Q-SCA: Incorporating QoS support into software communications architecture for SDR waveform processing

  • Authors:
  • Jaesoo Lee;Saehwa Kim;Jiyong Park;Seongsoo Hong

  • Affiliations:
  • Real-Time Operating Systems Laboratory, School of Electrical Engineering and Computer Science, Seoul National University, Seoul, Korea 151-742;Real-Time Operating Systems Laboratory, School of Electrical Engineering and Computer Science, Seoul National University, Seoul, Korea 151-742;Real-Time Operating Systems Laboratory, School of Electrical Engineering and Computer Science, Seoul National University, Seoul, Korea 151-742;Real-Time Operating Systems Laboratory, School of Electrical Engineering and Computer Science, Seoul National University, Seoul, Korea 151-742

  • Venue:
  • Real-Time Systems
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The Software Communications Architecture (SCA) defined by Joint Tactical Radio Systems (JTRS) is the de facto standard middleware currently adopted by the Software Defined Radio (SDR) Forum, and is widely accepted as a viable solution to reconfigurable component-based distributed computing for adaptive wireless terminals and base stations. While SDR is heavily involved in real-time signal processing, the current SCA lacks QoS capabilities in terms of both QoS specification and enforcement. In this paper, we propose Q-SCA (QoS enabled SCA) to address this problem. Specifically, we present an application model for SDR waveform software, and then extend the SCA core framework for QoS specification and enforcement. Q-SCA supports QoS capabilities by (1) providing a QoS descriptor that is backward compatible with SCA's original domain profiles, (2) offering services for admission control and resource allocation that are used throughout the application instantiation process, and (3) introducing a mechanism to enforce the result of the resource allocation. We have fully implemented Q-SCA and performed measurements to quantify its run-time performance. Our implementation clearly shows the viability of Q-SCA.