Clustering offers a general methodology and comes with remarkably rich conceptual algorithmic framework for data analysis interpretation. As one of the most representative algorithms fuzzy clustering, C-means (FCM) is widely used objective function-based clustering method exploited in various applications. In this study, virtual-based algorithm proposed to improve classification performance coming as result using clustering. This improvement achieved by forming virtual space based on...

Performance of direction arrival (DOA) estimation is one the most important issues in array signal processing. Subspace-based algorithms provide a good compromise between accuracy and computational complexity. However, these methods are exposed to performance breakdown at low SNR scenarios. A major reason for such subspace swap phenomenon (intersubspace leakage). In this article, we elaborate on novel modified model through exploiting supervised index DOA. With developed refine so as enhance...

The separation of signal-and-noise subspaces is a crucial step in many array signal processing applications, since the performance most high-resolution methods mainly depends on accuracy separated subspaces. If resulting subspace inaccurately estimated, subspace-based algorithms would likely fail. In this paper, we propose high-accuracy method to complete developed scheme, eigenvalues covariance matrix data received by uniform linear are first employed construct an ideal sample space using...

The traditional 2D MUSIC algorithm fixes the azimuth or elevation, and searches for other without considering directions of sources. A spectrum peak diffusion effect phenomenon is observed may be utilized to detect approximate Accordingly, a fast algorithm, which performs elevation simultaneous (henceforth referred as AESS) based on only three rounds search proposed. Firstly, AESS along circle source directions. Then, subsequent launched several straight lines these Finally, Direction...

In this article, an augmented subspace-based algorithm for detecting the direction of arrival (DOA) signals is presented. developed scheme, a uniform circular antenna array with full range lateral capacity first transformed into linear in which steering matrix has Vandermonde form. This kind structure can be exploited to formulate computationally efficient search-free estimation methods. process virtual transformation, novel DOA unit assumed sector where detected are located built and...

In this paper, a new sparse array structure, which further improves the degrees of freedom (DOFs) and enhanced DOA estimation performance, for fourth-order cumulant based direction arrival (DOA) is proposed. The new-formed hole-free can achieve large consecutive range in its difference coarray. By analyzing second-order sum coarray coarray, closed form expression physical sensor locations corresponding virtual configurations are derived. Compared with existing structures, such as FLNA...

As a new network paradigm, software-defined networking (SDN) technology has been increasingly adopted. Unfortunately, SDN-enabled networks are more prone to threats from DDoS attacks than traditional due the nature of centralized management. We propose an integrated defense framework detect and mitigate various types in networks. The proposed deploys two technical modules control plane SDN for defending against high-rate low-rate attacks, respectively. former module consists three...

The device-free localization (DFL), i.e., localizing target without requiring attached any devices, is attractive. Current methods, however, query a significant degree of pre-deployment effort, such as the transceivers' locations, transmission power, which cost huge human effort. In this paper, we present Alico, an accurate and low DFL method that does not require building detailed fingerprints or prior knowledge deployment. key intuition (i) distorted wireless links caused by target, even...

In this study, a new robust and fast beamformer is developed for multiple‐input multiple‐output radar to improve the robustness against steering vector (SV) mismatches. A convex model with magnitude response constraints (MRC) established by using conjugate symmetry characteristic of transmitted–received SV. To reduce computational complexity number samples required, full‐dimensional weight (WV) expressed as Kronecker product transmitted received array WVs. Thus, problem converted into...

We propose a method of bi-fuzzy clustering to enhance the classification performance fuzzy clustering. In proposed scheme, we first build supervised learning mode granulation-degranulation mechanisms. Both granulation and degranulation are considered as generalized rotation between data subspace prototype with membership matrix. A rotational invariance form is derived granulation-degranulation. The implies that improve mechanisms diagonalize invariant matrix subspace. Then, joint...

Wireless localization technology has become an important part of pervasive computing applications. This paper presents DE <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> , a wireless approach that is responsive to single beacon conditions. Unlike previous work needs priori knowledge the scenario during pre-deployment phase with significant human effort, or uses certain number beacons high energy consumption, accomplish localization,...

Many emerging applications and the ubiquitous wireless signals have accelerated development of WiFi-based indoor localization techniques. However, there is one primary issue, that environmental dynamics will greatly undermine fingerprint stability over time, hindering these techniques to be fully practical. Although many researchers dedicated deal with adaptation problem, it still remains an open problem without a wholesome approach. In this study, we propose TaLc, time adaptive low cost...

Joint diagonalisation (JD) is an effective tool in many signal processing applications. A non-unitary JD method proposed to estimate the two-dimensional (2D) frequencies. Experiments show that algorithm achieves good performance and estimated 2D frequencies can be paired automatically.