Claims:1. A processor-implemented method to assess fall risk of a person, the method comprising one or more steps of: collecting, via one or more hardware processors, a history of postural instability, a clinical diagnosis and gender of the person; receiving, via the one or more hardware processors, skeleton data of the person using a 3D motion sensor, wherein the person is performing a single limb stance (SLS) exercise; calculating, via the one or more hardware processors, a single limb stance (SLS) duration of the person using the received skeleton data of the person; analyzing, via the one or more hardware processors, dynamics of each movement of the person using a postural model; extracting, via the one or more hardware processors, one or more poses of the person from the received skeleton data through an eigenvector based decomposition; mapping, via the one or more hardware processors, each pose of the extracted one or more poses of the person with one or more predefined key poses in an eigenvector space representation to determine a deviation in the each pose of the person; determining, via the one or more hardware processors, a high and a low frequency fluctuation of joint coordinates of the person present in time series using an empirical mode decomposition based analysis, wherein a dimension of the joint coordinate is decomposed into at least first four intrinsic modes to determine the high and low frequency fluctuation present in each time series; and assessing, via the one or more hardware processors, the fall risk of the person to categorize him in to a high, medium and low fall risk group based on assessed deviation of the each mapped pose, determined the high and low frequency fluctuation of joint coordinates present in the time series, collected history of postural instability, the clinical diagnosis and gender of the person. 2. The method as claimed in claim 1, wherein the postural model is built using posture matrices of skeleton data of one or more healthy persons. 3. The method as claimed in claim 1, wherein the one or more predefined key poses are extracted from the plurality of skeleton data of one or more healthy persons through eigenvector based decomposition. 4. The method as claimed in claim 1, wherein only the Single Limb Stance (SLS) as a functional task of Berg Balance Scale (BBS) is used to collect the skeleton data of the person. 5. The method as claimed in claim 1, wherein the SLS duration is measured using an eigenvector based curvature analysis. 6. A system to assess fall risk of a person, wherein the system comprising: at least one memory with a plurality of instructions; one or more hardware processors communicatively coupled with the at least one memory to execute modules; a collection module configured to collect a history of postural instability, a clinical diagnosis and gender of the person; a receiving module configured to receive skeleton data of the person using a 3D motion sensor, wherein the person is performing a single limb stance (SLS) exercise; a calculation module configured to calculate a single limb stance (SLS) duration of the person using the determined skeleton data of the person; an analyzing module configured to analyze dynamics of each movement of the person using a postural model; an extraction module configured to extract one or more poses of the person from the received skeleton data through an eigenvector based decomposition; a mapping module configured to map each pose of the extracted one or more poses of the person with one or more predefined key poses in an eigenvector space representation to determine a deviation in the each pose of the person; a determining module configured to determine a high and a low frequency fluctuation of joint coordinates of the person present in each time series using an empirical mode decomposition based analysis, wherein a dimension of the joint coordinate is decomposed into at least first four intrinsic modes to determine the high and low frequency fluctuation present in each time series; and an assessment module configured to assess the fall risk of the person to categorize him in to a high, medium and low fall risk group based on assessed deviation of the each mapped pose, determined the high and low frequency fluctuation of joint coordinates present in each time series and collected history of postural instability, the clinical diagnosis and gender of the person. 7. The system as claimed in claim 6, wherein the postural model is built using posture matrices of skeleton data of one or more healthy persons. 8. The system as claimed in claim 6, wherein the one or more predefined key poses are extracted from the plurality of skeleton data of one or more healthy persons through eigenvector based decomposition. 9. The system as claimed in claim 6, wherein only the Single Limb Stance (SLS) as a functional task of Berg Balance Scale (BBS) is used to collect the skeleton data of the person. 10. The system as claimed in claim 6, wherein the SLS duration is measured using an eigenvector based curvature analysis. , Description:FORM 2 THE PATENTS ACT, 1970 (39 of 1970) & THE PATENT RULES, 2003 COMPLETE SPECIFICATION (See Section 10 and Rule 13) Title of invention: SYSTEM AND METHOD FOR ASSESSING FALL RISK OF A PERSON Applicant Tata Consultancy Services Limited A company Incorporated in India under the Companies Act, 1956 Having address: Nirmal Building, 9th floor, Nariman point, Mumbai 400021, Maharashtra, India The following specification particularly describes the invention and the manner in which it is to be performed. TECHNICAL FIELD The disclosure herein generally relates to the field of monitoring postural instability of a person, and, more particularly, but not specifically, a system and method for assessing fall risk of a person in terms of balance scores calculated using a single limb stance duration and a high and a low frequency fluctuation of joint coordinates of the person present in a time series. BACKGROUND Synchronized and coordinated activation of the postural muscles of the trunk and lower limbs is required for maintaining equilibrium and balance in human body. Balance problems introduce postural instability (PI) in an individual. Postural Instability (PI) is a major reason for fall in geriatric population as well as people with diseases or disorders like Parkinson’s, strokes etc. The patient who has survived the stroke disease are prone to fall due to imbalance. Therefore they require physiotherapy and other exercises. The assessment of fall risk of the person has become very critical, especially stroke survivors. The assessment of the fall risk will help the caregiver to design the training and physiotherapy for the stroke survivors. In addition to this, there is a huge need of a robust affordable arrangement to detect the early onset of impairments or functional decline to prevent falls from happening. It will further, helpful in continuous monitoring and assessment of stability over time. Generally, conventional stability indicators like Berg Balance Scale (BBS) require clinical settings with skilled personnel’s interventions to assess postural instability and finally classify the concerned person into low, medium, or high fall risk categories. Moreover, these conventional methodology to test postural instability demand a number of functional tasks to be performed by the concerned person for proper assessment of fall risk. SUMMARY Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one embodiment, a system to assess fall risk of a person is provided. The system includes at least one memory with a plurality of instructions and one or more hardware processors communicatively coupled with the at least one memory to execute modules. Further, the system comprises a collection module configured to collect a history of postural instability, a clinical diagnosis and gender of the person, a receiving module configured to receive a skeleton data of the person using a 3D motion sensor, a calculation module configured to calculate a single limb stance (SLS) duration of the person using the obtained skeleton data of the person, an analyzing module configured to analyze dynamics of each movement of the person using a postural model, an extraction module configured to extract one or more poses of the person, a mapping module configured to map each pose of the one or more poses of the person with one or more predefined key poses in an eigenvector space representation to determine a deviation in the each pose of the person. The one or more predefined key poses are extracted from the plurality of skeleton data of one or more healthy persons through eigenvector based decomposition. Furthermore, the system comprises a determining module configured to determine a high and a low frequency fluctuation of joint coordinates of the person present in a time series using an empirical mode decomposition based analysis and an assessment module configured to assess the fall risk of the person to categorize him in to a high, medium and low fall risk group based on assessed deviation of the each mapped pose, determined the high and low frequency fluctuation of joint coordinates present in the time series, collected history of postural instability, the clinical diagnosis and gender of the person. In another aspect, a processor-implemented method to assess fall risk of a person is provided. The method includes one or more steps such as collecting a history of postural instability, a clinical diagnosis and gender of the person, obtaining a skeleton data of the person using a 3D motion sensor, wherein the person is performing a single limb stance (SLS) exercise, calculating a single limb stance (SLS) duration of the person using the obtained skeleton data of the person, analyzing dynamics of each movement of the person using a postural model, wherein the postural model is built with a plurality of skeleton data of one or more healthy persons, extracting one or more poses of the person, mapping each pose of a person with one or more predefined key poses in an eigenvector space representation to determine a deviation in the each pose of the person, wherein the one or more predefined key poses are extracted from the plurality of skeleton data of one or more healthy persons through eigenvector based decomposition, determining a high and a low frequency fluctuation of joint coordinates of the person present in a time series using an empirical mode decomposition (EMD) based analysis, wherein each dimension of the joint coordinate is decomposed into at least first four intrinsic modes to determine the high and low frequency fluctuation present in the time series and assessing the fall risk of the person to categorize him in to a high, medium and low fall risk group based on assessed deviation of the each mapped pose, determined the high low frequency fluctuation of joint coordinates present in the time series, collected history of postural instability, the clinical diagnosis and gender of the person. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles: FIG. 1 illustrates an exemplary system to assess fall risk of a person, according to some embodiments of the present disclosure; FIG. 2 is a schematic diagram to show empirical mode decomposition (EMD) into four modes of a hip joint according to an embodiments of the present disclosure; and FIG. 3(a) & 3(b) is a flow diagram to illustrate a method to assess fall risk of a person in accordance with some embodiments of the present disclosure. DETAILED DESCRIPTION OF EMBODIMENTS Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the spirit and scope of the disclosed embodiments. It is intended that the following detailed description be considered as exemplary only, with the true scope and spirit being indicated by the following claims. Referring now to the drawings, and more particularly to FIG. 1 through 3(a) & 3(b), where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method. Referring FIG. 1, a system (100) is configured to assess fall risk of a person. This is machine learning based approach to assess the fall risk with minimal human intervention using only single limb stance (SLS) exercise. An analysis is done based on the spatiotemporal dynamics of skeleton joint positions obtained from Kinect sensor. It would be appreciated that the machine learning arrangement would help in assessing fall risk of both patients and healthy population. Further, it will be based only on skeleton data from only one functional task of a Berg Balance Scale (BBS) namely Single Limb Stance (SLS), however, there are fourteen exercises are needed for scoring using BBS. In the preferred embodiment, the system (100) comprises at least one memory (102) with a plurality of instructions and one or more hardware processors (104) which are communicatively coupled with the at least one memory (102) to execute modules therein. In the preferred embodiment, a collection module (106) of the system (100) is configured to collect a history of postural instability, a clinical diagnosis, and gender of the person. In one aspect, wherein three binary valued features are collected based the person specific data. The first feature is “1” or “0” based on whether the person is diagnosed with some balance related impairments or not respectively. The second feature is calculated from the last six months fall risk history of the person. If the person has suffered from fall in the last six months then the feature value is “1” otherwise “0”. Finally, gender of person is also considered as one of such features, wherein “1” and “0” are representing male and female respectively. In the preferred embodiment, a receiving module (108) of the system (100) is configured to determine skeleton data of the person using a 3D motion sensor, wherein the person is performing the SLS exercise. It is to be noted that a postural model algorithm, is built using one or more postural matrices of the skeleton data of the one or more healthy persons. In an example, wherein the posture information of a person is recorded as 3D joint coordinates (xti, yti, zti) using either a Kinect Xbox 360 or Kinect Xbox One while performing the SLS exercise. Herein, i varies from 1,2,…..N, wherein N is the total number of skeleton joints and t = 1,2,…..T denotes discrete time steps for the exercise duration T. The SLS is selected from the list of tasks performed in BBS as it captures detailed information regarding postural stability. For an individual, the recorded joint coordinates are stored in a posture matrix P ? RNxT, where each element Pij ? R3. Therefore, the one or more features are derived from P to categorize an individual in high, medium, and low fall risk groups. Moreover, it would be appreciated that the choosing SLS would save time as well as increases comfort level of the person. In the preferred embodiment, a calculation module (110) of the system (100) is configured to calculate a SLS duration of the person using the determined skeleton data of the person. It would be appreciated that the SLS duration is measured using an eigenvector based curvature analysis. In the preferred embodiment, an analyzing module (112) of the system (100) is configured to analyze dynamics of each movement of the person using a postural model of the system. It would be appreciated that the eigenvector based feature provides the indication of postural instability during SLS. Since the area of base of support is less in SLS than that of bipedal stance, individuals may suffer from instability during the exercise. It has been observed that the individuals with balance problem show more drastic postural adjustments and fluctuations compared to healthy population to maintain their balance. In the preferred embodiment, an extraction module (114) of the system (100) is configured to extract one or more poses of the person from the received skeleton data through an eigenvector based decomposition. In the preferred embodiment, a mapping module (116) of the system (100) is configured to map each pose of the person with one or more predefined key poses in an eigenvector space representation. The mapping would help in determining deviation in the each pose of the person. It would be appreciated that since every pose/posture can be mapped to a point in eigenvector space, therefore the sequential changes in the posture can be represented as a trajectory in eigenvector space. A trajectory of a healthy person would be different from the trajectory of a person with balance problems. Hence, in order to measure this deviation, one or more postures of the person is compared with the eigenvector based postural model generated from healthy population. It is to be noted that the one or more predefined key poses are extracted from the plurality of skeleton data of one or more healthy persons through an eigenvector based decomposition. In one aspect, wherein a postural model is built from posture matrices P1, P2,... PM of M stable healthy individuals with BBS score 56. Initially, the skeleton data are smoothed using moving average filter and normalized to unit height and unit distance with respect to Kinect. Then all the posture matrices Pi ? i = 1,… M are registered by a key pose based approach which will provide suitable eigenvector. The full time series is quite difficult to register as the initiation (Tstart = T) and completion (Tend = T) of SLS vary from person to person. In order to select key poses, the error between consecutive frames are computed from the full time series, ei = ||P (:, i) - P (:, i+1) ||2 for i = 1,… T-1 (1) where (Tstart, Tend) SLS duration is computed using curvature analysis algorithm and the values of ei are sorted in descending order between this SLS duration. In another aspect, wherein top 10 indexes are obtained based on ei and the corresponding columns of P are selected as ten key poses. This procedure is repeated for all Pi, ?i = [1….M] and the resulting columns are stored in the key pose matrices K1, K2,….. KM (Ki ? RNx10). This ensures registration of all the key-poses among the each other. Now the mean subtracted key-poses Ai = Ki – ?, where ? =1/M ?_(i=1)^M¦K_i are converted into vector form Fi and the postural model is formed by computing appropriates basis function of T = [F1, F2, F3…. FM]T ? RMxD. In yet another aspect, wherein D = 600 for 10 key-poses with 20 joints coordinates each. If WUT (W ? RMxK, U ? RDxK) with W = TU is the representation of T in terms of basis function matrix U, then to compactly represent T, the object function is represented as, min E [||T - T? || 2 ] = min E [||?_(j=K+1)^D¦?W_j U_j ?||2] (2) where E [.] stands for exprectation and Uj, Wi are the columns of basis function and weight matrix respectively. If j in ?_(j=K+1)^D¦?W_j U_j ? represents redundant features then (2) will result a compact representation of T? with minimum residual error. Furthermore, this can be extended as, minE [?_j¦?W_j U_j ? ?_i¦?U_i^T W_i^T ?]= min ?_(j=K+1)^D¦E[Wj2] = min?_(j=K+1)^D¦U_j^T CU_j, where C = E [TT T]. If the basis function Uj is considered to be the eigenvector of the covariance matrix C, then the minimization results in min?_(j=K+1)^D¦?_j , where ? is the eigenvalues of C. Hence the minimization of (2) reduces to the eigenvalue decomposition of covariance matrix of T and U is formed based on the eigenvectors corresponding to K largest eigenvalues ?. The final SLS posture model UTFi is the projection of all Fi on U. The mean square differences between the projection of the F vector of a new subject and UTFi is considered as feature vector for that particular subject. In the preferred embodiment, a determining module (118) of the system (100) is configured to determine a high and a low frequency fluctuation of joint coordinates of the person present in time series. Herein, each dimension of the joint coordinate is decomposed into at least first four intrinsic modes to determine the high and low frequency fluctuation present in the time series. In another embodiment, wherein to characterize the high and low frequency fluctuation of different joints during SLS, empirical based decomposition based analysis is carried out on hip, knee and ankle joints; coordinates. Each dimension of every join coordinate is decomposed into four intrinsic modes to observe different frequency components present in each time series. Only four modes are considered because the residue becomes insignificant beyond the same. Referring FIG. 2, for a hip joint, wherein first mode corresponds to very high frequency fluctuation which may correspond to noise and last component is smooth enough to capture minimum fluctuation during SLS phase. Further, third component is chosen as an intermediary mode for the calculation of Zero Crossing Rate (ZCR) and spectral entropy (EN). ZCR = 1/T ?_(i=1)^T|sign(x(i))-sign(x(i-1))| (3) EN = - ?_(i=1)^TPN (i) * log_2??PN (i)? (4) where PN is the normalized Power Spectral Density (PSD) and T is the total number of samples in the signal. The mean of the ZCR and EN for all the joints in all the coordinate axis are expected to capture the jerk and vibrations included in the joints during the performance of SLS and considered as two features in the final classification task. In the preferred embodiment, an assessment module (120) of the system (100) is configured to assess the fall risk of the person based on determined deviation of the each mapped pose, determined the high and low frequency fluctuation of joint coordinates present in the time series, collected history of postural instability, the clinical diagnosis and gender of the person. Further, it categorizes the person either into a high fall risk group or a medium or a low fall risk group. In an example, wherein a total of 224 persons data collection is done using Microsoft Kinect Xbox 360 in hospital for patients with neuro-physiological disorders e.g. stroke, Parkinson and Microsoft Kinect Xbox One in a community center settings for geriatric population. After extraction of features, class labels corresponding to low (40