Speech enhancement in strong noise condition is a challenging problem. Low-rank and sparse matrix decomposition (LSMD) theory has been applied to speech enhancement recently and good performance was obtained. Existing LSMD algorithms consider each frame as an individual observation. However, real-world speeches usually have a temporal structure, and their acoustic characteristics vary slowly as a function of time. In this paper, we propose a temporal continuity constrained low-rank sparse matrix decomposition (TCCLSMD) based speech enhancement method. In this method, speech separation is formulated as a TCCLSMD problem and temporal continuity constraints are imposed in the LSMD process. We develop an alternative optimisation algorithm for noisy spectrogram decomposition. By means of TCCLSMD, the recovery speech spectrogram is more consistent with the structure of the clean speech spectrogram, and it can lead to more stable and reasonable results than the existing LSMD algorithm. Experiments with various types of noises show the proposed algorithm can achieve a better performance than traditional speech enhancement algorithms, in terms of yielding less residual noise and lower speech distortion.
New equivalent conditions of the asymptotical stability and stabilization of positive linear dynamical systems are investigated in this paper. The asymptotical stability of the positive linear systems means that there is a solution for linear inequalities systems. New necessary and sufficient conditions for the existence of solutions of the linear inequalities systems as well as the asymptotical stability of the linear dynamical systems are obtained. New conditions for the stabilization of the resultant closed-loop systems to be asymptotically stable and positive are also presented. Both the stability and the stabilization conditions can be easily checked by the so-called I-rank of a matrix and by solving linear programming (LP). The proposed LP has compact form and is ready to be implemented, which can be considered as an improvement of existing LP methods. Numerical examples are provided in the end to show the effectiveness of the proposed method.
The paper presents the method of assessment of learning outcomes acquirement by students. The analysis is based on the results of the final matriculation exam in mathematics. For crisp and both types of fuzzy relations, cut scores (passing scores) can be defined along with the method of preparing rankings of students. The advantage of applying type 2 fuzzy relations is the lack of the necessity for experts to agree to one level (one number) of verification of learning outcomes by items created for the examination. Based on the results of the exam and experts’ knowledge, the decision support system for calculating the levels of learning outcomes acquirement, making decisions about passing the examination and preparing rankings of students, can be developed. Additionally, the rank reversal phenomenon does not burden the proposed method.
Priority wise channelization of resources is the key to successful environmental management, especially when funds are limited. The study in hand has successfully developed an algorithmic criterion to compare hazardous effects of Municipal Solid Waste (MSW) dumping sites quantitatively. It is a Multi Criteria Analysis (MCA) that has made use of the scaling function to normalize the data values, Analytical Hierarchy Process (AHP) for assigning weights to input parameters showing their relevant importance, and Weighted Linear Combination (WLC) for aggregating the normalized scores. Input parameters have been divided into three classes namely Resident’s Concerns, Groundwater Vulnerability and Surface Facilities. Remote Sensing data and GIS analysis were used to prepare most of the input data. To elaborate the idea, four dumpsites have been chosen as case study, namely Old-FSD, New-FSD, Saggian and Mahmood Booti. The comparison has been made first at class levels and then class scores have been aggregated into environmental normalized index for environmental impact ranking. The hierarchy of goodness found for the selected sites is New-FSD > Old-FSD > Mahmood Booti > Saggian with comparative scores of goodness to environment as 36.67, 28.43, 21.26 and 13.63 respectively. Flexibility of proposed model to adjust any number of classes and parameters in one class will be very helpful for developing world where availability of data is the biggest hurdle in research based environmental sustainability planning. The model can be run even without purchasing satellite data and GIS software, with little inaccuracy, using imagery and measurement tools provided by Google Earth.
Faithfull detection of non-utilized spectrum hole in available channel is a crucial issue for cognitive radio network. Choosing the best available channel for a secondary user transmission includes settling on decision of accessible choices of free frequency spectrum based on multiple objectives. Thus channel judgment can be demonstrated as several objective decision making (MODM) problem. An ultimate goal of this exploration is to define and execute a technique for multiple objective optimizations of multiple alternative of channel decision in Adhoc cognitive radio network. After a coarse review of an articles related to the multiple objective decision making within a process of channel selection, Multiple Objective Optimization on the basis of the Ratio Analysis (MOORA) technique is taken into consideration. Some important objectives values of non-utilized spectrum collected by a fusion center are proposed as objectives for consideration in the decision of alternatives. MOORA method are applied to a matrix of replies of each channel alternatives to channel objectives which results in set ratios. Among the set of obtained dimensionless ratios, all the channel alternatives are ranked in descending order. In MOORA, channel choices with moderate objectives can top in ranking order, which is hardly conceivable with linearly weighted objectives of the different channel by using different decision making technique.
Poland is characterized by a number of factors which adversely affect the agricultural economy, so this paper will aim to present the possibilities of using multi-criteria decision-making methods of Analytical Hierarchy Process (AHP) in the analysis of the spatial structure of rural areas. AHP is a widely used tool for making complex decisions based on a large number of criteria, such as, for example, land consolidation works on fragmented agricultural land. The first step is to formulate the decision-making process, then the assessment criteria and the solution variants guided by expert knowledge are determined.
A ranking, according to which the order of land consolidation and land exchange works in the studied area should be determined, will be defined by using decision-making models of the AHP method. The basis for calculations will be the weights received for the factors/parameters defined for the five thematic groups. Calculations for individual villages will bemade, and then the obtained results will allow creating a ranking for the studied commune, allowing for the effective (in terms of economic and socio-economic) spending of funds for this purpose. The presented method can be successfully used to conduct analogous analyses for any area.
One of the strategic decisions of any organization is decision making about manufacturing
strategy. Manufacturing strategy is a perspective distinguishing a company from other
present companies in that industry and creates a kind of stability in decisions and gives a special
direction to organizational activities. SIR (SUPERIORITY& INFERIORITY Ranking)
method and their applications have attracted much attention from academics and practitioners.
FSIR proves to be a very useful method for multiple criteria decision making in fuzzy
environments, which has found substantial applications in recent years. This paper proposes
a FSIR approach based methodology for TOPSIS, which using MILTENBURG Strategy
Worksheet in order to analyzing of the status of strategy of the Gas Company. Then formulates
the priorities of a fuzzy pair-wise comparison matrix as a linear programming and
derives crisp priorities from fuzzy pair-wise comparison matrices
Manufacturing levers (Alternatives) are examined and analyzed as the main elements of
manufacturing strategy. Also, manufacturing outputs (Criteria are identified that are competitive
priorities of production of any organization. Next, using a hybrid approach of FSIR
and TOPSIS, alternatives (manufacturing levers) are ranked. So dealing with the selected
manufacturing levers and promoting them, an organization makes customers satisfied with
the least cost and time.
Indian SMEs are going to play pivotal role in transforming Indian economy and achieving
double digit growth rate in near future. Performance of Indian SMEs is vital in making
India as a most preferred manufacturing destination worldwide under India’s “Make in India
Policy”. Current research was based on Indian automotive SMEs. Indian automotive SMEs
must develop significant agile capability in order to remain competitive in highly uncertain
global environment. One of the objectives of the research was to find various enablers of
agility through literature survey. Thereafter questionnaire administered exploratory factor
analysis was performed to extract various factors of agility relevant in Indian automotive
SMEs environment. Multiple regression analysis was applied to assess the relative importance
of these extracted factors. “Responsiveness” was the most important factor followed by
“Ability to reconfigure”, “Ability to collaborate”, and “Competency”. Thereafter fuzzy logic
bases algorithm was applied to assess the current level of agility of Indian automotive SMEs.
It was found as “Slightly Agile”, which was the deviation from the targeted level of agility.
Fuzzy ranking methodology facilitated the identification & criticalities of various barriers
to agility, so that necessary measures can be taken to improve the current agility level of
Indian automotive SMEs. The current research may helpful in finding; key enablers of agility,
assessing the level of agility, and ranking of the various enablers of agility to point out the
weak zone of agility so that subsequent corrective action may be taken in any industrial
environment similar to India automotive SMEs.