This paper presents two novel transmitter architectures based on the combination of radio-frequency pulse-width modulation and multiphase pulse-width modulation. The proposed transmitter architectures provide good amplitude resolution and large dynamic range at high carrier frequency, which is problematic with existing radio-frequency pulse-width modulation based transmitters. They also have better power efficiency and smaller chip area compared to multiphase pulse-width modulation based transmitters.

Discrete Fourier transforms of 3 and 5 points are essential building blocks in FFT implementations for standards such as 3GPP-LTE. In addition to being more complex than 2 and 4 point DFTs, these DFTs also cause problems with data access in SDR-DSPs, since the data access width, in general, is a power of 2. This work derives mappings of these DFTs to a 4-way SIMD datapath that has been designed with 2 and 4-point DFT in mind. Our instruction set proposals, based on modified Winograd DFT, achieves single cycle execution of 3-point DFTs and 2.25 cycle average execution of 5-point DFTs in a cost-effective manner by reutilizing the already available arithmetic units. This represents an approximate speed-up of 3 times compared to an SDR-DSP with only MAC-support. In contrast to our more general design, we also demonstrate that a typical single-purpose FFT-specialized 5-way architecture only delivers 9% to 25% extra performance on average, while requiring 85% more arithmetic units and a more expensive memory subsystem.

n/a

This paper presents the novel heterogeneous DSP architecture ePUMA and demonstrates its features through an implementation of sorting of larger data sets. We derive a sorting algorithm with fixed-size merging tasks suitable for distributed memory architectures, which allows very simple scheduling and predictable data-independent sorting time.The implementation on ePUMA utilizes the architecture's specialized compute cores and control cores, and local memory parallelism, to separate and overlap sorting with data access and control for close to stall-free sorting.Penalty-free unaligned and out-of-order local memory access is used in combination with proposed application-specific sorting instructions to derive highly efficient local sorting and merging kernels used by the system-level algorithm.Our evaluation shows that the proposed implementation can rival the sorting performance of high-performance commercial CPUs and GPUs, with two orders of magnitude higher energy efficiency, which would allow high-performance sorting on low-power devices.

Since the breakdown of Dennard scaling the primary design goal for processor designs has shifted from increasing performance to increasing performance per Watt. The ePUMA platform is a flexible and configurable DSP platform that tries to address many of the problems with traditional DSP designs, to increase  performance, but use less power. We trade the flexibility of traditional VLIW DSP designs for a simpler single instruction issue scheme and instead make sure that each instruction can perform more work. Multi-cycle instructions can operate directly on vectors and matrices in memory and the datapaths implement common DSP subgraphs directly in hardware, for high compute throughput. Memory bottlenecks, that are common in other architectures, are handled with flexible LUT-based multi-bank memory addressing and memory parallelism. A major contributor to energy consumption, data movement, is reduced by using heterogeneous interconnect and clustering compute resources around local memories for simple data sharing. To evaluate ePUMA we have implemented the majority of the kernel library from a commercial VLIW DSP manufacturer for comparison. Our results not only show good performance, but also an order of magnitude increase in energy- and area efficiency. In addition, the kernel code size is reduced by 91% on average compared to the VLIW DSP. These benefits makes ePUMA an attractive solution for future DSP.

Trellis codes, including Low-Density Parity-Check (LDPC), turbo, and convolutional code (CC), are widely adopted in advanced wireless standards to offer high-throughput forward error correction (FEC). Designing a multistandard FEC decoder is of great challenge. In this paper, a trellis application specified instruction-set processor (TASIP) is presented for multistandard trellis decoding. A unified forward-backward recursion kernel with an eight-state parallel trellis structure is proposed. Based on the kernel, a datapath for multialgorithm and a shared memory subsystem are introduced. The flexibility and the compatibility are guaranteed by a programmable decoding flow and the trellis decoding instruction set. Synthesis results show that the area consumption is 2.12 mm(2) (65 nm). TASIP provides trimode FEC decoding ability with the throughput of 533, 186, and 225 Mb/s for LDPC, turbo, and 64 states CC under the clock frequency of 200 MHz, which outperforms other trimode proposals both in area efficiency and recursion efficiency. TASIP provides high-throughput decoding for current standards, including 3rd Generation Partnership Project-Long Term Evolution, 802.16e, and 802.11n, with unified architecture and high compatibility.