Abstract This paper presents a design and an implementation of a hardware scheduler in VHDL for High-Speed VoQ using a modified i-SLIP algorithm. The aims for the scheduling algorithm is to match input queues with output queues to achieve the maximum throughput while maintaining stability and eliminating starvation. The N-input by N-output scheduler manages VoQ packets to avoid a HOL blocking.... This paper presents a design and an implementation of a hardware scheduler in VHDL for High-Speed VoQ using a modified i-SLIP algorithm. The aims for the scheduling algorithm is to match input queues with output queues to achieve the maximum throughput while maintaining stability and eliminating starvation. The N-input by N-output scheduler manages VoQ packets to avoid a HOL blocking. This implementation requires N Grant Arbiters, N Accept Arbiters, three input-to-output Swizzles and a FSM. The Modified i-SLIP arbiter requires perforce a state pointer for the highest priority input (or output). The simulation and synthesis results are shown for N = 8. The design was implemented in VHDL RTL, simulated with Modelsim 6.2 and synthesized using Xilinx-ISE and the Virtex-4 device XC4VFX100-12FFG1517 of 90 nm technology to achieve a maximum frequency of 265.88 MHz, a minimum slices utilization of 771 and a total estimated power consumption about 915 mW. Read More Read Less

Abstract Currently, Transmission Control Protocol, which was originally proposed for wired networks, is deployed in multi-hop wireless networks that have memory and energy constraints. The high energy consumption is a strong constraint for the deployment of TCP in low-power multi-hop wireless networks, and especially for low-power and lossy networks such as 6LoWPANs. Some researchers say that TCP is not the bestsuited... Currently, Transmission Control Protocol, which was originally proposed for wired networks, is deployed in multi-hop wireless networks that have memory and energy constraints. The high energy consumption is a strong constraint for the deployment of TCP in low-power multi-hop wireless networks, and especially for low-power and lossy networks such as 6LoWPANs. Some researchers say that TCP is not the bestsuited transport protocol for these kind of networks because of some TCP limitations such as end-to-end retransmissions and acknowledgment frequency. Recently, many works have been proposed to reduce significantly the amount of TCP acknowledgments in order to improve TCP performance. In this paper, we present a survey of recently TCP variants that were recently proposed to reduce the TCP acknowledgment. Thereafter, we present a comparative study of the most important solutions based on a metric that we have named energyefficiency, which takes into account not only the consumed energy but also the TCP throughput. Read More Read Less

Abstract In this paper, we investigate the influence of utilizing the distribution of multi-antenna interference (MAI) on the graph-based soft iterative receiver (GSIR) instead of applying the Gaussian approximation. It is shown Gaussian approximation is not only inaccurate at high signal-to-noise (SNR) ratio but also at low SNR with the availability of a priori information (during iterations). Consequently, it severely limits... In this paper, we investigate the influence of utilizing the distribution of multi-antenna interference (MAI) on the graph-based soft iterative receiver (GSIR) instead of applying the Gaussian approximation. It is shown Gaussian approximation is not only inaccurate at high signal-to-noise (SNR) ratio but also at low SNR with the availability of a priori information (during iterations). Consequently, it severely limits performance. It is demonstrated that implementing the distribution of interference at the data detection is able to double the bandwidth efficiency and improve the power efficiency dramatically. Even if the computational power is limited and a specific bandwidth efficiency is aimed, employing the distribution of interference can still achieve few dB gain with saving around 75 percentage of the available computational power compared to Gaussian approximation. Besides, a hybrid receiver that combines between employing the distribution of interference and applying the Gaussian approximation on interference is proposed. This combination is adjustable in the way that offers a real flexibility between performance and complexity. The performance characteristics of these kind of receivers are investigated via an Exit chart analysis. Additionally, Monte-Carlo simulation results showed that at BER = 10 − 3 the hybrid receiver needs only 1.9 dB to reach the performance of employing the distribution of interference with saving about 99.2 percentage of its computational power. The proposed method is applicable with any kind of GSIR with arbitrary modulation and channel coding schemes. Read More Read Less

Abstract We consider the problem of appropriate clusterhead selection in wireless ad-hoc networks where it is necessary to provide robustness in the face of topological changes caused by node motion, node failure and node insertion/removal. The main contribution of our work is a new strategy for clustering a wireless AD HOC network and improvements in WCA. We first derived a simple... We consider the problem of appropriate clusterhead selection in wireless ad-hoc networks where it is necessary to provide robustness in the face of topological changes caused by node motion, node failure and node insertion/removal. The main contribution of our work is a new strategy for clustering a wireless AD HOC network and improvements in WCA. We first derived a simple stability model and thereafter a load balancing clustering scheme. We showed that our algorithm outperforms the Weighted Clustering Algorithm (WCA) in terms of cluster formation and stability. One of the main ideas of our approach is to avoid clusterhead re-election and to reduce the computation and communication costs by implementing a non-periodic procedure for clusterhead election which is invoked on-demand. We strived to provide a trade-off between the uniformity of the load handled by the clusterheads and the connectivity of the network. Read More Read Less