Claims:We Claim: An Automatic Gain Controller (AGC) (204) in a baseband processing system (201) to determine gain values for a plurality of samples, said AGC (204) comprising: a synchronizer (501) configured to receive the samples and detect start of a packet based on the samples; a channel estimator (509) configured to estimate channel coefficients of a channel based on the samples of the packet; a channel characterizer (511) configured to estimate a type of the channel based on a coherence time of the channel; an a calculator (510) configured to determine a factor (a) where a is an exponential moving average factor; a power detector (502) configured to determine power levels of the samples of the packet; a variance calculator (503) configured to determine a variance of the power levels of the samples of the packet based on a in real-time; a mean calculator (504) configured to determine a mean of the power levels of the samples of the packet based on a in real-time; a comparator (505) configured to determine a maximum of the power levels of the samples of the packet; a gain value calculator (506) configured to determine all gain values corresponding to a range of power levels for the samples of the packet based on the variance, mean, and maximum of the power levels of said samples; a look-up-table (LUT) module (508) configured to store the determined gain values and the corresponding power levels of the samples, wherein the LUT module (508) provides a gain value and a time constant t based on a power level of each received packet; and a delay unit (507) configured to receive the gain value from the LUT module (508) and to provide the gain value after a predetermined time delay. The AGC (204) as claimed in claim 1, wherein the synchronizer (501) is configured to: correlate the received samples with a predefined preamble to obtain a correlation coefficient, detect the start of the packet when the correlation coefficient crosses a predefined threshold, and provide an indication of the start of the packet and the samples in the packet to the channel estimator (509) and the power detector (502). The AGC (204) as claimed in claim 2, comprising: an equalizer/demodulator (512) configured to receive the indication of the start of the packet and the samples from the synchronizer (501) and the estimated channel from the channel estimator (509) and demodulate the samples, and a packet validator (513) configured to receive the demodulated samples of the packet, detect whether the packet is valid or invalid, and provide an indication to the LUT module (508). The AGC (204) as claimed in claim 3, wherein the LUT module (508) is configured to: receive the indication from the packet validator (513), and update the gain value for all received packets until a first valid packet is detected and update the gain value only for valid received packets after the first valid packet is detected. The AGC (204) as claimed in claim 1, wherein the gain value calculator (506) is configured to calculate the gain values G for all power levels in a range [N_B, p_max] with variable gain resolution with respect to a reference power p_ref based on: g=f(p)= v(p_ref/p) Wherein g: Gain p: Power level p_ref: Reference power level. The AGC (204) as claimed in claim 5, wherein the gain resolution is based on a Probability Distribution Function (PDF) of the power levels of the samples. The AGC (204) as claimed in claim 1, wherein the time constant t is modified based on the estimated type of channel. The AGC (204) as claimed in claim 1, wherein the a calculator (510) is configured to calculate a based on the time constant t: a=1-exp((-T)/t) Wherein a: Exponential moving average factor T: Sampling time t: Time constant. The AGC (204) as claimed in claim 8, wherein the mean and the variance are calculated based on a: µ_n=(1-a) µ_(n-1)+ ap_n s_n^2=(1-a)(s_(n-1)^2+ a?(p_n-µ_(n-1))?^2) Wherein p_n: Power level of the received packet µ_n: Mean s_n^2: Variance. The AGC (204) as claimed in claim 1, wherein the type of channels is fast fading channel, medium fading channel, or slow fading channel. The AGC (204) as claimed in claim 1, wherein the power detector (502) is configured to compute the power levels of the samples of the packet based on a square law detector: p_n= 1/N ?_(i=0)^(N-1)¦(x_real^2 (i)+ x_imag^2 (i) ) Wherein N: Length of the packet x_real and x_imag: Real and Imaginary components of complex input signal x respectively. The AGC (204) as claimed in claim 1, wherein the samples are in-phase and quadrature-phase (I-Q) samples. A method for an Automatic Gain Control (AGC) in a baseband processing system (201) for determining gain values for a plurality of samples, said method performed by an AGC (204), the method comprising: receiving, by a synchronizer (501), the samples and detecting start of a packet based on the samples; estimating, by a channel estimator (509), channel coefficients of a channel based on the samples of the packet; estimating, by a channel characterizer (511), a type of channel based on a coherence time of the channel; determining, by an a calculator (510), a where a is an exponential moving average factor; determining, by a power detector (502), power levels of the samples of the packet; determining, by a variance calculator (503), a variance of the power levels of the samples of the packet based on a in real-time; determining, by a mean calculator (504), a mean of the power levels of the samples of the packet based on a in real-time; determining, by a comparator (505), a maximum of the power levels of the samples of the packet; determining, by a gain value calculator (506), gain values corresponding to a range of power levels for the samples of the packet based on the variance, mean, and maximum of the power levels of said samples; storing, by a look-up-table (LUT) module (508), the determined gain values and the corresponding power levels of the samples, wherein the LUT module (508) provides a gain value and a time constant t based on a power level of each received packet; and receiving, by a delay unit (507), the gain value from the LUT module (508) and providing the gain value after a predetermined time delay. The method as claimed in claim 13, comprising: correlating, by the synchronizer (501), the received samples with a predefined preamble to obtain a correlation coefficient; detecting, by the synchronizer (501), the start of the packet when the correlation coefficient crosses a predefined threshold; and providing, by the synchronizer (501), an indication of the start of the packet and the samples in the packet to the channel estimator (509) and the power detector (502). The method as claimed in claim 14, comprising: receiving, by an equalizer/demodulator (512), the indication of the start of the packet and the samples from the synchronizer (501) and the estimated channel from the channel estimator (509) and demodulating the samples; and receiving, by a packet validator (513), the demodulated samples of the packet, detecting whether the packet is valid or invalid, and providing an indication to the LUT module (508). The method as claimed in claim 15, comprising: receiving, by the LUT module (508), the indication from the packet validator (513); and updating, by the LUT module (508), the gain value for all received packets until a first valid packet is detected and updating the gain value only for valid received packets after the first valid packet is detected. The method as claimed in claim 13, comprising calculating, by the gain value calculator (506), the gain values G for all power levels in a range [N_B, p_max] with variable gain resolution with respect to a reference power p_ref based on: g=f(p)= v(p_ref/p) Wherein g: Gain p: Power level p_ref: Reference power level. The method as claimed in claim 17, wherein the gain resolution is based on a Probability Distribution Function (PDF) of the power levels of the samples. The method as claimed in claim 13, wherein the time constant t is modified based on the estimated type of channel. The method as claimed in claim 13, comprising calculating, by the a calculator (510), a based on the time constant t: a=1-exp((-T)/t) Wherein a: Exponential moving average factor T: Sampling time t: Time constant. The method as claimed in claim 20, wherein the mean and the variance are calculated based on a: µ_n=(1-a) µ_(n-1)+ ap_n s_n^2=(1-a)(s_(n-1)^2+ a?(p_n-µ_(n-1))?^2) Wherein p_n: Power level of the received packet µ_n: Mean s_n^2: Variance. The method as claimed in claim 13, wherein the type of channels is fast fading channel, medium fading channel, or slow fading channel. The method as claimed in claim 13, comprising computing, by the power detector (502), the power levels of the samples of the packet based on a square law detector: p_n= 1/N ?_(i=0)^(N-1)¦(x_real^2 (i)+ x_imag^2 (i) ) Wherein N: Length of the packet x_real and x_imag: Real and Imaginary components of complex input signal x respectively. The method as claimed in claim 13, wherein the samples are in-phase and quadrature-phase (I-Q) samples. Dated this 11th day of March 2021 For BHARAT ELECTRONICS LIMITED, By their Agent, D. MANOJ KUMAR (IN/PA-2110) KRISHNA & SAURASTRI ASSOCIATES LLP , Description:FORM 2 THE PATENTS ACT, 1970 (39 OF 1970) & THE PATENTS RULES, 2003 COMPLETE SPECIFICATION [SEE SECTION 10, RULE 13] AUTOMATIC GAIN CONTROLLER AND METHOD OF AUTOMATIC GAIN CONTROL Bharat Electronics Limited With Address: Outer Ring Road, Nagavara, Bangalore 560045, India THE FOLLOWING SPECIFICATION PARTICULARLY DESCRIBES THE INVENTION AND THE MANNER IN WHICH IT IS TO BE PERFORMED. FIELD OF INVENTION The present disclosure relates generally to digital communication systems and particularly to automatic gain control in packet-based radio communication systems. BACKGROUND Automatic Gain Control (AGC) is an integral part of a receive chain in a radio communication system. A receiver in a radio communication system caters to wide variations in incoming signal strengths of received signals. To cope up with large variations in power levels of the received signals, the AGC is used to scale up/down the received signal to a desired power level for further processing. The AGC determines power levels of the received signals and multiplies the signal with a required gain so as to meet constant output level. Figure 1 depicts a receive chain of a radio communication system. The receive chain includes an RF front end subsystem (104) and a baseband subsystem (101). The RF front end (104) down converts a signal received at an antenna (110) to an intermediate frequency (IF) level. A low noise amplifier (LNA) (109) increases sensitivity by amplifying weak signals and is an essential part of the RF front end subsystem (104) especially for higher frequency such as VHF and above. A mixer (107) beats the received signal with frequency generated by a local oscillator (108) to convert to IF. An IF-amplifier or VGA (105) scales up the received signal to a dynamic range of Analog to Digital Converter (ADC) (103) that is present in the baseband subsystem (101). The signal at the antenna (110) keeps varying due to movement of radio or due to disturbances caused by wireless channels. As the RF front end subsystem (104) is unaware of incoming signal power, there must be a way to perform gain adjustment in digital domain in processing system of the baseband subsystem (101) sub module. Figure 2 depicts a typical digital loop with an adaptive gain updated with respect to product of an error (205) and a step size (203). The error (205) is a difference between a reference power (202) and logarithm (or linear) of detected power (201). A fast or slow AGC is characterized based on the step size (203) chosen to update the gain (204). Such loop AGC requires dedicated number of symbols with sufficient duration for AGC attack time, to be added in the preamble of each packet. In a packet-based system, allocation of extra symbols for each packet results in reduced user throughput and increased latency which is not desirable for critical applications. US9490764 discloses reducing convergence of AGC by using predefined power levels computed by using statistics at the beginning of preamble and varying gain to analog VGA (Variable gain amplifier) based on pattern of received signal strength corresponding to predefined power levels. This method applies VGA gain during preamble portion of packet, and it is varied till SOP (Start of Packet) is detected. It does not consider fading channel characteristics into account and resolution of predefined power levels is not addressed. US6799023 discloses AGC implementation for cellular communication where AGC control data is computed and retained for random access burst (RACH) and fed back to AGC from processing system during normal burst (NB) so that the power level at ADC can be maintained at the reference level. This method exploits cellular packet structure to perform AGC for normal burst. Here the AGC control information is passed across different packets and does not consider fading parameters. US7386074 discloses a method of digital AGC by sending gain control information back to amplifiers present in the system like pre-amplifier and IF amplifier. Gain control is initialized and iterated multiple times until RMS value reaches reference level. This entire operation is performed during the preamble time of a data packet. This method uses combination of hardware and software to achieve the required gain instantaneously and does not consider fading parameters. Therefore, there is a need for an efficient AGC system that does not require allocating additional symbols for every packet, that enhances throughput and reduces latency. SUMMARY This summary is provided to introduce concepts related to an Automatic Gain Controller (AGC) and a method of automatic gain control. This summary is neither intended to identify essential features of the present invention nor is it intended for use in determining or limiting the scope of the present invention. In an embodiment of the present invention, an Automatic gain Controller (AGC) is provided. The AGC is in a baseband processing system to determine gain values for plurality of samples. The AGC comprises a synchronizer, a channel estimator, a channel characterizer, an a calculator, a power detector, a variance calculator, a mean calculator, a comparator, a gain value calculator, a look-up-table (LUT) module, and a delay unit. The synchronizer is configured to receive the samples and detect start of a packet based on the samples. The channel estimator is configured to estimate channel coefficients of a channel based on the samples of the packet. The channel characterizer is configured to estimate a type of the channel based on a coherence time of the channel. The a calculator is configured to determine a factor (a) where a is an exponential moving average factor. The power detector is configured to determine power levels of the samples of the packet. The variance calculator is configured to determine a variance of the power levels of the samples of the packet based on a in real-time. The mean calculator is configured to determine a mean of the power levels of the samples of the packet based on a in real-time. The comparator is configured to determine a maximum of the power levels of the samples of the packet. The gain value calculator is configured to determine all gain values corresponding to a range of power levels for the samples of the packet based on the variance, mean, and maximum of the power levels of said samples. The LUT module is configured to store the determined gain values and the corresponding power levels of the samples. The LUT module provides a gain value and a time constant t based on a power level of each received packet. The delay unit is configured to receive the gain value from the LUT module and to provide the gain value after a predetermined time delay. In an embodiment of the present invention, a method for an Automatic Gain Control (AGC) in a baseband processing system for determining gain values for a plurality of samples is provided. The method includes receiving the samples and detecting start of a packet based on the samples by a synchronizer. the method includes estimating channel coefficients of a channel based on the samples of the packet by a channel estimator. The method includes estimating a type of channel based on a coherence time of the channel by a channel characterizer. The method includes determining a by an a calculator. Here, a is an exponential moving average factor. The method includes determining a variance of the power levels of the samples of the packet based on a in real-time by a variance calculator. The method includes determining a mean of the power levels of the samples of the packet based on a in real-time by a mean calculator. The method includes determining gain values corresponding to a range of power levels for the samples of the packet based on the variance, mean, and maximum of the power levels of said samples by a gain value calculator. The method includes storing the determined gain values and the corresponding power levels of the samples by a look-up-table (LUT) module. The LUT module provides a gain value and a time constant t based on a power level of each received packet. the method includes receiving the gain value from the LUT module and providing the gain value after a predetermined time delay by a delay unit. In an embodiment, the synchronizer correlates the received samples with a predefined preamble to obtain a correlation coefficient. The synchronizer detects the start of the packet when the correlation coefficient crosses a predefined threshold. The synchronizer provides an indication of the start of the packet and the samples in the packet to the channel estimator and the power detector. In an embodiment, an equalizer/demodulator receives the indication of the start of the packet and the samples from the synchronizer and the estimated channel from the channel estimator and demodulates the samples. The packet validator receives the demodulated samples of the packet, detects whether the packet is valid or invalid, and provides an indication to the LUT module. In an embodiment, the LUT module receives the indication from the packet validator and updates the gain value for all received packets until a first valid packet is detected. The LUT module updates the gain value only for valid received packets after the first valid packet is detected. In an embodiment, the gain value calculator calculates the gain values G for all power levels in a range [N_B, p_max] with variable gain resolution with respect to a reference power p_ref based on: g=f(p)= v(p_ref/p) Wherein g: Gain p: Power level p_ref: Reference power level. In an embodiment, the gain resolution is based on a Probability Distribution Function (PDF) of the power levels of the samples. In an embodiment, wherein the time constant t is modified based on the estimated type of channel. In an embodiment, the a calculator calculates a based on the time constant t: a=1-exp((-T)/t) Wherein a: Exponential moving average factor T: Sampling time t: Time constant. In an embodiment, the mean and the variance are calculated based on a: µ_n=(1-a) µ_(n-1)+ ap_n s_n^2=(1-a)(s_(n-1)^2+ a?(p_n-µ_(n-1))?^2) Wherein p_n: Power level of the received packet µ_n: Mean s_n^2: Variance. In an embodiment, the gain resolution is based on a Probability Distribution Function (PDF) of the power levels of the samples. In an embodiment, the type of channels is fast fading channel, medium fading channel, or slow fading channel. In an embodiment, the power detector computes the power levels of the samples of the packet based on a square law detector: p_n= 1/N ?_(i=0)^(N-1)¦(x_real^2 (i)+ x_imag^2 (i) ) Wherein N: Length of the packet x_real and x_imag: Real and Imaginary components of complex input signal x respectively. In an embodiment, the samples are in-phase and quadrature-phase (I-Q) samples. BRIEF DESCRIPTION OF ACCOMPANYING DRAWINGS The detailed description is described with reference to the accompanying figures. Figure 1 illustrates a receive chain of a radio communication system. Figure 2 illustrates a typical digital loop with an adaptive gain. Figure 3 illustrates placement of an Automatic Gain Controller (AGC) in accordance with an embodiment of the present invention. Figure 4 illustrates a packet structure in accordance with an embodiment of the present invention. Figure 5 illustrates a schematic block diagram of an AGC in accordance with an embodiment of the present invention. Figure 6 illustrates a schematic diagram of gain update in accordance with an embodiment of the present invention. Figure 7 illustrates a flowchart depicting a method of updating Look-Up Table (LUT) in accordance with an embodiment of the present invention. Figure 8 illustrates a flowchart depicting a method of updating gain in accordance with an embodiment of the present invention. Figure 9 illustrates a Probability Distribution Function (PDF) of power levels and a variable resolution of gain in accordance with an embodiment of the present invention. DETAILED DESCRIPTION The various embodiments of the present disclosure provide an Automatic Gain Controller (AGC) and a method of automatic gain control. In the following description, for purpose of explanation, specific details are set forth in order to provide an understanding of the present invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced without these details. One skilled in the art will recognize that embodiments of the present invention, some of which are described below, may be incorporated into a number of systems. However, the systems and methods are not limited to the specific embodiments described herein. Further, structures and devices shown in the figures are illustrative of exemplary embodiments of the present invention and are meant to avoid obscuring of the present invention. Furthermore, connections between components and/or modules within the figures are not intended to be limited to direct connections. Rather, these components and modules may be modified, re-formatted or otherwise changed by intermediary components and modules. The appearances of the phrase “in an embodiment” in various places in the specification are not necessarily all referring to the same embodiment. The present invention provides a method for automatic adjustment of gain for packet-based radio system in a baseband processing system. The method uses dynamic Look-Up Table (LUT) digital soft feed-forward gain adjustment technique to control the difference in the received Intermediate Frequency (IF) power levels. The proposed AGC does not require extra over heads in the packet for gain adjustment. The AGC gain is selected from the LUT based on average of power calculated for valid packets. The LUT is updated based on parameters: mean, variance and maximum value of power received. The resolution of power levels to calculate gain values in the LUT is made variable proportional to the distribution of received power levels. Exponentially weighted moving average and variance is used in the system to have more weightage to most recent value. Time constant of exponential multiplier is computed based on channel characteristics. Time constant increases as channel varies from fast fading to slow fading and vice-versa. The gain is applied at the end of the packet as signal level cannot be changed while the packet is being demodulated. At the start-up of the AGC, to set into operation, the gain is updated every time a start of packet detected till the first valid packet is received. Thereafter the gain is updated only on valid packet reception. The time taken to settle the gain depends on channel characteristics and number of valid packets received as the AGC processes the packets. Referring now to Figure 3, a placement of an AGC in a processing system is shown in accordance with an embodiment of the present invention. Figure 3 illustrates a processing system (301), an Analog to Digital Converter (ADC) (302), a Digital Down Converter (DDC) (303), a digital AGC (304), and a demodulator (305). The AGC (304) receives a feedback (306) from the demodulator (305). The ADC (302) converts an analog IF signal into digital samples at required sampling rates and the digital samples are input to the processing system (FPGA/ DSP) (301) where the DDC (303) down converts the samples to 0-IF I and Q samples. The DDC (303) includes a DDS (Direct digital synthesizer) which functions as an LO (Local Oscillator) programmed to the intermediate frequency, multipliers to beat the LO with input samples to give 0-IF, band limiting filters for filtering out the out of band frequency components and down samplers for providing output I and Q samples at a lower sampling rate adequate for further processing. A combination of band limiting filter and down sampler is repeated to obtain multistage structure so as to reduce resources in the processing system (301). The 0-IF I and Q samples from the DDC (303) is input to the digital AGC (304) module where the gain is adjusted to get desired signal strength that is required for the demodulator (305). The demodulator (305) in the implemented system consists of various blocks that are required to decode transmitted data such as synchronization, channel estimation, equalization, FEC decoder etc. And the decoded bits are passed to upper layers like network layer, application layer etc. for further processing. In packet communication systems, the demodulator (305) decodes the packets and its boundaries. The AGC (304) in the present system is packet synchronized, and hence, a control signal (306) is fed back from the demodulator (305) to the AGC (304) so that the AGC (304) is evaluated only on the packets. Referring now to Figure 4, a packet structure is shown in accordance with an embodiment of the present invention. The AGC (304) is based on feed forward logic and does not require any additional symbol overhead per packet. A frame structure (401) of a packet includes a preamble (402) and a payload (403). The preamble (402) is used for synchronization (frame and carrier), channel estimation and equalization. The receive chain in the baseband is typically designed to process signals with constant power level and it is expected to be same for all the packets. Referring now to Figure 5, a schematic block diagram of the AGC (304) is shown in accordance with an embodiment of the present invention. The AGC (304) includes a synchronizer (501), a power detector (502), a variance calculator (503), a mean calculator (504), a comparator (505), a gain calculator (506), a delay unit (507), a Look-Up Table (LUT) module (508), a channel estimator (509), an a calculator (510), a channel characterizer (511), a demodulator/equalizer (512), a packet validator (513), and a logic unit (514, 515, 516, 517). Input to the AGC (304) is 0-IF I and Q samples corresponding to real and imaginary parts of the received signals, respectively. The synchronization module (501) correlates the received I-Q samples with preamble and start of the packet (SOP) is detected if the correlation coefficient crosses a threshold. In an example, the preamble sequence has high correlation properties and is long enough to give required processing gain in case of lower SNR and fading channels. Once the SOP is detected, the I-Q samples are output to the power detector (502), the channel estimator (509) and the equalizer/demodulator (512). The power detector (502) computes power? p?_n of the input samples x corresponding to the packet n using square law detector formulated as follows: p_n= 1/N ?_(i=0)^(N-1)¦(x_real^2 (i)+ x_imag^2 (i) ) where N is length of the packet and x_real and x_imag is real and imaginary components of complex input signal x respectively. The LUT module (508) containing gain values is updated dynamically by using statistical parameters mean, variance, and maximum value of power levels of the received I-Q samples. The resolution of power levels to compute gain values is a variable depending on the maximum power p_max , mean µ_p and variance s_p^2. In an embodiment, a weighted resolution is used to provide better resolution of gain values for the power levels that have more chances of occurrence. Under the assumption that probability density function (PDF) of power levels follows Gaussian distribution, weighted resolution will result in finer resolution surrounding mean power with a dispersion of standard deviation and resolution decreases towards extreme power levels. Ideally possible range of power levels is [0, p_max], but, in practical radio systems, lower limit of power is band limited noise power N_B under the assumption of AWGN noise. Hence, range of practical power levels can be defined as [N_B,p_max]. In the present system power levels with range [N_B,p_max] is partitioned into steps with weighted resolution wv. Here, w is a weight vector and v is a fundamental resolution which is a constant. Under the assumption of gaussian distribution (911), probability of power level follows 68-95-99.7 rule as shown below: pr(µ_p-s_(p ) =p = µ_p+s_(p ) ) ?0.68 pr(µ_p-?2s?_(p ) =p = µ_p+?2s?_(p ) ) ?0.95 pr(µ_p-?3s?_(p ) =p = µ_p+?3s?_(p ) ) ?0.99 Accordingly, a weight vector is chosen as: w= {¦(1/0.68 ? µ?_p-s_(p ) =p = µ_p+s_(p ) @1/0.14 ? µ?_p-?2s?_(p ) =p < µ_p-s_(p ); ? µ?_p+s_(p )

p?_max then p_n is assigned to p_max. Variance s_p^2 keeps changing over time for packet to packet, so it needs to be calculated real time. Output of the power detector is passed to both, the variance calculator (503) and the mean calculator (504). The mean µ_p and variance? s?_p^2 is computed as shown below: µ_p= 1/N ?_(i=0)^(N-1)¦p_i s_p^2= E[p^2 ]- µ_p^2 Where E[.] is expectation. However, calculation of the mean and variance is prone to loss of precision because of difference in magnitude between single sample and accumulated samples. To overcome this, an exponential moving average and variance is chosen in the system as shown below. Exponentially weighted moving average and variance gives more weightage to most recent value. µ_n=(1-a)µ_(n-1)+ ap_n s_n^2=(1-a)(s_(n-1)^2+ a?(p_n-µ_(n-1))?^2) Where a is effective length of incremental average which has bounds: 0