Parallel decoding of turbo codes using multi-point trellis termination and collision-free interleavers

The UMTS turbo encoder is composed of parallel concatenation of two Recursive Systematic Convolutional (RSC) encoders which start and end at a known state. This trellis termination directly affects the performance of turbo codes. This paper presents performance analysis of multi-point trellis termination of turbo codes which is to terminate RSC encoders at more than one point of the current frame while keeping the interleaver length the same. For long interleaver lengths, this approach provides dividing a data frame into sub-frames which can be treated as independent blocks. A novel decoding architecture using multi-point trellis termination and collision-free interleavers is presented. Collision-free interleavers are used to solve memory collision problems encountered by parallel decoding of turbo codes. The proposed parallel decoding architecture reduces the decoding delay caused by the iterative nature and forward-backward metric computations of turbo decoding algorithms. Our simulations verified that this turbo encoding and decoding scheme shows Bit Error Rate (BER) performance very close to that of the UMTS turbo coding while providing almost %50 time saving for the 2-point termination and %80 time saving for the 5-point termination.

Increasing the speed of parallel decoding of turbo codes

Turbo codes experience a significant decoding delay because of the iterative nature of the decoding algorithms, the high number of metric computations and the complexity added by the (de)interleaver. The extrinsic information is exchanged sequentially between two Soft-Input Soft-Output (SISO) decoders. Instead of this sequential process, a received frame can be divided into smaller windows to be processed in parallel. In this paper, a novel parallel processing methodology is proposed based on the previous parallel decoding techniques. A novel Contention-Free (CF) interleaver is proposed as part of the decoding architecture which allows using extrinsic Log-Likelihood Ratios (LLRs) immediately as a-priori LLRs to start the second half of the iterative turbo decoding. The simulation case studies performed in this paper show that our parallel decoding method can provide %80 time saving compared to the standard decoding and %30 time saving compared to the previous parallel decoding methods at the expense of 0.3 dB Bit Error Rate (BER) performance degradation.

Preconditioning 2D Integer Data for Fast Convex Hull Computations

In order to accelerate computing the convex hull on a set of n points, a heuristic procedure is often applied to reduce the number of points to a set of s points, s ≤ n, which also contains the same hull. We present an algorithm to precondition 2D data with integer coordinates bounded by a box of size p × q before building a 2D convex hull, with three distinct advantages. First, we prove that under the condition min(p, q) ≤ n the algorithm executes in time within O(n); second, no explicit sorting of data is required; and third, the reduced set of s points forms a simple polygonal chain and thus can be directly pipelined into an O(n) time convex hull algorithm. This paper empirically evaluates and quantifies the speed up gained by preconditioning a set of points by a method based on the proposed algorithm before using common convex hull algorithms to build the final hull. A speedup factor of at least four is consistently found from experiments on various datasets when the condition min(p, q) ≤ n holds; the smaller the ratio min(p, q)/n is in the dataset, the greater the speedup factor achieved.