirComp) of nomographic functions can be achieved simultaneously over common multiple-access channels. In particular, the proposed scheme allows for the integration of communication and computation in a manner that is asymptotically interference-free, by precanceling the computing symbols at the transmitters (TXs) using DPC principles. A simulation-based assessment of the proposed ICC scheme under a single-input multiple-output (SIMO) setup is also offered, including the evaluation of performance for data detection, and of mean-squared-error (MSE) performance for function computation, over a block of symbols. The results validate the proposed method and demonstrate its ability to significantly outperform state-of-the-art (SotA) ICC schemes in terms of both bit error rate (BER) and MSE.

Paper Structure

This paper contains 12 sections, 18 equations, 4 figures.

Table of Contents

  1. Introduction
  2. System and Signal Models
  3. Uplink ICC Signal Model
  4. SotA Transmit Signal Design
  5. Description of the AirComp Operation
  6. Computing on Dirty Paper
  7. Fundamentals on Dirty Paper Coding
  8. Fundamentals on Nested Lattice Codes
  9. Transmitter Design
  10. Receiver Design
  11. Performance Analysis
  12. Conclusion

Figures (4)

  • Figure 1: SIMO ICC system consisting of a RX with $N$ antennas and $K$ single antenna users.
  • Figure 2: Nested lattice structure with coarse lattice $\Lambda_c$ and fine lattice $\Lambda_f$.
  • Figure 3: Nested lattice structure with coarse lattice $\Lambda_c$ and fine lattice $\Lambda_f$.
  • Figure 4: Comparison of BER and MSE performances of SotA ranasinghe2025flexibledesignframeworkintegrated and proposed computing on dirty paperICC schemes.