Echo state network (ESN), which efficiently models nonlinear dynamic systems, has been proposed as a special form of recurrent neural network. However, most of the proposed ESNs consist of complex reservoir structures...
详细信息
Echo state network (ESN), which efficiently models nonlinear dynamic systems, has been proposed as a special form of recurrent neural network. However, most of the proposed ESNs consist of complex reservoir structures, leading to excessive computational cost. Recently, minimum complexity ESNs were proposed and proved to exhibit high performance and low computational cost. In this paper, we propose a simple deterministic ESN with a loop reservoir, i.e., an ESN with an adjacent-feedback loop reservoir. The novel reservoir is constructed by introducing regular adjacent feedback based on the simplest loop reservoir. Only a single free parameter is tuned, which considerably simplifies the ESN construction. The combination of a simplified reservoir and fewer free parameters provides superior prediction performance. In the benchmark datasets and real-world tasks, our scheme obtains higher prediction accuracy with relatively low complexity, compared to the classic ESN and the minimum complexity ESN. Furthermore, we prove that all the linear ESNs with the simplest loop reservoir possess the same memory capacity, arbitrarily converging to the optimal value.
In wireless cellular networks, the interference alignment (IA) is a promising technique for interference management. A new IA scheme for downlink cellular network with multi-cell and multi-user was proposed, in the ...
详细信息
In wireless cellular networks, the interference alignment (IA) is a promising technique for interference management. A new IA scheme for downlink cellular network with multi-cell and multi-user was proposed, in the proposed scheme, the interference in the networks is divided into inter-cell interference (ICI) among cells and inter-user interference (IUI) in each cell. The ICI is aligned onto a multi-dimensional subspace by multiplying the ICI alignment precoding matrix which is designed by the singular value decomposition (SVD) scheme at the base station (BS) side. The aligned ICI is eliminated by timing the interference suppression matrix which is designed by zero-forcing (ZF) scheme at the user equipment (UE) side. Meanwhile, the IUI is aligned by multiplying the IUI alignment precoding matrix which is designed based on Nash bargaining solution (NBS) in game theory. The NBS is solved by the particle swarm optimization (PSO) method. Simulations show that, compared with the traditional ZF IA scheme, the proposed scheme can obtain higher data rate and guarantee the data rate fairness of UEs with little additional complexity.
This paper characterizes the spatially-normalized degrees of freedom (DoF) of a 2-cell, 3-user per cell multiple input multiple output (MIMO) cellular networks with M antennas at each user and N antennas at each base ...
详细信息
Wireless sensor networks consist of a large number of sensor nodes with limited energy. Because wireless sensor networks are mainly deployed in complex environments, it is difficult to charge or replace nodes. Therefo...
详细信息
Physical layer network coding (PNC) exploits the broadcast nature of wireless channel to obtain transmission efficiency gains. This paper proposes a hybrid field physical layer network coding (HFNC) approach and apply...
详细信息
ISBN:
(纸本)9781849198196
Physical layer network coding (PNC) exploits the broadcast nature of wireless channel to obtain transmission efficiency gains. This paper proposes a hybrid field physical layer network coding (HFNC) approach and applys HFNC to the multi-user relay networks. The proposed method enables a reliable transmission with variable throughput under different signal-to-noise ratios (SNR). With given constellation rotations at sources, the points of the equivalent superpositioned (ESP) constellation cluster. The HFNC-based decoding can be divided into two layers. Point-level (PL) decoding over the complex-field achieves a higher throughput as complex field network coding (CFNC), while cloud-level decoding defined in the finite-field is able to resist bad channel with a lower throughput. Excellent performances of symbol error probability (SEP) and transmission efficiency are proved by both theoretical analysis and simulation results.
Traditional Viterbi decoding algorithm degrades performance under correlated noise. Inspired by the powerful learning from data of deep neural networks, we apply denoising convolutional neural network (DnCNN) to chann...
详细信息
The emerging dense heterogeneous networks comprising different types of small cells enable consumers to employ multiple radio access technologies to communicate. In such a dymica network environment, development of ro...
详细信息
This paper introduces the popular universal forum systems in domestic mainstream forum and analyzes the unique characteristics of these forum systems. Based on these unique characteristics, we propose the concept of s...
详细信息
For multi-source relay-based cooperative communications, it is widely believed that Complex Field network Coding (CFNC) achieves better throughput performance than Physical-layer network Coding (PNC). However, traditi...
详细信息
ISBN:
(纸本)9781849198455
For multi-source relay-based cooperative communications, it is widely believed that Complex Field network Coding (CFNC) achieves better throughput performance than Physical-layer network Coding (PNC). However, traditional CFNC schemes suffer a considerable loss on the performance of symbol to error ratio (SER) and peak-toaverage power ratio (PAPR) as the number of source nodes increases. In this paper, to tackle this problem in some extent, a novel CFNC scheme, which comprises of two SER and PAPR reduction algorithms, is proposed and analyzed. Particularly, by optimizing the complex coefficient vector at the source nodes at first, the performance of SER is improved;while by shrinking the amplitude of some signals selectively, the PAPR at the destination nodes is reduced. Simulations are provided to validate our analytical results. From the simulation results, it is observed that the proposed CFNC scheme significantly improves the performance of SER and PAPR.
This paper illustrates a semi-active handover algorithm on the integrated utilization of RSS and location, according to characteristics of a cellular network and WiFi. While connecting to a cellular network, MS (Mobil...
详细信息
暂无评论