Date: Wednesday 11 March 2015
Time: 11:00 - 12:30
Location / Room: Meije
Chair:
Muhammad Shafique, Karlsruhe Institute of Technology, DE
Co-Chair:
Marcello Coppola, STMicroelectronics, FR
This session presents hardware and software architectures that enable effective implementations of multimedia and consumer electronics systems.
| Time | Label | Presentation Title Authors |
|---|---|---|
| 11:00 | 6.5.1 | DRAM OR NO-DRAM? EXPLORING LINEAR SOLVER ARCHITECTURES FOR IMAGE DOMAIN WARPING IN 28 NM CMOS Speakers: Michael Schaffner1, Frank K. Gürkaynak1, Aljoscha Smolic2 and Luca Benini3 1Swiss Federal Institute of Technology in Zurich (ETHZ), CH; 2Disney Research Zurich, CH; 3Università di Bologna / Swiss Federal Institute of Technology in Zurich (ETHZ), CH Abstract Solving large optimization problems within the energy and cost budget of mobile SoCs in real-time is a challenging task and motivates the development of specialized hardware accelerators. We present an evaluation of different linear solvers suitable for least-squares problems emanating from image processing applications such as image domain warping. In particular, we estimate implementation costs in 28 nm CMOS technology, with focus on trading on-chip memory vs. off-chip (DRAM) bandwidth. Our assessment shows large differences in circuit area, throughput and energy consumption and aims at providing a recommendation for selecting a suitable architecture. Our results emphasize that DRAM-free accelerators are an attractive choice in terms of power consumption and overall system complexity, even though they require more logic silicon area when compared to accelerators that make use of external DRAM. Download Paper (PDF; Only available from the DATE venue WiFi) |
| 11:30 | 6.5.2 | A SMALL NON-VOLATILE WRITE BUFFER TO REDUCE STORAGE WRITES IN SMARTPHONES Speakers: Mungyu Son1, Sungkwang Lee1, Kyungho Kim2, Sungjoo Yoo1 and Sunggu Lee1 1POSTECH, KR; 2Samsung Electronics, KR Abstract Storage write behavior in mobile devices, e.g., smartphones, is characterized by frequent overwrites of small data. In our work, we first demonstrate a small non-volatile write buffer is effective in coalescing such overwrites to reduce storage writes. We also present how to make the best use of write buffer resource the size of which is limited by the requirement of small form factor. We present two new methods, shadow tag and SQLite-aware buffer management both of which aim at identifying hot storage data to keep in the write buffer. We also investigate the storage behavior of multiple mobile applications and show that their interference can reduce the effectiveness of write buffer. In order to resolve this problem, we propose a new dynamic buffer allocation method. We did experiments with real mobile applications running on a smartphone and a Flash memory-based storage system and obtained average 56.2% and 50.2% reduction in storage writes in single and multiple application runs, respectively. Download Paper (PDF; Only available from the DATE venue WiFi) |
| 12:00 | 6.5.3 | CLUSTERING-BASED MULTI-TOUCH ALGORITHM FRAMEWORK FOR THE TRACKING PROBLEM WITH A LARGE NUMBER OF POINTS Speakers: Shih-Lun Huang, Sheng-Yi Hung and Chung-Ping Chen, Graduate Institute of Electronics Engineering, National Taiwan University, TW Abstract Microcontrollers (MCUs) are extensively used in consumer devices for specific purposes because they are tiny, cheap, and low-power. Any time-consuming algorithm and any large-size program are not suited for MCUs. Recently, we found that the conventional multi-touch algorithm becomes computationally expensive to handle the applications of large-sized touch panels. Although a more high-end MCU can obtain an improvement on speed, it would increase manufacturing cost and operating power consumption as well. In the whole multi-touch algorithm flow, point tracking is the most computationally expensive part. Fortunately, touch point tracking is similar to the pin-assignment problem in EDA. To accelerate tracking, we employ EDA techniques, such as clustering, to speed up our multi-touch algorithm. Besides, we prove that the tracking problem would be solved in O(n) time for practical cases and without losing its accuracy after clustering. Furthermore, we apply computational geometry techniques to develop an efficient clustering method. Experimental results show that clustering is efficient and effective. For the necessary requirement of large-area touch panels having 20 touch points, we can reduce the runtime by up to 70%. Besides, our multi-touch algorithm may support up to 80 touch points accompanied by a low-cost MCU. Download Paper (PDF; Only available from the DATE venue WiFi) |
| 12:15 | 6.5.4 | A LOW ENERGY 2D ADAPTIVE MEDIAN FILTER HARDWARE Speakers: Ercan Kalali and Ilker Hamzaoglu, Sabanci University, TR Abstract The two-dimensional (2D) spatial median filter is the most commonly used filter for image denoising. Since it is a non-linear sorting based filter, it has high computational complexity. Therefore, in this paper, we propose a novel low complexity 2D adaptive median filter algorithm. The proposed algorithm reduces the computational complexity of 2D median filter by exploiting the pixel correlations in the input image, and it produces higher quality filtered images than 2D median filter. We also designed and implemented a low energy 2D adaptive median filter hardware implementing the proposed 2D adaptive median filter algorithm. The proposed hardware is verified to work correctly on a Xilinx Zynq 7000 FPGA board. It can process 105 full HD (1920x1080) images per second in the worst case on a Xilinx Virtex 6 FPGA, and it has more than 80% less energy consumption than original 2D median filter hardware on the same FPGA. Download Paper (PDF; Only available from the DATE venue WiFi) |
| 12:30 | IP3-5, 851 | DYNAMIC RECONFIGURABLE PUNCTURING FOR SECURE WIRELESS COMMUNICATION Speakers: Liang Tang1, Jude Angelo Ambrose2, Akash Kumar1 and Sri Parameswaran2 1National University of Singapore, SG; 2University of New South Wales, AU Abstract The ubiquity of wireless devices has created security concerns on the information being transferred. It is critical to protect the secret information in every layer of wireless communication to thwart any type of attacks. A dynamic reconfigurable puncturing based security mechanism, named RePunc, is proposed in this paper to provide an extra level of security at the physical layer. RePunc utilizes the puncturing feature of Forward Error Correction (FEC) to insert the secure information in the punctured positions of the standard information encoded data. The punctured patterns are dynamically changed and passed as a secret key from the sender to the receiver. An eavesdropper will not be able to detect the transmission of the secure information since the inserted secure information will be processed as channel noise by the eavesdropper's receiver. However, the rightful receiver will be able to successfully decode the secure packets by knowingly differentiating the secure information and the standard information before the FEC decoding. A case study of RePunc implementation for WiFi communication is presented in this paper, showing the extreme high security complexity with low hardware overhead. Download Paper (PDF; Only available from the DATE venue WiFi) |
| 12:31 | IP3-6, 662 | QR-DECOMPOSITION ARCHITECTURE BASED ON TWO-VARIABLE NUMERIC FUNCTION APPROXIMATION Speakers: Jochen Rust, Frank Ludwig and Steffen Paul, University of Bremen, DE Abstract This paper presents a new approach for hardware-based QR-decomposition using an efficient computation scheme of the Givens-Rotation. In detail, the angle of rotation and its application to the Givens-Matrix are processed in a direct, straight-forward manner. High-performance signal processing is achieved by piecewise approximation of the arctangent and sine function. In order to identify appropriate function approximations, several designs with varying constraints are automatically generated and analyzed. Physical and logical synthesis is performed in a 130nm CMOS-technology. The application of our proposal in a multi-antenna mobile communication scenario highlights our work to be very efficient in terms of calculation accuracy and computation performance. Download Paper (PDF; Only available from the DATE venue WiFi) |
| 12:32 | IP3-7, 1055 | IN-PLACE MEMORY MAPPING APPROACH FOR OPTIMIZED PARALLEL HARDWARE INTERLEAVER ARCHITECTURES Speakers: Saeed Ur Rehman1, Cyrille Chavet2, Philippe Coussy2 and Awais Sani1 1Lab-STICC / Université de Bretagne Sud, PK; 2Lab-STICC / Université de Bretagne Sud, FR Abstract Due to their impressive error correction performances, turbo-codes or LDPC (Low Density Parity Check) architectures are now widely used in communication system and are one of the most critical parts of decoders. In order to achieve high throughput requirements these decoders are based on parallel architecture, which results in a major problem to be solved: parallel memory access conflicts. To solve these conflicts, different approaches have been proposed in state of the art resulting in a lot of different architectural solutions. In this article, we introduce a new class of memory mapping approach that can solve the conflicts with an optimized architecture based on in-place memory mapping for any application. Download Paper (PDF; Only available from the DATE venue WiFi) |
| 12:30 | End of session Lunch Break, Keynote lectures from 1250 - 1420 (Room Oisans) in front of the session room Salle Oisans and in the Exhibition area Coffee Break in Exhibition AreaOn all conference days (Tuesday to Thursday), coffee and tea will be served during the coffee breaks at the below-mentioned times in the exhibition area. Lunch BreakOn Tuesday and Wednesday, lunch boxes will be served in front of the session room Salle Oisans and in the exhibition area for fully registered delegates (a voucher will be given upon registration on-site). On Thursday, lunch will be served in Room Les Ecrins (for fully registered conference delegates only). Tuesday, March 10, 2015Coffee Break 10:30 - 11:30 Lunch Break 13:00 - 14:30; Keynote session from 13:20 - 14:20 (Room Oisans) sponsored by Mentor Graphics Coffee Break 16:00 - 17:00 Wednesday, March 11, 2015Coffee Break 10:00 - 11:00 Lunch Break 12:30 - 14:30, Keynote lectures from 12:50 - 14:20 (Room Oisans) Coffee Break 16:00 - 17:00 Thursday, March 12, 2015Coffee Break 10:00 - 11:00 Lunch Break 12:30 - 14:00, Keynote lecture from 13:20 - 13:50 Coffee Break 15:30 - 16:00 |